View Full Version : will PS3's GPU be more modern than PS2's GS for its time?
Megadrive1988
25-Dec-2004, 22:33
I was just realizing something, but correct me if you feel my thinking on this is wrong. okay, Playstation2's Graphics Synthesizer was locked-down, feature-wise, in 1997, according to what some of people on this board who are PS2 developers have said. by late 1998, I am pretty sure GS was totally completed, regardless of any final clockspeed increases like EE got in early 1999. It seems to me that PS2 chipset, the GS in particular, sat around for while until Sony was ready to release PS2 in early 2000 in Japan and late 2000 in the western world.
With Playstation3, the GPU is still being worked on by at least Nvidia if not Sony-Nvidia. Jen-Hsun said that they expect to see it in production by the end or near the end of 2005. I assume that meant calendar-year 2005. Now if PS3 launches in Japan in the first half of 2006, that would probably mean PS3's GPU is still fairly new, at least compared to PS2's GS. I also realize that the underlying architecture of PS3's GPU, which is Nvidia's nextgen architecture, has been in development for several years by several hundred Nvidia architects. which is pretty normal for graphics architecture development. but it also seems to me that PS3 GPU is being worked on up until a much closer point to PS3's release to the public, or am I not seeing something?
so the way I am seeing things (and again correct me if im mistaken) is that PS3's GPU is probably going to end up being more modern than PS2's GS.
of course, I am not saying that PS3 GPU is much newer than PS2's GS, as that is a given, obviously. what I am saying is, PS3's GPU seems to be much newer for its intended release relative to PS2's GS and the release of PS2.
PS2's GS was probably in development from 1995 to 1997-98. feature-wise it sits around for awhile, locked down. PS2 released in 2000.
Nvidia's overall nextgen architecture (beyond NV40) was probably in development from maybe early 2002, then maybe it gets finished in late 2004 or early 2005. the actual PS3 GPU is still in development, until say early to mid 2005. goes into production in late 2005. PS3 releases in 2006.
the Xbox GPU was also fairly new when Xbox released in Nov 2001. the NV2A had only just been completed in early 2001, if I recall. it seems to me that PS3 GPU is going to get finished up much closer to PS3's release compared to PS2's GS. ...opps, sorry for repeating myself. hehe.
anyone agree or disagree?
have a peacful end of the year everyone :D
version
25-Dec-2004, 22:46
only speed is important
what is modern? (3 gigapolygon is modern?)
Megadrive1988
25-Dec-2004, 22:51
speed & raw performance is not the only thing that matters anymore. Sony already had that in 1994 with the Playstation and in 2000 with the PS2.
if speed & raw performance were the only things that mattered, Sony would've probably gone it alone again. but instead, they've partnered with Nvidia. obviously Sony likes Nvidia's rendering technologies,
version
25-Dec-2004, 22:58
speed & raw performance is not the only thing that matters anymore. Sony already had that in 1994 with the Playstation and in 2000 with the PS2.
if speed & raw performance were the only things that mattered, Sony would've probably gone it alone again. but instead, they've partnered with Nvidia. obviously Sony likes Nvidia's rendering technologies,
ps3' gpu is a nvidia-sony-rambus gpu , not pc gpu
nvidia to bring pixelshader technology, sony to add the embended memory , and rambus 's RedWood connection
version
25-Dec-2004, 23:03
gs is very fast but simple, pc's gpu was modern but slow in 2000
what is a modern gpu? pixelshader is modern? dont believe
raytracing,reyes? possible
Megadrive1988
25-Dec-2004, 23:11
well, i am hoping that PS3 GPU combines the best aspects of Graphics Synthesizer (very high speed thanks to parallalism and high bandwidth thanks to eDRAM) with the best aspects of PC GPUs (many hardwired features, pixel shaders) to make an awesome GPU.
version
25-Dec-2004, 23:18
well, i am hoping that PS3 GPU combines the best aspects of Graphics Synthesizer (very high speed thanks to parallalism and high bandwidth thanks to eDRAM) with the best aspects of PC GPUs (many hardwired features, pixel shaders) to make an awesome GPU.
in ps3' gpu:
eDram 100%
pixelshaders 100%
redwood 100%
but where and how is the vertexprocessing? cell?
Megadrive1988
25-Dec-2004, 23:24
vertex / geometry / lighting processing is probably done on the Cell-based CPU. I don't know what else hundreds of Gflops would be for other than game physics & a.i. processing.
but...
the PS3 GPU should be flexible enough to do its own vertex / geometry / lighting processing in addition to pixel processing & rasterizing.
or, accept v / g / l processing assistance from the CPU, then the GPU can devote all its resouces toward pixel processing & rasterizing.
modern without nvidia tech? I wouldn't think so.
version
25-Dec-2004, 23:37
vertex / geometry / lighting processing is probably done on the Cell-based CPU. I don't know what else hundreds of Gflops would be for other than game physics & a.i. processing.
but...
the PS3 GPU should be flexible enough to do its own vertex / geometry / lighting processing in addition to pixel processing & rasterizing.
or, accept v / g / l processing assistance from the CPU, then the GPU can devote all its resouces toward pixel processing & rasterizing.
my aspect:
CELL do vertex,lighting processing , but only nurbs
send it to gpu, GPU do triangle from nurbs, and clipping and rasterizing and pixel process
if in ps3' gpu would be pixeshaders +vertexshaders+eDram then it is too slow(less pipeline)
version
26-Dec-2004, 00:22
now, nvidia use 200 million transistors on own gpu(6800) on 0.13 micron process
sony use 0.65 with 500+ million transistor, possible 64 pixelpipeline and 32 MB eDram
london-boy
26-Dec-2004, 02:18
modern without nvidia tech? I wouldn't think so.
Errr, care to explain why? Is NVIDIA the only company in the whole universe capable to produce a "modern" GPU (Whatever that means)...?
Besides, the GS was considered "modern" in a very "abstract" kind of way, because whoever designed it decided to go full-on with the polygon pushing, which at the time was several times even the most powerful GPU or processor around. Sadly it had to do away with simple features that would have made everyone's life much easier.
That's how it was romantically "modern".
So, if I understand this well, Sony had the GS completed, with everything mayor locked down, in 1998? And sat on it for more than a year?? If that´s the truth then:
1. Wow, the GS really is one heck of a great graphics chip if it was really completed back then.
2. Wasn´t this kind of dumb?? To sit on a chip, no matter how good it is, for more than a year is not a very smart thing to do, IMO.
I´m not very knowledgable on this topic, but didn´t Sony had enough time to further upgrade the GS?? How difficult is it to do that?? :?:
version
26-Dec-2004, 03:18
GS is a very good and fast chip ,
with present-day technology such as 200M transistor and 500MHZ may make faster GS ,
5 GS on one chip = 20MB eDram , 500MHZ , 80 pixelunits , 600 GB/sec , 1 GigaPoly/s , 40 Gigapixel/s !!!!
if use sony 0,65 process(400M trans 800 MHZ) ,10 GS in one chip = 40MB eDram , 800MHZ , 160 pixelunits , 2 TERAByte/sec , 3.5 GigaPoly/s , 120 Gigapixel/s !!!!
ps3's GPU will be faster ?
overclocked
26-Dec-2004, 03:38
Megadrive1988 wrote
well, i am hoping that PS3 GPU combines the best aspects of Graphics Synthesizer (very high speed thanks to parallalism and high bandwidth thanks to eDRAM) with the best aspects of PC GPUs (many hardwired features, pixel shaders) to make an awesome GPU.
I think that´s a good "overall-view" of the collaboration.
You expressed earlier your thought´s of how "modern" in relation this GPU
will be versus the GS.
The thing that comes to my mind is the visual progress we have seen with the PS2`s second/third gen games. As said many times before that was because of the unfriendly software-tools that was/is the system´s "mainproblem".
I belive besides that nVidia will have a great Gpu the most important role is that they bring the software tools there and that´s great, many was afraid Sony was too "proud" i think, but they learned there lesson from
the PS2.
Qroach wrote
modern without nvidia tech? I wouldn't think so.
I would rather say "stronger" with nVidia.
version wrote
now, nvidia use 200 million transistors on own gpu(6800) on 0.13 micron process
On 90nm they can in theory atleast double that amount while still having the same diesize, as Sony will be manufacturing the chip and we still don´t know if they will fab the chip at 65nm or planning too do a quick transition from 90nm too 65nm it´s hard to guess the complexity of it.
london-boy wrote
Besides, the GS was considered "modern" in a very "abstract" kind of way, because whoever designed it decided to go full-on with the polygon pushing, which at the time was several times even the most powerful GPU or processor around. Sadly it had to do away with simple features that would have made everyone's life much easier.
That's how it was romantically "modern".
I like that description, i think there was a great vision and passion from the people designing it.
Megadrive1988
26-Dec-2004, 05:45
to re-clarify what I was trying to say/ask in my original post: I was thinking that maybe, the PS3 GPU will be newer/fresher by the time PS3 comes out, compared to the somewhat aging-though-powerful Graphics Synthesizer when PS2 came out. in otherwords: GS was getting old in 2000, whereas the PS3 GPU will be fairly fresh in 2006.
Inane_Dork
26-Dec-2004, 06:05
will PS3's GPU be more modern than PS2's GS for its time?Not exactly. The PS3's GPU will be the firs postmodern GPU: it's whatever you make of it.
Megadrive1988 if you mean , will the ps3 gpu follow the popular trends of its age where as the gs went another way.
I'd have to say yes and no .
In 1999 when the ps2 was release the other consoles like the dreamcast and the pc sector were all about better filtering and hardwired features.
Now it looks like ati will be going with unified shaders and it seems from beyond3ds news post that nvidia doesn't like that route . But it will be as feature rich as the gpus of the time in both xenon and pc sector.
GS is a very good and fast chip ,
with present-day technology such as 200M transistor and 500MHZ may make faster GS ,
5 GS on one chip = 20MB eDram , 500MHZ , 80 pixelunits , 600 GB/sec , 1 GigaPoly/s , 40 Gigapixel/s !!!!
if use sony 0,65 process(400M trans 800 MHZ) ,10 GS in one chip = 40MB eDram , 800MHZ , 160 pixelunits , 2 TERAByte/sec , 3.5 GigaPoly/s , 120 Gigapixel/s !!!!
ps3's GPU will be faster ?
In my very rough overview:
Sony's 250nm GPU = GS > RIVA TNT2 family = nVIDIA's 250nm GPU
nVIDIA-Sony 65nm GPU > 10 GS in 65nm one chip
But I'm not sure how nVIDIA-Sony PS3 GPU is better than 10 GS in one chip, at least one thing is sure that Sony determined it's better than a bunch of GS on steroids which they demonstrated in GSCube. Though you can expect there's a trade-off in exchange of raw power, it may be better in cost, in thermal design, in game-related fixed-function efficiency, in developer-friendlyness, I don't know.
Megadrive1988
26-Dec-2004, 10:34
Megadrive1988 if you mean , will the ps3 gpu follow the popular trends of its age where as the gs went another way.
I'd have to say yes and no .
In 1999 when the ps2 was release the other consoles like the dreamcast and the pc sector were all about better filtering and hardwired features.
Now it looks like ati will be going with unified shaders and it seems from beyond3ds news post that nvidia doesn't like that route . But it will be as feature rich as the gpus of the time in both xenon and pc sector.
jvd, that is not really what i meant. what i meant was, the GPU for PS3 seems to be in development right up until almost the last minute, before PS3 is released. whereas GS was finished long before PS2 was released.
version
26-Dec-2004, 11:41
GS is a very good and fast chip ,
with present-day technology such as 200M transistor and 500MHZ may make faster GS ,
5 GS on one chip = 20MB eDram , 500MHZ , 80 pixelunits , 600 GB/sec , 1 GigaPoly/s , 40 Gigapixel/s !!!!
if use sony 0,65 process(400M trans 800 MHZ) ,10 GS in one chip = 40MB eDram , 800MHZ , 160 pixelunits , 2 TERAByte/sec , 3.5 GigaPoly/s , 120 Gigapixel/s !!!!
ps3's GPU will be faster ?
In my very rough overview:
Sony's 250nm GPU = GS > RIVA TNT2 family = nVIDIA's 250nm GPU
nVIDIA-Sony 65nm GPU > 10 GS in 65nm one chip
But I'm not sure how nVIDIA-Sony PS3 GPU is better than 10 GS in one chip, at least one thing is sure that Sony determined it's better than a bunch of GS on steroids which they demonstrated in GSCube. Though you can expect there's a trade-off in exchange of raw power, it may be better in cost, in thermal design, in game-related fixed-function efficiency, in developer-friendlyness, I don't know.
raycasting or deferred technology?
if sony-nvidia use 256 rasterizer in chip and 1-2 pixelshaderpipeline
256 rasterizers work fast and write to memory pixel 's attribute(Z,normals, texture,shaderprogramaddress etc...)
after ,gpu read pixels,attributes and send to pixelpipeline to run current shaderprogram
with procedure accessible 6-200 Gigapoly/s
this is possible?
Errr, care to explain why? Is NVIDIA the only company in the whole universe capable to produce a "modern" GPU (Whatever that means)...?
You don't like what I said? well too bad. I simply don't think sony would have been capable of producing a modern graphics chip that could compare to what a company like Nvidia or ATI could produce in the time frame required for PS3.
Logically speaking, even if Sony is capable of doing it, there wouldn't be much benefit to using nvidia. Also don't give me the entire "nvidia is supplying software" arguement, because of course they are supplying software and tools, they do that when supplying tech, just like xbox.
btw, Merry Christmas!
MrSingh
26-Dec-2004, 16:01
Sony wasn't working on the in-house GPU, it was Toshiba.
Toshiba was contracted to provide the GPU for PS3, and it was as much a surprise for the folks at Toshiba as it was for everyone else when it was announced that NVIDIA would be providing the GPU. (yes.. the folks at Toshiba are Not Happy (tm) )
Sony wasn't working on the in-house GPU, it was Toshiba.
Toshiba was contracted to provide the GPU for PS3, and it was as much a surprise for the folks at Toshiba as it was for everyone else when it was announced that NVIDIA would be providing the GPU. (yes.. the folks at Toshiba are Not Happy (tm) )
Where's your source Mr Singh?
Your yendenay...Indian talk doesn't make sense at all.Why would Sony source the contract to Toshiba when it has adequate resources to build a GPU in the very first place(based on their experience with the GS).After all Nvidia did say that they were already working on the project for some time.
However their decision to go with Nvidia was clearly understandable since they were the experts in the graphics arena.Sony wouldn't have went with some unproven GPU designer like Toshiba.They can handle it all by themselves if they wanted to.How do you know that the Toshiba camp weren't happy about this?If you did have information about their Sony-Toshiba GPU development why weren't you the first to break the Sony-Nvidia alliance news to us?
Here's an OT Q.Does ATI employ lots of Indians as their engineers and PR officers?I know Nvidia does not.
If you did have information about their Sony-Toshiba GPU development why weren't you the first to break the Sony-Nvidia alliance news to us?
And so the argument falls. How do you know he didn't call it -- many here have been speculating on this in PMs for sometime, based off a few sources. And you never ask someone how they know something, tact man, tact. :P
If you did have information about their Sony-Toshiba GPU development why weren't you the first to break the Sony-Nvidia alliance news to us?
And so the argument falls. How do you know he didn't call it -- many here have been speculating on this in PMs for sometime, based off a few sources. And you never ask someone how they know something, tact man, tact. :P
You don't speculate when you have concrete sources to prove.If someone has access to insight news about SCEI like this you should already know about the Nvidia-Sony GPU deal in the first place.No one in this forum was able to confirm this until the offcial announcement was made.
I'll take what this guy said as just pure assumption.
http://www.digitimes.com/news/a20041207PR207.html
Sony Computer Entertainment Incorporated (SCEI) and Nvidia announced today that they have been working together to develop the custom graphics processing unit (GPU) for SCEI ‘s next-generation computer entertainment system, according to a joint press release.
The collaboration between SCEI and Nvidia includes chip development, and a variety of graphics development tools and middleware for content creation, stated Ken Kutaragi, executive deputy president and COO, Sony Corporation, and president and Group CEO, SCEI.
SCEI will use the jointly-developed GPU with the recently-announced Cell processor to power its next-generation entertainment systems. The Cell processor is being co-developed by IBM, Sony Group and Toshiba, and is a multicore chip comprising a 64-bit Power processor core and multiple synergistic processor cores, according to a November 29 press release. Cell is optimized for compute-intensive workloads and broadband rich media applications, including games, movies and other forms of digital content, the press release stated.
SCEI and Nvidia are working under a broad, multi-year, royalty-bearing agreement.
The custom GPU will be manufactured at Sony Group’s Nagasaki Fab2 as well as at OTSS (the joint fabrication facility of Toshiba and Sony), the press release explained.
Toshiba hasn't been left out in this.Their OTSS plant does have access to this GPU technology as well so you can't rule them out from being involved in the development of the GPU.So why would they be not happy when they can pick up new things also with the team?
You don't speculate when you have concrete sources to prove.If someone has access to insight news about SCEI like this you should already know about the Nvidia-Sony GPU deal in the first place.No one in this forum was able to confirm this until the offcial announcement was made.
You asked why he didn't state this was going to happen before it was made public if he knew, what you don't know is that he did state as much AFAIK. Try thinking about what I'm inferring before just firing back a responce.
And OTSS is a joint venture, Toshiba really don't have much of a choice then to go along with Sony on this one for a number of reasons, their in too deep in semiconductors with Sony to back-out. This doesn't mean that Fujita is happy with the SNE decision though.
Panajev2001a
27-Dec-2004, 00:33
Sony wasn't working on the in-house GPU, it was Toshiba.
Toshiba was contracted to provide the GPU for PS3, and it was as much a surprise for the folks at Toshiba as it was for everyone else when it was announced that NVIDIA would be providing the GPU. (yes.. the folks at Toshiba are Not Happy (tm) )
I do not think that SCE switched to nVIDIA to piss Toshiba off, but I certainly want to know more about the Toshiba side of things.
How close were they to be done ? How fast was their solution ? Etc...
Megadrive1988
27-Dec-2004, 01:12
It's interesting to know that Toshiba had designed or were designing a GPU for PS3. in the past, Sony did the graphics chips for their systems. PS1, PS2 and PSP.
whereas Toshiba provided much of the design for the PS2 CPU and Sony did the GS themselves, or pretty much themselves.
I too would like to know more about the PS3 GPU solution that was in development by Toshiba. did it follow in the footsteps of PS2' GS? was it more like a PC GPU, or what. and could elements of the Toshiba GPU still end up in the Nvidia GPU for PS3?
MrSingh
27-Dec-2004, 05:00
I do not think that SCE switched to nVIDIA to piss Toshiba off, but I certainly want to know more about the Toshiba side of things.
How close were they to be done ? How fast was their solution ? Etc...
check your MSN. :P
ultimate_end
27-Dec-2004, 05:10
Sony wasn't working on the in-house GPU, it was Toshiba.
Toshiba was contracted to provide the GPU for PS3, and it was as much a surprise for the folks at Toshiba as it was for everyone else when it was announced that NVIDIA would be providing the GPU. (yes.. the folks at Toshiba are Not Happy (tm) )
:?
Fafalada
27-Dec-2004, 05:29
How about you tell me your MSN too MrSingh :P
Here (http://www.pcworld.idg.com.au/index.php/id;768221236;fp;2;fpid;1) is a link to an article in regards to Toshiba’s involvement ‘Cell’ but no specifics that they were doing the GPU.
Toshiba's forthcoming Cell processor will lead to a "Cell World" of new products, Masashi Muromachi, president and chief executive officer of Toshiba's semiconductor company, said on Thursday. The Cell chip will power Sony Computer Entertainment's PlayStation 3 game console, among other products........
Here's an OT Q.Does ATI employ lots of Indians as their engineers and PR officers?I know Nvidia does not.
I don't know what you mean by lots, but I'd assume most tech companies in the states employ a number of Indians, Nvidia included. Many folks from India come to the US for grad school and stay here to work after they graduate.
Megadrive1988
27-Dec-2004, 06:28
I do not think that SCE switched to nVIDIA to piss Toshiba off, but I certainly want to know more about the Toshiba side of things.
How close were they to be done ? How fast was their solution ? Etc...
check your MSN. :P
my MSN is supergrafx@hotmail.com
I'd love to know too :wink:
Panajev2001a
27-Dec-2004, 08:07
How about you tell me your MSN too MrSingh :P
Hey, there are enough muscles on Singh for the both of us :P.
Panajev2001a
27-Dec-2004, 08:11
I do not think that SCE switched to nVIDIA to piss Toshiba off, but I certainly want to know more about the Toshiba side of things.
How close were they to be done ? How fast was their solution ? Etc...
check your MSN. :P
Did... nothing there :P... well, I hope the answer will be "yet" ;).
hey , anyone know something that i dont know ?
my mail fxtech@tiscalinet.it :shock:
Brimstone
27-Dec-2004, 20:00
Toshiba hasn't been left out in this.Their OTSS plant does have access to this GPU technology as well so you can't rule them out from being involved in the development of the GPU.So why would they be not happy when they can pick up new things also with the team?
Toshiba, the company thrown out into the cold by the Playstation 3. No HD-DVD disc format and no CELL based GPU after years of research.
StarShoe
27-Dec-2004, 20:13
Somebody please tell me more about the Toshiba ps3 gpu too.
Darkmorph@hotmail.com
Toshiba, the company thrown out into the cold by the Playstation 3. No HD-DVD disc format and no CELL based GPU after years of research.
Thanks for the insight Deadmeat. Toshiba, the company, is guaranteed volume production at their fabs for both logic and DRAM and they have access to the Cell architecture for their own CE. Yeah, they are really out in the cold. :roll:
Toshiba, the company thrown out into the cold by the Playstation 3. No HD-DVD disc format and no CELL based GPU after years of research.
Proof?
The HD-DVD one is true though.But it's merely a rivalry between the next generation media standards.It's got nothing to do with Toshiba's semiconductor department.The whole point was brought up in another thread before to stir things up.You are sure to spot this topic in one of those Xbox fan forums. :wink:
Your wishfull thoughts filled with fanboyism can really wander. :lol:
Toshiba has never ventured into commercial high end GPUs except for mobile products and the consumer market.It doesn't make sense that they were working on one and forcing Sony into buying it.The business world doesn't work that way.You make a product based on your customer's requestnot based on your own decision.SCEI does not need it anyway since they have their own graphics department.More so I wouldn't believe Toshiba was that stupid either to have come forward and suggest the idea to Sony.
Megadrive1988
28-Dec-2004, 00:47
I really am doubting the Toshiba-GPU thing.
I believe Sony had one or more of its own designs in the works. ever since they publicly mentioned GS3 in 1999. the GS3 or Visualizer. both one and the same, or two different concepts. by Sony regardless. maybe there was some collaboration between Sony and Toshiba on a GPU, but I doubt Toshiba came up with a GPU on its own for PS3. then again, I admit I could be wrong since I know next to nothing.
Panajev2001a
28-Dec-2004, 02:06
I really am doubting the Toshiba-GPU thing.
I believe Sony had one or more of its own designs in the works. ever since they publicly mentioned GS3 in 1999. the GS3 or Visualizer. both one and the same, or two different concepts. by Sony regardless. maybe there was some collaboration between Sony and Toshiba on a GPU, but I doubt Toshiba came up with a GPU on its own for PS3. then again, I admit I could be wrong since I know next to nothing.
In this case... you "might" be wrong ;).
Dave Baumann
28-Dec-2004, 02:11
I don't know what you mean by lots, but I'd assume most tech companies in the states employ a number of Indians, Nvidia included. Many folks from India come to the US for grad school and stay here to work after they graduate.
They just bought a company based in India with about 100 or so developers, but that was mainly on the DTV development side of things.
Brimstone
28-Dec-2004, 02:33
Toshiba, the company thrown out into the cold by the Playstation 3. No HD-DVD disc format and no CELL based GPU after years of research.
Thanks for the insight Deadmeat. Toshiba, the company, is guaranteed volume production at their fabs for both logic and DRAM and they have access to the Cell architecture for their own CE. Yeah, they are really out in the cold. :roll:
What DRAM? XDR? I got the impression from from this press release (http://biz.yahoo.com/bw/041020/205314_1.html) Toshiba thinks highly of DDR II. At this point I question Toshiba's commitment to XDR.
today announced that Toshiba Corporation has selected its DDR2 interface cells for next-generation high-volume consumer applications.
To me it seems Toshiba was into the XDR bandwagon because Sony wanted it for the PS3 and the GPU.
And thanks for the Deadmeat compliment. I wish I was as smart as him. Perhaps this is evidence that the holiday season has brought out a kinder gentler Vince.
Vince and Brimstone, stop bickering back and forth with the insults. It is degrading this thread.
With that said. Toshiba has no right to be bitter about not winning the GPU for PS3. Nvidia is much better equipped to handle such a task and will do a better job of achieving Sony's wants than Toshiba is. I don't care how powerful Toshiba's GPU is/was, I doubt it would be as feature rich or as fast as what Nvidia will come up with.
Toshiba has no right to be bitter about not winning the GPU for PS3. I dunno sonic . I would be bitter if a big pile of cash wasn't given to me and was given to someone else :-)
london-boy
28-Dec-2004, 06:03
Toshiba has no right to be bitter about not winning the GPU for PS3. I dunno sonic . I would be bitter if a big pile of cash wasn't given to me and was given to someone else :-)
It's not like Sony and NVIDIA are going "NAH NAH NAH NAH NAH" at Toshiba.
Toshiba lost the run for the commission to the PS3 GPU, it must be because Sony felt NVIDIA were more suitable for it. In perfect Japanese style, Toshiba will learn and move on. No reason to be bitter. They still have a big chunk in the PS3 development afterall.
ultimate_end
28-Dec-2004, 06:17
What DRAM? XDR? I got the impression from from this press release (http://biz.yahoo.com/bw/041020/205314_1.html) Toshiba thinks highly of DDR II. At this point I question Toshiba's commitment to XDR.
today announced that Toshiba Corporation has selected its DDR2 interface cells for next-generation high-volume consumer applications.
To me it seems Toshiba was into the XDR bandwagon because Sony wanted it for the PS3 and the GPU.
:roll: Oh where to begin?
In that quote you seem to have conveniently missed the part about "high-volume consumer applications". No company is likely to use something like XDR RAM for use in such products are they? DDRII is a little cheaper and runs at relatively low voltage, it is also extremely scalable to a point. That point being where DDRII simply runs out of performance potential. That's where XDR comes in. XDR is initially intended for high performance applications such as supercomputers, high-end PCs, game consoles, networks and graphics applications. They will also apparently use it in consumer applications that use Cell processors. All of these are not "high-volume consumer applications" (except for game consoles, which Toshiba obviously don't make). Just because Toshiba doesn't put XDR into every single product they happen to make does not mean that they have no interest in XDR. Pull your head out of your a**.
Also pull your head out of your a** and realise that where Toshiba will profit most from XDR is actually in manufacture. Toshiba is no different to any other memory manufacturing company, such as Samsung and Elpida (just off the top of my head) who produces a wide range of different memories including XDR. DDR memories are currently the most widely used, so it's natural that it would represent the majority of any company's production. But that doesn't stop these companies manufacturing XDR does it? BTW, this production isn't just for PS3 (see above examples), so don't try and pull that argument.
As for the Deadmeat comment, I myself never actually believed for a second that you were Deadmeat (it's all a running joke anyway). But I have followed practically every post that D has made here (and many on other forums) and D doesn't strike me as being particularly bright. His logic is consistantly flawed, almost in a way that seems he is doing it all just for fun. Often, it seems he has a wide knowledge base, but not always and it is in those times that the futility of his arguments become truly apparent. For you to openly praise his "smartness" above your own, when I'm not sure even his close friends would do so (if he actually has any), makes me suspect you do actually in fact have some connection afterall. It's not a good look.
ultimate_end
28-Dec-2004, 06:41
Toshiba has no right to be bitter about not winning the GPU for PS3. Nvidia is much better equipped to handle such a task and will do a better job of achieving Sony's wants than Toshiba is. I don't care how powerful Toshiba's GPU is/was, I doubt it would be as feature rich or as fast as what Nvidia will come up with.
I most certainly agree that Toshiba couldn't possibly compete with nVidia for GPU design (other than integrating some manufacturing technologies that SCEI/nVidia has access to anyway), there are no two ways about it. IMHO if PS3 had a Toshiba GPU it would turn out to be an absolute disaster.
What I wish someone would tell me is why did SCEI feel the need to get Toshiba to do it in the first place? I couldn't care less what the supposed specs were, I just want to know exactly what it was that made Toshiba look so attractive. I'm not so narrow minded that I don't want to fill gaps in my knowledge that might need filling. But then, most of us mere humans may never find out the truth and therefore whether there are actually gaps in our knowledge that need filling in the first place.
Well, Toshiba does have some experience in the graphics sector. Not as much as Nvidia but it can be believed that that Toshiba's efforts are better than Sony's own efforts. I guess it all boils down if you believe Sony is capable of making their own GPU for PS3 and Sony clearly felt they aren't up to the task. They went with Toshiba in the beginning for mostly obvious reasons, such as Toshiba's involvement in CELL and the Emotion Engine. It makes clear sense., but Sony is not one to rule out other competitors that would present much better opportunities. That's where Nvidia comes in.
Someone who made a comment about HD-DVD and Toshiba not getting it in PS3 has the completely wrong idea. HD-DVD is a Toshiba developed thing partly, whereas Blue-Ray or however it's spelled is a Sony developed format. Sony will go with their own before anyone else especially considering they will be able to lower the manufacturing costs of such drives considerably faster than others.
kenneth9265_3
28-Dec-2004, 09:20
I do not think that SCE switched to nVIDIA to piss Toshiba off, but I certainly want to know more about the Toshiba side of things.
How close were they to be done ? How fast was their solution ? Etc...
check your MSN. :P
MrSingh, I would also like to know kenneth9265@hotmail.com :D
Don't you noticed that Toshiba's notebooks are equipped with Nvidia's GoForce GPUs all these while?It seems that the 2 companies are quite close when it comes to the mobile PC GPU department.Toshiba was one of the first to announce the use of Nvidia's Geforce Go series of GPUs.But they have certain trust between each other not enemies.It's not like they don't know each other.
With this I would like to say Toshiba is not left out in the PS3 GPU effort.They may be probably the ones who had suggested to Sony instead that the GPU should be designed by Nvidia from their past good experience with Nvidia's products.FYI,the Japanese has always regarded Nvidia GPUs as their prefered choice in their systems.If you're doubting this you can head to most of the the cyber cafes located in Akiba(if you have the chance) and you're sure to spot one in their computers.That's because Nvidia cards are optimised for online games like FFXI.
darkblu
28-Dec-2004, 16:10
Don't you noticed that Toshiba's notebooks are equipped with Nvidia's GoForce GPUs all these while?It seems that the 2 companies are quite close when it comes to the mobile PC GPU department.Toshiba was one of the first to announce the use of Nvidia's Geforce Go series of GPUs.But they have certain trust between each other not enemies.It's not like they don't know each other.
toshiba has zero experience with pc-class graphics hardware development. being a big pc laptops vendor they're prone to aligning themselves with one or another big pc graphics vendor if they want to have a proven graphics subsystem in their laptops.
With this I would like to say Toshiba is not left out in the PS3 GPU effort.They may be probably the ones who had suggested to Sony instead that the GPU should be designed by Nvidia from their past good experience with Nvidia's products. FYI,the Japanese has always regarded Nvidia GPUs as their prefered choice in their systems.If you're doubting this you can head to most of the the cyber cafes located in Akiba(if you have the chance) and you're sure to spot one in their computers.That's because Nvidia cards are optimised for online games like FFXI.
you surely meant to say 'online games like ffxi are optimised for nvidia cards' : )
bottomline being, i am highly sceptical toshiba would have recommended anybody else if they had the potential gs2 deal in the first place, regardless of how high toshiba thinks of their pc graphics vendor.
toshiba has zero experience with pc-class graphics hardware development. being a big pc laptops vendor they're prone to aligning themselves with one or another big pc graphics vendor if they want to have a proven graphics subsystem in their laptops.
So do you think they can compare with the likes of ATI in the next generation console GPU war?I doubt it thus the whole idea of partnering Nvidia was a smart move.They could have the potential to develop the GS2 so does SCEI.Nvidia was the most suitable of the 3.Why would they suggest a plan when they are already comfortable with the Cell development all these while?
you surely meant to say 'online games like ffxi are optimised for nvidia cards' : )
No I was refering to online games that are played by the Japaneses,Koreans and Chinese.You didn't read the whole paragraph carefully did you?
bottomline being, i am highly sceptical toshiba would have recommended anybody else if they had the potential gs2 deal in the first place, regardless of how high toshiba thinks of their pc graphics vendor.
They did not.That could probably be the answer why Sony had to find another suitable partner instead.The Cell project alone took up almost every resources that they had.
darkblu
28-Dec-2004, 17:51
toshiba has zero experience with pc-class graphics hardware development. being a big pc laptops vendor they're prone to aligning themselves with one or another big pc graphics vendor if they want to have a proven graphics subsystem in their laptops.
So do you think they can compare with the likes of ATI in the next generation console GPU war?
depends what you mean by 'the likes of ati' - most pc grahics vendors would have little chances in the console graphics market. luckily ati has artx, so that places them on a bit different gorund. so to answer your question - yes, i do think toshiba can successfully compete with most of the pc-based graphics verndors when it comes to console gpu design.
They could have the potential to develop the GS2 so does SCEI. Nvidia was the most suitable of the 3. Why would they suggest a plan when they are already comfortable with the Cell development all these while?
better price/performance offered by nvidia's new rasterizer? cell eventually turned out to be too pricy to be used for rasterization purposes? and toshiba's version possibly being cell-based? and maybe you should think how come nv turned out to be "the most suitable of the 3" in the very last moment?
you surely meant to say 'online games like ffxi are optimised for nvidia cards' : )
No I was refering to online games that are played by the Japaneses,Koreans and Chinese.You didn't read the whole paragraph carefully did you?
japanese, korean and chinese online games should be a very special breed of software then : )
i am highly sceptical toshiba would have recommended anybody else if they had the potential gs2 deal in the first place
yes, i do think toshiba can successfully compete with most of the pc-based graphics verndors when it comes to console gpu design.
Your earlier post mentioned that you're highly sceptical but now you turned it around and say that you think they can compete.Let's say I take your second point which is that Toshiba can compete.Care to explain why Sony switched to Nvidia instead?To piss Toshiba because someone in there wants the PS3 to have a miserable launch?I get it now. :) Or was it because Toshiba themselves cannot handle it?If that is so why would they be pissed when it's due to their own fault?
japanese, korean and chinese online games should be a very special breed of software then : )
Nothing special but the mindset.Notice that SquareEnix's Online games has the motto "The Way It's Meant To Be Played" on them?
http://www.playonline.com/ff11us/index.shtml
Too bad ATI doesn't. :oops:
better price/performance offered by nvidia's new rasterizer? cell eventually turned out to be too pricy to be used for rasterization purposes? and toshiba's version possibly being cell-based? and maybe you should think how come nv turned out to be "the most suitable of the 3" in the very last moment?
Cell was always going to have pixel engine of somekind.
I don't think Toshiba is incompetent or anything. My hunch is NV in dire need to replace the loss of income from failing to get Xbox2, most likely outbid Toshiba, thus making PS3 or other cell device cheaper for Sony to produce.
darkblu
28-Dec-2004, 19:57
i am highly sceptical toshiba would have recommended anybody else if they had the potential gs2 deal in the first place
i'm affraid there's been a misunderstanding. the meaning of the above is: 'if toshiba had been originally approached by sony for the development of the gs2' (a premise) then 'toshiba would have hardly given up the deal and recommended another company in their own place' (my opinion based on the above premise). i never meant to, and i didn't imply toshiba was uncapable of developing gs2 for sony.
Care to explain why Sony switched to Nvidia instead? To piss Toshiba because someone in there wants the PS3 to have a miserable launch? I get it now. :) Or was it because Toshiba themselves cannot handle it?If that is so why would they be pissed when it's due to their own fault?
read my previous post again. firstly, it has some speculations answering exactly this question of yours (why sony switched), and secondly, it clearly lacks any statements in the sense of 'toshiba couldn't handle it'.
Nothing special but the mindset. Notice that SquareEnix's Online games has the motto "The Way It's Meant To Be Played" on them?
that means that the game was optimised for nv (or at least the marketing department wants us to think so). it doesn't mean that the hw was designed for this game.
darkblu
28-Dec-2004, 19:59
My hunch is NV in dire need to replace the loss of income from failing to get Xbox2, most likely outbid Toshiba, thus making PS3 or other cell device cheaper for Sony to produce.
that would have been my second quess ; )
I don't think Nvidia undercut Toshiba on pricing. I so think Nvidia was in a strange postion of being locked out of two of three console contracts, so they had nothing to loose.
I think sony made a decision to go with proven graphics technology instead of making everything from scratch (with or without toshiba). In other words, they had to bite the bullet and go for the more expensive choice. I just can't see how this would be cheaper than going with a partner like toshiba. So I think it came down to graphical features and support (drivers tools, etc) being the deciding factor.
I think sony made a decision to go with proven graphics technology instead of making everything from scratch (with or without toshiba). In other words, they had to bite the bullet and go for the more expensive choice. I just can't see how this would be cheaper than going with a partner like toshiba. So I think it came down to graphical features and support (drivers tools, etc) being the deciding factor.
NV probably offer them the price that is cheaper than what Sony could have done internally. This can be done, because it won't just be for PS3, unlike their contract with Xbox. This will cover the entire range of Sony's Cell devices. They are going for volume, since they're not fabbing them this time around.
Sony doesn't need to bite the bullet, they've got Playstation brand. They could have tack on any pixel engine to cell and it will be good enough.
NV probably offer them the price that is cheaper than what Sony could have done internally. This can be done...
I really don't see how going outside can be cheaper than doing it internally unless what was worked on internally was inferior or late in some way.
because it won't just be for PS3, unlike their contract with Xbox. This will cover the entire range of Sony's Cell devices.
...and you think the GS3 would be just for PS3? by your argument sony would be able to reduce the cost of the GS3 by using it in many different devices.
Sony doesn't need to bite the bullet, they've got Playstation brand. They could have tack on any pixel engine to cell and it will be good enough.
Sure they needed to "bite the bullet" if they are suddenly interested in paying royaltees for graphics hardware. When you say they could have used anything tacked on, and it would be good enough, obviously that's not true otherwise they would have done that.
See that's the thing, I don't think any of those pixel engines were good enough, and that's why sony is spending more money going outside to nvidia. They wouldn't spend the money if they didn't have to. Nvidia has already stated they think they can make as much money off each chip in royaltees as they do PC graphics chips. I have my doubts on that, but if true, i don't see how that could be less expensive than sony doing "everything" in house.
I really dont' see how goign outside can be cheaper than doing it internally unless what was worked on internally was inferior in some way.
Going outside can be cheaper, you cam go outside for several reasons, but as everything in business, cost is the main reasons. Strong Yen, going outside is cheaper. The same as how IT sector been farm out to India. Outside can be cheaper.
...and you think the GS3 would be just for PS3? by your argument sony would be able to reduce the cost of the GS3 by using it in many differing devices.
Exactly, so NV must have outbid them.
See that's the thing, I don't think any of those pixel engines were good enough and that's why they are spending mroe money. Nvidia has already stated they think they can make as much money off each chip in royaltees as they do PC graphics chips. I have my doubts on that, but if true,i don't see how that could be less expensive than sony doing everything in house.
It doesn't matter how much NV makes, NV gave them an offer they can't refuse from cost point of view. If not they could have gone with other graphic vendors like Toshiba, PowerVR or ATI even. But with Cell project and PSP, Sony doesn't have infinite engineers you know.
Ok take your pick, you have a bunch of choice here for the reasons they switched to Nvidia from internal or toshiba.
a) It's cheaper
b) The GS3 was behind on schedule or Nvidia is ahead.
c) The GS3 was inferior in some way to what nvidia had to offer.
d) the GS3 was superior to what Nvidia had to offer.
There's more choices I'm sure, but this is all I had time to type.
Going outside can be cheaper, you cam go outside for several reasons, but as everything in business, cost is the main reasons. Strong Yen, going outside is cheaper. The same as how IT sector been farm out to India. Outside can be cheaper.
Yeah but we're not talking about india here. We're talking about the US. The yen isn't that strong right now. Not only that if that changes it could have very drastic affects on how much sony pays out.
Exactly, so NV must have outbid them.
I still don't see how that would be possible.
It doesn't matter how much NV makes, NV gave them an offer they can't refuse from cost point of view. If not they could have gone with other graphic vendors like Toshiba, PowerVR or ATI even.
OR sony had to bite the bullet and pay nvidia instead of doing it internally for one reason or the other. I don't see how you can ignore that. just think about the last time sony thought about going outside for graphics hardware in a console, and the reasons they considered.
But with Cell project and PSP, Sony doesn't have infinite engineers you know.
The PSP design is already finished.
Ok take your pic, you have a bunch of choice here for the reasons they switched to Nvidia from internal or toshiba.
a) It's cheaper
b) The GS3 was behind on schedule or Nvidia is ahead.
c) The GS3 was inferior in some way to what nvidia had to offer.
d) the GS3 was superior to what Nvidia had to offer.
There's more choices I'm sure, but this is all I had time to type.
The only logical choice is, its cheaper, NV is the cheapest solution for what Sony needs. No question about it. When this technology has to go to a $300 console, that has to contains Blue ray and cell chip and sell it to mass market that knows little about what's superior or inferior, cost is not only a concern, its everything.
Why R&D when you can get it for cheap. They R&D cell because its not available elsewhere. Pixel engine, several companies already have that kind of technology, just go for the cheaper package.
Yeah but we're not talking about india here. We're talking about the US. The yen isn't that strong right now. Not only that if that changes it could have very drastic affects on how much sony pays out.
USD is really poor at the moment, its no brainer to go for US company instead of Japanese company. It doesn't matter if there is a swing, since PS3 are sold in the US too.
I still don't see how that would be possible.
Its the logical explanation, in the business world.
OR sony had to bite the bullet and pay nvidia instead of doing it internally for one reason or the other.
Why are you insisting Sony has to bite a bullet ?
I don't see how you can ignore that. just think about the last time sony thought about going outside for graphics hardware in a console, and the reasons they considered.
Did they ? Sony will leverage their core competence and leverage everything that is cheaper.
The only logical choice is, its cheaper, NV is the cheapest solution for what Sony needs. No question about it
That is clearly NOT the only logical choice.
That is clearly NOT the only logical choice.
For Sony, yes it is. MS probably not.
USD is really poor at the moment, its no brainer to go for US company instead of Japanese company. It doesn't matter if there is a swing, since PS3 are sold in the US too.
like I said before, that could change. Do you really think sony isn't smart enough to realize any value added by dollar exchange could just as quickly vanish as it did appear? I think dollar exchange rate can simply be ruled out. labor tends to cost more in the US anyways.
Its the logical explanation, in the business world.
Can to back up your oopinion with some more thought? I don't think your explanation holds water.
Why are you insisting Sony has to bite a bullet ?
Why are you insisting sony is invulnerable and like eveyr othe rcompany on the plant makes decisions based on timing and cost?
Did they ? Sony will leverage their core competence and leverage everything that is cheaper.
Yes, they did. I see what you're doing. You seem to be only considering whateve rway would make sony look like this was a smart bsuiness move. I think it is a smart biz move, but I certainly don't think it's a cheaper one.
For Sony, yes it is. MS probably not.
:roll: now that's pretty narrow minded V3. MS made thier decision to use nvidia over gigiapixel due to the time frame they wanted to launch in. They bit the bullet to pay more money so they could launch when they did. I think you'd have to be a fool not to consider sony could be making a similar decision.
Why R&D when you can get it for cheap. They R&D cell because its not available elsewhere. Pixel engine, several companies already have that kind of technology, just go for the cheaper package.
I agree with this. It's not that a Sony or Toshiba are in some way or form intrinsically incapable of matching the PC vendors in terms of technology, but that the cost to compete in this game is arbitrarily high due to IP and the established brain-trust these entities have. And, as I've posited before, the level of IP and patented technology concerning many of the more 'boring' rasterization aspects of the graphic pipeline are a likely barrier IMHO.
Wow vince, when i said that MS was doing things differently this time around with the licensing of ATI graphics tech instead of purchasing chips, you came out and said "how is this any different than last time" or something along those lines. i said it was cheaper to license, but not as cheap as developing things in house and you didn't have a problem with that.
Of course time may be a factor, but i still don't see how it could potentially cost more. not when you don't have to pay out royaltees on every chip produced after the R&D is compelte.
I agree with this. It's not that a Sony or Toshiba are in some way or form intrinsically incapable of matching the PC vendors in terms of technology, but that the cost to compete in this game is arbitrarily high due to IP and the established brain-trust these entities have.
Well cost and time go together don't they? if the cost and or time invested to compete is high, then it's more worth while to license if you need it in a resonable timeframe. yes I agree with part of that but I think the no royaltee route is always mroe favourable than paying royaltees.
like I said before, that could change. Do you really think sony isn't smart enough to realize any value added by dollar exchange could just as quickly vanish as it did appear? I think dollar exchange rate can simply be ruled out. labor tends to cost more in the US anyways.
You use this swing when you have it to your advantage. If it just to make one year worth of earning looks good, you would do it. And like I said PS3 are sold in the US too.
Can to back up your oopinion with some more thought? I don't think your explanation holds water.
I already did. Look its $300 device. Now your argument is harder to proof.
Why are you insisting sony is invulnerable and like eveyr othe rcompany on the plant makes decisions based on timing and cost?
They're not, I never said they are, I just said, that's what they base their decision on, cost. Now you on the other hand make this weird theory about Sony bitting a bullet for their decision, like how MS did once. MS has that luxury, Sony doesn't. And MS has proven that it doesn't work, taking a bullet.
Yes, they did. I see what you're doing. You seem to be only considering whateve rway would make sony look like this was a smart bsuiness move. I think it is a smart biz move, but I certainly don't think it's a cheaper one.
If it is a cheaper one, sure its a smart biz move for Sony. Not if its a more expensive one. For company that come out of no where to take over the consoles market, I assume they know what's cheaper for them more than you.
now that's pretty narrow minded V3. MS made thier decision to use nvidia over gigiapixel due to the time frame they wanted to launch in. They bit the bullet to pay more money so they could launch when they did. I think you'd have to be a fool not to consider sony could be making a similar decision.
Who's narrow minded ? I am just stating what MS had done.
And what Sony is doing. Sony doesn't have the luxury of MS, look at their restructuring and stuff, the company is under alot of pressure.
Under the pressure they are in, they won't bite a bullet.
V3 you're not making any sense imo, and you already contradicted yourself once. You said sony went with them because it was cheaper, but you also said they could tack on any pixel engine and it would have been good enough. If that were true, they could have stuck with the PS2 graphics tech because a) it's cheaper b) it would be good enough.
Sony chose to go with nvidia for a number of factors. I beleive thsoe to be timing, in that nvidia was more expensive but they could produce results far quicker.
I do not beleive it was Nvidia being cheaper than doing it internally there's a reason doing it internally won out with Sony before. When I say bite the bullet, I mean going outside the company to get what you want, and in almost all cases that will cost you more.
Wow vince, when i said that MS was doing things differently this time around with the licensing of ATI graphics tech instead of purchasing chips, you came out and said "how is this any different than last time" or something along those lines. i said it was cheaper to license, but not as cheap as developing things in house and you didn't have a problem with that.
I never said I didn't have a problem with it; I just didn't respond. It's a complex situation that's dependent on many factors, I just assume not get into it.
Of course time may be a factor, but i still don't see how it could potentially cost more. not when you don't have to pay out royaltees on every chip produced after the R&D is compelte.
It costs more because you can't just develop new techniques out of sheer serendipitous action. Ask yourself why the PC 3D field is consolidating and narrowing and why new start-ups are non-existent. For the same reason you don't see a huge field of start-up competitors taking on Intel or AMD. The barriers to entry are cost prohibitive, time is a cost and they are interchangable.
Ask Dave what the costs are to developing a new architecture for ATI or nVidia -- it's in the billions of dollars. And that's in a company which has the established IP base, the established engineering and brain-trust, that has the infastructure (computing, tools, etc).
V3 you're not making any sense imo, and you already contradicted yourself once. You said sony went with them because it was cheaper, but you also said they could tack on any pixel engine and it would have been good enough. If that were true, they could have stuck with the PS2 graphics tech because a) it's cheaper b) it would be good enough.
I said pixel engine. GS lacks any kind of pixel shader. Let alone pixel engine.
Sony chose to go with nvidia for a number of factors. I beleive thsoe to be timing, in that nvidia was more expensive but they could produce results far quicker.
NV is a cheaper solution.
I do not beleive it was Nvidia being cheaper than doing it internally there's a reason doing it internally won out with Sony before. When I say bite the bullet, I mean going outside the company to get what you want, and in almost all cases that will cost you more.
Not always the case, here people here even argue how much cheaper Dreamcast was compare to PS2.
But internally isn't always cheaper, you've go to accept that.
Sony will go anywhere that is cheaper to get what they need. External is cheaper so be it.
If that were true, they could have stuck with the PS2 graphics tech because a) it's cheaper b) it would be good enough.
Cheaper? Yes
Good enough? No
Sony went with nVIDIA for a number of reasons, the big ones being that nVIDIA already had a very capable next gen architecture built/almost completed, they already had all the tools built and the architecture fit in with Sony's visions, not to mention the solution obviously fills Sony's power requirements for what they wanted with the PS3 GPU.
So overall? It's cheaper, more developer friendly and will likely be more powerful than a in-house Solution that SCE+Toshiba could come up with.
I never said I didn't have a problem with it; I just didn't respond. It's a complex situation that's dependent on many factors, I just assume not get into it.
I seem to remember you having a big problem with it while saying it wouldn't be as cheap as what sony is doing internally. Now you say different, go figure. Yeah I can see why you wouldn't want to get into that. Moving along now...
It costs more because you can't just develop new techniques out of sheer serendipitous action.
ya know it's funny to see how you are you didn't think sony needed to licese tech before the Nvidia annoucement, but not that they did you embrace it like a hot girl.
Yes I do agree, it would take more TIME and money invested to reach the point where nvidia and ATI are at. which is why I said I think sony's decision was based more on timing and what they get for the money compared to what it costs.
Ask Dave what the costs are to developing a new architecture for ATI or nVidia -- it's in the billions of dollars. billions? I'm sure it costs in the hundreds of milllions. ...but billions seems a bit much don't you think?
I do not beleive it was Nvidia being cheaper than doing it internally there's a reason doing it internally won out with Sony before. When I say bite the bullet, I mean going outside the company to get what you want, and in almost all cases that will cost you more.
I seem to remember you having a big problem with it while saying it wouldn't be as cheap as what sony is doing internally. Now you say different, go figure. Yeah I can see why you wouldn't want to get into that. Moving along now...
This is just not true. As I said it's situation and context dependant. For example, it's immensely cheaper for ATI or nVidia to "go outside the company" and fab their parts at TSMC|UMC|IBM than it is for them to drop $4Billion on a bleeding-edge 300mm fab. Hell, why hasn't nVidia gone vertical?
Inversely, for a company like Sony that consumes over $8Billion in ICs per year, it's become immensely profitable for them to go vertical. This can be directly seen already in the PSP price, which someone from Sony has stated is only attainable since Sony internally fabs the majority of the ICs themselves. The benefits for them will only become more pronounced in the next half-decade as Sony, as a company, moves to a network-centric Cell world product line.
EDIT: Added second quote, since you're still not thinking about what's been said.
It costs more because you can't just develop new techniques out of sheer serendipitous action.
ya know it's funny to see how you are you didn't think sony needed to licese tech before the Nvidia annoucement, but not that they did you embrace it like a hot girl.
It's only funny because you don't understand what I'm saying, nor what V3 is for that matter. Pattern anybody? If you understood what I was saying, it was that they don't need nVidia or ATI for 'Shading' and the computationally intense functions that are unbounded in resource demands and is where the future lies. I've consistently maintained for over two years that the 'dumb' raster functions, which are hard-bounded, will never-the-less a problem for several reasons.
And secondly, ask Dave. You're wrong.
Also, you keep stating that it was Sony which was pushing for this deal. Yet, what Joe didn't post in his initial thread was that Jen-Hsun Huang stated:
<blockquote>"The fundamental difference, of course, between the two platforms is that Ken's vision for PS3 is much much broader and it's going to be the underlying platform for digital consumers, from media servers to digital televisions to the next generation PVRs to game consoles and his vision is far far greater than mine in terms of the types of things he wants to build with it. The CELL microprocessor is a four to five year undertaking and I know of no microprocessor -- modern microprocessor -- with a greater level of investment than the CELL microprocessor and architecturally it's innovating and it's applications are going to be quite amazing from those people. So I think that alone is worth the price of entry I would really be very surprised if all of us in the computer industry and consumer electronics industry aren't the first in line to buy a CELL processor, just so we can know what it can do. It's an amazing vision."</blockquote>
I guess it will all come out in the wash. however I don't think sony is really saving any money on graphics tech, based on all the "cell taking a revoloutionary approach to rendering graphics" nonesense people around here have been muttering for months.
Possibly they are doing this to save money on having to create thier own tech/ip in the long run. What I think that comes down to is them having inferior graphics technology at this stage of development and they realized they could better spend the money on licensing elsewhere. Whether it was cheaper or more expensive, is really something neither of us should be arguing about.
It's only funny because you don't understand what I'm saying, nor what V3 is for that matter. Pattern anybody?
The only pattern i see is the flips, then flops, and then flips again you are making my friend.
If you understood what I was saying, it was that they don't need nVidia or ATI for 'Shading' and the computationally intense functions that are unbounded in resource demands and is where the future lies. I've consistently maintained for over two years that the 'dumb' raster functions, which are bounded, are never-the-less a problem for several reasons.
Right vince. Sony is paying paying royaltees on every graphics chip produced, just so they have access to your "dumb raster functions"? I think you are VERY wrong there and don't give enough credit to SONY. I think sony alreayd had a number of dumb raster functions in the hardware they currently use in PS2. At the very least I think mroe time has been spent working on "shading tech" than anything else in graphics cards of the past 4 years.
IMO sony are using nvidia tech for pixel shading among other things. call it a hunch, but think you'll be pretty suprised just how many graphics functions are handed off to nvidias hardware (including computationally intense functions).
Also, you keep stating that it was Sony which was pushing for this deal. Yet, what Joe didn't post in his initial thread was that Jen-Hsun Huang stated
No I'm not stating that. I'm stating IMO reasons sony would choose to work with nvidia. Nvidia would have thier own reasons for being involed. I'm sure Nvidia was fine with this choice after being left out two competing consoles. It's not like they had anything to lose.
The only pattern i see is the flips, then flops, and then flips again you are making my friend.
:roll:
No I'm not stating that. I'm stating IMO reasons sony would choose to work with nvidia. Nvidia would have thier own reasons for being involed. I'm sure Nvidia was fine with this choice after being left out two competing consoles. It's not like they had anything to lose.
What I find interesting is that just based of the timing of announcements, people assume that it was nVidia who was left out. Yet, logically, nVidia was the front-runner for X2 and Sony's been dealing with nVidia since around the same time MS started looking for alternative IHVs.
Microsoft sent tenders out to the IHVs in mid-2002 because I was told "they are looking for alternatives" -- ask yourself why. Perhaps nVidia and Sony already had some level of communication (as we now know) and nVidia saw the potential revenues in Cell or because MS wouldn't play along, chose ATI as a second choice. As I posted way back when the talk first started, it would be foolish for nVidia to not get their IP in the Cell platform; in one swoop they achieve dominance and instantly overshadow ATIs attempts at moving into CE significantly. As Huang said, it's "worth the price of entry."
PS. And we'll see what's in the Media Processor and what's not...
Megadrive1988
29-Dec-2004, 00:33
Vince, nice summary of Nvidia's strategic moves.
This thread and the NV-Sony deal is full of Oxymorons! ;)
Even from a business prospective...Sony would have the leverage in negiotations on whatever deal is being struck. Sony has the dominate position and Nvidia wouldn't know what kind of deals Sony is making for their GPU development. So naturally Nvidia would have to bid in for business. IMO the outcome would be something like this...
Nvidia gave very good pricing on their tech. (to buy into the console market and be on a dominate platform). Sony saw diminused returns on there GPU prospect went with Nvidia (to get software tools and established tech.). In return for Nvidia's good deal, Sony gave them the ability to push their tech. to other devices (posibily to pick up new business from their low price Tech offer).
Dave Baumann
29-Dec-2004, 01:10
Yet, logically, nVidia was the front-runner for X2 and Sony's been dealing with nVidia since around the same time MS started looking for alternative IHVs.
Logically they were on the back foot because of arbitration issues and they were always making noises that the deal MS wanted (royalty based) was not profitable - they were, and still are for the most part, a revenue driven company as high revenues drive stock. ATI were already much more ammenable to such a model as they already had it in a few other areas, including GC; NVIDIA have only recently (since the Media-Q purchase actually) softend on the royalty.
Perhaps nVidia and Sony already had some level of communication (as we now know) and nVidia saw the potential revenues in Cell or because MS wouldn't play along, chose ATI as a second choice.
JHH's goal was to have all three, not one.
Oh, no doubt Dave. I'm just pointing out that while ATI may very well turn out to be a better choice for Microsoft's next console, there are likely (as you pointed out as well) many facets to the deal which people don't factor in when they state that nVidia was "left out" and that they had nothing "to lose".
Also, do you have a rough estimate of how many hardware engineers nVidia has now-a-days? I remember it was something like 350(?) a few years back, but the company has expanded alot since.
Brimstone
29-Dec-2004, 01:59
As I posted way back when the talk first started, it would be foolish for nVidia to not get their IP in the Cell platform; in one swoop they achieve dominance and instantly overshadow ATIs attempts at moving into CE significantly. As Huang said, it's "worth the price of entry."
ATI is fine. They've already licensed Tensilica's Extensa processor. Chris Rowen and his leadership of Tensilica will have a much greater impact than "CELL".
ATI is fine. They've already licensed Tensilica's Extensa processor. Chris Rowen and his leadership of Tensilica will have a much greater impact than "CELL".
Can anyone explain what is Tensilica's Extensa processor? :D
Thanks.
Brimstone
29-Dec-2004, 02:29
ATI is fine. They've already licensed Tensilica's Extensa processor. Chris Rowen and his leadership of Tensilica will have a much greater impact than "CELL".
Can anyone explain what is Tensilica's Extensa processor? :D
Thanks.
A recipie for instant chips (http://www.business2.com/b2/web/articles/0,17863,513408,00.html)
The devastating thing Tensilica has are software patents.
Tensilica Secures Key Patents (http://www.tensilica.com/html/pr_2002_11_11b.html)
Panajev2001a
29-Dec-2004, 04:25
NV probably offer them the price that is cheaper than what Sony could have done internally.
No... that is still not it... you are assuming that the internal solution was the one they set aside when they finally gave the contract to nVIDIA.
Panajev2001a
29-Dec-2004, 04:27
My hunch is NV in dire need to replace the loss of income from failing to get Xbox2, most likely outbid Toshiba, thus making PS3 or other cell device cheaper for Sony to produce.
that would have been my second quess ; )
My question is this: why cannot we maybe accept that nVIDIA's GPU is also the more featured and faster of the two ?
My question is this: why cannot we maybe accept that nVIDIA's GPU is also the more featured and faster of the two ?
I think this is tacit in their statements Pana. If you normalize the relative preformance to be equivalent, it would take a larger amount of capital to reach that static level with the internal Sony or Toshiba designs. Or, said in other words, the external design is cheaper.
MrSingh
29-Dec-2004, 05:54
My hunch is NV in dire need to replace the loss of income from failing to get Xbox2, most likely outbid Toshiba, thus making PS3 or other cell device cheaper for Sony to produce.
that would have been my second quess ; )
My question is this: why cannot we maybe accept that nVIDIA's GPU is also the more featured and faster of the two ?
no.. because italian wine and pasta is just so darned delish!
Microsoft sent tenders out to the IHVs in mid-2002 because I was told "they are looking for alternatives" -- ask yourself why. Perhaps nVidia and Sony already had some level of communication (as we now know) and nVidia saw the potential revenues in Cell or because MS wouldn't play along, chose ATI as a second choice.
Perhaps two years ago someone at MS had a 5800ultra and said "this is not the future". :lol:
I guess it will all come out in the wash. however I don't think sony is really saving any money on graphics tech, based on all the "cell taking a revoloutionary approach to rendering graphics" nonesense people around here have been muttering for months.
Qroach, if Sony ever going to bite a bullet, they would go with cheaper technology not an expensive one. In that way they may had more inferior pixel engine. That's Sony biting the bullets. Not MS style, like you thought.
We don't know anything about Toshiba solution (or NV), yet you thought of their solutions to be inferior because Sony went with NV instead. Sony will go with a cheaper solution. If it is better than its a plus isn't ?
Possibly they are doing this to save money on having to create thier own tech/ip in the long run. What I think that comes down to is them having inferior graphics technology at this stage of development and they realized they could better spend the money on licensing elsewhere. Whether it was cheaper or more expensive, is really something neither of us should be arguing about.
What you shouldn't be arguing is if the solutions is inferior or superior. Because that arguments doesn't hold. And your argument superior equates to something more expensive doesn't hold either.
Sony went which ever cheaper for them for their need.
Panajev2001a
29-Dec-2004, 06:36
My hunch is NV in dire need to replace the loss of income from failing to get Xbox2, most likely outbid Toshiba, thus making PS3 or other cell device cheaper for Sony to produce.
that would have been my second quess ; )
My question is this: why cannot we maybe accept that nVIDIA's GPU is also the more featured and faster of the two ?
no.. because italian wine and pasta is just so darned delish!
Mamma-miaaaaa..... Singh-ah... you know in the fAmily, you see the Pope-ah and eating past-ah while playing du mandulin-ah.
;).
No, that is not how I would talk to "Teflon" Ken ;).
No... that is still not it... you are assuming that the internal solution was the one they set aside when they finally gave the contract to nVIDIA.
Panajev2001a, I only assume that Sony went with cheaper solution, and they end up going with with NV. So NV is the cheaper solution.
In the past Sony always goes with cheaper solution too. Sony is cost concious. They'll do it internally or able to partner someone other than NV, if it was cheaper. NV just outbid everyone and won the contract. Sony would not go with a more expensive solutions just for the sake of having superior pixel engine compare to competitors, that's for sure. They would just get the cheapest pixel engine available, internally or externally.
Inane_Dork
29-Dec-2004, 07:35
Panajev2001a, I only assume that Sony went with cheaper solution, and they end up going with with NV. So NV is the cheaper solution.
In the past Sony always goes with cheaper solution too. Sony is cost concious. They'll do it internally or able to partner someone other than NV, if it was cheaper. NV just outbid everyone and won the contract. Sony would not go with a more expensive solutions just for the sake of having superior pixel engine compare to competitors, that's for sure. They would just get the cheapest pixel engine available, internally or externally.But that's only true to an extent. They could have used multiple GS's for less money, I'll bet. But that solution was unreasonable for feature set.
Panajev2001a
29-Dec-2004, 07:36
I guess it will all come out in the wash. however I don't think sony is really saving any money on graphics tech, based on all the "cell taking a revoloutionary approach to rendering graphics" nonesense people around here have been muttering for months.
Qroach, if Sony ever going to bite a bullet, they would go with cheaper technology not an expensive one. In that way they may had more inferior pixel engine. That's Sony biting the bullets. Not MS style, like you thought.
We don't know anything about Toshiba solution (or NV), yet you thought of their solutions to be inferior because Sony went with NV instead. Sony will go with a cheaper solution. If it is better than its a plus isn't ?
It is wrong on many levels this topic of "Sony wanted the cheapest solution".
It is really misleading and I am glad Vince caught the the argument by the head and understood the need of clarifying this.
If you normalize the relative preformance to be equivalent, it would take a larger amount of capital to reach that static level with the internal Sony or Toshiba designs. Or, said in other words, the external design is cheaper.
I think we need to read these lines again as they clear the confusion we are having here.
Time and money could bring a design from Toshiba or an internal one to be in parity or be superior to the design nVIDIA is proposing.
The problem is that time is not something they have an unlimited resource of and it is getting even shorter than money now: the console has a certain release date they have always planned to hit and miss it by 6 months to go with an internal or semi-internal solution is not an option as that would give MS too many months being the first next-generation console on the market.
If Toshiba would have had a fully PS 3.xx/4.xx by December 2006 then it is kinda worthless if the console must ship before that date.
Also, if money was no problem (meaning that they have an unlimited supply of money) they could simply put more Toshiba GPU cores there tille the nVIDIA solution is clearly outperformed and by far, but Sony/SCE does not have an infinite budget.
I think performance is one of the key that pushed the contract to nVIDIA, but I think that a lot of pressure is coming on the software side of things (PlayStation 3 SDK, GPU related tools, etc...) which I think nVIDIA was more prepared to provide than Toshiba GPU-wise (nVIDIA also has a very nice and estabilished Development Relations group and a lot of experience with PC developers [which I think SCE wants to bring in as well and not just leave all of them to the Xbox 2-side] and standard APIs like OpenGL) as well as the money side.
It is strange that Toshiba seems to have been caught off-guard about this announcement, a decision made recently (this does not mean that nVIDIA had not been following PlayStation 3's development and hadn't asked questions, etc...) must mean that nVIDIA made some very BIG and persuasive argument to SCE's management that changed their mind to the point of changing the contractor for the GPU.
I do not think that the GPU was supposed to be CELL even when they decided to go with the Toshiba GPU as well as now that they are going with the nVIDIA design.
This is not a loss for CELL, nobody said that CELL systems were not going to use any other chip: Toshiba is going to use CELL with thei MeP architecture.
I think that architecturally wise there are not many solutions to the perfomance and data flow problems for parallel-architectures in the context of multi-media driven solutions so it is not hard to see IMHO how forward looking architectures such as ATI R5xx, nVIDIA NV5X and S/T/I CELL might share some if not many ideas. I saw an nVIDIA patent not too long ago, it almost looked like a CELL system in one of the diagrams (that is before you started noticing differencies here and there).
With this said, I still say that IMHO the PlayStation 3 GPU is not CELL based, it does not have the SPUs/APUs.
Let's look at the stance ATI and nVIDIA are taking for the short-to-medium term (which includes the solutions that, time-frame wise, could ship in PlayStation 3, Xbox 2/Xenon and Revolution/NES 5).
ATI believes that developing a single unfiied shading unit is the key, that is what the GPU market needs to agree on as the best path to take for optimal performance and flexibility in 3D graphics.
This path leads to a situation in which Shading ALUs and Texture units are decoupled and we have a pool of Shading Units computing data, writing and reading from memory and interacting with a pool of Texture Units that process the requests from the Shading Units pool (separate scheduling and separate resources for each pool).
This is the path that CELL 2.0 might take, maybe in PlayStation 4.
This is the vision you have Vince: unified Shading Hardware (the SPU/APU) connected with a sea of fixed-function/hardwired units providing AA, Texture Sampling/Filtering, etc...
This is not what nVIDIA believs in.
nVIDIA believes that the time has not come yet for having the same Shading ALU hardware do both Vertex Shading work as well as Pixel Shading work, they believe that each unit should be heavvily optimized in a certain area (Vertex Shading or Pixel Shading) as they believe that the two areas, even though the ISA is converging, have still quite different needs that claim different hardware configurations.
I do not see the APUs in the Broadband Engine having instructions optimized to do texture sampling/filtering work or having separate TMU silicon. If the PlayStation 3 GPU ends up being a Pixel Shading only design as the other challenger design was then it means that Sony/SCE has a different view on Vertex Shading that includes versatility, but not features like Vertex Texturing (or at least not an use of Vertex Texturing so heavvy that it would require dedicatd hardware).
Vertex Shading might have been removed (as part of the customization work nVIDIA is doing with SCE on the GPU) and it would then be performed on the CELL based CPU. This path might be interesting as nVIDIA could re-utilize the silicon-space freed for e-DRAM and more Shading ALUs.
From what we hear, it is possible to guess that FP-wise the CELL based CPU in PlayStation 3 is aiming at an over-all higher-peak rating than what the Xbox 2/Xenon CPU is.
It would seem to me a not bad idea to assign all the Vertex Shading work to the CELL based CPU: we do not have tons of idle units if the application does not put much pressure on Vertex Shading (which would happen if we had a fixed number of Shaidng ALUs dedicated to Vertex Shading only) and we have lots of performance on the Pixel Shading side as we have space for more and very optimized Pixel Shading Units.
Panajev2001a
29-Dec-2004, 07:49
No... that is still not it... you are assuming that the internal solution was the one they set aside when they finally gave the contract to nVIDIA.
Panajev2001a, I only assume that Sony went with cheaper solution, and they end up going with with NV. So NV is the cheaper solution.
In the past Sony always goes with cheaper solution too. Sony is cost concious. They'll do it internally or able to partner someone other than NV, if it was cheaper. NV just outbid everyone and won the contract. Sony would not go with a more expensive solutions just for the sake of having superior pixel engine compare to competitors, that's for sure. They would just get the cheapest pixel engine available, internally or externally.
I disagree, they are cost conscious, but they also have a clear idea of what their performance and feature targets are which is what engineering is about.
It is wrong on many levels this topic of "Sony wanted the cheapest solution".
It is really misleading and I am glad Vince caught the the argument by the head and understood the need of clarifying this.
I know they have their performance target they need, (that's why they didn't reuse the old GS), but given options that meet their performance target, they will go with the cheapest option. That's just business.
If they have to bite the bullet, they will lower their performance target instead of increasing cost as suggested by Qroach. You don't increase your cost on something that will sell for $300 and less. Its really stupid, Sega and MS done it, it just doesn't work.
I disagree, they are cost conscious, but they also have a clear idea of what their performance and feature targets are which is what engineering is about.
Engineering is about meeting performance and features target at a given cost. That's engineering.
But that's only true to an extent. They could have used multiple GS's for less money, I'll bet. But that solution was unreasonable for feature set.
To re engineer GSs and produce Multiple GSs to meet what they wanted aren't exactly cheap for them. And they didn't went with it, they went with NV instead, its just a cheaper solution to Sony.
Panajev2001a
29-Dec-2004, 11:08
It is wrong on many levels this topic of "Sony wanted the cheapest solution".
It is really misleading and I am glad Vince caught the the argument by the head and understood the need of clarifying this.
I know they have their performance target they need, (that's why they didn't reuse the old GS), but given options that meet their performance target, they will go with the cheapest option. That's just business.
The other options did not meet their performance+features target IMHO (given the time constraints and the R&D available to fund the project... hey if they could afford 500 mm^2 chips for the GPU I do not doubt their internal solution would have been more than competitive enough ;)).
The other options did not meet their performance target (given the time constraints and the R&D available to fund the project... hey if they could afford 500 mm^2 chips for the GPU I do not doubt their internal solution would have been more than competitive enough
That's a speculation Pana :) Sony has their budget. They won't increase their budget, not on $300 product. They won't take an increase in budget for PS3. They will cut corners to meet that budget. Its just the real world, Pana.
Its nice though if we live in deamland of some sort, where 500 mm^2 chips grow on tree :D
Megadrive1988
29-Dec-2004, 11:33
I guess it will all come out in the wash. however I don't think sony is really saving any money on graphics tech, based on all the "cell taking a revoloutionary approach to rendering graphics" nonesense people around here have been muttering for months.
Qroach, if Sony ever going to bite a bullet, they would go with cheaper technology not an expensive one. In that way they may had more inferior pixel engine. That's Sony biting the bullets. Not MS style, like you thought.
We don't know anything about Toshiba solution (or NV), yet you thought of their solutions to be inferior because Sony went with NV instead. Sony will go with a cheaper solution. If it is better than its a plus isn't ?
It is wrong on many levels this topic of "Sony wanted the cheapest solution".
It is really misleading and I am glad Vince caught the the argument by the head and understood the need of clarifying this.
If you normalize the relative preformance to be equivalent, it would take a larger amount of capital to reach that static level with the internal Sony or Toshiba designs. Or, said in other words, the external design is cheaper.
I think we need to read these lines again as they clear the confusion we are having here.
Time and money could bring a design from Toshiba or an internal one to be in parity or be superior to the design nVIDIA is proposing.
The problem is that time is not something they have an unlimited resource of and it is getting even shorter than money now: the console has a certain release date they have always planned to hit and miss it by 6 months to go with an internal or semi-internal solution is not an option as that would give MS too many months being the first next-generation console on the market.
If Toshiba would have had a fully PS 3.xx/4.xx by December 2006 then it is kinda worthless if the console must ship before that date.
Also, if money was no problem (meaning that they have an unlimited supply of money) they could simply put more Toshiba GPU cores there tille the nVIDIA solution is clearly outperformed and by far, but Sony/SCE does not have an infinite budget.
I think performance is one of the key that pushed the contract to nVIDIA, but I think that a lot of pressure is coming on the software side of things (PlayStation 3 SDK, GPU related tools, etc...) which I think nVIDIA was more prepared to provide than Toshiba GPU-wise (nVIDIA also has a very nice and estabilished Development Relations group and a lot of experience with PC developers [which I think SCE wants to bring in as well and not just leave all of them to the Xbox 2-side] and standard APIs like OpenGL) as well as the money side.
It is strange that Toshiba seems to have been caught off-guard about this announcement, a decision made recently (this does not mean that nVIDIA had not been following PlayStation 3's development and hadn't asked questions, etc...) must mean that nVIDIA made some very BIG and persuasive argument to SCE's management that changed their mind to the point of changing the contractor for the GPU.
I do not think that the GPU was supposed to be CELL even when they decided to go with the Toshiba GPU as well as now that they are going with the nVIDIA design.
This is not a loss for CELL, nobody said that CELL systems were not going to use any other chip: Toshiba is going to use CELL with thei MeP architecture.
I think that architecturally wise there are not many solutions to the perfomance and data flow problems for parallel-architectures in the context of multi-media driven solutions so it is not hard to see IMHO how forward looking architectures such as ATI R5xx, nVIDIA NV5X and S/T/I CELL might share some if not many ideas. I saw an nVIDIA patent not too long ago, it almost looked like a CELL system in one of the diagrams (that is before you started noticing differencies here and there).
With this said, I still say that IMHO the PlayStation 3 GPU is not CELL based, it does not have the SPUs/APUs.
Let's look at the stance ATI and nVIDIA are taking for the short-to-medium term (which includes the solutions that, time-frame wise, could ship in PlayStation 3, Xbox 2/Xenon and Revolution/NES 5).
ATI believes that developing a single unfiied shading unit is the key, that is what the GPU market needs to agree on as the best path to take for optimal performance and flexibility in 3D graphics.
This path leads to a situation in which Shading ALUs and Texture units are decoupled and we have a pool of Shading Units computing data, writing and reading from memory and interacting with a pool of Texture Units that process the requests from the Shading Units pool (separate scheduling and separate resources for each pool).
This is the path that CELL 2.0 might take, maybe in PlayStation 4.
This is the vision you have Vince: unified Shading Hardware (the SPU/APU) connected with a sea of fixed-function/hardwired units providing AA, Texture Sampling/Filtering, etc...
This is not what nVIDIA believs in.
nVIDIA believes that the time has not come yet for having the same Shading ALU hardware do both Vertex Shading work as well as Pixel Shading work, they believe that each unit should be heavvily optimized in a certain area (Vertex Shading or Pixel Shading) as they believe that the two areas, even though the ISA is converging, have still quite different needs that claim different hardware configurations.
I do not see the APUs in the Broadband Engine having instructions optimized to do texture sampling/filtering work or having separate TMU silicon. If the PlayStation 3 GPU ends up being a Pixel Shading only design as the other challenger design was then it means that Sony/SCE has a different view on Vertex Shading that includes versatility, but not features like Vertex Texturing (or at least not an use of Vertex Texturing so heavvy that it would require dedicatd hardware).
Vertex Shading might have been removed (as part of the customization work nVIDIA is doing with SCE on the GPU) and it would then be performed on the CELL based CPU. This path might be interesting as nVIDIA could re-utilize the silicon-space freed for e-DRAM and more Shading ALUs.
From what we hear, it is possible to guess that FP-wise the CELL based CPU in PlayStation 3 is aiming at an over-all higher-peak rating than what the Xbox 2/Xenon CPU is.
It would seem to me a not bad idea to assign all the Vertex Shading work to the CELL based CPU: we do not have tons of idle units if the application does not put much pressure on Vertex Shading (which would happen if we had a fixed number of Shaidng ALUs dedicated to Vertex Shading only) and we have lots of performance on the Pixel Shading side as we have space for more and very optimized Pixel Shading Units.
Panajev, I enjoyed and savored your post like I would a nicely cooked steak. seriously, when i saw this was gonna be a substantial post, i ran downstairs to get a glass of coke to enjoy while I read 8) now don't get me wrong i amm not buttering you up or anything. i just enjoy your posts. :)
okay, can you re-explain your current thinking on how PS3 is going to render scenes and what PS3 is gonna be all about--are you no longer looking at a REYES style approach? Is PS3 going to be more conventional now, with texture mapping & pixel shading? was your eariler vision for PS3 now moved forward into PS4 or still pretty much the same?
take a look at this pre-rendered CG intro for Everquest 2.
http://www.gametrailers.com/player.php?id=21&type=mov (QuickTime)
http://www.gametrailers.com/player.php?id=21&type=wmv (Windows Media)
the first one-half or one-third of that CG sequence, especially the open landscape, do you feel that will be possible on the CELL-Nvidia based PS3 ?
london-boy
29-Dec-2004, 11:41
now don't get me wrong i amm not buttering you up or anything. i just enjoy your posts. :)
Hey hey hey! Let's leave the gay innuendos to the expert shall we...?
Anyway, to stay on topic, does anyone other than me think that now that NVIDIA is on board, the chances of PS3 having a "different" approach to rendering are getting slimmer?
Fafalada
29-Dec-2004, 11:51
If the PlayStation 3 GPU ends up being a Pixel Shading only design as the other challenger design was then it means that Sony/SCE has a different view on Vertex Shading that includes versatility, but not features like Vertex Texturing
Why? Obviously we don't know yet how easy we can deal with latencies in SPUs, but that problem would always remain, even if you bolted texture samplers directly onto each SPU.
....
With this said, I still say that IMHO the PlayStation 3 GPU is not CELL based, it does not have the SPUs/APUs.
....
It would seem to me a not bad idea to assign all the Vertex Shading work to the CELL based CPU:
....
To me the above two statements seem to contradict each other.. :?
Just to re-cap we exchanged a few pages of posts here in this thread,
http://www.beyond3d.com/forum/viewtopic.php?p=427444#427444
and continued a few exchages here,
http://www.beyond3d.com/forum/viewtopic.php?p=432187#432187
....
What does it mean having a "Consistent ISA without having the SPU/APU construct" ? Are you talking about on-the-fly JIT translation ? Emulation ?
...
I'm just looking at them as a bunch of SIMD units...the patents say you can have more or less SIMD units depending on performace...so what I'm thinking is that they have a Cell template, like the Toshiba MEP template if you remember?
...
Do you think nVIDIA is extending the ISA their Shading ALUs use to support the same ISA as the SPU/APU ?
...
Depends if you think there will be any shading going on in the CELL CPU? i.e vertex shading?
And of course they'll be eDRAM, Sony-Tosh love eDRAM! ...unless NV implement this turbo-cache thingy!
And at the end you asked me two questions which I answered with two more questions above! ;)
1. Basically, the way I see it is this...if you're gonna have vertex shading on the CELL CPU, then they're gonna be VS CELL threads, (aka software Cells), no?
If Hofstee was talking about two-way comms between CPU<=>GPU, then these VS CELL threads should run on the GPU also, no?
This should imply the GPU is CELL based.
2. OTOH, If you assume that these VS threads running on the CELL CPU aren't 'software Cells', i.e. VS non-CELL threads, then they're running in an evironment on the CPU with OTHER CELL threads. Why would you create a CELL architecture and run CELL and non-CELL threads on the CELL CPU?
Wouldn't this seriously fuck up the scheduling of these CELL threads (software Cells) from a load balancing POV in this environment?
3. However, if the VS is NOT done on the CELL CPU but on the GPU, then the entire GPU may OR may not be CELL based then, no?
This is how I see things and may be jadded due to the holiday season! :D
version
29-Dec-2004, 14:04
1. if on GPU dont run any vertex process, then 6 GigaPoly/s is the maximum without texture upload (1 poly about 16 byte,redwood 100GB/s)
about 3 Gigapoly/s with texture
32 pixelpipeline is enough on 800MHZ to exploit this
2. if on GPU use nurbs with displacementmapping then to require vertexprocess,
vertexshader or apulets or directhardware?
vertexshader not likely , this is nightmare for programming(cpu,APU, VS,PS, 4 different langugage and architect)
london-boy
29-Dec-2004, 14:17
Huh?
Here's my prediction on what has happened thus far with PS3 development
1. Many people working on PS3 hardware didn't know or expect this to happen.
2. The Nvidia's invovlement in PS3 did not start for one or two years before this annoucement. Nvidia's involvement was minimal, nothing more than NV execs talking on a high level to Sony execs, and sending out graphics cards for Sony to evaluate performance. Sony had been evaulating Nvidia tech for a while with zero cash exchanged between both parties. Nvidia didn't change what anything in thier roadmap to accomodate sony up until this annoucement.
3. Toshiba was left out in the cold with this change and didn't know it was going to happen. The work on the GS for PS3 will continue, but that graphics hardware won't be used in the PS3.
4. One of sony's offices led the charge to make this change, as internally issues were raised with the expected performance of the new GS and/or expected feature set/delivery time frame.
5. The nvidia licensing deal will cost Sony more money, but allow them to compete better on the graphics technology side and a later launch. The trade off was cost vrs performance & delivery time. Nvidia will incorporate support for the graphics API that Sony is jointly developing with other partners.
6. Nvidia will supply the entire GPU with minor modifications. It will be a custom version of a future Nvidia graphics chip (along the lines of Xgpu) however sony will be in charge of fabbing (as we already know). The Nvidia processor "can" handle all vertex and pixel shading, or the cell CPU can be used for vertex shading. They will not be incredibly integrated together.
7. Cell is a really good CPU, but NOT the second coming of christ.
8. There are few to zero actual Cell workstations available for developers to get cranking on. Sony is currently using emulators and PC's to perform their development, and the final devkits won't arrive until Nvidia has delivered their custom GPU late next year.
9. You will not see any realtime raytracying, radiosity lighiting, or any advanced rendering techniques beyond what is readily capable with the graphics hardware found on PC's of the same time frame. you might see real support for subdivision surfaces in hardware, but they will only be used on chracters.
Feel free to quote me on this or post this when we see real PS3 specs or actual performance out of developers. grab and save this post if you want I don't F'ing care. I think there's some people around here that are living in an imaginary dreamworld with regards to sony and PS3 and their plans for it. It's a game console, no company is invincible (MS could fuck up easily, just as sony or nintendo) and sometimes tradeoffs are made.
london-boy
29-Dec-2004, 14:36
I don't agree with many of the things you just said, or at least i think you (and everyone else) are in no position to make judgements on what we know very little about.
You don't have to agree with it, as I said I don't F'ing care. Btw, you don't know what information I may or may not know about what has happened thus far, so do NOT tell me I'm not in the postion to speculate or post my prediction. I can do what I please or are you my master :shock:
That's a common defense in forums when someone readons something they don't like, they say "nobody here is in a postion to speculate", but funny how they only say that right after they disagree with someone. Like I said, you can quote me on it, or throw it back in my face if I'm wrong for all I care.
london-boy
29-Dec-2004, 14:53
You don't have to agree with it, as I said I don't F'ing care. Btw, you don't know what information I may or may not know about what has happened thus far, so do NOT tell me I'm not in the postion to speculate or post my prediction. I can do what I please or are you my master :shock:
That's a common defense in forums when someone readons something they don't like, they say "nobody here is in a postion to speculate", but funny how they only say that right after they disagree with someone. Like I said, you can quote me on it, or throw it back in my face if I'm wrong for all I care.
Hey gee don't go all aggressive on me, remember i'm the Master, BOY.
All i'm saying is that your post screams "Sony don't know what they're doing, this is a rush job, Toshiba got stuffed" from every pixel. And you are not in the position to say that.
Speculation is fine and dandy, but it's not a medium to show us your personal preferences of a company over the other, like you usually do anyway so why am i even questioning your ways? :?
1. if on GPU dont run any vertex process, then 6 GigaPoly/s is the maximum without texture upload (1 poly about 16 byte,redwood 100GB/s)
about 3 Gigapoly/s with texture
32 pixelpipeline is enough on 800MHZ to exploit this
Something tells me we aren't going to see games pushing 6 billion polygons a second. I expect maybe at the maximum a magnitude under that somewhere.
2. if on GPU use nurbs with displacementmapping then to require vertexprocess,
vertexshader or apulets or directhardware?
vertexshader not likely , this is nightmare for programming(cpu,APU, VS,PS, 4 different langugage and architect)
I would assume the VS language would be similar to the PS language as has been in every shader language I've worked with. And really VS are pretty damn easy to write in comparison to trying to manage cell threads for doing vertex shading. The only time VS can be somewhat of a pain is when you are doing something completely different and kinda strange.
Yessir mastah
You can read whatever you want into my post, I can make a similar post about things MS has done lately that would make YOU think they don't know what they are doing. So what? This has nothing to do with personal preference as none of the next gen consoles have been released yet. I formulate my personal preference when I know somethign concreate about the systems. I do however have the opinion that sony isn't invincible and I think that clashes with some people vision of sony as a company doing everything right.
This is my opinion on what happened and as always there are tons of factors invovled.
london-boy
29-Dec-2004, 15:42
Yessir mastah
You can read whatever you want into my post, I can make a similar post about things MS has done lately that would make YOU think they don't know what they are doing. So what? This has nothing to do with personal preference as none of the next gen consoles have been released yet. I formulate my personal preference when I know somethign concreate about the systems. I do however have the opinion that sony isn't invincible and I think that clashes with some people vision of sony as a company doing everything right.
This is my opinion on what happened and as always there are tons of factors invovled.
Fine, and i said i don't agree on some things you said. That's it. You're the one who went all defensive.
London boy,
You're the one who went all defensive.
First, you attacked my right to express my opinion by telling me that i'm not in a position to make judgments, even when I clearly was not making any sort of judgments.
Second, you try to belittle my opinion by saying I'm showing my "personal preferences" when I feel this has nothing to do with any preferences at all.
I haven't been defensive until now. I mean, what would you expect after the two points above?
london-boy
29-Dec-2004, 16:03
London boy,
You're the one who went all defensive.
First, you attacked my right to express my opinion by telling me that i'm not in a position to make judgments, even when I clearly was not making any sort of judgments.
Second, you try to belittle my opinion by saying I'm showing my "personal preferences" when I feel this has nothing to do with any preferences at all.
I haven't been defensive until now. I mean, what would you expect after the two points above?
I was bored? And i saw your name and jumped to conclusions? Not a long jump to make anyway.
Not a long jump to make anyway.
Longer for some than others... yes.
pahcman
29-Dec-2004, 16:08
This is not what nVIDIA believs in.
nVIDIA believes that the time has not come yet for having the same Shading ALU hardware do both Vertex Shading work as well as Pixel Shading work, they believe that each unit should be heavvily optimized in a certain area (Vertex Shading or Pixel Shading) as they believe that the two areas, even though the ISA is converging, have still quite different needs that claim different hardware configurations.
Well said post, this portion i must ask something.
I was so caught up in Vince (was it Vince?) forceful "whats a cpu whats a gpu" arguement, that i was surprise Nvidia got the PS3 GPU. Because i do remember Nvidia is against a unified system soon. So GPU deal may mean Cell *is* so powerful that Nvidia see unified system finally doable on Cell than on peecee. Or i guess Vince was being overly forceful and optimisic that Cell is *so* powerful, "gpu" is no more, since Nvidia Cell-PU and NV50 will be conflicting r&d.
Does this mean tflops PS3 is also out? Remember too Vince expect PS3 to be tflops ahead of gflops rivals, that difference will be discernable and such raw power can do anything internally as Nvidia/Ati designs can.
Please know, my memory is on Vince because he strikes me as a very knowledgeable Cell hobbyist researcher, i could be wrong to quote Vince, sorry Vince if so. :)
london-boy
29-Dec-2004, 16:10
I have long legs.
Anyway, back to topic, kinda. I'm not sure i asked this already, but am i the only one who feels that now that NVIDIA is on board, this talk about "different rendering systems like REYES or whatever" is walking away steadily to the next next generation? Meaning, aint gonna happen this time around. I was skeptical before, and NVIDIA doesn't sound too different-rendering-y to me. It just sounds like more of the same, which is not a bad thing in itself, since we'll get much more of what we can see today on top PCs but hey...
...
Anyway, back to topic, kinda. I'm not sure i asked this already, but am i the only one who feels that now that NVIDIA is on board, this talk about "different rendering systems like REYES or whatever" is walking away steadily to the next next generation? Meaning, aint gonna happen this time around. I was skeptical before, and NVIDIA doesn't sound too different-rendering-y to me. It just sounds like more of the same, which is not a bad thing in itself, since we'll get much more of what we can see today on top PCs but hey...
FWIW, I still think it's a possibility! :P
In this recent thread there are plenty of hints if you read it all,
http://www.beyond3d.com/forum/viewtopic.php?t=18849&start=0
and these series of posts,
http://www.beyond3d.com/forum/viewtopic.php?p=430107#430107
FWIW, I still think it's a possibility! :P
In this recent thread there are plenty of hints if you read it all,
http://www.beyond3d.com/forum/viewtopic.php?t=18849&start=0
Niet..that stuff is not about realtime rendering!
I think any fancy rendering ideas should be thrown out and we'll be bound by what ATI and nvidia are capable of doing for a long while. I don't expect either company to achive a renderman like rendeirng system anytime soon. Although marketing may make people think they are capable of that.
london-boy
29-Dec-2004, 16:38
The thing is that the whole concept of Pixel Shaders (which will not be called Pixel Shaders on PS3 i guess as that is a DX term, but i could be wrong) was invented to fake effects offline renders render through raytracing. So, unless the same hardware can be used for a small-scale form of raytracing, we're still gonna see polygons and shaders for a long time.
Pixel shading was designed to provide a procedural method of representing surface texture among other things. It wasn't really designed to fake the effects of a raytracer specifically. You can do a whole lot with a scanline renderer.
FWIW, I still think it's a possibility! :P
In this recent thread there are plenty of hints if you read it all,
http://www.beyond3d.com/forum/viewtopic.php?t=18849&start=0
Niet..that stuff is not about realtime rendering!
I'm not talking about the patent...just remove the RT/GI stuff from the patents and your left with REYES. If you follow all the links in that thread, NV's goal is to make Offline rendering a Real-time prospect, which has also been Sony's goal with all the GScube R&D...
And scanline renderring isn't going to replaced by full RT, which alot people got the impression from that thread...their hearts still lie with Shaders! :)
I don't think we'll see anythign like reyes next gen.
I don't think we'll see anythign like reyes next gen.
If you can tell me that a GScube-64 (approx. 600-800 GFlops) was not capabable of real-time REYES because of lack of power, then I'll be more inclined to rule it out next-gen! ;)
If you can tell me that a GScube-64 (approx. 600-800 GFlops) was not capabable of real-time REYES because of lack of power, then I'll be more inclined to rule it out next-gen! ;)
it's not a matter of power.
You need a special hw support to make it (reyes) efficient and fast.
PURE sw reyes rendering is a suicide.
london-boy
29-Dec-2004, 17:11
I don't think we'll see anythign like reyes next gen.
If you can tell me that a GScube-64 (approx. 600-800 GFlops) was not capabable of real-time REYES because of lack of power, then I'll be more inclined to rule it out next-gen! ;)
GSCube had 64 GSs.
PS3 won't have a GS-like architecture.
I don't think we'll see anythign like reyes next gen.
If you can tell me that a GScube-64 (approx. 600-800 GFlops) was not capabable of real-time REYES because of lack of power, then I'll be more inclined to rule it out next-gen! ;)
GSCube had 63 GSs.
PS3 won't have a GS-like architecture.
GScube was a highly parallel architecture...What is PS3?
If you can tell me that a GScube-64 (approx. 600-800 GFlops) was not capabable of real-time REYES because of lack of power, then I'll be more inclined to rule it out next-gen! ;)
it's not a matter of power.
You need a special hw support to make it (reyes) efficient and fast.
PURE sw reyes rendering is a suicide.
If I was CTO and NV's goal was to make offline-rendering a real-time prospect and I was designing a new architecture, where would my GPU designs go without the bounds of Wintel? Why would I join allegiance with Sony who have a similar goal? And if you'd want to break away to a new paradigm, you don't do it in one big jump but allow the 'old' to exist with the 'new'...which seemed to be hinted by Hofstee when he said the architecture would allow you to 'try' new things! :)
Panajev2001a
29-Dec-2004, 18:25
If the PlayStation 3 GPU ends up being a Pixel Shading only design as the other challenger design was then it means that Sony/SCE has a different view on Vertex Shading that includes versatility, but not features like Vertex Texturing
Why? Obviously we don't know yet how easy we can deal with latencies in SPUs, but that problem would always remain, even if you bolted texture samplers directly onto each SPU.
First the latencies with dedicated TMUs would be much much lower IMHO: an SPU/APU bi-linear filtering a texture would waste quite a bit of the unit's performance.
Second, I do not see how this changes my point: if they saw Vertex Texturing and VS 3.xx/4.xx as NEEDED then they would push the most optimal way to support it (performance and cost wise) and that would be to have a non PS only GPU (which might still be what nVIDIA is delivering, we do not know if it will be PS only).
The other alternative, if they really saw Vertex Texturing as important would be to bolt TMUs close by each SPU (I think the SPUs will be able to deal with the latencies that texture fetching+filtering would imply... the stall would be quite lower than if te SPU would be filtering the texture in software and the power consumption would be lower as we would have the SPU idle a bit or switiching to another thread while the TMU is working).
Panajev2001a
29-Dec-2004, 18:31
....
With this said, I still say that IMHO the PlayStation 3 GPU is not CELL based, it does not have the SPUs/APUs.
....
It would seem to me a not bad idea to assign all the Vertex Shading work to the CELL based CPU:
....
To me the above two statements seem to contradict each other.. :?
Explain how.
Vertex Shading done by units like the SPUs/APUs or like the VUs in the EE can be done very effectively, while Pixel Shading might not (nVIDIA does not believe in using the same hardware for both and for now they might be right).
Splitting the VS and PS work between the CPU and the GPU is what some PS heavvy games on Xbox 2/Xenon will do: all the unified shading units would be dedicated to Pixel Shading work and Vertex Shading would then be done on the enhanced VMX units of the Xbox 2/Xenon's CPU.
Panajev2001a
29-Dec-2004, 18:43
1. Basically, the way I see it is this...if you're gonna have vertex shading on the CELL CPU, then they're gonna be VS CELL threads, (aka software Cells), no?
If Hofstee was talking about two-way comms between CPU<=>GPU, then these VS CELL threads should run on the GPU also, no?
Ideally that is what they would want, which is why I talked about the direction for CELL 2.0 for example. IBM, with CELL, sees a point in pushing towards the same direction ATI is pushing with unified shading hardware.
It makes sense as having two very integrated (very fast connection between CPU and GPU) ICs in which workload can be easily balanced between the two ICs leveraging CELL technology.
This is not the way nVIDIA sees things, not for theshort-to-medium term at least.
Two-way communication between CPU and GPU is not CELL dependent (although I can see how having a CELL based CPU and a CELL based GPU could work very well here): you will see it in Xbox 2/Xenon (with the GPU being able to read/write a portion of the CPU L2 cache) and future PCI-Express systems.
Megadrive1988
29-Dec-2004, 22:22
Here's my prediction on what has happened thus far with PS3 development
1. Many people working on PS3 hardware didn't know or expect this to happen.
2. The Nvidia's invovlement in PS3 did not start for one or two years before this annoucement. Nvidia's involvement was minimal, nothing more than NV execs talking on a high level to Sony execs, and sending out graphics cards for Sony to evaluate performance. Sony had been evaulating Nvidia tech for a while with zero cash exchanged between both parties. Nvidia didn't change what anything in thier roadmap to accomodate sony up until this annoucement.
3. Toshiba was left out in the cold with this change and didn't know it was going to happen. The work on the GS for PS3 will continue, but that graphics hardware won't be used in the PS3.
4. One of sony's offices led the charge to make this change, as internally issues were raised with the expected performance of the new GS and/or expected feature set/delivery time frame.
5. The nvidia licensing deal will cost Sony more money, but allow them to compete better on the graphics technology side and a later launch. The trade off was cost vrs performance & delivery time. Nvidia will incorporate support for the graphics API that Sony is jointly developing with other partners.
6. Nvidia will supply the entire GPU with minor modifications. It will be a custom version of a future Nvidia graphics chip (along the lines of Xgpu) however sony will be in charge of fabbing (as we already know). The Nvidia processor "can" handle all vertex and pixel shading, or the cell CPU can be used for vertex shading. They will not be incredibly integrated together.
7. Cell is a really good CPU, but NOT the second coming of christ.
8. There are few to zero actual Cell workstations available for developers to get cranking on. Sony is currently using emulators and PC's to perform their development, and the final devkits won't arrive until Nvidia has delivered their custom GPU late next year.
9. You will not see any realtime raytracying, radiosity lighiting, or any advanced rendering techniques beyond what is readily capable with the graphics hardware found on PC's of the same time frame. you might see real support for subdivision surfaces in hardware, but they will only be used on chracters.
Feel free to quote me on this or post this when we see real PS3 specs or actual performance out of developers. grab and save this post if you want I don't F'ing care. I think there's some people around here that are living in an imaginary dreamworld with regards to sony and PS3 and their plans for it. It's a game console, no company is invincible (MS could fuck up easily, just as sony or nintendo) and sometimes tradeoffs are made.
good post Qroach. I think I agree with all of it. I think most or many of your predictions & assumptions are most likely on target.
Johnny Awesome
29-Dec-2004, 22:44
I agree with QRoach in this thread. I think Sony did their best internally, but came up a little short on time or performance and went with NVidia to get something that will do PS3 justice. Certainly not the end of the world for Sony, but a small setback in terms of PS3 cost.
Dave Baumann
29-Dec-2004, 22:56
Splitting the VS and PS work between the CPU and the GPU is what some PS heavvy games on Xbox 2/Xenon will do: all the unified shading units would be dedicated to Pixel Shading work and Vertex Shading would then be done on the enhanced VMX units of the Xbox 2/Xenon's CPU.
Just to interject here (on a slightly unrelated note). Its my belief that MS expects the majority of VS work to be done via the CPU (in fact I believe their initial tendor guidelines suggested that they wanted no VS capabilities) having unified shaders just offers a convenient method of giving all the functionality and more flaxability as to where things are processed.
rabidrabbit
29-Dec-2004, 22:57
I agree with QRoach in this thread. I think Sony did their best internally, but came up a little short on time or performance and went with NVidia to get something that will do PS3 justice. Certainly not the end of the world for Sony, but a small setback in terms of PS3 cost.
How so?
I don't think Sony has ever made a console all by themselves, they always have had some others to help design them and licence technology: Toshiba, MIPS, Rambus, IBM.
I don't see how that deal with nVidia would be any different cost wise to SOny than the deals with other partners.
Had the graphics part been designed by Toshiba, would that have been more cost effective?
Vertex Shading done by units like the SPUs/APUs or like the VUs in the EE can be done very effectively, while Pixel Shading might not (nVIDIA does not believe in using the same hardware for both and for now they might be right).
But programmable vertex shading is only one component in the middle of a lot of fixed function geometry processing. This stuff while not being 'sexy' can be extremely expensive without custum hardware.
A simple example vertex indexation. A Modern GPU vertex processor reads a index list, and then that index list is used be read the actual vertex data from vertex pools.
In psuedo code a simplified triangle assembly is
Let I be a register with 3 16 bit tuples identified as 0,1,2
Let A be the address of the index stream
Let B be the address of the vertex stream
Let n be the current triangle number
Load I from n[A]
Load R0 from I.0[B]
Load R1 from I.1[B]
Load R2 from I.2[B]
VertexShade( R0 )
VertexShade( R1 )
VertexShade( R2 )
If we assume a memory latency of 200 cycles (thats very generous...) and that we can process 3+ memory requests at once we have 400 cycles to lose. Load I from n[A] is relatively easy, a predictor will notice the linear read and prefetch 200 cycles early, so when the load happens its hopefully already there. But the Load Rx from I.x[B] is much harder... I can't predict until the index register is filled, so I have to stall the shader for 200 cycles. The only serious option is a thread context switch.
Modern GPU's have this entire thing hidden, it just works. Unless a CPU is going to be fitted with a GPU style memory fetch system hows it going to hide it?
Of course you can rework you pipeline not to require things like this but its comes up several times (bone indices are the next obvious one).
Splitting the VS and PS work between the CPU and the GPU is what some PS heavvy games on Xbox 2/Xenon will do: all the unified shading units would be dedicated to Pixel Shading work and Vertex Shading would then be done on the enhanced VMX units of the Xbox 2/Xenon's CPU.
Just to interject here (on a slightly unrelated note). Its my belief that MS expects the majority of VS work to be done via the CPU (in fact I believe their initial tendor guidelines suggested that they wanted no VS capabilities) having unified shaders just offers a convenient method of giving all the functionality and more flaxability as to where things are processed.
I disagree. There is alot of magic behind the scene keeping the vertex throughput up, coming up with replacements on the CPU will be hard and I'd imagine why the idea was shot down by ATI.
Having had to write CPU geometry pipes, there is no doubt that FLOPs is largely irrelevant to geometry throughput. Almost every time you are memory bound, you either come up with wacky bandwidth saving methods or just do more vertex work... because you will have to have fairly complex vertex shaders to hide the cost of memory latency.
If we assume a memory latency of 200 cycles (thats very generous...) and that we can process 3+ memory requests at once we have 400 cycles to lose....[cut]
That's why we pre-encode data packets to fit in local memory on the PS2 and that's why we'll probably do the same even on the PS3.
Indexed primitives on the PS2 are not a problem at all when we need them (and most of time we don't need them)
No predictor would be effective in a situation like that, except in some very rare case (and probably you don't need indexed primitives in those cases)
I know you were just showing an example of the difficulties that can arise on non-customized hw, but it's also true one can think different ways to achieve the same goal, and sometimes exotic solutions works even better ;)
(sorry DeanoC, I know you know, it's just a ps2 pride post.. :) )
ciao,
Marco
Panajev2001a
30-Dec-2004, 00:50
Vertex Shading done by units like the SPUs/APUs or like the VUs in the EE can be done very effectively, while Pixel Shading might not (nVIDIA does not believe in using the same hardware for both and for now they might be right).
But programmable vertex shading is only one component in the middle of a lot of fixed function geometry processing. This stuff while not being 'sexy' can be extremely expensive without custum hardware.
A simple example vertex indexation. A Modern GPU vertex processor reads a index list, and then that index list is used be read the actual vertex data from vertex pools.
In psuedo code a simplified triangle assembly is
Let I be a register with 3 16 bit tuples identified as 0,1,2
Let A be the address of the index stream
Let B be the address of the vertex stream
Let n be the current triangle number
Load I from n[A]
Load R0 from I.0[B]
Load R1 from I.1[B]
Load R2 from I.2[B]
VertexShade( R0 )
VertexShade( R1 )
VertexShade( R2 )
If we assume a memory latency of 200 cycles (thats very generous...) and that we can process 3+ memory requests at once we have 400 cycles to lose. Load I from n[A] is relatively easy, a predictor will notice the linear read and prefetch 200 cycles early, so when the load happens its hopefully already there. But the Load Rx from I.x[B] is much harder... I can't predict until the index register is filled, so I have to stall the shader for 200 cycles. The only serious option is a thread context switch.
Modern GPU's have this entire thing hidden, it just works. Unless a CPU is going to be fitted with a GPU style memory fetch system hows it going to hide it?
Of course you can rework you pipeline not to require things like this but its comes up several times (bone indices are the next obvious one).
Well, what do you do on PlayStation 2 ? I do not expect SPUs/APUs to be any worse at Vertex processing than EE's VUs or any less flexible.
Also, we should keep in mind that the other solution was supposed to be PS only so it was understood that the SPUs/APUs would be working on Geometry Processing/Vertex Shading: I think Sony/SCE made sure IBM knew from the start what kind of code they would see runnign most of the time on the SPUs/APUs and that means also Geometry Processing/Vertex Shading.
A GPU doing PS only work was basically the idea since almost the start I believe (at least since the "other option" was taken): I do think Sony/SCE had the idea of maybe pushing more programmable shaders on the GPU, but to keep Vertex Processing and Pixel Processing resources separate.
I would think that they thought about this kind of problems.
What you are describing is basically SoE MTwhich the GPU do automatically: if what they look for is not in the caches (Texture or Vertex caches [pre or post T&L]) they switch to a different thread of execution (a different vertex).
You are presenting an interesting problem, which we have asked ourselves a lot: what will the SPUs/APUs do when the current thread stalls (for any reason) ?
Are SPUs/APUs in-order processors with no MT capability or are they Multi-Threaded ?
I think context-switching and MT can be "pushed" upon the SPUs/APUs on the PU side: the patents do mention how the PU can stop the SPU/APU execution, load a new Program Counter and a new context in the LS and re-start the execution on a new program using an APU RPC. I would assume that when the PU "forces its will" on the APUs that the APUs do save their context and can go back to it.
Maybe we will see this kind of multi-threaded handled by the CELL OS transparently enough to developers using the providd libraries (unless they want to go down to the metal).
IMHO, several things have changed since Sony first sought to use CELL technology to ease developers work.
I do believe that the PU might have changed in complexity and performance compared to earlier concepts.
Something I hear PlayStation 2 developers really hating is not VU programming or the GS having a limited feature-set or having to handle the DMAC to make sure VUs are working alongside the RISC core and the GS, etc...
What a lot of people hate is the fact that there is big need of people that take tons and tons of lines of C/C++ general purpose code (for physics, A.I. or other tasks) and convert it into R5900i friendly code working around the tiny D-Cache and using the SPRAM as much as possible: all things that PlayStation 2's GCC is not able to do.
We can have a simple and not too fast PU that does not recognizes the limits of compilers' technology or does not do much to work better with the compilers and thus push people again to re-write tons of lines of C/C++ code in ASM for all kinds of code.
Yes, I see the challenge... I see the glitter in your eyes when thinking of fighting the machine with the stenghts of your ASM knowledge, but this is not where game development is going IMHO: just like we stopped writing large portions of OS's in ASM (I seriously doubt that for example Longhorn is being written directly in ASM or even a significantly large portion of it) we will stop writing large portions of general purpose code in ASM for game programming.
Take something like the Athlon 64 and the MSVS 7.x compiler: I doubt it is worth writing huge portions of a PC game in ASM directly or to convert large portions of existing C/C++ code into x86 optimized ASM code because the code MSVS 7.x generates is way too crappy for the CPU it runs on.
Sure, an ASM programmer in most cases can still design faster and more efficient code, but we have to ask ourselves: how much faster is it ? how much time did this took ?
Modern micro-processors are designed to go towards compilers (an extreeme case of this is IA-64/EPIC): they contain larger on-chip caches (SPRAM is a very nice concept which I like, but it requires the programmer to manually take care of it and use it which means hand-writing ASM code) they are able to look at relatively large portions of the code and re-organize instructions to improove performance, hide latencies due to memory fetches or other kinds of problems creating stalls in the pipeline, switch thread of execution or even execute more than one thread at once.
If you compare how standard C/C++ code produced by the same version of GCC runs on PlayStation 2 and on Xbox, you will likely see an increase in performance that goes beyond the MHz difference.
In the patents I have seen recently we have shifted from e-DRAM on the CPU and in general from the use of large Work RAM buffers to caches and the rumors of a very fast PU core and a single PE with a certain number of APUs (instead of multiple PEs with smaller and less complex PU cores): PU's went from maybe L1 cache only to having an L2 cache which could be read/written by the SPUs/APUs and the e-DRAM on the CPU seems to have been taken out and instead of it we have, for the SPUs/APUs, a shared L1 cache.
IMHO for the APUs having a shared L1 cache is much better than having e-DRAM. Why ? Because the caches, being transparent to the applications, help performance without requiring major changes to the compiler and other high level libraries used or re-writing or large portions of code in ASM.
A cache-less system with lots of storage readable and writeable might be fast, but it would have to receive well optimized and hand-coded instruction streams and it would perform like crap if you just passed it code produced by GCC or MSVS's compiler.
The amount of responsability given to the PU has risen now (at least in PlayStation 3... the PU can be changed in performance, ISA, etc... if needed) and so have its complexity, frequency and over-all perfomance IMHO.
We expect the CELL OS and high-level libraries to handle more and more things transparently and we want to write more and more code using high-level languages: all of this requires a different idea of central CPU core that was present in the minds of the Emotion Engine's architects.
darkblu
30-Dec-2004, 01:01
Let I be a register with 3 16 bit tuples identified as 0,1,2
Let A be the address of the index stream
Let B be the address of the vertex stream
Let n be the current triangle number
Load I from n[A]
Load R0 from I.0[B]
Load R1 from I.1[B]
Load R2 from I.2[B]
VertexShade( R0 )
VertexShade( R1 )
VertexShade( R2 )
If we assume a memory latency of 200 cycles (thats very generous...) and that we can process 3+ memory requests at once we have 400 cycles to lose. Load I from n[A] is relatively easy, a predictor will notice the linear read and prefetch 200 cycles early, so when the load happens its hopefully already there. But the Load Rx from I.x[B] is much harder... I can't predict until the index register is filled, so I have to stall the shader for 200 cycles. The only serious option is a thread context switch.
i'm afraid a contex switch may not help either and you'd end up with things "just not working" even on a gpu if your indexing locality throughout the buffer is piss poor. so i'll second marco in this say that you don;t attacj brick walls head-on, you usually try to circumvent them : )
I don't deny it can be made to work, but my point is that ease of use just got thrown out the window.
The idea that a CPU is as good as a hardware vertex unit is false IMO. Its may do the job but to get VS3.0 functionality out of a CPU is hard, not because of FLOPs or functionality but memory latency.
As for Cell context switching I'm not convinced, we have to assume the 128K local RAM is shared among contexts (no way it could be saved/restored in time). We still have 2K of register to save/restore, assuming 128 bits per cycle read/write (to L1 cache I suppose) thats 32 cycles (16 per way) unless we have hardware contexts. Thats assuming hardware switching, if its actually software involving a PU interupt, then we have at least 10 cycles latency for the PU to react.
i'm afraid a contex switch may not help either and you'd end up with things "just not working" even on a gpu if your indexing locality throughout the buffer is piss poor. so i'll second marco in this say that you don;t attacj brick walls head-on, you usually try to circumvent them : )
At least one architecture has made the problem largely go away (of course there are pathologic cases but for most uses of index buffers and depedent texture reads its copes)
You just needs lots and lots of contexts (and a seriously good memory system). You have to stop thinking like a CPU where 8 contexts would be considered alot.
Panajev2001a
30-Dec-2004, 01:41
I don't deny it can be made to work, but my point is that ease of use just got thrown out the window.
The idea that a CPU is as good as a hardware vertex unit is false IMO. Its may do the job but to get VS3.0 functionality out of a CPU is hard, not because of FLOPs or functionality but memory latency.
True, but I expect developers to do some pre-packaging work for data and to work-around problems like this while still using a high level graphics library and not going to a PlayStation 2 kind of approach.
You just needs lots and lots of contexts (and a seriously good memory system). You have to stop thinking like a CPU where 8 contexts would be considered alot.
Like what? 32 or 64 or ... ?
version
30-Dec-2004, 02:09
I don't deny it can be made to work, but my point is that ease of use just got thrown out the window.
The idea that a CPU is as good as a hardware vertex unit is false IMO. Its may do the job but to get VS3.0 functionality out of a CPU is hard, not because of FLOPs or functionality but memory latency.
TRUE, but APUs run on 4.6 GHZ in cell, while in gpu VS run 800MHZ
cell will faster, in gpu more pixelshader will faster than ati's gpu
sony save transistor count with this method
Here's my prediction on what has happened thus far with PS3 development
1. Many people working on PS3 hardware didn't know or expect this to happen.
Companies always have Plan B, C, D, E....
2. The Nvidia's invovlement in PS3 did not start for one or two years before this annoucement. Nvidia's involvement was minimal, nothing more than NV execs talking on a high level to Sony execs, and sending out graphics cards for Sony to evaluate performance. Sony had been evaulating Nvidia tech for a while with zero cash exchanged between both parties. Nvidia didn't change what anything in thier roadmap to accomodate sony up until this annoucement.
That's the case too IMO. As we know, their concentration is on cell, the pixel engine was not something they make a big deal about. Even on the cell patent it was seems to be tacked on.
Anyway if they did something, you would already seen this annoucement earlier.
3. Toshiba was left out in the cold with this change and didn't know it was going to happen. The work on the GS for PS3 will continue, but that graphics hardware won't be used in the PS3.
Toshiba needs to develop their own pixel engine or license them from somewhere to be paired with their own Cell. I doubt they work specifically for PS3, and their exchange on this could be similar to what you said about NVIDIA. They would certainly make somekind of announcement earlier if they did otherwise.
4. One of sony's offices led the charge to make this change, as internally issues were raised with the expected performance of the new GS and/or expected feature set/delivery time frame.
That could be the case. Though it wasn't time frame or feature set, its most likely cost.
5. The nvidia licensing deal will cost Sony more money
Executives will not allow this. Executives will force the engineers to downgrade the spec either from NV part or Cell part to meet the budget. NV license deal was just a better value for money, that's why they chose it over their other options.
If I am speculating, for all we know, Sony owns solution could be superior to NV, but it broke the bank, and downgrading their solution, made NV solution a better performer. That could be why they go with NV.
But Sony would have a fixed range for PS3 budget, that they wouldn't break. That's not a speculation, that's just real world. Sony is building a PS3, not a bridge.
Panajev2001a
30-Dec-2004, 04:27
Look, sometimes budgets are increased if there is the need.
If you planned to spend $35,000 on a BMW 3 Series, but you see that for the model you want is either $1,000 more or a Yugo, you spend the extra $1,000.
According to your reasoning, V3, PSP would have never been launched in Japan this December, they would have delayed the launch as the budget they thought was good for it had to be majorly revised: PSP was pushed through the fabs in December because Son y wanted to hit the launch date even if it would cost them a lot of money (some hardware bugs had been recently fixed and to posh the amount of units through their manufacturing plants costed a premium in that period: the volume was quite low).
Sony will not lose an infinite amount o money on their hardware, but they still can adjust their decisions on how much they plan to lose per unit and how soon after launch these losses can be lowered.
Also, V3... neither Toshiba nor nVIDIA is providing ROPs/Pixel Engines to Sony: both of them were racing for a full GPU contract which includes Shaders, Texture Units, etc...
Look, sometimes budgets are increased if there is the need.
If you planned to spend $35,000 on a BMW 3 Series, but you see that for the model you want is either $1,000 more or a Yugo, you spend the extra $1,000.
No, not in this case. Consumers aren't in it to make money :), Sony is. Anyway, in business, luxury items are normally the first thing to get cut off. When you can't afford it, you can't afford it. Even the extra $1000.
According to your reasoning, V3, PSP would have never been launched in Japan this December, they would have delayed the launch as the budget they thought was good for it had to be majorly revised: PSP was pushed through the fabs in December because Son y wanted to hit the launch date even if it would cost them a lot of money (some hardware bugs had been recently fixed and to posh the amount of units through their manufacturing plants costed a premium in that period: the volume was quite low).
No, they already budgeted for that. Last minute hardware bugs are emergency, that sometimes happend. But those are bugs.
Sony will not lose an infinite amount o money on their hardware, but they still can adjust their decisions on how much they plan to lose per unit and how soon after launch these losses can be lowered.
No, those ranges are fix when they made the investment, and plans layed out. Options like abandoning Cell, going with Toshiba, NV, ATI, downgrading or upgrading spec, those are not fix though.
Also, V3... neither Toshiba nor nVIDIA is providing ROPs/Pixel Engines to Sony: both of them were racing for a full GPU contract which includes Shaders, Texture Units, etc...
That's what a GPU is isn't ? Pixel Engine is just Sony fancy name.
Look I am not saying PS3 is low budget or anything (its probably higher than the other two for all I know), but they wont increase the cost of it, just for the sake of increase in performance, that's futile in the eye of consumers. If it other features that may be visible, like Blu Ray, maybe they'll make allowance. But going to NV because their solution is expensive and will over budget the project but it gives better performance, thats just silly. They just won't go over budget. There are too much risk involved in doing that.
Like I said before going to NV is definitely the cheapest solution for Sony. Sony Plan A solution broke the bank so not feasible, Plan B look elsewhere for solutions that meet the cost. NV won the contract because it meets their cost, where others might not.
Panajev2001a
30-Dec-2004, 06:17
V3,
It is not going over budget, as you yourself said they already left wiggle room in the budget ;).
No, those ranges are fix when they made the investment, and plans layed out.
See ;) ?
That's what a GPU is isn't ? Pixel Engine is just Sony fancy name.
If you think that nVIDIA is in to provide only Pixel Engines/ROPs then you are mistaken.
Look at NV40, there is a lot more stuff that makes a GPU than just Pixel Engines/ROPs.
Like I said before going to NV is definitely the cheapest solution for Sony. Sony Plan A solution broke the bank so not feasible, Plan B look elsewhere for solutions that meet the cost. NV won the contract because it meets their cost, where others might not.
That is a fancy way of looking at things: it is not that nVIDIA's solution is superior, it is the most affordable.
Plan A might have not passed budgeting (internal... Visualizer), but I am not aware that plan B (which jad been the official plan for quite a while) went overbudget as the contractors for plan B seemed to be a bit surprised at the coup that nVIDIA managed.
I think nVIDIA solution might have out-performed and out-featured plan B, but Toshiba had the lower cost (being Sony/SCE's partner, having already done work with XDR, Redwood and CELL they could afford to push the envelope without needing too much R&D money [less money than what Sony/SCE deemed as maximum R&D budget for the GPU]).
I also think that nVIDIA might have convinced Sony/SCE because of their change on the royaltes issues: it might have been what Sony/SCE needed (plus a nice big argument by nVIDIA about where the future of GPUs was headed and how the GPU they selected so far was nowhere near, according to nVIDIA of course, what Sony/SCE needed for PlayStation 3) to finally change their minds.
nVIDIA did not even seem a likely winner of the contract in the two weeks before the announcement was made, not at all.
You just needs lots and lots of contexts (and a seriously good memory system). You have to stop thinking like a CPU where 8 contexts would be considered alot.
Like what? 32 or 64 or ... ?
Depends on the memory latency you want to hide...
But they sound reasonable figures for a GPU.
I assume Cell's DMA engines will support scatter/gather. Do the patents back this up?
I assume Cell's DMA engines will support scatter/gather. Do the patents back this up?
It seems so:
Method for asynchronous DMA command completion notification (http://appft1.uspto.gov/netacgi/nph-Parser?Sect1=PTO2&Sect2=HITOFF&p=1&u=%2Fnetahtml%2 FPTO%2Fsearch-bool.html&r=7&f=G&l=50&co1=AND&d=PG01&s1=hofstee&O S=hofstee&RS=hofstee)
[0019] With the flexibility of this approach, software can group DMA commands in order to manage them. For instance all commands for a particular "task" can be grouped into a single tag group. Alternatively, all DMA "get" commands can be placed in a group separate from an output group comprising all DMA "put" commands. In addition, hardware can provide additional command parallelism or ordering rules with respect to groups. The APU software can verify that a single group has completed, all groups have completed, or a specified set of groups have completed operations. In the current embodiment, tag group status is supplied by the APU reading a data channel, where each bit in the channel represents a tag group status. Bit 0 represents tag group 0 status, bit 1 tag group 1 status, and so on up to bit 31 for tag group 31. A 0 indicates the tag group is complete, a 1 in the corresponding position indicates the tag group has one or more outstanding commands not yet completed.
[0020] There are several variations on the above and a number of advantages associated with the different variations. In one embodiment, the DMA queue 135 can store up to 32 DMA commands. All DMA commands in the DMA queue 135 could have the same tag group number, they could all have different tag group numbers, or anything in between.
Auto prefetch is also supported:
DMA prefetch (http://appft1.uspto.gov/netacgi/nph-Parser?Sect1=PTO2&Sect2=HITOFF&p=1&u=%2Fnetahtml%2 FPTO%2Fsearch-bool.html&r=10&f=G&l=50&co1=AND&d=PG01&s1=kahle&OS =kahle&RS=kahle)
The load access pattern generally contains information on the data being transferred. The load access pattern can be used to predict future data transfers and prefetch data to the SL1 cache before the MFCactually requests the data. When the MFC actually requests the data, the MFC does not have to go all the way back to the system memory 212 to retrieve the data. Instead, the MFCaccesses the SL1 cache to retrieve the data and transfer the data to the local store
So a CELL CPU should support scatter and gather (with an authomatic prefetch mechanism) DMA operations, up to 32 operations at the same time. DMA ops can be grouped and prioritized and the SPU can check for single o groups completitions.
ciao,
Marco
....
With this said, I still say that IMHO the PlayStation 3 GPU is not CELL based, it does not have the SPUs/APUs.
....
It would seem to me a not bad idea to assign all the Vertex Shading work to the CELL based CPU:
....
To me the above two statements seem to contradict each other.. :?
Explain how.
Vertex Shading done by units like the SPUs/APUs or like the VUs in the EE can be done very effectively, while Pixel Shading might not (nVIDIA does not believe in using the same hardware for both and for now they might be right).
Splitting the VS and PS work between the CPU and the GPU is what some PS heavvy games on Xbox 2/Xenon will do: all the unified shading units would be dedicated to Pixel Shading work and Vertex Shading would then be done on the enhanced VMX units of the Xbox 2/Xenon's CPU.
I guess you've read the rest of my post for me not to explain it again! ;)
On a side note, what you've posted above would've been a natural question to ask after reading 'point 3' of my explanation which would lead back to 'point 1' again and imply the GPU should be CELL based. :P
1. Basically, the way I see it is this...if you're gonna have vertex shading on the CELL CPU, then they're gonna be VS CELL threads, (aka software Cells), no?
If Hofstee was talking about two-way comms between CPU<=>GPU, then these VS CELL threads should run on the GPU also, no?
Ideally that is what they would want, which is why I talked about the direction for CELL 2.0 for example. IBM, with CELL, sees a point in pushing towards the same direction ATI is pushing with unified shading hardware.
Unified graphics hardware != CELL.
By CELL 2.0, I presume you mean what CELL 1.0 should be from the patents? I.e. Run everything, all threads, scalar and vector, be it graphics or non-graphics threads in a distributed, broadband environment under a unified ISA?
IMHO, if they can't achieve that with CELL 1.0, then the architecture is just a glorified PS2 EE that uses distributed processing. The main addition would be for the VU equivalent S|APUs to work independantly. Don't get me wrong, this is still not bad, it's just a natural progression of tech. But it's not the vision laid out in the patents.
This is not the way nVIDIA sees things, not for theshort-to-medium term at least.
I disagree. I'm not sure what timescales you mean by short/medium term. But listening to that Huang interview/webcast with Morgan Stanley, he see's the graphics/media processor of the future as a programmable DSP. The CELL chip is basically a programmble DSP. The ATI R500 is basically a programmble DSP. I know on the VS/PS level ATI are looking at identical units but NV are differing on those being specialised. But this should not stop VS threads running on PS units and PS threads running on VS units.
In fact I see the NV approach capitalising on CELL and complemeting as mentioned in my other threads. If NV have two sets of shader units that are specilaised and optimised for their respective VS and PS parts, then for a future PCI-E PC's and PS3 IC's, we would see them employed ,
PC,
[CPU]<=>[VS<=>PS]
PS3,
[CELL]<=>[PS]
CELL Workstations perhaps,
[CELL<=>PS]
It will be interesting to see if they'll use Cg (very likely) and whether it will compile and run shaders on BOTH the CELL CPU and the NV GPU. It would make sense if they did. In my eyes, running Cg shaders would just become 'software renderring' executing on 'specilaised' SIMD hardware. :)
It will be interesting to see if they'll use Cg (very likely) and whether it will compile and run shaders on BOTH the CELL CPU and the NV GPU. It would make sense if they did. In my eyes, running Cg shaders would just become 'software renderring' executing on 'specilaised' SIMD hardware. :)
SCEA was hiring engineers for new shader compiler development... where does it fit in?
darkblu
30-Dec-2004, 22:39
It will be interesting to see if they'll use Cg (very likely) and whether it will compile and run shaders on BOTH the CELL CPU and the NV GPU. It would make sense if they did. In my eyes, running Cg shaders would just become 'software renderring' executing on 'specilaised' SIMD hardware. :)
SCEA was hiring engineers for new shader compiler development... where does it fit in?
it fits in as in 'they just hired some' ; )
Panajev2001a
30-Dec-2004, 23:24
IMHO, if they can't achieve that with CELL 1.0, then the architecture is just a glorified PS2 EE that uses distributed processing. The main addition would be for the VU equivalent S|APUs to work independantly. Don't get me wrong, this is still not bad, it's just a natural progression of tech. But it's not the vision laid out in the patents.
Yes, CELL 2.0 being the full vision of the patents and more of course (the road does not stop there).
I am not saying that CELL 2.0 is a new ISA, a totally different road.
Think CELL 1.0 as being the implementation of CELL as you will see in the WorkStation and in PlayStation 3 and CELL 2.0 to be a later implementation.
Like Treebeard would say, don't be so hasty master Hobbit ;).
The vision laid out in the patents was natural progression of existing technologies and ideas that have been theorized 10-15 or even 20 years ago in laboratory environments.
I disagree with saying that Sony did not try to experiment how you could build an entire system made of CELL ICs and few sections of custom logic/small dedicated ICs, but Sony/SCE is in the business of making of PlayStation 3 and the CELL WorkStations the best possible products they can make, not proofs of concepts.
On this I agree with V3, Sony/SCE does not make money prooving the a concept lke "use CELL for everything" works in a high volume machine if this means monumental losses.
This does not mean the architecture is not good: wait to see CELL based products like hopefully some renderfarms.
Maybe some customer will ask IBM and Sony to deliver a new renderfarm and toolset chain or maybe Sony ImageWorks itself willl ask for a few million dollars worth of a system and with that kind of budget they might decide to push CELL far enough to allow for something like the full Broadband Engine and its Visualizer chip.
I believe we will see that once Sony/SCE starts PlayStation 4 R&D, just like we saw the GSCube-32 and GSCube-64 machines.
I know on the VS/PS level ATI are looking at identical units but NV are differing on those being specialised. But this should not stop VS threads running on PS units and PS threads running on VS units.
It does as it would be really inefficient: your example of PS threads running on VS units is exactly what nVIDIA says they do not want right now as the unit would not be used optimally at all.
PC,
[CPU]<=>[VS<=>PS]
PS3,
[CELL]<=>[PS]
In order to share data and wor together the CELL CPU and the nVIDIA GPU do not need to run the same code, run the same Apulets: they just need to agee on what kind of interfaces you use to communicate between the two. More like a "you know if I leave this quad-word here that it is a Vertex because blah+blah is written at memory location blah-blah-blah, etc...".
Would'nt Cell 1.0 be 90NM and 2.0 Cell on 65NM?
Some quotes in regards to Nvidia and Sony rumors from CNN in 2003...
The reality is nVidia is not sitting in a vacuum," said Erach Desai, an analyst with American Technology Research. "They are in discussions with Sony for the PS3."
They realize, I think, that they cannot do it all," said Desai. "I have checked with a couple of very seasoned executives ... and the strong impression is that there is interest from Sony and interest from nVidia."
"We've always said we'd be happy to be in any game console," nVidia spokesperson Carrie Cowan told me recently.
"I would probably characterize it as a less than 50 percent chance that they win PS3," said Michael McConnell of Pacific Crest Securities. "However, we have talked with Sony and their take on it is they're considering an external vendor as well as an internal solution. So you can't rule it out, but you definitely can't say it's a sure thing."[/b]
Would'nt Cell 1.0 be 90NM and 2.0 Cell on 65NM?
Pana has been pre-gaming for tomorrow all week, don't pay attention to his rants on v1.0 and v2.0. :P
But yes, there are 90nm Cell processors and there will be 65nm fabricated ones as well. No, they don't coincide with his comments.
Panajev2001a
31-Dec-2004, 00:21
Would'nt Cell 1.0 be 90NM and 2.0 Cell on 65NM?
Pana has been pre-gaming for tomorrow all week, don't pay attention to his rants on v1.0 and v2.0. :P
:lol
Vince, I did explain what I meant.
I do not think PlayStation 3 will fully go the way the first patent showed: namely a wholly CELL based system with a CELL based Visualizer.
One day we might see CELL based set-ups going the way ATI seems to be going: an extension of the Visualizer concept. A single IC doing both general purpose processing as well as graphics processing with a sea of computational blocks exchanging data with a seoarate sea of dedicated silicon blocks (TMUs, ROPs, etc...).
Megadrive1988
31-Dec-2004, 00:29
Panajev, well, I hope that Sony fully funds smaller companies like the one that makes the SAARC raytracing processors. for Playstation4. I'm hopeful since Sony was willing to go with Nvidia for the PS3 GPU or at least a very very significant portion of it.
PS2 CPU: Emotion Engine
PS3 CPU: Broadband Engine (?)
I hope PS4 CPU: RayTracing Engine 8)
Panajev2001a
31-Dec-2004, 01:32
I am sorry, but I want the REYES Engine and always will ;).
Renderman forever :D.
Megadrive1988
31-Dec-2004, 02:12
can't we somehow have REYES, Raytracing & GI together ?
Dave Baumann
31-Dec-2004, 02:39
I do not think PlayStation 3 will fully go the way the first patent showed: namely a wholly CELL based system with a CELL based Visualizer.
Are we only just coming to this realisation?
Panajev2001a
31-Dec-2004, 02:53
I do not think PlayStation 3 will fully go the way the first patent showed: namely a wholly CELL based system with a CELL based Visualizer.
Are we only just coming to this realisation?
No, I am repeating what I have been saying for quite a bit actually.
The GPU was not going to be CELL based like the Visualizer even when the other contractor had the contract and I do not expect this to be different with nVIDIA.
Inane_Dork
31-Dec-2004, 09:59
I do not think PlayStation 3 will fully go the way the first patent showed: namely a wholly CELL based system with a CELL based Visualizer.
Are we only just coming to this realisation?:lol:
That got me laughing. Some, who will remain nameless, were giddy as schoolboys in hopes of a revolutionary, CELL-based rasterizer. I don't think they wanted to give that up.
version
31-Dec-2004, 14:33
http://pc.watch.impress.co.jp/docs/2004/1228/kaigai02l.gif
hmm?
Panajev2001a
31-Dec-2004, 19:21
I do not think PlayStation 3 will fully go the way the first patent showed: namely a wholly CELL based system with a CELL based Visualizer.
Are we only just coming to this realisation?:lol:
That got me laughing. Some, who will remain nameless, were giddy as schoolboys in hopes of a revolutionary, CELL-based rasterizer. I don't think they wanted to give that up.
Name me, I would have loved to see a REYES based set-up with a pure CELL based chipset :).
Would have it been cool ? Yes.
Is it the best thing for the PlayStation 3 taking timing, cost, etc... into consideration ? Probably not, hence the switch to the Toshiba solution and then to the nVIDIA solution.
...
Like Treebeard would say, don't be so hasty master Hobbit ;).
The vision laid out in the patents was natural progression of existing technologies and ideas that have been theorized 10-15 or even 20 years ago in laboratory environments.
It's not a matter of haste...they would've had 5yrs+ with access to immense resources! ;)
Anyway like you say it's been theorized for a while. IIRC, the TAOS operating system had these concepts of 'software Cells' for distributed graphics renderring and general parallel processing and in a heterogeneous environment, using an object oriented MPI programming model, compiled to a VM processor over TEN years ago...and it was a working model! :P
TAOS (http://www.byte.com/art/9407/sec6/art1.htm)
They have now formed into 'TAO Group' who have, interestingly enough, a link with SONY. They have a Realtime OS and middleware platform for CE devices...and Sony are a partner who have invested millions into that group. Perhaps a link with PS3 :?:
http://tao-group.com/about/about.php
PS. Notice all the playstation game images on their site ! ;)
...
This does not mean the architecture is not good: wait to see CELL based products like hopefully some renderfarms.
I'd like to know how this NV-Sony collaboration will impact their DCC market with NV's Gelato and Sony's CELL workstations. NV sell Gelato as a distributed HW/SW renderring package optimised for their GPUs.
If there are gonna be NV GPUs in CELL workstations, then these GPUs must be able to provide similar distributed renderring, alongside the CELL chips. Otherwise if NV continue selling PC based Gelato workstations in the same market as Sony's CELL workstations, then they aren't leveraging their collective strengths in distributed processing/renderring. One's just using CPUs and the other it's GPUs and NOT BOTH to their full potential. This is another reason to get CPU + GPU to work together on 'software Cells' and also for NV to license CELL processors, IMHO.
I know on the VS/PS level ATI are looking at identical units but NV are differing on those being specialised. But this should not stop VS threads running on PS units and PS threads running on VS units.
It does as it would be really inefficient: your example of PS threads running on VS units is exactly what nVIDIA says they do not want right now as the unit would not be used optimally at all.
Yes, I know it will be inefficient but the point is to allow a degree of flexibility for VS or PS heavy graphics. There will be an optimun VS and PS load on the CPU+GPU chipsets. But a degree of freedom allowed either way. It will be interesting to know this inefficiency factor if NV indeed have these optimised VS and PS units for their new architecture. :)
It will be interesting to see if they'll use Cg (very likely) and whether it will compile and run shaders on BOTH the CELL CPU and the NV GPU. It would make sense if they did. In my eyes, running Cg shaders would just become 'software renderring' executing on 'specilaised' SIMD hardware. :)
SCEA was hiring engineers for new shader compiler development... where does it fit in?
This was the Job Pana posted,
The successful candidate will develop a state-of-the-art shading language compiler for an advanced forth-generation graphics processing unit (GPU). With the assistance of other team members, the individual must be capable of designing and implementing the major components of the compiler backend.
To achieve this goal, the individual should have extensive recent experience with backend internal representations suitable for advanced code optimization. Detailed knowledge of modern code optimization techniques, register allocation, and code generation expertise is also required, as well as experience with programming language front ends, assemblers, linkers, and runtime libraries. Exposure to shading languages, such as nVidia CG, Microsoft HLSL, Brook, or StreamIt, and exposure to 3D graphics APIs, such as OpenGL and DirectX, is also desirable.
http://www.beyond3d.com/forum/viewtopic.php?p=423029#423029
I checked the link and the good news is that the vacancy has been filled!...pheeeww! :P
It mentions the major components of the compiler back-end being developed by this SCEA applicant. I would've thought NV would do most of that if they're supplying the GPU parts. :? ...still they could be extending it to the CELL CPU also...and/or there's custom Sony GPU parts involved too. :?:
I do not think PlayStation 3 will fully go the way the first patent showed: namely a wholly CELL based system with a CELL based Visualizer.
Are we only just coming to this realisation?
I'm sure you're aware of this but just to clarify, for the GPU to be Cell based, it just needs to run 'software Cells' and doesn't have to be the Visualizer from the patents or configured like a S|APU from the CELL CPU. They only need a consistent ISA to run 'software Cells'!
Just ask DeanoC about ISA's, G5's and Xenon CPU! ;)
http://pc.watch.impress.co.jp/docs/2004/1228/kaigai02l.gif
hmm?
Could someone translate the text in that diagram please. Thanks in advance!
Happy New Year! :D
Panajev2001a
31-Dec-2004, 22:00
I do not think PlayStation 3 will fully go the way the first patent showed: namely a wholly CELL based system with a CELL based Visualizer.
Are we only just coming to this realisation?
I'm sure you're aware of this but just to clarify, for the GPU to be Cell based, it just needs to run 'software Cells' and doesn't have to be the Visualizer from the patents or configured like a S|APU from the CELL CPU. They only need a consistent ISA to run 'software Cells'!
I give up ;).
:lol:
The NV5X architecture was not developed with CELL in mind or using the Apulet ISA/runnign the same Apulets and required several years of work and hundreds of engineers to be completed: wouldn't you think that nVIDIA would need more than 50 engineers for a bit more time to convert all their Shading cores to support the execution of Apluets/Software cells ?
You will see more and more collaboration between CPU and GPU in the PC when the I/O bus that connects them gets faster and faster (PCI-Express is already pushing towards it and PCI-Express 2.0 goes one step forward in the same direction delivering even more bandwidth).
This is done without having the PC GPU adopt the x86 ISA.
The point of having the GPU be CELL based was the point behidn the Visualizer: using the strengths of distributed processing well built in the CELL architecture to their full extent.
Eventually the CELL philosophy really leads to a BIG CELL IC which works on general purpose processing, rendering, A.I., Physics, etc... attached to a co-processor which only provides less and less hardwired functions (in the future you will want programmable AA, programmable Texture Filtering, etc...).
Dave Baumann
01-Jan-2005, 17:42
I'm sure you're aware of this but just to clarify, for the GPU to be Cell based, it just needs to run 'software Cells' and doesn't have to be the Visualizer from the patents or configured like a S|APU from the CELL CPU. They only need a consistent ISA to run 'software Cells'!
I’m fairly sure you know that’s not what Pana was talking about.
I do not think PlayStation 3 will fully go the way the first patent showed: namely a wholly CELL based system with a CELL based Visualizer.
Are we only just coming to this realisation?:lol:
That got me laughing. Some, who will remain nameless, were giddy as schoolboys in hopes of a revolutionary, CELL-based rasterizer. I don't think they wanted to give that up.
Name me, I would have loved to see a REYES based set-up with a pure CELL based chipset :).
Would have it been cool ? Yes.
Is it the best thing for the PlayStation 3 taking timing, cost, etc... into consideration ? Probably not, hence the switch to the Toshiba solution and then to the nVIDIA solution.
I wanted to reply to put my perspective down on where I've been coming from over the past year or so of discussions but I also wanted a wider discussion so I've opened a new thread here (http://www.beyond3d.com/forum/viewtopic.php?p=438761#438761).
passerby
01-Jan-2005, 18:19
Could someone translate the text in that diagram please. Thanks in advance!
The original article. http://pc.watch.impress.co.jp/docs/2004/1228/kaigai146.htm
Worth taking a look since it has a link to one Intel slide that sounds and looks very, very familiar.
The Intel presentation says how they are going 'many-core' looking at numbers up to 10-100. Should really click on that slide for the rest of the so-familiar 'key words'.
Rest of the article just briefly summarizes problems, solutions of going 'many-core' and TDP(Thermal Design Power). The slide version linked to is of course, a 'Cell example'. Note that Goto is just making 'educated speculations'.
I have trouble understanding what he is really trying to say there, since he refers to 'performance' and 'performance/TDP' seperately. I assume he means the ratio. In which case...
- single core: high performance, p/tdp low
- multicore, 'sophisticated cores' - cores more complex, high performance, p/tdp not too high.
- multicore, 'simpler cores' - high p/tdp, scalable, performance/efficiency suffers.
- 'customized' CPU core - many specialised multimedia CPU cores linked to a general purpose CPU. very high p/tdp if used properly.
I still think the Intel slide is more fun...
:D
It is not going over budget, as you yourself said they already left wiggle room in the budget.
If their solution can fit in that wiggle room, they would go with their own solution. But it didn't fit. That's why they go with NV, a cheaper solution that fit their budget.
If you think that nVIDIA is in to provide only Pixel Engines/ROPs then you are mistaken.
Look at NV40, there is a lot more stuff that makes a GPU than just Pixel Engines/ROPs.
Pixel Engines, is whatever it needs to shade pixels, so if it needs other stuff other than ROPs it will be there.
That is a fancy way of looking at things: it is not that nVIDIA's solution is superior, it is the most affordable.
Not exactly, superior and inferior is just speculation, but because Sony went with NV, NV solution is definitely meet their budget.
Plan A might have not passed budgeting (internal... Visualizer), but I am not aware that plan B (which jad been the official plan for quite a while) went overbudget as the contractors for plan B seemed to be a bit surprised at the coup that nVIDIA managed.
I think nVIDIA solution might have out-performed and out-featured plan B, but Toshiba had the lower cost (being Sony/SCE's partner, having already done work with XDR, Redwood and CELL they could afford to push the envelope without needing too much R&D money [less money than what Sony/SCE deemed as maximum R&D budget for the GPU]).
Being partner, doesn't always mean lower cost. NV undercut them. Like I said before, there are no motives to waste money on something better, performance or features wise. If they've contracted Toshiba to make something at their specs, it'll be good enough.
I also think that nVIDIA might have convinced Sony/SCE because of their change on the royaltes issues: it might have been what Sony/SCE needed (plus a nice big argument by nVIDIA about where the future of GPUs was headed and how the GPU they selected so far was nowhere near, according to nVIDIA of course, what Sony/SCE needed for PlayStation 3) to finally change their minds.
nVIDIA did not even seem a likely winner of the contract in the two weeks before the announcement was made, not at all.
That kind of argument won't win over Sony Pana. Like I said NV undercut Toshiba, to provide cheaper solutions, and won the contract. It might be better too, but that's speculation.
I do not think PlayStation 3 will fully go the way the first patent showed: namely a wholly CELL based system with a CELL based Visualizer.
Are we only just coming to this realisation?
I'm sure you're aware of this but just to clarify, for the GPU to be Cell based, it just needs to run 'software Cells' and doesn't have to be the Visualizer from the patents or configured like a S|APU from the CELL CPU. They only need a consistent ISA to run 'software Cells'!
I give up ;).
:lol:
Just trying to make my point not to discount a CELL based NV GPU under that definition too quickly. ;)
...
The NV5X architecture was not developed with CELL in mind or using the Apulet ISA/runnign the same Apulets and required several years of work and hundreds of engineers to be completed: wouldn't you think that nVIDIA would need more than 50 engineers for a bit more time to convert all their Shading cores to support the execution of Apluets/Software cells ?
...
Sure...but to put things into perspective, NV were using HUNDREDS of engineers to design the core GPU architecture and ALONGSIDE them, over the past 2 YEARS, have used 50 NV engineers to customize the GPU for the PS3 and ALONGSIDE them, an undisclosed number of SONY engineers. ;)
This is done without having the PC GPU adopt the x86 ISA.
Sure..but the x86 wasn't designed with the same CELL vision in mind. ;)
Any project and the PS3 is no different has a specification, a delivery time and allocated budget/resources. Of course, these will all be subject to change througout the course of the project. And all projects have several contingency plans to meet the spec and delivery time's at critical MILESTONES. But just because one of the resources is now OFFICIALLY NV, doesn't mean any of the project specs and delivery timescales have been altered. Re-allocation of resources are always being juggled in projects to MEET the spec and delivery date. Which noone here know without being NDA'd. ;)
Anyway, did you get a chance to read my link on the TAOS operating system?
http://www.beyond3d.com/forum/viewtopic.php?p=438551#438551
They get around the uniform ISA problem in a heterogeneous multi-processor environment by compiling to a 'virtual' processor with no overheads in translation due to a very efficient nano-kernel running on each processor. I know the CELL press releases mentions that the CELL processors can run multiple operating systems. If this means multiple nano-kernels or equivalent on each core, then they could be borrowing many ideas from TAOS for CELL. :)
Could someone translate the text in that diagram please. Thanks in advance!
The original article. http://pc.watch.impress.co.jp/docs/2004/1228/kaigai146.htm
Worth taking a look since it has a link to one Intel slide that sounds and looks very, very familiar.
The Intel presentation says how they are going 'many-core' looking at numbers up to 10-100. Should really click on that slide for the rest of the so-familiar 'key words'.
Rest of the article just briefly summarizes problems, solutions of going 'many-core' and TDP(Thermal Design Power). The slide version linked to is of course, a 'Cell example'. Note that Goto is just making 'educated speculations'.
I have trouble understanding what he is really trying to say there, since he refers to 'performance' and 'performance/TDP' seperately. I assume he means the ratio. In which case...
- single core: high performance, p/tdp low
- multicore, 'sophisticated cores' - cores more complex, high performance, p/tdp not too high.
- multicore, 'simpler cores' - high p/tdp, scalable, performance/efficiency suffers.
- 'customized' CPU core - many specialised multimedia CPU cores linked to a general purpose CPU. very high p/tdp if used properly.
Thanks for translation! :)
I still think the Intel slide is more fun...
:D
You mean this,
http://pc.watch.impress.co.jp/docs/2004/1228/kaigai03.jpg
:lol:
CELL inside, INTEL outside (TM)? :D
CELL inside, INTEL outside (TM)? :D
Intel won't make the mistake of putting a 1960s programming model on their system though (wrt. the memory model).
Cheers
Gubbi
Panajev2001a
02-Jan-2005, 21:29
If CELL APU's were considered by nVIDIA to be much better than their Vertex and Pixel Shader ALUs to spend a lot of effort and money replacing NV5X's ALUs with modified/enhanced APUs for a GPU contract they were not even sure of winning, it makes you think why NV5X is not CELL based ;).
Sure...but to put things into perspective, NV were using HUNDREDS of engineers to design the core GPU architecture and ALONGSIDE them, over the past 2 YEARS, have used 50 NV engineers to customize the GPU for the PS3 and ALONGSIDE them, an undisclosed number of SONY engineers. Wink
I do not think this is how it happened.
The GPU was not prepared in parallel with close collaboration of nVIDIA's engineers and SCE's engineers for that long (nVIDIA did follow closely PlayStation 3's development and started plannign for features and things that SCE might like in a console environment, but that is part of the "trying to win the contract" thing): the solution that was supposed to GO in PlayStation 3 officially was not nVIDIA's solution.
The custom GPU nVIDIA is providing to SCE started development when NV5X development was finishing down and nVIDIA was trying to win PlayStation 3's contract away from Toshiba.
I think that nVIDIA put their engineers to work on NV5X and if others were free during the ehavvy phase of R&D they were pout to work on NV4X revisions and early R&D for NV6X.
However what you talk about is not a simple customization of their architecture, something that would have required more engineers on their part.
There would the a good case for a lawsuit on the part of Toshiba if it was true that Sony was funding with engineers and resources (including CELL IPs) a customized NV5X while they had already told Toshiba that their GPU was going to be THE GPU used by PlayStation 3.
If CELL APU's were considered by nVIDIA to be much better than their Vertex and Pixel Shader ALUs to spend a lot of effort and money replacing NV5X's ALUs with modified/enhanced APUs for a GPU contract they were not even sure of winning, it makes you think why NV5X is not CELL based ;).
I thought you gave up! ;) I've already stated, I'm not bothered either way but I think I've made my point not to discount a CELL based GPU too quickly! Especially when we've come full circle with the above argument! ;)
For PC, the NV5x is NOT going to be CELL based. For PS3, it make's sense for it to be CELL based as you've already agreed to.
So we're now debating the CELL ISA and how that could be developed for a GPU for PS3 to the required spec and with the available budget/resources/engineers?
I've already given examples of ways for this ISA to work. Whether the contract remained in-house, went to Toshiba, ATI, or who ever...
1. In your above example, how do you know it won't be easier to modify NV ALUs than it would be to modify S|APUs for the CELL ISA? Would modified S|APUs be better at PS work than modified NV ALUs? They've chosen NV so...
2. I've already mentioned they could have a CELL template like the Toshiba MeP template here... (http://www.beyond3d.com/forum/viewtopic.php?t=16356&highlight=pixel+engine)
Toshiba has made a flexible MeP architecture to have modified ISA's. Why wouldn't CELL have something like this considering it's use in a similar market? Why wouldn't NV's next gen NV5x have a similar type of template to modify ISA's considering they're going into the same market as mentioned by Huang? Why are other companies looking at making customized ISA's, e.g. Tensilica's Extensa?
3. What about compiling to a 'virtual' processor CELL ISA as mentioned earlier like TAOS? In the press releases, STI have already stated that CELL is an 'OPEN' platform? How about making that CELL ISA 'OPEN' and introduce CELL to a wider market? How about encouraging HW/SW innovations for the CELL ISA to make inroads into the x86 ISA?
These are just my ramblings but you should get my gist! :D
...
There would the a good case for a lawsuit on the part of Toshiba if it was true that Sony was funding with engineers and resources (including CELL IPs) a customized NV5X while they had already told Toshiba that their GPU was going to be THE GPU used by PlayStation 3.
This is what lawyers are there for when signing contracts and sub-contracts! :P And remember PS3 project != CELL project. ;)
Panajev2001a
03-Jan-2005, 23:50
For PC, the NV5x is NOT going to be CELL based. For PS3, it make's sense for it to be CELL based as you've already agreed to.
I did not agree to that :P.
Panajev2001a
04-Jan-2005, 00:01
1. In your above example, how do you know it won't be easier to modify NV ALUs than it would be to modify S|APUs for the CELL ISA? Would modified S|APUs be better at PS work than modified NV ALUs? They've chosen NV so..
You knwo what is easier ?
Take the ALUs and the control logic and all the interfaces betwen Shading ALUs, TMUs, etc... that nVIDIA has developed and customize the architecture around those items :P.
There is more stuff to be done on the GPU than to modify it so that it becomes a CELL processor and share Apulets with the CELL based GPU.
Adding Shading ALUs and/or TMUs, adding e-DRAM, etc...
For PC, the NV5x is NOT going to be CELL based. For PS3, it make's sense for it to be CELL based as you've already agreed to.
I did not agree to that :P.
Well thanks for confirming this to me...now is the perfect opportunity to say we shall 'agree to disagree (TM)'! :P
Until now, I've read your post as , yes it makes sense to have a CELL based GPU, yes that would be cool...etc, etc, etc..and where we are disagreeing is that delivery time, resources/budget to meet that spec wouldn't be possible for PS3 but will be for PS4/ CELL 2.0? Hence my above post for various resolutions on the CELL ISA for PS3... ;)
And no, I'm not getting into a game of semantics here... :)
1. In your above example, how do you know it won't be easier to modify NV ALUs than it would be to modify S|APUs for the CELL ISA? Would modified S|APUs be better at PS work than modified NV ALUs? They've chosen NV so..
You knwo what is easier ?
Take the ALUs and the control logic and all the interfaces betwen Shading ALUs, TMUs, etc... that nVIDIA has developed and customize the architecture around those items :P.
There is more stuff to be done on the GPU than to modify it so that it becomes a CELL processor and share Apulets with the CELL based GPU.
Adding Shading ALUs and/or TMUs, adding e-DRAM, etc...
My preference for overall flexibility, simplicity and wider market penetraion is to make a CELL ISA on a 'virtual processor', a CELL VM, an Open standard. Like the Java VM but working closer to the metal for performance.
Then have various 'stream' programming languages, Cg, Brook etc. to compile to this CELL ISA, taking advantage of BOTH CPU+GPU being 'stream processors' to get a distributed 'software renderring' analogy. Making it open would let different stream languages develop in this CELL ISA environment :) Kinda like how C/C++ flourished under GNU but have the equivalent 'Stream-C/C++' flourish! :P
Panajev2001a
04-Jan-2005, 02:29
For PC, the NV5x is NOT going to be CELL based. For PS3, it make's sense for it to be CELL based as you've already agreed to.
I did not agree to that :P.
Well thanks for confirming this to me...now is the perfect opportunity to say we shall 'agree to disagree (TM)'! :P
Until now, I've read your post as , yes it makes sense to have a CELL based GPU, yes that would be cool...etc, etc, etc..and where we are disagreeing is that delivery time, resources/budget to meet that spec wouldn't be possible for PS3 but will be for PS4/ CELL 2.0? Hence my above post for various resolutions on the CELL ISA for PS3... ;)
I want to be clear.
I said that I would have liked a CELL based GPU... eventually I'd like to make of somethign like the Visualizer the only IC in the chipset if it were possible, fusing graphics and general purpose processing... no need of a Broadband Engine and a Visualizer... just a sea of shading ALUs and PUs with a pool of fixed function units as co-processor basically.
This is not what we can get IMHO in PlayStation 3: it is something we can look forward for the more distant future.
NV5X is not CELL based and at this point it would not make sense (financially and engineering wise) for the PlayStation 3 GPU to be CELL based as it is based on the NV5X architecture.
MrSingh
04-Jan-2005, 02:51
ohhh.. the RS would have sucked so badly compared to the NV offering.
For PC, the NV5x is NOT going to be CELL based. For PS3, it make's sense for it to be CELL based as you've already agreed to.
I did not agree to that :P.
Well thanks for confirming this to me...now is the perfect opportunity to say we shall 'agree to disagree (TM)'! :P
Until now, I've read your post as , yes it makes sense to have a CELL based GPU, yes that would be cool...etc, etc, etc..and where we are disagreeing is that delivery time, resources/budget to meet that spec wouldn't be possible for PS3 but will be for PS4/ CELL 2.0? Hence my above post for various resolutions on the CELL ISA for PS3... ;)
I want to be clear.
I said that I would have liked a CELL based GPU... eventually I'd like to make of somethign like the Visualizer the only IC in the chipset if it were possible, fusing graphics and general purpose processing... no need of a Broadband Engine and a Visualizer... just a sea of shading ALUs and PUs with a pool of fixed function units as co-processor basically.
This is not what we can get IMHO in PlayStation 3: it is something we can look forward for the more distant future.
NV5X is not CELL based and at this point it would not make sense (financially and engineering wise) for the PlayStation 3 GPU to be CELL based as it is based on the NV5X architecture.
Crytal clear...I'll just conclude with why, with all things said, the following still sounds contradictory for PS3, IMHO. ;)
http://www.beyond3d.com/forum/viewtopic.php?p=437597#437597
Panajev2001a
04-Jan-2005, 03:11
Your argument relies on this, basically:
Premise 1: If Hofstee was talking about two-way comms between CPU<=>GPU,
Premise 2 (a conclusion based on the validity of premise 1): then these VS CELL threads should run on the GPU also, no?
---------
Conclusion: This should imply the GPU is CELL based.
This is what you are using to say my argument is contraddictory, isn't it ?
I call it a non sequitur on the passage from P1 to P2 and especially from P1+P2 to the Conclusion.
Your argument might be valid, but it does not seem sound to me as your premises do not appear more known than the conclusion you are trying to proove.
Your argument relies on this, basically:
Premise 1: If Hofstee was talking about two-way comms between CPU<=>GPU,
Premise 2 (a conclusion based on the validity of premise 1): then these VS CELL threads should run on the GPU also, no?
---------
Conclusion: This should imply the GPU is CELL based.
This is what you are using to say my argument is contraddictory, isn't it ?
I call it a non sequitur on the passage from P1 to P2 and especially from P1+P2 to the Conclusion.
Your argument might be valid, but it does not seem sound to me as your premises do not appear more known than the conclusion you are trying to proove.
http://img65.exs.cx/img65/5219/hofstee45ti.jpg
http://www.beyond3d.com/forum/viewtopic.php?t=19108
Hofstee above...
And conversely, from my point 3, if CPU=>GPU, would imply GPU is NOT CELL based...but hey, it's been an interesting discussion either way. ;)
Panajev2001a
04-Jan-2005, 03:42
I do not see what Hofstee is saying is contraddicting what I am saying.
Whether CELL based or not, the GPU in PlayStation 3 will be programmable enough and with enough bandwidth between itself and the CPU to effectively act as a media co-processor: similarly to how it will work on Xbox 2.
The era of 1-way communication between CPUs and GPUs is ending IMHO.
Hofstee talked about things that were possible in the short term and then long term goals: he said that we would go away from textures (he wants an all procedural system [textures are still used in the most advanced off-line CG production today and the trend seems to continue]), do you believe that PlayStation 3 will go to a kind of rendering system that does not leverage textures (even if you limit this to saying "no bitmaps to fake detail" which would apply to normal mapping, detail texturing, displacement mapping , etc...) ?
You have a theory, you have certain premises: I ask you to expand on those premises :).
3. However, if the VS is NOT done on the CELL CPU but on the GPU, then the entire GPU may OR may not be CELL based then, no?
I am not saying VS will not be done at all on the APUs, even if the GPU had VS units.
I am disputing your premise that says "if Vertex Shading is done on the CPU then the GPU uses CELL technology".
Your argument relies on this, basically:
Quote:
Premise 1: If Hofstee was talking about two-way comms between CPU<=>GPU,
Premise 2 (a conclusion based on the validity of premise 1): then these VS CELL threads should run on the GPU also, no?
---------
Conclusion: This should imply the GPU is CELL based.
This is what you are using to say my argument is contraddictory, isn't it ?
I call it a non sequitur on the passage from P1 to P2 and especially from P1+P2 to the Conclusion.
Your argument might be valid, but it does not seem sound to me as your premises do not appear more known than the conclusion you are trying to proove.
I do not see what Hofstee is saying is contraddicting what I am saying.
Whether CELL based or not, the GPU in PlayStation 3 will be programmable enough and with enough bandwidth between itself and the CPU to effectively act as a media co-processor: similarly to how it will work on Xbox 2.
The era of 1-way communication between CPUs and GPUs is ending IMHO.
Hofstee talked about things that were possible in the short term and then long term goals: he said that we would go away from textures (he wants an all procedural system [textures are still used in the most advanced off-line CG production today and the trend seems to continue]), do you believe that PlayStation 3 will go to a kind of rendering system that does not leverage textures (even if you limit this to saying "no bitmaps to fake detail" which would apply to normal mapping, detail texturing, displacement mapping , etc...) ?
I'm not saying texture mapping will be replaced. Like you say, even offline rendering using REYES, still uses texture maps as you cannot procedurally create textures for all surfaces that look realistic. Perhaps they have a majic box/ fractal algorithm that converts real textures to a procedural one, who knows! ;)
And without sounding like a broken record, all I'm suggesting and Hofstee too, is you should be able to try 'new' things...whatever that means...your a developer...think up some stuff! :P
3. However, if the VS is NOT done on the CELL CPU but on the GPU, then the entire GPU may OR may not be CELL based then, no?
I am not saying VS will not be done at all on the APUs, even if the GPU had VS units.
I am disputing your premise that says "if Vertex Shading is done on the CPU then the GPU uses CELL technology".
Your argument relies on this, basically:
Quote:
Premise 1: If Hofstee was talking about two-way comms between CPU<=>GPU,
Premise 2 (a conclusion based on the validity of premise 1): then these VS CELL threads should run on the GPU also, no?
---------
Conclusion: This should imply the GPU is CELL based.
This is what you are using to say my argument is contraddictory, isn't it ?
I call it a non sequitur on the passage from P1 to P2 and especially from P1+P2 to the Conclusion.
Your argument might be valid, but it does not seem sound to me as your premises do not appear more known than the conclusion you are trying to proove.
I thought you gave up and we agreed to disagree! ;)
My premise is more known than my conclusion. Therefore if you accept my premise you should accept my conclusion...
But obviously not as you're now questioning my premise! :P I've supplied Hofstee's slides and video on his presentation...that is more known than my conclusion OR YOUR conclusion...
Anyway my FIRST EVER post, 6 Months ago was the SALC/SALP patent thread... (http://www.beyond3d.com/forum/viewtopic.php?t=12899&postdays=0&postorder=asc&sta rt=0), which really is more pertinent to what Dave is discussing in the other thread. But the relevance is that the patent was a replacement for fixed functionality with programmable functionality without incurring any extra cost in terms of die space/ transistors.
However, my SECOND ever post in the same thread,
but could this potentially be the patent relating to the much vaunted PixelEngine in the PS3's GPU aka Visualizer?
Taking a guess based on what I've read so far, this strikes me as an attempt at a fast and compact programmable replacement for conventionally hardwired circuitry (eg. rasterization, filtering, primitive setup etc.).
If I'm right, implications would be interesting - how about a programmable primitive processor? And jvd and co. should jump for joy about idea of having a configurable pixel feature set
That said, I don't think this is a patent related to entire Visualizer, just the blackbox parts previously thought as fixed hw yeah. (Then again that may have something to do with me refusing to accept any idea of having multiple ISAs for geometry and fragment computation parts... In my dream world we're still using APUs for all shading ops ).
Would it really matter if APUs doing all shading ops (vertex and pixel) be replaced by say, vertex shading via APUs and pixel shading via these SALPs in the PixelEngine? Indeed, the GPU maybe without APUs and replaced by these SALPs entirely...
mmm... I also see accepting one ISA for APU's and SALPs difficult from a Cell philosophy. Maybe the Cell ISA has extensions for graphics?
I'm not clear how software Cells will work on the SALPs. Or is there meant to be a consistent ISA for Cell graphics? I never really got the whole distributed graphics thing with Cell. E.g, an app written for a Cell PDA client work on a Cell PS3 (with a different GPU)? Would the Cell OS use a JIT type compiler to hide this from other types of GPUs on different Cell clients?
http://www.beyond3d.com/forum/viewtopic.php?p=297459#297459
I asked the same questions then...
I also suggested that the CELL CPU's ALUs consisting of S|APUs be complemented by a CELL based GPU who's ALUs consisted of SALC/SALPs. The VS work done on a CELL CPU and the PS work done on the CELL GPU.
That concept that I proposed still has not changed by having different contractors, whether it was Toshiba, SONY in-house, or NVIDIA providing the GPU...It just seemed to make more sense than the Visualizer.
And I can't freakin believe I'm still having the same discussion now! :D ...At least I've remained consistent! :P
Anyway, believe what you want... ;)
Panajev2001a
04-Jan-2005, 16:31
I apologize Jaws, but you called my statement "contraddictory" in that thread and I have a right to ask you to prove my contraddiction and to dispute the proofs that come forward (if it is possible to do so) or admit defeat and conceede that my argument was contraddictng itself (if this indeed happened).
What I see is that you want to have the GPU being CELL based.
I also suggested that the CELL CPU's ALUs consisting of S|APUs be complemented by a CELL based GPU who's ALUs consisted of SALC/SALPs. The VS work done on a CELL CPU and the PS work done on the CELL GPU.
How would that GPU be CELL based ? The SALCs/SALPs now are modified to support the CELL ISA ?
Would it really matter if APUs doing all shading ops (vertex and pixel) be replaced by say, vertex shading via APUs and pixel shading via these SALPs in the PixelEngine? Indeed, the GPU maybe without APUs and replaced by these SALPs entirely...
Then it is hardly CELL based, isn't it ?
In that case you could not run the same threads on the GPU as the SALPs/SALCs are not APUs, do not share their ISA and reconfiguring them to also to on-the-fly decoding of apulets and all that is required to mimic the APUs (DMA access functionality, use of LS, etc...).
The fact that Apulets can migrate over networks and be processed by any CELL based processor is possible because all those devices have units which have the same basic ISA, the SPUs/APUs's ISA, which was made clear IMHO by Suzuoki's patent. If you start adding layers of emulation to enforce compatibility with the APU ISA then you are defeating the point of CELL as stated in that very patent: the loss of efficiency due to complex software interfaces between CPUs with different ISAs that are networked together and are running the same software with say a Virtual Machine like the JVM.
Are you expecting the SALCs/SALPs to be modified to support the SPU/APU's ISA ?
IMHO, the point of SALCs was to connect serial ALUs in programmable pipelines (their structure would be flexible) to perform specific functions.
You could implement the execution logic of the APUs with SALCs: you could micro-code the entire SPU/APU ISA and have the SALCs do the work, but would that be more optimal than have the SPUs/APUs themselves do the work ?
If you used SALCs/SALPs in the GPU they would do a better job as programmable Pixel Engines and programmable TMUs.
I've supplied Hofstee's slides and video on his presentation...that is more known than my conclusion OR YOUR conclusion...
But his slides and his video do not mandate the logical step you make to generate your second premise which is a conclusion of your first premise.
Premise 1: If Hofstee was talking about two-way comms between CPU<=>GPU,
Premise 2 (a conclusion based on the validity of premise 1): then these VS CELL threads should run on the GPU also, no?
Replace the "should" with a "could".
Let's see the two arguments:
1.) Premise 1: Hofstee was talking about two-way comms between CPU<=>GPU (accepted assumption = VS threads are running on the CPU).
Premise 2 (a conclusion based on the validity of premise 1): then these VS CELL threads should run on the GPU also, no?
---------
Conclusion: The GPU is CELL based.
2.) Premise 1: If Hofstee was talking about two-way comms between CPU<=>GPU,
Premise 2 (a conclusion based on the validity of premise 1): then these VS CELL threads could run on the GPU also, no?
---------
Conclusion: This could mean that there is a point making a GPU that is CELL based to go along with a CELL based CPU, that is load balancing.
Your original conclusion was way too strong for its premises: "This should imply the GPU is CELL based" this basically goes to say that a non CELL based GPU makes no sense accroding to what Hofstee said.
Let's go back to your original post:
Premise 1: if you're gonna have vertex shading on the CELL CPU, then they're gonna be VS CELL threads, (aka software Cells).
Premise 1a: Hofstee was talking about two-way comms between CPU<=>GPU
---------------------
Conclusion 1 (or Premise 2 for Conclusion 2): these VS CELL threads run on the GPU also.
Conclusion 2: The GPU is CELL based.
Not only this commits a non sequitur (if you take, as I did while making of your post a series of Premises and Conclusions, the "should run on the GPU" as "the VS threads must be running on the GPU to make any sense" instead of a could run or "it would make some sense for them to run on the GPU too, for load balancing") going from the Premise 1 and Premise 1a to Conclusion 1 (the premises, which I am taking as true) do not necessarily support that conclusion (The GPU could not be CELL based and that conclusion would follwo from Premise 1 and Premise 1a), but you do beg the question going to conclusion 2 as conclusion 1, now your Premise 2, is not IMHO standing on solid ground thus you have a premise which is not necessarily better known than your Conclusion.
You used this logic to prove that my saying that VS could be done on the CPU while PS was done by a non CELL based GPU was contraddictory.
You did not say that we do not know which GPU we will have: CELL based or non CELL based, they both work...
You said:
Panajev2001a wrote:
....
With this said, I still say that IMHO the PlayStation 3 GPU is not CELL based, it does not have the SPUs/APUs.
....
It would seem to me a not bad idea to assign all the Vertex Shading work to the CELL based CPU:
....
To me the above two statements seem to contradict each other.. Confused
Can you please be patient with my stubborn head and go through the steps that show my logical contraddiction ?
Panajev2001a, is this really you....
http://www.hallmarkchannel.com/data/images/PG_USA_PerryMason.jpg
:wink: :lol:
Panajev2001a
04-Jan-2005, 17:14
More like...
http://image.guardian.co.uk/sys-images/Film/Pix/gallery/2003/05/20/lamberter3.jpg
;).
Pana,
I'm not frustrated at you or anyone in particular. It's the medium of communication in a discussion forum that tests one's patience once you've repeated yourself like a parrot and provided links to discussions etc. It's always like walking into a middle of a conversation without knowing what's been said etc. I'm more of a face to face people person! :P
By now you should know my general gist. I'm pretty sure you haven't really read all my posts and links in this thread because of repeating elements in your above post. Time permiting I'll try to answer your above post later...
But in the meantime, please re-read all my earlier posts in this thread and this Byte article from 1994 regarding the TAOS operating system (circa 1992) and the similarities it has to CELL concepts and what I'm trying to suggest for the CELL ISA. And it's not just an in-efficient JVM here...
http://www.byte.com/art/9407/sec6/art1.htm
Also ask yourself why NVIDIA and ATI will have a consistent ISA for WGF2.0 and why NV will have different VS and PS units while ATI will have unified shader units, US units, but BOTH NV and ATI will maintain a consistent ISA? Why would a consistent ISA make sense for PS3's CPU and GPU, bearing in mind they are both likely stream processors?
These are just rhetorical questions but you should get a theme of where I'm coming from that should answer most your questions by the time I reply to your above post! ;)
Panajev2001a
04-Jan-2005, 19:56
Pana,
I'm not frustrated at you or anyone in particular. It's the medium of communication in a discussion forum that tests one's patience once you've repeated yourself like a parrot and provided links to discussions etc. It's always like walking into a middle of a conversation without knowing what's been said etc. I'm more of a face to face people person! :P
By now you should know my general gist. I'm pretty sure you haven't really read all my posts and links in this thread because of repeating elements in your above post. Time permiting I'll try to answer your above post later...
I repeat certain things because all the links, quotes, etc... do not prove that the logical argument you presented is true. If they do, you should help me by guiding me as after-all I was told that my argument was contraddictory and I would like to be shown exactly how I contraddicted myself :).
But in the meantime, please re-read all my earlier posts in this thread and this Byte article from 1994 regarding the TAOS operating system (circa 1992) and the similarities it has to CELL concepts and what I'm trying to suggest for the CELL ISA. And it's not just an in-efficient JVM here...
http://www.byte.com/art/9407/sec6/art1.htm
Ok, I am reading it.
Also ask yourself why NVIDIA and ATI will have a consistent ISA for WGF2.0
Because Microsoft, nVIDIA and ATI have agreed that it makes sense for Vertex Shaders and Pixel Shaders to have the same capabilities.
And following to the next point...
ATI will have unified shader units, US units, but BOTH NV and ATI will maintain a consistent ISA? Why would a consistent ISA make sense for PS3's CPU and GPU, bearing in mind they are both likely stream processors?
Again (connecting this last quote with the orevious one just above), a non sequitur, in this form:
Definition:
Any argument of the following form is invalid:
If A then B
B
Therefore, A
Examples:
(i) If I am in Calgary, then I am in Alberta. I am in Alberta,
thus, I am in Calgary. (Of course, even though the premises
are true, I might be in Edmonton, Alberta.)
This is what you just said as far as I can understand:
Premise 1: VS and PS in their ISA will be unified come WGF 2.0.
Premise 2: A common ISA for VS and PS makes sense.
Premise 3: In PlayStation 3 VS work will be done by the SPU's/APU's on the CPU and PS work will be done by the GPU (quite probably).
--------------------------
Conclusion: In PlayStation 3, both CPU and GPU share the same ISA (CELL ISA... they can both run the same Apulets/Software cells directly).
One of the problems with this argument is that the APUs do not implement WGF 2.0's Vertex Shader ISA, they were designed to do more than process vertices. Saying that the SPUs/APUs run Vertex Shaders and thus implement the Vertex Shader ISA directly is a jump that should not be made. Take your PC... You can run games which use Vertex Shading even if you do not support/disable Vertex Shading on the GPU as programs coded for the Vertex Shader 1.0/2.0/3.0 ISA can be converted to SSE assembler code.
Also, it would make sense for a CPU and a GPU to be both CELL based because over-all the system's performance is increased thanks to the load-balancing work that can be done across all the SPUs/APUs in the system.
You say: if you can implement the Vertex Shader ISA (we are still very unclear if and how SPUs/APUs would be able to run Vertex Shader code in which Texture Look-ups are often made: did they add instructions that dela with Texture Sampling in the general purpose SPUs/APUs) with SPUs/APUs then you can also implement the Pixel Shader ISA (which is eaven heavier in Texture look-ups) and you can modify the hardware implementation of the SPU/APU used in the GPU to bear higher latencies (more common with Pixel Shaders' workload) or not.
I say, yes I agree, which is what I see myself for the future of CELL based systems. I disagree with the assumption that the GPU must be CELL based if Vertex Shading is run on the CPU by the SPUs/APUs: it is not the only possibility thus making me wonder how possibly could all of this make my argument (VS done on the CELL based CPU and PS done on the non CELL based GPU) contraddictory.
This does not mean that it is the only way to go: the premises you posted can lead to other conclusions. How can your argument prove the contraddictory nature of my argument ?
Panajev2001a
04-Jan-2005, 20:10
The Taos kernel translates VPcode into the native machine code of each real processor immediately before running it--there is little or no run-time penalty
Every processor in this network runs a copy of the Taos kernel and the translator from VPcode to its own native code. Whenever Taos creates a new object, it allocates the object to a processor and then starts a process to execute the object.
http://www.byte.com/art/9407/sec6/art1.htm
Thsi does not go well along with this:
The computers and computing devices of current computer networks, e.g., local area networks (LANs) used in office networks and global networks such as the Internet, were designed principally for stand-alone computing. The sharing of data and application programs ("applications") over a computer network was not a principal design goal of these computers and computing devices. These computers and computing devices also typically were designed using a wide assortment of different processors made by a variety of different manufacturers, e.g., Motorola, Intel, Texas Instruments, Sony and others. Each of these processors has its own particular instruction set and instruction set architecture (ISA), i.e., its own particular set of assembly language instructions and structure for the principal computational units and memory units for performing these instructions. A programmer is required to understand, therefore, each processor's instruction set and ISA to write applications for these processors. This heterogeneous combination of computers and computing devices on today's computer networks complicates the processing and sharing of data and applications. Multiple versions of the same application often are required, moreover, to accommodate this heterogeneous environment.
The Java model attempts to solve this problem. This model employs a small application ("applet") complying with a strict security protocol. Applets are sent from a server computer over the network to be run by a client computer ("client"). To avoid having to send different versions of the same applet to clients employing different ISAs, all Java applets are run on a client's Java virtual machine. The Java virtual machine is software emulating a computer having a Java ISA and Java instruction set. This software, however, runs on the client's ISA and the client's instruction set. A version of the Java virtual machine is provided for each different ISA and instruction set of the clients. A multiplicity of different versions of each applet, therefore, is not required. Each client downloads only the correct Java virtual machine for its particular ISA and instruction set to run all Java applets.
Although providing a solution to the problem of having to write different versions of an application for each different ISA and instruction set, the Java processing model requires an additional layer of software on the client's computer. This additional layer of software significantly degrades a processor's processing speed. This decrease in speed is particularly significant for real-time, multimedia applications. A downloaded Java applet also may contain viruses, processing malfunctions, etc. These viruses and malfunctions can corrupt a client's database and cause other damage. Although a security protocol employed in the Java model attempts to overcome this problem by implementing a software "sandbox," i.e., a space in the client's memory beyond which the Java applet cannot write data, this software-driven security model is often insecure in its implementation and requires even more processing.
In another aspect, the present invention provides a new programming model for transmitting data and applications over a network and for processing data and applications among the network's members. This programming model employs a software cell transmitted over the network for processing by any of the network's members. Each software cell has the same structure and can contain both applications and data. As a result of the high speed processing and transmission speed provided by the modular computer architecture, these cells can be rapidly processed. The code for the applications preferably is based upon the same common instruction set and ISA. Each software cell preferably contains a global identification (global ID) and information describing the amount of computing resources required for the cell's processing.
Since all computing resources have the same basic structure and employ the same ISA, the particular resource performing this processing can be located anywhere on the network and dynamically assigned.
http://makeashorterlink.com/?G16C2182A
TAOS still preaches run-time code-conversion (preferably optimization too) which is not what the CELL motto pushes: the idea of the ubiquitous CELL ISA was that a networked group of workstations and other devices that were CELL based required no code translation and/or complex software layers to allow the processing to be distributed over the netowrk and/or easy data sharing/communication between each device.
Pana,
READ MY POSTS!!! :D
Stop posting and replying to STATED RHETORICAL QUESTIONS before I've had time to answer your FIRST POST! :P
On a quick note on the above TAOS model. I said it was similar in concept to CELL. If you think about it, and stop jumping to prove differences, and take the best bits of JAVA, TAOS and the way I see it mapped to CELL, is that 'the-runtime-conversion' program that you've mentioned, would be a standard PROGRAM of EVERY 'software CELL/Apulet'. Which is different to JAVA... ;)
Panajev2001a
04-Jan-2005, 21:54
Pana,
READ MY POSTS!!! :D
Stop posting and replying to STATED RHETORICAL QUESTIONS before I've had time to answer your FIRST POST! :P
On a quick note on the above TAOS model. I said it was similar in concept to CELL. If you think about it, and stop jumping to prove differences, and take the best bits of JAVA, TAOS and the way I see it mapped to CELL, is that 'the-runtime-conversion' program that you've mentioned, would be a standard PROGRAM of EVERY 'software CELL/Apulet'. Which is different to JAVA... ;)
Can you compose this picture for me, please ?
I do not think that such a run-time conversion program should be part of any Apulet: it is time wasted that should not be needed in the Apulet as all APUs share the same basic ISA (extended or not).
cthellis42
04-Jan-2005, 21:56
...I think the mauve PS3 will have the most RAM. :P
Sorry. Had to break things up a bit. ;)
:?: Why would dev generate geometry on the fly insted offline (+++ qualitity) and use all those power to other things ( AI...) :?:
Inane_Dork
05-Jan-2005, 02:26
:?: Why would dev generate geometry on the fly insted offline (+++ qualitity) and use all those power to other things ( AI...) :?:You can save memory space and bandwidth by generating it on-the-fly. It might even be faster to render than to load it all in from memory and animate it, but I have no idea if that's correct or not.
Keep in mind we're not talking about generating every polygon every frame. Just some things (the patent gave the examples of leaves blowing in trees).
And it sounds like the CPU(s) are getting powerful enough that they are intended to do this. So if you saved the processing for AI (for instance), you'd have more than you need, likely.
...
I repeat certain things because all the links, quotes, etc... do not prove that the logical argument you presented is true. If they do, you should help me by guiding me as after-all I was told that my argument was contraddictory and I would like to be shown exactly how I contraddicted myself :) .
Okay...
I'll just re-cap here,
I am disputing your premise that says "if Vertex Shading is done on the CPU then the GPU uses CELL technology".
Your argument relies on this, basically:
Quote:
Premise 1: If Hofstee was talking about two-way comms between CPU<=>GPU,
Premise 2 (a conclusion based on the validity of premise 1): then these VS CELL threads should run on the GPU also, no?
---------
Conclusion: This should imply the GPU is CELL based.
This is what you are using to say my argument is contraddictory, isn't it ?
I call it a non sequitur on the passage from P1 to P2 and especially from P1+P2 to the Conclusion.
Your argument might be valid, but it does not seem sound to me as your premises do not appear more known than the conclusion you are trying to proove.
I thought you gave up and we agreed to disagree! ;)
My premise is more known than my conclusion. Therefore if you accept my premise you should accept my conclusion...
But obviously not as you're now questioning my premise! :P I've supplied Hofstee's slides and video on his presentation...that is more known than my conclusion OR YOUR conclusion...
Pana doesn't accept the logical jump from premise 1 to premise 2 as stated below,
...
I've supplied Hofstee's slides and video on his presentation...that is more known than my conclusion OR YOUR conclusion...
But his slides and his video do not mandate the logical step you make to generate your second premise which is a conclusion of your first premise.
So lets state premise 2 that Pana doesn't believe makes a logical step from premise 1.
Premise 2 (a conclusion based on the validity of premise 1): then these VS CELL threads should run on the GPU also, no?
HOWEVER,
...
Also, it would make sense for a CPU and a GPU to be both CELL based because over-all the system's performance is increased thanks to the load-balancing work that can be done across all the SPUs/APUs in the system.
...
http://www.beyond3d.com/forum/viewtopic.php?p=440106#440106
So Pana now believes the above makes sense? Hmmm...
This would mean VS CELL threads would run on GPU for premise 2,
Premise 2 (a conclusion based on the validity of premise 1): then these VS CELL threads should run on the GPU also, no?
Since premise 2 is a conclusion based on premise 1, then Pana accepts premise 1.
Premise 1 is more known than my conclusion, i.e.,
Because...
I've supplied Hofstee's slides and video on his presentation...that is more known than my conclusion OR YOUR conclusion...
My conclusion is,
This should imply the GPU is CELL based.
And imply does NOT exclude other conclusions.
Therefore...
Therefore if you accept my premise you should accept my conclusion...
Pana accepts my conclusion...
QED and Own3d! (http://www.short-media.com/forum/attachment.php?attachmentid=3701&stc=1)
;)
PS. I'll answer the other questions time permitting...
Panajev2001a
06-Jan-2005, 12:49
QED my arse ;).
"Vertex Shading is done on the CPU, but the GPU does not use CELL technology".
You said THIS was contraddictory.
You said this statement contraddicted itself.
You merely say that it makes sense for the GPU to use CELL technology (you spent half a post, not even looking at my latest recap of the argument which was more refined... ;)) which we agreed.
You forget that it also makes sense for the GPU not to use CELL technology (i.e. there are reasons for it not to use CELL technology).
I was trying to show you that you can only say that the GPU being CELL based is one of the possible reasnoable solutions, not the only one.
....
With this said, I still say that IMHO the PlayStation 3 GPU is not CELL based, it does not have the SPUs/APUs.
....
It would seem to me a not bad idea to assign all the Vertex Shading work to the CELL based CPU[...]
You said these two statements contraddicted each other and the sure way this would have been true would be if you take the IMPLY as "it is the necessary and only valid conclusion that..." or something along those lines.
If you look at that I attacked your argument based on that interpretation of your argument: not a problem, because as soon as you admit that it was not the point you wanted or could sustain then the shadow of contraddiction would disappear from those two statements of mine.
You trying to P0WN me, but you just 0WN3D your own argument which was trying to proove that my two statements there showed a clear contraddiction.
It seems like the trap worked ;).
Pana,
You are confused and stubborn.
I saw your trap a mile away... (http://www.short-media.com/forum/attachment.php?attachmentid=377&stc=1)
Because my premise and conclusion has ALWAYS remained consistent throughtout.
My conclusion is,
This should imply the GPU is CELL based.
And imply does NOT exclude other conclusions.
Therefore...
Therefore if you accept my premise you should accept my conclusion...
Pana accepts my conclusion...
QED and Own3d! (http://www.short-media.com/forum/attachment.php?attachmentid=3701&stc=1)
;)
My conclusion ALWAYS accepted other conclusions. It was suggestive and NEVER definitive. That's why I asked you to RE-READ MY POSTS...
Anyway, I'll re-iterate,
Believe what you want...It's been fun...I'm leaving this thread...bye... ;)
Shaddap my ears are bleeding :cry:
A suggestive conclusion. Kind of a oxymoron, don’t you think?
A suggestive conclusion. Kind of a oxymoron, don’t you think?
BINGO!
http://www.beyond3d.com/forum/viewtopic.php?p=437368&highlight=#437368
I've tried to conclude and leave this thread many times if you re-read earlier posts...apologies to anyone, it was always meant to be fun! :D
In a month, ISSCC 2005 comes and we'll know the (1st-gen) Cell details . In 90 days, in March, there'll be the PS3 unveiling event. Patience, dudes :D
Panajev2001a
06-Jan-2005, 19:14
I am stubborn, but I am not confused.
Your argument did fail to prove how my two sentences CONTRADDICTED each other.
The fact that you accepted that:
My conclusion ALWAYS accepted other conclusions. It was suggestive and NEVER definitive.
If it was not definitive, but accepted other conclusions youa re saying this:
"The GPU might not be CELL based."
You accepted this conclusion as one of the possible outcomes of your argument or of one or more of the premises you used.
One of those premises was Vertex Shading would be done using the APUs on the CPU.
Do you see you are saying the same thing that you deemed to be logically contraddictory ?
My argument was:
....
Statement 1: With this said, I still say that IMHO the PlayStation 3 GPU is not CELL based, it does not have the SPUs/APUs.
....
Statement 2: It would seem to me a not bad idea to assign all the Vertex Shading work to the CELL based CPU[...]
For these two statements to be contraddictory it would mean that if Statement 1 is true then Statement 2 cannot be true at the same time.
Your arguments accepts as plausible premise that "Vertex Shading is done by the APUs in the CPU".
Now if saying that "the GPU is CELL based" is not the ONLY conclusion, but one of the possible conclusions of an argument which was thought to analyze the nature of the GPU in PlayStation 3, then we might start to accept that another possible and plausible conclusion is that the GPU might not be CELL based.
If we accept that as valid conclusion then my two statements are not contraddicting each other.
3 : to contain potentially
4 : to express indirectly <his silence implied consent>
This is from the Dictionary definition of "imply"... when I was presenting the more "exclusive" version of your argument I used it in the meaning presented as #4 while you meant it as #3.
When Immanul Kant and his followers say "ought implies can" they mean that a moral duty (i.e. something you "ought to do/respect") makes sense only if the person has the capability to perform the duty, if it is possible to perform the duty. It would not be fair to mandate you to do something if you could not possibly perform the action.
So I was justified in not accepting the premises in that scenario which I admit was a bit warped considering what you wanted to say.
My point is that I do not believe you can say my two Statements contraddict each other, do not make sense together, etc... : if you rpooved that argument in my "exclusive conclusion" and my use of "implies", etc... then you would have shown how my two Statememnts contraddict each other as...
"Statement 2: It would seem to me a not bad idea to assign all the Vertex Shading work to the CELL based CPU"
could not lead to say anything different from "the GPU is CELL based" and therefore it would make a conclusion like "the GPU is not CELL based" as impossible and illogical.
Akumajou
07-Jan-2005, 07:03
So, if I understand this well, Sony had the GS completed, with everything mayor locked down, in 1998? And sat on it for more than a year?? If that´s the truth then:
1. Wow, the GS really is one heck of a great graphics chip if it was really completed back then.
The Matrox G200, the Nvidia Riva TNT/2, the Videologic PowerVR2 and ATI Rage Pro 128 GPUs were all impressive architectures that could render some trully amazing PS2 level graphics in a videogame if the game developers tools and code are written on an assembly level even thought they were all physically produced in 1998 on whatever process it was back then (.350nm?)
Unless I'm wrong, Nvidia released the Riva TNT on a .350nm process and then re-released the same GPU as the TNT2 on a .250nm raising the fill rate performance and if Sony would have chosen that chip instead of the GS we can argue that the graphics would have pretty much been the same since the PS2 does most of its software tricks dependant on its CPU (a possible reason why GS was finished first)
Fast foward to 1999 and Nvidia released the GeForce256 Nv10 GPU on a .220nm and in 2000 they released the GF2 Ultra Nv15 on 180nm then in 2002 they re-re-released it as the GF4 MX Nv18 on a .150nm process that is still basically a Nv10 just shrunk down and stably overclocked from 120Mhz all the way to the 300Mhz on the GF4 MX 460.
Now imagine that a 1999 GPU core having breakneck speed thanks to a company that allowed this to go on.
2. Wasn´t this kind of dumb?? To sit on a chip, no matter how good it is, for more than a year is not a very smart thing to do, IMO.
Not so, at least in the way Sony seems to think since they also neglected to add 4MBs to the already very limited 4MB of video ram in the PS2 to at least match up to Dreamcast's 8MB of video ram.
Besides there are reasons for this:
1: At the time 1998, 1999 there was no real XBox rumors based on fact.
2: The GS is very much like the Matrox G400, Nvidia Riva TNT2, etc in that they perform most of their functions dependant on a fast CPU therefore the wait for 1999 or 2000 EE production on a smaller die process for more speed in a custom chip 300Mhz. Compared to the way XBox works, the GPU handles most of the graphics tasks while the CPU is free to do things like A.I. something that PS2 games are not known for having in a good way.
3: the GS benefited in a RivaTNT2 or GeForce 2 way of getting a die shrink for more speed and therefore higher fill rate and polygon count.
4: Sony was going to destroy SEGA's credibility by buying off EA and having Square and Enix produce software only for the PS2 leaving the DC high and dry without those companie's IP and with a major head start on what they thought was going to be their competition in Nintendo.
5: Sony was basically promoting the use of assembly level software development kits so that game developers would be able to get very close to the hardware since Sony would preffer that a game developer devote all their time making games on their console leaving any port to another console a major software development effort farmed to either another team or outsource to another team if the 3rd party could afford it.
Basically Sony learned something that SEGA seemed to forget during the last PSX vs Saturn days in that Saturn games were really much more technologically impressive than anything on PSX if they were programed in assembly level code and the reason SEGA did that with some of their games was because they were not going to develop those games on a competing console platform.
SEGA's Saturn turned out some really technologically impressive 3d games during 1997/98 with titles like Burning Rangers Panzer Dragoon Saga, the Sonic World part of the Sonic compilation and even Nights into Dreams. But what is really more impressive is what AM2 was working on that very little got to see, a prototype Saturn game that requires the 4MB expansion cart, Project Berkerly aka Shenmue:
http://www.sega-saturn.com/news/112102.htm
If you watch that movie and you ever got to play the first version of VF4 on PS2 and compare it to the 3rd or fourth? version as VF4evo you will see that AM2 is really crunching the PS2 to nearly replicate the Naomi 2 arcade hardware on the PS2.
I´m not very knowledgable on this topic, but didn´t Sony had enough time to further upgrade the GS?? How difficult is it to do that??
I can assume that Sony may have had assembly level SDKs being written so that software developers would be able to start putting out and get a head start in PS2 game development and part of the reason as to why Metal Gear Solid 2 SoL pales in comparison to MGS 3 Snake Eater in the way the PS2 hardware is being used to display the much more complex graphics the same goes for GTAIII to GTASA.
Having a head start in assembly level SDKs and plenty of years of development time (PS2 is sales king world wide) has allowed and can allow game developers to produce amazing looking games that make people almost think that XBox games are technologically weak in comparison to a crippled 1998 technology like PS2.
As for the topic poster, I feel that I agree with you.
I see the same thing that happened to XBox happening with PS3 in that the technology will definetly be bleeding edge when released in comparison to PS2 mainly because:
1: Unlike PS2, PS3 will not be first DVD console anymore, it will be Blue Ray though and as Sony has proven, their pockets are deeper than SEGA's.
2: Assembly level development tools will be worked on however I feel that they will do the same thing they did with PSX and release a general C++ similar level SDK unless Nvidia is providing a custom version of CG to Sony. Expect alot of games to probably have choppy glitches and even some slight slowdown in the first and second wave/gen games released, followed by progressively improving games that even by the 3rd and 4th generation (unlike PS2) will not show the true potential of the console considering how games take an average of 2 years to make and even Halo 2 illustrates this even though its the developer's second effort.
3: I could be wrong and Sony and Nvidia could have a more powerfull assembly SDK ready for the 2nd gen of games to give full access to every transistor.
Now I personally am not a Sony fan, I am more of a SEGA fan but my taste in anime franchise based videogames that I personally find appealing like Mobile Suit Gundam, Dragon Ball Z, Transformers and others makes me already swing for PS3 over XBox Next since it seems like Microsoft has no idea that people watch anime and dig franchise related videogame products, it works for Star Wars but those anime games have and are being written specifically around PS2 hardware with a slim chance of a version or port on GameCube and no chance in hell for XBox even though the console's CPU is fully capable of emulating the PS2 hardware.
london-boy
07-Jan-2005, 10:55
The Matrox G200, the Nvidia Riva TNT/2, the Videologic PowerVR2 and ATI Rage Pro 128 GPUs were all impressive architectures that could render some trully amazing PS2 level graphics in a videogame if the game developers tools and code are written on an assembly level even thought they were all physically produced in 1998 on whatever process it was back then (.350nm?)
You're delusional. Sorry that's all anyone can say after that.
Not so, at least in the way Sony seems to think since they also neglected to add 4MBs to the already very limited 4MB of video ram in the PS2 to at least match up to Dreamcast's 8MB of video ram.
They work differently, i thought this argument was closed 3 years ago. You can only store 8MB of textures on DC, while on PS2, once u've finished with your game code, you have the rest of those 32MB to store textures.
3: the GS benefited in a RivaTNT2 or GeForce 2 way of getting a die shrink for more speed and therefore higher fill rate and polygon count.
The GS of today is exactly the same speed as it was in 1999. The die shrink was for costs reasons, heat dissipation and power requirements. It is also the reason why we now have a much smaller PS2 than it was at launch. No speed gains.
4: Sony was going to destroy SEGA's credibility by buying off EA and having Square and Enix produce software only for the PS2 leaving the DC high and dry without those companie's IP and with a major head start on what they thought was going to be their competition in Nintendo.
What does that have to do with the main subject?
5: Sony was basically promoting the use of assembly level software development kits so that game developers would be able to get very close to the hardware since Sony would preffer that a game developer devote all their time making games on their console leaving any port to another console a major software development effort farmed to either another team or outsource to another team if the 3rd party could afford it.
Or, Developers wanted the manufacturers to let them get closer to the metal after PS1, and Sony did just that. Obviously they didn't handle it the best way they could, but that's another matter.
Basically Sony learned something that SEGA seemed to forget during the last PSX vs Saturn days in that Saturn games were really much more technologically impressive than anything on PSX if they were programed in assembly level code and the reason SEGA did that with some of their games was because they were not going to develop those games on a competing console platform.
Meh... sounds to me like someone still has a chip on their shoulders...
SEGA's Saturn turned out some really technologically impressive 3d games during 1997/98 with titles like Burning Rangers Panzer Dragoon Saga, the Sonic World part of the Sonic compilation and even Nights into Dreams. But what is really more impressive is what AM2 was working on that very little got to see, a prototype Saturn game that requires the 4MB expansion cart, Project Berkerly aka Shenmue:
http://www.sega-saturn.com/news/112102.htm
NOT AGAIN!!!
If you watch that movie and you ever got to play the first version of VF4 on PS2 and compare it to the 3rd or fourth? version as VF4evo you will see that AM2 is really crunching the PS2 to nearly replicate the Naomi 2 arcade hardware on the PS2.
Sega was never that good at squeezing performance out of PS2. So i'm not sure why one should use them to prove PS2 or any other console's power. Konami and Sony's first parties teams are the best PS2 developers. Nintendo and their first party teams are the best devs for GC and Tecmo and some others have proved to be the best for Xbox.
Sega has not released anything exceptionally advanced for years and years. Panzer Dragoon Orta might be very pretty to look at, but it's hardly advanced, being an on-rail shooter.
I can assume that Sony may have had assembly level SDKs being written so that software developers would be able to start putting out and get a head start in PS2 game development and part of the reason as to why Metal Gear Solid 2 SoL pales in comparison to MGS 3 Snake Eater in the way the PS2 hardware is being used to display the much more complex graphics the same goes for GTAIII to GTASA.
Now Now... I wouldn't say MGS2 pales in comparison to MGS3 and the same for GTA3/GTASA... The later released ones obviously look better, but it's obvious that a game that comes out 2-3 years after another one looks better.
Having a head start in assembly level SDKs and plenty of years of development time (PS2 is sales king world wide) has allowed and can allow game developers to produce amazing looking games that make people almost think that XBox games are technologically weak in comparison to a crippled 1998 technology like PS2.
That's one way to put it i guess... :?
1: Unlike PS2, PS3 will not be first DVD console anymore, it will be Blue Ray though and as Sony has proven, their pockets are deeper than SEGA's.
Didn't really get that...
2: Assembly level development tools will be worked on however I feel that they will do the same thing they did with PSX and release a general C++ similar level SDK unless Nvidia is providing a custom version of CG to Sony. Expect alot of games to probably have choppy glitches and even some slight slowdown in the first and second wave/gen games released, followed by progressively improving games that even by the 3rd and 4th generation (unlike PS2) will not show the true potential of the console considering how games take an average of 2 years to make and even Halo 2 illustrates this even though its the developer's second effort.
Again, stating the obvious... Later gen games always run better than earlier ones... Also, it's a given this time around, with NVIDIA on board, that there will be plenty of libraries and documentation available to the developers straight away, at least for the GPU. It's the CPU part, and how Sony will inform devs on how to work with it that worries be a bit more... :wink:
3: I could be wrong and Sony and Nvidia could have a more powerfull assembly SDK ready for the 2nd gen of games to give full access to every transistor.
Sony, MS and N always release "new" SDKs or at least tools to help devs squeeze out performance out of their platforms. The Performance Analyser for PS2 helped a lot.
Now I personally am not a Sony fan, I am more of a SEGA fan
Who'd have guessed?!! :twisted:
...being written specifically around PS2 hardware with a slim chance of a version or port on GameCube and no chance in hell for XBox even though the console's CPU is fully capable of emulating the PS2 hardware.
Excuse me????? Now THAT i'd love to see in action. :lol:
Akumajou
07-Jan-2005, 19:05
london-boy wrote:
You're delusional. Sorry that's all anyone can say after that.
As the topic poster mentioned, the GS is 1998 technology, therefore comparable to graphic chips released in that year as far as capabilities and the fact that the GS cannot be compared to 1999 PC-GPUs because those chips introduced things like hardware transform & lighting (S3 and Nvidia), enviroment mapped bump mapping (Matrox G400), texture compression (S3), etc.
They work differently, i thought this argument was closed 3 years ago. You can only store 8MB of textures on DC, while on PS2, once u've finished with your game code, you have the rest of those 32MB to store textures.
Duh, Sony started shifting to streaming games of the media as opposed to just loading a level into memory since the PSX days, the only reason that became a standard for the console industry is because Sony was the name brand sales/marketing leader, NOT technological leader.
So what are you, a believer in that PS2 is so powerfull because you claim that the video ram argument was closed 3 years ago? you are wrong and I am part of the few who will say it.
Like I said, Sony's PS2 is NOT a technologically superior product compared to SEGA's Dreamcast, they are almost on similar power levels with the PS2 only being able to push more polygons.
All the 3d effects on PS2 are handled in software, if AM2 could take a 4MB cart and make a prototype Shenmue (however much you may hate that game) run on SEGA Saturn (1994 tech) using their own custom assembly level SDKs and their obvious experience then its pretty damn obvious that even the worst developer in the world, with plenty of experience using Sony's custom assembly level SDKs will be able to put out amazing 3d effects on either PSX or PS2.
Maybe you never knew that SEGA Genesis was not able to do the SNES "mode 7" effect, but a couple of years after the SNES was released there were Genesis games doing the same tricks like the infamous "mode 7" thanks to developer experience in using assembly level tools.
The GS of today is exactly the same speed as it was in 1999. The die shrink was for costs reasons, heat dissipation and power requirements. It is also the reason why we now have a much smaller PS2 than it was at launch. No speed gains.
Gee I wonder why the PS2 had to have heat sinks and a big huge fan suck all the hot air out of the console. (sarcasm)
First of all it was 1998, second we can assume that although the die shrink provided all the heat dissipation and lower voltage requirements that Sony desired to make sure that the GPU would be as fast as it could complement the EE CPU, again its worth mentioning that the GS is not a GPU like the Nvidia Nv10 and ATI R100 were with their on core geometry processors and therefore such a graphics chip required a fast CPU just like the Matrox G200, G400, Nvidia Riva TNT and even 3DFX Voodoo 2/3 to reach a higher performance.
That also explains how A.I. has not evolved much on PS2 because the EE is all tied up.
What does that have to do with the main subject?
It has to do with the "subject" because if you lived back in 1998 and 1999 prior to the 2000 release of the PS2 you would have been bombarded with all the hype Sony was spitting about how Square would be able to display games pushing 80 million polygons, you would have read interviews of EA and Sony dev teams dismissing the DC as a "failiure" months prior to the 9-9-99 DC launch, all creating a mindshare atmosphere controlled by Sony marketing evangelists.
Basically, having those "evangelists" and the suckers that believed it allows companies like Sony to release 1998 technology in the year 2000 and not get critisized or punished for it.
Or, Developers wanted the manufacturers to let them get closer to the metal after PS1, and Sony did just that. Obviously they didn't handle it the best way they could, but that's another matter.
I think my paragraph made more sense but...
Polyphony Digital in 96 (or was that 97) was already getting closer to the "PS1 metal" with both Gran Tourismo (and later GT2) and thats years before the Dreamcast and using BTW the same software based streaming of videogame levels on PSX hardware years before they would again use it on PS2, nothing new there.
Its obvious that Sony realized it was better to have every developer using "closer to the metal" assembly level tools so that way a dev could devote all their resources on one platform making a port become a financial risk unless the game was using general C++ level SDKs. That ensure all or the majority of games being made on a "hit console" even though the technological hardware being used is the equivalent of 1998 technology.
Meh... sounds to me like someone still has a chip on their shoulders...
And what the hell is wrong with that? rich boy Sony came, saw, stole and conquered the competition. Anyone can say anything about Sony, Sega or Nintendo just like the bunch of idiots who claim AMD is god and Intel/Microsoft is the devil.
Or you could see Sony as Walmart taking market share away from mom & pop game shops like SEGA & Nintendo eventually leading them to a financial corner. I never saw Nintendo making tv sets, SEGA selling DVD players, radios, cassette players, cd-players, etc.
Or maybe you just cannot imagine a world without Sony's involvement in the videogame field.
NOT AGAIN!!!
I guess a prototype Shenmue running on a Sega Saturn that was hyped by the competition as being 3d-deficient must be very offensive to you.
Sega was never that good at squeezing performance out of PS2. So i'm not sure why one should use them to prove PS2 or any other console's power. Konami and Sony's first parties teams are the best PS2 developers. Nintendo and their first party teams are the best devs for GC and Tecmo and some others have proved to be the best for Xbox.
Can you please go out and find, rent or borrow (or dl if you mod) VF4 and compare it to VF4Evo, then compare it to the VF4 and VF4Evo arcade machines, you just might notice how VF4-PS2 looks like garbage compared to the closer to the arcade VF4Evo.
I picked SEGA because they are NOT ass-kissing Konami or Sony first parties mainly because Sega dev teams had to start learning Sony SDKs much later than any other game developer so it makes sense that their first games as a third party would never look as good as a Konami effort, thats a big change from how Sega dev teams had SDKs months before any third party would or the same could be said for Sony first devs and second devs because its obvious Sony would want Square to have a FFXXX ready for console launch day if they could get away with it.
Now Now... I wouldn't say MGS2 pales in comparison to MGS3 and the same for GTA3/GTASA... The later released ones obviously look better, but it's obvious that a game that comes out 2-3 years after another one looks better.
Now it seems like you are just disagreeing or re-cycling the same mesage I originally posted just for the sake of finding something to do.
Gee I wonder why a game that comes out 2-3 years is always able to push more 3d effects+performance out of a console, hey maybe thats what makes consoles so great over PCs.
That's one way to put it i guess...
Maybe I could have said that the Sony driver has to learn how to drive a 6 speed manual transmision touring car and they figured that drawing the race track by hand in their backyard allow them to remember all the curves come race day even though the touring car was a tricked/pimped out Yugo racing against SEGA Supras and Nintendo RX-7s with drivers who had only heard of the track and assumed that with their experience they would win or complete the race.
Didn't really get that...
Do you remember the price of DVD players in the US back in 1999 and months prior to PS2's 2000 launch?
Basically the only new technologies being introduced in consoles was the DVD drive and enhanced backward (but flawed) compatability being built in and the sole reason alot of people rushed out to get PS2s, to watch movies and play old PSX games enhanced since the majority of first gen PS2 games were horribly below the then Dreamcast standard, something that should never have been.
Again, stating the obvious... Later gen games always run better than earlier ones... Also, it's a given this time around, with NVIDIA on board, that there will be plenty of libraries and documentation available to the developers straight away, at least for the GPU. It's the CPU part, and how Sony will inform devs on how to work with it that worries be a bit more...
So what?, the Topic poster asked for opinions on how the PS3 would compare to PS2 in its time of introduction, I gave my opinion, deal with it.
Sony, MS and N always release "new" SDKs or at least tools to help devs squeeze out performance out of their platforms. The Performance Analyser for PS2 helped a lot.
Gee I seem to remember that Genesis & SNES story about "mode 7", how about I tell you how both PSX and Saturn dev kits were buggy but because the Saturn used dual CPUs, it was critisized more for being more complex to develop a SDK for.
I have been very well informed on how SEGA, Sony, MS and N always in their historical times have always had to release updated SDKs, that is a standard that was beaten to death on SNK's Neo Geo 2d arcade hardware if you compare Fatal Fury to Mark of the Wolves when it comes to fluid 2d animation.
Who'd have guessed?!!
And I am to be punished for even remotely liking SEGA??
You missed the whole point of my post then if you just want to narrowly pre-judge me into a SEGA fanboy box.
Excuse me????? Now THAT i'd love to see in action.
If you are familiar with emulators you would know that Dreamcast was capable of running an emulator that would not only raise the resolution of certain PSX hit games, but breathe new life into them.
There are PSX emulators that work on PCs and the minimum requirement is an Intel Pentium II or I with MMX extensions for visual parity with PSX image/sound quality. Basically the emulator uses that old Intel CPU running at 200Mhz and the MMX extensions to emulate a 1994 console that had a 33Mhz cpu.
PSX emulators took a long time to develop, the console was released in 94 in Japan and the first actual working emulators I remember were released in either 1999 or 2000 with Bleem being one of the first that tried to go commercial (and they got sued for it) If you can count, it took hobbyists (the people who make emus) over five years to properly make accurate emulation and later enhanced emulation with the help of powerfull graphics cards and APIs like Direct X and OpenGL.
Now it did not help that Sony offered to hire (as in give jobs) to those hobbyists so as to prevent their fears from becoming a reality in a hobbyist made PS2 emulator being made much sooner thanks to the hobbyist's experience in making previous emulators.
Taking into account that the PS2's EE runs at 300Mhz and the GS runs at 150Mhz, it makes sense to say and assume that a properly written PS2 emulator could be made with a PIII as a minimum requirement for accurate emulation and a Dx+OGL graphics card to provide texture filtering enhancements.
As to when that will happen, hey its now 2005 and people are still working on it but since they do not get paid it depends on how much time they dedicate to it.
If we were to get a company to do it, like the company that made Bleem, we would have had a PS2 game emulator running on XBox 2 years ago.
darkblu
07-Jan-2005, 22:32
Taking into account that the PS2's EE runs at 300Mhz and the GS runs at 150Mhz, it makes sense to say and assume that a properly written PS2 emulator could be made with a PIII as a minimum requirement for accurate emulation and a Dx+OGL graphics card to provide texture filtering enhancements.
fastest p3 produced was the 1.4GHz tualatin. but that's hardly a commonly-available cpu, the consumer's p3 would be ~1ghz. so you're saying that a 1ghz p3 (sse) + a dx7 card can emulate the ps2, did i get you right?
The XCPU is actually weaker then the EE at least in some respect. And yet you're saying that it's fast enough to emulate it? Are you serious?? :shock:
Today's Windows PC can play a PS2 game on an emulator (http://www.pcsx2.net/) at whopping 0.99 fps :lol: It'll require a 15GHz or more CPU to run it at 60 fps.
http://www.pcsx2.net/cutenews/images/upimages/Sevensamurai_1.jpg
akira888
08-Jan-2005, 00:24
The reason there's no PS2 emulator out that runs commercial games is not for lack of trying or Sony "buying emucoders" but instead it is due the existence of real hardware issues: 128 bit GPRs, two independent floating point vector coprocessors, 64 bit integer instructions, the complex architecture and many others I forget.
I've spoken with two or three emu/plugin coders and the running consensus seems to be that we can expect a reasonably solid PS2 emulator that runs commercial games maybe sometime around a decade from now; certainly not on Xbox (or even on Xenon for that matter) and not on a P3 either
Akumajou
08-Jan-2005, 01:18
fastest p3 produced was the 1.4GHz tualatin. but that's hardly a commonly-available cpu, the consumer's p3 would be ~1ghz. so you're saying that a 1ghz p3 (sse) + a dx7 card can emulate the ps2, did i get you right?
I understand what you are saying about the P3, still what I meant is a properly written emulator that would provide accurate emulation and some texture filtering would have to be made so that it is taking full advantage of Direct X or OpenGL and the CPU+large memory (512MB to 1024MB as minimum).
Now this is just my theory, not a fact and I forgot to mention IMHO but I feel that an emulator that is probably written in an assembly or very close to the OS kernel+API+CPU & extensions (SSE/2/3, 3d Now, etc) basically the emulator has to work just like the PS2 was designed to work dependant on the EE+GS sharing the workload as opposed to XBox where the GPU (if a dev codes it) will handle the graphics load while the CPU does other tasks like A.I., etc.
The XCPU is actually weaker then the EE at least in some respect. And yet you're saying that it's fast enough to emulate it? Are you serious??
Like I mentioned I neglected to mention "IMHO" and I understand what you mean as I also know how slow the DC emulator Chankast and even that working Saturn emulator Cassini and the up coming SEGA Model 3 emulator are running on Pentium 4 at 2Ghz+ howeven I feel that it is possible to make the games run emulated if the coder is closer to how the graphics filtering API+cpu extensions+ large memory+OS kernel so IMO it would take a professional coder to do it, it is not impossible however.
akira888 wrote:
The reason there's no PS2 emulator out that runs commercial games is not for lack of trying or Sony "buying emucoders" but instead it is due the existence of real hardware issues: 128 bit GPRs, two independent floating point vector coprocessors, 64 bit integer instructions, the complex architecture and many others I forget.
I've spoken with two or three emu/plugin coders and the running consensus seems to be that we can expect a reasonably solid PS2 emulator that runs commercial games maybe sometime around a decade from now; certainly not on Xbox (or even on Xenon for that matter) and not on a P3 either
Still maybe the PS2 emulator could be limited like Bleem for DC was in that it was best to optimize the emulator for each specific game to be emulated and enhanced (just run at 640x480 with texture filtering to smooth out the edges)
I often wondered how Shenmue II, a game written specifically for a SH4+PowerVR2 DC (and I assume was written in second gen assembly level custom SDKs) and released in 2001 was ported to XBox so quickly with no real, major graphical enhancements, even on the textures since many still claim that the DC's textures were more accurate yet no major flaws.
Also I feel that although it is not a standard now since most emus are written in C++ for one CPU (Intel/AMD) alot of these console emulators could greatly benefit if they required a dual CPU with large memory (1024MB minimum) set up to compensate for the two independent floating point vector coprocessors, and 64 bit integer instructions. Maybe with upcoming dual core cpus it might be possible.
Also the Chankast story on it running XBox does give hope since those guys devoted so much hard work of their time:
http://www.ngemu.com/forums/showthread.php?t=58320
version
08-Jan-2005, 04:32
ps2 modern:)
http://web.axelero.hu/varga1973/ps2.JPG
see colon
08-Jan-2005, 05:09
I often wondered how Shenmue II, a game written specifically for a SH4+PowerVR2 DC (and I assume was written in second gen assembly level custom SDKs) and released in 2001 was ported to XBox so quickly with no real, major graphical enhancements, even on the textures since many still claim that the DC's textures were more accurate yet no major flaws.
the dreamcast (followed by the xbox) was really the first console to have an industry (IRT PC's) standard graphics subsystem. the xbox has a faster cpu and a faster gpu with basicly a superset of features compared to the pvr2dc. it's no surprise that shenmue ran so well.
Still maybe the PS2 emulator could be limited like Bleem for DC was in that it was best to optimize the emulator for each specific game to be emulated and enhanced (just run at 640x480 with texture filtering to smooth out the edges)
the dreamcast's sound subsystem alone is more powerful than the psx and has, what, 8 times the memory. the xbox and ps2 are comparable in speed and overall memory, with an obvious edge going to the xbox. you can't expect any system from this generation to emulate another, it's just not feasable.
ondaedg
08-Jan-2005, 05:56
I have read alot of talk about assembly level programming. Are console developers still using assembly for console games?
AlStrong
08-Jan-2005, 06:03
probably for shaders on the xbox.
Akumajou
08-Jan-2005, 06:34
I have read alot of talk about assembly level programming. Are console developers still using assembly for console games?
Assembly level SDK or Software Development Kits and custom versions and continually revised versions have been made to absolutely stay competitive against the stiffer competition of today's consoles.
Back in the Genesis vs SNES days it was used by SEGA to dispell the "mode 7" scaling and rotation of 2d sprites and fields.
It was used again with SEGA Saturn's in house dev teams like AM2 and other Amusement Divisions, Sonic Team and Team Andromeda to get closer to the Saturn's metal and produce game that dispelled the myth of Saturn being 3d deficient with games like Burning Rangers, Nights into Dreams, Virtua Fighters, Virtual On, Panzer Dragoon, etc just look at this movie of a prototype Project Berkerly aka Shenmue running on Saturn and requiring the 4MB expansion for video ram:
http://www.sega-saturn.com/news.htm
It was also used by Sony in house dev teams like Polyphony Digital to make Gran Turismo and GT2 possible, I think Konami used it to make Metal Gear Solid but I am not sure even though that game was exclusive to PSX and Namco used it too as well as other dev teams.
SNK has used it to push the 1989 Neo Geo 2d hardware to its limits in 2d graphics.
Nintendo has used it to make Super Mario 64 and other games from RARE on the N64.
It was used somewhat to a lesser extent on Dreamcast (unfortunatly) as SEGA was trying to get away from the supposed "nightmare to program" Saturn.
Sony absolutely had to used it at the start of the PS2's life cycle since they had no initial plans on doing online games like Sega was aiming at and as a way to get a major head start in development in pushing the PS2 to its limits in anticipation of the higher powered XBox and possible NGC threats.
N used it on GC to make dazzling 3d effects possible in many of their first party games.
As for Microsoft XBox, I am not sure, I believe it was confirmed that Tecmo tried or did used their own in-house developed assembly tools as opposed to using MS's SDK based on DirectX 8.0. However I believe there were two initial versions of the Dx SDK, one that was more like C++ and another that was closer to the metal but it was still Direct X, hence a possible reason why Tecmo and any other dev teams who want to prove their talents at pushing hardware.
However the main idea with XBox's SDK was to make it easier for devs to make a PC version, so I can see myself here that is good for some devs but bad for others who are not interested in making PC ports.
As for up coming consoles like XBox Next (some call Xenon), Nintendo Revolution and Sony PS3, I can only assume that Nintendo and Sony will immediately use assembly level SDKs instead of C++ like based to give them a head start unless I am wrong but if right they will be able to push those consoles to their limits much faster since they are not going to give their first party games to other competing consoles.
Thats also one of the reasons why current PS2 games like MGS3 Snake Eater look so impressive even compared to XBox specially when devs (some of them at least) have been gathering experience in knowing the Assembly level SDKs like the backs of their hands they would probably rap or sing assembly if dared.
Also keep in mind that is also the reason alot of 3rd party games are exclusive to only receiving one console version as re-programing a game, specially an assembly game takes alot of financial investment on the part of the dev company unless they can affort do have 3 or 2 dev teams each working on the 3 or 2 differing console versions. This also helps Sony and Nintendo if a 3rd party uses assembly level SDKs because it almost assures that the game will be exclusive and if its a good or great game, it helps to sell tons more consoles.
Now this is just my theory, not a fact and I forgot to mention IMHO but I feel that an emulator that is probably written in an assembly or very close to the OS kernel+API+CPU & extensions (SSE/2/3, 3d Now, etc) basically the emulator has to work just like the PS2 was designed to work dependant on the EE+GS sharing the workload as opposed to XBox where the GPU (if a dev codes it) will handle the graphics load while the CPU does other tasks like A.I., etc.
In others words, you want a PS2 emulator that goes close to metal, and you want the emulator to act like a "simple kind" of translator, also know as Compatibility Layer.
But there's a problem here, and it resides in the fact the closer you go to the PC architecture metal, and bigger the underlying and fundamental differences, with the PS2 architecture, are visible.
And the PS2 architecture, as others already told you, is really different than what you found in actual PC (Even futures dual cores ;) ).
The closer thing, in emulation that ressemble your ideafor obtaining accurate emulation is known as dynamic recompilation, Dynarec.
But here, again, you have to understand that a good and efficient Dynarec requires a lot available raw power.
Also I feel that although it is not a standard now since most emus are written in C++ for one CPU (Intel/AMD) alot of these console emulators could greatly benefit if they required a dual CPU with large memory (1024MB minimum) set up to compensate for the two independent floating point vector coprocessors, and 64 bit integer instructions. Maybe with upcoming dual core cpus it might be possible.
You're really suggesting that you can "compensate" the absence of the VU1 VU0, by having a lot of RAM available?
When it comes to PS2 emulation, RAM isn't an issue.
And lots of it won't help, at all, with accurately emulating the PS2's, really powerful, vector units. :wink:
I have read alot of talk about assembly level programming. Are console developers still using assembly for console games?
On today's game? Some developers might throw a few line of assembly here and there, in some hardware that lacks excellent compilers *cough*PS2*cough*.
But, generally, games are written using high-level language, and that for different reasons.
The main ones are readability, the game code in modern games is so complex and dev period are so short that you don't have time to loose hour looking for something.
There's also the portability, a lot of games, engines (graphic, sound, physic) are ported from a platform/game to another, and low level language are not the solution in this case.
And then you have "simplicity" (term used losely), as i said games are so complex, and the time available is so short that any technique that saves you some precious hours are welcomed.
The last one i cite would be the fact that today's compilers do an excellent job, making assembly obsolete for most of usage. For instance there's no need for low level language, AFAIK, in Xbox, since the machine embark an X86 processor, and seeing that compilers give excellent results.
Panajev - how many times can you mis-spell contradicted. :P
Inane_Dork
08-Jan-2005, 09:06
The last one i cite would be the fact that today's compilers do an excellent job, making assembly obsolete for most of usage. For instance there's no need for low level language, AFAIK, in Xbox, since the machine embark an X86 processor, and seeing that compilers give excellent results.I would guess that many Xbox games have x86 ASM for animation work. And almost all of the shaders written for the Xbox I would guess to be shader ASM.
There will always be a use for ASM. Compilers will, for the forseeable future, never support all the features everyone will want to use. Take the cache locking patent for the X2, for instance. Either a compiler will have to be specifically modified to handle that feature or developers will handle it through ASM (I hope I'm not presuming too much here). I guess it will be the latter.
Akumajou:
You quite don't quite grasp what was special about PS2. It is the fact that VU-units were really fast for it's time, same for the GS and memory speeds. GS does have 16 pipelines which could really work to full efficiency, albeit for texture mapping you had to combine pipelines so you had 8 pipelines for single textured surfaces, 4 pipelines for dual textured and so on. GS has the whopping 48GB/s bandwidth internally to really pull of some stuff not really possible with tnt2.
Based on nvidia pages(http://www.nvidia.com/page/tnt2.html) tnt2 ultra had 2.9GB/s memory bandwidth and had to use AGP2 to transfer data from main memory to graphics memory. Compare this to ps2 :) And yes, tnt2 had to do t&l on processor and transfer vertices via agp. PS2 has 3.2GB/s rdram as main memory, dedicated vector units with peak(not achievable) 6.4GFlops performance and so on. Try to emulate with tnt2, yeah, or even run same algorithms with same speed.
Compare PS2 to pc solutions available at the launch of PS2 and year after and you will see that even the peak specs of tnt2 are not anywhere near, much less the achievable results when you factor in memory speeds. I consider one of the biggest problems of the XBox the memory speeds, 6.4GB/s combined for cpu and graphics, not that much, luckily it does have nice caches to help, but STILL, MS had a WHOLE lot of time after PS2 launch and still so poor memory buses. Maybe some developer can give insight, if the memory bus in xbox is really as bad as I think it to be?
This is not to say that ps2 is god or anything, just that it was really, really fast, but simple. Also the complexity to code for it... well... not that bad anymore as better tools are available, hopefully sony has learnt its lesson from the past.
If you start thinking about emulating ps2 you would at ideal case need the 48GB/s internal bandwidth and 4MB memory to it. Also 3.2GB/s for main memory. Also factor in various caches and scratch pad memories in system, you have to emulate them probably cycle exact. Also games contain code that is to the metal and trust to strict timings that you would need to emulate in some compatible way adding to extra burden.
Considering previous and all other kinds of things that hinder efficiency I would guestimate needing at least twice the specs and performance of ps2 to emulate it. Really, where do you have such a pc, NOW? Sure 2007-8 or so there might be a fully working emulator for some not too expensive pc or similar machine... or maybe even earlier if sony puts soc implementation of ps2 into ps3...
I would guess that many Xbox games have x86 ASM for animation work. And almost all of the shaders written for the Xbox I would guess to be shader ASM.
There will always be a use for ASM. Compilers will, for the forseeable future, never support all the features everyone will want to use. Take the cache locking patent for the X2, for instance. Either a compiler will have to be specifically modified to handle that feature or developers will handle it through ASM (I hope I'm not presuming too much here). I guess it will be the latter.
You're perfectly correct, and i agree like i said HLL, in Xbox, is perfect for most of cases, not all the cases, of course.
I just forgot the word "usually" in the second phrase you quoted. :D
london-boy
08-Jan-2005, 16:05
As the topic poster mentioned, the GS is 1998 technology, therefore comparable to graphic chips released in that year as far as capabilities and the fact that the GS cannot be compared to 1999 PC-GPUs because those chips introduced things like hardware transform & lighting (S3 and Nvidia), enviroment mapped bump mapping (Matrox G400), texture compression (S3), etc.
Yeah, only it's not comparable at all. As we have discussed ages ago a loooong time ago, those PC chips might have features that are absent on GS (since everything is absent from GS, pretty much), but the GS is so much faster than them it's not even funny.
So, no, they're not comparable. Not from a preformance point of view, not from an architectureal point of view (just the fact that one has 4MBeDRAM is a big hint), not from a targetted audience point of view, not from a functionality point of view, even the targetted resolution is completely different.
Duh, Sony started shifting to streaming games of the media as opposed to just loading a level into memory since the PSX days, the only reason that became a standard for the console industry is because Sony was the name brand sales/marketing leader, NOT technological leader.
Streaming levels never became a standard, and there is no "reason" behind it. It just happened because on consoles it's a more clever design, which sadly has its limitations anyway. I never praised Sony as being the first to do that (although they are, in the end). Seems you're a bit defensive when it comes to Sony.
So what are you, a believer in that PS2 is so powerfull because you claim that the video ram argument was closed 3 years ago? you are wrong and I am part of the few who will say it.
I'm not a "believer". You said something, it's wrong. Over.
Like I said, Sony's PS2 is NOT a technologically superior product compared to SEGA's Dreamcast, they are almost on similar power levels with the PS2 only being able to push more polygons.
Well then maybe you need to look at old threads on this forum with posts from people who you might believe more than you believe me. Although i see the Force is strong in you, not much will convince you of what at the end of the day is true.
All the 3d effects on PS2 are handled in software, if AM2 could take a 4MB cart and make a prototype Shenmue (however much you may hate that game) run on SEGA Saturn (1994 tech) using their own custom assembly level SDKs and their obvious experience then its pretty damn obvious that even the worst developer in the world, with plenty of experience using Sony's custom assembly level SDKs will be able to put out amazing 3d effects on either PSX or PS2.
I don't hate that game, why would you think that? Because i'm making some point in favour of Sony, now i hate everything Sega-related? Hell, 2 days ago i was the daily Nintendo Fanboi for defending them, the day after i was the daily Nintendo Hater for slating them, today i'm the Sega Hater....
Maybe you never knew that SEGA Genesis was not able to do the SNES "mode 7" effect, but a couple of years after the SNES was released there were Genesis games doing the same tricks like the infamous "mode 7" thanks to developer experience in using assembly level tools.
Again, your point being? I had a Genesis and i know what it could and what it could not do. Not sure what this has to do with the current discussion, or are you gonna take more rabbits out of your hat, and mention all the great things Sega ever did in their life, just so they look better in your eyes? I love Sega, the most fun i've had in my life was with my Genesis, i'm just sad they haven't been able to keep up with their standards lately. Does that make me a Sega hater, just for pointing it out? If anything, it makes me more of a Sega fan for demanding more of them, considering the quality of their old products.
Gee I wonder why the PS2 had to have heat sinks and a big huge fan suck all the hot air out of the console. (sarcasm)First of all it was 1998, second we can assume that although the die shrink provided all the heat dissipation and lower voltage requirements that Sony desired to make sure that the GPU would be as fast as it could complement the EE CPU, again its worth mentioning that the GS is not a GPU like the Nvidia Nv10 and ATI R100 were with their on core geometry processors and therefore such a graphics chip required a fast CPU just like the Matrox G200, G400, Nvidia Riva TNT and even 3DFX Voodoo 2/3 to reach a higher performance.That also explains how A.I. has not evolved much on PS2 because the EE is all tied up.
Again, your point? My point was that the die shrink helped Sony reduce the size and cost of the PS2, because the smaller chips now run cooler than the early ones which needed lots of (noisy) active cooling. So, what's the problem here?
It has to do with the "subject" because if you lived back in 1998 and 1999 prior to the 2000 release of the PS2 you would have been bombarded with all the hype Sony was spitting about how Square would be able to display games pushing 80 million polygons, you would have read interviews of EA and Sony dev teams dismissing the DC as a "failiure" months prior to the 9-9-99 DC launch, all creating a mindshare atmosphere controlled by Sony marketing evangelists.
Sounds to me like you're another one of the boys who bought into the hype, got burnt, and now are Sony haters. For your information, i began playing videogames on my Commodore64. Actually even before, with those stupid Tiger portable videogame things.
So, thank you, i was around very much at the time PS2 was released, i listened to all the hype and knew what was hype and what was reality. Seems like you didn't.
Basically, having those "evangelists" and the suckers that believed it allows companies like Sony to release 1998 technology in the year 2000 and not get critisized or punished for it.
PS2 was released in 1999. And now we need to punish Sony for having the design of a chip completed a few months before actually releasing it? Jesus, someone really has a chip on their shoulders then!
There are many reasons why Sony couldn't release PS2 sooner, one of the biggest was that they weren't getting good enough yields on their chips, especially the GS if i remember correctly.
And what the hell is wrong with that? rich boy Sony came, saw, stole and conquered the competition. :lol: :lol: MOMMY MOMMY BIG BAD SONY CAME AND STOLE THE MARKET!!! Get the f**k over it, Sega were broke, they couldn't stand a chance against anyone. Sony only did what they had to do, they didn't steal anything from anybody, they worked hard to get where they are, and it's still paying off now against 2 newer consoles.
Or you could see Sony as Walmart taking market share away from mom & pop game shops like SEGA & Nintendo eventually leading them to a financial corner. I never saw Nintendo making tv sets, SEGA selling DVD players, radios, cassette players, cd-players, etc.
Well that's bad for them. Sony obviously has better advisers who saw a market for all-in-one boxes and they aggressively pursued their objective. If Sega and Nintendo are stuck with their ideals, that's too bad for them. Or now we have to punish Sony for pursuing their objective and kmake products that are bought by hundreds of milliosn of people around the world? If anything, let's punish the lazy ones who think they can give us the same formula of the last 20 years over and over again.
Or maybe you just cannot imagine a world without Sony's involvement in the videogame field.
Why would you say that? I was playing games before Sony came into the market and i will play games well after they go bust.
I guess a prototype Shenmue running on a Sega Saturn that was hyped by the competition as being 3d-deficient must be very offensive to you.
Offensive? Not at all, i just went "not again!!" cause not too long ago SegaR&Deadmeat opened a thread on the subject. And it got very nasty obviously.
Can you please go out and find, rent or borrow (or dl if you mod) VF4 and compare it to VF4Evo, then compare it to the VF4 and VF4Evo arcade machines, you just might notice how VF4-PS2 looks like garbage compared to the closer to the arcade VF4Evo.
I never doubted that. I used to own VF4 and V4EVO. I know what they look like.
I picked SEGA because they are NOT ass-kissing Konami or Sony first parties mainly because Sega dev teams had to start learning Sony SDKs much later than any other game developer so it makes sense that their first games as a third party would never look as good as a Konami effort, thats a big change from how Sega dev teams had SDKs months before any third party would or the same could be said for Sony first devs and second devs because its obvious Sony would want Square to have a FFXXX ready for console launch day if they could get away with it.
Good god!! Your spite for Sony really is deep, jesus!! You could have said "WEE WEE BIG BAD SONY GAVE THE SDKS TO OTHER PEOPLE FIRST WEE WEE KONAMI ASSLICKER SQUARE ASSLICKERS WEE WEE" and it would have made the same impact.
I still fail to see your point, you keep stating the obvious, then when i tell you u've stated the obvious, u go "Well DUH!"... :? :?
You're not PC-Engine by any chance are you? Just checking...
Now it seems like you are just disagreeing or re-cycling the same mesage I originally posted just for the sake of finding something to do.
Gee I wonder why a game that comes out 2-3 years is always able to push more 3d effects+performance out of a console, hey maybe thats what makes consoles so great over PCs.
As i said, you keep stating the obvious, then when i ask you why you're doing that, you just go "Gee i wonder why...!"
Maybe I could have said that the Sony driver has to learn how to drive a 6 speed manual transmision touring car and they figured that drawing the race track by hand in their backyard allow them to remember all the curves come race day even though the touring car was a tricked/pimped out Yugo racing against SEGA Supras and Nintendo RX-7s with drivers who had only heard of the track and assumed that with their experience they would win or complete the race.
Whatever.
Do you remember the price of DVD players in the US back in 1999 and months prior to PS2's 2000 launch?
In the US? No. In the UK, yes i do.
Basically the only new technologies being introduced in consoles was the DVD drive and enhanced backward (but flawed) compatability being built in and the sole reason alot of people rushed out to get PS2s, to watch movies and play old PSX games enhanced since the majority of first gen PS2 games were horribly below the then Dreamcast standard, something that should never have been.
YAWN. Oh it's the "PS2 was just a DVD player that could play PS1 games!! UNACCEPTABLE!!" argument...
So what?, the Topic poster asked for opinions on how the PS3 would compare to PS2 in its time of introduction, I gave my opinion, deal with it.
You're replying like this to a totally neutral statement i made. Meh...
Gee I seem to remember that Genesis & SNES story about "mode 7", how about I tell you how both PSX and Saturn dev kits were buggy but because the Saturn used dual CPUs, it was critisized more for being more complex to develop a SDK for.
Getting bored now... Your point? My point was, obviously early dev kits are buggy, the important thing is that there are always constant revisions. Why do you have to pick up the Saturn-PS1 issue? Who cares? We're talking about soemthing else! Keep you Sega-Sony love-hate thing out of this! Or are you gonna play the Saturn card and the Genesis card every time you feel like you have to prove a point? :roll:
I have been very well informed on how SEGA, Sony, MS and N always in their historical times have always had to release updated SDKs, that is a standard that was beaten to death on SNK's Neo Geo 2d arcade hardware if you compare Fatal Fury to Mark of the Wolves when it comes to fluid 2d animation.
Your point? Of course it's a standard! Do you really expect those companies to have fully functional, 100% bug-free SDKs at launch, when they even know very little about their own new architecture?! What is your point?!!
And I am to be punished for even remotely liking SEGA??
No one's being punished, it's your attitude that's very irritating, and i've pointed it out already.
You missed the whole point of my post then if you just want to narrowly pre-judge me into a SEGA <bleep> box.
There was no point in your post, other than to tell us how Sega is a magnificent company and other aren't, that's why.
If you are familiar with emulators you would know that Dreamcast was capable of running an emulator that would not only raise the resolution of certain PSX hit games, but breathe new life into them.
There are PSX emulators that work on PCs and the minimum requirement is an Intel Pentium II or I with MMX extensions for visual parity with PSX image/sound quality. Basically the emulator uses that old Intel CPU running at 200Mhz and the MMX extensions to emulate a 1994 console that had a 33Mhz cpu.
PSX emulators took a long time to develop, the console was released in 94 in Japan and the first actual working emulators I remember were released in either 1999 or 2000 with Bleem being one of the first that tried to go commercial (and they got sued for it) If you can count, it took hobbyists (the people who make emus) over five years to properly make accurate emulation and later enhanced emulation with the help of powerfull graphics cards and APIs like Direct X and OpenGL.
Now it did not help that Sony offered to hire (as in give jobs) to those hobbyists so as to prevent their fears from becoming a reality in a hobbyist made PS2 emulator being made much sooner thanks to the hobbyist's experience in making previous emulators.
Taking into account that the PS2's EE runs at 300Mhz and the GS runs at 150Mhz, it makes sense to say and assume that a properly written PS2 emulator could be made with a PIII as a minimum requirement for accurate emulation and a Dx+OGL graphics card to provide texture filtering enhancements.
As to when that will happen, hey its now 2005 and people are still working on it but since they do not get paid it depends on how much time they dedicate to it.
If we were to get a company to do it, like the company that made Bleem, we would have had a PS2 game emulator running on XBox 2 years ago.
Thankfully that's already been clarified. The Xbox or the Xbox2 will never ever be able to emulate the PS2 at decent speeds. The software is far too different to make it workable at decent speeds on anything but a real monster architecture, and it has to be a monster not because PS2 is a monster, but becasue it will need a lot of overhead to translate Arab into Mongolian on-the-fly.
cthellis42
08-Jan-2005, 17:49
I don't hate that game, why would you think that? Because i'm making some point in favour of Sony, now i hate everything Sega-related? Hell, 2 days ago i was the daily Nintendo Fanboi for defending them, the day after i was the daily Nintendo Hater for slating them, today i'm the Sega Hater....
Quiet, l-b. I believe he has proven quite handily that you are the Great Satan. FIE!
;)
And while I'm not touching the rest of this mess:
PS2 was released in 1999. And now we need to punish Sony for having the design of a chip completed a few months before actually releasing it?
March 6th, 2000. At least I'm pretty sure it was the 6th. First week of March, at any rate. When the various stages of chip design were locked down is still debated by folks on here even now, though.
london-boy
08-Jan-2005, 17:58
^^ OK then 2000 it is. Not sure why i had 1999 in mind...
Anyway, i just don't see the huge deal, there were obviously reasons why Sony had to sit on the design for so long. It is questionable whether they should have updated the design in that long period of time, but really, in the end it all come sdown to cost and updating it would have probably pushed the cost a bit too far, already being quite huge.
london-boy
08-Jan-2005, 18:09
Quiet, l-b. I believe he has proven quite handily that you are the Great Satan. FIE!
;)
:evil:
function
08-Jan-2005, 18:42
Streaming levels never became a standard, and there is no "reason" behind it. It just happened because on consoles it's a more clever design, which sadly has its limitations anyway. I never praised Sony as being the first to do that (although they are, in the end). Seems you're a bit defensive when it comes to Sony.
There were console games streaming data off the media long before Sony entered the market. There were M-CD and CD32 games that did this and no doubt other machines had them too (like the Commodore CDTV and the PC-Engine CD). The Saturn was paricularly capable in this regard, having a processor dedicated to CD access. If anything I think this supports your point ...
All the 3d effects on PS2 are handled in software, if AM2 could take a 4MB cart and make a prototype Shenmue (however much you may hate that game) run on SEGA Saturn (1994 tech) using their own custom assembly level SDKs and their obvious experience then its pretty damn obvious that even the worst developer in the world, with plenty of experience using Sony's custom assembly level SDKs will be able to put out amazing 3d effects on either PSX or PS2.
Shenmue didn't use the 4MB expansion cart. That rumour was created from assumptions made by internet types. Infact, the game was probably canned for the Saturn and switched over to DC before the 4MB expansion cart even got the go ahead.
Gee I wonder why the PS2 had to have heat sinks and a big huge fan suck all the hot air out of the console. (sarcasm)
Probably the same reason the DC needed a fan and complex cooling system (it produced a lot of heat).
london-boy
08-Jan-2005, 18:49
There were console games streaming data off the media long before Sony entered the market. There were M-CD and CD32 games that did this and no doubt other machines had them too (like the Commodore CDTV and the PC-Engine CD). The Saturn was paricularly capable in this regard, having a processor dedicated to CD access. If anything I think this supports your point ...
There u go, even better. I never even mentioned this cause i wasn't too sure to begin with, and had nothing to do with the discussion, but just out of curiosity, what games on Saturn had no loading levels? That's quite neat.
ondaedg
08-Jan-2005, 18:54
Thanks for all the ASM info, but I really didn't think that there was too much ASM being used today. As a matter of fact, wasn't it John Carmack who stated that with today's modern development platforms, that there really is no need to use ASM? Maybe that doesn't apply to the PS2 since its architecture is much different than the XBox and GC. I do recall the wonders that Sega had done for Saturn using ASM, but that proved to be one of its downfalls as well since developing in that type of environment is much more difficult. With this next round of consoles, I can't imagine any of them using assembly.
function
08-Jan-2005, 19:21
There were console games streaming data off the media long before Sony entered the market. There were M-CD and CD32 games that did this and no doubt other machines had them too (like the Commodore CDTV and the PC-Engine CD). The Saturn was paricularly capable in this regard, having a processor dedicated to CD access. If anything I think this supports your point ...
There u go, even better. I never even mentioned this cause i wasn't too sure to begin with, and had nothing to do with the discussion, but just out of curiosity, what games on Saturn had no loading levels? That's quite neat.
I can't think of any games that had no loading screens at all, but that's just a matter of implementation. Any time you're playing a level and the game dumps data it doesn't need any more and loads new stuff in (in a seamless fashion) it's the same process going on. You can do this based on position within a map, time passed, opponents defeated, whatever.
You could quite reasonably say that crappy fmv games like the MCDs Thunderhawk do this: the UI, cursor and overlayed effects are being manipulated real time, but the background image is being streamed in, decompressed and having "collision" detection calculated on it at the same time.
My favourite example on the Saturn was Panzer Dragoon Zwei. The (awesome) music was chip generated unlike in the first game, so you continued travelling through the levels without interruption as new data (such as for the rather impressive bosses) was loaded up.
akira888
09-Jan-2005, 00:25
I'm not even sure that geometry and AI would even impact each other that much. AI is mostly parsing through various types of data structures (binary trees, directed graphs) which require scalar integer operations, while geometry is mostly vector pointing point. Since floating point [VU] and integer [R5900] run in parallel on the EE I don't see why this would constitute a trade-off situation.
I think the reason for the poor integer performance on PS2 has more to do with a relatively archaic though fast (for 99-00) CPU with high latency memory and a tiny cache. That combination is pretty deadly...
vBulletin® v3.8.4, Copyright ©2000-2010, Jelsoft Enterprises Ltd.