nVidia building the PS3 GPU in its "entirety"

Status
Not open for further replies.
ultimate_end said:
one said:
ultimate_end said:
Sony behind 3D labs? Judging from patent applications, Sony has been looking very hard at antialiasing technology and Sony does understand pixel shading (PSP), so it isn't so rediculous to assume that with enough money and work (they have that research department remember) that the PS3 implementation would have been acceptable. Then there is the SALCS/SALPS theory that would have been interesting in this regard too.

IMO Sony thought it's enough to beat competitors in the handheld space with the Sony in-house PSP GPU. Plus cost-wise handheld is tighter.

In the console space, they had to secure sure win over competitors with capable graphics technology nVIDIA and ATi demonstrated in this gen, especially when in so-called 'this generation' (though with unfair 2 years interval) PS2 is behind in hardware power. Consumer may wink it away once, but won't twice.

Will it really matter if the PS3's pixel shaders are only PS2.0 equivelent? (Unified Shader Model notwithstanding). What if the proprietry pixel shading technology actually turned out superior to Xenon's, much like how PS2's VU1 is functionally superior to Xbox's Vertex Shaders? In any case, why does everyone assume that Sony is so far behind nVidia/ATI? PS2 was designed in an older graphics technology era than Xbox for ****s sake!

one said:
Sony might choose nVIDIA programmable shader only to lure developers into the PS3, as a bridge to more avant-garde things avaliable in the PS3. If silicon space permits, they can even cram different graphics paradigms in 1 die. To save silicon space, Sony need optimized implementation of 3D algorithms nVIDIA put into their GPU. Though they already gambled big money into the Cell R&D as they are confident in winning, at the same time they have to be realistic without complacence to continue to be the market leader.

I'm very curious how Xbox2 turns out in the next month more than PS3, as so many PS3 info are disclosed in this timing. Can those Xbox executives and Nintendo executives sleep well these days or sh*tting in their pants? If I'm Steve Ballmer I'll put back Xbox 2 release to 2006 like Xbox 1 release was delayed several times to complete its spec than taking headstart with only a half year window advantage.

Yes, I do think that PS3 will still be a beast. I do not see eye to eye with those who somehow think that nVidia/Sony cannot make a graphics chip to compete with ATI's Xenon chip. That seems rediculous to me. I agree that Microsoft/ATI are probably concerned by the potential of PS3.

As for what this chip turns out to be, it's just that I keep getting the impression that it won't be anything truly special. Yes I know it doesn't actually matter, but what can I do, I'm a tech freak :oops: .

I predict a chip that is close to 1 billion transistors and is fabbed initially at 90nm and then moved immediately to 65nm (a repeat of the GS basically). If I hypothetically place an NV50 for example, a sound processor and 64MB eDRAM on a die then that would require the chip to have been designed strictly for 65nm as a 90nm chip would be horrifyingly big. Maybe I am under the wrong impression, but from what has been said through all of this, I still keep thinking that the final chip won't stray too radically from this. But according to my own advice, I will attempt to assume nothing. At least I may be pleasently surprised when the final chip features considerable SCEI input. I just feel that there is still going to be a lot of Sony ideas/research that will be going to waste.

Lazy8s said:
The design of the PS2 was finished only shortly before launch, like most systems -- even though Sony had originally planned to launch in 1999, some of the chips hadn't been ready and had required a layout to a new process size.

Designing a system is more than locking down the specs; there's also the whole fabrication side that determines what's possible. There's nothing hard about drawing up a system that can't be effectively produced until many years later, and that's really a sign of bad R&D scheduling -- not forward thinking -- since the goal is to have readiness from the target fabrication process and from the feature design of the chip to coincide as closely as possible so that neither part is left waiting around for the other and becoming outdated.

Yes, I in fact know all of this. My original comment was very loose and intended to support an argument that I am making. I should have been more specific about what it was exactly that I was referring to. Now I end up going off on a tangent, questioning when the GS was implemented in final silicon. But as I said before, the final hardware date is irrelevant considering my argument. No matter how long it took for Sony to fine tune their manufacturing process, how much work it took to implement edge antialiasing, or what compatability problems they had, the point is that, like the Dreamcast, the basic architecture of PS2 was conceived in a time before pixel shaders and before graphics chip integrated T&L. For Sony to suddenly drop all their work and throw an NV15 into the PS2 at the last minute would have been impossible IMO, and it still would not have had any pixel shaders!

I should have brought all this up months ago, but now it surfaces here and I end up involving innocent people such as Megadirive1988, One and yourself.

My point is that Sony is not as retarded as a lot of people like to believe. My point is that Sony has been carrying out graphics research before companies such as nVidia practically even existed. Why is it so hard for people to believe that Sony might actually be able to create an incredible GPU for PS3 on their own, or that the reasons why Sony went with nVidia might have nothing at all to do with Sony's technical failure to do so. That is my argument to everyone here.

And now it looks as though I have some other posts to write *looks at two previous posts*...


Why in the heck would go sony go with NVidia unless the NVidia GPU wiped the floor with sony's internal GPU? This deal is about last second as it can get as far as I can see. That would lead me to believe sony gave every opertunity to its internal team to improve its GPU to get with in the ball park of a NVIdia one and when it was apprent they could not they signed a deal with NVidia. Also if the Powervr team had access to the sony fabs and transitor buget the dreamcast would of wiped the floor with the PS2. If sony had to play with budget contrants everyone else does they would be dead in the water.
 
PC-Engine said:
CELL living room? :LOL:

Get back to me when Matsushita, Pioneer, Mitsubishi, Sanyo, Yamaha, Denon, JVC, Hitachi, Marantz, Runco, etc. supports CELL...

FYI, Cell will in some ways an open platform. Ken Kutaragi said this some years back. The recent press release reiterated this. Lol, the irony is that most of those companies have openly accepted Blu-ray, which can be considered to be a Sony thing. Not to mention, look at CDs (Sony/Philips). Are you going to tell me that CDs aren't everywhere now?
Do you recall x86? That seems to be pretty common these days doesn't it?
Small things can grow, that's all I'm saying.

Anyway. Regardless of whether Cell becomes the default standard, there will be enough support on the ground to sustain it as a viable enough alternative to anything else. That's all that matters in the end.

The weirdest thing about all of this is that I'm defending Cell, when I couldn't even give a damn what happens to it. Well, that is as long as I still get to play GT6 :p.
 
jvd said:
quest55720 said:
In any case, why does everyone assume that Sony is so far behind nVidia/ATI? PS2 was designed in an older graphics technology era than Xbox for ****s sake!



The logical answer would be if they had better they would use it in the PS3 and not go running to NVida at the last second for a GPU. Also considering that the GS of the PS2 has about as many features as a voodoo 1.
Right , had the features of a voodoo 1 when geforce 2s were hitting the market .

Only advantage the gs had was its fillrate .

A bit like saying the only advantage the P4 has over the original pentium is it's clock speed.
The high alpha blended fill rate of the GS is it's strongest point by far.
 
Most of the companies I listed above design and/or manufacturer their own chips. Why would they suddenly drop what they're currently doing and adopt CELL? I think MIPS and ARM make more sense in CE device than CELL unless CELL offers something extremely compelling for those CE companies to switch. We know why SONY and Toshiba will use CELL, ie they invested heavily in it.
 
PC-Engine said:
Most of the companies I listed above design and/or manufacturer their own chips. Why would they suddenly drop what they're currently doing and adopt CELL? I think MIPS and ARM make more sense in CE device than CELL unless CELL offers something extremely compelling for those CE companies to switch. We know why SONY and Toshiba will use CELL, ie they invested heavily in it.


"Matsushita, Sony to produce Linux for media devices

That is, a Linux platform for "Digital Home Electronic Devices".

Matsushita Electric Industrial Co. Ltd. (Panasonic) and Sony Corporation today announced their agreement to work in partnership on the development of an enhanced Linux®*1 platform for digital home electronic devices. The companies will aim to apply the fruits of their collaboration to their AV (audio visual) products and to make the source code*2 available to the public for free in order to encourage its broader use throughout the industry.

"
 
version said:
PC-Engine said:
Most of the companies I listed above design and/or manufacturer their own chips. Why would they suddenly drop what they're currently doing and adopt CELL? I think MIPS and ARM make more sense in CE device than CELL unless CELL offers something extremely compelling for those CE companies to switch. We know why SONY and Toshiba will use CELL, ie they invested heavily in it.


"Matsushita, Sony to produce Linux for media devices

That is, a Linux platform for "Digital Home Electronic Devices".

Matsushita Electric Industrial Co. Ltd. (Panasonic) and Sony Corporation today announced their agreement to work in partnership on the development of an enhanced Linux®*1 platform for digital home electronic devices. The companies will aim to apply the fruits of their collaboration to their AV (audio visual) products and to make the source code*2 available to the public for free in order to encourage its broader use throughout the industry.

"

You don't need CELL to run embedded Linux :LOL:

http://neasia.nikkeibp.com/wcs/frm/leaf/CID/onair/asabt/news/345387

NEC-Matsushita Alliance Officially Announces Linux Mobile Phone
November 23, 2004 (TOKYO) -- NEC Corp, NTT DoCoMo Inc and Panasonic Mobile Communications Co Ltd officially announced that they have co-developed a Linux-based OS for third-generation (3G) mobile phone handsets, FOMA 901i series.

NEC's FOMA N901i and Panasonic Mobile's FOMA P901i use MontaVista Linux Consumer Electronics Edition 3.1 as the OS. NEC's N900iL, a handset with a built-in wireless LAN released from NTT DoCoMo on November 16, 2004, also uses the same Linux-embedded OS. NEC, Matsushita Electric Industrial Co Ltd and Panasonic Mobile Communications have jointly been developing this OS since the announcement of their business partnership in August 2001. It took three years before they finally accomplished the OS.

According to NEC and Matsushita group, they have also developed application software such as device drivers, terminal emulators for PCs, schedulers and image viewers, along with the OS. "Although we cannot specify the names, we have previously licensed out our jointly developed software to other companies," a Panasonic Mobile spokesperson said. COSMOBIC Technology Co Ltd, a joint venture formed by NEC, Panasonic Mobile and Huawei Technologies Co Ltd of China, which designs 3G handsets, seems to have offered the software licenses.

The co-developed OS is currently used only in the FOMA handsets manufactured by NEC and Panasonic Mobile. The alliance, however, intends to aggressively license out the OS for mobile phones excluding FOMA handsets across the world. "The original purpose of the joint development was to create a global platform for 3G handsets," Panasonic Mobile said. "We need to further develop the OS to be used for more purposes, because the Linux used in it is largely specialized for FOMA handsets." The companies aim to secure stable license revenues by increasing sales to overseas 3G mobile phone makers in the future.

[NE Asia Online]

So it has come down to NEC needing CELL in it's cellphones (pun intended) to run Linux? :LOL:

http://www.celinuxforum.org/
 
"CELF was formed a little more than a year ago to spread the usage of Linux in consumer electronics devices. It includes Hitachi, NEC, Panasonic, Royal Philips Electronics, Samsung Electronics, Sharp, Sony and Toshiba Corporation, as well as more than 50 other member companies. "

50 member is enough? :)
 
If CELL was the only architecture to support embedded Linux then you might have a point, but as of right now wishing, hoping, dreaming itsn't going to make it true...
 
PC-Engine said:
If CELL was the only architecture to support embedded Linux then you might have a point, but as of right now wishing, hoping, dreaming itsn't going to make it true...

and why CELL? because cell will the best, fastest ,cheap ...

100 mills ps3 users
100-200 mills PSP2 users use it and more
 
version said:
PC-Engine said:
If CELL was the only architecture to support embedded Linux then you might have a point, but as of right now wishing, hoping, dreaming itsn't going to make it true...

and why CELL? because cell will the best, fastest ,cheap ...

100 mills ps3 users
100-200 mills PSP2 users use it and more

Uh..yeah..probably explains why my DVD player uses a dedicated MPEG2 decoder chip and a Faroudja scaling IC instead of a Emotion Engine. :LOL:
 
I give you the kudos, ultimate_end. I think you have a very level-headed outlook on the topic, and you are able to articulate it well in your posts w/o being overly verbose or obtuse. I think you make a great addition to B3D and stick around for a long time (at the least, not get frustrated and leave).

Your diligence in providing responses to subsequent posts is commendable, but at some point, you have to realize that certain "characters" here will deliberately draw you out with inane digressions and irrelevance just for the sake of arguing (it's what they do as a matter of identity). So don't feel you have to spend too much time replying over and over to those individuals. I think after a while, you will catch on to those who sincerely wish to carry a genuine discussion with you and others who only seek to bend your ideas out of context simply to tangle you into irrelevant arguing over nothing.

I do look forward to your future posts, and good luck here at these forums. :)
 
quest55720 said:
Why in the heck would go sony go with NVidia unless the NVidia GPU wiped the floor with sony's internal GPU? This deal is about last second as it can get as far as I can see.

Wiped the floor? That seems a little strong don't you think?

In my previous posts in this thread (which you still seem to not have read) I gave just a couple of examples of alternative explanations for the nVidia deal. There are probably others too.

That's just the problem In life, there are always alternative explanations to what is the first, most obvious one. People assumed that the world was flat, because they didn't know any better. We cannot assume that because an apple is red on the outside, this is also the case on the inside.

quest55720 said:
That would lead me to believe sony gave every opertunity to its internal team to improve its GPU to get with in the ball park of a NVIdia one and when it was apprent they could not they signed a deal with NVidia

Yes I agree with this in the sense that Sony may have ran out of time in the development of the GPU, considering that regardless of what BS Sony PR may say, they still realise that they must not allow Microsoft to get much of a head start. That would be a third explanation to "nVidia wiping the floor with the internal GPU". They may have run into delays.

quest55720 said:
Also if the Powervr team had access to the sony fabs and transitor buget the dreamcast would of wiped the floor with the PS2. If sony had to play with budget contrants everyone else does they would be dead in the water.

And this proves what exactly, other than you are a Dreamcast "enthusiast"?

I too would have liked to have seen what the Dreamcast would have been like if it had a PS2 vector unit and eDRAM. But it didn't, because it came out earlier than the PS2 and also because Sony and Toshiba developed their eDRAM process. You can't do what you can't do, son. That's exactly why lambasting the PS2 for not having pixel shaders and a 733MHz CPU is so rediculous.

Oh and BTW, how exactly is it that using a lot of transistors is Sony's fault? And FYI the PS2 is actually pretty darn transistor efficient compared to the Xbox. Don't forget that in terms of overall transistor count (including memory), the PS2 is actually pretty slim. Pretty good for a console that doesn't look that much worse than Xbox, wouldn't you agree?

And please don't bring Dreamcast into this, the whole point of PowerVR's technology was (and is) to make extremely efficient use resources that you have. Please come back and tell me when Microsoft or Nintendo announce they are using TBDR in their next-gen console.
 
PC-Engine said:
version said:
PC-Engine said:
If CELL was the only architecture to support embedded Linux then you might have a point, but as of right now wishing, hoping, dreaming itsn't going to make it true...

and why CELL? because cell will the best, fastest ,cheap ...

100 mills ps3 users
100-200 mills PSP2 users use it and more

Uh..yeah..probably explains why my DVD player uses a dedicated MPEG2 decoder chip and a Faroudja scaling IC instead of a Emotion Engine. :LOL:

DVD player? :), CELL is the future not present
 
version said:
PC-Engine said:
version said:
PC-Engine said:
If CELL was the only architecture to support embedded Linux then you might have a point, but as of right now wishing, hoping, dreaming itsn't going to make it true...

and why CELL? because cell will the best, fastest ,cheap ...

100 mills ps3 users
100-200 mills PSP2 users use it and more

Uh..yeah..probably explains why my DVD player uses a dedicated MPEG2 decoder chip and a Faroudja scaling IC instead of a Emotion Engine. :LOL:

DVD player? :), CELL is the future not present


Well first your going to have to prove that a cell cpu can do things faster , cooler and less expensive than a chip to say ohhh put a menu on your tv screen.
 
jvd said:
version said:
PC-Engine said:
version said:
PC-Engine said:
If CELL was the only architecture to support embedded Linux then you might have a point, but as of right now wishing, hoping, dreaming itsn't going to make it true...

and why CELL? because cell will the best, fastest ,cheap ...

100 mills ps3 users
100-200 mills PSP2 users use it and more

Uh..yeah..probably explains why my DVD player uses a dedicated MPEG2 decoder chip and a Faroudja scaling IC instead of a Emotion Engine. :LOL:

DVD player? :), CELL is the future not present


Well first your going to have to prove that a cell cpu can do things faster , cooler and less expensive than a chip to say ohhh put a menu on your tv screen.


future TV is online with internet connection with fast CPU
 
aaronspink said:
one said:
:rolleyes:
You should note that it's easier and cheaper to maintain Cell and software that run on it than maintaining several thousands of different chips. Also, note that you can adjust the number of cores and clockspeed freely in Cell.

Not really. See, the other software already exists. Hmm, imagine that, the world gets on just fine without cell...

As far as changing the number of cores freely... If free is several tens of millions of dollars minimum, then yeah, you can do it freely.

PC-Engine said:
Most of the companies I listed above design and/or manufacturer their own chips. Why would they suddenly drop what they're currently doing and adopt CELL? I think MIPS and ARM make more sense in CE device than CELL unless CELL offers something extremely compelling for those CE companies to switch. We know why SONY and Toshiba will use CELL, ie they invested heavily in it.

So aaronspink misses the trend in CE.
Matsushita(Panasonic) is also running towards the integration of their ICs by developing the UniPhier platform by 3 types (1. high power for car/home 2. middle for personal use 3. low power for cellphone) of one-chip system LSI that consists of a custom RISC CPU or an ARM CPU and a media processor.

While the answer of Sony Group CEO Nobuyuki Idei (at the presentation from Nomura Investment Forum, Dec. 09, 2004) to the question about the difference between Cell and UniPhier is not clear enough,

Q3: How does Sony's Cell strategy differ from that of Matsushita's UniPhier semiconductor?
A: We are aware of differences between the two concepts. The Cell chip possesses new processing technology, is optimized for software design, and was implemented to generate new business. Investments in Cell are investments in a new business.

* For further information related to Cell, please see the following press releases:
"IBM, Sony, Sony Computer Entertainment Inc. and Toshiba Unveil Cell Processor"
"IBM, Sony and SCEI Power-On Cell Processor-Based Workstation Prototype"

you can safely assume that Cell differs from UniPhier by its network-aware nature and scalability beyond CE use.
 
Status
Not open for further replies.
Back
Top