View Full Version : nVidia building the PS3 GPU in its "entirety"
Joe DeFuria
16-Dec-2004, 23:07
http://www.beyond3d.com/forum/viewtopic.php?t=19053
I only had a chance to listen to the first 15 minutes or so, but it was made quite clear that for all intents and purposes, nVidia is designing the GPU itself in PS3.
In other words, not really a "collaboration" with Sony on the design...not giving Sony some IP, etc, to integrate into a Sony design... nVidia is designing the GPU like they would any other PC chip.
The reason why the deal has a royalty structure, is because "Sony" is fabbing the chips.
It is interesting to hear that Jen-Husn say that X-box was a "massive" undertaking in part because nVidia basically designed the whole thing (GPU, MPU....). Whereas with the PS3, it's "only the GPU", so it's not as big of a deal.
Another little factoid: they expect to have a production GPU by the end of the year.
Why send a woman to do a man's job? nVidia specializes in this field, I'm not surprised that they're fully in control.
Dave Baumann
16-Dec-2004, 23:21
Joe, I've not listended to it yet (they didn't have a replay last I tried) -when you say produced by the end of the year is that this calender year (i.e. a few weeks), next calender year or end of the financial year? Was there any clarification?
In other words, not really a "collaboration" with Sony on the design...not giving Sony some IP, etc, to integrate into a Sony design... nVidia is designing the GPU like they would any other PC chip.
The reason why the deal has a royalty structure, is because "Sony" is fabbing the chips.
Is that what he explicitly clarified, or what you deduced from what he said?
Did he deny someone else who asked about Sony's involvement?
http://www.beyond3d.com/forum/viewtopic.php?t=19053
I only had a chance to listen to the first 15 minutes or so, but it was made quite clear that for all intents and purposes, nVidia is designing the GPU itself in PS3.
In other words, not really a "collaboration" with Sony on the design...not giving Sony some IP, etc, to integrate into a Sony design... nVidia is designing the GPU like they would any other PC chip.
Hello, Vince, did you read that?
Panajev2001a
16-Dec-2004, 23:52
I doubt there is no involvement from SCE, nVIDIA has MUCH less experience in e-DRAM design than SCE and I think e-DRAM makes sense for a console chip as they target a limited range of rwsolutions (even ATI pushed for e-DRAM for Xbox 2/Xenon's GPU if rumors are true)... really 480p and 720p.
Still, nVIDIA would not be lying when they said they designed the GPU by themselves (with SCE letting know some of the features they really wanted to have): e-DRAM is not the GPU, is part of the memory and the memory interface blocks ;).
Joe DeFuria
16-Dec-2004, 23:57
Joe, I've not listended to it yet (they didn't have a replay last I tried)
It's available now...if you're having trouble with the PC media, they also have a phone-in replay number:
U.S. Toll Number: 646-827-3701
U.S. Toll-Free Number: 866-369-8596
U.K. Toll-Free Number: 0-800-032-3672
International Number: 001 646 827 3701
Pin 173
...when you say produced by the end of the year is that this calender year (i.e. a few weeks), next calender year or end of the financial year? Was there any clarification?
Next calendar year if memory serves...I'm trying to listen to it again...
Megadrive1988
16-Dec-2004, 23:57
ironicly, I have not been able to listen to the conference myself.
interesting.
I could see that Nvidia is handling the entire rasterization portion. like, Panajev, I do not believe SCEI is not involved. they must be working with Nvidia to help Nvidia make the GPU work well with the Cell-CPU. and then there's the eDRAM.
hmmm.
MightyHedgehog
17-Dec-2004, 00:01
So, it sounds like PS3 GPU = NV2A-like deal.
*am clueless*
OK here's the direct link (http://www.talkpoint.com/content/17720C7F-49B7-4601-9993-DF7181F618CB/70A3801F-92B6-4636-95A6-047F9384D9CB/F89C7480-5118-44DA-8015-8A06C93D9206/3/StreamingMetaFiles/slide001.asx)
Joe DeFuria
17-Dec-2004, 00:06
ironicly, I have not been able to listen to the conference myself.
interesting.
I could see that Nvidia is handling the entire rasterization portion. like, Panajev, I do not believe SCEI is not involved. they must be working with Nvidia to help Nvidia make the GPU work well with the Cell-CPU. and then there's the eDRAM.
hmmm.
Well, I wouldn't say Sony has ZERO involvement, but the "PS3 GPU = NV2A-like deal" seems to be correct.
On an odd note...this first 15 minutes I'm listening to now is entirely different than before...now it's all PC stuff. I wonder if this was split into two parts....though there's no way to tell how to listen to one part or another.
Megadrive1988
17-Dec-2004, 00:09
I'm now listening to the conference... its PC stuff so far (5 min in)
Joe DeFuria
17-Dec-2004, 00:14
Via the direct link...it's at about the 40:00 mark...
Via the direct link...it's at about the 40:00 mark...
The talk about Sony starts from 35:00
darkblu
17-Dec-2004, 00:24
In other words, not really a "collaboration" with Sony on the design...not giving Sony some IP, etc, to integrate into a Sony design... nVidia is designing the GPU like they would any other PC chip.
then again, one could say that NV don't have lots of liberties when designing a PC chip. neither would i expect them to have those with the PS3 GPU design, no matter how Jen-Husn may make it sound like, or the audience may interpret.
It is interesting to hear that Jen-Husn say that X-box was a "massive" undertaking in part because nVidia basically designed the whole thing (GPU, MPU....). Whereas with the PS3, it's "only the GPU", so it's not as big of a deal.
precisely for that reason NV has possibly had way greater freedom designing the xgpu than it is going to have now with the ps3 gpu -- there's a certain architecture there they need to tailor that GPU for, and they have little control over that architecture, unlike last time.
Joe DeFuria
17-Dec-2004, 00:32
In the case of the xbox 1, we designed the entire system......it was a very large project proportional to our size....We're not building the entire system [for PS3], we're building the custom GPU....
...Although it's structured ultimately as a royalty deal, the reason for that of course is because Sony has its own fabs....and its important to them that we manufacture it at Sony....this is a capability that we'd like to leverage. So the easiest way for us to of course struture a deal like that.... is to structure it as a royalty.
However the way I think about it is no different than a chip that I'm building. Building a chip in its entirety...and I think about the gross profits that is consistent with a device like that... The economics that we ought to get from something like this should be consistent with gross profits we get from one of our GPUs.
So, upon a second hearing, it's not as clear cut (integration of technology) as I originally made it out to be. ;) But still, it's pointing toward nVidia building the chip. The royalty nature of the deal ceratainly seems to be there not based on any integration / sharing, but because of Sony fabbing and consuming the chips themselves.
Why send a woman to do a man's job?
http://www.ilovebeams.com
Basically, he defends the royalty way as something as fruitful as manufacturing a card, something like that. So if you are an nVIDIA stockholder, he want you to think virtually as if
In other words, not really a "collaboration" with Sony on the design...not giving Sony some IP, etc, to integrate into a Sony design... nVidia is designing the GPU like they would any other PC chip.
The reason why the deal has a royalty structure, is because "Sony" is fabbing the chips.
, right? :wink:
Megadrive1988
17-Dec-2004, 01:05
I'm still having problems with this webcast. had to start over again, and I cannot skip ahead. have to listen to it from start to finish. cannot get the full functioning version because it keeps detecting pop-up blockers even though Ive disabled all of mine. anyway, lets not jump to any conclusions about whether Nvidia is designing the whole GPU or parts of it..
He talks about the GPU becoming a programmable DSP...hmm...isn't that CELL?
Apparently 50 engineeres are working on it as opposed to 200 engineers for the xbox but they pretty much designed the whole system as opposed to the GPU.
Around 1 billion dollars of nVidia investement's have been made since the xbox gpu that are going into the PS3 gpu.
Still not clear how much influence SCE are having but he emphasises custom gpu.
He's extremely proud that Sony has chosen nVidia and how visionry the whole PS3 team are. The gpu will form the basis of media centric roles and apparently the PS3 will be the 'centre of gravity' for CE products.
And boy does he BIG-UP CELL! ;)
london-boy
17-Dec-2004, 01:09
Intriguing. PR gurus at their best from now till release...
To the console-exclusive gamers:
If you thought Sony or MS were bad, wait till u see NVIDIA spinning around in their golden hotpants.
Megadrive1988
17-Dec-2004, 01:09
Jaws wrote:
Apparently 50 enginneres ar working on it as opposed to 200 enginners for the xbox but they pretty much designed the whole system ass opposed to the GPU.
that is probably a clue.
I kinda doubt only 50 engineers are working on Playstation3 GPU. maybe only 50 Nvidia engineers, which means probably that Sony is also working on it. I just don't see a simply Nvidia-designed (Nvidia only) GPU in PS3.
Apparently 50 enginneres ar working on it as opposed to 200 enginners for the xbox but they pretty much designed the whole system ass opposed to the GPU.
hmmm that's probably a clue. I kinda doubt only 50 engineers are working on Playstation3 GPU. maybe only 50 Nvidia engineers, which means probably that Sony is also working on it. I just don't see a simply Nvidia-designed (Nvidia only) GPU in PS3.
Yes...50 NV engineers but total from Sony not disclosed...
london-boy
17-Dec-2004, 01:24
So there's gonna be a GPU with NVIDIA's name on it, on which NVIDIA worked on with Sony...
What's new?
AlStrong
17-Dec-2004, 01:25
The PS3 will be black and green.
:lol:
Megadrive1988
17-Dec-2004, 01:26
as far as I am concerned, it's still a custom Sony-Nvidia, or, Nvidia-Sony GPU for Playstation3. not exactly an 'NV5A' 8)
Trawler
17-Dec-2004, 01:27
So there's gonna be a GPU with NVIDIA's name on it, on which NVIDIA worked on with Sony...
What's new?
It confirms that Sony wanted more from NVIDIA than just IP for starters... :wink:
I'm still having problems with this webcast. had to start over again, and I cannot skip ahead. have to listen to it from start to finish. cannot get the full functioning version because it keeps detecting pop-up blockers even though Ive disabled all of mine. anyway, lets not jump to any conclusions about whether Nvidia is designing the whole GPU or parts of it..
Oh, doesn't the direct link I posted above work for you? As the URL contains a UUID in it, it may work only for me though. I can open it in Windows Media Player and can skip in the stream ahead.
Intriguing. PR gurus at their best from now till release...
Paraphrasing the same PR in 37337 ways until the PS3 release day.
Fafalada
17-Dec-2004, 01:31
Still not clear how much influence SCE are having but he emphasises custom gpu.
Translation - no hardware clip! 8)
Megadrive1988
17-Dec-2004, 01:32
Jen first mentions PS3 using "one of our GPUs" at around 31:15 mark
So there's gonna be a GPU with NVIDIA's name on it, on which NVIDIA worked on with Sony...
What's new?
Nothing new really except of the level of investment ~ $ 1 billion from NV gone into it and they're looking to recoup $ 300 miilion+ from the deal.
And the GPU is basically a programmable DSP processor that would be suitable for non graphics tasks also...
Joe DeFuria
17-Dec-2004, 01:46
Nothing new really except of the level of investment ~ $ 1 billion from NV gone into it and they're looking to recoup $ 300 miilion+ from the deal.
Ummm....I hope nVidia doesn't make too many deals like that. ;)
Seriously, IIRC, the 1 Billion dollar statement was made in reference to not direct PS3 R&D costs...but overall investment for all their tech / GPUs since the x Box chip. (I'll have to go back and find that statement again to be sure.)
And the GPU is basically a programmable DSP processor that would be suitable for non graphics tasks also...
Where was this said / implied? (I haven't heard the entire thing yet, or I may have missed it while trying to do about three things at once here....)
So the gpu will be based on Nvidia's next-generation gpu. But which one? Is it NV5X?
In one corner, ATI R500. In the other corner NV5X. Fight!
Nothing new really except of the level of investment ~ $ 1 billion from NV gone into it and they're looking to recoup $ 300 miilion+ from the deal.
Ummm....I hope nVidia doesn't make too many deals like that. ;)
Seriously, IIRC, the 1 Billion dollar statement was made in reference to not direct PS3 R&D costs...but overall investment for all their tech / GPUs since the x Box chip. (I'll have to go back and find that statement again to be sure.)
Yeah! ;) ...I already mentioned that in my earlier post...just too lazy to re-type! ;) ...but that ROI was for PS3 and not for further CE returns...iirc.
And the GPU is basically a programmable DSP processor that would be suitable for non graphics tasks also...
Where was this said / implied? (I haven't heard the entire thing yet, or I may have missed it while trying to do about three things at once here....)
31 mins+
'GPU is the digital signal processor of the 21st century'...
This seems WAY more in line with what I thought.
Sure it's a custom GPU (so is NV2A) because sony will have some input, but for the most part this isn't some radical new technology only developed for Sony, I see it as a combination of technology Nvidia planned for PC.
I'm kinda curious to see what vince is going to say regarding this info.
Brimstone
17-Dec-2004, 02:26
How many engineers does Nvidia use when they make a variant of their cores?
I'm guessing that the NV40 would take more people than say a NV47. So if the Sony GPU is based off a NV50 design it wouldn't be as labor intensive.
Jaws wrote:
Apparently 50 enginneres ar working on it as opposed to 200 enginners for the xbox but they pretty much designed the whole system ass opposed to the GPU.
that is probably a clue.
I kinda doubt only 50 engineers are working on Playstation3 GPU. maybe only 50 Nvidia engineers, which means probably that Sony is also working on it. I just don't see a simply Nvidia-designed (Nvidia only) GPU in PS3.
Yeah, but he also said the next-gen gpu architecture took _hundreds_ of engineers to design. That 50 number refers to the number of engineers it took to design that (for the PS3) particular implementation of the gpu architecture.
Megadrive1988
17-Dec-2004, 03:07
bbot wrote:
Yeah, but he also said the next-gen gpu architecture took _hundreds_ of engineers to design. That 50 number refers to the number of engineers it took to design that (for the PS3) particular implementation of the gpu architecture.
now that I've heard the whole conference call thingy, I say you're right, I agree that is basicly true. Nvidia pours alot of resources into the next-gen desktop architecture, but only a small amount to actually implement the PS3 version of it.
PC-Engine
17-Dec-2004, 03:17
Why do people automatically assume Nvidia needs help from SONY or Toshiba if the design requires eDRAM? How difficult is it to add an embedded frame buffer into a GPU design? You do realize that IBM has manufactured TigerSHARC chips for Texas instruments with 24Mb of eDRAM right? What makes TI any different from Nvidia?
Then again you're not going to hear... by the way Sony here's our GPU go ahead and plug it right in. They're both going have to work at it to balance out the GPU/CPU combination.
Joe DeFuria
17-Dec-2004, 03:31
'GPU is the digital signal processor of the 21st century'...
Oh...OK.. He's talking about the future of GPUs in general though, not the PS3 GPU as far as I can tell.
This isn't anything really new or earth shattering. Basically, GPUs are getting more and more general purpose, and that, combined with "two way" busses like PCI-E (or closed console busses) will converge to make GPUs do much more than simply "3D Graphics."
Nothing.....
Let's just say that Nvidia is probably doing 90% of the GPU and it is based off of one of their future designs. We don't know how flexible Nvidia's future graphics chip architectures will be and it could be that they are adaptable to a console solution with great ease. After that then they can incorporate what Sony wants to see in the GPU. That probably does mean high fill rate, great shader power/throughput, and insane geometry levels. I imagine Nvidia will make a fully competent GPU that Sony envisions for the PS3, not what Sony is capable of making on their own.
Megadrive1988
17-Dec-2004, 03:35
Sonic wrote:
Let's just say that Nvidia is probably doing 90% of the GPU and it is based off of one of their future designs. We don't know how flexible Nvidia's future graphics chip architectures will be and it could be that they are adaptable to a console solution with great ease. After that then they can incorporate what Sony wants to see in the GPU. That probably does mean high fill rate, great shader power/throughput, and insane geometry levels. I imagine Nvidia will make a fully competent GPU that Sony envisions for the PS3, not what Sony is capable of making on their own.
probably.
I cannot wait to find out the whole truth on PS3. but I suppose these revalations, at least to a large degree, probably kills the Graphics Synthesizer 3 or 'Visualizer' with on-GPU Cell-based components like A|SPUs. (right?)
*wants to know what Panajev is thinking now*
on second thought, maybe GS3 or Visualizer is not dead. maybe the GPU is still Cell-based at the front end with A|SPUs and Nvidia is making the entire backend.
'GPU is the digital signal processor of the 21st century'...
Oh...OK.. He's talking about the future of GPUs in general though, not the PS3 GPU as far as I can tell.
This isn't anything really new or earth shattering. Basically, GPUs are getting more and more general purpose, and that, combined with "two way" busses like PCI-E (or closed console busses) will converge to make GPUs do much more than simply "3D Graphics."
I got the hint that it's the architecture for PS3/NV50 GPU after listening to it again...and it's not anything new like you say. But I bet the main attraction would be the compilers and dev tools that would be optimised for the hardware to make Cg shine! ;)
Sonic wrote:
Let's just say that Nvidia is probably doing 90% of the GPU and it is based off of one of their future designs. We don't know how flexible Nvidia's future graphics chip architectures will be and it could be that they are adaptable to a console solution with great ease. After that then they can incorporate what Sony wants to see in the GPU. That probably does mean high fill rate, great shader power/throughput, and insane geometry levels. I imagine Nvidia will make a fully competent GPU that Sony envisions for the PS3, not what Sony is capable of making on their own.
probably.
I cannot wait to find out the whole truth on PS3. but I suppose these revalations, at least to a large degree, probably kills the Graphics Synthesizer 3 or 'Visualizer' with on-GPU Cell-based components like A|SPUs. (right?)
*wants to know what Panajev is thinking now*
on second thought, maybe GS3 or Visualizer is not dead. maybe the GPU is still Cell-based at the front end with A|SPUs and Nvidia is making the entire backend.
Megadrive,
Cell is basically a programmable DSP...NV are hinting that the GPU is a programmble DSP. 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 CPU. They only need an consistent ISA to run 'software Cells'! ;)
Panajev2001a
17-Dec-2004, 03:54
Sonic wrote:
Let's just say that Nvidia is probably doing 90% of the GPU and it is based off of one of their future designs. We don't know how flexible Nvidia's future graphics chip architectures will be and it could be that they are adaptable to a console solution with great ease. After that then they can incorporate what Sony wants to see in the GPU. That probably does mean high fill rate, great shader power/throughput, and insane geometry levels. I imagine Nvidia will make a fully competent GPU that Sony envisions for the PS3, not what Sony is capable of making on their own.
probably.
I cannot wait to find out the whole truth on PS3. but I suppose these revalations, at least to a large degree, probably kills the Graphics Synthesizer 3 or 'Visualizer' with on-GPU Cell-based components like A|SPUs. (right?)
*wants to know what Panajev is thinking now*
on second thought, maybe GS3 or Visualizer is not dead. maybe the GPU is still Cell-based at the front end with A|SPUs and Nvidia is making the entire backend.
I still think the GPU migh not be CELL based as that was basically plan "a0": plan "a1" already was not CELL based AFAIK and had won over plan "a0", before plan "a2" was finally chosen.
Megadrive1988
17-Dec-2004, 03:54
Megadrive,
Cell is basically a programmable DSP...NV are hinting that the GPU is a programmble DSP. 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 CPU. They only need an consistent ISA to run 'software Cells'!
Jaws: okay, I see :o I've been somewhat confused on that for while now
Trawler
17-Dec-2004, 03:56
If it's Cell based why didn't they just say so? It makes no sense. Everything from the CC points towards this being a traditional nv design with modifications to suit Sony (ala the XBox deal).
Megadrive1988
17-Dec-2004, 03:57
Sonic wrote:
Let's just say that Nvidia is probably doing 90% of the GPU and it is based off of one of their future designs. We don't know how flexible Nvidia's future graphics chip architectures will be and it could be that they are adaptable to a console solution with great ease. After that then they can incorporate what Sony wants to see in the GPU. That probably does mean high fill rate, great shader power/throughput, and insane geometry levels. I imagine Nvidia will make a fully competent GPU that Sony envisions for the PS3, not what Sony is capable of making on their own.
probably.
I cannot wait to find out the whole truth on PS3. but I suppose these revalations, at least to a large degree, probably kills the Graphics Synthesizer 3 or 'Visualizer' with on-GPU Cell-based components like A|SPUs. (right?)
*wants to know what Panajev is thinking now*
on second thought, maybe GS3 or Visualizer is not dead. maybe the GPU is still Cell-based at the front end with A|SPUs and Nvidia is making the entire backend.
I still think the GPU migh not be CELL based as that was basically plan "a0": plan "a1" already was not CELL based AFAIK and had won over plan "a0", before plan "a1" was finally chosen.
so basicly, I take it you are thinking the GPU is mostly Nvidia-based, with input from Sony. it seems to be the best choice, considering that Sony was quite alot behind Nvidia when PS2 was developed, and now that Nvidia has grown in resources and IP by leaps & bounds, Nvidia was the only really good choice for Sony..
If it's Cell based why didn't they just say so? It makes no sense. Everything from the CC points towards this being a traditional nv design with modifications to suit Sony (ala the XBox deal).
AFAIK, the interviewer didn't ask him whether it was Cell based or not but I definitely would've! ;)
Trawler
17-Dec-2004, 04:07
AFAIK, the interviewer didn't ask him whether it was Cell based or not but I definitely would've! ;)
This is Huang we're talking about. If it were Cell based he'd be screaming it from the rooftops.
Unless he was smoking something hallucinogenic?
AFAIK, the interviewer didn't ask him whether it was Cell based or not but I definitely would've! ;)
This is Huang we're talking about. If it were Cell based he'd be screaming it from the rooftops.
Unless he was smoking something hallucinogenic?
Well, indirectly he was pretty much ecstatic about Cell (and so was the interviewer!) by the end of the interview! :P
For the defn of Cell based, both CPU and GPU need a consistent ISA to run software Cells, no big deal! ;) I don't see by that logic why a 'Cell' based GPU would be suitable for x86 PC's. Afterall, their CPUs aren't Cell based anyway! :P
AFAICS, it's more a software issue than a hardware issue! ;)
Panajev2001a
17-Dec-2004, 04:19
What does it mean having a "Consistent ISA without having the SPU/APU construct" ? Are you talking about on-the-fly JIT translation ? Emulation ?
Do you think nVIDIA is extending the ISA their Shading ALUs use to support the same ISA as the SPU/APU ?
The GPU they are working on was in development already to be their next-generation PC GPU with or without Sony even though they were interested in PlayStation 3.
If you look at some nVIDIA patents you will appreciate how at the end good answers to the parallel processing problem are not that different after-all ;).
With that said, I do not see nVIDIA swapping their shading ALUs with APUs and then modifying the APUs to do the same job their own Shading ALUs were designed to do: that would be a MAJOR departure... something I think is beyond the scope of this collaboration.
Maybe future nVIDIA parts, after PlayStation 3, might include some CELL technology if nVIDIA decdes to license the technology.
I am thinking there will be e-DRAM, but with nVIDIA eyeing XDRI wonder if the GPU will use its own pool of XDR RAM at very high speeds rather than e-DRAM. It would be difficult to match e-DRAM speeds, but I think nVIDIA's and SCE's engineers know what their performance+features+cost target (they want high performance and advanced shading features, but a $600 to manufacture GPU will be a bit to extreeme for PlayStation 3, but a part of the job of good engineers is to use the R&D money the best way possible to maximize speed while minimizing, where possible [I do not think that destroying the design to save $5 on each chip is something either nVIDIA or SCE want], costs) is and how to get there.
snakejoe
17-Dec-2004, 04:21
...when you say produced by the end of the year is that this calender year (i.e. a few weeks), next calender year or end of the financial year? Was there any clarification?
Next calendar year if memory serves...I'm trying to listen to it again...
any update on this time? thanks.
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! :?
Megadrive1988
17-Dec-2004, 04:49
one wrote:
Oh, doesn't the direct link I posted above work for you? As the URL contains a UUID in it, it may work only for me though. I can open it in Windows Media Player and can skip in the stream ahead.
naa, it didn't work for me. I'll try it on another PC later. thanks for trying though.
edit: ohhh I tried it again on the same PC, now it's working. again, thanks very much. this is so much more convienient :)
Joe DeFuria
17-Dec-2004, 04:53
...when you say produced by the end of the year is that this calender year (i.e. a few weeks), next calender year or end of the financial year? Was there any clarification?
Next calendar year if memory serves...I'm trying to listen to it again...
any update on this time? thanks.
At about 38:50
The next gneration game console will consist of 2 devices that Sony has talked about. One of them is the Cell microprocessor....the second device that will be a companion device to it is the graphics processor and the graphics processor will also be the image processor and there's all kinds of exciting features that will(?) come out. It's based on the next gneration GPU technology and our expectation is to try and put it into production this year. Our next generation GPU has been in development for quite some time as you'd imagine and is something that's near completion.
In the video game makret historically companies want to have their own flavor so it's got to be somewhat customized for that application so will that be the case here and if so can you give us a sense as to how many engineers it takes and what sort of time frame to actually complete the design and get it to the point where you're prototyping and getting ready for pre-production?
It is going to be a custom GPU, and its going to be a custom GPU to architect it and optimizie to work specifically with the Cell microprocessor. Ultimately if you think about the architecture inside this GPU...it probably was a......this next generation architecture took several hundred people several years to go build, but this specific implementation of that architecture should take about 50 engineers and is something we're running full throttle on and I have every expectation that we will be able to see final production silicon later on in the year.
Megadrive1988
17-Dec-2004, 05:10
ot:
heh, listening to it again from the 41 min mark, I heard the questioner say:
"you were doing something for Sega which didn't actually go anywhere"
that was the NV2. worked on during the 'dark' period between NV1 in Diamond Edge 3D cards and Riva128
JH didn't say anything about that :)
From, how the things are going, Sony seems to have a locked down on the spec of PS3 for quite sometimes now. Maybe they'll launch as soon as Xbox 2 launched.
The NVPS3 will be a graphics processor built from the ground up(using NV50 tech) to operate and compliment the Broadband Engine(The PS3 CPU).
To think that SCE engineers have no say, or are not working with nVIDIA engineers on this thing is insane.
And that's all I have to say.
Panajev2001a
17-Dec-2004, 06:20
The NVPS3 will be a graphics processor built from the ground up(using NV50 tech) to operate and compliment the Broadband Engine(The PS3 CPU).
It was not built from the ground-up (or from scratch) to compliment CELL based MPUs: if Sony went with the other GPU, nVIDIA would have still used the architecture in PCs.
The NV50's architecture has been completed after several years of work with hudreds of engineers working very hard on it.
Now, we are talking about NV5X and more specifically NV5X-PS3 (if we want to make a nick-name for it).
We know that nVIDIA has assigned 50 engineers and a non small budget to work with SCE's engineers to customize the NV5X architecture to work optimally in the CELL+Redwood+XDR based PlayStation 3.
It will not be NV6X, NV5X-PS3 will be based on the NV50's architecture.
We know that nVIDIA has assigned 50 engineers and a non small budget to work with SCE's engineers to customize the NV5X architecture to work optimally in the CELL+Redwood+XDR based PlayStation 3.
Exactly what I wanted to say, "ground up" wasn't the best choice of words on my part.
What I'm trying to say is that this thing isn't just a nvidia GPU thrown in the PS3, it's custom, designed to work with the Cell based Broadband Engine.
Panajev2001a
17-Dec-2004, 06:28
We know that nVIDIA has assigned 50 engineers and a non small budget to work with SCE's engineers to customize the NV5X architecture to work optimally in the CELL+Redwood+XDR based PlayStation 3.
Exactly what I wanted to say, "ground up" wasn't the best choice of words on my part.
What I'm trying to say is that this thing isn't just a nvidia GPU thrown in the PS3, it's custom, designed to work with the Cell based Broadband Engine.
Of course, nobody thought it was something just thrown-in: the NV2A was not just thrown-in either and I am sure SCE has all the intentions to get the best GPU possible for PlayStation 3.
So we've come to a full circle.
The thing is going to be rediculously fast, I look foward for what the future brings.
So we've come to a full circle.
The thing is going to be rediculously fast, I look foward for what the future brings.
Can we expect nV5x ~ nV4x SLI [+ new features + GHz on a core] putting it simply? Or are we expecting much much more?
Megadrive1988
17-Dec-2004, 07:02
well now my dream is this: take NV50/NV5X architecture and parallal-ize it massively. obviously a single desktop NV50 GPU is (or would have been) already a parallel design like all modern GPUs. but for PS3 GPU, I'd like a multi-NV5X core super-GPU.
(I might go off-topic slightly)
like how Graphics Synthesizer has 16 parallel pixel pipes or pixel engines. each of those 16 pixel pipes/engines is thought to be an enhanced version of PS1's graphics chip with some additional features. (z-buffer, bilinear filtering, alpha blending, a poor mip-mapping etc. and no CLIPPING!)
now with a 65 nm process, put in 4,8,16 or...(back to within reason) even just two NV5X cores. perhaps minus the vertex shaders since the Cell-based Broadband Engine CPU can take care of all the vertex processing.
another way of looking at it: PS3 GPU, I hope, is like a 64 or 128 pixel pipeline NV5X based GPU with lots of eDRAM. whereas a desktop NV5X GPU might have had 32 pipes.
in other words, NV5X-SLI on a chip. plus eDRAM.
the NV40 was 222 million transistors on 13nm. the NV50 desktop CPU might have been 400 to 500 million transistors on 90nm (Nvidia roughly doubles transistors every real architecture change) the PS3's NV5X based GPU could have a billion or more transistors on 65nm. fill it up with parallel implementation of NV5X plus eDRAM, all optimized for Cell-Broadband Engine, Redwood interconnects, XDR-DRAM, etc.
.... just thinking out loud.....I know there's ALOT i do not understand and Nvidia-Sony are having to work within the limits of process technology & cost among other things.
So we've come to a full circle.
The thing is going to be rediculously fast, I look foward for what the future brings.
Can we expect nV5x ~ nV4x SLI [+ new features + GHz on a core] putting it simply? Or are we expecting much much more?
ghz on core would 2.3x the clock speed of nvidia's currently top of the line process pushing low yielding gpu .
I think 1 ghz would be the most we would see from it if it launches 2006
Panajev2001a
17-Dec-2004, 07:14
I do not see the need of 32-128 pipe-lines.... that's way too many for now: fill-rate is important, but more important seems to be Shader ops execution performance/throughput.
Better to have 16 pipelines and X Shading+Texture ALUs per pipeline or 64 pipelines and X/4 Shading+Texture ALUs per pipe-line ?
I'd rather have the 16 pipes version until we can fully move to a REYES based solution in the long-term future.
Can we expect nV5x ~ nV4x SLI [+ new features + GHz on a core] putting it simply? Or are we expecting much much more?
ghz on core would 2.3x the clock speed of nvidia's currently top of the line process pushing low yielding gpu .
I think 1 ghz would be the most we would see from it if it launches 2006
2.3x the current top-of-the-line due to current fabbing and process size, but when fabbing at 65nm/SOI/Lo-K/etc... they might have more room to up the GHz.
If the Cell BE ~ 4.6+ GHz, then the GPU will likely be 1/4 (worst) ~ 1/2 (best) that speed.
Megadrive1988
17-Dec-2004, 07:27
Panajev:
I do not see the need of 32-128 pipe-lines.... that's way too many for now: fill-rate is important, but more important seems to be Shader ops execution performance/throughput.
Better to have 16 pipelines and X Shading+Texture ALUs per pipeline or 64 pipelines and X/4 Shading+Texture ALUs per pipe-line ?
I'd rather have the 16 pipes version until we can fully move to a REYES based solution in the long-term future.
so a REYES like rendering system might not be possible or might not happen until PS4 or PS5? I'd imagine that's when true raytracing & GI *might* be possible. with PS5, I wonder if silicon chips will still be used. will optical processors be feasible by then? I'm thinking PS4 might be the last Playstation to use normal silicon chips. and there needs something (optical processors?) to fill the gap between silicon chips (PS3 probably PS4) and biochip processors that Sony says might be used in PS6 or PS7 :shock:
PS4 predictions: 25 nm process or smaller. still silicon chips. many tflops. some raytracing. several GBs of external memory @ several hundred GB/sec bandwidth or even around 1 TB/sec bandwidth. plus several hundred MB of eDRAM @ many TB/sec bandwidth. visuals like current best CG intro/cut scenes in videogames. (i.e. Onimusha 3)
PS3 will probably realize the lofty goals of PS2. that is, PS1 CG in games. then the PS4 will probably realize the lofty goals of PS3.
actually PS2 did surpass Ken K's public predictions that he made in 1995 which were 10 million polygons achieved on .25 micron (250 nm) chips.
*puts always his crystal ball which cannot see too clearly*
.... just thinking out loud.....
Yes. Just from the public information in the press release:
"The powerful custom GPU will be the graphics and image processing foundation for a broad range of applications from computer entertainment to broadband applications. The agreement will encompass future Sony digital consumer electronics products."
Sony's plans for Cell based products are vast - most likely emcompassing the entire range of consumer electronics from low to very high in cost. The notion that Sony is buying a pc video card and 'tweaking' it for the PS3 is absurd, although I can certainly see how some would need to see it that way.
Just like much of the Cell patent talk was overly focused on just the PS3, the Sony-NV relationship talk is overly focused on just the PS3's GPU. Just like the Cell architecture itself, the graphics technology Sony is designing for NV to implement will likewise need to be at its core fundementally scalable for the wide range of products it will be used in.
Panajev2001a
17-Dec-2004, 07:44
I do not think nVIDIA's next-generation "architecture" is so "un-scalable" by design ;).
Megadrive1988
17-Dec-2004, 07:46
Tuttle, I do not disagree with what you wrote there
Megadrive1988
17-Dec-2004, 07:49
I do not think nVIDIA's next-generation "architecture" is so "un-scalable" by design
exactly why I am hoping for a scaled up implementation of nVIDIA's next-gen architecture for PS3 GPU, for a console that'll have to last 5-7 years, until early next decade when PS4 is likely to arrive (2011-2013)
.... just thinking out loud.....
Yes. Just from the public information in the press release:
"The powerful custom GPU will be the graphics and image processing foundation for a broad range of applications from computer entertainment to broadband applications. The agreement will encompass future Sony digital consumer electronics products."
Sony's plans for Cell based products are vast - most likely emcompassing the entire range of consumer electronics from low to very high in cost. The notion that Sony is buying a pc video card and 'tweaking' it for the PS3 is absurd, although I can certainly see how some would need to see it that way.
Just like much of the Cell patent talk was overly focused on just the PS3, the Sony-NV relationship talk is overly focused on just the PS3's GPU. Just like the Cell architecture itself, the graphics technology Sony is designing for NV to implement will likewise need to be at its core fundementally scalable for the wide range of products it will be used in.
Agreed, and in summary:
Sony/nVidia PS3 GPU deal != MS/nVidia Xbox GPU deal [not including the fabbing component]
It should not have been compared in the 1st place.
The XGPU would have been [alot] easier to develop for the Xbox given the over architecture is x86 based/centric (i.e. CPU, GPU and Chipset).
With the PS3, we are talking an entirely different type of fish. It’s not even in the same architectural paradigm.
And to think Sony has little or no input into the design of the M/V/GPU is ludicrous.
Also given it’s a new design and never done before (which probably rules out the NV4x SLI idea), I will not be surprised by the end of this how Cell like the GPU will end up to be, even though not Cell based (base – building upon).
joe emo
17-Dec-2004, 08:21
The XGPU would have been [alot] easier to develop for the Xbox given the over architecture is x86 based/centric (i.e. CPU, GPU and Chipset).
That's a fairly flawed assumption, given Apple uses NVIDIA boards in G5s. I would estimated the PowerPC architecture has more in common with cell than x86 chips do; and for all intents and purposes the NV GPUs used in Apple products are the same exact chips used for PC boards.
Megadrive1988
17-Dec-2004, 08:31
found this piece on xbitlabs
http://www.xbitlabs.com/articles/editorial/display/nvidia-editors04_3.html
System-on-a-chip is the future of computing. Right now GeForce 6 has 250million transistors. Next generation will be from 0.5-1 billion transistors. These systems are going to come one day, the only question is how they become relevant in the market.
Image processing is also picking up as one of the most popular tasks nowadays. The media processor business is also very important for the consumer electronics. And already now we see a lot of steps undertaken in this direction. On Tuesday night, for instance, next generation Sony Playstation was announced. So far NVIDIA refused to disclose any more details, but we know that it is going to be a very powerful platform. NVIDIA is building a custom GPU on the architecture beyond GeForce 6 for Playstation 3. Graphics technology is certainly one of the most important things NVIDIA is working on regarding this platform, but there is the whole variety of other things including software tools and other technologies, that are also worth paying special attention to. You can also find more info about this upcoming gaming console in this news story of ours.
other than the fact that 'GeForce 6' (NV40) has "only" 222 million transistors, not 250 million, I agree that nextgen Nvidia chips should see 0.5 to 1 billion transistors. hmmm nextgen desktop GPU with 0.5 billion and PS3 GPU at 1 billion sound about right to me. a good chunk of that likely 1 billion trannies in PS3 GPU is gonna be eDRAM.
The XGPU would have been [alot] easier to develop for the Xbox given the over architecture is x86 based/centric (i.e. CPU, GPU and Chipset).
That's a fairly flawed assumption, given Apple uses NVIDIA boards in G5s. I would estimated the PowerPC architecture has more in common with cell than x86 chips do; and for all intents and purposes the NV GPUs used in Apple products are the same exact chips used for PC boards.
I am aware of that... was spending hours looking at the HD Cinema Display with the G5. For the 30" display it requires the GeForce 6800 Ultra DDL version card, but what they both have in common (PC and G5) is AGP (the glue between CPU and GPU). Unless the Cell arch includes AGP, then it’s less in common then your comments will imply.
Further more, the integration between the CPU, GPU and Chipset in a console is a lot tighter compared to the PC/Mac (or any other System using AGP/PCI – e.g Sun, HP) systems.
For example at the time of release, the Xbox had the CPU & GPU sharing the RAM for a top end graphics solution. Don't even compare this example with today's chipsets which includes a low-end gpu using main memory. :lol:
pahcman
17-Dec-2004, 09:06
Sony/nVidia PS3 GPU deal != MS/nVidia Xbox GPU deal
It should not have been compared in the 1st place.
That is obvious.
We can extend the list even further Sony/nVidia PS3 GPU deal != MS/nVidia Xbox GPU != MS/ATi Xenon GPU != Nintendo/ATi Revolution GPU != Nintendn/ATi GC GPU != Sega/ImgTech DC GPU != Tiger/Nvidia Gizmondo GPU...
The next gneration game console will consist of 2 devices that Sony has talked about. One of them is the Cell microprocessor....the second device that will be a companion device to it is the graphics processor and the graphics processor will also be the image processor and there's all kinds of exciting features that will(?) come out. It's based on the next gneration GPU technology and our expectation is to try and put it into production this year. Our next generation GPU has been in development for quite some time as you'd imagine and is something that's near completion.
It is going to be a custom GPU, and its going to be a custom GPU to architect it and optimizie to work specifically with the Cell microprocessor. Ultimately if you think about the architecture inside this GPU...it probably was a......this next generation architecture took several hundred people several years to go build, but this specific implementation of that architecture should take about 50 engineers and is something we're running full throttle on and I have every expectation that we will be able to see final production silicon later on in the year.
Sounds from this interview, PS3 is not entirely Sony Cell designed (why a break of 2? this next gen architecture is for peecee right? what is nvidia usual defination for "next gen architecture"? i dont think they will spend years and manpower on cell architecture, nvidia intermediate market is still peecee right? that why the 50 is for cell version of NVXX right?), and so peecee Nvidia have a much larger say in the PS3 design. Right Vince, tuttle?
What are the all kind of exciting features he said for the companion device? Is it restricted to new graphics or will PS3 have nvidia media processor like MCP2? It is good if so. I am using an Xbox and nForce2 PC. Love both of them.
Sounds from this interview, PS3 is not entirely Sony Cell designed(why a break of 2?), and peecee Nvidia have a much larger say in the PS3 design. Right Vince, tuttle?
From this interview, I still can't tell how the NV GPU going to be paired with Cell, JH still very vague. Because this time NV GPU won't house the main memory controller, Cell will have that part instead.
[quote=Jov]Can we expect nV5x ~ nV4x SLI [+ new features + GHz on a core] putting it simply? Or are we expecting much much more?
ghz on core would 2.3x the clock speed of nvidia's currently top of the line process pushing low yielding gpu .
I think 1 ghz would be the most we would see from it if it launches 2006
2.3x the current top-of-the-line due to current fabbing and process size, but when fabbing at 65nm/SOI/Lo-K/etc... they might have more room to up the GHz.
If the Cell BE ~ 4.6+ GHz, then the GPU will likely be 1/4 (worst) ~ 1/2 (best) that speed.[/quote
So your telling me they are going to go from 400mhz to possible 2.3 ghz while making the gpu even bigger than it currently is all the while going from a drop of 110nm to 65nm .
Find it hard to believe.
cybamerc
17-Dec-2004, 10:33
I'm no engineer but I reckon that switching to Cell tech is more than a simple modification and would require more than 50 engineers. Surely none of you think that the PC part will be Cell based?
The PS3 GPU is obviously going to be a customized part in the same sense that NV2A was.
london-boy
17-Dec-2004, 10:44
A very very small part of me tells me that the smaller involvement Sony has in the GPU, the better things will be. I know i shouldn't, but u know whent you have that little voice in your brain....
Still, i think they'll butt in big time, if only to make sure Cell and the GPU speak the same language.
regarding EDram: since Sony is goning to fab the gpu, nvidia is obviously designing it using their design libraries. Since integrating Dram into a logic IC is mainly a manufacturing challenge, i don't see how that should give nvidia a major sweat. Having two memory controllers (BE & GPU) with corresponding distinct external memory banks is neither space efficient (redundant data) nor cheap as it will significantly increase the costs for manufacturing the system board during the whole product lifecycle (guess that is why no one went that way this generation). BTW how are Nvidia's / ATI's / etc. approaches to designing GPUs x86 centric?
BTW how are Nvidia's / ATI's / etc. approaches to designing GPUs x86 centric?
Or (legacy) DirectX-centric?
london-boy
17-Dec-2004, 11:08
NVIDIA and ATI GPUs seem to be quite x86 independant, seeing how video cards for Macs are pretty much the same as the ones for PC.
So i guess it doesn't matter, to a certain extent, what CPU they are coupled with, as long as the software side of things works properly.
Will be interesting to see how CELL+NVGPU will turn out to be.
so can somebody tell me, how does NV50x compare to the ATI tech Msoft are using in xenon?
london-boy
17-Dec-2004, 11:13
so can somebody tell me, how does NV50x compare to the ATI tech Msoft are using in xenon?
Would be nice to know. Not sure anyone knows anything though.
Oh, i think you need to fix that link in your sig. :D
eh yeah... but i cant seem to fix it.
london-boy
17-Dec-2004, 11:16
eh yeah... but i cant seem to fix it.
Copy this (quote and copy the whole thing):
Done?
Oh i took the liberty to fix the spelling too :oops: :wink:
As it seems Nvidia are trying to get their next gen GPU in production this year, doesn't that mean that the ATI GPU for the Xbox2 are going to fairly evenly matched? So in terms of graphical output there can't be that much difference between each even if the Cell processor is say 3-4 times more powerful than that in the Xbox2? I don't know if these question are right but it seems to me that if your graphics card can only do X, whats the worth of having Y loads of CPU, won't that just be used for AI and phyics like calculations and if so surely the Xbox2 cpu will be powerful enough to preform similar functions to that of whatever the Cell processor is. Or is there a lot more to it?
BTW how are Nvidia's / ATI's / etc. approaches to designing GPUs x86 centric?
x86-centric is, IMHO, an aphorism of sorts for the principles behind the generalized PC architecture that has evolved due to the multi-vendor landscape and (in)ability to create usable and farsignted standard between each component.
I suppose a distinction can easily be drawn between the current PC architecture and programming model and that put forth in Keith Diefendorff and Pradeep Dubey's seminal paper (http://portal.acm.org/citation.cfm?id=620800) on the necessary shift in processor design to conform to the emergence of dynamic workloads. This setiment was most recently echoed by Peter Hofstee during his presentation in San Fransisco and will likely be raised again at ISSCC.
Many will undoubtedly reply back that PCI-Express will allow for this and point towards X2 -- especially those who are working on it actively without similar knowledge of PS3 -- but this isn't analogous and X2's steps are small and evolutionary, which ERP once agreed to.
'm kinda curious to see what vince is going to say regarding this info.
Concerning my comment on this Qroach, while I'd love to give you and your twin Johnny more comments to take out of context and use as a sig; I'll wait for tangible information before jumping on the bandwagon that many here so love to do: Cell != PS3, Cell != 65nm, Cell != 2005, but 2007, Cell =! 4GHz, et al. I like Joe, but he jumped the gun (maybe pulled the trigger ;)) on his comments quite a bit if you listen to the actual call. Besides, I'm sitting pretty so far in the grand scheme of things, there's no need to comment.
What is most interesting to me is how Jen-Hsun commented again that this will be used in all Sony CE devices, which is what Cell is intended and was designed for -- We have independent confirmation from both Sony group, IBM and Toshiba on this. So, how would one make these statements logically compatable?
PS. While I'm without a doubt an ass (no pun intended), can we ditch the link to Deadmeat's picture? Thanks.
Dave Baumann
17-Dec-2004, 11:44
x86-centric is, IMHO, an aphorism of sorts for the principles behind the generalized PC architecture that has evolved due to the multi-vendor landscape and (in)ability to create usable and farsignted standards between each component.
Funny that these unusable and short sighted standards have resulted in the most profilic form of (high performance) computing device on the planet with little sign of abatement. I guess all the industry drivers must be smacking their foreheads at their stupidity (but at least they have bundles of cash to fall back on, eh?).
:?
Funny that these unusable and short sighted standards have resulted in the most profilic form of (high performance) computing device on the planet with little sign of abatement. I guess all the industry drivers must be smacking their foreheads at their stupidity (but at least they have bundles of cash to fall back on, eh?).
Dave, before commenting in typical knee-jerk fashion to whatever I say, take a few seconds and mull over what I posted:
A closed-set/design will, almost intrinsically, surpass the open PC paradigm given invarient computation components and a variable integration and communications model.
The "most prolific" comment is a Red Herring as you attempt to disguise the actual topic concerning relative preformance with sales and prominence. Which, in fact, is an utterly pathetic comment as it reinforces what I previously stated if you logically think about it and what it entails. Of course it'll be the most prolific form of computation on the planet because it's an open model, which allows for the free-market to partake in the system and sets the foundation for the OEM PC buisness.
Yet, in accordance with what I stated in my post, systems that utilize this model will lose out to custom systems that are closed-sets in relative preformance: Witness the Top-10 super computers and the dominence of closed sets such as IBM's BlueGene, SGI and NEC's vector systems.
A prime example is the XBox which used off-the-shelf components (relatively speaking) and due to it being a closed-set, allowed for superior integration and all that is derived from this.
So, next time, try to respond to what was posted. I'd appreciate it.
When Diefendorff and Dubey's paper was written, the desktop GPU revolution hadn't started.
They picked one route to multi-media integration, the world picked another. Even Sony agree, a specialised vector processor (a GPU) will be 'standard' on all media centric devices for the foreseeable future.
When Diefendorff and Dubey's paper was written, the desktop GPU revolution hadn't started.
They picked one route to multi-media integration, the world picked another. Even Sony agree, a specialised vector processor (a GPU) will be 'standard' on all media centric devices for the foreseeable future.
I think the key problem that you've overlooked is the phrase "the desktop."
Let me ask you think, if the PC landscape wasn't populated by multiple vendors that compete in graphics and image processing and several others who compete in CPU design and yet others who design interconnections and the storage hierarchies and even more who deal with all the components that go into a contemporary PC -- but, instead you had, say, three monolithic companies that produced the entire system and competed against eachother for sales in the holistic view of a PC, would the industry have went on the same road that it has?
The current paradigm is fabulous for the PC, no question about it. It's clearly hovering around the point of equilibrium for greatest net benefits per investment, I'm not saying it's not. The current affordable PC wouldn't exist without it. But, that doesn't mean for a second that this equilibrium point is in the region of greatest relative preformance.
That's an assumption that you make in your argument, tacitly or not, and one which I clearly and emphatically refute.
Dave Baumann
17-Dec-2004, 12:14
The current paradigm is fabulous for the PC, no question about it.
A prime example is the XBox which used off-the-shelf components (relatively speaking) and due to it being a closed-set, allowed for superior integration and all that is derived from this.
You appear to agree that that same paradigm for PC hardware design can be applied elsewhere and work equally as well.
The current paradigm is fabulous for the PC, no question about it.
A prime example is the XBox which used off-the-shelf components (relatively speaking) and due to it being a closed-set, allowed for superior integration and all that is derived from this.
You appear to agree that that same paradigm for PC hardware design can be applied elsewhere and work equally as well.
Yeah, it worked so well in the console market that it only took a few billion dollars in losses to keep it afloat in head to head competition with older but dedicated hardware.
Three cheers for the mighty x86 pc hardware makers!
The current paradigm is fabulous for the PC, no question about it.
A prime example is the XBox which used off-the-shelf components (relatively speaking) and due to it being a closed-set, allowed for superior integration and all that is derived from this.
You appear to agree that that same paradigm for PC hardware design can be applied elsewhere and work equally as well.
You see Dave, this is what happens when you selectively quotes me as if you just want to diuspute something. If you'd read my entire post as is intended, you'd see that I'm quite clearly stating that the current PC paradigm is fabulous for the niche it fills, namely I highlighted the cost to consumers and achieving good equilibrium in the region for greatest net benefit per investment.
I then went on to state that if the world didn't see the emergence of the PC paradigm, which seems inevitable with free market dynamics and competition driving price reduction, but instead we were concerned with sheerly preformance, that there is another region on the hypothetical landscape that will yeild higher preformance. The XBox shows this principle to be true by following the conditions I layed out before and will list below. It also refutes your dumb post from earlier defending the current model on the grounds that it's "prolific" (totally unrelated to the argument at hand), by showing that the same components when taken out of the competitive PC paradigm will outpreform it. As I already stated:
A closed-set/design will, almost intrinsically, surpass the open PC paradigm given invarient computation components and a variable integration and communications model.
This argument can then be extended, quite easily, to the entire system: the components within it and the processing model when you close the set and make it invarient -- which is what Diefendorff and Dubey did in their paper. This is the guiding principle that I believe in and, appearently, so does STI. Understand? Or are you just going to chop it up and keep going in circles.
EDIT: Formatting.
london-boy
17-Dec-2004, 12:29
The current paradigm is fabulous for the PC, no question about it.
A prime example is the XBox which used off-the-shelf components (relatively speaking) and due to it being a closed-set, allowed for superior integration and all that is derived from this.
You appear to agree that that same paradigm for PC hardware design can be applied elsewhere and work equally as well.
Yeah, it worked so well in the console market that it only took a few billion dollars in losses to keep it afloat in head to head competition with older but dedicated hardware.
Three cheers for the mighty x86 pc hardware makers!
It's not like "the hardware" won. Ps2 didn't succeed over the Xbox because of the EE+GS.
Let me ask you think, if the PC landscape wasn't populated by multiple vendors that compete in graphics and image processing and several others who compete in CPU design and yet others who design interconnections and the storage hierarchies and even more who deal with all the components that go into a contemporary PC -- but, instead you had, say, three monolithic companies that produced the entire system and competed against eachother for sales in the holistic view of a PC, would the industry have went on the same road that it has?
But they don't exist, because no single company can compete in RnD versus the whole PC business (with 10's of billions of dollars RnD each year). The monopoly's have to external PC company to get the latest/greatest tech.
If you want a fast moving industry you basically require the way the PC business works.
Thats why the PC model, applies through-out all IT today. Mobile phones, super-computers, consoles are all built according to the PC model. You (as the platform holder) mix and match N company product and produce the product that fits your budget and target performance.
Its just the company wide model of plugging in PC express cards.
The current paradigm is fabulous for the PC, no question about it. It's clearly hovering around the point of equilibrium for greatest net benefits per investment, I'm not saying it's not. The current affordable PC wouldn't exist without it. But, that doesn't mean for a second that this equilibrium point is in the region of greatest relative preformance.
That's an assumption that you make in your argument, tacitly or not, and one which I clearly and emphatically refute.
We will have to agree to disagree, I have no doubt the PC market pushs performance like no other. Else why is Sony using an IBM designed CPU (based on work of BlueGene and PowerPC both based on the PC model) and a GPU by NVIDIA...
Of course a closed design can win the short term but lose in the long term. Indeed STI know this, by trying to get Cell into into its own market economy model.
It's not like "the hardware" won. Ps2 didn't succeed over the Xbox because of the EE+GS.
Don't be silly. That's just a strawman.
The superior PS2 hardware design was just one of many reasons why Sony destroyed MS in the marketplace.
PC-Engine
17-Dec-2004, 12:42
It's not like "the hardware" won. Ps2 didn't succeed over the Xbox because of the EE+GS.
Exactly, not to mention MS didn't lose $3 billion on hardware. Also don't forget that Xbox had a built-in HDD that was basically thrown in for free.
The superior PS2 hardware design was just one of many reasons why Sony destroyed MS in the marketplace.
What hardware in PS2 is superior to Xbox?
london-boy
17-Dec-2004, 12:43
It's not like "the hardware" won. Ps2 didn't succeed over the Xbox because of the EE+GS.
Don't be silly. That's just a strawman.
The superior PS2 hardware design was just one of many reasons why Sony destroyed MS in the marketplace.
Errrr... Ok then...? :? If you say so... I can feel the heat already.
Guden Oden
17-Dec-2004, 12:48
The superior PS2 hardware design was just one of many reasons why Sony destroyed MS in the marketplace.
You're not serious, right? Please don't troll threads on purpose, thanks...
But they don't exist, because no single company can compete in RnD versus the whole PC business (with 10's of billions of dollars RnD each year). The monopoly's have to external PC company to get the latest/greatest tech.
Please don't evade my questions, I know the answer and this is getting old. I'm not disputing the benefits on the current paradigm, I already stated it's fabulous for the PC marketplace. No argument there, I fully support your employment.
But, this doesn't mean that it's the best paradigm to extract the greastest preformance. In fact, it's clearly not and there is objective evidence in the XBox (vis-a-vis the PC) and high-end Supercomputing to support this proposition. It's nigh impossible for you to disagree to this on a theoretical ground.
Secondly, it is possible to closely approximate the theoretical. I do believe STI is doing it and the console platform is the perfect catalyst as it is invarient and is a closed-set.
Thats why the PC model, applies through-out all IT today. Mobile phones, super-computers, consoles are all built according to the PC model. You (as the platform holder) mix and match N company product and produce the product that fits your budget and target performance.
The top 3 Supercomputers are all customly built using custom ASICs designed for a specific closed-set enviroment. Consoles aren't built according to the PC paradigm as it's a closed-set since there is no inhomogeneity when you view the platform cross-sectionally.
We will have to agree to disagree, I have no doubt the PC market pushs performance like no other. Else why is Sony using an IBM designed CPU (based on work of BlueGene and PowerPC both based on the PC model) and a GPU by NVIDIA...
If the PC model was superior, why aren't Sony and Microsoft analogous to Dell and HP-Compaq? Your argument is logically inconistent as there are differences and these are demonstrated by Sony spending several billion dollars for a clean-sheet architecture and Microsoft doing the same with IBM to a lesser extent.
Of course a closed design can win the short term but lose in the long term. Indeed STI know this, by trying to get Cell into into its own market economy model.
So, you spend most of the post dissagreeing and then undermine it all by agreeing with me in the short-term. Which is all that's relevent to this discussion? *confused*
rabidrabbit
17-Dec-2004, 12:49
It's not like "the hardware" won. Ps2 didn't succeed over the Xbox because of the EE+GS.
Exactly, not to mention MS didn't lose $3 billion on hardware. Also don't forget that Xbox had a built-in HDD that was basically thrown in for free.
I was under the impression that Sony is no longer losing money on hardware :?
How much Sony are still on minus?
And I also thought I read somewhere that Microsoft would not be able to get the xbox profitable this gen :?
Dave Baumann
17-Dec-2004, 12:50
The XBox shows this is true, and refutes your dumb earlier post defending the current model on the grounds that it's prolific (totally unrelated to the argument at hand), by showing that the same components when taken out of the competitive PC paradigm will outpreform it.
Its not dumb Vince, your statement that the PC vendors have an "(in)ability to create usable and farsignted standard between each component" is clearly incorrect as you have proved yourself by the very fact that it can be applied beyond the scope of the PC environment and used to a full context. I don't take issue with the idea can equally be applied to hardware as well, but I just don't agree with some of the things you speak about - so far the PC has been the longest lasting platform available purely on the basis that it has also had to be one of the most adaptible and quick to adopt and even define and push new technologies and standards.
The curious element about your last point is that the Cell paradigm is that it is designed to be used in a multitude of devices so the hardware contructs it employs are going to have to be less specific by design, in some areas, than the pure device its going to be utilised in.
The superior PS2 hardware design was just one of many reasons why Sony destroyed MS in the marketplace.
You're not serious, right? Please don't troll threads on purpose, thanks...
Go find another thread to play in.
PC-Engine
17-Dec-2004, 12:57
It's not like "the hardware" won. Ps2 didn't succeed over the Xbox because of the EE+GS.
Exactly, not to mention MS didn't lose $3 billion on hardware. Also don't forget that Xbox had a built-in HDD that was basically thrown in for free.
I was under the impression that Sony is no longer losing money on hardware :?
How much Sony are still on minus?
And I also thought I read somewhere that Microsoft would not be able to get the xbox profitable this gen :?
That's because MS decided not to shrink the motherboard, cpu, gpu, down in size. If they did it would be cheaper to manufacture. Again with the HDD built-in and the technology being newer, it would take more time for Xbox hardware to be profitable. Don't forget that MS lowered the price to $150 at a sooner time than SONY did with PS2 relative to their respectivel console lifetime. It's not as simple as SONY makes better cheaper hardware while MS can't. :lol:
Dave Baumann
17-Dec-2004, 12:59
That's because MS decided not to shrink the motherboard, cpu, gpu, down in size. If they did it would be cheaper to manufacture.
The components aren't theirs, so its would probably take more money to request the component makers to do this than it would just to put up with the loss.
london-boy
17-Dec-2004, 13:00
I think that the main difference is that for some reason, God has stated that there should be a new cutting-edge PC graphics card out every 6-8 months. With a completely new architecture every 12-18 months.
PC therefore can afford to take smaller steps, one at a time. Whereas consoles come out at ~5 years intervals, and when they do, technology has progressed so much that the jump makes people go WOW, as they compare those to the last generation of consoles. those same people might, at the same time, look at PC and think that PC CPU and GPU manufactueres might lack "foresightness". I don't think that's the point. PC manufacturers can afford to take smaller steps, maybe delving a little bit deeper in what they're doing at the moment, cause they know in 6 months time they'll have a new fresh product out of the door.
With consoles, companies take big risks, trying to figure out the best way to make that console last for the next 5 years without looking embarrassingly outdated at the end of their lifetime, which is just tough if you ask me. That might look like "pioneering" in some people's eyes.
I just say... It's all about the gameplay :twisted:
pahcman
17-Dec-2004, 13:00
Hi Vince,
You sound like a smart guy who thinks abreast far and wide.
I find there is a problem for us mere readers to communicate with your excitement with Cell. Instead of talking numbers, papers, theories, revolution, deadend peecee etc, perhaps you can share with us your envisioned type of exciting new entertainment that this Cell project will exclusively create. We can start with the PS3. ;)
This way its better for all, now we can visualize something common and easier. Everyone agree?
Titanio
17-Dec-2004, 13:01
As it seems Nvidia are trying to get their next gen GPU in production this year, doesn't that mean that the ATI GPU for the Xbox2 are going to fairly evenly matched?
NVidia won't have it in production this year, they're looking to have final silicon towards the end of next year. So the answer to your question is possibly/probably not, but there are other factors asides from time.
Of course a closed design can win the short term but lose in the long term. Indeed STI know this, by trying to get Cell into into its own market economy model.
So, you spend most of the post dissagreeing and then undermine it all by agreeing with me in the short-term. Which is all that's relevent to this discussion? *confused*
No Sony is embracing a version of the PC architecture.
Cell is hoping to become the next x86. A standard that works o.k. in most cases.
Moving from a custom designed processor for every generation to using a basically off-the-shelf part thats compatible with the lots of already written software.
All thats really new is fitting more advanced vector units than usual for a general processor.
What I'm saying is that there isn't yet an inherit advantage of Sony's stragegy (except to Sony) than the normal IT industry (which uses x86, PowerPC or ARMs as a rule). Sony is hoping for one processor to rule them all, but I don't see any reason why Cell will do any better than any other plan for global domination.
x86 isn't getting SPUs because its doesn't need them, it will use seperate GPUs for when you need lots of floating point maths.
I fail to see why you see Cell as so revolutionary, its just a CPU with a few independent vector units. Of course Sony will tell you its a revolution, just the same as every new CPU architecture is "revolutionary". Alpha, Itanium, POWER, etc. Whats Cell doing that these didn't, its not money Itanium has had a fortune spend on it.
So why didn't x86 die to Itanium?
Because Intel misguessed the future... They guessed that x86 would run out of steam, they guessed wrong. I personally beleive that KK is complelety wrong when he says PC architecture has run out of steam...
Indeed its likely to first cheap TFLOP will be in a PC (thanks the GPUs) not a PS3....
Vince could you please give your definitions for "PC paradigm" and "Cell paradigm".
IMO in all the various engineering disciplines open standards and modular architectures have quite consistently outperformed (performance/costs) closed-set approaches (e.g. computer hardware, software, cars, avation, etc..). All three current generation consoles are (to me) just variations of the pc architecture (whether they are using MIPS,x86 or PowerPC ISAs). The same holds for mobiles and other embedded devices. Requiering hardware equality instead of software compatability (platform compatability) seems to be a step back to the 70s.
I suppose I should mention that I quite like Cell/PS3 its just I don't get the argument that the future is Cell. That its inheritantly better than the other architectures. Its got some cool ideas, that look fairly good for a games console, but as a Sony employee told me "Its just a good CPU".
To me its just a high performance float version of ARM. Fast, cheap and cool for techies but not really making much of a difference to the rest of the world outside embedded systems..
while I'd love to give you and your twin Johnny more comments to take out of context and use as a sig;
It's a quote guy, I really don't see how it's taken out of context. I just thought it was funny how conflicts with your more recent arguments.
Besides, I'm sitting pretty so far in the grand scheme of things, there's no need to comment.
Riiight, you have a selective memory my friend. ;) that's fine, I care not.
What is most interesting to me is how Jen-Hsun commented again that this will be used in all Sony CE devices, which is what Cell is intended and was designed for -- We have independent confirmation from both Sony group, IBM and Toshiba on this. So, how would one make these statements logically compatable?
I'm not really suprised by this, as sony intends to leverage what they are doing in many projects, but once again I really don't care what they are doing outside of the PS3.
PS. While I'm without a doubt an ass (no pun intended), can we ditch the link to Deadmeat's picture? Thanks.
What link are you talking about??
Its not dumb Vince, your statement that the PC vendors have an "(in)ability to create usable and farsignted standard between each component" is clearly incorrect as you have proved yourself by the very fact that it can be applied beyond the scope of the PC environment and used to a full context.
No, you just don't understand what's being stated. Very simple, and you're going to do all the thinking:
I have GPU Gn, CPU Cn, Storage Sn, which of the following would more closely approximate the eq. point of greastest preformance:
[list:4fc65be220] Closed System A: [GPU G1, CPU C1, Storage S1] with custom interconnections and system optimization; or:
Open System B: Random combination of [GPU Gn, CPU Cn, Storage Sn] |n=1...1E5[/list:u:4fc65be220]
The PC will allways lose out to a closed-set enviroment, especially in the PC paradigm we have in which there is little thinking of the system on a holistic level. Each vender is concerned primarily with their given component and there is little cooperative work. Hell, look at the history of PCI, AGP and - finally - PCI-Express. AGP texturing anyone?
The curious element about your last point is that the Cell paradigm is that it is designed to be used in a multitude of devices so the hardware contructs it employs are going to have to be less specific by design, in some areas, than the pure device its going to be utilised in.
I question the convention wisdom that applied, and both of us talked about, back in the 1999 period when DX7 was around with DX8 on the horizon and all the buzz was about the move to more programmability and the questions of fixed-functionality verse programmability and what preformance delta there would be. I feel that we were correct for out time, but the paradigm (I need a new word... situation) has changed with the influx of logic that's been happening and will only accelerate in the next two years tremedously.
When we were talking about a 20M-odd transistor NV15, the balance is alot finer than it is when we're talking about a Broadband Engine or R500 that could be approaching a Billion transistors. We've, IMHO, reached a point where the bounds are on sheer computation and bandwith in dynamic applications. It's possible to design around a modular architecture that's focused on these types of applications which can be scaled down to the low-end apps which have a low resource budget, while retaining cost effeciency due to process advances.
For example, as the ATI guys have told me and you've mentioned on the site from time to time, general computation is moving to the GPU. Not to mention names, but somone here and I were discussing sequencing on them. A GPU is, I'm betting you're going to say, highly tuned to graphic applications but it can still run anything. With the move to a unified shader (I don't know but would guess the ALU they'd use is more akin to a current VS [over current PS constructs, which is why I stated this (http://www.beyond3d.com/forum/viewtopic.php?p=430726#430726)] which is akin to an APU) this just becomes more and more feasible.
london-boy
17-Dec-2004, 13:19
PS. While I'm without a doubt an ass (no pun intended), can we ditch the link to Deadmeat's picture? Thanks.
What link are you talking about??
Link dutifully ditched.
What :?: WTF are you talking about?
if he thinks I've been linking to deadmeat, or anything that idiot says, he can think again!
london-boy
17-Dec-2004, 13:25
if he thinks I've been linking to deadmeat, or anything that idiot says, he can think again!
Nooooo... there was a link in my post with a picture, it was Hey69's link i fixed for him, so he could copy it and paste it to his sig. It had a very nasty looking picture.
Now i took it off. :D
Bring the thread back on topic.
x86 isn't getting SPUs because its doesn't need them, it will use seperate GPUs for when you need lots of floating point maths.
Yeah x86 is unchanged (or changed a bit in x86-64), but the world is changing. What the people in the world want in computers is changing. Without this change in demand, there was no 'multimedia'-focused enhancement in the PC such as NetBurst. Quoting the exact comment of KK from the press release (http://www.sony.net/SonyInfo/News/Press/200411/04-1129E/),
"Massive and rich content, like multi-channel HD broadcasting programs as well as mega-pixel digital still/movie images captured by high-resolution CCD/CMOS imagers, require huge amount of media processing in real-time. In the future, all forms of digital content will be converged and fused onto the broadband network, and will start to explode," said Ken Kutaragi, executive deputy president and COO, Sony Corporation, and president and Group CEO, Sony Computer Entertainment Inc. "To access and/or browse sea of content freely in real-time, more sophisticated GUI within the 3D world will become the 'key' in the future. Current PC architecture is nearing its limits, in both processing power and bus bandwidth, for handling such rich applications."
No Sony is embracing a version of the PC architecture.
Cell is hoping to become the next x86. A standard that works o.k. in most cases.
You think? I, personally, don't see Cell advancing as x86 did. I wonder if we'll see such open and differentiated development happening around it as we do between companies such as AMD and Intel and Cyrix. But, none of us can answer that so...
I do disagree on the standard that works O.K. comment. I think it's pretty clear that at the core it's targetted at a particular area of Computation/Bandwith heavy applications (harking back to Diefendorff and Dubey) that will work O.K. when tasked with other things whose resource requirements scale alot alower or have basically stopped for all intents and purposed. This being said, I recognize that many things assocated with gaming don't mesh well, namely the fixed-functional back-end.
All thats really new is fitting more advanced vector units than usual for a general processor. What I'm saying is that there isn't yet an inherit advantage of Sony's stragegy (except to Sony) than the normal IT industry (which uses x86, PowerPC or ARMs as a rule).
The inherient advantage is holistic, it's in the preformance delta between Sony's PS3 console and the PC at the time it comes out. Do you think the PC will be comparable, I sure as shit don't.
x86 isn't getting SPUs because its doesn't need them, it will use seperate GPUs for when you need lots of floating point maths.
And back to my origional argument we go. If you could do away with the arbitrary divides between vendors and standards and the open platform and the industry was entirely vertical, would we have the current CPU + GPU paradigm as you just articulated?
I'd say no, and I'd think Microsoft agrees with me if X2 is any indication of a trend.
And, please, you don't need to go after the revolutionary speel. I'm not coming from that direction, I'm not saying it is. I'm saying it's going to cure world hunger and get deadmeat laid; I'm just saying it's a departure from the PC and it's significant in it's holistic view towards a computing system and, beyond that, I think it'll have alot of potential for digital content distibution and other non-gaming but still important for Sony tasks.
pahcman
17-Dec-2004, 13:37
KK is complelety wrong when he says PC architecture has run out of steam...
KK said the same think with PS2 :lol:
MrSingh also said Cell is just PS2 on steriods. He has his japan sources, first to break the Nvidia news, right Panajeva and Fafalada?
Just think of SPU's as VU's on steroids.
In fact, the best way to describe Cell is to think of it as a PS2 on steroids. (1 + 2 -> 1 +8)
Panajeva share with us how much more info you manage squeeze out lately? ;)
london-boy
17-Dec-2004, 13:39
Isn't any new architecture the "last one on steroids" plus some new features?
KK is complelety wrong when he says PC architecture has run out of steam...
KK said the same think with PS2 :lol:
MrSingh also said Cell is just PS2 on steriods. He has his japan sources, first to break the Nvidia news, right Panajeva and Fafalada?
Just think of SPU's as VU's on steroids.
In fact, the best way to describe Cell is to think of it as a PS2 on steroids. (1 + 2 -> 1 +8)
Panajeva share with us how much more info you manage squeeze out lately? ;)
Hello? Are you OK?
PS2 succeeded commercially. What's wrong with that? :lol: :lol: :lol:
Dave Baumann
17-Dec-2004, 13:45
I have GPU Gn, CPU Cn, Storage Sn, which of the following would more closely approximate the eq. point of greastest preformance:
[list:0901d49371] Closed System A: [GPU G1, CPU C1, Storage S1] with custom interconnections and system optimization; or:
Open System B: Random combination of [GPU Gn, CPU Cn, Storage Sn] |n=1...1E5[/list:u:0901d49371]
The PC will allways lose out to a closed-set enviroment, especially in the PC paradigm we have in which there is little thinking of the system on a holistic level. Each vender is concerned primarily with their given component and there is little cooperative work. Hell, look at the history of PCI, AGP and - finally - PCI-Express. AGP texturing anyone?
Actually I fail to see how currently available PC inteconnect devices will fundamentally differ from anything in closed box systems of a similar size/performance/ilk within a similar timeframe right now. You may talk derisively about AGP texturing but that was a technology development many, many years ago (and it still being used) - what type of graphics were closed home box home entertainment devices producing at the time? The PC indsutry has moved on to the PCI Express interface, removing the legacy issues behind a parallel interface - what "closed computing systems" are currently, right now, on the market that matches it? In what ways is this not pushing technologies available today to a high degree and in what way does this mean that other steps won't come in similar timeframes to other competing technologoes used elsewhere? The PC has evolved to suite the very changing needs of its environment and the techologies it employs will continue to evolve along the same lines - JHH believes this to be the case as well as he pretty much stated it as well.
And, no, its not "finally" PCI Express, its "currently" PCI Express.
For example, as the ATI guys have told me and you've mentioned on the site from time to time, general computation is moving to the GPU. Not to mention names, but somone here and I were discussing sequencing on them. A GPU is, I'm betting you're going to say, highly tuned to graphic applications but it can still run anything.
And it is, thats the point - we've mentioned in serveral threads about providing much functionality for hiding texture latency as a clear example of that. The other point being is that we are seeing CPU functionality move to the graphics processor, as yet we're not talking about it going the other way.
Vince could you please give your definitions for "PC paradigm" and "Cell paradigm".
Sure, I don't know if labeling it a 'Cell paradigm' is appropriate, bit I was referring to the difference between an open-set of potential combinations taken from a multiple vendor pool that's dictated by market dynamics (PC-esque) and a model which is a closed-set and holistically designed as a closed-set (Cell-esque).
While the open system is advantaged when it comes to reaching a low price point, I've questioned how closely it approximates the region of optimum preformance. Following from here, it's clear that a holistically built device (component invarient for argumetn sake) will have certain intrinsic benefits when it comes to preformance over an open system.
I feel like I'm repeating (because I am), but of course in a competitive enviroment a closed-set will lose out due to marketplace dynamics. Which is why the CE industry and, in particular, the Console field is so unique, it's inheriently closed and it's a good catalyst for a move into the CE field. Sony is even more unique as they are a company that can make this work, controlling the entire pipeline from front-end content creation to the back-end CE devices, they can use an architecture like Cell and Broadband to create synergistic value for the consumer.
pahcman
17-Dec-2004, 14:01
Sure PS2 succeed as a simple game console, like others before it. ;)
The inherient advantage is holistic, it's in the preformance delta between Sony's PS3 console and the PC at the time it comes out. Do you think the PC will be comparable, I sure as shit don't.
I dont get it, if PC has reached limit, and this Cell revolution will remedy these limits, then shouldnt the Cell advantage be persistent? Why stop a the time it comes out? Shouldnt PS3 process undisputed displayed power till the next PS4 revolution?
One thing i sure PS3 will not beat PC at its time, is more RAM more Storage more Functions. :D
Panajev2001a
17-Dec-2004, 14:04
regarding EDram: since Sony is goning to fab the gpu, nvidia is obviously designing it using their design libraries.
SCE's and Toshiba's design libraries ? That would make sense, but they started the design with other tools so the transition process to the ST (Sony/SCE-Toshiba) design libraries will take part of the development time of the custom GPU.
I think 50 of nVIDIA's best engineers (major NV50 work has prolly finished already and they can re-use what they develop working with SCE on the NV60 or try to converge the R&D on the custom parts of NV5X-PS3) engineers should be able to deliver quite a fast architecture.
I think that nVIDIA and SCE can adn will collaborate quite well together: they both are competent designers of complex ICs with a good amount of skilled engineers at their service.
They mention how they see the architecture they are developing for/with Sony/SCE as being present in a multitude of CELL based devices (he mentions how the whole PlayStation 3 platform would become the basis for most of Sony's CE products [scaled according to the product's needs of course]).
I think this extends beyond simply taking the chip and disabling "quads" or removing "quads" from the chip: I think when they designed NV5X they gave a hard thought about the whole range of devices they would want to see the architecture be present in. I think they understood that their goals and what Sony/SCE wanted were not that different from each other.
I think nVIDIA did win the GPU contract because they managed to convince Sony/SCE's management that the future was Shaders, but also because the philosophy of their next-generation architecture went along quite well with the philosophy Sony/SCE had in regards to the CELL architecture.
Actually I fail to see how currently available PC inteconnect devices will fundamentally differ from anything in closed box systems of a similar size/performance/ilk within a similar timeframe right now. You may talk derisively about AGP texturing but that was a technology development many, many years ago (and it still being used) - what type of graphics were closed home box home entertainment devices producing at the time?
Dave, ask a console developer here if they have greater flexibility in using resource sharing and balancing processing between components on the PC or on a console.
The PC indsutry has moved on to the PCI Express interface, removing the legacy issues behind a parallel interface - what "closed computing systems" are currently, right now, on the market that matches it? In what ways is this not pushing technologies available today to a high degree and in what way does this mean that other steps won't come in similar timeframes to other competing technologoes used elsewhere?
Dave, any open-system will need standards; and standards that are mediated by committee will be slower that what a single vertical organization can do. For example, if Sony wants XDR or RaZor, it gets it. AGP was launched in what, 1998... so it's been 6 years or there-abouts?
Blue-Gene's interface blows it away AFAIK. As will Cells in a a few months.
And it is, thats the point - we've mentioned in serveral threads about providing much functionality for hiding texture latency as a clear example of that. The other point being is that we are seeing CPU functionality move to the graphics processor, as yet we're not talking about it going the other way.
Wow! Now do you get why I've been asking you what seperated an X2 ALU from an APU core? Cell isn't built around the PUs Dave... at it's heart it's a large concurrent SIMD Vector processor in the same vein as a unified shading architecture, but a hell of alot faster in clock and more flexible.
Secondly, an SPU complex addresses these concerns to some extent and the basic concept can be extended. That's what you're not understanding, the SIMD cores are very close (Faf answered for you), the difference is between the complex built around the computational resources (eg. SPU or in the Shaders). AND, we don't know fully how their DMAC system works and how the PU/DMAC arbitrate. Your comments are premature and, likely, will come around to get you.
I think nVIDIA did win the GPU contract because they managed to convince Sony/SCE's management that the future was Shaders, but also because the philosophy of their next-generation architecture went along quite well with the philosophy Sony/SCE had in regards to the CELL architecture.
Future is shaders? I don't think so. PC shaders are nothing more than an artifact of the hardware layout of desktop computers.
Future is shaders? I don't think so. PC shaders are nothing more than an artifact of the hardware layout of desktop computers.
Someone just touched the third rail... do I jump on or not? :P
Panajev2001a
17-Dec-2004, 14:23
The XBox shows this is true, and refutes your dumb earlier post defending the current model on the grounds that it's prolific (totally unrelated to the argument at hand), by showing that the same components when taken out of the competitive PC paradigm will outpreform it.
Its not dumb Vince, your statement that the PC vendors have an "(in)ability to create usable and farsignted standard between each component" is clearly incorrect as you have proved yourself by the very fact that it can be applied beyond the scope of the PC environment and used to a full context.
Dave, look at the NV2A, look at the NV20... I think that some times there are clearly cases in which nVIDIA (or ATI) saw performance best attainable one way, saw the future being in a certain direction, requiring certain customizations... yet, let's say, Microsoft (being lobbied by ATI and other companies other than just nVIDIA) disagreed and pushed DirectX in another direction.
Under DirectX, nVIDIA is limited in what they can expose of their innovative ways and if OpenGL for gaming went *puff* and DirectX were to be the main and basically only commercially viable solution in the PC space then you would see IMHO a model that would have to follow Microsoft's pace and not the industry's pace.
The PC model tends also to prefer ease of implementation and ease of backward-compatibility to performance (compatibility in general takes a front-seat compared to performance) as indicated once again by the industry as they chose the slowest of the three standards proposed for PCI-Express 2.0 (5.0 Gbps).
A console, something like PlayStation 3 or Xbox 2/Xenon, not being a Desktop PC and not having to be as compatible as another PC made by Dell or HP, but being its own closed platform can afford to do much more than simply mixing and matching off-the-shelf PC parts with no customization.
Look at the Xbox 2 architecture or a bit more in the future the PlayStation 3 architecture: they run custom software, developed for highly customized hardware (that does not need to employ the technologies which are available and are used in Desktop PCs) and customized APIs (DirectX for Xbox 1's inner-workings were completely re-written and optimizeed for the Xbox 1 architecture and specific customizations present in the NV2A were uncovered... customizations that DirectX on PC would have not uncovered).
Joe DeFuria
17-Dec-2004, 14:33
What is most interesting to me is how Jen-Hsun commented again that this will be used in all Sony CE devices, which is what Cell is intended and was designed for -- We have independent confirmation from both Sony group, IBM and Toshiba on this. So, how would one make these statements logically compatable?
Don't get your question.
Why is this interesting, any more than ATIs and nVidia's already known plans (and already existing products) of pushing their GPUs into other consumer devices, like TVs, set-top boxes, cell phones, etc?
Panajev2001a
17-Dec-2004, 14:39
Of course a closed design can win the short term but lose in the long term. Indeed STI know this, by trying to get Cell into into its own market economy model.
So, you spend most of the post dissagreeing and then undermine it all by agreeing with me in the short-term. Which is all that's relevent to this discussion? *confused*
No Sony is embracing a version of the PC architecture.
Cell is hoping to become the next x86. A standard that works o.k. in most cases.
Moving from a custom designed processor for every generation to using a basically off-the-shelf part thats compatible with the lots of already written software.
All thats really new is fitting more advanced vector units than usual for a general processor.
What I'm saying is that there isn't yet an inherit advantage of Sony's stragegy (except to Sony) than the normal IT industry (which uses x86, PowerPC or ARMs as a rule). Sony is hoping for one processor to rule them all, but I don't see any reason why Cell will do any better than any other plan for global domination.
x86 isn't getting SPUs because its doesn't need them, it will use seperate GPUs for when you need lots of floating point maths.
I fail to see why you see Cell as so revolutionary, its just a CPU with a few independent vector units.
We agree it is not a revolution, but an evolution of the existing theories and practices in micro-processors' design it is.
I really dislike this... bah it takes a general processor and slaps on top of them some vector units.
This is overly simplicistic and off-the-spot IMHO: hey, ya know what... there is not much difference between a BMW M5 and a Yugo... I mean one just has a fast engine slapped on, but both have a chassis, 4 wheels, suspensions and seats... the same car... :roll:
If you break down the basic ideas that CELL was designed with... you'll see that they are not brand new... they have been theorized and experimented mostly "separately" for a long time in the industry.
I see CELL as an architecture that finally, thanks also to the advances in manufacturing technology that make this possible, brings all those ideas together and glue them quite efficiently.
It is like a sport team (of any sport really): the winning team does not just have great players, it has a great coach that knows how to keep the team playing the same game efficiently and see how things should be arranged on the field to allow each player to perform at their best.
It is like spitting on BlueGene/L systems or Itanium 2 based systems like SGI's Columbia because they are based on general-purpose MPUs: go build a comparable system... I bet that after you make the massive newegg.com order for the processors that you will have to sit with tons of empty boxes as you miss the foundations on which those processors rely on and that allows them to communicate, reach data and share data at the highest possible speeds.
I see CELL as an architeture in which each major component was thought and designed both independently (we want a fast and advanced Vector/Scalar unit design) and as parts of a CELL based processor (What does CELL need ? How can we maximize the whole MPU's performance ?).
It's interesting because Sony and Toshiba want to use Cell in CE applications to support their broadband distibution of digital media. Cell, from the patents, was suited towards this movement of data across an IPv6 network fabric; their recent ISSCC press releases mentioned it's use in CE again. What good would an NV50 derivative do in Sony's CE equiptment?
I think you stated:
nVidia is designing the GPU like they would any other PC chip.
Panajev2001a
17-Dec-2004, 14:47
I suppose I should mention that I quite like Cell/PS3 its just I don't get the argument that the future is Cell. That its inheritantly better than the other architectures. Its got some cool ideas, that look fairly good for a games console, but as a Sony employee told me "Its just a good CPU".
To me its just a high performance float version of ARM. Fast, cheap and cool for techies but not really making much of a difference to the rest of the world outside embedded systems..
I would like to see it on PCs... I do think that your Pentium 4's, your Atholn 64's are pushing performance where most PC users (sorry, I should specify... users who use 3D applicatios and multi-media software on their PCs extensively... and these seems to be the driving force behind a lot of the money spent also in the PC sector as nVIDIA's CEO pointed out in the conference call) do not need and not delivering on what they do need: I prefer an approach like the one CELL and Xbox 2/Xenon's CPU are taking. Less optimized single thread/scalar performance and high-speed multi-thread/multi-processing/vector and multi-media processing performance.
Yes, this would be a "gaming/multi-media" kind of PCs, but I ahve no problem seeing future PCs differing greatly from corporate PCs to Server-oriented PCs to Gaming-oriented PCs.
I would like to see IA-64 and parallel processing/multi-cores approaches to replace x86 in the consumer-space as soon as possible as well as I see it as a much better step forward compared to x86's evolution.
Future is shaders? I don't think so. PC shaders are nothing more than an artifact of the hardware layout of desktop computers.
Shader are just code.
The only difference between a shader and 'normal' code is thats is designed to be executed by N (where N is 1 to infinity) processors simultanously. Shaders don't even have to run on GPU, the run well on at least 2 CPUs I know off (one being the Emotion Engine).
Stop looking at PC vs Cell and look at the parellel problem. Suddenly you see two slightly different approaches to the same problem, both seem quite valid to me, both have advantages and disadvantages. I don't see either having a killer blow to the other...
<Tip toes into thread>
Guys WTF are you talking about for the last 5 pages?
It seems we get a Dave and Vince show every other day now and I dunno if anyone else feels the same but I have no clue what your discussing? It seems to be a round robin game of semantics and circular BS!
Can someone please state the question clearly and concisley that is being discussed without fuffy words and verbal diahroea?
/Sorry coffee not working.
<Tip toes out of thread>
london-boy
17-Dec-2004, 14:55
<Tip toes into thread>
Guys WTF are you talking about for the last 5 pages?
It seems we get a Dave and Vince show every other day now and I dunno if anyone else feels the same but I have no clue what your discussing? It seems to be a round robin game of semantics and circular BS!
Can someone please state the question clearly and concisley that is being discussed without fuffy words and verbal diahroea?
<Tip toes out of thread>
There is a God then!!!!
Shifty Geezer
17-Dec-2004, 14:56
I don't get it. If nVidia's effort is based on the NV50 architecture, why'd they announce they're dropping the NV50?
Panajev2001a
17-Dec-2004, 14:58
It's interesting because Sony and Toshiba want to use Cell in CE applications to support their broadband distibution of digital media. Cell, from the patents, was suited towards this movement of data across an IPv6 network fabric; their recent ISSCC press releases mentioned it's use in CE again. What good would an NV50 derivative do in Sony's CE equiptment?
The same thing as Toshiba's MeP would do for Toshiba's CE devices in which they plan to pair CELL with their own scalable MeP architecture.
I do not think NV50 will be so limited in its scalability as you might think: nVIDIA has had the scalability issue in mind when they started NV50 development for the first time IMHO.
london-boy
17-Dec-2004, 14:58
I don't get it. If nVidia's effort is based on the NV50 architecture, why'd they announce they're dropping the NV50?
And the wheel goes round and round and round....
(I asked the same thing on the 2nd page, and in another 3 threads...)
Panajev2001a
17-Dec-2004, 14:59
I don't get it. If nVidia's effort is based on the NV50 architecture, why'd they announce they're dropping the NV50?
1.) Show the official announcement.
2.) Think about Xbox 2 and their use of a R400 derivative which will probably appear on the PC space as R600 later on and that was substituted in the PC space by the more conservative R420 (which was based on the R3XX architecture).
The same thing as Toshiba's MeP would do for Toshiba's CE devices in which they plan to pair CELL with their won scalable MeP architecture.
I do not think NV50 will be so limited in its scalability as you might think: nVIDIA has had the scalability issue in mind when they started NV50 development for the first time IMHO.
Maybe, but the differential in complexity between what I'd assume to be NV50 and MeP is just a bit. heh.
Shifty Geezer, London-boy -- if Dave says it's cancelled I'll believe it, untill then the the Inquirer is still a piece of shit, B3D want-to-be.
Joe DeFuria
17-Dec-2004, 15:12
It's interesting because Sony and Toshiba want to use Cell in CE applications to support their broadband distibution of digital media. Cell, from the patents, was suited towards this movement of data across an IPv6 network fabric; their recent ISSCC press releases mentioned it's use in CE again. What good would an NV50 derivative do in Sony's CE equiptment?
I think you stated:
nVidia is designing the GPU like they would any other PC chip.
Yes, I did. I would revise that to say "nvdia is desiging the PS3 GPU like they would any other GPU, including their PC based ones."
What good would an nV50 derivative do in Sony's CE equipment? That all depends on what NV50 is designed to do (how flexible, etc.), doesn't it?
pahcman
17-Dec-2004, 15:16
The small 50 engineers mean either,
Nvidia NV-Next architecture is set so they can easily custom slap NVPS onto with Cell CPU or
Sony architecture is set, share same philosophy as Nvidia, so only small number is needed to help professionally optimise the Cell VS?
Is this what you think panajeva?
Dave Baumann
17-Dec-2004, 15:46
Dave, ask a console developer here if they have greater flexibility in using resource sharing and balancing processing between components on the PC or on a console.
Developers are only just coming to grips with the ideas of what they can do with the point to point serial nature of PCI Express, so they are only just beginning to explore the possabilities. Curiously its the IHV's that have taken the bigger step by producing hardware that now renders directly to/from system RAM as the performance is good enough to warrent it.
Dave, any open-system will need standards; and standards that are mediated by committee will be slower that what a single vertical organization can do. For example, if Sony wants XDR or RaZor, it gets it. AGP was launched in what, 1998... so it's been 6 years or there-abouts?
And standards can also shift within without large platform changes - in the 6 years of AGP we've had 4 revisions; PCI Express is already scheduled to double bandwidth within a few years, and thats per lane. I don't see that the open nature of of the workgroups has inhibitied the introduction of PCI Express; if so, how is this behind anything else that is realistially affordable for the market now?
IMO, Funadamentally we're not really talking about issues with the approach, but the plain old bog standard difference between Consoles and PC's - PC's have to go for the technologies that are realistically affordable for the market at the time of introduction; consoles can use more exotic elements at the time of introduction becuase they know they are playing a longer game with the hardware and costs will drop. PC's will inevitably catch-up and exceed consoles during their lifetime as the costs for the technologies come into line with being feasible.
Cell isn't built around the PUs Dave... at it's heart it's a large concurrent SIMD Vector processor in the same vein as a unified shading architecture, but a hell of alot faster in clock and more flexible.
And its still has no specialised hardware for removing texture latencies and are the instruction set tunied to pixel shading type functionality?
Megadrive1988
17-Dec-2004, 15:54
this thread has taken a different turn from when I last looked at it, which was around 3 or 4 am central time.
basicly it's (this thread is now about) the high performance custom closed architecture not bound by PC restraints with massive performance that PC should not be able to attain at any given timeframe - VS the PC architecture that cators to compatibility and only high performance in certain pieces, that are unable to work together in absolute harmony without inefficiencies or major bottlenecks. PCs take baby steps every few months. high performance custom closed architectures like SGI visualization supercomputers, IBM super computers, NEC supercomputers and custom non-PC based consoles like Playstation to Playstation2 take massive leaps by introducing a new, fresh architecture at each generation shift. it is true that at the time of their debut and release, the PS1 (1993, 1994-1995) and PS2 (1999, 2000) were unmatched by the best PCs that could be put together with off the shelf components. pretty much the same with SGI visualization super computers (to a much larger degree) and other supercomputers.
although Vince might not acknowledge my post here, I do agree with alot of the things he says.
the Playstation3 and CELL were envisioned to finally smash the PC architecture and deem it irrelivant or dead. with a massive upping of transistors. totally fresh architecture. totally surpassing Intel (and AMD). at least publically that is what Sony was talking about in the early days of 1999 and 2000 with Playstation3 before Cell was announced in 2001. back then it was all about the Emotion Engine 3 and Graphics Synthesizer 3 with drastiically changed architectures Sony would break Moore's Law with a processor that had 500 million transistors. far more than what Intel would likely have for consumers in 2005-2006. at that time (1999-2000) the Pentium III was the highend in PC computing, the Athlon (K7) was getting into gear and the Pentium 4 was on the near-term horizon. PC graphics chips were advancing steadily, but still very much constrained by the worst aspects of the PC architecture (AGP, etc). the Graphics Synthesizer 3 would have a massively parallel design, like the highly parallel design of GS1 in PS2. Much like Intel (and AMD) would not be able to compete with EE3, I think Sony hoped that Nvidia, 3Dfx, ArtX (and others) would not be able to compete with GS3. the EE3+GS3 combined with a decent set of tools, a good API and competent OS would be able to obliterate Wintel in both the living room and on Wintels own turf which was always the desktop and more recently the workstation & server markets. as well as supercomputing. all kinds of CE and computing devices could be built using the new EE3 and GS3 architectures. much like the Cell architecture that came to light in 2001. (I personally believe that the drastically changed architectures of EE3 and GS3 where in fact the Cell based Broadband Engine CPU and Cell based Visualizer).
[post not completed yet, work in progress, kinda rambing as i collect my thoughts]
I happen to believe in the non-PC centric approach. when it's done right, it shows the weakness of PC architecture very well.
even well-put together systems that are not all that radical, and that only somewhat move away from the PC architecture, can show how bad the PC is. like Sega's Model 3 board of 1996. PowerPC CPU + 2 Lockheed Real3D GPUs. there is no way that a properly equiped PC of 1996 could compete with Model 3. you had Pentium I CPUs, Pentium Pro at best (I don't think PII was out then) and 3Dfx Voodoo graphics. Model 3 slaughtered such PC systems. the PC was hard pressed to rival the 2.5 almost 3 year old Model 2 board in late 1996.
Dave Baumann
17-Dec-2004, 15:59
Dave, look at the NV2A, look at the NV20... I think that some times there are clearly cases in which nVIDIA (or ATI) saw performance best attainable one way, saw the future being in a certain direction, requiring certain customizations... yet, let's say, Microsoft (being lobbied by ATI and other companies other than just nVIDIA) disagreed and pushed DirectX in another direction.
Under DirectX, nVIDIA is limited in what they can expose of their innovative ways and if OpenGL for gaming went *puff* and DirectX were to be the main and basically only commercially viable solution in the PC space then you would see IMHO a model that would have to follow Microsoft's pace and not the industry's pace.
Thats a fairly poor example really since capabilitiy wise (not performance, but features) NV25 and NV2A were very, very similar, if not identical, so thats hardly a case for DX stifling innovation at the hardware level, since it was designed in. Its presupposing that MS is "dicatating" the standard, which I'mn sure you know isn't the case - MS have to be driven by the where the IHV's see the capabilities in terms of feasible design implementation in the time frame, and the R&D that the IHV's undertake, alongside MS's own reasearch; all of this is also driven by the feedback MS are reciving from developers wants / needs directly and through the IHV's. MS's DX introductions are also usually time with simialar timescales that the IHV's are inputting to them for these types of features to be affordable - you can see that even though XBox 1, Gamecube and XBox 2 had graphics that were designed specifically for closed box systems the mainstay of the functionality doesn't significantly differ from the functionality on the PC at similar timescales because that was waht was sensible to implement; the areas of difference are a fairly small proprotion of the gates in comparison to the main featureset - I'm taking a stab to say that PS3 will now probably fall in a very similar line, especially if it is based on NV50.
The PC model tends also to prefer ease of implementation and ease of backward-compatibility to performance (compatibility in general takes a front-seat compared to performance) as indicated once again by the industry as they chose the slowest of the three standards proposed for PCI-Express 2.0 (5.0 Gbps).
What makes sense from a performance/cost standpoint is adopted. That is what Dave already pointed out ("And, no, its not "finally" PCI Express, its "currently" PCI Express."). There is currently little need for a (imaginary) 100Gbps bus architecture.
A console, something like PlayStation 3 or Xbox 2/Xenon, not being a Desktop PC and not having to be as compatible as another PC made by Dell or HP, but being its own closed platform can afford to do much more than simply mixing and matching off-the-shelf PC parts with no customization.
Look at the Xbox 2 architecture or a bit more in the future the PlayStation 3 architecture: they run custom software, developed for highly customized hardware (that does not need to employ the technologies which are available and are used in Desktop PCs) and customized APIs (DirectX for Xbox 1's inner-workings were completely re-written and optimizeed for the Xbox 1 architecture and specific customizations present in the NV2A were uncovered... customizations that DirectX on PC would have not uncovered).
So, since you'll now getl a PS3 with a GPU that is designed from the ground up to solve a spefic problem without any need to adhere to an all-purpose ISA such as Cell, you should be a happy camper *takes cover*
I really dislike this... bah it takes a general processor and slaps on top of them some vector units.
This is overly simplicistic and off-the-spot IMHO: hey, ya know what... there is not much difference between a BMW M5 and a Yugo... I mean one just has a fast engine slapped on, but both have a chassis, 4 wheels, suspensions and seats... the same car... Rolling Eyes
While for you current solutions might compare to cell like a yugo to a m5, others (me included) just do not see such a large gap. To me current PCs already sport general purpose CPUs for control and specialized ASICs for computing intensive tasks (think your Pentium/Athlon with graphics-, sound and whatever subsystems). Cell brings more integration, but looking at other embedded systems you'll see this has been the norm rather then exception.
I prefer an approach like the one CELL and Xbox 2/Xenon's CPU are taking. Less optimized single thread/scalar performance and high-speed multi-thread/multi-processing/vector and multi-media processing performance.
Take a look at major IHVs roadmaps. You'll probably have at least dual-core solutions in desktop PCs before the first next-gen console hits retail. SMP with special interconnection schemes has been a common sight in workstation/server markets for years.
Dave Baumann
17-Dec-2004, 16:17
The PC model tends also to prefer ease of implementation and ease of backward-compatibility to performance (compatibility in general takes a front-seat compared to performance) as indicated once again by the industry as they chose the slowest of the three standards proposed for PCI-Express 2.0 (5.0 Gbps).
What makes sense from a performance/cost standpoint is adopted. That is what Dave already pointed out ("And, no, its not "finally" PCI Express, its "currently" PCI Express."). There is currently little need for a (imaginary) 100Gbps bus architecture.
I find it quite ironic that "backwards compatibility" and "ease of implemententation" can be thrown up against PC development in light of the PCI Express transition occuring at the moment - on the graphics front there is neither hardware compatibility or an ease of implementation. And they did so to increase the bandwidth available by a factor of 4 fold.
Megadrive1988
17-Dec-2004, 16:17
btw, I am still hoping that Playstation 3, dispite it now having an Nvidia designed GPU with Sony's input, I am hoping that it is far LESS PC-like than Xbox1, Dreamcast, or even Gamecube. as well as the forthcoming Xenon. even though Xenon is a slight shift away from PC architecture that Xbox1 was, Xenon is still a less radical shift than PS3.
I hope STI-N, Sony-IBM-Toshiba and now Nvidia, are able to combine the very best aspecs of a clean-sheet non-PC architecture with Nvidia's strengths in rasterization & shaders.
The Xbox GPU is essentially a beefy version of the GeForce 3. Will the PS3 (or whatever it will be called) GPU be based on forthcoming desktop GPU architecture or will it be its own entity entirely?
David Roman: It is a custom version of our next generation GPU
Sounds like they actually are doing something for the sony folks i.e. making the chip, and it sounds like if we find out about the GPU in the PS3 we will know more about their next product...
Megadrive1988
17-Dec-2004, 16:33
Panajev and Vince, what do you guys think of my PS4 and beyond predictions? ...that I made back on page 4.
Vince, the way I see PS3 now is, the best hope for it, is that it is indeed a massively powerful super computer CPU architecture from STI, not bound by X86 legacy or even the limitations of MIPs, combine with an SGI-based graphics subsystem. nVIDIA of course being mainly ex-SGI people, as well as people from other competent graphics companies (3DLabs, E&S, Real3D etc).
Sony and Nvidia's vision of graphics probably dove-tail quite nicely.
Regarding the PC architecture and something alternative, along with the KK comment in the press release I quoted above (he says the current PC architecture nears its limit), I'd like to quote this speech (mms://a373.m.akastream.net/7/373/5372/1/gamespot.download.akamai.com/5372/netshow/gslive/2004/12/cellchip_conf_300.wmv) about Cell by Peter Hofstee in which he talks about 'System Trends toward integration' (around 3:10). Probably it can differentiate the current PC architecture from something beyond.
Dave, ask a console developer here if they have greater flexibility in using resource sharing and balancing processing between components on the PC or on a console.
Developers are only just coming to grips with the ideas of what they can do with the point to point serial nature of PCI Express, so they are only just beginning to explore the possabilities. Curiously its the IHV's that have taken the bigger step by producing hardware that now renders directly to/from system RAM as the performance is good enough to warrent it.
This is exactly my point. Developers are "only just coming to grips" with things that console developers have been doing for some time concerning processing balancing and access to system RAM. As a closed set, there are intrinsic benefits. Why is this so hard to agree with?
IMO, Funadamentally we're not really talking about issues with the approach, but the plain old bog standard difference between Consoles and PC's - PC's have to go for the technologies that are realistically affordable for the market at the time of introduction; consoles can use more exotic elements at the time of introduction becuase they know they are playing a longer game with the hardware and costs will drop. PC's will inevitably catch-up and exceed consoles during their lifetime as the costs for the technologies come into line with being feasible.
I fully agree that they are overlapping concepts, they are distinct but fundimentally have analogous causation, which is why I can't for the life of me understand why you're arguing. It's like you're doing it just to say something.
Cell isn't built around the PUs Dave... at it's heart it's a large concurrent SIMD Vector processor in the same vein as a unified shading architecture, but a hell of alot faster in clock and more flexible.
And its still has no specialised hardware for removing texture latencies and are the instruction set tunied to pixel shading type functionality?
It doesn't? What's an SPU (not an APU) Dave, outside of an acronym we toss around and you ignore? Seriously, lets end this stupid thing once and for all... just have a go at it man.
And then you can answer the question of what's the difference between an S|APU SIMD pathway and that in an ALU of contemporary VShaders; which, as I stated Here (http://www.beyond3d.com/forum/viewtopic.php?p=432415#432415), I'd most anticpate a unified shading complex to have commonality with. But correct me if I'm wrong.
Pinky, your comment articulates exactly what I'm talking about:
What makes sense from a performance/cost standpoint is adopted. That is what Dave already pointed out ("And, no, its not "finally" PCI Express, its "currently" PCI Express."). There is currently little need for a (imaginary) 100Gbps bus architecture.
Actually, I stated the preformance/cost considerations verse sheer preformance maximization Here (http://www.beyond3d.com/forum/viewtopic.php?p=432377#432377) & Here (http://www.beyond3d.com/forum/viewtopic.php?p=432370#432370). This is exactly my point, ask yourself why isn't there a need for a 100Gbps bus on the PC. Why isn't the PC designed holistically with more attention payed towards the data flow and computational fluidity; Why does the interconnection/communications lag behind as had been the case with utilization of AGP (vis-a-vis consoles) and will likely be true on PCI-E as well.
Could it be intrinsic to the PC paradigm of multiple vendors in competition horizontally all concerned with tuning their one part based on a cost/preformance equilibrium (and not looking vertically) instead of a more vertical company which is looking at the system holistically. Could it be that there are global regions which yeild higher relative preformance than on the open-set PC, and could it be that a closed-set/system can reach these points?
Maybe that coulkd be why X2 is designed why it is? ;) I'm not talking just about Cell, it applies to X2 as well concering these aspects; I just feel they were stressed more prominently in STI's architecture than MS's.
Megadrive1988, I usually only acknowledge that which I disagree with. It doesn't mean I don't read it.
EDIT: Links sucked.
Megadrive1988
17-Dec-2004, 17:08
Its not dumb Vince, your statement that the PC vendors have an "(in)ability to create usable and farsignted standard between each component" is clearly incorrect as you have proved yourself by the very fact that it can be applied beyond the scope of the PC environment and used to a full context.
No, you just don't understand what's being stated. Very simple, and you're going to do all the thinking:
I have GPU Gn, CPU Cn, Storage Sn, which of the following would more closely approximate the eq. point of greastest preformance:
[list:edf7651e8e] Closed System A: [GPU G1, CPU C1, Storage S1] with custom interconnections and system optimization; or:
Open System B: Random combination of [GPU Gn, CPU Cn, Storage Sn] |n=1...1E5[/list:u:edf7651e8e]
The PC will allways lose out to a closed-set enviroment, especially in the PC paradigm we have in which there is little thinking of the system on a holistic level. Each vender is concerned primarily with their given component and there is little cooperative work. Hell, look at the history of PCI, AGP and - finally - PCI-Express. AGP texturing anyone?
The curious element about your last point is that the Cell paradigm is that it is designed to be used in a multitude of devices so the hardware contructs it employs are going to have to be less specific by design, in some areas, than the pure device its going to be utilised in.
I question the convention wisdom that applied, and both of us talked about, back in the 1999 period when DX7 was around with DX8 on the horizon and all the buzz was about the move to more programmability and the questions of fixed-functionality verse programmability and what preformance delta there would be. I feel that we were correct for out time, but the paradigm (I need a new word... situation) has changed with the influx of logic that's been happening and will only accelerate in the next two years tremedously.
When we were talking about a 20M-odd transistor NV15, the balance is alot finer than it is when we're talking about a Broadband Engine or R500 that could be approaching a Billion transistors. We've, IMHO, reached a point where the bounds are on sheer computation and bandwith in dynamic applications. It's possible to design around a modular architecture that's focused on these types of applications which can be scaled down to the low-end apps which have a low resource budget, while retaining cost effeciency due to process advances.
For example, as the ATI guys have told me and you've mentioned on the site from time to time, general computation is moving to the GPU. Not to mention names, but somone here and I were discussing sequencing on them. A GPU is, I'm betting you're going to say, highly tuned to graphic applications but it can still run anything. With the move to a unified shader (I don't know but would guess the ALU they'd use is more akin to a current VS [over current PS constructs, which is why I stated this (http://www.beyond3d.com/forum/viewtopic.php?p=430726#430726)] which is akin to an APU) this just becomes more and more feasible.
Vince I agree with almost everything you've said.
the Broadband Engine will, I think, be well north of 500M transistors, maybe even approaching 1 billion.
but I am less certain about the R500.
the current R420/R480 is only ~160 million transistors. the R520 PC VPU is rumored to be ~300M and I 'm certain the Xenon's R500 will be more than 300M ....but... approaching 1 billion? I'd be surprised. I was thinking more around ~500M. It's got some eDRAM but not as much as Broadband Engine is expected to have. I would like to believe R500 has near 1 billion. could you show me how you figure it might?
...
And then you can answer the question of what's the difference between an S|APU SIMD pathway and that in an ALU of contemporary VShaders; which, as I stated Here (http://www.beyond3d.com/forum/viewtopic.php?p=432415#432415), I'd most anticpate a unified shading complex to have commonality with. But correct me if I'm wrong.
Is this the question that you guys are trying to resolve with pages of posts? Just curious...I'd really like to know because my eyes have strated to bleed from the insisde...
Dave Baumann
17-Dec-2004, 17:21
I fully agree that they are overlapping concepts, they are distinct but fundimentally have analogous causation, which is why I can't for the life of me understand why you're arguing. It's like you're doing it just to say something.
I take issue with your assertion that they have an "(in)ability to create usable and farsignted standard between each component" Vince, or the implication that the PC is somehow fundamentally EOL, or near, because of it; the concepts that have been put in place so far have been fairly forward things and when they have reached the end of their useful lifetime, the industry PC has the capability to evolve on to newer platforms that better suites its needs at the time. It has been that way for the past 20 years and if it wasn't then there is no way its could have survived that long and they know that if they don't in the future that they won't have an industry to feed on.
You should note that its the entire console (as well as, rapidly becoming, handheld and consumer display device) instrusty that, in the absense of SGI or sufficient inhouse capabilities, that are now reliant on PC IHV's in order to provide their graphics processing capabilities, not the other way around. You have to wonder why this this is the case given the consertavite, shortsighted, and backard environment they operate in! ;)
The PC industry has given them the freedom to innvovate, something that wouldn't be possible along in the consumer industry, but now they have proven technologies they are able to sell them back into the consumer space. If it wasn't for the depth of strength, innvovation that the collective groups of companies contribute and the ability to change and retarget difference uses then you may end up with looking at much worse graphics on your TV screen next year.
I take issue with your assertion that they have an "(in)ability to create usable and farsignted standard between each component" Vince, or the implication that the PC is somehow fundamentally EOL, or near, because of it
I didn't say it's EOL, you're not out of a job, atleast not yet atleast (;)); which is why I like it when you respond to what I've written. I think you're taking issue with something which is a non-issue, I think we'd both agree that each type of platform has specific benefits: which I've stated several for each as have you.
My intentions were to state that a closed-set console has an intrinsic advantage (component invarient) as it's designed holistically, something which is manifested as a particular weakness in the otherwise strong PC model due to the actual market dynamics which are more fragmented between vendors in both axis. I don't know how this didn't get across to you, but I think it's because you're English and I'm American -- so we don't share a common language.
You should note that its the entire console (as well as, rapidly becoming, handheld and consumer display device) instrusty that, in the absense of SGI or sufficient inhouse capabilities, that are now reliant on PC IHV's in order to provide their graphics processing capabilities, not the other way around. You have to wonder why this this is the case given the consertavite, shortsighted, and backard environment they operate in! ;)
Haha. Save this and we'll revisit it in 10 years, will be interesting to see what the industry is like then.
Megadrive1988
17-Dec-2004, 17:35
slightly off-topic. I am waiting to see an announcement that Sony, or perhaps Nvidia or ATI, has acquired that relatively small research company or group (edit: SAARC ?) that has been working on real-time raytracing architectures a using very very small amount of transistors running at very low frequency. edit: the AR350 chip 8)
If that happens, that would probably signal that a future architecture (i.e. PS4, NVxx, Rxxx) would be moving toward realtime raytracing, aside from their own R&D on raytracing.
I'd expect this more of Sony or ATI rather than Nvidia since Nvidia's David Kirk seemed to downplay the significance of this small company's work on raytracing in a debate... oh yeah, Prof. Slusallek
http://www.beyond3d.com/forum/viewtopic.php?t=12881&highlight=raytracing+kirk
*imagines a Cell-2 powered PlayStation4 with a fully realized AR350 or saarCOR based raytracing graphics engine on a massively, massively larger scale than those tiny prototype chips.* ....please don't wake me from this dream unless it happens.... :wink:
Johnny Awesome
17-Dec-2004, 17:42
Vince: Under Seige - the movie. :)
Anyway, Sony promised to break Moore's Law with PS3. If they don't come through, then PS3 will be just another nice piece of hardware. Can't blame them for trying though...
Dave Baumann
17-Dec-2004, 17:44
My intentions were to state that a closed-set console has an intrinsic advantage (component invarient) as it's designed holistically, something which is manifested as a particular weakness in the otherwise strong PC model due to the actual market dynamics which are more fragmented between vendors in both axis.
But this isn't down to some fundamental differences in computing directions or issues with the PC industry, its just the plain, age-old difference between consoles and PC's. Consoles shoot for the stars now because they have to last 4-5 years and they'll make back what they loose on hardware cost reductions and software licenses, PC's implement the technologies when it become feasibles to do so in a cost effective manner.
However, even the timescales at which consoles are exceeded by PC's capabilities these days are are lessening, even more so now the same graphics vendors are being used.
london-boy
17-Dec-2004, 17:45
Hasn't Moore law already been broken?
Megadrive1988
17-Dec-2004, 17:51
still, if PS3 has even a reasonable fraction of a tflop of compute performance I don't see PCs coming close for a long time, as far as Intel, AMD CPUs. when (if) PCs do come close to PS3's compute power, the PS4 will be near, or launching, with many tflops of compute performance.
as for graphics, things on the PS3 side might be a bit closer to Xenon and PCs, as far as hardwired features. but even though PS3 is going to have NV5X based graphics, does not mean PS3 cannot totally eclipse Xenon and PC graphics for a long long time. the 65nm process that Sony, Toshiba and IBM will have long before TSMC has it, could provide PS3 GPU will a large or even massive advantage. 65nm means alot more transistors than 90nm and higher clock speeds. PS3 could have a GPU that is 4-8 times the performance of GeForce 6800 Ultra. not to mention that PS3 GPU will be fully utilized over its 5-7 year life, unlike any given PC GPU which is never fully targeted because of nature of PC game development.
Imagine if you take the best that nVIDIA has to offer and take that out of a restricted platform like the PC and put it into a streamlined, massively parallel, nearly bottle-neck free platform like PS3 or an SGI visualization system. Imagine if you take the best that Nvidia has and combine that with the best aspects of Sony's Graphics Synthesizer chip and GS family.
(including GS I-32, the never-seen GS2, and the GS3 that was probably developed for PS3)... so you combined the best of Nvidia (rasterization, shaders, harewired features, image quality, etc) with the best of Graphics Synthesizer (eDRAM, parallism, high bandwidth) and we could be seeing the birth of something that PS2 almost was in 2000.
of course I could be wrong and PS3 is merely a powerful CPU plus a slightly customized PC-based GPU like NV2A. we'll see! 8)
But this isn't down to some fundamental differences in computing directions or issues with the PC industry, its just the plain, age-old difference between consoles and PC's.
Well, according to what you've stated it can be a fundimental difference in computing directions. I do believe you previously stated:
This argument can then be extended, quite easily, to the entire system: the components within it and the processing model when you close the set and make it invarient -- which is what Diefendorff and Dubey did in their paper. This is the guiding principle that I believe in and, appearently, so does STI.
I don't take issue with the idea can equally be applied to hardware as well, but I just don't agree with some of the things you speak about
But it's not an "issue" with the PC industry anymore than the 4-5 year gap between console revisions is an "issue" as you stated. They are both sources of their respective strengths and their respective weaknesses. You guys really need to stop thinking I'm calling for the death of the PC in every post I make. When I do, I'll be sure to label the post as such.
I do question your last comment though, lets wait 3 months and revisit it. Graphic vendors control the 3D specific IP, which is why they're sought, but they don't have the bleeding-edge process technology and ability to manufacture the parts which is what will ultimately bound preformance in todays world which is increasingly computation and bandwith bound.
still, if PS3 has even a reasonable fraction of a tflop of compute performance I don't see PCs coming close for a long time, as far as Intel, AMD CPUs. when (if) PC do come close to PS3's compute power, the PS4 will be near, or launching, with many tflops of compute performance.
Except a PC today already have ~100GFLOPs (24 (16+6) pipes, 8FLOPS at 500Mhz = 96 GFlops). Thats not even counting the CPU or all the GPU ops (that assumes only 1 4D Dot per cycle).
If GPU keep doubling ever 6 months, they will pass a TFLOP in 2 years...
A PC today is a CPU + GPU. The FLOPs come from the GPU on a PC not the CPU.
london-boy
17-Dec-2004, 18:00
I'm still not so sure how Sony will get a CPU that's so much more powerful than the competition, without making it extremely expensive.
I'm hopeful and all, but in the end, unless this is a miracle machine, everything comes at a cost.
The BE might be the fastest processor at doing some things, with very large shortcomings somewhere else.
It's always like a balancing scale, put something on one side and the other will be inversely affected, and my view is that ther is only so much cash they can pump into a BE, put it on one side of the scale and the other will be inversely affected.
Obviously Sony will target specific tasks they think will be needed and ignore tasks they feel will never be used in a PS3 game.
still, if PS3 has even a reasonable fraction of a tflop of compute performance I don't see PCs coming close for a long time, as far as Intel, AMD CPUs. when (if) PC do come close to PS3's compute power, the PS4 will be near, or launching, with many tflops of compute performance.
Except a PC today already have ~100GFLOPs (24 (16+6) pipes, 8FLOPS at 500Mhz = 96 GFlops). Thats not even counting the CPU or all the GPU ops (that assumes only 1 4D Dot per cycle).
If GPU keep doubling ever 6 months, they will pass a TFLOP in 2 years...
A PC today is a CPU + GPU. The FLOPs come from the GPU on a PC not the CPU.
How much does it cost?
How much does it cost?
About $299 including RAM and profit.
Edit: The 16 pipe cards are RRP $299
Panajev2001a
17-Dec-2004, 18:11
The PC industry has given them the freedom to innvovate, something that wouldn't be possible along in the consumer industry
What kind of Pravda is this ? Are you taking Vince's comments and out of spite inverting them to make the other side shine ?
The PC sector with its focus on backward-compatibility with previous architectures and the other components which the GPU interacts with, with a dominant API that proceeds out of the complete control of the hardware vendors (MS seems to switch, every once in a while, who its favorite IHV is and this is seen in the DirectX evolution and which IHV influenced it and the other adapting and trying to beat the competition at their own game)...
Also, in PCs high-end parts which is where you expect performance and new features to appear ship in very limited quanities and developers, who cater to the lowest common denominator, will not take full advantage of what these architecture expose for a good while (in most cases).
In console chipsets, especially ones with quite guarantee of high volumes of production, like PlayStation 3, Xbox 2 and GCN 2/Revolution, manufacturers can afford more risks engineering their hardware.
PlayStation 3 and Xbox 2 will in fact receive technology which is the bleeding-edge (and it will be further customized) of what nVIDIA and ATI are cooking-up in their own R&D labs and all developers for each platform will have (we can hope, being a closed environment) those features and performance characteristics documented and exposed which is the first step towards taking advantage of them.
Megadrive1988
17-Dec-2004, 18:13
still, if PS3 has even a reasonable fraction of a tflop of compute performance I don't see PCs coming close for a long time, as far as Intel, AMD CPUs. when (if) PC do come close to PS3's compute power, the PS4 will be near, or launching, with many tflops of compute performance.
Except a PC today already have ~100GFLOPs (24 (16+6) pipes, 8FLOPS at 500Mhz = 96 GFlops). Thats not even counting the CPU or all the GPU ops (that assumes only 1 4D Dot per cycle).
If GPU keep doubling ever 6 months, they will pass a TFLOP in 2 years...
A PC today is a CPU + GPU. The FLOPs come from the GPU on a PC not the CPU.
yeah, the GPUs on the PC side are rapidly increasing in their computational performance. but the PC GPUs are not completely general purpose enough for anything but graphics applications, yet. Whereas the CELL CPU in PS3 should be better for general purpose usage than GPUs in PCs are today. The CPUs on the PC side are pretty stagnate. they're not growing in performance. even less growth the last few years than they used to in the 1990s. at least until we get dual-core CPUs from Intel and AMD on the desktop. but even then, dual-core CPUs will not rival highend implementations of CELL like that which will be used in PS3 and other highend computing platforms. not counting the lower-end performance of CELL based DVD players, televisions, PDAs etc.
now back to the PC graphics performance increases: yes they are happening. but highend PC GPUs are hardly being used because ~90% or more of all PCs in the USA & the world are using lowend to midrange GPUs that have been released over the last 5-6 years. GeForce2 MX or GeForce 4 MX or GeForceFX 5200 anyone? very few people will have the newest highest-end GPUs from Nvidia or ATI. whereas, everyone who has a PS3 (or Xenon or Revolution) will have a modern highperformance 3D graphics chip, and in the case of PS3, a very very highperformance CPU. the PS3 will raise the minimum bar, i mean the lowest commen denominator, of 3D graphics in homes. like PS2 did, but probably to a somewhat greater extent, specially because of that CPU.
in this coming cycle, process technology and the massive amount of R&D behind PS3 is probably going to show us a more pronounced difference between PC and console.
this is probably true. the GPUs on the PC side are rapidly increasing in their computational performance. but this is not completely general purpose enough for anything but graphics applications. and the CPUs on the PC side are pretty stagnate. at least until we get dual-core CPUs from Intel and AMD on the desktop. but even then, dual-core CPUs will not rival highend implementations of CELL like that which will be used in PS3 and other highend computing platforms, not counting the lower-end performance for CELL based DVD players, televisions, PDAs etc. now back to the PC graphics performance increases: yes they are happening. but highend PC GPUs are hardly being used because 90~98 % of all PCs in the USA/the world are using lowend to midrange GPUs that have been released over the last 5-6 years. very few people will have the newest highest end GPUs from Nvidia or ATI. whereas everyone who has a Playstation3 (or Xenon or Revolution) will have a modern highperformance 3D graphics chip. the PS3 will raise the minimum bar, i mean the lowest commen denominator of 3D in homes. like PS2 did, but probably to a somewhat greater extent.
Yes exactly, just like the PC business. A range of FLOPs targets based on whats it used for.
Of course 3D gaming needs lots of flops whereas watching a DVD doesn't. Same for PC, same for Cell.
What exactly are we argueing here? Most PC aren't used for games, those that are damn powerful. Who really plays PC games with less than a GFFX today?
When hasn't PC gaming pushed the envelope (as anybody who has Doom3 or HL2 knows, you need a good PC).
Of course the luxery of console refresh is great as it offers a subsized jump in tech, but after a little while the PC market equalises and then beats til the next console appears.
Repeat and rinse. Thats the point Cell/PS3 is 'just' another console, not different PS2 and PS1 before. Better than most PCs when released.
Thats great as a developer and a consumer, just not very revolutionary.
I think we been round the houses with this discussion, best we agree to disagree and see if the PC gaming business is here in 5 years or if we all have PS3 gaming systems :-)
in this coming cycle, process technology and the massive amount of R&D behind PS3 is probably going to show us a more pronounced difference between PC and console.
What like the billion dollars NVIDIA have spend in RnD in the last 5 years (Some of which will help PS3 of course).
Sony have outspent the combined RnD budgets of MS, Intel, AMD, ATI, etc?
Sorry but lots of money gets spend on research in PC land. Its just spread out among N companys.
Sorry but lots of money gets spend on research in PC land. Its just spread out among N companys.
Which is exactly his point, N companies aren't responcible for a single IC or a single system. I don't think it's to much of a stretch to see that investment into Sony's single device/architecture outstrips the others significantly.
Megadrive1988
17-Dec-2004, 18:35
I have no doubt the PC will be around as will PC gaming, in 5 years. it just might shrink a bit more though, as the console industry grows ever larger.
what I am hoping for though, is a competitive computing (not console) platform based on Cell, maybe based on modified PS3 components, seperate from Mac, that competes with Wintel & Winamd PCs
Cell based workstations. Cell based desktop computers. maybe that will replace Mac. or maybe not. but regardless, something that competes with PCs, something that is upgradable every 2 years or so. less often than PCs but more often than console.
something along the lines of Commodore AMIGA in the USA & Europe or Sharp X680X0 and Fujitsu FM Towns in Japan. these computing/gaming platforms were superior in almost every aspect to PCs of their time. I'd like to see something modern, based on CELL, or even some other architecture if not CELL. something that addresses the weakness of PCs.
yet also has its strengths. something that also takes from the best of the console industry (fixed or semi-fixed platform)..
Dave Baumann
17-Dec-2004, 18:36
Well, according to what you've stated it can be a fundimental difference in computing directions. I do believe you previously stated:
I said that hardware designed to a specific task can be more effective than generalised hardware (something that has been a common theme). In the case of bus both are looking at addressing fairly common needs of making high bandwidth upstream and downsteam interfaces.
I do question your last comment though, lets wait 3 months and revisit it. Graphic vendors control the 3D specific IP, which is why they're sought, but they don't have the bleeding-edge process technology and ability to manufacture the parts which is what will ultimately bound preformance in todays world which is increasingly computation and bandwith bound.
The will generally aquire soon after if they don't have it immediately. ATI have 90nm silicon running inhouse now and I suspect they will have 80nm next year and 65 in 2006 - even if they aren't on the same process at the point in time the console vendors (actually, specifically Sony in this case), they will not be that far behind.
A PC today is a CPU + GPU. The FLOPs come from the GPU on a PC not the CPU.
How much does it cost?
Actual ASIC cost for a highend graphics processor is in the order of $150 (with revenues for the IHV), the rest of it is board BOM and vendor profits.
The PC industry has given them the freedom to innvovate, something that wouldn't be possible along in the consumer industry
What kind of Pravda is this ?
Unless you are looking at a different industry than I am I would say it’s a simple truth. With the demise of SGI, and others like them, the PC vendors are becoming, if not already are, the single most prolific driving force in 3D – they are the only force in the PC space, they are the pixel generators for the entire next generation console industry, they are rapidly increasing their presence in the mobile phone company, PC’s with ATI/NVIDIA hardware are becoming more frequent in arcade environments, ATI and NVIDIA hardware are used by high end rendering / visualisation / simulation companies, they are used more frequently in television and film companies and they are increasingly becoming more important in the off-line rendering processes. The only way they have achieved this is by generating the $1-2Billion (and increasing) revenues from the PC market – it simply didn’t happen from the consumer space and now the consumer space is increasingly turning to them.
The PC sector with its focus on backward-compatibility with previous architectures and the other components which the GPU interacts with, with a dominant API that proceeds out of the complete control of the hardware vendors (MS seems to switch, every once in a while, who its favorite IHV is and this is seen in the DirectX evolution and which IHV influenced it and the other adapting and trying to beat the competition at their own game)...
Again, MS don’t, and can’t just stamp their authority on things – this is generated from the directions all the vendors are looking towards anyway; that’s why MS have talked about DX Next for the past two years but only recently finalised on Rev 1.0 of the specification.
Also, in PCs high-end parts which is where you expect performance and new features to appear ship in very limited quanities and developers, who cater to the lowest common denominator, will not take full advantage of what these architecture expose for a good while (in most cases).
ATI sold 4 Million 9700’s alone, that’s before we look at the countless millions of 9800’s on top of that – it is actually a bit of a fallacy. Many titles are constrained more by the performance of current console on a wide variety of graphics cards; the $70 board that NVIDIA have just released has a roughly equivalent rendering power as the Xbox.
I said that hardware designed to a specific task can be more effective than generalised hardware (something that has been a common theme). In the case of bus both are looking at addressing fairly common needs of making high bandwidth upstream and downsteam interfaces.
Uh, reread the conversation.
I do question your last comment though, lets wait 3 months and revisit it. Graphic vendors control the 3D specific IP, which is why they're sought, but they don't have the bleeding-edge process technology and ability to manufacture the parts which is what will ultimately bound preformance in todays world which is increasingly computation and bandwith bound.
The will generally aquire soon after if they don't have it immediately. ATI have 90nm silicon running inhouse now and I suspect they will have 80nm next year and 65 in 2006 - even if they aren't on the same process at the point in time the console vendors (actually, specifically Sony in this case), they will not be that far behind.
Sony had 90nm shipping last fall, they have a bleeding-edge 90nm SoC shipping in Japan and from what I've heard have recieved back 65nm SOI samples which are running in house. I suspect their announcement of 45nm was correct and we can expect that in 2006.
And their far enough behind to fuck them due to the [fixed] launch window. This isn't going to be close, but think as you wish. Things are inevitable at this point.
PS. The PC marketplace isn't driving 3D, the userbase of high-end accelerators isn't expanding at a rate that's even close to Consoles. 4M 9700's compared with 75M PS2s -- that countless millions better be alot. Lets cut the shit, the PC isn't the future for the 3D firms... Jen-Hsun just said as much in his conference when he stated they'e targeting the high-volume markets on their quest to be be a 5-10 billion dollar company.
Megadrive1988
17-Dec-2004, 18:44
DaveBaumann wrote:
With the demise of SGI, and others like them, the PC vendors are becoming, if not already are, the single most prolific driving force in 3D – they are the only force in the PC space, they are the pixel generators for the entire next generation console industry, they are rapidly increasing their presence in the mobile phone company, PC’s with ATI/NVIDIA hardware are becoming more frequent in arcade environments, ATI and NVIDIA hardware are used by high end rendering / visualisation / simulation companies, they are used more frequently in television and film companies and they are increasingly becoming more important in the off-line rendering processes. The only way they have achieved this is by generating the $1-2Billion (and increasing) revenues from the PC market – it simply didn’t happen from the consumer space and now the consumer space is increasingly turning to them.
I find no fault with these statements, generally speakng.
Vince wrote:
Sony had 90nm shipping last fall, they have a bleeding-edge 90nm SoC shipping in Japan and from what I've heard have recieved back 65nm SOI samples which are running in house. I suspect their announcement of 45nm was correct and we can expect that in 2006.
btw Vince, what are the chances that PS3 will launch on 45 nm since you mentioned 2006 ?
None, people underestimated the PS3 timetable.
Unless you are looking at a different industry than I am I would say it’s a simple truth. With the demise of SGI, and others like them, the PC vendors are becoming, if not already are, the single most prolific driving force in 3D – they are the only force in the PC space, they are the pixel generators for the entire next generation console industry, they are rapidly increasing their presence in the mobile phone company, PC’s with ATI/NVIDIA hardware are becoming more frequent in arcade environments, ATI and NVIDIA hardware are used by high end rendering / visualisation / simulation companies, they are used more frequently in television and film companies and they are increasingly becoming more important in the off-line rendering processes. The only way they have achieved this is by generating the $1-2Billion (and increasing) revenues from the PC market – it simply didn’t happen from the consumer space and now the consumer space is increasingly turning to them.
So they became the today's form by M&A. Buying talents from SGI and other failed companies. Just like Microsoft buying their way. It's just market economy at work. Once they have become such form, they try to become something much more than a PC component provider. Are they still the champions of the PC architecture? When this discussion started, 'x86-centric' was mentioned, and I supposed it mainly suggested Wintel rather than component vendors. Is it correct?
Take a look at major IHVs roadmaps. You'll probably have at least dual-core solutions in desktop PCs before the first next-gen console hits retail. SMP with special interconnection schemes has been a common sight in workstation/server markets for years.
True, SMP has been around for years, but its also have not taken off compared to upping the GHz of the x86. It remained in the mid to high-end servers where price is 2nd to performance.
Now Chip and Server vendors are pushing massive multicore on a chip, or system on chip solutions as pushing the Hz is more costly for the value returned. Why just go faster when you can go wider, large [load], smarter, etc?
Cell is in the same shift in paradigm, where you're not just relying on a fat CPU, but instead simpler, but faster units to share the load in a distributed environment. SMP is still poor in distribution of [granular] load as the work units are too heavy/large per core.
You can say Cell and Sun Niagara are evolutions of SMP, but the revolution is bringing these solutions to a cost effect level where it can be use where x86 implementations are norm. That said software will be the key for their success as it’s always been the case.
Dave Baumann
17-Dec-2004, 19:05
Sony had 90nm shipping last fall, they have a bleeding-edge 90nm SoC shipping in Japan and from what I've heard have recieved back 65nm SOI samples which are running in house. I suspect their announcement of 45nm was correct and we can expect that in 2006.
And a year’s difference isn’t that great when you look at the lifecycle of the consoles.
PS. The PC marketplace isn't driving 3D, the userbase of high-end accelerators isn't expanding at a rate that's even close to Consoles. 4M 9700's compared with 75M PS2s -- that countless millions better be alot. Lets cut the shit, the PC isn't the future for the 3D firms... Jen-Hsun just said as much in his conference when he stated they'e targeting the high-volume markets on their quest to be be a 5-10 billion dollar company.
Both ATI and NVIDIA are itching to get into consumer land more and more because that’s the lucrative end of the market, but the simple fact is that it’s the PC market that has enabled them to develop and innovate to such a point that their offerings are attractive to even those that could produce consumer 3D graphics before. Realistically, with the fixed time spans for consoles, it will still be the PC market that is the major driver for constant change and development because that market will demand better graphics when it is cost effective to produce it; the only other market that will have a similar rate of change for the time being will be the handheld mobile/phone market, but this is low end revenues comparatively and that can’t afford a rate of development (and they’ll mainly just be retracing the steps that already have with PC graphics for the past several years).
Panajev2001a
17-Dec-2004, 19:09
The PC model tends also to prefer ease of implementation and ease of backward-compatibility to performance (compatibility in general takes a front-seat compared to performance) as indicated once again by the industry as they chose the slowest of the three standards proposed for PCI-Express 2.0 (5.0 Gbps).
What makes sense from a performance/cost standpoint is adopted. That is what Dave already pointed out ("And, no, its not "finally" PCI Express, its "currently" PCI Express."). There is currently little need for a (imaginary) 100Gbps bus architecture.
What makes sense for "who" ?
It is a big body that decided the new specs and there is more than a couple of vendors who are not happy about the decisions (including nVIDIA which was pushing for faster PCI-Express 2.0 speeds than 5.0 Gbps as they want to get ready for 10 Gbps Ethernet and new SCSI and SATA speeds).
So, since you'll now getl a PS3 with a GPU that is designed from the ground up to solve a spefic problem without any need to adhere to an all-purpose ISA such as Cell, you should be a happy camper *takes cover*
I'll get a GPU whose designed to fit in the CELL platform and not to be compatible with every fricking PC hardware, tons of different motherboards, etc...
Closed environments like consoles do have their advantages: as I said, if the manufacturing vollume is good enough (and in the case of all three next-generation consoles I think it will be) then PC vendors can take risks and push in volume parts that will take longer to arrive and be utilized in the PC market (when they approach the common denominator staus in terms of performance and features more). PRogramming-wise, you have to worry about different hardwares compatibility less and can push the target platform more.
darkblu
17-Dec-2004, 19:17
The PC industry has given them the freedom to innvovate, something that wouldn't be possible along in the consumer industry
What kind of Pravda is this ?
<snip>
ok, apparently there are two mindsets expressed here:
people of the first mindset think PC industry (a synergism for both its hw and sw) is thriving and flourishing, and all other consumer computing sectors can only hope to be like the PC sector. they have their resons for that.
people of the other mindset, in contrast, see the PC industry as a well-entranched behemoth, hardly providing any innovation, and hence falling badly behind the price/performance curve of the day. which may not be necessarily bad for those few entrenched PC IHV/ISV's, but is clearly bad for both the consumer and any (small) innovative players on the market.
and the reason i'm saying this is not to beat a dead horse, but rather to hint at the prospect that the dead horse' beating won't stop any time soon, pardon any possible lack of eloquence on my side.
The next gneration game console will consist of 2 devices that Sony has talked about. One of them is the Cell microprocessor....the second device that will be a companion device to it is the graphics processor and the graphics processor will also be the image processor and there's all kinds of exciting features that will(?) come out. It's based on the next gneration GPU technology and our expectation is to try and put it into production this year. Our next generation GPU has been in development for quite some time as you'd imagine and is something that's near completion.
After finally listening to JH, at this point my thinking is that NV GPU will be integrated into Cell. So I won't be suprised if that's what happend. Who knows it might be on a single chip too.
Like this: 64b Power - Synergistic Processors - NV GPU on one chip.
Sony had 90nm shipping last fall, they have a bleeding-edge 90nm SoC shipping in Japan and from what I've heard have recieved back 65nm SOI samples which are running in house. I suspect their announcement of 45nm was correct and we can expect that in 2006.
And a year’s difference isn’t that great when you look at the lifecycle of the consoles.
Are you serious in alot of your responces? A years worth of difference (one lithography generation) in launch window is equivalent to an economically sustainable 2X increase in logic budget for the 5 year fixed cycle over the competition. It's HUGE.
PS. Say whatever you want about the PC as a platform, defend it for some bizzare reason with arguments about what it's done yesterday, the proof is in the numbers and where they money is tomorrow.
PPS. It was quite elequent darkblu. :)
Are you serious in alot of your responces? A year difference in launch window is equivalent to an economically sustainable 2X increase in logic budget for a 5 year fixed cycle over the competition.
We'll all see if that matters to any of the consumers. Unless they can see a drastic difference in the end result, a 2 times budget in logic won't equate to anything a customer will care about. Even if I doubt it will be 2 times.
Even if I doubt it will be 2 times.
It has to be anyway, if not how in the world ATI gonna expect MS to manufacture that graphic chip for Xbox 2 ?
ATI isn't manufacturing the graphics chip for X2.
ATI isn't manufacturing the graphics chip for X2.
That beside the point and you know it. :D
2.3x the current top-of-the-line due to current fabbing and process size, but when fabbing at 65nm/SOI/Lo-K/etc... they might have more room to up the GHz.
If the Cell BE ~ 4.6+ GHz, then the GPU will likely be 1/4 (worst) ~ 1/2 (best) that speed.
So your telling me they are going to go from 400mhz to possible 2.3 ghz while making the gpu even bigger than it currently is all the while going from a drop of 110nm to 65nm .
Find it hard to believe.
Did anyone expect Cell to be 4.6+GHz? Especially given the top G5 [PPC] is only 2.5GHz.
hmmm I guess there's a reason people "in the know" on the hardware can't stand comments like that. Anyway...
Dave Baumann
17-Dec-2004, 20:39
Are you serious in alot of your responces? A years worth of difference (one lithography generation) in launch window is equivalent to an economically sustainable 2X increase in logic budget for the 5 year fixed cycle over the competition. It's HUGE.
Eh? It can potentially from one console to another, but I'm talking about the difference between a PC and a consoles (even then, thats not neceearily the case for all elements of the PC industry).
Panajev2001a
17-Dec-2004, 20:46
The PC industry has given them the freedom to innvovate, something that wouldn't be possible along in the consumer industry
What kind of Pravda is this ?
<snip>
ok, apparently there are two mindsets expressed here:
people of the first mindset think PC industry (a synergism for both its hw and sw) is thriving and flourishing, and all other consumer computing sectors can only hope to be like the PC sector. they have their resons for that.
people of the other mindset, in contrast, see the PC industry as a well-entranched behemoth, hardly providing any innovation, and hence falling badly behind the price/performance curve of the day. which may not be necessarily bad for those few entrenched PC IHV/ISV's, but is clearly bad for both the consumer and any (small) innovative players on the market.
and the reason i'm saying this is not to beat a dead horse, but rather to hint at the prospect that the dead horse' beating won't stop any time soon, pardon any possible lack of eloquence on my side.
Thank you Darkblu, very eloquent post.
TheMightyPuck
17-Dec-2004, 21:28
Sony had 90nm shipping last fall, they have a bleeding-edge 90nm SoC shipping in Japan and from what I've heard have recieved back 65nm SOI samples which are running in house. I suspect their announcement of 45nm was correct and we can expect that in 2006.
And a year’s difference isn’t that great when you look at the lifecycle of the consoles.
Are you serious in alot of your responces? A years worth of difference (one lithography generation) in launch window is equivalent to an economically sustainable 2X increase in logic budget for the 5 year fixed cycle over the competition. It's HUGE.
PS. Say whatever you want about the PC as a platform, defend it for some bizzare reason with arguments about what it's done yesterday, the proof is in the numbers and where they money is tomorrow.
PPS. It was quite elequent darkblu. :)
This is why MS is willing to lose so much on the Xbox. When the home PC gives way to a converged consumer device they want to be there. Everyone knows where things are going but the trick is predicting critical mass and mass adoption. I think the PC survives a couple more console generations as an entertainment medium. PC's are so amazingly adaptable. Thousands of small developers are inventing new businesses and the pc connected internet is the perfect beta environment. Think about P2P. Biggest thing ever. Skype and other VOIP apps will end POTS in a few years. It will gain critical mass on PC's before going to a converged device. Online gaming gains critical mass on PC's before going to a converged device (xbox live is a start but PC's are where online gaming happens). PC's are cool and allow individuals and small groups of people to create amazing things that simply are not possible on proprietary systems. People use PCs to enhance productivity and will continue to do so for a long time. The bifurcation will happen eventually between production and consumption devices but it ain't going to happen as soon as you think ;)
2.3x the current top-of-the-line due to current fabbing and process size, but when fabbing at 65nm/SOI/Lo-K/etc... they might have more room to up the GHz.
If the Cell BE ~ 4.6+ GHz, then the GPU will likely be 1/4 (worst) ~ 1/2 (best) that speed.
So your telling me they are going to go from 400mhz to possible 2.3 ghz while making the gpu even bigger than it currently is all the while going from a drop of 110nm to 65nm .
Find it hard to believe.
Did anyone expect Cell to be 4.6+GHz? Especially given the top G5 [PPC] is only 2.5GHz.
Is cell 4.6+ ghz ? I dunno i don't have a ps3 in my hands to you ? But thats besides the point
Nvidia's clock speeds have been steadly going down in clock rate since the geforce fx 5800ultra.
I don't see them producing a nv50+ lvl chip much higher than 1 ghz . Certianly not 2.3 ghz Esp when you see them almost double thier transitors generation. Which means they will be at around 444million transistors by the time the ps3 comes out . Hell they will be at that with thier next gen chip after the nv40( 222m at 130nm)
I also don't see ati going past 800-1ghz by xbox 2 launch
Megadrive1988
17-Dec-2004, 23:07
DaveBaumann wrote:
the PC vendors are becoming, if not already are, the single most prolific driving force in 3D – they are the only force in the PC space, they are the pixel generators for the entire next generation console industry
I had to take a look at this statement (a true one) again.
last console generation, graphics chip wise, we had:
3DO M1's graphics: 3DO Systems in-house
Saturn's VDP1 & VDP2: Sega in-house
PS1's 'GPU' (GPU in name only) Sony in-house
NEC PC-FX's graphics: Hudson in-house
N64's Reality Co-Processor: SGI
3DO M2's BDA (BullDog ASIC): 3DO Systems in-house - not used in console.
this generation:
Dreamcast's PowerVR2DC aka CLX2: Videologic, NEC
Playstation2's GS: Sony in-house
Gamecube's Flipper: ARTX a splinter of SGI that got bought by ATI, NEC
Xbox's NV2A: Nvidia
the coming generation
Xenon's R500: largely ATI, maybe some MS stuff
Revolution's GPU: ATI (+NEC?)
PS3's GPU: largely Nvidia, maybe some Sony stuff
I don't really concider Gamecube's Flipper GPU to be from the PC space. more from SGI since ARTX was really completely ex-SGI people. ATI bought ArtX by the time Flipper was nearing completion. of course, Revolution's GPU is fully ATI thus could be counted as coming from the PC GPU space.
indeed, for the first time, the entire next generation of consoles, as far as we know, are, for the most part, getting their pixels generated by PC GPU vendors. the major holdout was, of course, Sony.
On the subject of PC architecture and GPU's...
Nvidia:
"However, he took pains to note that the new console GPU is bringing no PC-related baggage with it. "It's nothing to do with Windows, it doesn't use any of the Windows features, and it's not about driving Windows," he said. "It's not about the PC at all."
Cell or Cell light based?
Megadrive1988
17-Dec-2004, 23:42
yeah that's true as well. although Nvidia is first and formost a PC GPU vendor, its tenticles reach out into almost every area of 3D processing, many that do not have any PC baggage with them.
Dave Baumann
17-Dec-2004, 23:45
Taken literally, driving "Windows" is actually done from a tiny portion of the ASIC valled the VGA engine; all graphics vendors are hopping up and down to dump this at the earliest possible opportunity.
Brimstone
17-Dec-2004, 23:47
2.3x the current top-of-the-line due to current fabbing and process size, but when fabbing at 65nm/SOI/Lo-K/etc... they might have more room to up the GHz.
If the Cell BE ~ 4.6+ GHz, then the GPU will likely be 1/4 (worst) ~ 1/2 (best) that speed.
So your telling me they are going to go from 400mhz to possible 2.3 ghz while making the gpu even bigger than it currently is all the while going from a drop of 110nm to 65nm .
Find it hard to believe.
Did anyone expect Cell to be 4.6+GHz? Especially given the top G5 [PPC] is only 2.5GHz.
Is cell 4.6+ ghz ? I dunno i don't have a ps3 in my hands to you ? But thats besides the point
Nvidia's clock speeds have been steadly going down in clock rate since the geforce fx 5800ultra.
I don't see them producing a nv50+ lvl chip much higher than 1 ghz . Certianly not 2.3 ghz Esp when you see them almost double thier transitors generation. Which means they will be at around 444million transistors by the time the ps3 comes out . Hell they will be at that with thier next gen chip after the nv40( 222m at 130nm)
I also don't see ati going past 800-1ghz by xbox 2 launch
I have the same question. Is "CELL" clocked at 4.6 Ghz? The NEC SX-8 supercomputer has vector processors running at 2.0 Ghz.
Further-enhanced single-chip, vector processor The vector processor (vector and scalar units) is integrated into a single chip by applying leading-edge CMOS technology with 90-nanometer (nanometer: 10-9 meter) copper interconnects and the most advanced LSI design technology. Pipelines of the vector unit, the central part of a vector processor, operate at a 2GHz clock frequency, which is double the speed of the SX-6, and realize a peak vector performance of 16GFLOPS per CPU. Moreover, hardware support of the vector square root operation achieves a sustained performance six times higher than that of the SX-6.
http://news.nanoapex.com/modules.php?name=News&file=print&sid=5158
Megadrive1988
17-Dec-2004, 23:52
off-topic: the above post about NEC supercomputers reminded me of.....
I had once hoped that NEC would build a parallel CPU from its supercomputer & vector processor experience, for Nintendo. something to rival CELL or at least attempt to do so. there were rumors of just that, but they faded. I doubt what IBM is doing for Nintendo is nearly on the same scale as CELL. the Revolution CPU is likely to be a single core or maybe dual-core PowerPC solution. at best, on par with Xenon's CPU, but probably not even. I'm in no way downplaying Revolution though. but thinking more optimistically, the time frame would have been there, or is there, for IBM to build a CPU that's beyond the Xenon CPU, for Nintendo.
cybamerc
18-Dec-2004, 00:41
DaveBaumann:
> However, even the timescales at which consoles are exceeded by PC's
> capabilities these days are are lessening, even more so now the same
> graphics vendors are being used.
Because the prices for high end consumers cards have increased drastically over the last couple of years. A top of the line graphics card costs twice as much as PS2 and Xbox did when they launched.
Vince:
> PS. The PC marketplace isn't driving 3D, the userbase of high-end
> accelerators isn't expanding at a rate that's even close to Consoles. 4M
> 9700's compared with 75M PS2s
The number of consoles versus those of PC graphics cards isn't relevant. ATI's and Nvidia's businesses don't revolve around consoles. I don't think anyone can deny the impact PC gaming has had on the evolution of 3d graphics. The fact that the three big console/set top box makers are going with graphics solutions from ATI and Nvidia should be enough to tell you that.
> Lets cut the shit, the PC isn't the future for the 3D firms...
In the near future it is. Who knows what the market looks like in 10 years from now. The console as a gaming only device could be a thing of the past if Sony and Microsoft have their ways.
ultimate_end
18-Dec-2004, 00:42
so basicly, I take it you are thinking the GPU is mostly Nvidia-based, with input from Sony. it seems to be the best choice, considering that Sony was quite alot behind Nvidia when PS2 was developed, and now that Nvidia has grown in resources and IP by leaps & bounds, Nvidia was the only really good choice for Sony..
Um, how did you come to this conclusion?
PS2 was developed between 1994 and 1998. Now what exactly was the nVidia company at this time? Not much, that's for sure.
As for post PS2 nVidia innovations, video card integrated T&L a la GeForce 256, although having a somewhat revolutionary effect at the time, is not that incredible, considering it was merely a common sense evolution for graphics technology. Programmeable pixel shaders didn't arrive until GeForce 3, with a 2001 release. Irrespective of how long this technology was in development, the fact remains that it was given physical implementation almost 2 years after PS2, which shows that it is a much later development than anything SCEI could have utilised when the GS was locked down in 1998. In the years since then, I can't see how Sony could currently be any more than one PC GPU generation behind in pixel shader technology. Given Sony's graphics related patent applications, even that itself may be a moot point.
And now for the general rant (I've been holding this in for quite some time)...
I find the suggestion that Sony is vastly behind nVidia almost as outragious as the belief, of many on these boards, that Sony needed Toshiba to develop the PS3 GPU.
Emotion Engine notwithstanding, since when has Toshiba had anything to do with Playstation graphics? Even the Emotion Engine is essentially a general media processor, not a graphics processor. The only remotly graphics related development (based on general purpose computing technology BTW) related strictly to graphics is one vector unit and the GIF (which isn't strictly "graphics technology" either). OK, so the R5900 core has a couple of extra graphics related instructions. Big ****ing deal!
Sorry to burst everyone's bubble, but regardless of a few mediocre graphics, patents Toshiba is not the graphics technology powerhouse that people here seem to believe they are! I respect Toshiba more than almost anyone, for their semiconductor technology and SOC expertise. I even suggest that Toshiba had a far larger role in the development of CELL than anyone (including B3D Forums' Toshiba zealots) have given them credit for. I too, dream of an X-Architecture implementation of CELL. But I laugh in the face of anyone who suggests that they have graphics knowledge approaching that even of Sony.
Advanced eDRAM process? That's a joint development between Sony and Toshiba and was intended not only for graphics applications.
Toshiba designs graphics processors? Yes, but then Sony Corporation set up a R&D graphics research department in the early 90's, which still very much operates today and has contributed to SCEI's projects (possibly even PS3).
I probably need to state this again. Sony does not need Toshiba to help them with graphics technology!
.
Yes, that was off-topic. No, I don't give a damn.
Here, back on topic...
I was disappointed when this nVidia deal was announced and that certainly hasn't changed now. Do not assume that this deal came about because Sony isn't capable of making a powerhouse GPU of their own. Do not assume that this nVidia GPU is some kind of radical technology designed specifically for Sony. Do not even assume that PS3's GPU will be much more than a GeForce 7800 Ultra, manufactured with Sony/Toshiba technology. Finally, do not assume that this deal is an indication of some kind of failure of Sony to implement their own design. Alternative explanation random example? If Sony ran out of funds after planning to plow $3 billion into CELL, the "buy now, pay later" licensing approach would have looked a feasible way to produce a powerful, fully featured GPU, without spending many millions of dollars.
Look, I'm not trying to tell you what's what. I'm trying to tell you that we can assume nothing about PS3, positive or negative. In spite of what we think we know *cough*Vince/Qroach/PCengine*cough*, we are all assuming what PS3 will become.
There is nothing wrong with this. Just accept the fact that we may be wrong.
I realise that to some, I sound like I'm very much stating the obvious here, but it had to be said.
BTW Megadrive, please don't think anything in this post is actually directed at you. I realise that you have been very humble in your comments so far. That is more than some might say of myself.
I accept that you may have worded your original comment the way it was to not offend anybody. If that is the case, then please accept my apology for using it to air some of my feelings on this matter.
*edit* And yes of course the nVidia GPU will be customised by nVidia to work with the PS3 CPU.
Megadrive1988
18-Dec-2004, 00:55
ultimate_end, why do you think Sony is going with an Nvidia-based solution for PS3 graphics processor?
I don't recall ever saying anything to the effect that Toshiba is a graphics powerhouse. I've never heard much about Toshiba's graphics teams. since Sony designed the PS1 'GPU' (gpu in name only) and the PS2 Graphics Synthesizer, I would believe that Sony's graphics expertise is far beyond that of Toshiba's but not in the same leauge as Nvidia or ATI, or even 3Dlabs.
admittedly, knowing that the Graphics Synthesizer was locked down in 1997 or 1998 makes me think that Sony could have developed a decent graphics processor on its own for PS3. but it seems they have choosen not to do so this time, or are at the very least getting major help from Nvidia.
ok thanks ultimate_end for you comments at the end of your post. yeah I am trying my best not to be overly against any one company or overly for another. it's hard sometimes.
ultimate_end
18-Dec-2004, 01:44
ultimate_end, why do you think Sony is going with an Nvidia-based solution for PS3 graphics processor?
I don't recall ever saying anything to the effect that Toshiba is a graphics powerhouse. I've never heard much about Toshiba's graphics teams. since Sony designed the PS1 'GPU' (gpu in name only) and the PS2 Graphics Synthesizer, I would believe that Sony's graphics expertise is far beyond that of Toshiba's but not in the same leauge as Nvidia or ATI, or even 3Dlabs.
admittedly, knowing that the Graphics Synthesizer was locked down in 1997 or 1998 makes me think that Sony could have developed a decent graphics processor on its own for PS3. but it seems they have choosen not to do so this time, or are at the very least getting major help from Nvidia.
As I said at the bottom of the post, most of the rant isn't related to what you have said. I just hope now that other people will read it.
Why the nVidia deal? As I said, I'm just saying that we shouldn't assume the reason. I mentioned the cost issue. Another possibility is ease of portability between PC games and PS3. If anyone hadn't noticed, a lot of PC developers don't want to port their games to PS2. The PS2 is too different from a PC, simple as that. Even if PS2 was easy to program for, the lack of PC-style features would kill off the idea anyway. Hell, call it developer lazyness if you want, but it shows a perceived need for such PC commonality to be in PS3. As I also said, Sony would still be a generation behind in pixel shaders as well. But it's not for me to say what Sony feels is needed for PS3.
That's just it, we just don't know enough about any of this.
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.
Don't forget the brick based etc rendering patents, or the fact that a possibly customised high speed CELL GPU implementation, as suggested by the original cell patent, would have also been interesting. Again I will say, I am disappointed with what it seems the PS3 GPU will become.
Megadrive1988
18-Dec-2004, 01:48
just a quick reply before i'm off to work: yeah on some levels I am disappointed that Sony seems to be going for a customized implementation of Nvidia's next-gen desktop architecture. I would have liked to see the other GPU solutions Sony was working on internally, and perhaps with other partners. maybe we'll see a combination of the best of Sony's work and Nvidia. but that may or may not happen. it's true, we don't know enough yet.
Dave Glue
18-Dec-2004, 01:52
Taken literally, driving "Windows" is actually done from a tiny portion of the ASIC valled the VGA engine; all graphics vendors are hopping up and down to dump this at the earliest possible opportunity.
Why, exactly? I can't see this taking up much die space - what hassles does it bring?
ummm PS2 development didn't stop in 1998. It went down to the wire with the early launch titles in japan.
ultimate_end
18-Dec-2004, 02:07
ummm PS2 development didn't stop in 1998. It went down to the wire with the early launch titles in japan.
I assume you are referring to my post.
Now I know you are a game developer, so you probably know about the Graphics Synthesizer's development. But when you say PS2, would you be referring to the changes made to the EE? Such as increasing clock speed and removing the 128bit VU0-VU1 bus? I find it hard to believe that something as relatively simple as the Graphics Synthesizer would require continued development long after the initial specs were released.
*edit* OK wait, I did say PS2 development didn't I? Well I can't really edit it now, but basically I just meant PS2 development on the whole, but really, GS development in particular.
I find it hard to believe that something as relatively simple as the Graphics Synthesizer would require continued development long after the initial specs were released.
I don't see any reason to "not" beleive it. Anyway, I've said what I had to say on it, I'm not getting into a big dicussion where one tries to prove it over another. if you don't agree that's fine...
ultimate_end
18-Dec-2004, 03:27
I find it hard to believe that something as relatively simple as the Graphics Synthesizer would require continued development long after the initial specs were released.
I don't see any reason to "not" believe it. Anyway, I've said what I had to say on it, I'm not getting into a big discussion where one tries to prove it over another. if you don't agree that's fine...
I'll take your word for it then.
In any case, it doesn't change much as far as my original point is concerned. But I appreciate your correction all the same.
As for not getting into a discussion, well most of my post was off topic anyway so...
Is cell 4.6+ ghz ? I dunno i don't have a ps3 in my hands to you ? The NEC SX-8 supercomputer has vector processors running at 2.0 Ghz.
Clock speed doesn't matter in super computers much. Fast interconnect does.
NEC is behind in the process shrink race and has no strength to mass-produce those smaller processed chips because of the economic recovery from the failure of GameCube which couldn't reach the expected shipment target.
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.
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.
PC-Engine
18-Dec-2004, 09:49
the Playstation3 and CELL were envisioned to finally smash the PC architecture and deem it irrelivant or dead. with a massive upping of transistors. totally fresh architecture. totally surpassing Intel (and AMD). at least publically that is what Sony was talking about in the early days of 1999 and 2000 with Playstation3 before Cell was announced in 2001. back then it was all about the Emotion Engine 3 and Graphics Synthesizer 3 with drastiically changed architectures Sony would break Moore's Law with a processor that had 500 million transistors. far more than what Intel would likely have for consumers in 2005-2006. at that time (1999-2000) the Pentium III was the highend in PC computing, the Athlon (K7) was getting into gear and the Pentium 4 was on the near-term horizon. PC graphics chips were advancing steadily, but still very much constrained by the worst aspects of the PC architecture (AGP, etc). the Graphics Synthesizer 3 would have a massively parallel design, like the highly parallel design of GS1 in PS2. Much like Intel (and AMD) would not be able to compete with EE3, I think Sony hoped that Nvidia, 3Dfx, ArtX (and others) would not be able to compete with GS3. the EE3+GS3 combined with a decent set of tools, a good API and competent OS would be able to obliterate Wintel in both the living room and on Wintels own turf which was always the desktop and more recently the workstation & server markets. as well as supercomputing. all kinds of CE and computing devices could be built using the new EE3 and GS3 architectures. much like the Cell architecture that came to light in 2001. (I personally believe that the drastically changed architectures of EE3 and GS3 where in fact the Cell based Broadband Engine CPU and Cell based Visualizer).
Intel sells their cpus for profit. SONY sells their consoles including the cpu and gpu in them at a loss. Intel could EASILY design a cpu to destroy all cpus, but the market for that type of cpu is very small unless it's put into a console that loses hundreds of dollars at launch. Why would Intel want to lose money on cpus? If someone like MS paid Intel hundreds of millions to design a SUPER CPU from the ground up then yeah Intel might do it. :lol:
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.
aaronspink
18-Dec-2004, 11:55
So why didn't x86 die to Itanium?
Because Intel misguessed the future... They guessed that x86 would run out of steam, they guessed wrong. I personally beleive that KK is complelety wrong when he says PC architecture has run out of steam...
Old joke: Never invest in a venture based on the theory that the sun won't rise. If you win, you lose.
Pretty much that same with any company betting against x86. There is simply too much money involved. Sony hopes that they'll sell 100 million units over a 5 year span. AMD and Intel sell more than that in a single year.
The field is littered with challengers from AIM PowerPC (which is very very similar to Cell, lots of pr, no results), Transmeta, Alpha, MIPS, 68K.
In the end it all comes down to this: Anything you can do, they can afford to do good enough. They may not end up faster, but they will end up fast enough to bankrupt you.
Aaron Spink
speaking for myself inc.
aaronspink
18-Dec-2004, 12:21
A PC today is a CPU + GPU. The FLOPs come from the GPU on a PC not the CPU.
How much does it cost?
Cost? Or Price? There is a fairly sizable difference. Mostly because PC compenents are actually valuable.
As far as cost goes, consoles are generally don't have a cheaper BOM than PCs.
Aaron Spink
speaking for myself inc.
aaronspink
18-Dec-2004, 12:26
GPUs in PCs are today. The CPUs on the PC side are pretty stagnate. they're not growing in performance. even less growth the last few years than they used to in the 1990s. at least until we get dual-core CPUs from Intel and AMD on the desktop. but even then, dual-core CPUs will not rival highend implementations of CELL like that which will be used in PS3 and other highend computing platforms. not counting the lower-end performance of CELL based DVD players, televisions, PDAs etc.
Two things. For the majority of applications for which money is made, the current PC processors will exceed the performance of CELL/PS3.
Two, people are living in a dream world if they think that Sony has any influence over the direction of DVD players, PDAs, and televisions. CELL will be found in PS3 for about 99.9999% of its volume.
Aaron Spink
speaking for myself inc.
enture based on the theory that the sun won't rise. If you win, you lose.
Pretty much that same with any company betting against x86. There is simply too much money involved. Sony hopes that they'll sell 100 million units over a 5 year span. AMD and Intel sell more than that in a single year.
You forget Cell is deployed not only in the PS3.
The field is littered with challengers from AIM PowerPC (which is very very similar to Cell, lots of pr, no results), Transmeta, Alpha, MIPS, 68K.
Why is Transmeta included there and AMD not?
A PC today is a CPU + GPU. The FLOPs come from the GPU on a PC not the CPU.
How much does it cost?
Cost? Or Price? There is a fairly sizable difference. Mostly because PC compenents are actually valuable.
Valuable? Or wasted in terms of gaming?
aaronspink
18-Dec-2004, 12:57
You forget Cell is deployed not only in the PS3
That remains to be seen. Sony can't afford to lose that much money and Cell has no value proposition outside of PS3. Sure, you'll see small lot and one off "workstation" systems, but that will be about it.
Why is Transmeta included there and AMD not?
Because AMD didn't think they had a better mouse trap than x86. Because AMD has actually made money from the x86 market, whereas Transmeta hasn't, ever.
Aaron Spink
speaking for myself inc.
pahcman
18-Dec-2004, 13:28
Bit lost...
I thought the dis/agreeing was about how cutting edge SCE technologies are, how Cell will create new worlds impossible without it, leaving competitors in dust? That why Nvidia involvment have cause feathers ruffled.
If its about PC vs set top box vs console, i think really a stretch. I understand cpu power will mature, that a standard box will be enough for years of digital entertainment, without need for messy pc upgrades.
But this is console forum, PS3 will be just another game console. Its no way to challenge PC or set top box. Sony already say there will multiple versions of PS3. Cost will stop them.
Set top box may or not win in the end. PC we know today may lose to boxes, but who to say PC wouldnt become such standard box? Longhorn is coming down with stricter config and if Xenon PC(succesful XPC cannibalises sales? how about MS license Xenon blueprint, like they always wanted to Dell/HP etc) is true, i think PC guys will be in good stead to fight traditional CE groups.
But PS3 Xenon and Revolution, where bulk of sales will be for respective companies, are going be just simpe game consoles imho.
So i ask the Cell lobbyists, do you think eg Cell PS3, because of revolutionary design, will absolutely destroy competition time for time, price for price, until next big Cell hardware? If no, i got no disagreement.
You forget Cell is deployed not only in the PS3
That remains to be seen. Sony can't afford to lose that much money and Cell has no value proposition outside of PS3. Sure, you'll see small lot and one off "workstation" systems, but that will be about it.
Sony doesn't lose money. They only change who they pay for in-house semiconductor needs.
Sony will sell HDTV with Cell, Blu-ray HD recorder with Cell, PDA with Cell, HD digicam with Cell, robot with Cell, and on and on. Toshiba too. This connection with CE demand was previously not there for the companies in your non-Intel&AMD list. Today Intel challenges CE with XScale and a new IA-32 chip. Sony & Toshiba challenges it with Cell, ditching MIPS and ARM. Apart from those CE products and creative workstations, Cell is suitable also for internet edge servers that process massive numbers of I/O requests, with its low power consumption and higher parallelism like Sun Niagara, which nicely fits in IBM product lines.
Cell has no value proposition outside of PS3. Sure, you'll see small lot and one off "workstation" systems, but that will be about it.
Wow.
Perhaps it's time for you to actually do some research on Sony, IBM, and Toshiba's plans for the Cell architecture.
AIM PowerPC (which is very very similar to Cell, lots of pr, no results)
Hey Aaron, plan on getting really quiet in a few months?
Cell has no value proposition outside of PS3. Sure, you'll see small lot and one off "workstation" systems, but that will be about it.
Wow.
Perhaps it's time for you to actually do some research on Sony, IBM, and Toshiba's plans for the Cell architecture.
Lots of companies have plans for lots of things. Microsoft "plans" to put Windows into mobile phones, this doesn't mean the vast majority of people would agree there is any value proposition there.
Likewise for putting a Cell CPU in a DVD player to replace a $3 DVD decoder chip.
By Vince's same argument that a closed system (PS3) will always outperform an open one (PC) due to more opportunities for optimization, a Cell will always be too big, burn too much power, and have too little performance per sq. mm. in the vast majority of CE devices (TV's, DVD players, etc.). The ISSCC results support this.
Cell has no value proposition outside of PS3. Sure, you'll see small lot and one off "workstation" systems, but that will be about it.
Wow.
Perhaps it's time for you to actually do some research on Sony, IBM, and Toshiba's plans for the Cell architecture.
Lots of companies have plans for lots of things. Microsoft "plans" to put Windows into mobile phones, this doesn't mean the vast majority of people would agree there is any value proposition there.
Likewise for putting a Cell CPU in a DVD player to replace a $3 DVD decoder chip.
By Vince's same argument that a closed system (PS3) will always outperform an open one (PC) due to more opportunities for optimization, a Cell will always be too big, burn too much power, and have too little performance per sq. mm. in the vast majority of CE devices (TV's, DVD players, etc.). The ISSCC results support this.
:roll:
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.
Brimstone
18-Dec-2004, 19:04
Cell has no value proposition outside of PS3. Sure, you'll see small lot and one off "workstation" systems, but that will be about it.
Wow.
Perhaps it's time for you to actually do some research on Sony, IBM, and Toshiba's plans for the Cell architecture.
Lots of companies have plans for lots of things. Microsoft "plans" to put Windows into mobile phones, this doesn't mean the vast majority of people would agree there is any value proposition there.
Likewise for putting a Cell CPU in a DVD player to replace a $3 DVD decoder chip.
By Vince's same argument that a closed system (PS3) will always outperform an open one (PC) due to more opportunities for optimization, a Cell will always be too big, burn too much power, and have too little performance per sq. mm. in the vast majority of CE devices (TV's, DVD players, etc.). The ISSCC results support this.
I don't think a Power PC CPU (970) and the SPU's (Power PC 450)combination will be in CE devices like phones, DVD players and PDA's. CE devices will just have SPU's.
Megadrive1988
18-Dec-2004, 19:24
thinking a bit more about Nvidia doing the GPU or at least a very significant portion of it..... today I am not feeling disappointed too much. I think Nvidia handling the rasterizer(s) is ultimately a good thing. while Sony could've done it themselves as they did with PS2's GS, Nvidia can do it better. I still want the Nvidia-Sony GPU to take full advantage of all of the strong aspects of the Graphics Synthesizer / Visualizer concept. massive internal bandwidth. massive parallalism--moreso than what Nvidia would do by itself for an NV50 GPU for PCs. of course PS3 has the opertinuity to be, visually-graphically, the best Playstation yet. I don't just mean the best, as oviously PS3 is going to be better than PS2 no matter what. I mean, the best Playstation for its time. better than PS1 was in 1994-1995 and better than PS2 was in 2000.
Johnny Awesome
18-Dec-2004, 19:32
I would postulate that open systems breed more and better technology, but closed systems produce better implementations of said technology. Sony is admitting this by taking PC tech from IBM and NVidia and adopting it for optimal use in PS3, just like MS is doing with Xenon using IBM, SiS, and ATI tech. There isn't much difference really, other than that Cell is a new CPU from Sony/IBM/Toshiba joint venture. It's not going to change the world IMO. It'll just be a kick ass console.
darkblu
18-Dec-2004, 20:09
I would postulate that open systems breed more and better technology, but closed systems produce better implementations of said technology.
although i would generally agree with you, i think we need to distinguish what qualifies as an "open" platform and what doesn't. would the PC qualify as an "open platform"? why?
in this line of thought, consider the following case study:
year is 1997, we've got two competing graphic technologies at hand: the pvr2 and the voodoo1/2 - both serve the same ends but utilise largely different means. now, the "closed platform" DC endeavor by sega shows the first competitor to be basically wiping the floor with the second, cost- and performance-wise. and yet, the "open platform" pc is effectively dominated by the second up until 1999-2000. why? how did the "open platform" promote the technologically and cost-effective better competitor (as we can safely assume the pvr2 was one)? was it because an open platform would not necessarily promote the better technology or was it because the pc was not exactly an "open" platform? or maybe because the IHV behind the better competitor happened to be jinxed?
akira888
19-Dec-2004, 00:05
PowerVR's failure to penetrate the PC industry is to no small measure due to the fact they allowed Voodoo2, Rage 128 and TNT 12-18 months of free reign until they could release Neon 250, and then they shipped it with immature drivers.
PVR certainly had significant hardware advantages over the antiquated 3dfx tech, but that mattered little if you couldn't play the games you wanted or if you couldn't even buy one. The market, I hate to say, made the right choice.
In any event ATI and Nvidia's superior technology eventually won anyway.
Entropy
19-Dec-2004, 01:10
So i ask the Cell lobbyists, do you think eg Cell PS3, because of revolutionary design, will absolutely destroy competition time for time, price for price, until next big Cell hardware? If no, i got no disagreement.
Of course it won't, for the vast majority of codes.
It's not even certain that it will be vastly superior even in the limited console niche.
However.
It would seem to be a much better thought through platform for high performance media centric computing than x86, particularly when integration levels go 4 times or more beyond the limits of this first implementation. Small wonder, not being the result of decades of evolution of an architecture meant to manipulate bytes in a clerical environment.
Success in the marketplace has little to do with technical merit though, being more related with economic (software) inertia, and market manipulation. The cell concept could very well be a success for the PS3 and PS4, but be irrelevant everywhere else. I certainly don't see the x86 market as a whole clamoring for higher media performance, so even if this first attempt turns out wildly successful, it doesn't really give it much leverage in the x86 market as a whole. If, as seems more prudent to assume, the first cell processor is a mixed success with some strengths and some weaknesses, computing life is likely to go on much as we know it for at least another half decade or more.
Realistically though, going forward, who really gives a damn what ISA Word runs on? Clerical computing is not performance critical anymore, and can be emulated just fine if need be. This creates a small opening where an architecture that is strong in those particular areas where performance still matters can rock the boat.
I doubt the world is interested in having its boat rocked though - x86 forever.
a bit off topic but Gigabyte is going to launch a pc videocard with dual nvidia gpus on one card
----
The card integrates two Nvidia GeForce 6600 GT graphics processors and is the first 6600 GT card on the market to offer a total of 256 MByte DDR 3 memory and 256 Mbit of memory bandwidth, according to the manufacturer. The card is cooled by two on-board fans.
The 3D1's two processors communicate through Nvidia's SLI interface and achieved 14,293 points in 3DMark2003, sources at Gigabyte said. This would not only be almost twice the performance of a regular 6600 GT card, but also more than ATI's Radeon X850 XT Platinum Edition, which achieved in Gigabyte's test environment 13,271 points and Nvidia's GeForce 6800 Ultra, which posted 12,680 points.
----
you guys think we will see this kind of customisation on the ps3 ? multicore GPU
Hi, Im reading B3Ds Forums now and then and decided to push some more Speculations in this Thread :D
There are a few reasons Cell could be more than just a "good CPU". Anyway I`m like most others basing them assumpitons, so dont think im talking about anything else than my interpretations.
The first time I seen the Network-Diagram, I looked over to my Rack: holding HIFI Components, a TV, a couple Consoles and other Devices, all strung together in a mess of Cables, Scart-Switchers and obscure Adapters.
Now think Cell: There would be a single (optical) Wire for each Device, the only additional thing you may need is a hub or switch. Future devices like Pseudo3D TVs wont need yet another Type of Cable, theres no restriction where you route your Music/Video/whatever to, you could even decide via a TV-Remote if you want to play against a pal in splitscreen or on 2 TVs in different rooms. Let alone the fact that everything is transfered digital and therefore free of any noise.
Thats my guess of Cell an how Sony ( and Toshiba ) are gonna use it - it makes perfect sense, as Cell is network-ready and Devices could communicate with each other after exchanging "Software-Cells"- which tell how to de/encode the Data.
There will of course be "legacy" Ports for quite a while, but given a good marketing this could certainly be a huge success (and a very good thing IMHO). The only way to get a digital connection between a PS3 would be to get a Cell-ready-TV.. and so you already got 2 of these Chips ;). In the end Sony could have its IP in all new Media-Devices.
Given those odds, Cell would certainly be a big thing and a big platform to stay for years - a Cell-Desktop doubling as a Mediaserver for the whole family wouldnt be that unrealistic. OSes are gonna be needed anyways, be it for embedded Systems or IBMs Servers.
Additionaly if Cell should have the ability to run multiple OS`s at once and the existence of a few Wintel-Emulators available for Unix( good enough for Apps atleast ) could ease the transistion.
you guys think we will see this kind of customisation on the ps3 ? multicore GPU
multicore (on one Chip) - surely not, instead they would simply double the Pipes, more efficient than having two GPUs to "talk to".
Two physically seperate Chips - possible, but unlikely
aaronspink
19-Dec-2004, 02:32
Sony doesn't lose money. They only change who they pay for in-house semiconductor needs.
And they will end up paying more for a cell based solution.
Sony will sell HDTV with Cell, Blu-ray HD recorder with Cell, PDA with Cell, HD digicam with Cell, robot with Cell, and on and on.
PDA? we'll see. HDTV? again we'll see. HD digicam? Again we'll see. I'll predict that the vast majority (95%+) of all "cell" devices will be in PS3.
Aaron Spink
speaking for myself inc.
aaronspink
19-Dec-2004, 02:33
Wow.
Perhaps it's time for you to actually do some research on Sony, IBM, and Toshiba's plans for the Cell architecture.
Sony = PS3. IBM = licensing and semi revenue. Toshiba = still maintains some relevancy.
Aaron Spink
speaking for myself inc.
aaronspink
19-Dec-2004, 02:34
AIM PowerPC (which is very very similar to Cell, lots of pr, no results)
Hey Aaron, plan on getting really quiet in a few months?
Sure, plan on stepping out of your hole and confronting reality?
aaronspink
19-Dec-2004, 02:36
:roll:
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.
Aaron Spink
speaking for myself inc.
:roll:
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.
Let's wait for a product to be brought to market based on Cell before we start noting how easy and cheap it is. You're very idealistic.
This idea of cell everywhere is a bit overblown. I would have to agree with Aaron that it is well suited for a console but overkill in other applications. The consumer electronics market is very cost sensitive. Unlike consoles the cost (plus profit margin) must be paid for up front. There is no continuing revenue stream which could be used to subsidize the COM.
ultimate_end
19-Dec-2004, 05:51
I take issue with a lot of what has been said in this thread.
Correct me if I'm wrong, but I don't recall anyone saying that PS3 is going to replace the PC! That is utterly rediculous. For Cell to replace the PC, it will take many years. I don't think anyone is claiming otherwise.
And no, Cell is not just a "kick ass console". Sony corporation is investing the billions that it is because they are R&Ding for the future of their entire consumer electronics business. Same goes for Toshiba, who I heard are even supplying GM with cell based processors for use in motor vehicles.
What's the total STI investment? More than $10 billion probably. What kind of fuckwit company spends that kind of money on the CPU of a stupid games console? I cannot believe the relentless stupidity that festers around these boards! Cell probably isn't suitable for general purpose computing, but the scalable architecture sure is suitable for the kinds of applications that Sony and Toshiba are best known for!
Also, who are the morons in this thread who somehow think that PC's aren't moving towards media and entertainment applications? I hate to break it to you, but the "media revolution" already began years ago. Today, even companies like Microsoft, Intel etc openly admit that media and entertainment will soon become the key application for PCs. Open your eyes people. Why do you think that the PC industry relentlessly pursues more power? If you try to tell me that it is to speed up your Microsoft Word, I will laugh in your face! Why do you think things like SSE, 3DNow, MMX, Altivec have appeared? Do you understand?
It wasn't that long ago that PCs didn't even have FPUs. Now, not only are FPU execution units an important part of the CPU, we see endless optimisations and extensions to improve media performance, even in spite of rapidly improving GPU technology. PCs are constantly required for CGI modeling, gaming, photo viewing, music playing, video/TV watching. Everyone wants entertainment and recreation. This is the media revolution. Deal with it.
Media/entertainment and the PC industry is converging. Microsoft wants to put the PC at the centre of your life. Consumer electronics companies want to do the same with their eqiuipment. The PC will become the "home server" if Microsoft get their way. If Sony gets their way, your whole house will essentially be your home server. Which will win out? Don't ask me. But the PC and the consumer electronics industries are converging, simple as that. Where have you people been?
Microsoft wants everything to come from a PC. They have said this already. They want your videos, photos, games and everything else to be stored on your PC, which then serves the user with "entertainment". They want appliances to feature Windows CE type operating systems. I thought all of this was common knowledge by now. But consumer electonics companies do not share this view. They want "ubiquitous computing". They want to conserve their own influence on the market that concerns them the most: The Home. Companies such as Sony envisage that media and entertainment, along with everything else, will come not from PCs but from TVs, sterios and games consoles. At the very least, companies such as Sony do not believe that a single home server should be PC-based. Why do think Cell is so important?
Now, I also read people defending the PC paradigm. In the face of the media revolution, the lack of industry unity and the requirement for legacy software support has held back progress in the PC industry. The PC industry is not as fast moving and flexible as some people *cough*DaveBaumann*cough* seem to think. If it was so quick to react, then why are we stuck in some kind of archaic x86 dark ages. This is the media/entertainment era, but x86 is far from multi-media friendly. A common argument I see is that "but teh GPU cans do all da work!" And while yes, we will see everything move onto the GPU, why then will I want to fork out multiple hundreds of dollars for a useless x86 processor to sit on my motherboard? I don't need 500 million transistors of pentium what-the-hell-ever to run Microsoft Word!
We'll be buying x86 and its derivitives for years to come, just so we can subsidise Intel's survival and backwards compatability with Microsoft software! You have to look at the bigger picture. How should we spend 2 billion transistors total overall system budget? retaining x86 architecture and excessive scalar capability? Think about this a little harder people! The PC "paradigm" is neither cheap nor efficiant, especially in light of the media revolution. How is that so hard to grasp?
vBulletin® v3.8.4, Copyright ©2000-2010, Jelsoft Enterprises Ltd.