^^ so true

Can you point me to some reputable, unbiased source giving a detailed history of PS3 development? That would settle most of my questions. Until then I have a hard time believing that Sony ever thought that a 4-Cell design could be remotely cost-conscious.

To respond to your reply #2: If Sony was trying to close a deal with ATI and it fell through, forcing them to turn to Nvidia at the last minute, any multi-Cell designs that may or may not have existed before the attempted ATI deal are irrelevant to the fact that the Nvidia-GPU PS3 design was rushed because of the failed ATI deal.

To respond to your reply #3: In what way is the fact that the Cell is good at things GPUs are good at evidence that a Cell or Cells was/were meant to do the job of a GPU in the PS3?

I watched a video not too long ago with John Carmack talking about how sony had planned on going with an all cell design and he said they were smart to change their setup back to a more general design, as he thought it would have been a very bad idea to have a multi-cell console as was their first intentions. I hope that helps.

WEWdeadeye said: I watched a video not too long ago with John Carmack talking about how sony had planned on going with an all cell design and he said they were smart to change their setup back to a more general design, as he thought it would have been a very bad idea to have a multi-cell console as was their first intentions. I hope that helps.

It does constitute evidence that Sony was considering a multi-Cell design at one point, but says nothing as far as how long into the design process that idea lasted.

Final-Fan said: 2. How does that take care of the rushed GPU design? The GPU in the PS3 is a nvidia G70 variant put in the console at the last moment, it was rushed. All unless what your getting at is that Sony never seriously considered 4-Cell solution; which is odd because they were working on this until a couple months before they went to Nvidia. You dont work on something for that long, with no real alternative, if you arent seriously considering it. Remember that in the past Sony had created their own graphics solutions for their consoles including systems as recent as the PSP. The move to an external vendor for the GPU is quite the change in Sony strategy. 3. It doesn't imply that it was designed to replace the GPU, although it does support the idea that they wanted a 4-cell solution. Can you point me to some reputable, unbiased source giving a detailed history of PS3 development? That would settle most of my questions. Until then I have a hard time believing that Sony ever thought that a 4-Cell design could be remotely cost-conscious. To respond to your reply #2: If Sony was trying to close a deal with ATI and it fell through, forcing them to turn to Nvidia at the last minute, any multi-Cell designs that may or may not have existed before the attempted ATI deal are irrelevant to the fact that the Nvidia-GPU PS3 design was rushed because of the failed ATI deal. To respond to your reply #3: In what way is the fact that the Cell is good at things GPUs are good at evidence that a Cell or Cells was/were meant to do the job of a GPU in the PS3?

One of the higher ups at ATI had a semi-well known interview where he talked about Sonys all cell solution and some of the other things they attempted before they came to ATI. Plenty of developers at Beyond 3D have confirmed this, talking about the original specifications for the PS3. There were rumors from years ago about the PS3 using multiple Cells, for example this and hundreds of other forum post; http://www.dvinfo.net/conf/showthread.php?t=11321 

Sorry, I can't find a "reputable" source of the PS3 development history with a timeline - those things usually dont become clear until years after a system is dead. Still, Sony did think that a 4-Cell system would be cost effective and you can find various write ups talking about cost reduction with the Cell and how it would eventually reach cell phones. A big part of the Cells current cost comes from poor chip yields, which is something that can be hard to predict before you start chip production.

But keep in mind this was Kutaragi's idea of the right price - it was under his lead that the PS3 project ran hundreds of millions of dollars overbudget, and resulted in major problems from the people above him. Supposedly, he just didn't care that much about how much the system was going to cost as all he wanted to do was push his idea of the future, and thought that the PS brand would still be profitiable. Really, just look at rumored production cost of the PS3 now shows they didn't factor in production cost that much in the current model - because it wasn't a major priority of the design.

Also keep in mind that GPUs advanced faster than many people were expecting - Sony in particular.

2. Uh, ok? 

3. Its circumstantial evidence for a reason; no hard proof, but it makes sense. A multi Cell PS3 without a real GPU was rumored for ages, you have a CPU that can do GPU functions - its not rocket science.

Why the hell would you make a CPU be capable of handling GPU like functions if you intended to have a real GPU in the system - one that can do those things much better. Why have it do those functions when it comes at the detriment of AI and other CPU functions? Why not create a simpler, more convential CPU that would be easier on developers - unless you meant for those SPUs to fullfil a GPU like role. There is a certain elegance to an all Cell PS3 design, but GPUs just advanced too fast to make it competitive from a performance standpoint.

ok..Ive been following this thread from the begining (can't believe I'm actually posting arg!).

There is no point-counterpoint discussion from the individual who created this thread.  Everytime someone comes up with an intelligent, well-thought out, technical counter it is completely ignored.  The only things addressed by said person are the disputable portions of others threads without every adressing legitimate points.

The only comparison I can make is politicians, ignoring the legitimate arguments from one another and focusing in on how they can discredit the person thus drowning the message.

This is PR at best and a vain attempt to blind anyone willing to question at worst.  Your debate methods are flawed and the validity of anything you say quickly became overshadowed by your exremely weak methodologies for addressing legitimate counterpoints.  Any credibility you may or may not have had becomes irrelevent under such circumstances. 

You expect people to buy into your ethos, but all you use is pathos masked as logos. Using intrinsic information to derive extrinsic value is falwed logic and rhetoric. Selective data and misreprsentative comparisons do not a vaild point make. If casual reasoning and ambiguities are the only arguments that can be made then they can be dismissed out of hand.

AHHHH. I'm done, this is pointless.  Both individuals have to be willing to listen for discussion to take place and I'd rather hear nothing back than more formal and informal flawed logic arguing with what I've said.

2. The thrust of the point I had made which you had been arguing was that the multi-Cell design didn't have to exist for the GPU design to be rushed. That's all I was trying to say.

3. The Cell is not designed exclusively for the PS3, correct? Perhaps the GPU-like qualities of the Cell are simply a natural result of the processor's novel take on CPU design. It would be plausible to say that that property would cause Sony to consider a Cell graphics solution, but not to say that the intention to use it for such is necessarily the reason why the Cell was designed the way it was.

And GPUs aside, wouldn't a 4-Cell (or EIGHT-CELL, as the thread speculates) design have been prohibitively expensive? Look at how much one Cell cost! "PS3 is expected to sport multiple Cell chips" -- I can easily interpret that as being the result of people taking the fact of the Cell's existence and basically speculating the PS3's design by saying "how would I use this to build an awesomely powerful console?" That may not be so, but nothing about that post tells me it's not. In fact, for all I know, it could be the result of someone misunderstanding the PPU/SPU design!

From Ars Technica: "each SPE is like a very small vector computer, with its own "CPU" and RAM." It wouldn't be hard for someone to misinterpret that comment and others like it.
http://arstechnica.com/articles/paedia/cpu/cell-1.ars/2
http://www.pcstats.com/articleview.cfm?articleID=1727

Please try again, I really want to settle this question of multi-Cell design, I won't be sad to be proven wrong.

Regarding the ATI interview you mention: I presume you mean this one: "With the PS3 my understanding of what happened is that they had three different internal hardware solutions - at one point, for example, as I understand it there was a proposal to use multiple Cell processors just to handle the graphics. And towards the end of the process, as the story goes, they took a look at the three internal tenders and decided than none of them would actually do".

That is entirely consistent with my theory that the multi-Cell design was an early idea that got thrown out because it was impractical -- one of many proposed designs, but not one that necessarily made it very far into the design process. The ATI guy probably used that example because it was memorable and remarkable.  And ATI had no reason to love Sony at that point.  

OnlyIsrael said: ok..Ive been following this thread from the begining (can't believe I'm actually posting arg!). There is no point-counterpoint discussion from the individual who created this thread.  Everytime someone comes up with an intelligent, well-thought out, technical counter it is completely ignored.  The only things addressed by said person are the disputable portions of others threads without every adressing legitimate points. The only comparison I can make is politicians, ignoring the legitimate arguments from one another and focusing in on how they can discredit the person thus drowning the message. This is PR at best and a vain attempt to blind anyone willing to question at worst.  Your debate methods are flawed and the validity of anything you say quickly became overshadowed by your exremely weak methodologies for addressing legitimate counterpoints.  Any credibility you may or may not have had becomes irrelevent under such circumstances.  You expect people to buy into your ethos, but all you use is pathos masked as logos. Using intrinsic information to derive extrinsic value is falwed logic and rhetoric. Selective data and misreprsentative comparisons do not a vaild point make. If casual reasoning and ambiguities are the only arguments that can be made then they can be dismissed out of hand. AHHHH. I'm done, this is pointless.  Both individuals have to be willing to listen for discussion to take place and I'd rather hear nothing back than more formal and informal flawed logic arguing with what I've said.

ding ding we have a winner

@ OnlyIsrael

I mostly correct incorrect statements, personally I don't see anything wrong with that. The main reason for this thread was to dispell untruth and give counter opinions and perspectives against all the negative bias.

I predict we will see way more game engine advancements on the PS3 than we will see for the XBox 360. The techincal capabilities of the XBox 360 is well understood by developers, the PS3 is still more in the experimental phase.

Looking back at the original post the only points disputed seem to be CPU/GPU perfomance related, I think I have gone into great depth explaining why the RSX/Cell is more powerful for the long run than the Xenon/Xenos combination.

All that's left are some people who don't really understand these clarifications, I am sorry but they will have to wait a couple of years, as only seeing is believing when you're not an expert.

At least a lot of misinformation has been corrected in this thread, so if we see less techical FUD and misinformation here on VGChartz in the future I consider my mission to have been successful.

My last post for this thread, if you want to continue discussing something feel free to PM me.