PS4 to be based on Cell?

You can have two smaller chips that way. At some point along the line you can intergrate the two. However to start with you'd pay more for the more complex motherboard , however you'd be able to split the heat load amongst two chips. You could also add more cache to each of the chips. Perhaps put the ppu into the south bridge portion of the system along with the tranistors for the sound and external bus controllers. It seems this is the direction the pc market is going . Sure the gpu can assist with physics but nvidia is pushing to have you buy a second gpu just for physics .

 

Only if you can shrink them, which doesn't look like it's going to be an option for next-gen chips.

 

Very true for Sonys position. Every one is having a hard enough time with 45nm transition. Let alone beyond 32. It amazes me that this situation will end up boiling down to Intel vs the world. Eventually every major fabrication tech firm will join into a mega-conglomerate consortium tech sharing group, and will compete in unison against lone ranger Intel. Mental !

 

Couldn't Sony "Larrabee-ise" Cell? Put multiple small dedicated texturing units in there, use lots of SPEs and forget about a third-party's licensed GPU.

Sure, that's a hell of a lot of new software that needs to be written, to get graphics working again. But aren't PS3 programmers already writing a hell of a lot of "close to the metal" "graphics" code to run on Cell, before handing over final rendering to RSX?

If Sony's Cell differ's from IBM's Cell and Toshiba's Cell, right now, why wouldn't it differ in the future?

Jawed

 

TBDR will help but I'm speaking more to hw solutions than software approaches. Specific programming paradigms and approaches is a large and different discussion I'm sure B3D will have in the not so distant future...again.

I expect RAMBUS can pull it off but I somehow doubt it will be "cheap" especially compared to the offerings from competing vendors. We will see I suppose.

 

"Who" writes that software would make that a plausible idea or downright suicidal. Note, most would rather NOT write all of this "Cell Code" "just" to get RSX rendering to its potential and probably would prefer to NOT have to write a software renderer from scratch before the console launches.

If Sony were to provide a well optimized software renderer for Cell that is extendable and customizable then that would likely be welcomed.

However, Cell itself needs a lot more work than slapping a few texture units in and calling it a day. Major work would be needed to address latency alone. I would say the resulting chip after alterations would be a Cell chip in name alone. I would go with LRB vs making pretty much that same thing on my own if I could mitigate the risks. I just don't see the risks being so easily surmounted at all.

 

Yeah, you have two smaller chip with the introduction of a third. GPUs are inherently good at physics given the right tweaks thus the migration to GPGPUs and Cell was designed for heavy computational tasks in mind. Thus given the circumstances, a PPU would be redundant and its introduction would add unneccessary cost with minimal benefit.

Nvidia is pushing to have a second card for physics simply because they want people to buy into SLI in their discrete card market. That doesn't mean that a PPU in needed with a Nvidia based GPU in the console market.

 

Considering there are no 32 nm CPUs produced today AFAIK, I would say there is quite some risk involved if you setup such a short timeline. There is probably a 32 nm Cell already being developed with the PS3 in mind and I think it would make sense if this new rumoured Cell would be designed/produced sometime afterward since it usually take time before a process matures and provides reasonable yields.

Do we know if 45 nm Cells are produced in volumes yet? I don´t think it is likely we will see a 32 nm Cell reach volume production < 2 years from that the 45nm Cell reaches volume production. If 2011 is an important timeline for a new Cell I think they would opt for the 45 nm process to have some safety margins.

The rumour that Sony is evaulating the possibility of using 3d-stacked memory chips, which also is an unproven technology, also make me think Sony is considering a development cycle longer than 3 years.

 

How about combining the chips in layers like the new 3d chips they are talking about.

 

The most important thing for Sony next gen is not to be late in the party against MS, and the second most important thing is to keep costs reasonable against the competition, that means that they have to have atleast one viable option for PS4 tech ready and decided pretty soon.

 

Larrabee handles this with a combination of hardware threads and software threads.

From what I've seen of the discussions on this topic, merely increasing the LS and using purely software threading in each SPE prolly wouldn't be enough. So I agree it would be a major alteration...

When Sony goes for any third party chip of the importance of a CPU or GPU they have the problem of developing software/APIs/libraries/support for it. I suppose this becomes more of a question of balance: can a third party improve this software/bring-it-to-market faster, than Sony working alone?

Jawed

 

everyone is speculating in this thread. SSD may be cost effective in time for the PS4 release.

 

Anything is possible, just not likely

 

I would have to conclude the answer is yes. It Sony were to procure another GPU from Nvidia/AMD then that chip will come with a wealth of tools, common APIs, and a generated wealth of knowledge about how to leverage the part on day one without much intervention at all from Sony. This is assuming the GPU isn't something completely unlike what Nvidia/AMD sells in the PC space.

Sony having to develop a GPU on its own or a CPU that can replace one is a tall order especially given the timeline. Intel is way ahead of them and would have much better tools and nearly everything else if we normalize things. It would be smarter to go with LRB. Of particular note is that the pressure would be on Intel not Sony to ensure the chip performed well and was easily accessible.

That said I am still unconvinced LRB is or will be better than a chip from AMD/Nvidia in more than one regard.

 

Well the tread are discussing on the merit of Cell in PS4. So you probably want to know the alternatives and what it will be paired with.

 

The dilemma for Sony is that pairing Cell with something like LRB is just silly. Two different architectures that have their jobs overlap. Sure you can load balance but it is just silly making devs deal with two complex architectures like that.

 

Hrmm I agree with Shifty's first post in this thread. The assumption that Wii is this gen's "clear winner" being simply because of low cost (at the cost of graphical power) is a mistake Sony need to be wary of. A cheap PS4 "reboot" of the PS3 with better graphics and fully backwards compatability sounds like good business sense, but it still needs that magic draw-in factor the Wii has to make it actually sell.

 

They'd basically be hoping for a market psychology reboot. It can work but you need a strong hook. Wii did that with motion control this time around. What will Sony have? Cell, Blu Ray, new GPU etc are all just enthusiast and hard core talk. It's not the hook for the masses. And frankly you don't know if the hook now, will still be alive and well in 2011. What if the PS4 becomes "wii like" but the market had enough of it and are looking for something else?

Market psychology is the most difficult thing to predict and only becomes known when it's too late.

The key for Sony will be a few things:
1. Stick to gaming. That should be the identity of the PS4. The hardcore will be there for you already but don't waste the precious marketing pre release window by being all over the place like the did with the PS3. Once the market perception is set, it takes a LOT of effort to change.

2. Price of entry. I don't think Sony will make the mistake of a $600 console again. They just won't.

3. Better development tools. Don't let your exotic tech and lack of support tools be the reason your version of the game lags behind routinely. Even when parity is achieved, market perception overrides actual results.

4. Talk shit, less and deliver more. Quit the boasting and grandstanding. Come out and show people that the PS4 will make for FUN GAMES. Don't sit there and brag about power numbers. Like I said, the hard core will be there for you. You've already established the Playstation brand in the minds of the hard core and the casuals don't care for all the tech bragging. Show the games and deliver them ON TIME.

They need to get back to the fundamentals of making a gaming system for a great gaming experience and convince the people that they've done so.

 

Making the software backwards compatible might be something they'd consider too. The idea that software during the transition peroid could have a PS3 and a PS4 version on the same disk. You have a PS3 you can play the game, you have a PS4 and you play a cooler better version of the game. Though really this sort of idea is a pipe dream, it would be cool, but i don't think we'd see it happen. The only reason I'd suggest its possible is because PS3 and PS4 would both use bluray which is huge so even if the two systems aren't completely architechturally compatible, you could just stick both versions on one disk.

 

i never said otherwise.