[Florin]

Originally Posted by deepbrown

http://www.driverheaven.net/news.php?newsid=344



Not the best source - but neither is Heiss Online in the first place that brought this IBM cancelling Cell thing.

Generally, Heise is first class reporting.

[Lazy8s]

Eventually, the winning designs in the mobile sector -- META someday -- will spread out to supercomputing, whereas mini-Cell never would've stood a chance in the mobile space.

[draconian]

Does this mean that the ps4 will most likely not be backwards compatible with the ps3?

Because of diminishing returns of graphics. This gen's graphics will, I think, compete well with even next gen's graphics.

And because of this Sony pretty much lost next gen. This generation's game will have long legs and people will still want to play uncharted or alan wake on next gen systemts.

[semitope]

You should read the rest of the thread. All i am seeing is that the 32spe 4PPE cell is no longer going to be made, not that cell is completely off for IBM. I would like to see a form of cell in the future ps. Seeing the things they are doing with the current 8 spe cell it would be great to see something like a 4+Ghz PPE 5Ghz spe x 64 cell processor in the future ps. (are varying clock speeds possible?)

I believe sony would be quite secure with improved cell variants for at least 2 more playstation generations with ati dx11 graphics as GPU.

There is mention here of NLM and whatever other GPGPU but why would sony need to take yet another risk with their playstation line? Its time to mature what they have built and I dont see the need to integrate everything yet. The cell can already do some work for the gpu and dx11 gpus can be used to do some tasks outside of graphics so I am thinking work could be done to build an archictecture where even though both are separate they could be made to work in a system in which either could help the other (without bandwidth limitations and with low latency memory access both ways). Basically a system where the GPU can help the CPU and the CPU can help the GPU but neither compromises in being able to carry out its own traditional task.

/end fantasy

[draconian]

This is what I would like for sony to do.

Keep cell architecture.
go from single core ppe to 3 core ppe. Increase on die cache from 512kb to 1 mb. Keep the spus the same # but increase the sram from 256kb to 512 kb. Maybe even give some spus access to outside ram(I'm dreaming aren't I?).

Then simply increase ram from 512megs to 4 gigs. and get a better backwards compatible gpu from nvidia.

That's it. Maintain backwards compatibility and everyone will be happy.

[liolio]

Originally Posted by Shifty Geezer

Perhaps it's due to this matter that I disagree with that explains why I feel Cell etc. can be successful PC processors. My processor is 95% idle 95% of the time. Compressing a huge amount of data will be limited more by storage speed then processing speed. A Cell core could handle this with ease. Applying filters in Photoshop will be realtime on a Cell optimised platform on a couple of SPEs in most cases, and anything more significant will still only be a matter of seconds. And how many people compile large projects at home?

So sure, it wouldn't be optimal, and if running a task like an encode, perhaps a good percentage of cycles will be wasted as they're reserved for other systems. It'll still be a very viable platform though, comfortable to use and very fast in the heavy workloads. At the end of the day, a Cell PC running a Cell-targetted OS would feel every bit as responsive and flexible as an x86 Windows PC and still be able to number crunch supremely well.

Make me wonder about something, Nvidia is trying to tackle the GPGPU by the high end, do you think (you and other members) that Nvidia and others may have first trying to grow GPU as more general purpose accelerator in the low hand? I mean there is not a lot of pressure on IGP graphic performances and it could have been a good place to introduce a compute oriented GPU, cryptic graphics performance (perf/mm² obviously) may have gone mostly unnoticed.

Originally Posted by semitope

You should read the rest of the thread. All i am seeing is that the 32spe 4PPE cell is no longer going to be made, not that cell is completely off for IBM.

Well IBM statement is unclear " the heart of cell will continue to live in many products,etc."
I stated it as a joke but IBM could really be speaking of the power PC, IBM relevance as a CPU manufacturer is going down, the cell no longer has a roadmap, X86 are gaining ground on PowerPC chips, statement could be about reassuring IBM engagement on the hardware front.

[Weaste]

Originally Posted by liolio

the cell no longer has a roadmap

We don't know that. They could simply have modified it.

[liolio]

Originally Posted by Weaste

We don't know that. They could simply have modified it.

And not release it? secrecy in this regard serve no purpose imho.

[deepbrown]

Originally Posted by Carl B

The difference is that Heiss Online was quoting the head of IBM's deep computing division - and then later got further clarification to boot - and DriverHeaven is simply quoting a PR person on the extent of what has officially been canceled. I've spoken to IBM PR people before, and believe me, anything from them is like nothing at all.

Not only that, but a part of me is worried that Kotaku and DriverHeaven and all these folk spoke to the 'same' PR person... which is to say that only one of them did, and they're all simply quoting each others findings and claiming that they also contacted IBM in tandem.

I know which is why you find my report of Heiss Online at the top of this thread...and not a report from DriverHeaven

[Carl B]

Originally Posted by liolio

Well IBM statement is unclear " the heart of cell will continue to live in many products,etc."
I stated it as a joke but IBM could really be speaking of the power PC, IBM relevance as a CPU manufacturer is going down, the cell no longer has a roadmap, X86 are gaining ground on PowerPC chips, statement could be about reassuring IBM engagement on the hardware front.

Well they definitely don't mean Power; Power has a life to live anyway, and Power7 is going to be a big deal for IBM, so that's not Cell related in and of itself. Whether there is a true continuance of Cell or not, what they were referring to was definitely IMO the notion of many-core, small/efficient, high throughput design, potentially also heterogeneous. If and when such a chip shows up, we'll recognize it immediately.

Originally Posted by deepbrown

I know which is why you find my report of Heiss Online at the top of this thread...and not a report from DriverHeaven

Fair.

[V3]

I think at this point if IBM moves forward with the next iteration of Cell, it is most likely won't be backward compatible like how 32i version was going to be. It'll be more like spiritual successor rather than something that is next in line.

[semitope]

Originally Posted by V3

I think at this point if IBM moves forward with the next iteration of Cell, it is most likely won't be backward compatible like how 32i version was going to be. It'll be more like spiritual successor rather than something that is next in line.

It better be backwards compatible otherwise that is another nail in sonys Playstation coffin. The ps3 would not be cheap enough to excuse that omission when the new playstation comes out (not likely anyway)

[liolio]

Originally Posted by semitope

It better be backwards compatible otherwise that is another nail in sonys Playstation coffin. The ps3 would not be cheap enough to excuse that omission when the new playstation comes out (not likely anyway)

STI no longer is, IBM is free to do what they want. BC may be important for them too tho, but they'll have to weight
different factors mostly the relative strength of competing architecture.

[V3]

Originally Posted by semitope

It better be backwards compatible otherwise that is another nail in sonys Playstation coffin. The ps3 would not be cheap enough to excuse that omission when the new playstation comes out (not likely anyway)

It should be possible for Sony to put together Cell for PS4 with more PPUs and SPUs and new north and southbridge. They have the building blocks. It won't be like IBM 32i where it was going to feature enhanced SPUs.

It will be just like PS3 Cell except with more cores. Cell can already scale to multiple chips, it should be possible to put those chips onto one die in the future.

Sony problem is not CPU, it's the GPU they've got none and no one is making a suitable GPU to sit along mutli cores Cell. They need to get Toshiba working on Cell base GPU or something.

The path that IBM is taking with future Cell, won't be suitable for PS4. It'll target server and supercomputer not consoles or consumer electornics.

[Weaste]

In all likelihood, Sony will already have their base architecture for the PS4, and what V3 suggests regarding the CPU is probably what they have done. I don't know if it needs to have 32 SPUs (or four Cells connected on a bus), it's probably overkill, but two of them duct taped together should be very cheap for them to do. The GPU on the other hand is another question entirely.

[JPT]

I am not sure how important the backwards compatibility is, if you look at the numbers of PS3 sold, I think most of those are not backwards compatible.

[Weaste]

I don't necessarily think it's about sales in that regard, but rather allowing developers to hit the ground running and reuse the technology and engines that they have developed.

[Prophecy2k]

Originally Posted by JPT

I am not sure how important the backwards compatibility is, if you look at the numbers of PS3 sold, I think most of those are not backwards compatible.

I'd wager BC will be much more important next generation when you factor in the online platforms.

What will people do if they're no longer able to play their PSN games on their PS4, yet people with a nextbox will be happily playing their xbla content on their nextbox?... they'l likely rage and switch sides. I'm convinced BC will be much more of a deal breaker next gen than it was this gen.

Again, one has to factor in that this gen we went from SD gaming to HD. While next will likely be 720pHD to 1080pHD, and so many may still care about playing thier old games as they will still look good on their HD sets against their new games.

Sony would do well to preserve thier use of the CELL architecture in my book. If only for the BC and ability to leverage existing tools/libraries.

[Crossbar]

BC is still important for Playstation 3 in this generation. We know from the leaked SEGA meeting notes that PS2 SW BC is coming to the PS3 in some shape.

[ShadowRunner]

Originally Posted by Prophecy2k

I'd wager BC will be much more important next generation when you factor in the online platforms.

What will people do if they're no longer able to play their PSN games on their PS4, yet people with a nextbox will be happily playing their xbla content on their nextbox?... they'l likely rage and switch sides. I'm convinced BC will be much more of a deal breaker next gen than it was this gen.

Again, one has to factor in that this gen we went from SD gaming to HD. While next will likely be 720pHD to 1080pHD, and so many may still care about playing thier old games as they will still look good on their HD sets against their new games.

Sony would do well to preserve thier use of the CELL architecture in my book. If only for the BC and ability to leverage existing tools/libraries.

Why would they switch sides? they still wouldnt be able to play thier old games if they did that. Sure BC is a nice extra feature that could encourage a few more sales but if the machine is great and has a great software lineup it will do well on its own merits, regardless of BC. If the forced inclusion of BC effects the consoles in a negative way it could be a mistake. If BC inclusion comes easy then fine but it has a negative impact on consoles design then leave it out.

Originally Posted by Crossbar

BC is still important for Playstation 3 in this generation. We know from the leaked SEGA meeting notes that PS2 SW BC is coming to the PS3 in some shape.

Its important to sony because they want to be able to make money from selling thier old games, im not sure consumers are that fussed beyond it being a nice little feature. How well do XBox originals do? and is it really something that effects consoles sales significantly?

[Carl B]

The BC discussion aside, remember that ease of programability/approachability is a free-floating variable independent of Cell's existing platform maturity. If Sony were to continue with Cell, yes the experience of devs so far and the increasing maturity of tools would be of benefit to the system. But that's not to say that if Sony switched architectures, that the system wouldn't also be approachable... and in fact in spite of the switch might still become moreso. Like, what if they went straight x86 OOE for example? Not saying that's what I think would happen, but just trying to get people to stop thinking about the investments already made in the architecture as irrevocably tied to PS4 ease of approachability.

[specwarGP2]

Originally Posted by ShadowRunner

Why would they switch sides? they still wouldnt be able to play thier old games if they did that. Sure BC is a nice extra feature that could encourage a few more sales but if the machine is great and has a great software lineup it will do well on its own merits, regardless of BC. If the forced inclusion of BC effects the consoles in a negative way it could be a mistake. If BC inclusion comes easy then fine but it has a negative impact on consoles design then leave it out.

With physical media, you can always sell or give away your old games. With DD you're stuck with it, and I think the perception is that if I'm stuck with it, it better work in perpetuity. I know if I bought a movie from itunes, and some time down the line it stopped working, I'd be really annoyed and I'd never buy anything from itunes again.

[Prophecy2k]

Originally Posted by specwarGP2

With physical media, you can always sell or give away your old games. With DD you're stuck with it, and I think the perception is that if I'm stuck with it, it better work in perpetuity. I know if I bought a movie from itunes, and some time down the line it stopped working, I'd be really annoyed and I'd never buy anything from itunes again.

Thanks That's the point I was in fact trying to make.

[Prophecy2k]

Originally Posted by Carl B

The BC discussion aside, remember that ease of programability/approachability is a free-floating variable independent of Cell's existing platform maturity. If Sony were to continue with Cell, yes the experience of devs so far and the increasing maturity of tools would be of benefit to the system. But that's not to say that if Sony switched architectures, that the system wouldn't also be approachable... and in fact in spite of the switch might still become moreso. Like, what if they went straight x86 OOE for example? Not saying that's what I think would happen, but just trying to get people to stop thinking about the investments already made in the architecture as irrevocably tied to PS4 ease of approachability.

You're definitely correct with your point there. I guess I'm just thinking that it would by far allow exising devs to hit the ground running with existing tools/libraries if Sony went CELL once more, rather than having to tweak their existing code bases (however significant a tweak that might be).

I guess for the most part, the benefit of choosing CELL in PS4 means you can leverage developer past experience/tools/libraries etc and you would also be in a much better position regarding BC (however important Sony decides it to be).

[ShadowRunner]

Originally Posted by specwarGP2

With physical media, you can always sell or give away your old games. With DD you're stuck with it, and I think the perception is that if I'm stuck with it, it better work in perpetuity. I know if I bought a movie from itunes, and some time down the line it stopped working, I'd be really annoyed and I'd never buy anything from itunes again.

Its a good point, though i dont see BC as the best solution. Id rather they alowed me to trade/gift/sell my content on a built-in marketplace so that DD isnt always at a disadvantage in this respect. If im stuck with something i dont want anymore, and would have given it away or sold it, BC doesnt help at all.