1337 Gamer said: wow big deal the cell with its 7 SPEs can beat the I7 at folding@home thats not that surprising but the I7 rapes the cell (or the 360 CPU for that matter) at whats improtant. Gaming and real world applications

Thats not completely true. In the medical field, which is a real world application, the CELL destroys the i7 because it can render 3d medical images in a fraction of the time it would take even the Core i7 275. The CELL is also faster at encoding and decoding video which is why Toshiba created the Spur Engine which contains 4 SPEs to assist a general processor like the i7s.

dahuman said: I use Linux for server purposes(most efficient form of servers) since a lot of the software that I use are not well supported or even on Linux. on my work machine, if I need something done on Linux, Virtual Box is generally my answer. I do think it's a good OS and it has a good amount of features that neither Windows or OS-X can bring, but it's not as good for multimedia purposes(EVR is superior when counting playback quality abilities, and a lot of editing software are simply lacking,) I also need DX(10) for obvious reasons =/. and since you have experiences, goto doom9 forums and tell them what you said and I'll bet you'll get destroyed very fast =P(about encoding and decoding that is.) real video was never a good format for quality distribution though, they've improved over the years but they still have lousy audio codecs and the video quality is only better because they've updated to 264 type video codecs recently, but the options are terribad(not a real word =P.)

All I put was the general text book definition. IF you don't like it bitch at the definition.

@ Deneidez

Maybe you lack the talent, but the PS3 is much easier to program for than the PS2, you can just do a whole lot more. Any coder I know could get started with the Cell SDK right away. It's far from abra cadabra.

However what using the Cell does point out well is bad design, if you have to deal with porting a lot of badly designed legacy code that may cause a few headaches if you want to harvest great performance.



One thing that's well illustrating, is that the weakest gaming hardware platforms (mobile phones) heavily rely on Java. If what you say would be true, this would not be feasible. There are JIT engines, there are techniques to translate Java into native CPU code during loadtime rather than during runtime.

Deneidez said: Kynes said: CGI-Quality said: I'm baffled by the some of the negative reception to this. Why is this news so bothersome to some? Not cheerleading the Cell is not saying that the processor is bad. What we need is some perspective, and realize the highs and lows of the chip. It's not the second coming, only that. Having a good FOH performance is not what we, as gamers, need. FOH doesn't translate directly to performance in games, that's all. Well, I am not saying either that CELL is bad. Its excellent when it comes to heavily parallel FP operating. Thats why they are using CELL for scientific stuff etc. Unfortunately games aren't that simple and actually CELL is more encumbrance than useful in games. (To be honest XENON isn't either that good for games. Its in order like CELL and that can actually nearly halve its efficiency in games. Both of them encourage to make simple games, because they can't really handle anything complex.)

To design a chip with OOE you must necessarily reduce the amount of cache chip has as it becomes dedicated to your instruction pool. Really you only want it for badly optimized applications.

@ Deneidez

What a nonsense, I thought Crysis was much (most) hailed for its technical excellence by PC gamers and was originally fully designed and geared at PC technology. Crytek devs state they don't find the PS3 hard to develop for at all and their engine is running just fine on both consoles already. I expect their gaming engine to advance a lot over the years, tapping more and more SPE resources, their PC gaming engine was already very advanced and highly multi-threaded.

All you lovers of the Cell should love the latest GPU's

They are 4 times faster than the Cell

CommunistHater said: All you lovers of the Cell should love the latest GPU's They are 4 times faster than the Cell

As long as your computations are all lock step... and you don't need to do division.

@ Soleron

Actually XDR provides well lower latencies, that must have been Sony's reason for going with more expensive main CPU Ram (yes, the GPU can tap it as well increasing graphics bandwidth, but it does not need to within the PS3 design).

MikeB said: @ Deneidez And I really would want to see PS4 with more CELL, that would mean pretty much end of playstation. If programming gets any harder than it is already, nobody would use it. Maybe you lack the talent, but the PS3 is much easier to program for than the PS2, you can just do a whole lot more. Any coder I know could get started with the Cell SDK right away. It's far from abra cadabra. However what using the Cell does point out well is bad design, if you have to deal with porting a lot of badly designed legacy code that may cause a few headaches if you want to harvest great performance.

I also would like to add that moving PS3 Cell code over to a more advanced future PS4 Cell would in fact be a lot easier than having to deal with moving legacy PC code or PS2 code to the PS3 Cell. It's more likely they will add more convenieces rather than take away what's already there in the core design.

IMO the PS3 provides a great stepping stone towards future Cell based systems. And going Cell all the way will not hamper moving code over towards other processors at all, in fact the code will be cleaner and should yield better performance with other multi-core or multi-CPU architectures as well.

alephnull said: Deneidez said: Kynes said: CGI-Quality said: I'm baffled by the some of the negative reception to this. Why is this news so bothersome to some? Not cheerleading the Cell is not saying that the processor is bad. What we need is some perspective, and realize the highs and lows of the chip. It's not the second coming, only that. Having a good FOH performance is not what we, as gamers, need. FOH doesn't translate directly to performance in games, that's all. Well, I am not saying either that CELL is bad. Its excellent when it comes to heavily parallel FP operating. Thats why they are using CELL for scientific stuff etc. Unfortunately games aren't that simple and actually CELL is more encumbrance than useful in games. (To be honest XENON isn't either that good for games. Its in order like CELL and that can actually nearly halve its efficiency in games. Both of them encourage to make simple games, because they can't really handle anything complex.) To design a chip with OOE you must necessarily reduce the amount of cache chip has as it becomes dedicated to your instruction pool. Really you only want it for badly optimized applications.

Yes, you are right. OoOE make chips bigger and more expensive. Anyway, it haven't been invented for nothing. Also are you implying that most advanced AIs are badly optimized, because they might have too many branches (and way too big reserved memory block) Cell to handle? :)

(And before MikeB is going to dump some stupid crap about Cell simulating human brains. I know how neural networks work and what they are good for. You really can't use them for AI. Teaching would take forever and it would still be very simple vs traditional methods.)

MikeB:

I don't lack talent to use CELL. Unfortunately many others do and I can understand why.