By using this site, you agree to our Privacy Policy and our Terms of Use. Close
Staude said:
Chark said:
Staude said:
One one hand, I'd prefere they keep it, so you'll be able to play ps3 games on the machine.. And because I think the structure, when used properly, is badass.

On the other hand. A lot of people who are not sony, have problems using the structure properly, and a more common structure, will likely allow developers to invest less into making games on the machine.. Since they wont have to figure out how it works first :P


Would the lack of Cell prevent ps3 games from playing?

That is a good question.

The Cell processor is structured uniquely, in that it has 1 main core, and 8 sub cores. Of these subcores 1 is disabled for a higher yield in production.
The way the subcores are structured, means the processor is very good at some things, but very bad at others.
The things it's very good at however, it does far better than ordenary processors and even some powerful GPUs (which is the reason their original plan was to use the cell as a GPU aswell as the CPU, instead of including a standalone GPU component.)

Now, while you would assume that you would theoretically be able to multithread emulate the playstation 3, the fact that it moved things so uniquely, and so fast between it's cores would mean that you would have to have a much much more powerful system capable of not only moving more since it would be running software designed to make the processor emulate a different processor, it would also have to be much faster, as it would need to know that it had to emulate, while emulating, at the same speed.


I really like the concept of the CPU, I think that they could have had a machine with 4 main CPUs and an additional 16 sub CPUs (SPUs) for a powerful next generation system :P .. With larger L2 cache, and more XDR ram.. Well, or XDR 2.. It probably wouldn't be very cost effective though.

The biggest issue with the Cell was that most multiplatform game developers, used to working on X86 systems (32bit pc architecture) decided to simply use only the main CPU, or at best, very little of the additional CPUs. It's a shame though, because when used properly, you could have the CPU do a lot of the GPU work, which is why sony has produced better looking games on the machine compared to the competition, despite the slightly inferior GPU in the ps3.


I suppose this would sum up why I personally hope they would retain the cell processor in the PS4.  Given that Sony has established a team of talanted developers willing and perhaps better at developing under a cell architype, it would make sense.  Even the more "traditionalist?" developers have been delving into Cell to provide multi platforms, it wouldn't make too much sense for it to just be dropped.  IBM even created the 32nm chips that haven't been taken advantage of.  There will be an incentive for 3rd parties to learn the architype so they can acess the full market and having a non-uniform architype will also provide Sony with a unique edge.  The only downside I see, since the cell's introduction has already been handled, would be a fear that having 2 other systems run on lower quality cpu/gpu next gen would isolate Sony too much from the main market.  Personally I would be pissed if Microsoft released a budget, lower end console to match what appears to be the concentration of Nintendo's Wii U.  It might make sense for Microsoft, if Nintendo releases something next gen that seems current gen, to release something that's not that big of a step up so they can keep all of the developers working on last gen software releases.  It would choke out Sony if they were to push ahead with technology while no one else did.  It would upset me if that happened and Sony loss out on it and it would upset me if Sony played into that concept as well just to say itself because it had to. In fact so much anger just thinking about it.  10 year life cycle shouldn't turn into a facade 20 year life cycle.



Before the PS3 everyone was nice to me :(