Reports out of Japan suggest that the Cell chip could be used to power Sony’s Playstation 4 console. Using Cell would help reduce development costs for Sony, which spent billions on R&D for the Playstation 3. In addition, the move would also help Sony bring the PS4 to market sooner, a plan both Microsoft and Nintendo seem to have.
According to the article, Sony could increase the number of processing cores to 24 or even 32 in order to improve performance. Higher clock speeds could be in the cards as well. Add a next-generation ATI or NVIDIA GPU, Blu-ray optical drive, and DDR3 RAM, and the PS4 could be easily introduced in time to compete with Microsoft and Nintendo.
But with the complications and difficulties that quite a couple of developers have had with while designing for cell, will it get better? It seems to me like this move would just make the process of creating games for the system even more difficult, with the added cores and such. If that is the case, I would have thought Sony would want to learn from their mistakes and make the Playstation 4 easier to develop for and not build on what is an existing problem for them. There’s no point having the most powerful console if no one can take full advantage of it’s potential.
Last Updated: October 2, 2008
Lupus
October 2, 2008 at 14:14
Yes there is, so people can wave it around going, I is more powerful then you.
baba
October 2, 2008 at 14:35
I would have to disagree with you Gareth. 3-4 Years from now developers would know the cell inside out.
Just think about it: The PS2 was also a beast to develop for with the emotion engine. Compare the games from 7 years ago to the titles on the market now (for PS2). Now imagine they kept the emotion engine for the PS3, obviously with improved technology…
I think it will be an excellent move for Sony, they can just build and refine their current development kits, and when the PS4 arrives developers will have familiar technology and tools they would have been working with for years.
Nick
October 2, 2008 at 14:44
You do make a good point however wouldn’t it still be an issue if the next xbox console has even simpler and better tools than now. So while those xbox developers are creating technically amazing games with even easier tools, the playstation tools will only be reaching a point of adequacy.
Im not saying that this is the way things will happen, but you must wonder if Sony arent shooting themselves in the foot by not concentrating on making easier tools and simpler architecture.
ewie
October 2, 2008 at 14:58
There is also the issue of Sony/Nintendo using opengl and Microsoft using directx, where dirext x are wholy integrated
within the development cycles of the the card manufacturers, opengl is lagging behind because all role players can not agree on the way to implement it. So the developers on Sony is restricted by opengl as well.
Milesh Bhana ZA
October 2, 2008 at 15:19
I think in the “tools development” side of things MS are definitely the leader, purely because they leverage the IPs they have from the work they do with Visual Studio.NET.
I really think that the success of the PS4 is going to rely more on the software/tools than the hardware.
Let’s face it, it’s easy to throw lots of hardware in a box, up the RAM, get nVidia/ATi to make a phat gfx card etc.
easy
October 2, 2008 at 15:36
opengl is the only api available for multiplatform development, so it actually makes sense from a developers point of view.
opengl is not bad by any stretch, but its just not as future proof as dx. maybe opengl 3 will sort that out.
ewie
October 2, 2008 at 15:50
Interesting Article.
on Opengl 3 vs directx 11
Scotty777
October 2, 2008 at 17:53
well… I suppose it’s better to have a known devil then an unknown one… Which would be why Sony will stick with the cell… They would much rather stick with a chip, refine it, drop the fab process(to say 32nm) and then add more “cells”, simply because they know that jumping to a new architecture has a greater chance of failure… And as stated before, the developers will be very accustomed to the ps3’s tools, so when they are given more hardware to play with, they’ll know how to use AT LEAST 7 spe’s, and the others will be added bonuses, and things to play with…
Now, i think that sony is WAY to far into their cpu’s. Honestly, having a powerful gpu is better then a powerful cpu…
gazza
October 3, 2008 at 02:08
Regardless of weather developers are acustomed to deving for PS3 or not the main point here is that its still takes time, way more time than it take for the same dev to make a game on 360. Yes the PS3 has potential, cell is good, it’s just not right for gaming imho, and for them to make the cell more complicated would add more time. Would MGS 5 for instance still be worth it if it took 10 years to dev?
baba
October 3, 2008 at 11:04
If you understand the unique problems associated with multicore game development you will know why development takes longer on the PS3. I personally feel that is the way to go in the fututre, and a lot of CPU manufactures feel the same way. As programmers learn how to approach problems and code efficient code on multi core CPU’s, the development time will shorten.
And I think the Cell is more suited to gaming than most of us realize. It has unmatched floating point performance which is essential in modern 3D gaming. IBM can just bump up the cache on the PPE and perhaps add branch prediction while they’re busy and they’ll have an absolute killer gaming CPU.