PS3 vs PC

[alephnull]

 

HappySqurriel said:

MikeB said:

The PC and PS3 aren't easily comparable, one runs Vista and the other a far more optimised embedded OS, one will be optimised for to the bone the other will not.

You are correct that the full potential of the Cell processor hasn't been shown yet. IMO that's not a surprise when dealing with regard to radically new innovative hardware. For example early Atari ST (which was good compared to Macs and PCs of the time) to Amiga 500 ports ran better on the Atari ST, small enhancements started to pop up in the coming years though, after 2 years the first eye few eye opener titles started to appear:

http://it.youtube.com/watch?v=e5ic3Fy-tVY

But it took until around 6 years to see the Amiga 500's gaming potential fully realized. Personally I expect a similar timespan with regard to the PS3, like I stated a year before the PS3's launch. If we talk about other advantages like Blu-Ray disc and default harddrive they should show much quicker.

I think the PS3 will be viewed as a cutting edge gaming system for a long time due to this growth. It also potentially paves the way for a Cell based PS4, which should thus be much easier to adapt to for PS3 developers.

Since the days of the Amiga, NeoGeo and later the BeBox we have said heavy multi-processing will be the future. The success of the x86 IBM PC this delayed progress for a decade but now it's being realized.

There have been developers entirely devoted to the cell processor for nearly 4 years and it hasn't really broken away from the performance of the XBox 360 or a 3 year old PC. Its about time that you admit that the Cell is not a magical, law of physics breaking, processor and it is very similar in performace to the other processors. Certainly, there will be improvements over time as developers work with it, but it isn't a cutting edge gaming system anymore and it never will be again.

 

I've given up arguing with people making statements of the form A is "more powerful" than B on this forum, and so im going to ignore the thread owner. I will pick on you however :P.

If IBM's linux on cell division was having trouble gettting it's hands on workstations for cell developement when I talked to it's manager about a year and a half ago now, then I don't think it's unreasonable to suspect that other developers might have had some difficulties. Most of the people for the majority of those 4 years you are talking about were doing most of their dev work on simulators. With a buggy compiler.

@ZenfoldorVGI -- I fear the ignorance seen in so many PC gamers recently of current major architecture problems yet to be definitively overcome -- such as, the problems that occur as you increase the number of cores and insist on maintaining cache-coherence, potential bus congestion issues, etc -- has likely played a decisive part in this tragi-comic spectacle whereby so many people waste money on these quad-core chips which are more likely to cause performance decreases for most games compared to many less expensive chips. I think a better understanding of the problems the CBE was intended to address (even if you don't believe it was successful) would have at least saved you.