Omac said: Well I'm kinda new here and I'm not going to look for every post you have made. I still think you are wrong on the subject. Why don't you find me some article on what you preach and post about them so I can look at it. I think you are trying to weasel out of the subject at hand.

If you make a valid point he'll ususally just ignore it, never responding and just changing subject.

@ Omac

I've linked to various articles and have given various sources including IBM.

IBM's early testing regarding efficiency:

"Number of SPUs 1
SPEsim (GFLOPS) 25.12
Hardware (GFLOPS) 25.01
Accuracy (%) 99.6%

Since operations in each data block are independent from those in other blocks, the matrix multiplication algorithm is easily parallelized to all eight SPUs. Figure 5 shows that the matrix multiplication performance increases almost linearly with the number of SPUs, especially with large matrix sizes. Using eight SPUs, the parallel version of matrix multiplication achieves 201GFLOPS, very close to the theoretical maximum of 204.8GFLOPS. "

"Not the best of analogies, but this will teach everyone to be cautious when they see either side throwing around their Megahertz, dot products and GFLOPS. I’m not saying the numbers are 100% meaningless as there are numbers that are actually trustworthy, but its getting you all ready for what I’m about to tell you

#

GFLOPS is something that gets thrown around a lot, but it should be clear that the peak theoretical GFLOP numbers for both these CPUs are:
# 115GFLOPS Theoretical Peak Performance for 360 CPU
# 218GFLOPS Theoretical Peak Performance for PS3 CPU.

These CPU theories will not be achieved in real world performance. What IBM did when testing for theoretical peaks on both CPUs can't really be considered as representative of how the processors would actually perform in real world situations, because of the type of testing done is too controlled. It’s a much too perfect of an environment and game development is going to involve an unforgiving environment that doesn’t cater so well to the perfect environment the CPUs were tested under.

The GFLOP numbers for the PS3 were calculated based on 8 running SPE, so the fact that the PS3 uses only 6 SPE for game applications lowers the peak theoretical even further, as majority of the floating point work on the PS3’s CPU is done via the SPE. Each SPE has a peak theoretical of 25.6GFLOPS. So the total peak theoretical performance for all 6 SPE would be 153.6GFLOPS, but why is that number also not achievable?

In IBM’s controlled testing environment, their optimized code on 8 SPE only yielded a performance number of 155.5GFLOPS. If it took 8 SPE to achieve that, no way 6 will be able to and that testing was done in a fashion that didn’t model all the complexities of DMA and the memory system. Using a 1Kx1K matrix and 8 SPE they were able to achieve 73.4GFLOPS, but the PS3 uses 6 SPE for games and these tests were done in controlled environments. So going on this information, even 73.4GFLOPS is seemingly out of reach, showing us that Sony didn’t necessarily lie about the cell’s performance as they made clear the 218GFLOPS was “theoretical.” But just like Microsoft they definitely wanted you to misinterpret these numbers into believing they were achievable.

Even while taking all of this into consideration, the CPUs can’t reach those crazy performance numbers; the PS3’s cell still comfortably comes out on top in terms of overall floating point capability, but it should be known that the available power on the PS3’s cell will be significantly more difficult to harness than the available power on the 360’s CPU.

It’s also worth mentioning that even the PS2 CPU had more than twice the GFLOPS of the original Xbox’s CPU, but it didn’t necessarily lead it to being the performance winner. This time around, while the cell has the GFLOPS advantage, its advantage isn’t quite as big as the PS2 CPU had on the Xbox. This teaches us that there is more than one meter of real world performance."

@ Omac

I read that in the article and they are clearly wrong according to IBM.

And GFlops is one of the most objective means to determine peak performances. Clearly the tests indicate that the Cell can actually achieve these GFlops is the real world.

However for the Xenon there are no real tests made public. IMO it will probably be a lot harder to reach its peak ( 77 GFlops), especially in combination with the fact the CPU is sharing its memory bus with the Xenos as apposed to each SPE having dedicated memory and both GPU and CPU having dedicated memory busses on the PS3.

On the XBox 360 the CPU shared 22.4 GB/s together with the GPU.
On the PS3 the CPU has 25.6 GB/s all for itself (and the GPU 22.4 GB/s) (taking out of account the GPU can access XDR memory).

MikeB said: @ Omac I read that in the article and they are clearly wrong according to IBM. And GFlops is one of the most objective means to determine peak performances. Clearly the tests indicate that the Cell can actually achieve these GFlops is the real world. However for the Xenon there are no real tests made public. IMO it will probably be a lot harder to reach its peak ( 77 GFlops), especially in combination with the fact the CPU is sharing its memory bus with the Xenos as apposed to each SPE having dedicated memory and both GPU and CPU having dedicated memory busses on the PS3. On the XBox 360 the CPU shared 22.4 GB/s together with the GPU. On the PS3 the CPU has 25.6 GB/s all for itself (and the GPU 22.4 GB/s).

Obvious; PS2 did more gigaflops than the Xbox

Obvious; Xbox had better looking games

@ sieanr

I don't know, but I was most impressed by the God of War games for the PS2.

Like I pointed out in the original post, CPU Flops aren't important alone for instance bandwidth and other things also comes into play. The Cell has been proven to be a graphics powerhouse in addition to performing more ordinary CPU tasks. With the XBox being released a year later than the PS2 I am sure it has technological strongpoints as well. Dev tools being the most obvious, as the original XBox was more like a cutdown PC than a traditional console.

MikeB said: @ sieanr Obvious; PS2 did more gigaflops than the Xbox Obvious; Xbox had better looking games I don't know, but I was most impressed by the God of War games for the PS2. Like I pointed out in the original post, CPU Flops aren't important alone for instance bandwidth and other things also comes into play. The Cell has been proven to be a graphics powerhouse in addition to performing more ordinary CPU tasks. With the XBox being released a year later than the PS2 I am sure it has technological strongpoints as well. Dev tools being the most obvious, as the original XBox was more like a cutdown PC than a traditional console.

 Impressed is not the same thing as technically superior. You think using a fixed camera had nothing to do with the game having resources to achieve its high texturing and framerate?* Also, a lot of the cut scenes are done through FMVs, although they are well disguised.

 This does not mean the games had bad graphics. It does mean that those were great artistically, and you are discussing NUMERICAL graphical output. What you think of the graphics is irrelevant.

 Plus the actual point, which you seem to have missed, was about theorectical maximum FLOPS, and simply used the PS2 vs the Xbox as an example. You used an irrelevant point, which did NOT disprove the actual point.

 Frankly, I haven't seen you give ONE FACTUAL DSIPUTE of the gotfrag article. You claiming it's wrong or false is not a dispute.

 *To compare, FFXII and RE4 had hight texturing, and a free camera, but ran at 30fps instead of 60fps.

@ LordTheNightKnight

I love the camera in the God of War games, I hope they won't change this approach with God of War 3.

Technically I am most impressed with the bosses, IMO they've really showed off how powerful the PS2 hardware was for its time. Technically as well as artistically these are the most impressive games I have seen on "last gen" hardware.



Like I said, the CPU and GPU performance are relevant, but like I stated in the original post:

"A great specced CPU and GPU would be useless if the system has to deal with slow memory and not enough bandwidth to move around the data quickly enough."

I haven't digged into the PS2 hardware specifications, so I wouldn't state with certainty which approach is more powerful. I only know the PS2 marginalized the XBox market share and the XBox hardware was more powerful for FPS games.



Reread the thread.

@ sieanr

If you decide to quote me in your footnote, then please put the quote in context:

"But that would be a mammoth undertaking and involve a complete game engine redesign from scratch. Maybe 3rd generation PS3 (SPE optimised, harddrive caching and texture streaming enabled) first and second party game engines will start to perform at such levels."

Yeah, maybe in late 2010 we'll see a big PS3 game run it's shaders on the Cell instead of the GPU (something like God of War 4 probably). But in those days all the talk will be about the Xbox 3 and PS4.

Until then the graphics battle on next gen will be all about GPU only, and you know very well that the X360's GPU is at least as good as the RSX in the PS3.