BTOA said:
We have seen PS3 demostrating Cell's physic capabilities with their tech demos shown at E3 and Sony's Summer event, but I don't recall much or any Xbox 360 demo showing off Xenon's physic capabilities.
I'm not basing that on demos, I'm basing that on what AGEIA said. What I wrote in the post you quoted is a pretty accurate reflection of AGEIA's stance I think.
To be very explicit, look at AGEIA's own slide from CEDEC:
Translation from one:
" Scalabilities for Advanced Hardwares
Xbox 360
+ Just like dual-core PC, but uses 3 PPC cores with HT
+ Shared L2 cache
+ Probably shares the cores with other Xbox 360 libraries
PS3
+ PPC processor
+ Multiple "Synergistic Processing Elements" (SPE)
+
Greater simulation potential than PC or Xbox 360
+ All simulation classes are available"
Look at what SenatorMonkey reported from GDCE:
"Having attended this talk today, it was slightly different than noted above. The comparison was that on a single-core system, it ran at about 5fps and occupied the full processor. On a dual-core system with the PhysX chip, it ran at about 40fps, and took only 25% of one core.
Regarding having the fluid dynamics and cloth simulation be more suited to PS3, this was mainly due to cache issues according to the presenter. They're still looking at the X360 architecture to see exactly what they can get out of it, but right now they're concerned that they'd take up too much CPU to do these things through Novodex, since they'd like up likely one full core or more processing power. I believe that on PS3, they figure in 1 SPU to do the same, since it's massively iterated calculations on smaller datasets, which is perfectly suited to the SPU architecture."
So they certainly did make quite explicit comparative statements between these systems at these conferences, even if they "weren't meant to". And they don't deny that they made these statements either.
Add in the PR quotes from this statement:
"The presentation uses implied assumptions on the relative power of the platforms from single core, dual core, console and PC-with-PhysX platforms, based on publicly-available information. The only platforms for which there is actual comparison data at AGEIA are the single core PC, dual core PC and PhysX platforms. There is no current data on the PS3 or Xbox360 on relative performance.
The implied difference in performance was from assumptions about the number of compute elements and memory architectures, and how well these might fit to various simulation algorithms as enabled by typical game developers. The difficulty in predicting performance across platforms is that physics is inside a game loop where many other game-related processing is taking place. Again, we don’t have a simple metric like fill-rate that is greatly dependent on the capabilities of the graphics chip and less dependent on what is going on in the game."
And I'll add another quote from Andy Keane in a follow up email to my own further queries:
"Any analyses on differences that I have seen to date are analyses of the architectures from specifications released by the console maker."
Note that they did not refute these analyses - in fact they recognise and acknowledge that they were there - they now simply class them as being based on architecture etc. rather than benchmark data, and threw in the variability of the game loop as an example of how difficult it would be to benchmark (even though this did not stop them from doing so with their own application on PhysX and dual/single core PCs
).
So putting all that together I think it's reasonable to
at least say - as I had earlier:
"They indicated at the conferences that PS3 had more capacity/potential when running novodex than X360 and/or that it had a more suitable architecture for fluid/cloth dynamics. The official PR line now is that any suggestions made at these conferences was prediction/expectation based on architectural/system analysis rather than benchmark data. So..draw your own conclusions I guess."