PhysX: PS3 can handle it; 360 features limited

[scooby_dooby]

They may not have benchmarked the platforms to determine relative speeds, but I'm sure they have comparative results which no amount of pressing will get them to release.

i'm sure they do too, but the question is how meaningful would the numbers be at this point anyways?

it's not an in-game scenario, it's not tailored for either of the platforms, I dunno I think this is being blown a little out of proportion.

 

[Nemo80]

i'm sure they do too, but the question is how meaningful would the numbers be at this point anyways?

it's not an in-game scenario, it's not tailored for either of the platforms, I dunno I think this is being blown a little out of proportion.

One could also say they ran unoptimized code on both platforms, just to see how it works. Since the Xbox is easier too develop for and handles GP code better in first place, that could give a little clue of what to expect ...

 

[Shifty Geezer]

i'm sure they do too, but the question is how meaningful would the numbers be at this point anyways?

Probably only a little more meaningful than FLOPs . But even though they're not games and aren;t indicative of what to expect in-game, I'd like to see what the hardwares are capable of, replative to PC as well as each other. PS3's s'posed to be great at physics. I'd like to se some comparative numbers to show how much so.

 

[Titanio]

Maybe it's just possible that what they are saying is the actual truth, did anyone consider that?

Certainly did, but I found other conclusions more probable Well certainly the questions I asked myself were:

1) Do they have data internally?
2) Would they make the explicit comparisons in these presentations if those comparisons were being contradicted by internal data?

The answers I found were 1) Probably (for dev kits only of course, not finished hardware) and 2) Probably not. Other questions could also be pertinent, like for example, whether AGEIA qualified their statements at the presentations themselves or if they're only doing so after the fact, and how likely you think it is MS was pissed at unflattering comparisons from a partner.

Also, on benchmarking, it's odd that Keane discusses its difficulty on the one hand, but on the other AGEIA are more than happy to provide such benches themselves on other platforms. Just the consoles coming in for special treatment. Wonder why?

Anyway, it feels like we're raking over well worked-ground at this stage. Nothing Keane has said here differs from his previous statements, just a different report of the same stuff.

 

[scooby_dooby]

Meh, it seems fairly probable to me that these statements were made based on assumptions not supported by hard data, and maybe on just the initial performance of the ported SDK, or even assumptions based simply on the differing architectures.

I mean, a statement like the x360 cannot do FD is obviously not true, and based only on assumption, how much time have they devoted to making their SDK run FD on the x360? It sounds like they've haven't really done much.

I think the most probable explanation, and what the Ageia guys were trying to say, is that the PS3 will be better at phsyics than the x360, and most especially in the area of FD. However, this is nothing new to us, and is exactly what we expected from CELL. Furthermore it's only their initial assumptions and has yet to be proven IRL.

 

[ps2xboxcube]

I think that people who say Ageia is lying about their explanation just want to beleive and convey to others that the PS3 is better than the XBOX360.

Wednesday evening, Ageia vice president of marketing Andy Keane acknowledged that including those comparisons in its presentation was a mistake, and that they were based on speculation from public specifications released on Microsoft. Although the company has tested its software on Xbox 360 development kits, Keane said, the tests were of extremely limited scope, and weren't designed to test the limitations of the hardware.

"The summary of the information below is that AGEIA would like to go on record that we do not have data to support performance comparisons for the PS3 or Xbox360 that would impact any of our physics features," Keane said in an email, which was also sent to other news organizations."Specifically, statements that the Xbox360 cannot run fluid simulations are not correct. In addition, conclusions about relative performance should not have been stated or implied in our presentations."

The benchmarks run on the PlayStation 3 and Xbox 360 simply tested to see if the SDK would run, and do not provide performance data, unlike the performance tests run on the PC systems.

http://www.extremetech.com/article2/0,1697,1856641,00.asp

 

[hugo]

Well the PS3 's GPU hasn't been taped out and you've got a CPU which has 7SPUs/1PPU don't you think it's better?
The Xbox360 is expected to hit the shelves soon there's no way it can be better in spec wise.

 

[Shifty Geezer]

I think that people who say Ageia is lying about their explanation just want to beleive and convey to others that the PS3 is better than the XBOX360.

I don't think people are saying Aegia are lying so much; just being conservative with the truth for diplomatic reasons. I'm sure they tried their SDK on the different platforms and pushed them to see what the could do. After all they've got new toys to play with. What red-blooded male wouldn't throw more at the hardware and see how long before it broke?

"Okay, the rock simulation runs fine on the PS3."
"I've marked that down. How quick was it by the way?"
"Kept at 60 fps."
"6,000 rocks. Cool. How many do you think it could take?"
"Dunno. Let's up it to 10,000 and see what happens..."
"Whoa, still 60 fps!"
"Right, let's try 20,000 rocks."
"Hah. That's a bit much. Try a bit less..."
(After some playing later)
"So PS3's maxxed out on 12,000 rocks at 60 fps. Pretty darned impressive."
"Anyway we'd better test this XB360 SDK now."
"Yeah, that runs fine."
"Okay, I've logged that. 60 fps."
"Sure."
"Reckon it'll keep up with Cell?"
"Dunno. Let's give it a try..."

 

[_phil_]

just being conservative with the truth for diplomatic reasons.

You're right.Even if we're not allowed to extend generalisation to: XCPU<CELL,tthis is what their midleware solution is showing.

 

[Lysander]

truths are so simple, aren`t they

 

[_phil_]

truths are so simple, aren`t they

never enought for those who expect to find it in mainstream,orPR stuff