Nvidia Responds to AMD's Claim of PhysX Failure

Page 2 - Seeking answers? Join the Tom's Hardware community: where nearly two million members share solutions and discuss the latest tech.
Status
Not open for further replies.
G

Guest

Guest
Wait, does this mean that PhysX isn't a waste of time and money?
 

ljbade

Distinguished
Oct 28, 2009
19
0
18,510
Sure the PhysX SDK may allow developers to choose to make their PhysX apps multi-threaded, and many will (3dmark).

But the real question is, does NVIDIA pressure The Way It Is Meant To Be Played game developers to not multi-thread their PhysX apps.
 
To me physx is like the US economy both are rigged to be in favor of a select few while every one else either pays the price or miss out. Personally I hope that Nvidia suffers for a few years in order to learn from their on going mistakes and become a better company to its competitors as well most of all us the consumer plus if it wasn't for gamers this company would have gone belly up during the .com bubble. If they had ATI or 3DFX would be the main players with Matrox offering a few consumer cards that didn't cost a fortune but offered much of the same quality as they do today. Perhaps if some other player came on the scene with out Intel sucking every thing up if Nvidia failed then maybe a little Renaissance for PC gaming would pop up like what happened with 3dfx but with out better long lasting results. I wish that Amiga was still around and popular like they were.
 

liquidsnake718

Distinguished
Jul 8, 2009
1,379
0
19,310
Being an avid PC and game player I 100% fully support Nvidias PhysX. Since replacing my former Nvidia GPU with a 5850 I am not as pleased with the Ati cards' physics code. In several games, the physics preformance is poor when comparing to a lessor Nvidia card. Even though there is an increase in FPS from my 9800gtx+ in games like Crysis, Batman AA, GTA4, and Far Cry2, the physics somehow feels "off" or the calculations allow the realism in the Ati card feels faster. Example, the trees sway way too fast and unrealistically in FC2 while in Crysis the barrels and throwing of an object and the way it "bounces" and "rolls" of a rocky edge is not as fluid and realistic as the PhysX run engine.

Anyone else notice this?
 
[citation][nom]liquidsnake718[/nom]Being an avid PC and game player I 100% fully support Nvidias PhysX. Since replacing my former Nvidia GPU with a 5850 I am not as pleased with the Ati cards' physics code. In several games, the physics preformance is poor when comparing to a lessor Nvidia card. Even though there is an increase in FPS from my 9800gtx+ in games like Crysis, Batman AA, GTA4, and Far Cry2, the physics somehow feels "off" or the calculations allow the realism in the Ati card feels faster. Example, the trees sway way too fast and unrealistically in FC2 while in Crysis the barrels and throwing of an object and the way it "bounces" and "rolls" of a rocky edge is not as fluid and realistic as the PhysX run engine.Anyone else notice this?[/citation]

I don't think you know what PhysX is. Crysis, Farcry 2, GTA4 do not do physics on the video card. It plays no part in the physics. You are experiencing a placebo effect, or maybe now that you see a lot more FPS, things react differently.

The only game you mentioned there that has physX and the video card played a role in the physics, is Batman.
 

randomizer

Champion
Moderator
[citation][nom]liquidsnake718[/nom]Example, the trees sway way too fast and unrealistically in FC2 while in Crysis the barrels and throwing of an object and the way it "bounces" and "rolls" of a rocky edge is not as fluid and realistic as the PhysX run engine.Anyone else notice this?[/citation]
What's a game that uses PhysX for more than special effects? If I knew one, I could probably tell you if it looks good or not. As yet, I can only think of Batman but that's just smoke and flying paper. It's hardly a game changer.
 

AsAnAtheist

Distinguished
Sep 15, 2009
790
0
19,060
I found the manipulation of words by Nadeem Mohammad.
The theme of the whole controversy was Nvidia skimping performance with a competitor's card.
Nadeem's main theme or point was "...there have been no changes to the SDK code which purposely reduces the software performance of PhysX or its use of CPU multi-cores."
What does this mean? That Physx will use multicores. Whether or not it will use multi cores at the same efficiency as Nvidia cards vs ATI card is a different issue. Notice how they never tackle the main issue which was Nvidia's anti competitive acts. They also say it's up to the developer to program the threading. We already saw with Batman Asylum that Nvidia locked AA out of ATI cards because of "copyrights", we can already guess they added incentives or threats to forget to optimize physx for ATI cards.

Shame one you Nvidia, you really need to step up your game or at least use PR as well as Apple that you can steal hundreds or thousands of dollars and still have happy clients. First ludicrous Fermi claims, then fake Fermi GPU's, now anti-competitive actions. I understand you been bullied already by Intel's anti-Nvidia campaign against the Ion, but seriously can't you learn by their anti-trust mistakes?
 

matt87_50

Distinguished
Mar 23, 2009
1,150
0
19,280
I agree that an i7 (and alike) could easily handle even the most intensive physics in games I've seen. and IF nvidia were ditching the PC and trying to convince us that it can "only run on GPUs (our GPUS of course)" then that would be shit, and I wouldn't put it past them at the moment. however I DO doubt that this is the case, and I say nvidia should focus on getting it running on the gpu more than the cpu, because that is the future, anything that can run on a gpu ABSOLUTELY SHOULD BE. if that leaves nothing intensive left to run on the CPU then SO MUCH THE BETTER, thats progress. cpus can't compete in terms of processing power with GPUs, GPUs are far ahead and continue to outpace the growth of cpus. the only ace up CPU's sleeve is that it can do some forms of processing that GPUs really can't, physics is NOT one of those things.

of course I DO NOT support nvidia's dedication to PhysX and CUDA because these things ONLY work on their cards, its anti competitive, its not that they couldn't work just as well on other cards, its just nvidia don't allow it. They should focus more on OpenCL and DirectCompute ect.
 
I wouldn't be surprised to one day see a Physics SDK that takes advantage of Direct Compute in order to compete with PhysX. While one might make a case that nVidia could proliferate PhysX before that happens, they haven't yet. Considering that developers will be moving to DX11 and nVidia has yet to put out their DX11 card, getting developers to pay attention to PhysX in the near term isn't looking all that good :D.
 

DjEaZy

Distinguished
Apr 3, 2008
1,161
0
19,280
... how many popular physx titles are out there? And why nVidia disables phiysx in drivers when ATi is running as a primary card? What physics ir DiRT2 using? And with OpenCL, DX11 and Direct Compute is physx fading anyway?
 
Sorry but PhysX is a fial to me. HavoK owns the physics API market and probably will continue to since they actually utilize more than just proprietary hardware which gives a larger base meaning the game company can design a game that will run better with more physics effects on more machines.

nVidia is just in denial.
 

stridervm

Distinguished
Jan 28, 2008
645
0
19,010
If I remember correctly PhysX is now "optimized" for nVidia GPUs and now even the old Ageia PPU doesn't work with the latest PhysX runtimes. I think there are also some discussions in other forums that mention degraded CPU PhsyX performance when comparing old and newer PhysX runtimes as well.
 

spinaltap11

Distinguished
Apr 23, 2008
3
0
18,510
I'm quite startled by the large number of comments here that completely miss the point. I think that Nadeem Mohammad pretty clearly rebuts the original statement.

If PhysX implementations are not running well on multicore, it's because the game developers did not implement multicore code for PhysX. He even cites one example of a developer who has implemented multicore PhysX code properly (Futuremark). It's not up to NVIDIA to enforce programming practices at game companies for CPUs in the "The Way it's Meant to be Played" campaign, and it's silly to assume that they would do that.
 

Leopardos

Distinguished
Oct 8, 2008
179
0
18,690
Nadeem Mohammed .. im Proud of u. Holding an Dir. Product Management at NVIDIA , Nice to see a brother hold this positions .. beat them all :p
 

anonymous x

Distinguished
Apr 22, 2008
121
0
18,680
[citation][nom]spinaltap11[/nom]I'm quite startled by the large number of comments here that completely miss the point. I think that Nadeem Mohammad pretty clearly rebuts the original statement. If PhysX implementations are not running well on multicore, it's because the game developers did not implement multicore code for PhysX. He even cites one example of a developer who has implemented multicore PhysX code properly (Futuremark). It's not up to NVIDIA to enforce programming practices at game companies for CPUs in the "The Way it's Meant to be Played" campaign, and it's silly to assume that they would do that.[/citation]
+1
Is it so hard to read people?
 
Please people, NVIDIA hasn't locked AMD from supporting the PhysX API, AMD has simply refused to support it themselves.

And I'm tired of people saying "The CPU can run Physics fine". All current physics implementations STINK! Physics should determine if a bullet passes through an object, how much force it has when it hits a person, whether or not is passes through armor, and finally, how much damage it does. Nope, instead we get X damage for one size bullet, and y for another.

I want the power to use Physics to play the way I want to play; I want to create avalanches on the fly, destroy key enemy positions dynamically, etc. Not determined by what scenereos the devs though of, but done entirly through the API itself.

A true implementation of physics will be an even more demanding task then Rasterization, which currently requires discrete cards to take the load off the CPU.

Also, NVIDIA is right; like most API's (DirectX, etc), implementation is left up for the developer. The API itself is still parallel in nature, and can be scaled properly with minimal effort. I'm chalking that batman AA numbers on sloppy coding on the devs part.
 

lradunovic77

Distinguished
Dec 20, 2009
405
0
18,780
Physx is available for AMD to implement but they refuse it. It's AMD/ATI problem and seems that they more bitch around. CPU can't handle physx they way GPU can cause in term of raw power computing CPU is joke.
 

techguy911

Distinguished
Jun 8, 2007
1,075
0
19,460
[citation][nom]elie3000[/nom]Were mad gamers because PhysX is actually Slowing the Multicore game optimization process. Look how bad games are multicore optimized while they can run perfectly on a Quad Core if done efficiently. They would rather let the GPU make the physics calculations and forget about the CPU.[/citation]


Most games don't take advantage of multi-cores and 99.50% of software out there is dx9 they still dont use dx10 much and dx11 is non-existant.

Don't forget PhysX is for nvidia cards only anyways and is optimized for gpu's not cpu's although they could let the cpu do it with multiple but your missing the point letting the gpu take load off the cpu, that is what the original board did as well.
 

phantomtrooper

Distinguished
Apr 17, 2008
194
0
18,690
If there is code that limits the CPU then it was there when Ageia was running Physx. I had an old PC with a really slow CPU that could not run physics stuff. When Nvidia bought them out and they incorporated Physx with the GPU suddenly I could handle the physics stuff.
 

juliom

Distinguished
Jul 21, 2009
84
0
18,630
[citation][nom]liquidsnake718[/nom]Being an avid PC and game player I 100% fully support Nvidias PhysX. Since replacing my former Nvidia GPU with a 5850 I am not as pleased with the Ati cards' physics code. In several games, the physics preformance is poor when comparing to a lessor Nvidia card. Even though there is an increase in FPS from my 9800gtx+ in games like Crysis, Batman AA, GTA4, and Far Cry2, the physics somehow feels "off" or the calculations allow the realism in the Ati card feels faster. Example, the trees sway way too fast and unrealistically in FC2 while in Crysis the barrels and throwing of an object and the way it "bounces" and "rolls" of a rocky edge is not as fluid and realistic as the PhysX run engine.Anyone else notice this?[/citation]

What I noticed is that you have no idea of what you're talking about. Crysis does not use PhysX. Far Cry 2 does not use PhysX. Maybe the faster tree shaking is due to the much better frame rate you're getting? Please, stop trolling the forum.
 

Regulas

Distinguished
May 11, 2008
1,202
0
19,280
[citation][nom]Murissokah[/nom]The response sounded quite well founded. Don't think Nvidia is to blame on this one.[/citation]
I agree with your post. By your flames (Thumbs Down) I would say this is a AMD/ATI fanboy site.
Quote from Nvidia in article, "Our PhysX SDK API is designed such that thread control is done explicitly by the application developer"
I agree with you. Nvidia says it lets the game developers handle the code but for some reason theses tards on this site still want to blame Nvidia.
 

deanjo

Distinguished
Sep 30, 2008
113
0
18,680
Looks and sounds like AMD is taking one last swing at nvidia hoping it would sustain sales until Fermi is released and they get knocked down back to #2.
 
Status
Not open for further replies.