NVIDIA Responds To GPU PhysX Cheating Allegations

Massman · 25th June 2008, 22:05

During the benchmark install, a runtime library is updated to allow the test to run on the GPU and then during the test, it addresses the benchmark DLLs to the GPU instead of the PPU or CPU. Nothing within the benchmark is changed at all. No software libraries or even a line of code changes in the benchmark whatsoever. The only thing that changes is that installer, nothing else. It has been said that the tests results look different on the screen when running with PhysX enabled on the GPU. And of course this is true, just as the screen results look different when you test on a dual-core CPU versus a quad-core. This isn't a graphics test; it's a physics test. 3DMark Vantage specifically scales more complexity into a scene to take advantage of additional physics compute resources, which of course is why it looks so different/better on a test run with PhysX processing on our GPUs. This is by design in the benchmark and if the folks accusing us took the time to run it, they would know that.

http://www.hothardware.com/News/NVID...ng_Allegation/

jmke · 25th June 2008, 22:21

Did they even read what the "cheat" was all about? Their driver release changes the CPU test. This is not allowed according the Futuremark guidelines.

"no VIDEO CARD driver release should impact the CPU test scores"

that's what the rules say. NVIDIA's driver changes CPU test 2 score. Thus.... it's not a valid driver for that benchmark

We have followed them to a tee and this new beta driver has not been submitted for consideration.

they are just responding to the "header" used by TheInq of their article

------

Mark Rein for UT3

Quote:

We couldn't care less about synthetic benchmarks. Real players play real games and in this case there is a real performance gain for a large number of users playing our game. The updated drivers that people are talking about allow properly-equipped Unreal Tournament 3 players users to use some amazing UT3 content that was specifically designed to take advantage of hardware-accelerated physics. So now that content works on high-end Nvidia graphics cards, as opposed to Ageia PPUs. This is a great thing as high-end Nvidia cards number in the millions, or tens of millions, compared to Ageia PPUs which obviously numbered far, far less. So Nvidia has done a cool optimization that allows their customers to get more value and performance out of their graphics cards. How anyone could possibly confuse this for a bad thing is beyond me.

How can it be beyond him that this driver releases changes the VIDEO CARD benchmark Vantage in an incorrect way not sanctioned by those who developed the benchmark? What's there so difficult to understand? Did TheInq ever say that allowing PhysX on the GPU was a bad thing for games? http://www.theinquirer.net/gb/inquir...ats-3dmark-177 I can't find it? It's obvious with the release of GT200 and HD4800 that Vantage will matter for e-penis contests and the PhysX enabled driver makes NVIDIA stand out nicely in result charts, right?

26th June 2008, 01:46

The constant optimization war is the main problem with 3DMark. It's gotten to the point that performance in 3DMark has almost nothing to do with actual game performance other than as a vague rule-of-thumb. This is just next step on the path of it becoming completely irrelevant to enthusiasts.

piotke · 26th June 2008, 08:59

This might not be a valid driver for futuremark. But for the end user it is. Better performance, WITHOUT loosing quality. Not as before, when a cheat caused degradation of the images.

T'is is just using the hardware in a better way.

jmke · 26th June 2008, 09:36

I agree that for games this driver rocks!

only FM needs to make a clear stand of whether or not they will allow GPU based acceleration in CPU tests for their benchmark. From what I've read, if they "ok" the PhysX driver, they are halfway there. THis will make Vantage a NVIDIA Tech demo

25th June 2008, 22:05	#1
Massman [M] Reviewer Join Date: Nov 2004 Location: Waregem Posts: 6,466	NVIDIA Responds To GPU PhysX Cheating Allegations During the benchmark install, a runtime library is updated to allow the test to run on the GPU and then during the test, it addresses the benchmark DLLs to the GPU instead of the PPU or CPU. Nothing within the benchmark is changed at all. No software libraries or even a line of code changes in the benchmark whatsoever. The only thing that changes is that installer, nothing else. It has been said that the tests results look different on the screen when running with PhysX enabled on the GPU. And of course this is true, just as the screen results look different when you test on a dual-core CPU versus a quad-core. This isn't a graphics test; it's a physics test. 3DMark Vantage specifically scales more complexity into a scene to take advantage of additional physics compute resources, which of course is why it looks so different/better on a test run with PhysX processing on our GPUs. This is by design in the benchmark and if the folks accusing us took the time to run it, they would know that. http://www.hothardware.com/News/NVID...ng_Allegation/ __________________

26th June 2008, 08:59	#4
piotke [M] Reviewer Join Date: Jun 2002 Posts: 5,003	This might not be a valid driver for futuremark. But for the end user it is. Better performance, WITHOUT loosing quality. Not as before, when a cheat caused degradation of the images. T'is is just using the hardware in a better way. __________________ .

26th June 2008, 09:36	#5
jmke Madshrimp Join Date: May 2002 Location: 7090/Belgium Posts: 79,021	I agree that for games this driver rocks! only FM needs to make a clear stand of whether or not they will allow GPU based acceleration in CPU tests for their benchmark. From what I've read, if they "ok" the PhysX driver, they are halfway there. THis will make Vantage a NVIDIA Tech demo __________________

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
NVIDIA Removes Restriction on ATI GPUs with NVIDIA GPUs Processing PhysX	jmke	WebNews	5	30th May 2010 11:13
NVIDIA Forceware 257 & Heterogeneous GPU PhysX: It's a Bug, Not a Feature	jmke	WebNews	0	28th May 2010 22:28
Nvidia Responds to AMD's Claim of PhysX CPU MultiCore Issues	jmke	WebNews	1	21st January 2010 09:24
AMD Accuses Nvidia of Disabling Multi-Core CPU Support in PhysX API	jmke	WebNews	0	20th January 2010 21:24
Driverheaven Nvidia Mobility Modder Released V1.1.0.0	jmke	WebNews	0	6th August 2009 13:12
Intel Responds to AMD, NVIDIA USB 3.0 Allegations	jmke	WebNews	5	14th August 2008 00:31
Futuremark: NVIDIA GPU PhysX not allowed	jmke	WebNews	0	21st July 2008 09:42
NVIDIA Launches PhysX!	jmke	WebNews	3	21st June 2008 10:47
NVIDIA Unleashes World's Fastest Mobile GPU	Sidney	WebNews	0	25th February 2005 00:45
NVIDIA Is the Only GPU Manufacturer	Sidney	WebNews	0	24th August 2004 05:44

26th June 2008, 01:46	#3
phlegm Posts: n/a	The constant optimization war is the main problem with 3DMark. It's gotten to the point that performance in 3DMark has almost nothing to do with actual game performance other than as a vague rule-of-thumb. This is just next step on the path of it becoming completely irrelevant to enthusiasts.