Futuremark releases 3DMark03 patch - alleges NVIDIA cheats - Page 3

Moose · 05-23-03, 10:10 AM

I think this is a GREAT move by Futuremark.

They FOR SURE have caught Nvidia cheating and they may have caught ATI as well and will be investigating further.

This is good for the entire computer industry as well as all computer enthusiasts.

We need to have valid benchmarks so we know how good the products we are going to buy, REALLY are.

I think Futuremark has done a good service to help legitimize the benchmarking process.

Kudos.

Although I think their work in this area will need to be ongoing to prevent future cheating.

Typedef Enum · 05-23-03, 10:13 AM

I don't have the time to look this up, but it just goes to show what a bunch of freakin' liars nVidia have turned into...

The aforementioned quote went something like this...

"We don't think that 3DMark2003 is a good benchmark. Do you really want our engineers to spend X amount of time optimizing for a synthetic benchmark that we don't feel is representative of real games, or spend it optimizing our drivers for real/actual/shipping games?"

And they went on to say something like...

"nVidia doesn't optimize for specific titles, rather, general pipeline optimizations..."

I know these quotes aren't word-for-word, but off the top of my head, they're very close...

Once you think about the total BS that has been coming from nVidia, think about this...Because of all this crap, guess who has to spend the time to correct the code to account for such practices? Yep, not nVidia...FutureMark has to then dedicate some of their resources to addressing this crap that probably should be doing other more important things.

Skinner · 05-23-03, 10:18 AM

Get ready:

almost 1000 points less !!

5600 with 43.51 and

4742 with 44.03 and patch 330,

well,.. this will shake the world

I cann't publish it.

Sazar · 05-23-03, 10:19 AM

Quote:

Originally posted by Skinner
Get ready:

almost 1000 points less !!

5600 with 43.51 and

4742 with 44.03 and patch 330,

well,.. this will shake the world

I cann't publish it.

wow... thats my score... sort of

Moose · 05-23-03, 10:23 AM

Quote:

Originally posted by Skinner
Get ready:

almost 1000 points less !!

5600 with 43.51 and

4742 with 44.03 and patch 330,

well,.. this will shake the world

I cann't publish it.

Ouch!!!!!!!!!!!!!

Skinner · 05-23-03, 10:25 AM

This didn't realy come as a suprice for me, ATI's R300 core is superior towards dx9 in mosts tests.

rokzy · 05-23-03, 10:27 AM

has anybody decided to not buy a FX card after seeing these results? or decided to get one anyway?

I personally wouldn't feel comfortable buying from a company who would try to cheat me out of 24.1% of it's performance, especially with something as expensive as the FX's...

schuey74 · 05-23-03, 10:29 AM

Compare the 44.03s with and without the patch. This will tell us exactly what performance gains they got from whatever cheats' were in the 44.03s.

Skinner · 05-23-03, 10:31 AM

Quote:

Originally posted by rokzy
has anybody decided to not buy a FX card after seeing these results? or decided to get one anyway?

I personally wouldn't feel comfortable buying from a company who would try to cheat me out of 24.1% of it's performance, especially with something as expensive as the FX's...

Well, I like to adjust the lod and like the AF of nVidia, I don't use AA, I prefere to play at high res and there are no dx9 games yet, so that brings me by the GFFX, but I'm considering a R9800, because the fan is just to loud, I thought that I get used to it, but that didn't happen

digitalwanderer · 05-23-03, 10:31 AM

...has nVidia responded at all yet?

Morrow · 05-23-03, 10:32 AM

After officially knowing now that both nvidia and ATI cheat in 3dmark03 (anyone surprised?), what do we learn from this incident?

We learn of course that we can no longer trust synthetic benchmarks, isn't it obvious?

Cheating in games is certainly easier to hide but not as easy to implement because the cheats found in 3dmark03 do not work in games where the camerapath is random.

Another thing we learn from this is that FutureMark now has also officially stated that their shader routines are inefficient! They say that nvidia managed to implement nvidia hw optimized shaders which are sometimes more than twice as fast as the shaders used by FutureMark. What does this say about FutureMark's credibility having released a benchmark optimized for future hardware! Well, nothing positive in any case...

SurfMonkey · 05-23-03, 10:33 AM

Quote:

In game test 4, the water pixel shader (M_Water.psh) is detected. The driver uses this
detection to artificially achieve a large performance boost - more than doubling the early
frame rate on some systems. In our inspection we noticed a difference in the rendering when
compared either to the DirectX reference rasterizer or to those of other hardware. It appears
the water shader is being totally discarded and replaced with an alternative more efficient
shader implemented in the drivers themselves. The drivers produce a similar looking
rendering, but not an identical one.

Maybe Futuremark should use the more efficient version

I wonder if that explains all the shader code you can find in the drivers if you look with a text tool (there's a good one at AnalogX)

I would print it all here but I's probably get my ass sued.

05-23-03, 10:10 AM	#25
Moose Cheating is for losers!!! Join Date: Jul 2002 Posts: 241	I think this is a GREAT move by Futuremark. They FOR SURE have caught Nvidia cheating and they may have caught ATI as well and will be investigating further. This is good for the entire computer industry as well as all computer enthusiasts. We need to have valid benchmarks so we know how good the products we are going to buy, REALLY are. I think Futuremark has done a good service to help legitimize the benchmarking process. Kudos. Although I think their work in this area will need to be ongoing to prevent future cheating. Last edited by Moose; 05-23-03 at 10:18 AM.

05-23-03, 10:18 AM	#27
Skinner Registered User Join Date: Aug 2002 Location: Rotterdam, Netherlands Posts: 1,128	Get ready: almost 1000 points less !! 5600 with 43.51 and 4742 with 44.03 and patch 330, well,.. this will shake the world I cann't publish it. __________________ Intel i7@3,8gHzAsus P6T Deluxe* 2x Sapphire HD7970 3GB Crossfire 6 GB Corsair Dominator 1600C8OCZ Vertex 120 GB SSDadaptec 19160U160Intel X25 80 GB Quantum Atlas 15KII SCSI U160 147Gig* WD Raptor 300 GBApple 24" LED Cinema X-Fi TitaniumLogitech Z5500Coolmaster RP 1000WW7 64 Home

05-23-03, 10:25 AM	#30
Skinner Registered User Join Date: Aug 2002 Location: Rotterdam, Netherlands Posts: 1,128	This didn't realy come as a suprice for me, ATI's R300 core is superior towards dx9 in mosts tests. __________________ Intel i7@3,8gHzAsus P6T Deluxe* 2x Sapphire HD7970 3GB Crossfire 6 GB Corsair Dominator 1600C8OCZ Vertex 120 GB SSDadaptec 19160U160Intel X25 80 GB Quantum Atlas 15KII SCSI U160 147Gig* WD Raptor 300 GBApple 24" LED Cinema X-Fi TitaniumLogitech Z5500Coolmaster RP 1000WW7 64 Home

05-23-03, 10:31 AM	#34
digitalwanderer Join Date: Jul 2002 Location: Highland, IN USA Posts: 4,944	And the $64,000 question-o-the-moment.... ...has nVidia responded at all yet? __________________ [SIZE=1][I]"It was very important to us that NVIDIA did not know exactly where to aim. As a result they seem to have over-engineered in some aspects creating a power-hungry monster which is going to be very expensive for them to manufacture. We have a beautifully balanced piece of hardware that beats them on pure performance, cost, scalability, future mobile relevance, etc. That's all because they didn't know what to aim at." -R.Huddy[/I] [/SIZE]

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Need Help Installing NVIDIA Tesla M2070Q in Linux RHEL5	Ferianto85	NVIDIA Linux	0	05-18-12 08:35 PM
Current NVIDIA FreeBSD graphics driver releases	zander	NVIDIA FreeBSD	0	01-27-09 05:22 PM
Current NVIDIA Linux graphics driver releases	AaronP	NVIDIA Linux	0	11-06-08 04:39 PM

05-23-03, 10:13 AM	#26
Typedef Enum Join Date: Aug 2002 Posts: 191	I don't have the time to look this up, but it just goes to show what a bunch of freakin' liars nVidia have turned into... The aforementioned quote went something like this... "We don't think that 3DMark2003 is a good benchmark. Do you really want our engineers to spend X amount of time optimizing for a synthetic benchmark that we don't feel is representative of real games, or spend it optimizing our drivers for real/actual/shipping games?" And they went on to say something like... "nVidia doesn't optimize for specific titles, rather, general pipeline optimizations..." I know these quotes aren't word-for-word, but off the top of my head, they're very close... Once you think about the total BS that has been coming from nVidia, think about this...Because of all this crap, guess who has to spend the time to correct the code to account for such practices? Yep, not nVidia...FutureMark has to then dedicate some of their resources to addressing this crap that probably should be doing other more important things.

05-23-03, 10:27 AM	#31
rokzy Join Date: May 2003 Posts: 158	has anybody decided to not buy a FX card after seeing these results? or decided to get one anyway? I personally wouldn't feel comfortable buying from a company who would try to cheat me out of 24.1% of it's performance, especially with something as expensive as the FX's...

05-23-03, 10:29 AM	#32
schuey74 GTX 280 (675/1458/2500) Join Date: Feb 2003 Location: South Florida Posts: 819	Compare the 44.03s with and without the patch. This will tell us exactly what performance gains they got from whatever cheats' were in the 44.03s.

05-23-03, 10:32 AM	#35
Morrow Atari STE 4-bit color Join Date: May 2003 Posts: 798	After officially knowing now that both nvidia and ATI cheat in 3dmark03 (anyone surprised?), what do we learn from this incident? We learn of course that we can no longer trust synthetic benchmarks, isn't it obvious? Cheating in games is certainly easier to hide but not as easy to implement because the cheats found in 3dmark03 do not work in games where the camerapath is random. Another thing we learn from this is that FutureMark now has also officially stated that their shader routines are inefficient! They say that nvidia managed to implement nvidia hw optimized shaders which are sometimes more than twice as fast as the shaders used by FutureMark. What does this say about FutureMark's credibility having released a benchmark optimized for future hardware! Well, nothing positive in any case...