Futuremark Announces Patch for 3DMark03

[Nick[FM]]

FYI:

Press Release: http://www.futuremark.com/pressroom/pressreleases/?111103
Verified & Approved 3DMark03 Drivers: http://www.futuremark.com/community/drivers/?approved
Download: http://www.futuremark.com/download/
News: http://www.futuremark.com/news/?newsarticle=200311/2003111108#200311/2003111108

 

[Dave Baumann]

As a member of the Futuremark Beta program, we've had acess to the release candidate of this patch for a little while. here's a peak:

                      Patch Driver      3DMarks GT1    GT2   GT3   GT4 
GeForce FX 5950 Ultra 330   52.16       6412    205.7  46.6  37    37.3 
GeForce FX 5950 Ultra 333   52.16       5538    205    39.6  33.1  26.3 
Radeon 9800 XT        330   3.8         6435    209.7  45.4  38.6  36.3 
Radeon 9800 XT        333   3.8         6436    210.5  45.4  38.6  36.3

http://www.beyond3d.com/articles/3dmark03/340/

 

[volt]

Oh my, GT4

I can see some people raise hell over that one

 

[tEd]

As a member of the Futuremark Beta program, we've had acess to the release candidate of this patch for a little while. here's a peak:

                      Patch Driver      3DMarks GT1    GT2   GT3   GT4 
GeForce FX 5950 Ultra 330   52.16       6412    205.7  46.6  37    37.3 
GeForce FX 5950 Ultra 333   52.16       5538    205    39.6  33.1  26.3 
Radeon 9800 XT        330   3.8         6435    209.7  45.4  38.6  36.3 
Radeon 9800 XT        333   3.8         6436    210.5  45.4  38.6  36.3

More a later...

52.16= trilinear in 3dmark03?

 

[Dave Baumann]

No, that doesn't change since that DX specific rather than app specific.

 

[Entropy]

Good to see concrete results.
What is odd is that 52.12 is an approved driver, even though it (according to Daves data) obviously attempts to cheat. Thus drivers whose cheats are blocked by the new patch are apparently regarded as OK.

 

[tEd]

No, that doesn't change since that DX specific rather than app specific.

i just thought maybe nvidia would make a special case for 3dmark03 to get trilinear working to get approved drivers. But then that would be a app. specific (optimisation lol) which would be against futuremark rules

..just kidding

 

[vb]

As a member of the Futuremark Beta program, we've had acess to the release candidate of this patch for a little while. here's a peak:

                      Patch Driver      3DMarks GT1    GT2   GT3   GT4 
GeForce FX 5950 Ultra 330   52.16       6412    205.7  46.6  37    37.3 
GeForce FX 5950 Ultra 333   52.16       5538    205    39.6  33.1  26.3 
Radeon 9800 XT        330   3.8         6435    209.7  45.4  38.6  36.3 
Radeon 9800 XT        333   3.8         6436    210.5  45.4  38.6  36.3

More a later...

52.16= trilinear in 3dmark03?

If the answer is yes how it is done?

3DMark detection?

And what about aniso on all stages? (when requested)

Ps 340 BTW

 

[Dave Baumann]

333 is the release candidate of 340

 

[AzBat]

Worm,

Thanks for the heads-up on the new patch and certified drivers. I'll have more comments on my thoughts and opinions on Futuremark and 3DMark later.

Dave,

Thanks for the peak. It sure is interesting. Something tells me we will see new drivers from both IHVs that increases their score. It's a good thing we now have this guideline/certification process in place to keep it from getting out of hand.


What else is there to talk about? I think the press release had some interesting info...

Technical Details Associated with the New Patch:

* Patch changes the program's build number to 340;

* Maximum point sprite size is now limited to 256x256 instead of the maximum supported by the hardware;

* Incorporates new version of the Entech library (used to detect the clock frequency of the graphics chip core and memory);

* Provides more detailed and accurate system information and improved detection of hyperthreaded CPUs

Anybody care to explain or comment any of these changes? I understand the first and last one, but not so much the other two.

FYI, there was one more change that was only listed on their website...

http://www.futuremark.com/products/3dmark03/?03patch340

* Fixed all reported and reproduced minor issues

Looking great so far. Anybody see reports from other web sites about this?

Tommy McClain

 

[vb]

333 is the release candidate of 340

seen that, but I don't have 333 to test for myself, do I?

How were 331 and 332 scores?

 

[Bouncing Zabaglione Bros.]

As a member of the Futuremark Beta program, we've had acess to the release candidate of this patch for a little while. here's a peak:

                      Patch Driver      3DMarks GT1    GT2   GT3   GT4 
GeForce FX 5950 Ultra 330   52.16       6412    205.7  46.6  37    37.3 
GeForce FX 5950 Ultra 333   52.16       5538    205    39.6  33.1  26.3 
Radeon 9800 XT        330   3.8         6435    209.7  45.4  38.6  36.3 
Radeon 9800 XT        333   3.8         6436    210.5  45.4  38.6  36.3

More a later...

Hang on, I'm a little confused here. It seem that the Nvdia 52.16 drivers cheat in 3DMark v330, but because 3DMark v333 disables those cheats, these same drivers now become "approved"?

Yes, the score is now lower for the 52.16 drivers, but surely Nvidia should be taken to task for *attempting* to cheat and getting caught? What if not all the cheats have been detected by Futuremark and the Nvidia scores are still being artificially inflated?

Is this what it means to be an approved driver - Nvidia impelment more cheats in every release, and Futuremark "approve" the drivers by attempting to block those cheats?

 

[Joe DeFuria]

As a member of the Futuremark Beta program, we've had acess to the release candidate of this patch for a little while. here's a peak:

                      Patch Driver      3DMarks GT1    GT2   GT3   GT4 
GeForce FX 5950 Ultra 330   52.16       6412    205.7  46.6  37    37.3 
GeForce FX 5950 Ultra 333   52.16       5538    205    39.6  33.1  26.3 
Radeon 9800 XT        330   3.8         6435    209.7  45.4  38.6  36.3 
Radeon 9800 XT        333   3.8         6436    210.5  45.4  38.6  36.3

More a later...

Wow...I'm surprised!

.....Not.

What is odd is that 52.12 is an approved driver, even though it (according to Daves data) obviously attempts to cheat. Thus drivers whose cheats are blocked by the new patch are apparently regarded as OK.

Yes, that seems to be the case. FM highlights that these drivers are aproved only with the 440 patch. Unfortunately, I don't like this approach.

I would much rather that FM simply not approve any drivers when a "patch" is required to force the driver to follow the guidelines. They should still release the patch though...so that the drivers give the "correct score" to the best of FM's knowledge.

The way it seems to be now, there is little incentive for the vendor not to cheat! Now, all the burden seems to be on FM to "circumvent" the cheats in order to have a set of "approved drivers" to test. Significant burden should be on the IHV.

Seems very much backwards to me.

 

[Deathlike2]

A couple of thoughts...

Although internal builds of 3DMark are not accessible to anyone... how do these drivers fare with the Catalysts 3.8 and 3.9... and the Det 52.16s with builds 330 and 340...

If preliminary results hold true with the new Dets (though it should be tested with older Dets including Det 45.23 as Det 44.03 was the first set of drivers nullified by build 330), then it would be safe to say that NVidia has not stopped its stance with "aggressive optimizations"...

If the above is true, what now?

 

[Deathlike2]

A more important question to ask yourselves is this:

How did these drivers able to get "Futuremark Approved" status if they ARE cheating?

 

[Joe DeFuria]

Don't want to sound too harsh on FM.

The fact that they are releasing patches to circumvent cheats is great.

I can now very much recommend 3DMark03 to be used in benchmark tests...as long as "approved" drivers are used with the patch. FM is to be commended STRONGLY for taking an active stand against cheating in benchmarks.



It's just that to satisfy me almost 100%, any driver that is "caught" cheating, should not be labeled as approved at all IMO.

 

[Joe DeFuria]

A more important question to ask yourselves is this:

How did these drivers able to get "Futuremark Approved" status if they ARE cheating?

Presumably, because when the 340 patch is applied, it defeats all the "known" cheating attempts that the Dets are trying.

(edit...patch 340, not 440. )

 

[AAlcHemY]

Is 52.16 the only affected driver? ( Not 52.14, 52.13? )

 

[Exxtreme]

As a member of the Futuremark Beta program, we've had acess to the release candidate of this patch for a little while. here's a peak:

                      Patch Driver      3DMarks GT1    GT2   GT3   GT4 
GeForce FX 5950 Ultra 330   52.16       6412    205.7  46.6  37    37.3 
GeForce FX 5950 Ultra 333   52.16       5538    205    39.6  33.1  26.3 
Radeon 9800 XT        330   3.8         6435    209.7  45.4  38.6  36.3 
Radeon 9800 XT        333   3.8         6436    210.5  45.4  38.6  36.3

More a later...

52.16= trilinear in 3dmark03?

If the answer is yes how it is done?

3DMark detection?

And what about aniso on all stages? (when requested)

Ps 340 BTW

I looks like an typical application detection for me. I don't know which part of 3dmark the drivers are using to detect them but it seems that FM has removed/changed this part. The drivers are not able to detect the benchmark anymore.

 

[Deathlike2]

Joe... it's 340

I guess that can be the case, but one of two things should happen then.

1) Futuremark should let everyone be aware of this fact. If these drivers are approved because they stop "unauthorized optimizations" in this set of drivers, then that's fine.

2) Futuremark should remove/revoke their approval to this set of drivers.

I'd suspect NVidia has already been informed of this (hopefully).