More 'Application-Specific' Optimizations in NVidia Drivers

EconolineCrush writes "Futuremark and NVIDIA have been embroiled in a spat over various cheat/optimizations in 3DMark03 for several weeks now. Last week, the soap opera appeared to be over; Futuremark and NVIDIA released a joint statement in which Futuremark clarified that NVIDIA was optimizing its drivers for 3DMark03 rather than cheating. This story, however, appears to be far from over. Tech Report has uncovered a new series of optimizations in NVIDIA's Detonator FX drivers that affect image quality in even Futuremark's latest 3DMark03 build. What's more, if you rename the 3DMark03 executable, the optimizations disappear."

Comments

Re:Make NVIDIA drivers Open Source!

[iainl]

Lovely idea in theory.

In practice, nVidia have made it painfully aware on numerous occasions that they CANNOT do this. Its not just them being nasty closed-source meanies. The driver binaries contain licensed tech from numerous third parties that their license doesn't let them reveal the source to.

Re:It's the reviewers' fault

[Anonymous Coward]

The problem is that there are no real world applications yet that use the DX9 features the latest batch of graphic cards are offering. On current "real world" applications, the speed gain by those cards over the generation before them is not really that big....

Re:High or low level strategy?

[somberlain]

Well... All I know is, that a lot of companies do this... This is also the case for CD-R/RW-writers. When using tools like Nero CD-Speed etc, this problem is always solved by 'cheating' on the program. Been there for a while in this sector of hardware testing...

Re:Who cares about benchmark software?

[Toraz Chryx]

bzzt, quake 3's physics are tied to the framerate, you ca jump slightly higher with a higher framerate. ergo, with 125fps+ you can get to locations via jumppads that are inaccessible otherwise.

and the fps thing.. well, you're just outright wrong on that, I can easily see a difference between 30fps and 120fps :)

Re:Who cares about benchmark software?

[Tipsy McStagger]

the q3 physics when jumping works best with fps's where the rounding errors in calculating the path through discrete points are maximised.

43 76 & 125 all produce similar results.

http://ucguides.savagehelp.com/Quake3/FAQFPSJump s. html

Comment removed

[account_deleted]

Comment removed based on user account deletion

For a good explanation...

[epicstruggle]

as to why synthetic benchmarks are useful please see the following:Beyond3D. This website is probably the best site for info on 3d hardware.

later,

Re:Renaming

[grub]

In your pants? Yahoooo! 6 days of getting karma to "Excellent" (nee: "50") just to FUCK IT ALL UP IN ONE DAY so For The Record:

Linux sucks
Windows sucks
RMS sucks
Linus sucks
Microsoft really sucks
Apple sucks
*BSD sucks (and is also dying)
slashdot sucks
everyone on this fetid, shithole of a planet sucks.

Thank you.

Re:Image Quality

[Anonymous Coward]

That's because you've got them confused. The "optimized" one is 3dmark, the renamed one is 3dmurk . 3dmurk has the smooth carpet and the hints of text on the book. 3dmark has the jaggies.

Re:Image Quality (MOD PARENT UP)

[TheMeuge]

Actually, Tech-Report did check what happens to the ATI drivers as well, just visit their website and see. And like they said in the article, the differences they found were negligible for ATI (1-2%) and I would not be surprised if image quality remained the same. After all, there ARE legitimate optimizations one can write for a graphics engine to run better on a given video card architecture, just look at Doom3. The optimisations for GFFX are endless when it comes to the new Doom engine, and I doubt that Id would want to get caught reducing game image quality to let a certain card have an fps advantage.

P.S.I've been visiting Tech-Report for a while now and I find no reason to doubt their honesty quite yet, but then again - i semi-trusted nVidia too.