“Trust, but verify.”
- Damon Runyon
More pages: 1 2
The state of ethics in 3D graphics
Friday, May 23, 2003 | Permalink

We live in sad times. Around the time of the release of Radeon 8500 news spread on the web the ATi applied cheats to gain performance in Quake3. SiS has been busted occasionally for lowering IQ for performance in benchmarks. Not too long ago Trident was accused of cheating too. Recently nVidia officially joined the crew when their cheats was exposed. This time it went as far as to Futuremark having to issue a statement and provide a patch to disable the cheats and a detailed report on the cheats applied. No less than 8 different cheats has been applied by the driver which added no less than 24% more performance. This puts the NV35 into a less pleasing light. Instead of being the chip that brings nVidia's floating point performance up on par with ATi the painful truth again after disabling the cheats looks to be that the floating point performance still is crap. At this time I also have no hope that any NV3x well ever perform well in that area, the hardware is just that slow, as patched 3dmarks proves and other applications has proven for some time. It should also be said that Futuremark found oddities in one test where ATi seems to get a small but suspicious performance drop with the patch applied. The jury is still out on that one though, but in all likelyhood the truth may be less beautiful than one might hope.

It's sad. It's really sad. But as usual, there's something to learn. The lesson is NOT that 3dmarks is useless as a benchmark that some sites, which I refuse to name and link to, has implied. And it's absolutely not that one's cheating justifies another one's as endless fanboys are ranting. One may have have objective opinions on 3dmarks, but the real lesson is that reviewers need to spend more time looking at Image Quality and look for rendering anomalities. Reviewers need to break away from traditional benchmarks and look for alternatives. Benchmark games that are less known. Write your own benchmarking tools. I have little hope on the last one though given the enourmous amount of ignorance out there. Most review sites can hardly discern a pipeline from a pixel. The last, and maybe the most important point though is that we as consumers need to be aware of this. We need to demand that this does not happend. We all vote with our wallets. We much make it so that it's not worth it to cheat.

Name

Comment

Enter the code below



NeoKenobi
Saturday, May 24, 2003

Well said Humus, I totally agree. I think there should be an independent benchmarking suite that measures performance. Speaking in terms of OpenGL, it should use the standard ARB extensions instead of the vendor specific ones. This way developers will learn (mostly nvidia) to focus on the ARB standard rather than developing their own extensions.

davepermen
Saturday, May 24, 2003

i totally agree. why isn't cheating actually illegal? nvidia should get sued for not providing actually what they tell their customers they do. since the radeon9700pro, nvidia has NOT provided any high quality drivers, or hardware anymore. and they cheated, used tons of marketings (nude dawn's etc) to regain the people, spreading out false-statements.

i'm very dissapointed myself, too. currently, i'm happy running realtime the dawn demo on my radeon with no problem faster than any nv30 chip as well, and more beautiful

FXO
Saturday, May 24, 2003

I thought that cheating was illegal due to some licenseagreement in 3Dmark2003.
If this is true, I hope Futuremark sues the cheaters.

btw. good article, I totally agree about the IQ-issue.

NeoKenobi
Saturday, May 24, 2003

Isn't it true that ATI cheated too? In the new build of 3dmark03 the performance of the Radeon 9800 Pro dropped with approximately 2%.

Edit by Humus: Fixed the comment, will need to fix my php-script later to deal with this

NeoKenobi
Saturday, May 24, 2003

Whoa! My last post seems to have been corrupted by a bug of some sort... What I said was this: Isn't it true that ATI cheated too? The Radeon 9800 Pro dropped about 2 percent in the new build of 3dmark03.

gibber
Saturday, May 24, 2003

This is very sad.

I am very disappointed in nVidia, they seem to have gone much further with their cheating than ATI.

Until now, nVidia's real weapon has been the strength and reliability of their drivers.

I will be voting with my wallet on this one.

Massimo
Saturday, May 24, 2003

I waited nv35 hoping it would be fast. Looks like it's not. No matter, I was still holding for them.
Now, I know they joined the cheating bandwagon. I no more feel them right. I feel they made a very bad marketing move by doing this, their reputation will probably be badly hurt.
BTW, I begin to think that GeForces are getting plagued by their own success. There's simply too much things to support... I guess this is a reason because companies rise and fall.
It would be really good if one can get an ARB extensions from the beginning. Too bad it's not the case. Look at EXT_frag_shad. It is equivalent to ARB_v_p but the interface is bad. Sometimes nvidia had to go for NV extensions, it is the case of NV_tex_shad and NV_reg_comb. If only they allowed other vendors to use them...

BlackSilver
Sunday, May 25, 2003

I personally don't agree with Nvidia decision, but I don't think that is so bad. The clipping planes question is very bad, but there isn't so sad since there isn't really gameplay on 3DMark, so there isn't anything that will be render slower than the benchmark, because don't exist a really game

Anyway, is unfair because ATI doesn't do the same thing

But in relation to the reorganization of the code I don't see anything wrong. This is the compiler driver job.
The remake of the code I think that should be unfair, but isn't really wrong since the 3DMark code never was fair with any Nvidia card.

More pages: 1 2