Looking Back: ATI's Catalyst Drivers Exposed
by Ryan Smith on December 11, 2005 3:22 PM EST- Posted in
- GPUs
3dMark 2003
3dMark is not a benchmark that we routinely bring you here at AnandTech, as our editorial policy is to bring you benchmarks from real-world games and engines, and not synthetic metrics. That said, it would be inappropriate to leave out 3dMark in this case due to the significant cheating incidents with it. And, as a flashy, system draining benchmark backed by a unique database for comparisons, it's still an important title in the eyes of many consumers, OEMs, and the GPU makers looking for bragging rights.
With 3dMark, its importance in this regression is not so much the performance improvements as a sign of what happened with the card - the improvements were most certainly exaggerated due to in part by the synthetic nature of the benchmark - but rather a possibility of what can happen when ATI dedicates its resources to a game/benchmark that it considers most important. We should note that ATI has admitted to "cheating" on 3dMark 2003; however, these were what we consider honest shader-replacement optimizations (same mathematical output) that ATI voluntarily removed, though they were apparently re-introduced at some point. We used the latest version of 3dMark 2003, so this "cheat" was not activated in the older drivers.
For these benchmarks, 3dMark was run at its default resolution of 1024x768.
Overall then, 3dMark is much like Halo, a benchmark that received a slow, but steady improvement, without any fixes.
3dMark is not a benchmark that we routinely bring you here at AnandTech, as our editorial policy is to bring you benchmarks from real-world games and engines, and not synthetic metrics. That said, it would be inappropriate to leave out 3dMark in this case due to the significant cheating incidents with it. And, as a flashy, system draining benchmark backed by a unique database for comparisons, it's still an important title in the eyes of many consumers, OEMs, and the GPU makers looking for bragging rights.
With 3dMark, its importance in this regression is not so much the performance improvements as a sign of what happened with the card - the improvements were most certainly exaggerated due to in part by the synthetic nature of the benchmark - but rather a possibility of what can happen when ATI dedicates its resources to a game/benchmark that it considers most important. We should note that ATI has admitted to "cheating" on 3dMark 2003; however, these were what we consider honest shader-replacement optimizations (same mathematical output) that ATI voluntarily removed, though they were apparently re-introduced at some point. We used the latest version of 3dMark 2003, so this "cheat" was not activated in the older drivers.
For these benchmarks, 3dMark was run at its default resolution of 1024x768.
Catalyst 5.11 versus 3.00 (mouse over to see 3.00)
Overall then, 3dMark is much like Halo, a benchmark that received a slow, but steady improvement, without any fixes.
Facebook
Twitter
RSS
58 Comments
View All Comments
WileCoyote - Monday, December 12, 2005 - link
Easy, ATI was a Halo for PC launch partner. This was before the "Best Played with ATI" or "Insist on NVidia" days but ATI was the graphics card sponsor for the game. So they had a committment to Bungie/Microsoft.... not really to the customer. I'm not complaining because they're businesses and they want to make money. I just consider it cheating. Halo benchmark explained. Next?GameManK - Monday, December 12, 2005 - link
well done, but it did feel like a bit of a waste of time readingsomething like farcry or half life 2 i think would be a more useful test
Googer - Monday, December 12, 2005 - link
Nice article, it must have taken a lot of time and effort to do this. Ryan how long did it take you to do all of these driver installs (then reboot) and benchmark them 72 Time?Thanks for the effort!
Googer - Monday, December 12, 2005 - link
Could you also test 3rd party drivers like Omega and others I have forgotten about? Then could you compaire them to Stock ATI drivers?nullpointerus - Monday, December 12, 2005 - link
Aren't the Omega drivers just a mix of different official ATI driver files?Humble Magii - Monday, December 12, 2005 - link
Seriously another craptastic article on drivers? Guys please sit down and think before you post an article and give it actual thought maybe ask some people around you god forbid.This site is sucking huge.
If you are going to do an article such as this use both competitors and go through each revision or at least a major revision to the drivers on each core and card.
Again stop posting worthless articles someone at Anandtech please take control and scrutinize what your people write and do before posting. Don't they have a set process there?
Cygni - Tuesday, December 13, 2005 - link
"If you're disappointed with the free entertainment on this site, fine, write about it on your shitty Angelfire Dragonball Z site or send AIM messages to the other Korn fanclub members."Cygni - Tuesday, December 13, 2005 - link
Ok, so what ive learned is you are reading a site that you think is "sucking huge", pretty making you a retard.Please sit down and think before you write such a fucking pointless post. God forbid there are people out there who are actually interested in video card driver performance.
Do you have any idea how long it takes to run through these tests? "Oh, just do both companys! And do all their cards! And do every CPU/motherboard/memory/timings setting too!... oh, and give it to me for free!"
What a joke dude. Go find your cave asshole, or go to some other hardware website.
VIAN - Sunday, December 11, 2005 - link
Is this article that important? I didn't think there was enough content in the article to make it worth reading. Plus, the way you built it up in the introduction seemed to give less meaning to the article when we found that there wasn't that much of an increase in half the games you tested. It also seemed like most of the big performance boosting optimizations took place within the first few drivers for the R300. To prove your point, it might have been better to make a shorter article covering various games, but use only 2 drivers, the current and the earliest.And where's that long lost image quality article we were promised about a year ago?
Jedi2155 - Monday, December 12, 2005 - link
But then, it wouldn't show the slight improvements of the driversets like the 3.00 tothe 3.04.