• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

Nvidia cheating... again?

A

Actaeon

Diamond Member
Linky for you.

We ran the GeForceFX 5900 through 3DMark03, first with the current 44.03 driver, and then with this new 44.67 driver. The performance increases seen in 3DMark03 were substantial, almost double on one of the tests. We also tested three of the games we usually use for 3D testing. We found essentially no performance difference between the two builds there.
Click to expand...
 
Hmmmm, seems they were puzzled by the lack of game performance increase after seeing such large performance increases in 3Dmark tests, especially Game 4.
Seems pretty obvious to me.
 
Originally posted by: rbV5
Hmmmm, seems they were puzzled by the lack of game performance increase after seeing such large performance increases in 3Dmark tests, especially Game 4.
Seems pretty obvious to me.
Click to expand...

Yep, drivers written intentionally to make performance in 3DMark high. Sounds like someones scared over at Nvidia.
 
god, whats it gonna take for nvidia to learn its lesson?
Click to expand...

What it will take, for ALL companies, not just nVidia (and this assumes they did "cheat" which I am not saying, nor do I care) is this: EVERY (and I may be a cynic, but I tend to think they all do it) companpy will stop tweaking to perform better on the synthetic benchmarks when we, the consumers, stop caring about synthetic benchamrks. When their people visit forums and sites such as this and see threads/articles going on and on and on ad nauseum about how many such-and-such points or how many so-and-so marks said hardware scored or people basing the speed of their system on the scores they are getting (are the systems really faster than mine even if I were to get a lower score? No. Not IMO). My point is that synthetic tests don't mean much of anything. At least not to me. I don't base my purchases on how many "points" a product scores on some synthetic benchmark. I base it on features, price, extras, service etc. That and I use the product. If it is fast enough, it is fast enough. I don't need some software to tell me I am getting good framerates in games. I can see for myself if I am or not. My feeling is (to pick on video cards here) if the human eye can't pick up over a certain number of frames per second, any over that is a waste. Why should I care if your PC gets 190 fps when my "slow" PC gets "only" 70 at the same settings? I don't. When I read reviews, I skip right over the benchmarks to get the overall information about a product. That's just me, and I really think that once we stop putting so much faith in these tests, the comapnies will be forced to manufacture a product that performs kick-ass across the board, instead of one that is tweaked for a certain tool.</end rant>

\Dan

 
If they are cheating again, its kinda lame they'd do it twice in a row. ATI learned there lesson after the Quake incident.
 
Originally posted by: EeyoreX
god, whats it gonna take for nvidia to learn its lesson?
Click to expand...

What it will take, for ALL companies, not just nVidia (and this assumes they did "cheat" which I am not saying, nor do I care) is this: EVERY (and I may be a cynic, but I tend to think they all do it) companpy will stop tweaking to perform better on the synthetic benchmarks when we, the consumers, stop caring about synthetic benchamrks. When their people visit forums and sites such as this and see threads/articles going on and on and on ad nauseum about how many such-and-such points or how many so-and-so marks said hardware scored or people basing the speed of their system on the scores they are getting (are the systems really faster than mine even if I were to get a lower score? No. Not IMO). My point is that synthetic tests don't mean much of anything. At least not to me. I don't base my purchases on how many "points" a product scores on some synthetic benchmark. I base it on features, price, extras, service etc. That and I use the product. If it is fast enough, it is fast enough. I don't need some software to tell me I am getting good framerates in games. I can see for myself if I am or not. My feeling is (to pick on video cards here) if the human eye can't pick up over a certain number of frames per second, any over that is a waste. Why should I care if your PC gets 190 fps when my "slow" PC gets "only" 70 at the same settings? I don't. When I read reviews, I skip right over the benchmarks to get the overall information about a product. That's just me, and I really think that once we stop putting so much faith in these tests, the comapnies will be forced to manufacture a product that performs kick-ass across the board, instead of one that is tweaked for a certain tool.</end rant>

\Dan
Click to expand...

not to say that i put much stock in synthetic benchmarks myselfs, but that still doesnt justify that cheating or optimizing or whatever you want to call it. do synthetic benchmarks mean much? no, not in my opinion. should ANY company cheat on synthetic benchmarks? absolutely not. it is a matter of trying to decieve the consumer that i have a problem with. and a wide majority of computer buyers go on a website, see a 3dmark score, and make a decison on what card to get based on that score, not taking into consideration that it is a synthetic benchmark

 
honestly, nvidia does not deserve our hard earned cash. It is quite sad how they went from underdogs to this, a company that cares not about quality and integrity. I know ATI's not innocent either, but at least they're showing some improvement and making some effort to remove these optimizations.
 
Originally posted by: Sudheer Anne
honestly, nvidia does not deserve our hard earned cash. It is quite sad how they went from underdogs to this, a company that cares not about quality and integrity. I know ATI's not innocent either, but at least they're showing some improvement and making some effort to remove these optimizations.
Click to expand...

Agreed, ATi is the MUCH lesser of two evils
 
god, whats it gonna take for nvidia to learn its lesson?
Click to expand...

What it will take, for ALL companies, not just nVidia (and this assumes they did "cheat" which I am not saying, nor do I care) is this: EVERY (and I may be a cynic, but I tend to think they all do it) companpy will stop tweaking to perform better on the synthetic benchmarks when we, the consumers, stop caring about synthetic benchamrks. When their people visit forums and sites such as this and see threads/articles going on and on and on ad nauseum about how many such-and-such points or how many so-and-so marks said hardware scored or people basing the speed of their system on the scores they are getting (are the systems really faster than mine even if I were to get a lower score? No. Not IMO). My point is that synthetic tests don't mean much of anything. At least not to me. I don't base my purchases on how many "points" a product scores on some synthetic benchmark. I base it on features, price, extras, service etc. That and I use the product. If it is fast enough, it is fast enough. I don't need some software to tell me I am getting good framerates in games. I can see for myself if I am or not. My feeling is (to pick on video cards here) if the human eye can't pick up over a certain number of frames per second, any over that is a waste. Why should I care if your PC gets 190 fps when my "slow" PC gets "only" 70 at the same settings? I don't. When I read reviews, I skip right over the benchmarks to get the overall information about a product. That's just me, and I really think that once we stop putting so much faith in these tests, the comapnies will be forced to manufacture a product that performs kick-ass across the board, instead of one that is tweaked for a certain tool.</end rant>

\Dan
Click to expand...

not to say that i put much stock in synthetic benchmarks myselfs, but that still doesnt justify that cheating or optimizing or whatever you want to call it. do synthetic benchmarks mean much? no, not in my opinion. should ANY company cheat on synthetic benchmarks? absolutely not. it is a matter of trying to decieve the consumer that i have a problem with. and a wide majority of computer buyers go on a website, see a 3dmark score, and make a decison on what card to get based on that score, not taking into consideration that it is a synthetic benchmark
Click to expand...

That was my point, we need to educate the "majority of computer buyers" about this issue. As long as people put too much faith in synthetic benchmarks, the companies (ALL of them, I bet ATI, Intel, AMD ALL of them do it somehow) will continue to try and skew the tests in their favor. I don't agree that nVidia doesn't deserve our money. If I were to agree with that, I would not have a computer since I firmly believe every company does it to some degree or another. Just because they haven't gotten caught or are more subtle or aren't doing it at this moment, I think they all do it. That's why I ignore benchmarks and buy what is good (IMO) based on other information in reviews and comments from people at places such as AT. I was not trying to justify the "cheats" "optimizations" or whatever in any way. I meant just to explain why I think they are done and how we can try to stop it.

\Dan
 
anyone remember that nvidia said these were just enhancements? they NEVER said they were gunna stop!!!!!! lets quit our bitching, and get on with life, who here buys a video card based on 3dmark scores ALONE? if you do, im gunna build a cheap video card, that always displays 1000000 3dmark points, due to software optimizations, and sell it for 50$ when it costs me 10$ to make, and get around 1000000$ pure profit from it. JESUS ppl!!!!

NVIDIA NEVER THOUGHT THEY WERE CHEATING, AND NEVER SAID THEY WERE GUNNA STOP "OPTOMIZING THEIR DRIVERS!!

MIKE
 
Originally posted by: nourdmrolNMT1
anyone remember that nvidia said these were just enhancements? they NEVER said they were gunna stop!!!!!! lets quit our bitching, and get on with life, who here buys a video card based on 3dmark scores ALONE? if you do, im gunna build a cheap 1$ video card, that always displays 1000000 3dmark points, due to software optimizations, and sell it for 50$ when it costs me 10$ to make, and get around 1000000$ pure profit from it. JESUS ppl!!!!

NVIDIA NEVER THOUGHT THEY WERE CHEATING, AND NEVER SAID THEY WERE GUNNA STOP "OPTOMIZING THEIR DRIVERS!!

MIKE
Click to expand...

So you're going to build a $1 videocard that cost you $10 to make.

Ok.
 
Originally posted by: Actaeon
Originally posted by: nourdmrolNMT1
anyone remember that nvidia said these were just enhancements? they NEVER said they were gunna stop!!!!!! lets quit our bitching, and get on with life, who here buys a video card based on 3dmark scores ALONE? if you do, im gunna build a cheap 1$ video card, that always displays 1000000 3dmark points, due to software optimizations, and sell it for 50$ when it costs me 10$ to make, and get around 1000000$ pure profit from it. JESUS ppl!!!!

NVIDIA NEVER THOUGHT THEY WERE CHEATING, AND NEVER SAID THEY WERE GUNNA STOP "OPTOMIZING THEIR DRIVERS!!

MIKE
Click to expand...

So you're going to build a $1 videocard that cost you $10 to make.

Ok.
Click to expand...

ROFL
 
As far as your thread title of Nvidia cheating... again? The article clearly said NO.


"After we tested the two drivers, we fired up the developer version of 3DMark03, ran Game Test 4 ? Mother Nature, and enabled the free-camera mode. We wanted to see whether the same corner-cutting we observed with 44.03 was still present in this new driver. It wasn't. As far as we can tell, nVidia achieved these impressive performance gains without any "on-rail" optimizations."

 
If the drivers don't improve gaming performance or IQ, who cares how high a score is in 3DMark 2003. I don't understand their reasoning on this.
 
Originally posted by: jeffrey
As far as your thread title of Nvidia cheating... again? The article clearly said NO.


"After we tested the two drivers, we fired up the developer version of 3DMark03, ran Game Test 4 ? Mother Nature, and enabled the free-camera mode. We wanted to see whether the same corner-cutting we observed with 44.03 was still present in this new driver. It wasn't. As far as we can tell, nVidia achieved these impressive performance gains without any "on-rail" optimizations."
Click to expand...

What of image quality? Did they turn off anisotropic filtering or something like that?
 
i just dont understand why they cant include a true QUALITY setting that would disable all their little hacks and cheats, and give good image quality. Then they could have a performance mode and a high performance mode like they do that would include these hacks. It is the simple solution to the problem.
 
I also am really getting sick of these threads, but let me point your attention to something:

Why do you think optimizations in 3dmark03 would show up in any current game? 3dmark03 was designed to show how video cards will handle future games, so your 3dmark03 score is obviously not going to be indicative of how the card will perform in any other benchmark. For example, my 3dmark03 score for my Radeon 9700 is 3x what it was for my GF4 ti4400, but the ATI is not 3x as fast as the nVidia card in any games.

Also, as was mentioned before, these optimizations come from Game 4, which is purely dependent on the Advanced Pixel shaders. nVidia did the same thing with the Detonator 40 Drivers in 3dmark01, so I don't see why so many of you are surprised.

 
You must log in or register to reply here.

TRENDING THREADS

Back
Top