Nvidia and Futurmark are now blood brothers.
- Thread starter Rogozhin
- Start date
"Both NVIDIA and Futuremark want to define clear rules with the industry about how benchmarks should be developed and how they should be used. We believe that common rules will prevent these types of unfortunate situations moving forward."
This one really got me laughing!
Just who I want to decide how benchmarks should be developed and used!
Christ!!!
Just like I want the Oil companies to decide how hybrid cars should be designed!
Rogo
This one really got me laughing!
Just who I want to decide how benchmarks should be developed and used!
Christ!!!
Just like I want the Oil companies to decide how hybrid cars should be designed!
Rogo
Schadenfroh
Elite Member
- Mar 8, 2003
- 38,416
- 4
- 0
c/p for the lazy
"Futuremark Statement
For the first time in 6 months, as a result of Futuremark's White Paper on May 23rd, 2003, Futuremark and NVIDIA have had detailed discussions regarding NVIDIA GPUs and Futuremark's 3DMark03 benchmark.
Futuremark now has a deeper understanding of the situation and NVIDIA's optimization strategy. In the light of this, Futuremark now states that NVIDIA's driver design is an application specific optimization and not a cheat.
The world of 3D Graphics has changed dramatically with the latest generation of highly programmable GPUs. Much like the world of CPUs, each GPU has a different architecture and a unique optimal code path. For example, Futuremark's PCMark2002 has different CPU test compilations for AMD's AthlonXP and Intel's Pentium4 CPUs.
3DMark03 is designed as an un-optimized DirectX test and it provides performance comparisons accordingly. It does not contain manufacturer specific optimized code paths. Because all modifications that change the workload in 3DMark03 are forbidden, we were obliged to update the product to eliminate the effect of optimizations identified in different drivers so that 3DMark03 continued to produce comparable results.
However, recent developments in the graphics industry and game development suggest that a different approach for game performance benchmarking might be needed, where manufacturer-specific code path optimization is directly in the code source. Futuremark will consider whether this approach is needed in its future benchmarks.
NVIDIA Statement
NVIDIA works closely with developers to optimize games for GeForceFX. These optimizations (including shader optimizations) are the result of the co-development process. This is the approach NVIDIA would have preferred also for 3DMark03.
Joint NVIDIA-Futuremark Statement
Both NVIDIA and Futuremark want to define clear rules with the industry about how benchmarks should be developed and how they should be used. We believe that common rules will prevent these types of unfortunate situations moving forward."
i for one am GLAD THIS IS OVER
, i was getting sick of the repost blitz
"Futuremark Statement
For the first time in 6 months, as a result of Futuremark's White Paper on May 23rd, 2003, Futuremark and NVIDIA have had detailed discussions regarding NVIDIA GPUs and Futuremark's 3DMark03 benchmark.
Futuremark now has a deeper understanding of the situation and NVIDIA's optimization strategy. In the light of this, Futuremark now states that NVIDIA's driver design is an application specific optimization and not a cheat.
The world of 3D Graphics has changed dramatically with the latest generation of highly programmable GPUs. Much like the world of CPUs, each GPU has a different architecture and a unique optimal code path. For example, Futuremark's PCMark2002 has different CPU test compilations for AMD's AthlonXP and Intel's Pentium4 CPUs.
3DMark03 is designed as an un-optimized DirectX test and it provides performance comparisons accordingly. It does not contain manufacturer specific optimized code paths. Because all modifications that change the workload in 3DMark03 are forbidden, we were obliged to update the product to eliminate the effect of optimizations identified in different drivers so that 3DMark03 continued to produce comparable results.
However, recent developments in the graphics industry and game development suggest that a different approach for game performance benchmarking might be needed, where manufacturer-specific code path optimization is directly in the code source. Futuremark will consider whether this approach is needed in its future benchmarks.
NVIDIA Statement
NVIDIA works closely with developers to optimize games for GeForceFX. These optimizations (including shader optimizations) are the result of the co-development process. This is the approach NVIDIA would have preferred also for 3DMark03.
Joint NVIDIA-Futuremark Statement
Both NVIDIA and Futuremark want to define clear rules with the industry about how benchmarks should be developed and how they should be used. We believe that common rules will prevent these types of unfortunate situations moving forward."
i for one am GLAD THIS IS OVER
, i was getting sick of the repost blitzI think a lot of us argued way too much about this issue, and that 3dmark is still a peripheral issue for me.
My guess is FutureMark was more or less legally forced to make some type of reparative statement.
Doubtlessly nVidia would have been quick to involve their lawyers in the dispute.
At the very least inserting static clipping planes between frames has been pretty much clearly proven at that definitely isnt applicable to a typical gaming environment.
In any case I'm not sure it matters whether it was a cheat or an optimization.... FutureMark prohibits application specific optimizations, so even doing such would be cheating.
Anyone want to bet that despite this press statement FutureMark will NOT call back the 3.30 patch, and nVidia will NOT input the same cheats/optimizations again.
I'd be willing to bet such was part of whatever agreement nVidia/FutureMark came to.
Doubtlessly nVidia would have been quick to involve their lawyers in the dispute.
At the very least inserting static clipping planes between frames has been pretty much clearly proven at that definitely isnt applicable to a typical gaming environment.
In any case I'm not sure it matters whether it was a cheat or an optimization.... FutureMark prohibits application specific optimizations, so even doing such would be cheating.
Anyone want to bet that despite this press statement FutureMark will NOT call back the 3.30 patch, and nVidia will NOT input the same cheats/optimizations again.
I'd be willing to bet such was part of whatever agreement nVidia/FutureMark came to.
It's pretty clear that Nvidia told FutureMark "Our lawyers can beat up your lawyers." It was obvious to nearly everyone that it was a cheat and Nvidia still has a black eye. This is just damage control on Nvidia's part.
I got couple o' words of wisdom:
"Aaaaah, fuk this!" Who cares about what these fools got goin on there? I didnt pay for their useless tests. I just ran the useless test to see if the performance (that satidfies me) is comparable to other users. I dont make a cult out of it. Thats friggin nerd stuff.
"Aaaaah, fuk this!" Who cares about what these fools got goin on there? I didnt pay for their useless tests. I just ran the useless test to see if the performance (that satidfies me) is comparable to other users. I dont make a cult out of it. Thats friggin nerd stuff.
nvidia doesn't MAKE people run 3DMARK on their products. and since they certainly didn't advocate it,
I can see how they may have had a legal argument against futuremarks claim.
but they are so weaving their will into the industry. they minced words when it came down to the number of onboard pipes, they released an outright cheat on an influential benchmark, and now they strongarm a tiny scandinavian software firm trying to protect its product.
what will we talk about now...this forum was so alive!
I can see how they may have had a legal argument against futuremarks claim.
but they are so weaving their will into the industry. they minced words when it came down to the number of onboard pipes, they released an outright cheat on an influential benchmark, and now they strongarm a tiny scandinavian software firm trying to protect its product.
what will we talk about now...this forum was so alive!
Since "application specific optimization" can blatently ignore IQ, maybe I should try my homebrew driver optimization that detects the 3DMark03 splash screen and puts up a 640x480 8-bit screen that says "F*** you" for the entire test. Since I can't code worth a crap, 3DMark03 only returns a constant 500fps for each test which, according to Futuremark's formula, gives me 61550 3DMarks on a GeForce2. Man, ignoring IQ really allows a lot of freedom.
Megatomic
Lifer
- Nov 9, 2000
- 20,127
- 6
- 81
3DMark has been a louzy benchmark ever since the 2001 version. This whole soap opera, yes you are all involved in the most pathetic intarweb geek soap opera ever, has been ridiculous. I'm hoping this is drawing to an end soon, you are all wasting bandwidth (I guess I am now, too ).
Time to get a life.
).Time to get a life.
Jeff7181
Lifer
- Aug 21, 2002
- 18,368
- 11
- 81
I don't know about you people, but I look at benchmark results as a way of distinguising which card performs better so I know which one to buy. If you disable the optimizations the card uses in normal games for the benchmark, that's not an accurate depiction of it's performance in games. Yeah I know... blah blah blah...
Here's a solution to the problem... forget about 3DMark and rely on AnandTech's reviews =)
Here's a solution to the problem... forget about 3DMark and rely on AnandTech's reviews =)
i actually prefer extremetech.com game gauge, they use the standard games like ut2003, splinter cell, and quake3 but then they have a set of about 8 games besides those mentioned earlier that they run and give a composite score. the most games I've seen tested by a review site
rogo
rogo
I don't think you do know. It's not normal optimizations in games that are the problem. The "optimizations" (ie -cheats) that Nvidia included in their driver will work ONLY in 3dMark03. Since these "optimizations" only work in the benchmark and not in ANY games, it's a cheat to simply inflate their score.
Megatomic
Lifer
- Nov 9, 2000
- 20,127
- 6
- 81
Can I download this benchmark? I thought it was only for their internal use... I am really only interested in benchmarks I can run myself, at this point I'm not inclined to place much stock in internet reviews.Originally posted by: Rogozhin
i actually prefer extremetech.com game gauge, they use the standard games like ut2003, splinter cell, and quake3 but then they have a set of about 8 games besides those mentioned earlier that they run and give a composite score. the most games I've seen tested by a review site
rogo
Schadenfroh
Elite Member
- Mar 8, 2003
- 38,416
- 4
- 0
Originally posted by: GoHAnSoN
i can smell 100% free and open source "Open3Dmark" is comming our way....
w00t:beer:
yes, they are. They release a pdf documenting in detail why this is a cheat, only to take it back a week or so later.
megatomic
If you have the games then you can figure out how to run it with fraps to compile it but it's alot of work.
rogo
If you have the games then you can figure out how to run it with fraps to compile it but it's alot of work.
rogo
TRENDING THREADS
-
Discussion Zen 5 Speculation (EPYC Turin and Strix Point/Granite Ridge - Ryzen 9000)- Started by DisEnchantment
- Replies: 25K
-
Discussion Intel Meteor, Arrow, Lunar & Panther Lakes Discussion Threads
- Started by Tigerick
- Replies: 21K
-
Discussion Intel current and future Lakes & Rapids thread- Started by TheF34RChannel
- Replies: 23K
-
-
Facebook
Twitter