Futuremark Responds to Nvidia 3DMark03 Criticism

[NFS4]

Saw over at Rage3D:

NVIDIA bitchin'

Futuremark response

 

[John]

ATi and Futuremark's Response to NVIDIA's Report

 

[SgtZulu]

Just sounds like Futremark is trying to justify their 177MB piece of bloatware

 

[yllus]

I suppose we'll have to wait and see if 3DMark03 really turns out to be an accurate synthetic benchmark when considering performance in the new generation of (DX9) games. Though my personal bias is heavily tilted to the "hellz no" side.

 

[human2k]

its just a benchmark, no need to pee in your pants, use something else....

 

[NFS4]

The fact that NVIDIA released new drivers specifically to increase the score in 3DMark03 (and not in any other program) gives me the impression that they are full oh crap...or at least hypocritical.

 

[Trader05]

Yes, this bench is just a way of saying your stuff is outdated and go buy new stuff...if it works to your liking, no need to upgrade in my own opinion... i'm not going to upgrade my gf4 4200 cause it sucks in a benchmark... whatever happened to benching your system of video games, like quake 3/unreal 2

 

[Megatomic]

Originally posted by: Trader05
Yes, this bench is just a way of saying your stuff is outdated and go buy new stuff...if it works to your liking, no need to upgrade in my own opinion... i'm not going to upgrade my gf4 4200 cause it sucks in a benchmark... whatever happened to benching your system of video games, like quake 3/unreal 2

Agreed. My GF4 Ti 4200 128MB card has lots of life left in it. Especially since it now thinks it's a Ti 4400.

 

[Insidious]

yep, yep, yep....

Not quite ready to ditch the GF4s yet..... not even close!

 

[ElFenix]

i didn't put much stock in 3dmark2k1, i'm not gonna put too much stock in 3dmark2k3. the only performance that matters is games that i actually play.

 

[AtomicDude512]

Originally posted by: SeekingTao

Originally posted by: Trader05
Yes, this bench is just a way of saying your stuff is outdated and go buy new stuff...if it works to your liking, no need to upgrade in my own opinion... i'm not going to upgrade my gf4 4200 cause it sucks in a benchmark... whatever happened to benching your system of video games, like quake 3/unreal 2

Agreed. My GF4 Ti 4200 128MB card has lots of life left in it. Especially since it now thinks it's a Ti 4400.

Mt Ti4200 didnt score to well in 3DMark2003, but it still makes 70fps at 1200x1000 (something like that) with details maxed. I am happy with it and wont replace it just because some stupid benchmark ditches it as "last generation" hardware.

 

[GTaudiophile]

Originally posted by: NFS4
The fact that NVIDIA released new drivers specifically to increase the score in 3DMark03 (and not in any other program) gives me the impression that they are full oh crap...or at least hypocritical.

And do I hear ANYONE screaming bloody murder like they did with ATi's Quake/Quack issue? No!

 

[cnicol]

Right no reason to go and ditch your Geforce 4's just yet. Those cards will be fine for a year or two more.

But this benchmark does now show the GF4mx cards for what they are! Garbage!
I can't help but laugh when I see people getting mad because they are just now finding out that their GF4mx card is only a directx 7 card!

 

[ElFenix]

Originally posted by: cnicol
Right no reason to go and ditch your Geforce 4's just yet. Those cards will be fine for a year or two more.

But this benchmark does now show the GF4mx cards for what they are! Garbage!
I can't help but laugh when I see people getting mad because they are just now finding out that their GF4mx card is only a directx 7 card!

heheh its really a GF2 on steroids

 

[NFS4]

And another thing...

I've NEVER put much stock in 3DMark...in fact on many occassion of I have shown my utter hatred towards the benchmark...

But 3DMark03 is no more "bogus" or "irrelevant" than 3DMark01/3DMark01SE & THAT is my main beef with NVIDIA and their "run to mama" antics now.

 

[Spicedaddy]

To people whining: what do you suggest reviewers use to benchmark DX9 video cards at this moment?


The only other DX9 programs I can think of are the nVidia and ATI demos, you guys want to use that instead? If you're gonna complain that it's useless, then at least suggest something better to test DX9 hardware.



And about nVidia's whining, I think it's obvious they're trying to discredit 3DMark03 because, at this moment, their cards from every market segment are getting spanked in the benchmark. (GF4MX vs 8500/9000, GF4Ti vs 9500, FX beats the 9700 slightly but isn't available...)

 

[Pariah]

But 3DMark03 is no more "bogus" or "irrelevant" than 3DMark01/3DMark01SE & THAT is my main beef with NVIDIA and their "run to mama" antics now.

Yes it is.

2001SE Results

Rankings in this order: 9700Pro,9700,Ti4600,9500Pro,Ti4200,9000

Most people would agree that is they way they would be ordered in real games as well with even the percentage difference pretty close. Now you look at 2003 and you find the 9700Pro way out it in front followed by the 9500Pro way in front of everything else. No games available in any way duplicate these results at 1024 with no AA or AF which is what this benchmark defaults to. The almost complete irrelevance of the CPU speed is also not indicative of any real world gaming situation.

To people whining: what do you suggest reviewers use to benchmark DX9 video cards at this moment?

Can you name one announced DirectX 9 game? What's the point of benchmarking something that doesn't even have announced support let alone available support? There's nothing wrong with benchmarking DirectX 9 support, but it should not account for 50% of the score of a major industry benchmark this early in the game. DirectX9 benchmarks should be an extra test like the pixel shader tests and bump mapping were in 2001. Once DirectX 9 games start approaching (which at this point looks like no time soon), an update to 3DMark2003 should be released to reflect the changes like 3DMark2001SE was to 3DMark2001

 

[Spicedaddy]

Can you name one announced DirectX 9 game? What's the point of benchmarking something that doesn't even have announced support let alone available support?

Do you buy a video card every time you buy a new game? Most people keep their video cards for about 2 years, can you tell me there won't be any DX9 games during that time?

There's nothing wrong with benchmarking DirectX 9 support, but it should not account for 50% of the score of a major industry benchmark this early in the game. DirectX9 benchmarks should be an extra test like the pixel shader tests and bump mapping were in 2001. Once DirectX 9 games start approaching (which at this point looks like no time soon), an update to 3DMark2003 should be released to reflect the changes like 3DMark2001SE was to 3DMark2001

Huh, remember the Nature test in 2001? What DX8 games were announced when 3D Mark 2001 came out that warranted the Nature test?? (which only one card ran at that time)

And where's your suggestion for an alternative to 3DMark03 for testing DX9 hardware?

 

[Vernor]

But this benchmark does now show the GF4mx cards for what they are! Garbage!
I can't help but laugh when I see people getting mad because they are just now finding out that their GF4mx card is only a directx 7 card!

Yeah, the market is just saturated with games that require shaders.


Most people buying these cards will probably never know they're missing something.

 

[Wolfsraider]

Originally posted by: Spicedaddy
To people whining: what do you suggest reviewers use to benchmark DX9 video cards at this moment?


The only other DX9 programs I can think of are the nVidia and ATI demos, you guys want to use that instead? If you're gonna complain that it's useless, then at least suggest something better to test DX9 hardware.



And about nVidia's whining, I think it's obvious they're trying to discredit 3DMark03 because, at this moment, their cards from every market segment are getting spanked in the benchmark. (GF4MX vs 8500/9000, GF4Ti vs 9500, FX beats the 9700 slightly but isn't available...)

i own a 9700 and i think this benchmark sucks

no i can not give a recomendation for another benchmark as there are none
but by the same token i am not going to use something just because it came first?
if this was a valid benchmark then great
but try this

set your overclock program to the lowest settings
start this benchmark.after it starts abort the benchmark
reset your overclock program to max for your card
restart this benchmark and let it run
when it gets to the results page compare that score to this

leave your card overclocked
run 3dmark2003 till it ends

compare scores then see the difference

does it matter ...no
but it isn't an accurate test now is it?

thanks but i will wait for a better app or game to compare my system to.
it is free so why am i bitching? well hmmm... how many posts are "my system isn't getting a high enough score in 3dmark2003 so something must be wrong."


just because 1.5 million people downloaded it in the first 72 hours doesn't mean its not a bs benchmarking tool.

how many people bought crap (insert product name here) just because of the name?

am i whining?maybe or maybe i am just stating obvious reasons that this kind of "tool" is useless.

mike

 

[ViRGE]

Originally posted by: GTaudiophile

Originally posted by: NFS4
The fact that NVIDIA released new drivers specifically to increase the score in 3DMark03 (and not in any other program) gives me the impression that they are full oh crap...or at least hypocritical.

And do I hear ANYONE screaming bloody murder like they did with ATi's Quake/Quack issue? No!

Quack'd drivers were shipped as part of MS qualified drivers, and there was a noticable quality decrease in order to get those higher rates. For the time being, Nvidia's "funny" drivers, the 42.68's are neither qualified, or officially up for grabs for the public yet, so I'm giving them the benefit of a doubt at this point. If they send out qualified drivers that purposely induce a quality loss however, then they will be in the same boat as ATI.

 

[Pariah]

Do you buy a video card every time you buy a new game? Most people keep their video cards for about 2 years, can you tell me there won't be any DX9 games during that time?

Within 2 years there will be a new 3DMark making this one obsolete before it becomes useful.

Huh, remember the Nature test in 2001? What DX8 games were announced when 3D Mark 2001 came out that warranted the Nature test?? (which only one card ran at that time)

And remember how that card didn't more than triple (see 9700Pro vs any non DX9 card) the score of the cards that didn't? And remember how when you changed your CPU, it actually affected the score too?

 

[OS]

I got 18 3dmarks!

I don't really care about my score tho, I just want to be able to see the pretty demos and my GF2 GTS-V can't show it.

I dunno, on one hand 3dmark tells me my system is crap, but it plays UT2k3 just fine.

 

[Adul]

http://www.hardocp.com/article.html?art=NDMx

ATI's take on the whole orderdea,

 

[Spicedaddy]

And remember how that card didn't more than triple (see 9700Pro vs any non DX9 card) the score of the cards that didn't? And remember how when you changed your CPU, it actually affected the score too?

I see your point, you're right about the scale being questionable. That's why I see it as a good way to test DX9 hardware, nothing more. (and you don't have to use the scale, just looking at the FPS gives a better idea IMO)


But I disagree about the CPU comment. If you want to see how your CPU performs, look at the CPU Marks... The whole point of a 3D benchmark is to isolate the video card's performance, being CPU limited would make it useless as a 3D benchmark.