New 3dmark03 patch - nVIDIA cheating ... again???

[chsh1ca]

Originally posted by: Jeff7181
Since I'm just me, why don't you ask HardOCP, which has a much better reputation than I do. Apparently they think the same thing I do.

One day it gives me one score, then the next day it gives me another score for the same exact hardware and software configuration? Futuremark seems to have a whim about something and all of sudden the scores change drastically without any true explanation.

*EDIT* By the way... I hate being misquoted and having my words twisted...

Originally posted by: Jeff7181
Looks like a bunch of BS to me... they're showing the 9800XT scores changing by only 1 point... that's HIGHLY unlikely... run 3DMark twice in a row without changing ANYTHING and I guarantee you won't come within 1 point of the previous test.

Twisted how? You yourself *guaranteed* that two runs in a row without changing anything means not within 1 point of the previous test. I had to run it more than twice consecutively in order for any variance to come into play, and thereafter it was consistently identical. That is not a misquote, you just misspoke and then were quoted on it. Please learn to deal with that. I notice you also refused comment on my numbers, disputing your theory.

 

[Jeff7181]

Originally posted by: chsh1ca

Originally posted by: Jeff7181
Since I'm just me, why don't you ask HardOCP, which has a much better reputation than I do. Apparently they think the same thing I do.

One day it gives me one score, then the next day it gives me another score for the same exact hardware and software configuration? Futuremark seems to have a whim about something and all of sudden the scores change drastically without any true explanation.

*EDIT* By the way... I hate being misquoted and having my words twisted...

Originally posted by: Jeff7181
Looks like a bunch of BS to me... they're showing the 9800XT scores changing by only 1 point... that's HIGHLY unlikely... run 3DMark twice in a row without changing ANYTHING and I guarantee you won't come within 1 point of the previous test.

Twisted how? You yourself *guaranteed* that two runs in a row without changing anything means not within 1 point of the previous test. I had to run it more than twice consecutively in order for any variance to come into play, and thereafter it was consistently identical. That is not a misquote, you just misspoke and then were quoted on it. Please learn to deal with that. I notice you also refused comment on my numbers, disputing your theory.

You never even posted your scores. You posted frames per second... I was talking about the scores posted on that site.

 

[chsh1ca]

Originally posted by: Jeff7181
You never even posted your scores. You posted frames per second... I was talking about the scores posted on that site.

Yes, I did:

Originally posted by: chsh1ca
Overall: 1013 | 1013 | 1015

Those are the overall 3DM03 scores I am getting on my MSI GF3Ti200 Pro-VT128.

 

[Jeff7181]

Originally posted by: chsh1ca

Originally posted by: Jeff7181
You never even posted your scores. You posted frames per second... I was talking about the scores posted on that site.

Yes, I did:

Originally posted by: chsh1ca
Overall: 1013 | 1013 | 1015

Those are the overall 3DM03 scores I am getting on my MSI GF3Ti200 Pro-VT128.

You're running 2003 on a GeForce3?

 

[Pete]

Originally posted by: DefRef

IS FUTUREMARK SABOTAGING NVIDIA BY ADDING CODE THAT WORKS LIKE "IF DRIVER=NV THEN FUX0R THEM!"?

A logical question and one that I shared, but it has since been answered in two ways:

1. xbit-labs reports that nV's PR people know less (or have less compunction about lying) than FM's, and that it's not possible for 3DM to disable nV's new compiler.
2. Tridam at B3D has done the following:

I've tried v330 and v340 shaders in another application. They showed exactly the same execution speed. So NVIDIA is doing replacement of these shaders ONLY in 3Dmark03. If a game uses the same shaders it won't have the enhancement.

So this is just the same behavior we've come to expect from nV. Whether they are cheating per 3DM's rules is not at issue, though whether their cheating is of any consequence is still able to draw a crowd.

Originally posted by: Jeff7181
Looks like a bunch of BS to me... they're showing the 9800XT scores changing by only 1 point... that's HIGHLY unlikely... run 3DMark twice in a row without changing ANYTHING and I guarantee you won't come within 1 point of the previous test.

chsh1ca proved you wrong, above.

Originally posted by: Jeff7181
As someone else pointed out... then why are the scores for the CPU tests effected as well?

Aren't the CPU tests also 3D, just at lower resolutions or with much less happening on-screen?

LOL, that's a great comeback. The scores are invalid because the hardware is too slow?

 

[BFG10K]

IS FUTUREMARK SABOTAGING NVIDIA BY ADDING CODE THAT WORKS LIKE "IF DRIVER=NV THEN FUX0R THEM!"

Nonsense. There is absolutely nothing FutureMark can do about nVidia's compiler nor their instruction shuffler. All FutureMark can do is prevent application detection and stop nVidia from cheating and this is precisely why nVidia's scores are going down but ATi's aren't. If nVidia were employing genuine and generic optimisations like they promised after the first fiasco then FutureMark's actions wouldn't have any effect on them.

As someone else pointed out... then why are the scores for the CPU tests effected as well?

Presumably because any GPU bottleneck still present in those tests is getting relieved by nVidia's driver cheats in a similar fashion that the GPU tests are, just to a lesser degree.

"This reminds me of the early days of CPU's, when for some weird reason the industry felt the need to run "unoptimized code" through fear of the "new" optimizing compilers.

Except nVidia's code isn't optimised, it's cheating.

MMX, SSE, 3DNow, SSE2... hmmmmm... I wonder what those do?

They perform genuine optimisations through SIMD, something completely different to nVidia's cheating. The two are nothing alike.

 

[Rogodin2]

I was on the road to damascus and "god" blinded me with the truth.

"nvidia is a cheating shoddy corp and radioshack puts them to shame."

I spit into the sunlight and "god" took my vision-but I still believe.

I'll sell you some 4 grade robusta for $50 a lb and label it "blue mountain", unless you're intelligents or a coffee person you won't tell the difference.

Analogy hold true.

Nvidia is scaming the public (not powerusers because they know) by selling them cards based on spurious PR that uses benchmark numbers that are rigged.

What would you do if you thought you were buying a $3 double quarter pounder and only got one tiny patty with some cheese? You'd take it back.

But nvidia can fleece you into paying 200-500$ for a video card that doesn't run up to it's specified potential-and your reaction is justification?

I"m just dumbfounded.

rogo

 

[OpTiX]

damn you people are hurting, how they hell do you cheat anyways???, oh wait so if i go out and buy a 2.1 ghz cpu and over clock it does that mean im cheating, god damn get over it already.....
-OpTiX

 

[ronnn]

Why does everyone think Nvidia is cheating? They are just making it possible for people to add nice scores to their tag line.

 

[PliotronX]

They just can't give up the corked bats...

 

[5ayle]

Forget all this Nvidia/ ATi crap... i'm getting me a Volari Duo!!!

 

[OpTiX]

Originally posted by: 5ayle
Forget all this Nvidia/ ATi crap... i'm getting me a Volari Duo!!!

LOL same here or i'll get a Wildcat 4

 

[SilverLock]

Guys what really matters, not to sound Clintonesque, is determining what a "cheat" is. Two companies have a different approach to solving the problem.

For me the issue is simple... if Nvidia is putting an optimization on their drivers that will be active in ALL games then I have no issue with it. If they build one that detects certain games or benchmarks then I have an issue with it.

I was a SOLID staunch supporter of Nvidia when it was 3dFX versus Nvidia... When ATI started to comeback I wanted to like them (They're Canadian) but they had GOD awful driver support. I tried ATI twice and was frustrated severly each time.. First time I got a card that booted up with no color - nice shades of grey though. Then I bought one that wouldn't boot on system unless I unplugged all the fans (with a 430W PS - which I then had tested and it was NOT defective). Now, I am going to try them again - 9600Pro but I am seriously considering getting the 5700Ultra or the 5900 non-Ultra.

 

[DefRef]

I didn't bother responding - though I did start drafting a reply, but the nested quotes got FUBARed and I was busy at work - to the wounding chimp howls of misinformation and FUD in reply to my last missive, but after seeing Kyle's piece at and his ability to ask the same question I had - What is FM doing to fuX0r over Nvidia? - while maintaining his jilted little girl whimpering about how Nvidia lied to him and broke his heart and shot his dog and drank the last of the milk and made his underwear ride up, I find it amusing that the ATI cabal here are still barfing up the same rationales for their bigotry, despite the fact that ATI STILL blows for compatability and just added KOTOR to the list of games it can't handle.

Amazing how the ATI faithful proclaim their superiority because they "follow the rules", yet they can't run commercial software without visual errors, if at all.

Carry on with the nonsense, y'all. ATI needs your blind loyalty as much as Apple does.

 

[UCSDHappyAsian]

hmm... i doubt the credibility of this test....
about nvidia's cheatin thinging.... it has to do with nvidia's 16bit and 32 bit vs ATI's 24 bit, and some people believe that one of nvidia's official driver was always use 16-bit even if the user manual set it to 32bit??
oh what exactly was it?

 

[Matthias99]

despite the fact that ATI STILL blows for compatability and just added KOTOR to the list of games it can't handle.

Amazing how the ATI faithful proclaim their superiority because they "follow the rules", yet they can't run commercial software without visual errors, if at all.

NVIDIA is not exactly flawless in terms of software compatibility and weird graphical bugs either.

And KOTOR a) just came out on PC, and b) is a port from an XBox game (ie, developed originally exclusively for NVIDIA hardware). Give ATI at least one driver release before you start bashing them...

 

[Genx87]

Defref I do find it kind of funny ATI supposedly follows the spec to a T and yet thier hardware has the most compatibility problems. Should we blame ATI or Microsoft?

 

[UCSDHappyAsian]

there are a lot well-known websites that does write reviews. They tried to be fair during the test. However, if one of the company not only send the reference card but also donate some cash to these websites. guess wut.... these websites staffs most likely would be biased.... or intentionally change a few settings, so the other company's product would struggle!!



mang... i read enough B$ reviews and some of the test results were really funny. hmm... therefore, i learned how to save my time and go to a website that is very trustworthy and popular---- ADANDTECH.com
hehe... i loved it when they are doin their tests and giving us a detail result.....

THANX ANANDTECH

 

[Rogodin2]

Whatever you nvidiots need to justify your acceptance of a sordid slimy vga company that is fleecing its' consumers is something I will never have to use or see-and NONE of my customers have to worry about "slimy" buisness practices when they buy my guanenteed high grade arabicas.

rogo

 

[chsh1ca]

Originally posted by: DefRef
I didn't bother responding - though I did start drafting a reply, but the nested quotes got FUBARed and I was busy at work - to the wounding chimp howls of misinformation and FUD in reply to my last missive, but after seeing Kyle's piece at ]H[ and his ability to ask the same question I had - What is FM doing to fuX0r over Nvidia? - while maintaining his jilted little girl whimpering about how Nvidia lied to him and broke his heart and shot his dog and drank the last of the milk and made his underwear ride up, I find it amusing that the ATI cabal here are still barfing up the same rationales for their bigotry, despite the fact that ATI STILL blows for compatability and just added KOTOR to the list of games it can't handle.

LMAO, this gave me a really good laugh. You think FM is 'screwing' NVidia -- without any evidence to that fact, mind you. Coincidentally, you bash the ATI 'faithful', and 'faith' is defined as "Belief that does not rest on logical proof or material evidence."
I also found your description of Kyle's complaints amusing.

 

[BFG10K]

how they hell do you cheat anyways???,

By boosting performance in an invalid fashion.

oh wait so if i go out and buy a 2.1 ghz cpu and over clock it does that mean im cheating,

Not at all.

 

[Blastman]

It?s funny, some people who work in the industry have a difference view about hardware compatibility, NV and ATI ?

link ...Actually, I find the discussion of driver bugs on the list to be often useful, as I then log them on my list of 'driver bugs to make sure we are not effected by..." and to test our own code against. With drivers changing so often and the fact the new drivers always seem to introduce new bugs (unfortunately), this has become a necessary evil.

A few years back, ATI bugs filled most of my list. Now nVidia seems to have taken the title by quite a wide margin.... and as has been mentioned recently, WHQL testing is often more concerned about avoid BSOD than it is about 'accurate' operation of the driver.

-- George. - George Geczy - Lead Programmer, BattleGoat Studios - www.supremeruler2010.com - "Designing the Next Generation of Intelligent Strategy Games"

 

[spam]

What I would like to understand more of, is how the Microsoft WHQL process works. What level of knowledge would MS have on these video drivers? Would they know if there is application detection? or any of the other questionable practices? Could they request that the drivers be unencrypted so that these practices could be spotted and expurgated?

 

[spam]

More grist for the mill, the Det 53's reactivate illegal optimizations. For some reason I can't link to it so here is the url, just copy and paste. http://www.elitebastards.com/page.php?pageid=2647&head=1&comments=1

 

[Matthias99]

AFAIK, the WHQL process has nothing to do with visual quality, optimization, etc. It's just Microsoft testing it to make sure it doesn't interfere with their fixes or cause system instability. And making yet more money, but no big surprise there.