GFX 5900 Caught Cheating

"We have now established that nVidia's Detonator FX drivers contain certain detection mechanisms that cause an artificially high score when using 3DMark03. We have just published a patch 330 for 3DMark03 that defeats the detection mechanisms in the drivers and provides correct results." this is how Futuremark introduce their latest patch for their popular 3DMark03 software.

Saratoga, Calif.-based Futuremark issued a statement claiming that nVidia tweaked software needed to run its new GeForce FX 5900 chip, in order to distort performance in Futuremark's 3DMark 03 testing application. Futuremark is one of the leaders in software and services for performing PC benchmark tests.

According to Futuremark when the patch was applied a drop of as much as 24.1 per cent was observed in certain nVidia products, while competition products performance-drop stayed within the margin of error of 3 per cent.

An nVidia spokesperson said Since nVidia is not part of the Futuremark beta program (a program which costs of hundreds of thousands of dollars to participate in), we do not get a chance to work with Futuremark on writing the shaders like we would with a real applications developer. We don't know what they did, but it looks like they have intentionally tried to create a scenario that makes our products look bad.

ATI meanwhile are not too keen to capitalize on another, possible, embarassment for their rivals. ATI's Dave Baumann, partly in response to nVidia's statement said:
Despite still being a full Futuremark Beta member, ATI did not make it out of the report entirely unscathed either. There is a performance difference of about 8 per cent in Game Test 4, that accounts for about a 2 per cent difference in the final 3DMark03 score, between the new and old versions, indicating that although not visually different something was occurring on this particular test.

In order to add further validity to the importance of benchmarking tests ATI's Christ Evenden stated:
The 1.9 per cent performance gain comes from optimization of the two DX9 shaders (water and sky) in Game Test 4 . We render the scene exactly as intended by Futuremark, in full-precision floating point. Our shaders are mathematically and functionally identical to Futuremark's and there are no visual artifacts; we simply shuffle instructions to take advantage of our architecture. These are exactly the sort of optimizations that work in games to improve frame rates without reducing image quality and as such, are a realistic approach to a benchmark intended to measure in-game performance. However, we recognize that these can be used by some people to call into question the legitimacy of benchmark results, and so we are removing them from our driver as soon as is physically possible. We expect them to be gone by the next release of CATALYST.

What does all this mean however? Most knowledgeable gamers do not buy a card based, solely, on the score achieved in a benchmarking test. It is reliability, performance and value for money which usually determine which card is bought. Benchmark tests are mainly utilized in order to claim the performance crown, giving the bearer increased prestige and an advantage created by the publicity. All that cheating on tests will achieve is to render such tests useless and to create consumers who do not trust the manufacturers. Unfortunately, this time, the reputation of the biggest graphics chip manufacturers has been questioned and there is little chance that they will make it out of it without some doubt remaining in the publics mind. Doubt which can only be cleared by the release of solid, reliable and high quality products.

Add new comment

CAPTCHA
This question is for testing whether you are a human visitor and to prevent automated spam submissions.

Comments

GFX 5900 Caught Cheating

sscrew - you are correct. as i have said (adinfanitum) not drawing objects that are not onscreen is not a cheat. drdeath DOES try and make it sound otherwise.drdeath - yes nvidia was a supporter of the 3dmark program UNTIL they decided 3dmark 2003 tests were so far removed from what REAL games will be programmed like that they became of no real use. that is they are not reprasentative of current or future dx9 games.

GFX 5900 Caught Cheating

sscrew - correction, FM did say they thought ati was cheating aswell but they put it in small text in the pdf file, while NVIDIA CHEATING was done big and bold on the upgrade page. funny huh ?drdeath - now you want to critisize anyone who says 3dmark is not a valid benchmark ? are you on their payroll ???? for someone who perports to be "unbiased" you sure like to put nvidia down and scream praise on ati. even when nvidia is faster you cry "its only a little" and "ati win some too".

GFX 5900 Caught Cheating

oh and drdeath since you seem to "know it all" i am still waiting for you to tell me how a crappy little review site like extremetech got the developers version of 3dmark 2003 when the higher profile hardware/review sites dont have it.or how about explaing why FM did the nvidia cheating big and bold but hid the ati part in the pdf ?and nvidia said it may be a smart clipping problem waaaay back when it first came out. the bug showing up in 3dmark 2001 is also a sure sign there is a bug in the drivers that could be related to that. dont want to remember any of that though do you.

GFX 5900 Caught Cheating

"Its just they removed offscreen items, but that doesnt affect the experience at all.... i thought you would of known that..."SSCREW you know that's cheating plain and simple. Even John Carmacks definition of cheating vs. Optimization clearly states this. To say otherwise is plain foolish. And the effected camera path that they took DID effect the out-put of IQ. That has been proven! Why are you guys making all this up that it did not. At least 6 sites have proven this as FACT. When they did this IQ was lost and artifacts in the test were present all over. ATI had no such images. Not one. IQ was never lost. NOR DID THEY REDUCE THE PRECISION TO RUN THE TEST TO GET FASTER SPEEDS. When they compared pictures to each other. IQ quality on the 5900 was degraded. This has been proven."as i have said (adinfanitum) not drawing objects that are not onscreen is not a cheat. drdeath DOES try and make it sound otherwise."Sorry, that's just not my opinion. That is practically EVERYONES opinion that has not been swayed by Nvidia. You should read articles about this at Beyond3d. Theinquirer. Techreport. Geek.com These are unbiaised news reports. Why do i say unbiased? Because Nvidia won't talk to them anymore because they talk straight from the hip."smart clipping problem waaaay back when it first came out."We all know that was a lie. That has been proven, not to be a "driver" bug. Because Nvidia still has not released a patch for that "driver" bug. When you purposely do not render images off screen and reduce the precision to what is less then the standard dx9 spec, it is cheating.

GFX 5900 Caught Cheating

"Rewriting shaders behind an application's back in a way that changes the output under non-controlled circumstances is absolutely, positively wrong and indefensible.Rewriting a shader so that it does exactly the same thing, but in a more efficient way, is generally acceptable compiler optimization."That was John Carmacks definition of cheating vs. Optimization. The output was changed. Images were not rendered off screen to take the work load off the card. That is a cheat. Plain and simple."ATi didnt cheat, they admitted to doing so, in fear Futuremark would reveal there opitimisations as cheats..."FutureMarks already revealed the method inwhich ATI took to get that 1.9% increase. And in doing so found out that ATI took the exact same path AS PRESCRIBED BY FutureMarks and lost no IQ whatsoever. They only reorganized it's instructions and improved performance by 1.9% And the final outcome was exactly the same. That is what Tim Sweeny and John Carmack call OPTIMIZATION. Read their article again carefully."the bug showing up in 3dmark 2001 is also a sure sign there is a bug in the drivers that could be related to that. dont want to remember any of that though do you."I have to bring that up agin. That excuse was the joke of the week when Nvidia released that PR spin. So far they have released NO PATCH OR FIX FOR THAT SO-CALLED BUG. No mention of that has ever come up again. 43 web-sites PROVED this was not a bug. Clipping planes. Objects not rendered are not "bugs" That was targeted. Are you people reading your info from Nvidia's website?"its only a little" and "ati win some too".Is that not the case??? Like i said also. This was the same scenerio with the 9700pro vs. 5800Ultra. 9800pro vs. 5800Ultra. All cards were close.

GFX 5900 Caught Cheating

"Futuremark claims that Nvidia is not cheating. If we were Nvidia's lawyer, we'd never have allowed the sentence that includes the phrase "and not a cheat" to appear in a statement, because it sort of implies something, doesn't it?We're not entirely sure if Nvidia has threatened Futuremark with legal action, because Nvidia doesn't talk to us any more, and has also threatened us with legal action too. [What, for calling them Nvidiots? Eva.] Nvidia is not the first multinational corporation that's done that. Nor the last, we reckon.But if any more pixels are wasted on this futile Futuremark-Nvidia topic, we'll get the Green Police on their respective cases. [Note to columnists. No more column pixels about this crap, OK? Ed.]The joint Nvidia-Futuremark statement uses some very weasel-like words in what's obviously supposed to be a face-saving exercise, but reads more like a soap opera.Any joint statement that uses the phrases "unfortunate situations" and "moving forward" indicates clearly to us that m'learned fiends have been involved. No marketing person would ever employ such terms.And so it transpires that both Futuremark and Nvidia have egg on their respective faces, yet again...."That was from the Inquirer. Any site that has been posting things about this issue about Nvidia has been black listed by them. Truth or not.But FM saying we will consider allowing NVidia to drop precison below the DX9 spec! if their Cinematic Rendering 128 bit capable cards can't keep up seems like such a payoff."Both companies have an understanding. And Nvidia didnt lower the quality onscreen, the quality was equvialent of the 9800pro."They did lower the quality onscreen. The lower precision from the standard fp24 was reduced to 12 or 16. When they compared screen shots. It was proven that IQ was lost! Just the fact that Nvidia lowered the precision(which is not a driver "bug" anonymous) in a dx9 test is cheating. Because Nvidia KNOWS the required

GFX 5900 Caught Cheating

.....Because Nvidia KNOWS the required specification of dx9 applications. It allowed them to run it faster.With IQ being lost in the process. Like i have said before. That has been proven! If you want to believe the "trumped" up company saving statement FM released, then you people need help. Anyone with a brain knows that whole statement was said to save itself from a lawsuit. SSCREW i know you don't think FM said all this because they just found out that they were wrong. It's like 2 people running a race on a set race track. And one of them rides a horse across the field to win... Nvidia lowered precision to less then what the REQUIRED dx9 spec allowed.(FP24) ATI never did that. Then they ALTERED the camera set path. ATI never did that. They DID NOT render objects off screen to ensure the work load was taken off the card to improve speed. ATI never did that. They rendered all objects. End result? IQ was lost by the 5900,5800.

GFX 5900 Caught Cheating

"As you know, 3DMark03's overall score is derived from its four game tests. You can see my results for Game 1, Game 2, Game 3, and Game 4.As you can see, ATI's optimizations—which they claim didn't change image output—barely affect the overall score. NVIDIA's, however, make a substantial difference. The most striking difference between builds 320 and 330 is in the Game 4 test, where performance drops dramatically once the cheats and optimizations are disabled. This test, the "Mother Nature" scene, makes the most extensive use of DirectX 9 and pixel shader 2.0.Those are the numbers. I don't have time here to dig into all of the related issues, but Dave from Beyond3D sent me a note about a couple of things you should check out on his site. First, Dave has captured images from ATI and NVIDIA cards in 3DMark03 to show the image quality differences between builds 320 and 330. The NVIDIA drivers clearly have more impact on image quality. Next, to help you sort out what that fact means, have a look at Unreal guru Tim Sweeney's take on cheating versus optimization. The basic principle he outlines seems like a good guide in this case."I like the people that comment and never leave a id name. You want to call me a fan-boy, go ahead. Like i care. But will never defend something with great zeal and effort that is obviously wrong. That is what i call a fanboy. Someone who says someone or something can do no wrong even though the evidence is plain as rain. The evidence is there in spades regarding this cheating issue with Nvidia. It has been proven. Clipping plains. Not clearing buffers. Lowered precision that defies the standard specifications. Altered camera paths that cause major artifacts when the test is running.(That ladies is lost IQ)Just read Carmacks and Tim Sweeny's article on it. I am just posting facts about this whole ordeal. FACTS. Not conclusions from Nvidia's or FM's PR department.

GFX 5900 Caught Cheating

Here's something for you f**king Nvidia fan-boys. Looks like FM just wanted the threats to end, so they gave in to NVIDIA, but then said afterwards that they were still cheating...Q: Does this mean what you called originally as "cheats" actually were acceptable "optimizations", and that you made a wrong decicion in releasing Patch 330 and the Audit Report?A: By the definition of our benchmark and process, the optimizations are not acceptable. 3DMark scores are only comparable if drivers perform exactly the work 3DMark instructs them to do.The statement also says: Quote: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.As earlier stated, we recommend using the latest build 330 of 3DMark03, with the 44.03 (or 43.51 WHQL) Nvidia drivers, or the Catalyst 3.4 ATI drivers. This way obtained 3DMark03 results are genuinely comparable as far as we know.Q: What is the reasoning behind this statement?A: Both companies want to end the public dispute that has been going on since we launched 3DMark03 in mid-February this year.Q: Did NVIDIA pay you any money to make this statement?A: No, they did not. Our companies had a mutual desire to end this dispute, and we are very pleased that we reached this agreement.Q: Does this mean that in the future you will not make patches for 3DMark03 (or 3DMark2001) in order to reveal cheating?A: We might release further patches to 3DMark03, if a need for preventing driver optimizations appear in the future._________________Patric Ojala - 3DMark Producer

GFX 5900 Caught Cheating

'yes nvidia was a supporter of the 3dmark program UNTIL they decided 3dmark 2003 tests were so far removed from what REAL games will be programmed like that they became of no real use.'Are you f**king serious Anonymous? 3DMARKS NEVER WAS A TRUE GAMING BENCH TEST. 3Dmarks2000,2001 or otherwise. Never was. It was always a synthetic test. A test EVERY MOTHERf**kER supported until now. Why? Becuase Nvidia f**king cried? Because their cards cannot run a test as fast as the competition can and then they reduce the f**king floating point precision to match speed which does f**k up quality. Case in f**king point. My card runs at a higher level of performance. Yours at a lower performance. Both look the same. How the f**k is that possible??????? Because if that was true and it's f**king not, then ALL companies would run their cards at a lower floating point precision to get that extra speed because Quality stayed the same. Sorry assholes. Quality did change for the 5900. That's why this whole f**king issue is about. Sorry. dr.death may talk a f**king lot, but he's right on the mark. Nvidia ran the f**king tests that altered the f**king end result. That's cheating anyway you f**king slice it. You assholes want to defend that? Go ahead fanboys. If ATI did that i would say the same f**king thing about them too.

GFX 5900 Caught Cheating

If any of you nvidia lemmings want to believe this "joint" statement was not forced upon futuremarks needs to stop smoking hard core drugs. This was all done to save futuremarks from a legal lawsuit that futuremarks would win, but would cost them millions in litigation and court costs which the small company cannot afford and probablly bankrupt them. Because the cheating aspect inregards to 3dmarks has been proven.Kiss my ass Nvidia, posted the Q&A from futuremarks and it's a funny read. It still claims that they are cheating but not coming out and saying it. What a f**king joke. And all this crap coming right after a remark of support for 3dmarks from the industries largest OEM. Dell. Another reason Nvidia wanted to make amends and give futuremarks a reason to get onboard(lawsuit included) It's all about the money. Dell supporting futuremarks and nvidia does not? Guess nvidia cards would not be in Dell boxes. Money. Money. Money. Money. Money. Money. And the end result of this whole bullshit fiasco is Nvidia and futuremarks look like little kids that are trying to save face amoung all the lies and pr bullshit. And sccrew. This does not look good despite what you say. Both asshole companies lot major credibility. ATI must be laughing like crazy

GFX 5900 Caught Cheating

----drdeath, god you talk sh*t. how much are ati paying you to babble on ? try ready the article about not using 3dmark again moron[url removed] you will see that dell are basing their decisions on this 1 unreliable test. ye thats real smart.----Hey Anonymous i did read that link. And it sounds to me that hardocp is 100% behind nvidia to even state that no one should use this benchmark. Now that is biased beyond belief as i have not seen anyother site saying this. It looks plain to see that they are involved someway with nvidia or does not want to get in the bad books with them. Oh well, that's the last time igo to a site tht seems to be too afraid to print facts and truth.

GFX 5900 Caught Cheating

Firstly, FM does not offer a PS1.4 fallback for PS2.0. There is no fallback for PS2.0 - either a card can run Game 4 - Nature (the only PS2.0/DX9 untilising part of 3DM03) or it can't. There is a PS1.1 fallback for PS1.4, as the GeForce 4 Ti 4x00 cards don't support PS1.4, but that's only relevant for Games 2 & 3.Secondly, a card should not use a lower precision than 24-bit Float/channel for Game 4 - 24 bit FP is the minimum spec for Direct X 9. If you use FP16 or FX12, then you are not performing a DX9 test - I'm surprised if Microsoft would let through a driver that did that in WHQL. Blame nvidia for being stupid enough to support FP16 (which is below DX9 minimum specs) and FP32 (which is above DX9 minimum specs, but definitely permissible), yet have both FP paths be so much slower than the competition - so slow in fact that John Carmack had to use the FX12 path on the FX chips to get decent performance (but in that case, it's okay since OGL has no minimum specified bit depth. At least that is what most people think. It still has to run at the spec of FP24 minimum). This may sound pedantic, but that's what standards are there for (and nvidia has had longer to meet those standards than the competition, since the FX line was over 6 months later than the R300 series). If for example some (unnamed) company put out a lousy integrated graphics chipset but cleverly wrote drivers to fool applications into thinking it was DX9 capable by letting the CPU do the PS/VS emulation, we'd all be screaming bloody murder.BTW a Platonic ideal is one where there is no sex involved. Maybe you're thinking of a utopian ideal?The actions of FM and nvidia both cast doubt onto the integrity of the industry.By first blowing a whistle and providing detailed information on an alleged (let's face it, the evidence was pretty damn convincing) "cheat", and then backtracking on the issue, FM has undermined its own integrity in providing an impartial and ultimately useful benchma

GFX 5900 Caught Cheating

....ultimately useful benchmarking too and/or standard and cast doubt on its commitment and ability to police its own product. This means that it is no longer an industry "standard". In other words, they have gone from Touchstone to Keystone Cops...By showing the public that it would rather bully someone into silence than to address allegations of cheating, nvidia has cast doubt into every single standard game benchmark run on its cards. This means that they are openly supporting the policy of doctoring their device drivers for given benchmarks. In the most extreme case, this could mean that every single Q3, UT2K3, commanche, RTCW etc benchmark run on an FX card could be utterly worthless, as there is no way of knowing just how much a game benchmark has been optimised or flagrantly cheated with. Note that it is possible to optimise a game benchmark without optimising the actual game to a similar degree, and obviously any cheating done on a game benchmark would not translate to real-world performance gains. The real-world situation is probably not as extreme as the "extreme" case I outlined, but there is a huge area of uncertainty cast over nvidia right now.Of course this is nothing new - even back in the day of 2d grafx and VESA cards, vendors were known for cheating drivers.The stupid thing that nvidia is doing is that by undermining the public confidence in its own products, they will flock to the only other viable competitor out there at the moment - ATI.This is why I remarked earlier that I felt this "joint announcement" will be ultimately injurious to both nvidia and FM - and that they were boneheaded to think that they could pull off a cheap shyster trick on the entire Internet community.

GFX 5900 Caught Cheating

nvidia laughing at us?, maybe but alot people wont buy there card so whos going to be laughing now.doesnt matter, nvidia does suck and will suck and will always suck.There products are too crappy for my high end pc so I go with ATI.

GFX 5900 Caught Cheating

You (DRDEATH) are wasting your time posting your message here. Of all the people, you are the most retarded person who wasted his time posting facts. I find most of your posts are pure BS. You didn't provided any source to backup your post.No concrete proof = PURE BULLSH*TCome up with better source or don't post at all.

GFX 5900 Caught Cheating

...dork...what do you worship hardware makers like they are rock bands or something? go sit in the corner with your nvidia hat, t-shirt and fake tattoo and STFU!ati is doing well and has earned some respect in the market. nvidia has been snoozing after the success of the gf4 ti series. ati apparantly has been hard at work. i hope nvidia picks up the pace and blows away ati with thier next card....and then i hope ati does the same....and then nvidia....get it? if there is only one company around, then they will make shit progress. there must be TWO GOOD COMPANIES, that both make GOOD CARDS, and DO SHIT LIKE THEY ARE SUPPOSED TO, for us to see really great technology gains, while not having to pay out the ass for them. if either company "wins" the corparate war...we lose. we win as long as they stay roughly tied.do yourselves a favor, have no loyalties except to performance. buy the better shit. period.

GFX 5900 Caught Cheating

Dude, [H]ardOCP must be on crack.First of all the hundreds of thousands of dollars to be in the Beta program is completely inaccurate. nVidia were part of the program and they had the moneyto continue being in the Beta program, but they decided their card wasn't doing well in the shader tests, so they pulled out and tried to discredit the test.It's sites like [H]ardOCP that stand behind these kind of bully tactics that damage the graphics enthusiast community. Many readers visit their site and take what they say as fact. It's too bad they side with the playground bully who's a little slower, but has the monetary might to overcompensate for it.

GFX 5900 Caught Cheating

QUOTE:> "DeQuosaek, try using that single brain cell before YOU start spouting off. yes there isnt any directx9 games yet and there wont be for a while. but if you bother to read futuremarks statement they say they use unoptimized routines for gfx while they use different for athlon and P4 cpu's. hmm considiering that both ati and nvidia usually work with game developers to get the most out of their cards then that shows that 3dmark is in no way representative of real games development."nVidia made the choice not to be part of development, so it's their own fault if they didn't have more input. And do you believe they couldn't afford it. Yeah. Right. Trident can afford it but nVidia can't? PR BS.

GFX 5900 Caught Cheating

I like ATII like NvidiaBut I have to say after the release of the 40.xx drivers, and then the huge flop of the 5800 ultra, and now the obvious cheating by nvidia to get better scores.. It's painfully obvious who I am liking more now a days.

GFX 5900 Caught Cheating

Not drawing objects offscreen is not cheating. It's smart culling.Ignoring a whole set of shader instructions and replacing them with a completely rewritten one that changes image quality is. And there is proof. I don't feel like linking it, but if you've been reading you've seen the pictures at beyond3d's forums.Read the statements FutureMark before nVidia decided to use it's position to threaten shutdown of a small HONEST company. They stated that nVidia's drivers ignored the shaders for the sky and water and completely replaced them. Thereby changing the images. That's more than just not drawing offscreen stuff.

GFX 5900 Caught Cheating

QUOTE:> "oh and drdeath since you seem to "know it all" i am still waiting for you to tell me how a crappy little review site like extremetech got the developers version of 3dmark 2003 when the higher profile hardware/review sites dont have it."I'll field this one. You don't have to be a know it all, just read a little. They are a member of the Beta program. Easy answer.

GFX 5900 Caught Cheating

I wouldnt call it cheating anyhow. If you read what ATI had to say then the drivers are clearly doing there job. Cutting down the crap and achieving performance with practicaly nill visual loss. Nvidia's drivers obviously work differently to ATI's so one of them was bound to come off worse than the other. Shame it had to be Nvidia :( 20%'s big but when u take in to consideration that the patch forced the drivers to do something unnecessary that won't be used anywhere else, its not so bad.

GFX 5900 Caught Cheating

Anonymous, most of my posts are facts from direct quotes. If you had any knowledge of progamming then you would know this. As in relation to pixel shader info i posted, obviously you know absolutley nothing on the issue as you cannot challange what i say. Just that i am full of shit. Prove that then? Prove what i am saying is false tough guy Anonymous. Can you? Do you have any knowledge in this area? Or are you just talking crap. Prove me wrong then. Prove to everyone here that Nvidia did not cheat. That clipping planes and those other issues were "done" by accident. The one area which FX chips are week in. Shader performance.How much more proof do you people need in relation to this issue. Go to any harware review site and the proof is right there in black and white. Even the heavy weights in this industry have said so. Direct quotes from Carmack and Sweeny is not enough for the Nvidia fanboy? Screen shots at beyond3d is not enough? Degraded visual quality being detected is not enough?And this is a forum. Which people can render their OPINIONS as well as facts. And for this issue at hand, the facts are clear. Nvidia cheated. Got caught. And strong armed a smaller company that for weeks stood fast with their findings only NOW to do a 360 degree turn about that has stunned the entire internet community. And when you read about this in EVERY site that runs the story, they all come to the same conclusion: This was a forced statement. And the statement still claims Nvidia cheated but does not come out and say it. For the obvious legal reasons. Check the major sites on this. It's right there...

GFX 5900 Caught Cheating

QUOTE:> "And it sounds to me that hardocp is 100% behind nvidia to even state that no one should use this benchmark. Now that is biased beyond belief as i have not seen anyother site saying this. It looks plain to see that they are involved someway with nvidia or does not want to get in the bad books with them. Oh well, that's the last time igo to a site tht seems to be too afraid to print facts and truth."I agree 100%. Looks like [H]ardOCP should now be called [s]foftOCP since they just rolled over for nVidia instead of printing fact. Not very bold.

GFX 5900 Caught Cheating

>All that cheating on tests will achieve is to render such tests useless and to create consumers who do not trust the manufacturers. Unfortunately, this time, the reputation of the biggest graphics chip manufacturers has been questioned and there is little chance that they will make it out of it without some doubt remaining in the publics mind. Doubt which can only be cleared by the release of solid, reliable and high quality products.<<<That sums it all up. 24.1% drop in performance. That's cheating. Nvidia got caught with their hands in the cookie jar. 1.9% makes little difference. You know OEM companies like Dell and Compaq, IBM and the rest look at all this FutureMarks crap regardless how much the benchtest is so-called "not a true gaming test" And Nvidia and ATI know this. Regardless what Nvidia says that they no longer see it as a viable tool. OEM companies still do. And that means...CASH lost or GAINED.

GFX 5900 Caught Cheating

DRDEATH won't admit he is an ATI Fanboy. He always have something against Nvidia. This is why I consider he is a troll in this forum & deserved to be BANNED!!!!! We don't need you DRDEATH trash talking Nvidia. Go back to your ATI forum you retard Fanboy.

GFX 5900 Caught Cheating

One thing i find interesting is the fact that ALL dx9 ATI cards ran the Dawn demo perfectly and faster then the 5900, or the 5800 Ultra(9800pro) also in OpenGL wrapper. Which we all know degrades video presentation. So Nvidia's PR department that praised the FX Chip and the Cg graphics that the FX chips will employ(something that they said would give them impressive rendering power over ATI's chips.) turns out to be a major dud. ATI cards can render the same images and shaders with their harware and do it faster. And do it with OpenGL wrapper at that! Good PR campaign though at Nvidia for fooling everone that Cg graphics is the way of the future. We all now know that was just more PR crap from Nvidia. :-P

GFX 5900 Caught Cheating

Not a fanboy. Just stating facts inregards to this issue. And the real sad part about all this is the fact that the 5900 Ultra is a amazing card. And what Nvidia has done is take some of the spotlight off the card. Right now the 5900 is the fastest card on the market when it reaches retail in late June. I am running a Nvidia card in my Second PC MSI4200TI. No favorites. It's just that since Nvidia brought out this fx line,(5800)they have been so far from truthful about alot of things. IE: How many pipe lines the card had. Availibility. Quanity. Lied to consumers how the card was going to be the 9700pro killer when they knew it would not be. But made consumers wait. And wait they did.If i would upgrade right now and had the cash. I would purchase a 9800pro over the 5900 Ultra. And for the same reasons Hardocp choose the 9800pro over the 5900 Ultra. (have to wonder if they did another review would they think this way again...) Not because i am a fanboy. But when you stack both cards against each other, the 9800pro still runs quieter, cooler, the card is smaller. And it still is only a 1 slot solution video card. Not 2 like the 5900. And the performance is basically the same as the 5900pro. The 9800pro has better implementaion of FSAA. Which i want. AA is done better on the 5900 Ultra though. So it's a trade off. I want IQ and less noise and size. The 9800pro for now would be my choice. But both are sweet cards.And i am not trash talking Nvidia. They done this all by themselves. NO ONE FORCED THEM TO CHEAT TO DECIEVE CONSUMERS. DeQuosaek, myself and others are just stating facts that have developed since the fx line of chips have come out. And now this.

GFX 5900 Caught Cheating

There you go f**ktards...lmao. This must be a new asshole to this forum. That news release is what everyone has been talking about you f**k head. lol. A day late pal. And that joint statement if you have not been reading on every f**king website is so full of shit it's scary. Everyone knows that it was a forced statement to avoid possible lawsuits. And the fact that the statement came right after Dell released theirs proves this to a f**king tee. Since you are a f**ktard yourself for posting shit that EVERYONE here has been talking about for the last god knows how long, look at this "statement' from FutureMarks at beyond3d right after asshole. FM still utters that Nvidia did cheat. But weasles the words not to say it in fear of f**king lawsuits.Update: Reinforcing AJ's statement 3DMark Producer's, Patric Ojala, has written a small Q&A in our forum in relation to the joint statement. Here's a snip:Q: Does this mean what you called originally as "cheats" actually were acceptable "optimizations", and that you made a wrong decicion in releasing Patch 330 and the Audit Report?A: By the definition of our benchmark and process, the optimizations are not acceptable. 3DMark scores are only comparable if drivers perform exactly the work 3DMark instructs them to do.As earlier stated, we recommend using the latest build 330 of 3DMark03, with the 44.03 (or 43.51 WHQL) Nvidia drivers, or the Catalyst 3.4 ATI drivers. This way obtained 3DMark03 results are genuinely comparable as far as we know.Q: Does this mean that in the future you will not make patches for 3DMark03 (or 3DMark2001) in order to reveal cheating?A: We might release further patches to 3DMark03, if a need for preventing driver optimizations appear in the future.

GFX 5900 Caught Cheating

QUOTE:> "I will paste the link DeQuosaek. ;)[url removed] anon. This is the one I was referring to. You need to download the zip file of the images. The difference is very small, but it's definately different. Especially the shading of the turtle. The water is slightly different too.

GFX 5900 Caught Cheating

I agree with some of you guys. Competition will keep low prices.The thing is, even with the cheating, the 5900 is somewhat faster than the 9800 in games, BUT as a owner of a 9700 PRO I can tell you that if you got one of these cards (9700, 9800, 5900) you don't play games at 1024 with 2x AA and 4x AS, I know for sure I don't. I play my games at 1600 with 6x AA and 16x AS, and I only got a 1.3GHz Athlon! :D The difference between the 9800 and 5900 is quite small at high Anti Alianing and Antisoptric levels. And in my opinion, that's what counts the most. I don't care about 1024, or no AA or even 2x AA because it's stupid to use such low settings when the card is capable of so much more.So, yes, the 5900 probably still come out on top, but at those res/aa/as settings it's a matter of 10 FPS or less. I got my 9700 PRO and it kicks ass, and even if it's not as good as the 9800 or the 5900, it's cheap and it's definitely up there with the others.ATI - Well done, now get a new card out. ;)nVIDIA - At last, a good card out, but earn your performance, get your act together, don't cheat. Cheating really doesn't go down well with me.

GFX 5900 Caught Cheating

Stating facts about Nvidia tells people you(DRDEATH) are nothing more than a FANBOY. You got something against Nvidia. Every graphic card you post, you always praise ATI & trash Nvidia. Admit it DRDEATH, you are nothing but a troll. You hate Nvidia so much that you have to start providing prove stating ATI is honest & Nvidia is a dirty cheater. You should be ban here for trash talking. No one need your trash talking opinion.

GFX 5900 Caught Cheating

HZcore. Nvidia did cheat. Plain and simple. Are you that f**king stupid not to see it? Even after FM released that bullshit turn around statement, they still hint that Nvidia still cheated. And even Megagames just added a new post about this but as the f**king norm, they stay neutral on the issue and play it safe. The truth be f**king told. Nvidia cheated. I just saw the link below. Yup. Quality has been f**king changed. See for your f**king selves Nvidia fanboys. It's not the same. iq has changed so it is no longer a so called optimization. THE f**kING END RESULT WAS DIFFERENT. The same way carmack mentioned when he gave his half assed explanantion. Are you Nvidiots that blind? Wether drdeath lists facts about it or other people do, it's all true. Nvidia has done this to themselves. No one else. They cheated. Got caught. And in a f**king baby like fashion, are trying to blame others for the apparnt cheat. And i will say this for drdeath, DeQuosaek, Lex, and some others. If they have something against Nvidia for cheating and lying about it and all the bullshit they have forced us to eat over the last 8-10 months, so do I. And many others also i bet. I hate f**king companies that lie and decieve the public to push their product. And then pull a f**king Bill Clinton or Richard Nixon move and bullshit it. THE f**kING PROOF IS THERE. And that is what Nvidia has done this time. Only stupid Nvidia asshole fanboys can say otherwise. They have lied to you dumbasses and you just eat up everything they say. "it's a bug in the drivers...""It's not a good test"For all you assholes that doubt this as cheating. Ask Nvidia why they rendered the directx9 test out of spec? When they f**king very well knew their card runs 32 floating point slower, so they ran it at 12 or 16 to speed it up. A no no for directx9 test. 24bit precsion is the min. requirement. CHEAT RIGHT THERE. How the f**k can they pull that off and tell everyone it's ok to do that. Ask Microsoft about that. lol. Se

GFX 5900 Caught Cheating

to anon below,how many oem companies actually use video cards these days, even if they did, to get the best out of any board you have to pair it with the brand of ram and motherboard taht works best with that card,for example msi makes their video cards to be optimized with their motherboards and vice versa, its the same ol story with all these companies, so they make optimazations to make their stuff better than everything else but in reality they are just making driver optimazations.Futuremark Co. have obviously made a patch to even things out with the ati, the true test lies in game benchmarks with the 5900 it scored much better with some games (mainly the opengl based) over the ati and it was a very little margin between directx

Pages

Add new comment