Futuremark has released a new patch and...

[nRollo]

Rollo

Nvidia fanboy

Cesar- last year I bought a Radeon 9700Pro and a Radeon 8500 at my local Best Buy, the 8500 at the very beginning of the year, the 9700 at the very end. The year before that I bought a Radeon VIVO and a Radeon 32DDR at my local Best Buy. Before that I bought a Radeon MAXX, and a Rage 32.

I've given well over $1000 of my money to ATI in the last couple years. How about you chief? How much have you supported ATI that you call me a nVidia fanboy?

While buying and using a company's cards doesn't make you not a fan of another company, it would sort of indicate you're not exactly against the other company, wouldn't you say?

 

[Rogozhin]

Rollo

Admit that a large majority of joe six packs buy their video cards based soley on 3dmark merit -score- (just look at the mainstream computer periodicals).

I know you say "fook them, their stupid ' but it's simply bad business.

You will have to think of an equivalent analogy for yourself.

Here's mine

Our coffee bar sells the highest grade arabicas but as we need more capital (like nvidia seems to need) we decide to mix in 1/3 robusta into every batch (and only hardcore taste testers will ever be the wiser, certainly not your average joe coffee drinker) and sell it as "high grade arabica."

Not excatly the same but pretty damn close.


It's advertising a product that ISN'T what it is claimed to be-this is FRAUD.

I wouldn't do it to my customers and nvidia shouldn't do it to theirs.

Rogo

 

[Rogozhin]

You should brush up on your ethics to rollo.

rogo

 

[RobsTV]

I did not think nvidia cheated, or if they did cheat, I thought that only indicates how great their driver techs are.
Now I will eat crow.

Still prefer my nvidia cards over my ATI card, even though they did cheat to get higher scores in that benchmark.

I remember a few years back, when 3dfx was still around, general thought was that 3dm had only one useful purpose,
and that was to see how your video card compared to other identical systems running your video card. It was generally
accepted that you could NOT compare systems with different video cards. This was because 3dm was not a true indicator
of real world gaming performance. Why has this all of a sudden changed now? 3dm still only serves one useful purpose,
and that is to see or confirm that your system is running as good as others with identical systems. Of course if I had the
fastest card in 3dm, I would also mention that every chance I could, and boast about how it beats other cards in 3dm.

EDIT:
nvidia cheats gained 24%. ATI cheats (if they were cheats) maybe gained 8% in one test.
How on earth does that make ATI any better?????????
Because they did not have the driver techs good enough to modify the drivers to beat the other in cheating?
Own an ATI card, and this driver tech issue is no surprise.
You don't think if ATI had qualified driver techs they would not have done the same thing?

 

[nRollo]

It's advertising a product that ISN'T what it is claimed to be-this is FRAUD.

So nVidia is somehow different than ATI, who gave drivers with a Q3 cheat in them out with the preview copies of their 8500s? That's not FRAUD?
Or nVidia is somehow different than ATI on the 3mark think because they cheated on all the games instead of just one? People like to point to the 1.9% difference overall for ATI, they don't seem as happy to admit that it comes from an over 8% difference on Game 4, and that ATI has already agreed to remove their "optimization".

My position is what it's always been:
1. Both companies have been caught cheating on drivers, so you can only look at the ones they haven't been caught cheating on.
2. I don't care about 3dmark, don't play it.
3. Company "ethics" are not the reason I buy VGAs of one sort or another. I don't have the time or means to research a company's ethics. (e.g. How have they treated the employees? What companies do they have full/partial ownership of, and how ethical are they? Have their factories negatively impacted the environment? Have their financial dealings adversely affected stock holders? etc ad infinitum)
4. When someone shows me they cheated on a game benchmark I play, which I'm sure many are trying to do, I'll pay attention.

This whole thing is a mountain out of a molehill. What if everyone bought 5900 Ultras due to inflated 3dmark scores? How would they lose out? At very WORST case, they'd have the second fastest VGA, and a little less AA IQ. They wouldn't be injured in any way, they'd suffer no inconvenience.

You need to stop drowning in minutia Rogo.

 

[stonecold3169]

Originally posted by: RobsTV
I did not think nvidia cheated, or if they did cheat, I thought that only indicates how great their driver techs are.
Now I will eat crow.

Still prefer my nvidia cards over my ATI card, even though they did cheat to get higher scores in that benchmark.

I remember a few years back, when 3dfx was still around, general thought was that 3dm had only one useful purpose,
and that was to see how your video card compared to other identical systems running your video card. It was generally
accepted that you could NOT compare systems with different video cards. This was because 3dm was not a true indicator
of real world gaming performance. Why has this all of a sudden changed now? 3dm still only serves one useful purpose,
and that is to see or confirm that your system is running as good as others with identical systems. Of course if I had the
fastest card in 3dm, I would also mention that every chance I could, and boast about how it beats other cards in 3dm.

EDIT:
nvidia cheats gained 24%. ATI cheats (if they were cheats) maybe gained 8% in one test.
How on earth does that make ATI any better?????????
Because they did not have the driver techs good enough to modify the drivers to beat the other in cheating?
Own an ATI card, and this driver tech issue is no surprise.
You don't think if ATI had qualified driver techs they would not have done the same thing?

You sir, are misinformed. The technique nvidia used did not involve "qualified driver techs". For the most part they made it so that when certain calls were made to the card, under said program, it ignored them. That takes no talent at all. Rendering things in high speed with a complicated piece of hardware, like normal drivers for videocard do, takes talent. Setting up if loops and catch statements to ignore other commands is overly simple. Not only did nvidia cheat, they used the "qualified tech" of a freshman computer science major. Go Nvidia!

 

[b4u]

Originally posted by: gururu
HardOCP regarding 3dmark fix:

To us, things like this just solidify our belief that 3DMark03 overall score is useless as a real world benchmark.

are these guys stupid? futuremark is trying to improve their bench and hardocp is slamming them harder.
and this is after coming to the defense of nvidia's 'driver optimisations'. I hate it when sites lose sense of
objectivity.

Check this out DOOM 3 Benchmarks, read the first lines ... well, I'll post them here anyway

Posted by: HardOCP

Many Thanks

Let me start by saying, "Thank you!" to id Software and NVIDIA for making this article possible. It is truly an honor to be included in an opportunity of this magnitude. John Carmack and Tim Willits of id Software went the extra mile this last Friday to help us make sure we could show this to you today. Also, Jim Black of NVIDIA extended himself above and beyond the call of duty in order to make this all possible.

Reading this, I do believe they're on nVidia side ... come on ... any of us would if we where to see DOOM3 mooooving (joke!)

 

[b4u]

"Since NVIDIA is not part in 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. This is obvious since our relative performance on games like Unreal Tournament 2003 and Doom 3 shows that the GeForce FX 5900 Ultra is by far the fastest graphics on the market today."

- from HotHardware


Comment:

They don't belong to the beta program, that's why they didn't received the development version, which would allow them to change camera view point and check the errors from the cheats


I really like nvidia, and I'm quite scheptical now for future product performances ... of course optimizations can be done in benchmarking, that's why I prefer the experience of you folks who have tried the product

Shame on Nvidia! ... Well, not really ... it just shows that there are no benchmarks as safe as running the games

 

[SilverTrine]

You people trying to excuse Nvidia are hilarious, some glaring flaws in your arguments.

3dmark2001 was so based in unreality that they made Max Payne off the 3dmark engine! There goes the entire argument that 3dmark isnt based on reality. Its just as much reality as any other rail benchmark like the Ut2k3 flyby.

Nvidia was in the 3dmark program right up until it was released in December, they quit when they knew Futuremark was going to use Dx9 techniques and not their proprietary Cg crap. Can anyone say sour grapes? I can SOUR GRAPES.

Also 3dmark2003 is such a worthless benchmark Nvidia decided to spend dozens of developer hours making cheats for it. Its so irrelevant we spend tens of thousands of dollars on wages monkeying with it!

Basically you boys need to get another schtick, you cant fight reality(well u can only if you're stupid). But I must say thanks to Nvidia their whopping 24% cheat made the decision to buy a 9800pro much easier, when down the road its proven that the 9800pro is better in Dx9 games I expect to see you guys here to eat your crow about 3dmark2003.

 

[Cesar]

Originally posted by: Rollo

Rollo Nvidia fanboy

Cesar- last year I bought a Radeon 9700Pro and a Radeon 8500 at my local Best Buy, the 8500 at the very beginning of the year, the 9700 at the very end. The year before that I bought a Radeon VIVO and a Radeon 32DDR at my local Best Buy. Before that I bought a Radeon MAXX, and a Rage 32. I've given well over $1000 of my money to ATI in the last couple years. How about you chief? How much have you supported ATI that you call me a nVidia fanboy? While buying and using a company's cards doesn't make you not a fan of another company, it would sort of indicate you're not exactly against the other company, wouldn't you say?

Rollo you are a fanboy stop foolling around

 

[nRollo]

Rollo you are a fanboy stop foolling around

I noticed you didn't answer my questions though did you Cesar? I'll guarantee you this:
If you haven't spent as much money as I have on ATI products, ATI wants "nVidia fanboys" like me for customers a LOT more than they want your misguided loyalty.

What you guys don't seem to understand is that these people are not "friends" you're defending. They're big corporations whose interest in you goes exactly as far as the moment they get the money out of your wallets, at which point they could care less about you until it's time to ask for money again.

I think a lot of the ATI loyalty is based on the old 3dfx wars in which the 3dfx faithful were left saying "Wait-you'll see" until nVidia swallowed 3dfx whole and made everything the faithful had posted a lie. I think a lot of those people resent nVidia and point to things like that ridiculous internal sales memo or the ridiculous 3dmark driver "cheats" as a reason to not buy nVidia.

If you think the 3dmark thing is a big deal, I urge you not to buy nVidia. Me, I laugh at the 3mark thing because there aren't any DX9 games, and by the time there are, I'll have a different card anyway. I live in teh present, and have no loyalty beyond what runs the games I play fastest at the settings I want to use. Last October till now, that was ATI. Looks to me like it will be nVidia in June. When I want to upgrade again, I'll buy whoever is fastest, even if it's Jaton.

 

[SilverTrine]

If you were truly in the now you would realize purposefully falsifying any benchmark by 24% is pretty damn significant. And not signifcant in a good way.
Personally I could not spend a huge wad of cash on an Nvidia product until I saw a full suite of benchmarks that I knew contained no cheating.

Since sites like Anandtech havent even addressed the issue yet it leaves people like me who were about to buy a high end card in the lurch. Great job Anandtech.

 

[gururu]

a nice summary of nvidia's current tactics.

ETC.

did any of you see these articles:
DOH!!
DOH!

management at nvidia seem to be using the Enron guide for partying.

 

[Creig]

Rollo, (and all other NVidiots), slow down and listen up.

If Nvidia thinks the 5900 is the fastest card currently available, then WHY CHEAT IN THE FIRST PLACE? You only cheat if you HAVE SOMETHING TO HIDE!

And no, I'm not a dyed-in-the-wool ATI fan. I have two GF256, a GF2 and a GF3 Ti200. My modded 9500 is the first ATI video card I've owned. I'll buy the best bang for my buck when it comes time for my next card, too.

But if Nvidia will cheat on benchmark programs to make their card look faster than the nearest competitor, what's to stop them from doing it in games to keep themselves on top. I'm sorry, but I don't want my video card skipping frames on my games just so that its creator can say "We're the fastest!". I want to see EVERYTHING! Nvidia has been putting a lot of emphasis on Doom3. It would be pretty embarrassing to them if it turns out the 9800 is faster in Doom3 than 5900. If this happens I can see their "qualified driver techs" going back to work to "fix" the drivers for another 24% speed increase. At the expense of the game itself.

 

[SilverTrine]

Nvidia's argument would be this: 3dmark2003 is a flawed benchmark but its used by companies to determine which cards to buy and put in thier systems, Nvidia decided to cheat and justified their cheating as necessary because paranoidly they decided 3dmark2003 is biased against them.

To see how justifiable this go I suggest all College Students on this board blatantly cheat their asses off in a test, and when the Professor kicks you out of school tell the Professor you only cheated to point out how unfair and biased his test is. hahaha.

 

[nRollo]

Trine:

If you were truly in the now you would realize purposefully falsifying any benchmark by 24% is pretty damn significant.

Really? I use my VGA to run office apps, surf the net, and play UT2003 online mainly. I haven't downloaded 3dmark 2003. So just how is nVidia cheating on 3dmark 2003 significant to me Trine? I don't care if my VGA will run it at all, and never will. Why should I care? How am I damaged?

Creig:

If Nvidia thinks the 5900 is the fastest card currently available, then WHY CHEAT IN THE FIRST PLACE? You only cheat if you HAVE SOMETHING TO HIDE!

LOL, really? Do you think it's possible they're the fastest at games currently available, but not at DX9, so they cheat on it because some people (foolishly) care about their 3dmark score? D'oh.

"But if Nvidia will cheat on benchmark programs to make their card look faster than the nearest competitor, what's to stop them from doing it in games to keep themselves on top"
Who will save us?! LOL I guess "what's to stop them" is the fact that drivers and benchmarking are obviously scrutinized for such things because there's lots of money at stake? Why don't you point us at a site that shows some proof of nVidia cheating on game benchmarking, instead of saying, "they're probably cheating on this benchmark because they cheated on the other" which is nonsense.

It would be pretty embarrassing to them if it turns out the 9800 is faster in Doom3 than 5900. If this happens I can see their "qualified driver techs" going back to work to "fix" the drivers for another 24% speed increase.

ROFLMFAO
Of course, you're totally talking out of your a$$, because you have NO WAY whatsoever to predict what nVidia will do in the future?????

 

[Rogozhin]

Rollo

All one needs to counter this proposition-

"Really? I use my VGA to run office apps, surf the net, and play UT2003 online mainly. I haven't downloaded 3dmark 2003. So just how is nVidia cheating on 3dmark 2003 significant to me Trine? I don't care if my VGA will run it at all, and never will. Why should I care? How am I damaged?"

-is ONE person that DOES use 3dmark to bench their card and system.

This whole argument is based on ethics and since you proclaim that ethics doesn't matter you're falling into the abyss of nihilism. The vga tech world does not revolve around your perception of the relevance of benchmarks, it is a huge operation that relies upon a standard where websites and periodicals can trust that benchmarks are done fairly and actually show the real performance of a card.

If you believe that these standards are irrelevant and don't apply then you might as well let nividia or ati chose what software you should run and at what settings you need to run it.

My opinion here is that this could be the trigger that sets off a landslide of vga hardware companies being able to influence consumer perception by manipulating standardized benchmarking programs.

With most other goods and services there is a standard of performance that has to be passed in able to enter the world of public consumption-futuremark has ostensibly filled this role within the gpu hardware community-even though you might not like it.

I'm tired and need to go roast some "robusta" beans so as to "shaft" my customers that expect me to give them some of the best high grade arabica beans in my region (what nvidia is doing).

good night.

rogo

 

[Shamrock]

Why do you guys continue to bash NVidia, when ATI came forth and admitted they cheated too? as for Ethics, neither has morals now, right? It doesnt matter if it's 24% or 8%, or 1.9% it's still cheating.

Go steal a Yugo, and then go Steal a Mercedes...chances are you'll get put in prison for Grand Theft Auto either way.

both cheated...question is, did ATI cheat on anything as you guys claim NV did? hmm...

 

[RobsTV]

Exam Time.
A good student sees that a poor student has access to the "special insider" exam books, because he paid extra.
The good student also sees the bad student glancing around at other good students, and sees cheating, which is common.

So, the good student, thinking this is not right, and because he is such a good smart student,
also cheats a little on the exam to get that edge. He knows he has always produced the best
before, has the best background training, and to let someone lesser take top spot because of
a "special advantage" that money bought the lesser student, was not right, and not going to happen.

Results come back, and even though the bad student cheated, he was still a bad student, and even though
he had "special insider help", was only able to gain 2% in the final score. The good student , that had no
insider help, and using his brainy brawn, scorched the other cheater by 24%.

Now the exam giver also happens to be the same guy that the poor student paid off, and he too
was shocked that even with his help, the poorer student couldn't beat the smarter student. After
investigating, exam giver finally saw that they smarter student was too smart for him, and even without
the insider book, the smarter student was able to see flaws in the exam. Poorer student was outraged,
and the exam giver was also annoyed by not getting extra cash from the smart student, so exam giver,
said it would tell the world that the smarter student cheated, and was not smart at all.


Hey, it could have happened that way.

 

[Creig]

Well Rollo, obviously you are simply enjoying your 15 seconds of fame in this thread and will continue to refute any and all arguments about Nvidia cheating on the benchmark just to keep it going.

 

[ginfest]

Why do you guys continue to bash NVidia, when ATI came forth and admitted they cheated too? as for Ethics, neither has morals now, right? It doesnt matter if it's 24% or 8%, or 1.9% it's still cheating.

You won't see this addressed. Or if you do it will be a "qualified answer".
For me, worse than the cheating is the concerted campaign started by a couple guys at Rage and spread thruout the web like wildfire.
For instance check out the sign-up dates of the most fervent ATI supporters.
Guys like Rogoa(zhin or din?)who altered his Rage3D name, comes here and pretends to be "helping" people while actually pushing the ATI party line while downing NV.
Funny, his "facts" seem to lose thier weight when a guy like Ben S answers with real 3D knowledge that can't be refuted.
A read a lot of HW forums and newsgroups every day and it just never ends.

I was thinking of going with a 9700 or 9800 but when I go to Rage to do some research I see so many NV bashing threads by the "bigshots" but when people genuinely have a problem and ask for help these same "know it alls" either don't reply or offer the standard "reformat, bad mainboard, memory, power supply, monitor, lousy OS, lousy DX, lousy game coding" and on and on. It's never the card or drivers? Doesn't make any sense!
What they don't realize is that most people can make thier own informed decisions and are turned off by the constant bashing and twisting of facts.
I for one will not buy ATI not because the hardware is no good (it is in fact kicka**) but because IMHO thier drivers stink and go from bad to worse.
My $$$, my decision and no amount of FUD will change my experience.

Mike G

 

[WetWilly]

Sorry, RobsTV - the analogy is WAY off. Here's the corrected one:

Two good students have access to the "special insider" exam review books. One of the students gets pissed because even though he has access to the "special insider" exam review books, he knows he still can't score higher than the other students with special access, and that's unacceptable to him. So he starts bashing the exam as unfair and biased, starts rumors that the Professor likes the other students better than him, and gets some of his ass-kissing [ H ]angers-on to join in.

But despite all his ranting, Pissed Off Student STILL has to take the exam - there's no way around it. Since he's determined that the "special insider" knowledge is useless to him, he rather publicly dumps his "special insider" access privileges and does the only thing he can do - cheat. The results are posted and Pissed Off Student has scored WAY above everyone else. Two students who had "special insider" access but didn't take the exam think something doesn't look right because several of Pissed Off Student's results are blatently incorrect but he got full credit. So they report their observations to the Professor. The Prof reviews the results and totally busts Pissed Off Student, documenting multiple instances of cheating. Pissed Off Student's only response is "I don't know why, but the Professor is trying to make me look bad." The Prof also finds evidence that the second good student MAY have cheated. The second good student immediately 'fesses up and states that he did things a different way than the Prof said to do them, but he - unlike Pissed Off Student - gets the exact same result as the Prof. These results are the talk of the campus.

Two highly respected experts, Carmack and Sweeney, review the Prof's evidence. They state unequivocably that Pissed Off Student cheated and it's totally unacceptable. They also state the second good student's "cheat" is NOT unequivocably a cheat but may be an acceptable optimization depending on the context. Unlike Pissed Off Student, the other good student at least got the results the Prof expected.

That IS the way it happened.

 

[DAPUNISHER]

Originally posted by: Shamrock

Go steal a Yugo, and then go Steal a Mercedes...chances are you'll get put in prison for Grand Theft Auto either way.

Naw, you could get off on the Yugo charges by pleading temporary insanity

 

[Rogozhin]

Ginfest

If you want to question my motives then do it over a PM.

As it stands I don't have to justify my posts to you or anyone else.

And as I've stated in another thread I am ROGOZHN at anandtech.com and nvnews.net but ROGODIN over at rage3d and IL-2sturmovik.

I've not tried to contend with ben's amount of vga knowledge but the little I do know I try to use efficaciously.

Rogo

 

[Rogozhin]

PS

ati has said they cheated and so they did. my arguments with rollo have been about the problems with cheating (broadly) across all vga makers, not either~or.

Rogo