i notice this too both on my 8800 cards and my 2900 xt. its almost makes it unplayable for me.
![[H]ard|Forum](/styles/hardforum/xenforo/logo_dark.png){kind=logo}
Nvidia caught cheating; latest drivers 169.xx cheat in Crysis demo
- Thread starter LawGiver
- Start date
i notice this too both on my 8800 cards and my 2900 xt. its almost makes it unplayable for me.
phide said:
I have heard from enough people to know that this issue isn't isolated to Hanners. I have faith that Hanners "does his homework" before posting any article (especially a high profile one), something certain websites have been failing to do as of late. It just upsets me when a good man has his name unneccessarily dragged through the dirt.
Ya, so it wasnt you who originally posted it, but you attempted to correct me, and you were wrong. I was never contesting this cheating, I was simply correcting the guy stating that the cheat was some FoV cheat which simply isn't true, as all hardware does this. You posting links to the articles was semi-pointless, since that's not what was being debated.
I've heard from numerous people who have confirmed this issue with various ForceWare 169.0x versions, although admittedly it's far more apparent in the fly-by benchmark than during normal gameplay. It isn't just me though, that's one of the first things I checked.
But Hanners, what you did was really irresponsible.. you logically linked the "definitive performance difference" (which is contestable and within a margin of error anyway) to this corruption you saw. Just because you see a fix to corruption and a slight framerate drop when you change the .exe name doesn't mean the corruption is what is buying the performance.
It's a logical fallacy to assume correlation = causation. But your article made that logical link, incorrectly and assumptively.
There are TONS of titles that come out with ATI having massive corruption in it. Does anyone cry foul and claim they're cheating? No... it's just clear that they don't have every bug nailed before the game/demo is released.
- Joined
- May 18, 1997
- Messages
- 55,634
I think doing homework would have included talking to NVIDIA about this since it is their driver, their part, and a TWIWMTBP game. Instead he only talked to AMD about it. He talks about treating NVIDIA fairly in the conclusion, but I don't think that is what happened.
OK, lets mince words. Does anyone assume AMD is hacking up the driver for some "optimization" and getting too aggressive with it when they have corruption in a game?
Happy?
King_Weaver
Gawd
- Joined
- Nov 26, 2006
- Messages
- 620
what...again...quite spitting out nonsense
Brent_Justice
Moderator
- Joined
- Apr 17, 2000
- Messages
- 17,755
He made a connection that this "bug" was what was causing the performance difference. Changing the EXE name might be disabling other optimizations, and this could be the reason for a 1-3 FPS perf difference (or it could just be in the margin of error.)
I'm not sure if that changes anything. All he said was that the optimization might not be working as Nvidia intended.
Which I did refer to in the article, and it's a fair point to some extent, although its equally fair to at least suggest the change in rendering workload that this bug introduces is an obvious culprit for the performance change.
More importantly, it's useful for gamers to be aware of issues which may negatively impact their gaming experience, and how to work around them - Personally, I see it as an important part of our review work, and I always try to point out driver issues in reviews whenever I spot them. Simply brushing them under the proverbial carpet helps neither the graphics IHVs nor consumers.
If I was running benchmarks on a graphics board with a 7% margin of error between runs, I'd start getting worried in all honesty. We should also remember here that the in-built Crysis benchmark performs three runs of the timedemo (four if you're being picky, but the first run is discarded as it's used to 'cache' all the necessary data), producing a result for each one. There was not a 7% margin of error between each test run at the same settings, I can say that for certain.
Also, I'll mention again that I received no e-mail communication regarding this issue from ATI/AMD whatsoever. Hopefully this will clear up Kyle's obvious confusion about this point, as I'm sure such a respected web site owner wouldn't want to damage the reputation of a fellow site erroneously.
- Joined
- May 18, 1997
- Messages
- 55,634
Hanners discussed the NVIDIA driver issue at length with AMD employees but not NVIDIA employees.
*cough*
Well, I guess that pretty much squares things away. Out of curiosity though, who did you hear about this from?
Ferrariman60
n00b
- Joined
- Mar 31, 2006
- Messages
- 63
Well, for me, the important thing they fixed with 169.04 was they made my 8800 GT work. When I tried running it with the 169.01, or even the drivers they submitted to WHQL, it would freeze at the "Don't copy this demo or we'll be pissed" screen. I grabbed the 169.04 drivers, no problem, and I'm not complaining about water update rates. There's enough other beauty in the game to look at, I say don't whine about little shit like this. Just be frickin happy the game runs and is finally coming out after all the delays.
ChrisRay Confrimed it's fixed in the next driver, this is a bug not a cheat no matter how much ATI wants it to be.
- Joined
- May 18, 1997
- Messages
- 55,634
So Hanners is "fair" to NVIDIA when he does not discuss the issue with them before publication? And I am unfair for pointing out that this story was delivered to him from AMD? AMD came to HardOCP with this story as well last week....
As far as EB being "stand up guys," I really don't see it that way especially with this story.
Hmmm, maybe Hanners will post here since he is active in this thread and state that he has not worked with AMD on this particular issue? I would like to see that happen.
- Joined
- May 18, 1997
- Messages
- 55,634
Sorry I missed this post....
AMD called me, they did not communicate via e-mail on this to me. I find it very interesting that AMD wanted me to do a story on this, and you just so-happened to have the same exact story up a couple days later. And I have heard specifically that the story was delivered to you. Maybe my source is wrong. Maybe he could clear it up?
But I am not "confused" about what was related to me on this issue. I am not making guesses about it, or simply coming up with some crazy ideas. The story is sloppy at best and I think you know that. That is my take on it along with my opinion. I figure you would not be posting in our forum if you did not want to hear my thoughts.
AMD called me, they did not communicate via e-mail on this to me. I find it very interesting that AMD wanted me to do a story on this, and you just so-happened to have the same exact story up a couple days later. And I have heard specifically that the story was delivered to you. Maybe my source is wrong. Maybe he could clear it up?
But I am not "confused" about what was related to me on this issue. I am not making guesses about it, or simply coming up with some crazy ideas. The story is sloppy at best and I think you know that. That is my take on it along with my opinion. I figure you would not be posting in our forum if you did not want to hear my thoughts.
Oh, so the post right above mine in which Kyle beat me to linking the same article was pointless as well? It's what the poster was referring to, wether or not he used the exact correct terminology.
We all knew what he meant and that is what the article linked dicussed.
- Joined
- May 18, 1997
- Messages
- 55,634
It was an accusation, but it was far from baseless. Anyway, glad to see you directly reply.
digitalwanderer
Long Time No See
- Joined
- Aug 21, 2004
- Messages
- 141
Back it up Kyle, on what do you base your accusation? This un-named "source" of yours? :|
SEALTeamSix
2[H]4U
- Joined
- Aug 15, 2003
- Messages
- 2,434
Quoted for Justice... Brent Justice! (I've been waiting a while for that setup)
Seriously, though... the "story" was leaked by the competition, and is retarded considering both the drivers and game are pre-release. Kudos to the [H] for reserving judgment.
Are you saying that demos shouldn't be used for benchmarking. This particular one is used quite often. Maybe [H] should remove the Crysis results from its 8800gt review. as the game and the drivers are both pre-release.
Why is everyone getting their panties in a bunch? This is a DEMO! Not the full release, something YOU HAVE TO PAY FOR! So isn't this just a little tad pointless?
If it was based on the full blown release of Crysis, then I can see some stink about it. This article is getting out of hand. If the retail version comes out and everything is fine and dandy, then this article and all this chat is pointless...
We all know that ATI/NVIDIA will be releasing drivers for this game. Can't we just fricken wait until the game is released in what? a week and a half the most? Sheeesh!
If it was based on the full blown release of Crysis, then I can see some stink about it. This article is getting out of hand. If the retail version comes out and everything is fine and dandy, then this article and all this chat is pointless...
We all know that ATI/NVIDIA will be releasing drivers for this game. Can't we just fricken wait until the game is released in what? a week and a half the most? Sheeesh!
I think what's Seal's trying to reaffirm is that some might have the feeling that this story was premature. The demo is pre-release; the drivers are beta. Ergo, bad shit is destined to happen. That being said, the article itself isn't charged either way, as Moofasa's been hammering, and it's really just a performance report/bug report, though some of the conclusions are a little iffy, and the circumstances are a bit...coincidental.
I don't think there's anything wrong with benchmarking with the demo per se, though it makes more sense to wait until the retail version is out.
I wonder if anyone would have gotten their knickers in a twist had the thread title not been so unbelievably misleading and, well, just plain stupid. The article says something akin to "the Earth is 1F hotter", yet the thread title is akin to "MAN-MADE GLOBAL WARMING CONFIRMED -- SEE SUCH-AND-SUCH ARTICLE".
SEALTeamSix
2[H]4U
- Joined
- Aug 15, 2003
- Messages
- 2,434
And Kyle jumped all over the 5800 when it was released too, and the 9800pro still stomped all over it... If I recall correctly, at release time the 2900xt under-performed compared to its competition, cost more, used more power, yadda yadda... the review seemed justified to me.
But, this is all secondary - the fact is the card was an official release. In contrast, these drivers and this game are not.
And for the record, I think canned timedemos should NOT be used for benchmarking... actual in-game experience should be the standard (and has been at the [H] for a long time).
THIS article isn't getting out of hand. It is this thread and the pointless mudslinging that is over the top. The article is pretty neutral, lest you believe the 2 guys that don't think so in this thread.
SEALTeamSix
2[H]4U
- Joined
- Aug 15, 2003
- Messages
- 2,434
Indeed.
Absolutely agreed. I have nothing more to say here, I'd like to think that everybody here knows my thoughts on the entire situation, as well as the series of events which led to it.
- Joined
- May 18, 1997
- Messages
- 55,634
And we have done stories on pre-release / beta games and drivers in the past, but we did not do 3 page articles with tons of benchmarks that basically pointed out Team Green or Team Red doing something underhanded. It is a waste of time. We decided to spend our time seeing how Gears of War for the PC was doing.
That said, we used Crysis and UT3 to overall evaluate the 8800 GT because I think that is what our readers care about. But spending our time and focus on something so minuscule that implies cheating is not a good way for us to spend our resources. (Would have been great to see them try to pull off something of the magnitude of Quake/Quack then we could have unloaded both barrels!) Taking 3 hours and throwing together some benchmark numbers does not prove much anything now days, at least that is what I think. Playing around with the Crysis internal benchmarks has so far left me unimpressed as well but we will keep using them on the CPU side to see if we see processor scaling with their physics engine, but I am not sure we will.
- Joined
- May 18, 1997
- Messages
- 55,634
I can't tell you how many times over the years I have seen your comments about our content and thought the same thing....guess its Karma.
Bottom line is your talents could have been at a much better use elsewhere IMO.
I never stated the article isn't neutral, I said it's a little premature to raise the question when the release of the game is days away. It's not like we are waiting months for the game to arrive. It's what? 2 weeks before the launch of the game the demo appeared?
As a gamer, I'm smart to understand that this is a demo and not the full version. If their are glitches, I can accept that and wait to see if the glitches are in the final version. When the game is available and the drivers that are available are producing these image screwup's then I'll be up in arms. This article just tells me that the demo and the drivers out now for the demo are not a match made in heaven. I can deal with that because it's a demo. If the problem occurs in the retail version then get back to me...
ATIIsBetter
[H]ard|Gawd
- Joined
- Feb 19, 2005
- Messages
- 1,202
Kyle and Hanners are gonna start making out if this type of argument keeps going between them.
Solid Gears of War article though, I hadn't read it until it was mentioned here.
Solid Gears of War article though, I hadn't read it until it was mentioned here.
It's a "bug" now that it has been exposed and rectified, isn't it? If no one noticed and no one complained, would it be a cheat?
It doesn't really matter. Those of us with top hardware still aren't going to get high framerates with this game .
We've both gotta bring in the hits to our respective sites on Valentine's Day somehow...
Okay, that really is my closing post before things get way too weird.
- Joined
- May 18, 1997
- Messages
- 55,634
We've both gotta bring in the hits to our respective sites on Valentine's Day somehow...
Okay, that really is my closing post before things get way too weird.
I am bringing the tongue with my sloppy kisses.