ati cheating in crysis benchmark?

[cowie]

Convenient? How is it convenient? A driver glitch that potentially lowers IQ without any evidence of enhancing performance? Sounds more like a PR disaster.

And since when was timing irrefutable proof that something nefarious is going on?

How about we look at this logically and objectively, for a change?

thuo you do have to question the launch date of the drivers or you would be burying your head in the sand.
i'm sure it was make sure you use these drivers type thing for reviewers but that just me with a tinfoil hat on
its more like they pushed more towards performance and have topped out the card so far driver wise,what else could they do?

[freeloader]

I'm getting old. My 40 year plus eyes can't tell the difference in the pictures on the first page of this thread.

Good thing I don't stop to look around at scenery in my games. LOL....

[Serpentarius]

I'm getting old. My 40 year plus eyes can't tell the difference in the pictures on the first page of this thread.

Good thing I don't stop to look around at scenery in my games. LOL....

same here.

to me ... it's 2 different model from 2 different manufacturers, with 2 different specs .... wasn't it supposed to have atleast 2 different results?

[Cybercat]

I dont realy understand your post. I said its convenient that this bug apears right now, i never said that that is proof of ATI doing it on purpose.

Further as i said in my post we stil dont know how much extra perf this bug gives.
Seeing this bug forces less processing it would only be logical to assume that it increases performance.

And I'm saying that any convenience this may provide for ATI is negated with the problems it might cause, particularly in regard to public relations.

And assuming it gives a performance boost is just that, an assumption. Crysis is probably not particularly texture performance bound.

[trinibwoy]

The responses in this thread crack me up.

Yep, hilarious

Also, the substitution of FP16 render targets by ATI has been confirmed in quite a few games. DoW 2, NFS: Shift and a few others are the most obvious candidates.

You referring to substituting in FP10 where the app requested FP16? I thought that little rumour was still under wraps.

[Zayras]

Coldon's posts make the most sense to me, but regardless... if ATI did this, then shame, shame .

I don't see why people want to blame someone here. It seems some ATI "fanboys" are blaming Nvidia for previous mishaps and some Nvidia "fanboys" (and perhaps they are right) are complaining that if this was Nvidia everyone would've gone crazy about it.

I did the name change and found no difference something asked for a few times in this thread.

So until someone provides some real evidence I'm going to call this BS and will trust Eastcoasthandle's response.

[ElSel10]

It's ATI they get a free pass

The sad part is, that is basically true.

[Macadamia]

This is bad if true. Why? nVidia gets a free pass consequently on driver optimizations, and that's considering what y they generated with the 480 (480 only) in terms of TDP/Power they can be very desperate.
If ATI is playing with render targets and texture formats, what would nVidia try? Tinkering with water reflection rate again?

But again, moral high ground on the last 2.5 generations didn't sell ATI cards. Image quality seems decent, nothing of the really big shoddy hacks like GF FX/Quack.exe . That's purely from another perspective.

[RejZoR]

I don't see how that "problem" with water in the link above is cheating. It looks horrible and you can't miss it. All these are just bugs. And if we jump into the renaming war, both AMD and NVIDIA do per-application profiles, in this case they just broken something related to Crysis. Renaming it to "iamnotcheating.exe" solves it because that profile is not linking to the broken profile anymore. Though i'd say NVIDIA actually cheating would be more probable. After all, it's not AMD that's late and desperate, it's NVIDIA at the moment. And as for the current gen of cards, NVIDIA has other problems like high power draw and low availability, not performance. AMD on the other hand has lower prices, lower power usage and high availability. So frame or two less, who really cares?

[Svnth]

This is bad if true. Why? nVidia gets a free pass consequently on driver optimizations, and that's considering what y they generated with the 480 (480 only) in terms of TDP/Power they can be very desperate.
If ATI is playing with render targets and texture formats, what would nVidia try? Tinkering with water reflection rate again?

But again, moral high ground on the last 2.5 generations didn't sell ATI cards. Image quality seems decent, nothing of the really big shoddy hacks like GF FX/Quack.exe . That's purely from another perspective.

That wasn't tinkering. It was a driver bug where an SLI path in the driver was triggering in single GPU mode. It was immediately fixed.

[Svnth]

Coldon's posts make the most sense to me, but regardless... if ATI did this, then shame, shame .

I don't see why people want to blame someone here. It seems some ATI "fanboys" are blaming Nvidia for previous mishaps and some Nvidia "fanboys" (and perhaps they are right) are complaining that if this was Nvidia everyone would've gone crazy about it.



So until someone provides some real evidence I'm going to call this BS and will trust Eastcoasthandle's response.

PCGH is not claiming it's an issue with Cat AI or exe specific.. they simply say that it's happening when you diff in image from a GeForce at 16x AF and an ATI at 16x AF. The banding is happening on the ATI, despite their supposedly slightly better AF (as seen in the directed tests with AF the test with the circles).

The question is why is there this difference between the directed AF tests and the games?

[cowie]

I don't see how that "problem" with water in the link above is cheating. It looks horrible and you can't miss it. All these are just bugs. And if we jump into the renaming war, both AMD and NVIDIA do per-application profiles, in this case they just broken something related to Crysis. Renaming it to "iamnotcheating.exe" solves it because that profile is not linking to the broken profile anymore. Though i'd say NVIDIA actually cheating would be more probable. After all, it's not AMD that's late and desperate, it's NVIDIA at the moment. And as for the current gen of cards, NVIDIA has other problems like high power draw and low availability, not performance. AMD on the other hand has lower prices, lower power usage and high availability. So frame or two less, who really cares?

oh come now its nvidia fault and there the ones that are cheating omfg i heard it all now

[Decami]

lol at you guys not believing it cause its ATI, how about looking up the history of both companies, both have been caught in many occasions cheating in all sorts of ways, definitely including ATI, infact they started a big cheat war way long ago.

[mrcape]

I don't have a brand preference, but I do notice a ton of ATI lovin in this xtreme oven. I think most fanboyism stems from the fact that a lot of people tend to love what they just bought or want, and ATI cards were standing on top of the mountain for the past six months.

[Solus Corvus]

lol at you guys not believing it cause its ATI, how about looking up the history of both companies, both have been caught in many occasions cheating in all sorts of ways, definitely including ATI, infact they started a big cheat war way long ago.

It's great when the shoe is on the other foot isn't it?

People are always trying to blame these two companies for cheating. When it may have just have been a minor side effect of a driver. I remember not too long ago someone was blaming Nvidia for cheating because their driver release had "ghost" images inserted into video content to enhance quality, when this is actually a natural use of picture enhancement.

Both companies have done everything in their power to get ahead. So when it comes to this kind of stuff, I could care less. They both cheat, not really anymore though, they dont really need to, both companies are pretty solid atm. Anything nowadays is pretty much an un-expected side effect or looked at the wrong way. ATI has some catching up to do, for shure, though.

If Nvidia says its a bug, I would believe them right now. They obviously have a good distance lead to ATI, atm. Why would they purposely cheat? for fun? These guys are people like us, they make mistakes. Somtimes things (even big) are hugely looked over. Thats what makes these people human, and thats also what betas are for..

It's a bug plain and simple...

These kind of things commonly occur with both companies, it's just a matter of if a site takes the time to complain about it before it's fixed or not.

As for "cheats" both companies were VERY guilty of it during the FX/6800 days. For the most part those days are gone, and now all that remains is drivers bugs generally.

As for image quality by both companies, on a technical level NVidia's is better, that much is a fact. Which you prefer is a matter of personal opinion, but going by paper facts nvidia has better IQ this gen.

QFT LOL something I was trying to say a page ago LOL. Cant stop the fanboys!

Which is why I stick with Nvidia usually. See us Nvidia buyers are pretty nit picky. That way we keep Nvidia on their toes. Which in turn forces them to work harder and release better. Maybe the ATI guys should go on strike and hold signs in front an AMD building they havent sold yet! LOL jkjk

But anyway i do agree with you, too many people are too quick to call cheats on things!

I agree with you, the old you, that it is most likely a bug...in both cases.

[RejZoR]

oh come now its nvidia fault and there the ones that are cheating omfg i heard it all now

What if you'd ACTUALLY READ, you'd see i never said that... but since ppl never read stuff properly...

[GoldenTiger]

The responses in this thread crack me up. If the roles were reversed, I am sure the responses would be FAR different. Just shows how long so many here have been smoking the good stuff.

Also, the substitution of FP16 render targets by ATI has been confirmed in quite a few games. DoW 2, NFS: Shift and a few others are the most obvious candidates.

Well said... I too love how because it's ATI no one cares, but if nVidia did this the SAME people would be grilling them and flaming nV to cinders.

I think you may need eye glasses.

Actually I noticed poor AF with noticeable lines between texture filtering areas in some games I played as well... I chalked it up to just being an ATI/nV difference when I had my 5870, now I guess we know they were cheating to gain some extra performance, eh?

I don't have a brand preference, but I do notice a ton of ATI lovin in this xtreme oven. I think most fanboyism stems from the fact that a lot of people tend to love what they just bought or want, and ATI cards were standing on top of the mountain for the past six months.

Definitely... normally people would clamor over the gains a GTX 480 brings over the 5870, but because the 5870 is the "cool kid" in town, everyone rushes to slam the other brand's solutions.

[Amorphous]

I was informed that ATI drivers were cutting out some IQ in a "minor way" to improve performance. I wasn't planning on sharing that tidbit here because I lack an ATI card to actually prove it to anyone with. As far as I know, it affects more than just Crysis. My understanding is that it defaults to some lower IQ settings that affect commonly benchmarked games.

Anyone want to loan me a 5870? :P


Amorphous

[insurgent]

I agree with you, the old you, that it is most likely a bug...in both cases.

Hahaha! Hilarious! "Can't stop the fanboys!"

[saaya]

to all the people claiming you dont see it on the screenshots, use the mouseover script on pcgh... much better to compare it that way, and youll see how it goes from sharp to blurr and from aniso to bilinear

and this is a still... if you move the textures will flicker a lot...

and if crysis is no big deal and nobody plays it, then why did ati cheat and risk ruining their pr karma?

[Coldon]

this may also be why BFBC2 still flickers and flashes on ATI cards, even with the new drivers BFBC2 still has graphical issues with the 5series cards...

[Sn0wm@n]

this whole thread is one giant *ROFL*

[eric66]

this thread is giant fail imho nobody tested with other drivers to see if its really optimization or not

[trinibwoy]

The question is why is there this difference between the directed AF tests and the games?

Because the directed test checks for angle independence, not filtering quality. Everyone looks at that little circle test and thinks everything is honky dory.

[annihilat0r]

this thread is giant fail imho nobody tested with other drivers to see if its really optimization or not

Why does this have to be brought with the "latest" drivers to be an issue? Maybe it was this way from the beginning. And I think it DOES have to bring a performance improvement, looking at the screenshots one looks like bilinear filtering from a certain point, the other is AF all the way.

this whole thread is one giant *ROFL*

why so? because ati can't do wrong?