Testing Nvidia vs. AMD Image Quality

[AnandThenMan]

All people need to do is look at the original post, the link that is posted, and then consider what is going on here. Nvidia has found a perceived weakness on AMD hardware and jumped all over it, and the focus group members post the info and take every opportunity to highlight it.

Is that fair? Sure. But as a wise man once said, consider the source. If and when AMD does the same thing, take it with a grain of salt, just like people need to do here. And make up your own mind, vote with your wallet. Simple.

 

[Keysplayr]

Guru3D is disputing those other sites findings, they are not disputing Guru3D, which is what I asked for. Their past like or dislike from "ATI brethren", as you label them, is not the subject. It's a straw man. No point in linking to past threads. They would be irrelevant to the discussion here.

Strawman. Ok, if you say so.

You might want to try to understand, that I bring up Computerbase's past glories in the eyes of ATI fans for a reason. The reason is you discard not 1, not 2, but 4 German sites instantly if favor of 1 other site. I'm reminding you that you should not, at the very least discard computerbase so easily.

And it has every bit as much to do with this subject as anything else.
You're calling Computerbase's findings into question, and effortlessly advocating Guru. Makes ya think, ya know?

 

[Keysplayr]

All people need to do is look at the original post, the link that is posted, and then consider what is going on here. Nvidia has found a perceived weakness on AMD hardware and jumped all over it, and the focus group members post the info and take every opportunity to highlight it.

Is that fair? Sure. But as a wise man once said, consider the source. If and when AMD does the same thing, take it with a grain of salt, just like people need to do here. And make up your own mind, vote with your wallet. Simple.

Sounds like you're on a soap box looking for witches. Nvidia blogged about it, but it wasn't them who found the IQ discrepancy. You know who actually FOUND the issue. 4 sites worth. And by the way, it's not just focus group posting about it. It's all over the web being posted by everyone. So, while you sit there and try to create reasonable doubt about the OP and it's contents, the rest of the world can actually click the provided link and the following 4 web site links and read for themselves, WITHOUT your guidance, thank you very much. :hmm:

 

[SirPauly]

I'm not. Computerbase, PCGH, 3DCenter.org, and TweakPC are. Don't be so quick to forget what a forum favorite Computerbase was ESPECIALLY when they used 8xAA in all their benches showing ATI in a pretty light when at the time, NV took a bigger performance hit at those AA levels. Boy oh boy did this forum love them then. Not now though? But why?
They were loved SPECIFICALLY for testing in ways other review sites didn't, like using 8xAA when not a lot of other sites used more than 4x. It's that Computerbase went the extra mile that they were heralded one of the best review sites on the web by ATI bretheren. I can link to threads here from that time to prove it after a bit of searching. They are still here..

Computerbase.de is one of my favorite sites based on they used x8 AA, even though ATI had an advantage and was good to see nVidia improve here as well over time because it was important for some and to me, too. When one cherry picks and cherry chooses quality or immersion, it usually offers a camp mentality in their posts.

But, one of the points I use to raise about image quality was even though nVidia's hit was more with x8 AA in the past, they raised the bar with a very efficient CSAA feature that allowed impressive performance over-all with their x16Q and made it very viable to use compared to the CFAA features.

What I am trying to say, it is good to see AMD offer an efficient CSAA feature as well for their up and coming generation and to see their MLAA. Hopefully, nVidia reads this and may offer MLAA or a feature similar for their customer base.

When one fights with each other really clouds the constructive views that IHV's may read.

 

[AnandThenMan]

...the rest of the world can actually click the provided link and the following 4 web site links and read for themselves, WITHOUT your guidance, thank you very much. :hmm:

And the rest of the world can read and decide for themselves, without your "guidance". It works both ways. Like I said, vote with your wallet.

BTW, the "witches and soapbox" comment is very unprofessional IMO.

 

[Keysplayr]

And the rest of the world can read and decide for themselves, without your "guidance". It works both ways. Like I said, vote with your wallet.

BTW, the "witches and soapbox" comment is very unprofessional IMO.

Well thank goodness I'm not a professional then.

 

[SirPauly]

All people need to do is look at the original post, the link that is posted, and then consider what is going on here. Nvidia has found a perceived weakness on AMD hardware and jumped all over it, and the focus group members post the info and take every opportunity to highlight it.

Is that fair? Sure. But as a wise man once said, consider the source. If and when AMD does the same thing, take it with a grain of salt, just like people need to do here. And make up your own mind, vote with your wallet. Simple.

Of course it is fair. These are findings by really fine web-sites and many threads on the Internet. AMD and nVidia are big boys and gals and can take care of themselves. Consumers need web-sites to test hard and test hard for quality because it is important to some.

 

[Xarick]

"I don't know much German so I decided to post this feedback here in English because I feel it's pretty important. I do not know much German so I decided to post this feedback here in English because I feel it's pretty important.

When using 3DCenter Filter Tester and NVIDIA card it seems that even when using renamed .exe HQ filtering quality is always forced on. When using filter tester 3DCenter NVIDIA card and it seems that even when using renamed. Exe HQ filtering quality is always forced on. Here's the proof: Here's the proof: "

From the 3dcenter link I posted before. Has anyone else looked into this? Apparently in that filter test that shows AMD HQ on par only with Nvidia Quality, that Nvidia Quality is really HQ because somehow Nvidia forces HQ on this test. They seem unsure if it is a driver bug.

 

[HurleyBird]

I'm with the Nvidia fans this time. Whenever AMD is in the right (or at least when they believe they are) they back up their statements with technical details (Batman AA, FP16 demotion, Hawx2/Tessellation). This is not the case here. Saying that their IQ is good enough to deliver a "very good gaming experience" is a non-answer, it's beating around the bush, and if anything it reminds me of classic "we believe our business practices are fair and honest" Intel.

 

[Stoneburner]

Of course it is fair. These are findings by really fine web-sites and many threads on the Internet. AMD and nVidia are big boys and gals and can take care of themselves. Consumers need web-sites to test hard and test hard for quality because it is important to some.

YOu are correct that it's a good thing that there are good people over in the fatherland doing such things. The problem is that for the fanboys (and shills) tend to spin the issue like a political campaign instead of letting things get sorted out. This is why the baffling team mentality that afflicts some people on this forum is COUNTERPRODUCTIVE.

Point in fact: I read through this entire thread and I have learned nothing concrete. Instead we have keys throwing out cryptic remarks about how he's not making certain claims and is not responsible for their content. The lady doth protest overmuch.

So here's what I Need answered:

1. As an owner of a 6870, I would like to know exactly (if anything) is flawed.
2. I would like to know what games, if any, suffer image degradation
3. I would like to know the exact cause of the degradation.

Since it's keys that's vicariously making these claims, it would be nice if he would vicariously answer my questions as the Vicar of Nvidia.

 

[Skurge]

I'm with the Nvidia fans this time. Whenever AMD is in the right (or at least when they believe they are) they back up their statements with technical details (Batman AA, FP16 demotion, Hawx2/Tessellation). This is not the case here. Saying that their IQ is good enough to deliver a "very good gaming experience" is a non-answer, it's beating around the bush, and if anything it reminds me of classic "we believe our business practices are fair and honest" Intel.

What exactly do they have to answer to? 4 German sites that found the same issue with the same games? All of them being 5yrs old or so?

Other sites see no image degradation Xbit and Guru3d being some of them and they tested newer games. I don't really see what all the fuss is about. I could start another thread where nV has filtering issues too, but I won't cause those are Isolated cases in some old games. Does that mean nV is cheating too? I don't think so.

 

[busydude]

2. I would like to know what games, if any, suffer image degradation

They could 'prove' it for Half life 2 and Oblivion.

 

[MrK6]

Since we're being unprofessional, I enjoy seeing this blow up in Keys' face. That said, I think there have been some great points made in this thread. To recap and add my two cents, a few sites found some discrepancies in image quality when testing old games on the new AMD hardware. My first question is why were they testing these specific games out of all the games out there in the first place. Seems rather odd. Furthermore, there was only a noticeable difference, to me anyway, in Trackmania. The other games looked identical in the screenshots, so what's the fuss about?

Then, other sites tested these claims, especially in newer games, and found there to be zero difference. It seems to me like a case of AMD's newer driver tweaks being applied oddly in a few older games; this happens all the time in both camps, so why is it news? Oh, right, AMD just released a great set of of cards that crushes the competition so the Green Team PR squad is out in full force. Why would AMD "cheat" on its drivers in five year old games that no sites use to benchmark? Think about that for a second to see how ludicrous the conjecture proposed in this thread is.

 

[Skurge]

They could 'prove' it for Half life 2 and Oblivion.

To be fair, the HL2 issues have always been there long before even DX10 hardware came out for both vendors. I even saw it on my old 9600XT.

 

[thilanliyan]

if this is not affecting newer games why does it matter so much? just to be sure, this is not affecting newer games right?

 

[Keysplayr]

YOu are correct that it's a good thing that there are good people over in the fatherland doing such things. The problem is that for the fanboys (and shills) tend to spin the issue like a political campaign instead of letting things get sorted out. This is why the baffling team mentality that afflicts some people on this forum is COUNTERPRODUCTIVE.

Point in fact: I read through this entire thread and I have learned nothing concrete. Instead we have keys throwing out cryptic remarks about how he's not making certain claims and is not responsible for their content. The lady doth protest overmuch.

So here's what I Need answered:

1. As an owner of a 6870, I would like to know exactly (if anything) is flawed.
2. I would like to know what games, if any, suffer image degradation
3. I would like to know the exact cause of the degradation.

Since it's keys that's vicariously making these claims, it would be nice if he would vicariously answer my questions as the Vicar of Nvidia.

Stonedude, there isn't anything "Cryptic" about, "I'm not saying it, but 4 German sites are.". You don't need a Runestone to decipher what that means. Please don't take what I say and blow it out of proportion. As I mentioned, I am certain those 4 German sites are continuing their investigations with other titles and other cards. I'm inquiring about it myself to get details.

And at MrK6, I appreciate that you would love to see this blow up in my face, but it would not be my face this blows up in. I will admit, I am glad that these 4 sites are probing into this issue (you know, the issue that doesn't exist) and lets the public know what is going on. Other sites will jump on this eventually, or at the very least when the new 69xx series launches. If they don't bother to test IQ when those release, especially after this issue, then the cred goes down the tubes. IMHO, the next round of benches, whether Nvidia or AMD releases another new card first, should include in depth IQ testing to see if the cards being pitted against one another are in fact doing the same amount of work.

Now if you would put your personal glee at my demise aside, lets keep the technical convo going. Thanks.

 

[Keysplayr]

if this is not affecting newer games why does it matter so much? just to be sure, this is not affecting newer games right?

You may want to ask BFG10K to comment on that one.

 

[Larries]

Just wondering, does ATI cards suffer from this same HAWX bug too???

http://www.kitguru.net/components/g...-gtx570-be-caught-in-new-app-detection-cheat/

So, in addition to driver optimisation, driver bugs, and now, we have application bugs that may affect image quality too!

 

[Keysplayr]

Just wondering, does ATI cards suffer from this same HAWX bug too???

http://www.kitguru.net/components/g...-gtx570-be-caught-in-new-app-detection-cheat/

So, in addition to driver optimisation, driver bugs, and now, we have application bugs that may affect image quality too!

You mean the Hawx application calling for the highest sample quality AA from the driver when it doesn't have the fix? In other words, runs 16CSAA by default instead of 4xAA by default?

 

[notty22]

http://benchmarkreviews.com/index.php?option=com_content&task=view&id=12845&Itemid=47
This web site has noted shimmering issues in newer games. Which don't show up in screen shots.
TH has made a news post about this issue, but has not commented in editorial fashion.

EDITOR'S NOTE: This is a disturbing article, and the sources here are critical for legitimacy. NVIDIA is a direct competitor to AMD and is the author of this article, which may lead some readers to ignore the message. However, it was several independent review website's that first brought this issue to the forefront, and proved it exists. I personally trust these websites, particularly 3DCenter.org, and have found them to be unbiased over the years.
Benchmark Reviews can confirm that issues with filtering still exist, and pointed this out in our Radeon HD 6850 and Radeon HD 6870 launch articles. We also made it public that certain AMD partners were sending 'juiced' video card samples to reviews sites, ours included, with details published in our 1120-Core "Fixed" Radeon HD 6850 Review Samples Shipped to Media article. So could this be AMDs last ditch effort to compete with NVIDIA by manipulating performance?

 

[Keysplayr]

http://benchmarkreviews.com/index.php?option=com_content&task=view&id=12845&Itemid=47
This web site has noted shimmering issues in newer games. Which don't show up in screen shots.
TH has made a news post about this issue, but has not commented in editorial fashion.

Very cool. Now maybe Benchmark Reviews will join in with their own new extensive testing.

 

[Larries]

You mean the Hawx application calling for the highest sample quality AA from the driver when it doesn't have the fix? In other words, runs 16CSAA by default instead of 4xAA by default?

Yes. Not sure if ATI also suffer from this Hawx bug. If yes, then all sites comparing 4xAA are basically comparing 4xAA from Nvidia vs whatever is highest setting from ATI. Which doesn't sound very much like apple to apple comparison.

(Granted, it is not Nvidia's responsibility to fix it for ATI.)

But then, this just adds further support that simply setting the same quality setting at driver and/or application level is not sufficient (all review sites should be doing this in the first place anyway). Driver bugs and application bugs may also affect image quality, distorting the true comparison.

So, reviewers have to find some way to actually comparing the image instead of trusting the application / drivers

 

[Xarick]

can someone point to where benchmark reviews shows it showing up in newer games. They only indicate it shows up in the filter tests. Which is true. However this is the same test that nvidias default quality is actually high quality (see my previous link).

 

[Stoneburner]

You may want to ask BFG10K to comment on that one.

Yo BFG10k, drop some science on us!

(I did it for him)

 

[Stoneburner]

Stonedude, there isn't anything "Cryptic" about, "I'm not saying it, but 4 German sites are.". You don't need a Runestone to decipher what that means. Please don't take what I say and blow it out of proportion. As I mentioned, I am certain those 4 German sites are continuing their investigations with other titles and other cards. I'm inquiring about it myself to get details.
.

This is the glenn beck method of "not" saying something.

"Now people! I can't prove that Keysplayer has gout. But I can't disprove it either. I'm just asking questions here!"

I mean, you posted this thread, you to hold a neutral "Let's wait for more information" stance, but everybody else who has actually stated anything neutral has gotten shot down (by you) without any technical explanation.

IN the end, this is a technical forum. We want answers, not spin. And definitely not fried rice. Not today.