Testing Nvidia vs. AMD Image Quality

[Keysplayr]

http://blogs.nvidia.com/ntersect/2010/11/testing-nvidia-vs-amd-image-quality.html

Looks like several review sites across the pond are taking IQ a bit more seriously.

This is related to the Catalyst 10.10s and I believe 10.11's. Also related to the FP16 to R11G11B10 render targets that was talked about in another thread here. When I find it, I'll link it up here.

 

[Qbah]

You mean this one?

http://forums.anandtech.com/showthread.php?t=2117420&page=7

As for the actual blog post... I'm a bit disappointed with the way it was handled by AMD. On the other hand, I'm running at default settings and all my games look great (mostly thanks to forcing SSAA), so *shrug*. I do notice the crawling things quite often though (mostly when I can't use SSAA). It's funny though, the comments say it's a biased article, since it's hosted by nVidia... And they completely miss the fact that the poster just gathered articles from regular sites.

 

[Rezist]

It's the same German site linked to originally this just a nvidia blog stating it.
It seems they are only getting this on there own program and not in games. Or they aren't able to catch it in games.

 

[Arkadrel]

Looks like several review sites across the pond are taking IQ a bit more seriously.

Yes... 1 site... a nvidia forum, where nvidia did the testing and wrote a artical. Subjective?

Also theres already a thread on this matter.

Also it seems all the review sites are running at highest quality settings, possibly higher than nvidias, instead of at the default value, so the quality of amd cards when benched where atleast equal to or better than nvidias (to have a fair benchmark).

they probably do things differntly... maybe AMD has some areas where it can show nvidias cards are lacking? would that get any news if so? The old thread smelled of nvidia fanboyisme, claiming that amd purposefully manipulated the benchmark results by doing this switch.

yet every review site seems to have compensated and used high quality settings. Where is the problem?

 

[Keysplayr]

Yes... 1 site... a nvidia forum, where nvidia did the testing and wrote a artical. Subjective?

Look before you leap. All testing was done via 3rd party editors. Nvidia did NOT do the testing. However, they did VERIFY the results of the 4 tech sites who conducted the actual tests.
I think you should just take a moment and look at what this is. Not what you think it is.

 

[Qbah]

It's actually 4 different sites. All German though. And it's an nVidia blog, what did you all expect? Praise for the competitor? There's a change in AMD drivers that impacts default image quality. Some people don't notice it, some do (and once you do, you will see it, always). Why wouldn't nVidia report something negative about their main competitor?

 

[videogames101]

Look before you leap. All testing was done via 3rd party editors. Nvidia did NOT do the testing. However, they did VERIFY the results of the 4 tech sites who conducted the actual tests.
I think you should just take a moment and look at what this is. Not what you think it is.

You could have avoid the immediate criticism by linking the actual articles, rather than the Nvidia blog post. =D

In any case, if you image quality difference isn't noticed by the end-user, and the extra frames are, who is to say that the frames should be given up for greater default image quality? I don't have an answer to that.

 

[Arkadrel]

Who wrote the artical?

"Nick Stam
Technical Marketing Director"

Im sorry but this artical reads like a marketing trick, saying competitors cheat, buy ours instead.

In the old thread, there was 1 old game, and 2-3 old benchmarks that showed differnces in quality. No one ever bothered to try it on a newer game? why not? could it be because it only shows up in 1 old game?


A leak about 7th december being launch date for 6970's, and the same day nvidia's marketing director writes a artical, and its posted here. Its spin, which I dont really blame nvidia or amd for doing, if it helps them make more sales, why not. However people should be able to see it for what it is when either do it.

 

[Keysplayr]

You could have avoid the immediate criticism by linking the actual articles, rather than the Nvidia blog post. =D

In any case, if you image quality difference isn't noticed by the end-user, and the extra frames are, who is to say that the frames should be given up for greater default image quality? I don't have an answer to that.

Oh, no. The criticism would be there regarless. And that's ok.
You can easily see the 4 sites links in the blog.

 

[Keysplayr]

Who wrote the artical?



Im sorry but this artical reads like a marketing trick, saying competitors cheat, buy ours instead.

In the old thread, there was 1 old game, and 2-3 old benchmarks that showed differnces in quality. No one ever bothered to try it on a newer game? why not? could it be because it only shows up in 1 old game?

Nah. Nick Stam wrote the blog ABOUT the other 4 tech site findings. And I don't think this a marketing trick Arkadrel. I think Nvidia just might be fed up with it.

 

[Arkadrel]

My belief is that its a isolated issue, that they stumbled upon in oblivon, which is why its the only game we hear of it in. Then there are a few old benchmark programs or such that show it too.

Meh.. does this happend in all games? or just a few games and then nvidia make a artical about it twisting it to their agenda? There is a differnce.

 

[WelshBloke]

Isn't this actually a repost of the earlier thread.

Theres nothing new in that blog post.

 

[Keysplayr]

My belief is that its a isolated issue, that they stumbled upon in oblivon, which is why its the only game we hear of it in. Then there are a few old benchmark programs or such that show it too.

Meh.. does this happend in all games? or just a few games and then nvidia make a artical about it twisting it to their agenda? There is a differnce.

What is an isolated issue? I suddenly got the feeling that you didn't read anything and assumed it was only about the FP16 to R11G11B10 render targets thread that Qbah linked to?

 

[Arkadrel]

Isn't this actually a repost of the earlier thread.

Theres nothing new in that blog post.

Yes its a repost (that doesnt seem to have anything new in it) ^-^ other thread was dead, and the higher up this one is and the more that see it the more effect a scare campagne it is.


Thread crapping is not acceptable.

Moderator Idontcare

 

[apoppin]

yet every review site seems to have compensated and used high quality settings. Where is the problem?

Is there a problem? Do other reviewers use Quality for either AMD or Nvidia CP settings?


i always use the highest quality setting in each CP

 

[Rezist]

Well it seems AMD was pretty lucky it only showed up in oblivion, HL2, and the racing game. And a ground texture program.
Although the oblivion screen shots look to me like AA.

The F1 shots are clearly AF.
HL2 I couldnt see any difference.

 

[Keysplayr]

Isn't this actually a repost of the earlier thread.

Theres nothing new in that blog post.

There isn't? There are now comparison videos, AF testers, and proof of testing application detection in AMD's drivers. For starters?

 

[Arkadrel]

I was looking at the HL2 screenshots, couldnt notice anything either.

Its clear in the oblivion game, which is why I think its more of a isolated issue with a few games than something that happends in every game.

There isn't? There are now comparison videos, AF testers, and proof of testing application detection in AMD's drivers. For starters?

arnt the comparison videos and the af testing program, the same thing your thinking off? from Tweakpc.de. Thats not new, that was mentioned early on in the old thread. ComputerBase is baseing those claims again off the same test program, and their subjective about it ("appear to" instead of saying "does", why the differnce? if they actually had proof why not say it did for certrain?). As far as I can tell theres nothing new in that artical that wasnt in the old thread.

 

[WelshBloke]

There isn't? There are now comparison videos, AF testers, and proof of testing application detection in AMD's drivers. For starters?

Well you'll forgive me for not following all the links as they are all in German.

I didnt know you spoke it.

 

[Keysplayr]

Well you'll forgive me for not following all the links as they are all in German.

I didnt know you spoke it.

If you really wanted to read the articles, you would have run them through google translator or the like already. If you're not interested, you're not interested. I'm not forcing you to read anything.

 

[WelshBloke]

If you really wanted to read the articles, you would have run them through google translator or the like already. If you're not interested, you're not interested. I'm not forcing you to read anything.

I clicked through a few of the photos on the first link and couldn't really see any difference except for the racing game, and that wasn't really that noticeable.

If I really wanted to read them I'd get my wife to translate them (she's a professional translator), given that there appears to be nothing that wasnt said before I'm not going to subject myself to the pain of reading something translated by google.

 

[MentalIlness]

Im sorry but this artical reads like a marketing trick, saying competitors cheat, buy ours instead.

+5

 

[Keysplayr]

I clicked through a few of the photos on the first link and couldn't really see any difference except for the racing game, and that wasn't really that noticeable.

If I really wanted to read them I'd get my wife to translate them (she's a professional translator), given that there appears to be nothing that wasnt said before I'm not going to subject myself to the pain of reading something translated by google.

Understood. And very kewl your wife being a translator. Must be fun on vacations.

 

[WelshBloke]

Understood. And very kewl your wife being a translator. Must be fun on vacations.

Its a nightmare, everyone assumes I speak their language as well.

I can look non-committal and shrug in about 5 languages. :\

 

[Sind]

My 10.11s for my 5870 came via Windows Update and the setting remained on the "High Quality" the slider didn't adjust itself. He states further that even on the 5xxx series there was adjustments and that I would have to disable CAT AI to achieve the same as NVs default quality. It's still on standard and I have noticed no difference in games versus the previous 9.8s that I had installed.

Maybe there is a difference, maybe there isn't. Personally I couldn't tell between the two driver sets. Further until I see what I consider, and I don't mean to come off the wrong way here, english language in-depth testing of image quality from reputable sites such as AT, HardOCP I'm inclinded to believe that any differences discovered are negligable or are marketing speak.

I'm looking seriously at a GTX 580 vs a 6970 and it would be nice to have a seriously indepth non bias review of image quality between the two competitors and their latest and greatest and get the exact meat of if there is any difference, and if there is - does it really matter. I also find it strange that 4 german sites are seemingly all over this and the english language review sites seem to either not care, or do not have an issue so far.