Is it just me, or do the rest of you detect bias in maxpc's conclusion:

"That’s good news for AMD, too, because now the company need only worry about catching up on one performance metric: frame rate. Unfortunately, we don’t think CrossFireX is going to be a panacea in the interim. Running four moderately powerful videocards in one box will never be as cost effective as building a rig with one super-powerful GPU—especially if the CPU in that box is an Intel quad core. Sorry, Phenom."

"That leaves Nvidia in the catbird seat—again. But it won’t have the perfect solution either until it replaces the 8800 GTX and 8800 Ultra with parts that support HDCP on both links of their dual-link connectors and that are capable offloading all HD video decoding from the host CPU. "

Maybe it's just because I know that today AMD wins the framerate contest as well with the HD3870X2, while also being lower cost than the ultra.

Man, the picture that comes up on this in the featured article rotation keeps surprising me. I'm used to coming to TR and seeing technological eyecandy, not the other kind.

It's finally coming together...the French manicure, the pink MP3 player, and now this picture...Geoff Gaisor really is a hot girl!

All I know is that my "old" X1900XT, used with Windows XP MCE, scales a video better than my "newer" 8800 GTS used with Vista. Why this is the case? Well I'm not really certain...

The key is, however, that I'm ready to try replacing my 8800GTS with at least an HD 3870 to see what, if anything, improves.

I'm glad that nVidia are serious about image signal quality. I've mainly used ATI because of their decent signal quality, though still not as good as Matrox. On my last CRT monitor, I stayed with my Matrox G400 for years because that monitor was particularly sensitive to crap signals. I remember trying a GeForce4 MX and getting a nasty blurry picture, not long after that I got a Radeon 9000Pro and at last I had hardware T&L and a decent signal. It's not so bad these days as I'm using DVI but it's still appreciated that they took the trouble as I see it as a sign of a quality product.

BTW The Mach32 on my 386 has much better signal quality than that MX did.

AMD vs. Nvidia? They must mean ATI vs. Nvidia.

Digital Vibrance FTW!!11

(for the loss or tie, according to the above :)

http://www.nvidia.com/object/feature_dvc.html

I still miss Matrox..........alas, my poor G400Max, I knew you.....

One question :

May I have the higher res file for this girl ?

I say the difference is minute (at very best), and from that point I look at frames per second. Go nVidia.

The control group section of the article reminds me of some experiences at the Chanhassen Dinner Theatres. I sell souvenir photographs and I always present two photos. People are more eager to buy when I give them a choice. There are a few times when one photo is absolutely terrible -- sometimes my fault, most of the time theirs -- and I'd have the lab print two of the good photo. On more than one occasion , I've had to argue -- not screaming or anything, more of a laughing matter -- with customers about the photos being identical. The customers I argued with had always been female come to think of it.

I've got one problem with their gaming tests: They were using 30" monitors with a native resolution of 2560 x 1600, but were running the game at 1280 x 800. No where do I see them mention if they used the graphic driver's scaling, the monitor's scaling, or no scaling at all.

If they're going to do a test like this, it would have really been preferable for them to have run the games at the monitor's native resolution. Failing that, turn off the monitor's scaling and use the graphic drivers scaling modes.

I'm scandalized. Shocked. I distinctly remember that TR's policy was to display zero "booth babes" on its site. It was clearly printed on one of the TR t-shirts in a handy bar graph. All of a sudden, I surf on over to TR's main page and lo and behold, there's a babe from this article staring me in the face.

I mean this somewhat in jest, of course, but it makes it especially hard for me to swallow because I despise Minimum PC Maximum BS Maximum PC's exaggerated, artificial machoness that they use to try and hide bias and laziness. It's why I stopped reading their magazine a long time ago.

Of course, since I mean it somewhat in jest, that means that it's mostly a serious concern. Can I expect this otherwise respectful, tech-oriented site to bathe my screen with eye candy that distracts me from the content?

Interesting stuff. This is somewhat similar, with GPUs, to the listening tests we've done in our sound card reviews, like so:

http://techreport.com/articles.x/13874/8

Using a panel of observers and getting some subjective impressions is a cool idea for GPU IQ. Wish I'd thought of it!

The results also pretty much confirm my impressions of the state of GPU image quality. Here's what I said about it in my Radeon HD 3800 series review:

http://techreport.com/articles.x/13603/10

And, as complex as these GPUs are, the issues really do boil down to price, performance, and adequacy at the end of the day. The DirectX 10 spec has firmed up the requirements for image quality, and partially as a result, the latest Radeons and GeForces produce very similar, great-looking output. Although I think AMD has something of an edge in terms of HD video playback filtering and noise reduction in some ways, even that is largely a wash. Most HD movies are going to look gorgeous on either card, regardless.

There are still issues here in 3D rendering, finer points of difference in things like texture filtering quality, but the DX10 generation of GPUs really cleaned up the worst offenses. Frankly, this is one reason we haven't talked terribly often about image quality in our more recent GPU reviews: there's simply less to complain about, and both solutions do a good job overall, with very similar image output. I'd still like to have more time to pay attention to these things in an ideal world, but things don't change much when you move a half-generation ahead (i.e., from G80 to G92 or R600 to RV670) and keep the same technological foundation.

Most of this:

http://techreport.com/articles.x/12458/5

and this:

http://techreport.com/articles.x/12458/6

..and the following pages still apply.

Incidentally, back in the day when analog VGA connections roamed the earth, ATI cards had a warmer color temperature and more color saturation. GeForces were cooler and less saturated. I would expect more people to prefer the warmer, more colorful default settings of the Radeons. The difference was pretty striking. Things seem to have gotten closer with the move to DVI, so that I don't really notice it anymore.

I can never see the difference in the high-end cards...

I read this article a week ago when I bought the magazine and wondered why they kept going off topic.

MaximumPC was comparing specific cards, but kept rambling on about how the faster, higher priced G-Force cards would trounce the AMD card being reviewed, which I thought was going against what the article was about and showing quite a bit of bias.

I threw the magazine in the garbage.

i'm sure the statisticians and the scientists out there will have a field day with these results.

One thing this can do is give a better sense of the IQ for objects in motion. For gaming IQ depends quite a lot on avoiding things like texture crawl, sparkles, and popping -- all things that you don't see in a static screen shot -- and some of these are more objectionable to some people than others.

Ah ! finally an experiment to give the better result rather than some editor saying what he thinks has better quality.