R420 IQ Comparison

If you have a ATI card then you have seen it in motion and guess what...for the last year ATI has won the IQ contest using it. Give your head a shake.

 

I agree, the second link is every so slightly sharper in the ATi card but it is VERY SLIGHT. It is within 0.1% margin of difference IMO.

 

Compared to relatively low quality AA and hacked brilinear filtering on the FX cards, why am I not surprised?

 

After changing to 800x600 again i see what might appear to be a slightly sharper image on the ATi. But Im not even sure if im seeing this because its there or because you are saying it is.

 

thats not the question at hand, what im asking for is why does he feal compelled to go this far out of his way to for the aligations thrown at ati but he didnt seem to go this far for nvidia

 

ok different algoritm, got it. Thanks .
Now what are the significant differences ?

 

Then what is the debate about, image quality is excellent, AA is great..go play games people.

 

BEcause he didnt have too. Everyones grandma could see it.

 

ATI Slogan?

 

I linked this pic because I left it in it's original form:

Other Pic

I mean, if this is gonna be how difficult it is to spot... then great job ATi. Though if I notice it gaming, I still want the ability to turn on/off full Trilinear. I am paying so much for a freaking card, please allow me that right... at least.

 

There's one thing I'd like to know if it's possible: ATi have said this is an "adaptive" or "intelligent" algorithm, capable of scaling itself giving certain conditions; would it be possible for them to exceed the filtering levels that see on the coloured mip tests? I play Day Of Defeat almost exclusively, I don't need performance tweaks there, but I don't care if they don't mess my output -- however, if they could possibly increase the quality further (at the expense of performance) in games like this, I'd be all for it!

 

If I turn my desktop gamma up near max, I'm sure I can find at least 4 or 5 pixels in that region that are different. Also one of the shadows is lighter. How does polygon alignment, and shadows effect trilinear filtering?

 

Because it didn't take nearly this much work to spot nvidia's implementation of brilinear. Read his 5900Ultra review if you think he didn't look into the matter.

 

Ok the grass is different. is this was your referring to?

 

well, I'm not to versed on the issue, but I'd say: NVIDIA's brilinear was a fixed algorithm that was could not scale/adapt like ATi's algorithm. I'd further wager, that there are conditions when ATi's algorithms adapts to be the SAME as NVIDIA's, as NVIDIA's may be perfect given very specific conditions.


Just speculation really -- we do know that ATi's algorithm is different than NVIDIA's, however.

 

Dave,

I hate to be a PITA but while those shots like fantastically similar, could you do one with a bit more range ? Both of those have limited scene depth.

It would make me happy

 

I think your seeing things. If you cover up the areas of the screen away from the top square then do the mouseover there is no difference. If you leave the sky in view, the area appears to change brightness, with no change in texture quality what so ever. Unless, I have crap eyes, I think your seeing an optical illusion, or "other".

 

I really can't see a difference no matter how hard I try, is there any really obvious examples of the difference?