toms hardware puts up article on ati's filtering game

[UCSDHappyAsian]

well.. at least Nvidia is coming out with something most of us would buy...

 

[sandorski]

Originally posted by: CaiNaM

Originally posted by: sandorski

Originally posted by: phreaqe
bfg10k, have you even read this thread, there is an owner of an x800 pro saying that he notices lowered image quality during game play. and yet you still say that ati is not lowering the image quality. you are one of the most blatently obvious fanboys i have ever seen

Doesn't mean anything. He may have things setup wrong, the game may have a bug with ATI drivers, who knows? Just because he may see some issue, it doesn't mean it has any relation to the subject at hand.

yea.. ati lowers the filtering quality, but it doesn't affect the IQ (and fp16 is as good as fp24) :roll:

"bri"linear is lower quality than trilinear, unless of course you're an ati fanboy, in which case brilinear is only lower quality on nvidia hardware; on ati it's equal quality, even tho the lower quality yields the same performance improvement on both cards. ya gotta love fanboy logic

then of course, there's always "user error" if the facts don't coincide with what you want to hear

What facts? No one has shown anything.

Call me a fanboy if you must, I just ask for proof from anybody. So far there is nothing.

 

[CaiNaM]

Originally posted by: sandorski
What facts? No one has shown anything.

Call me a fanboy if you must, I just ask for proof from anybody. So far there is nothing.

heh.. like i said, "logic".. yes, "ati lowers the quality of the filtering, but it doesn't lower the quality of the image - nvidia on the other lowers the image quality when they lower the filtering.."

that statement speaks for itself.

secondly, i am not repeating hearsay or wishful thinking as you are, rather first hand experience. why shouldn't the burder of proof be on you?

Originally posted by: chsh1ca
CaiNaM, could you use something like Fraps to capture a short video of the effect?

i did a 15-20 sec movie capture in fraps which shows this. it doesn't stand out quite like it does in the game as the image is sized down, however you can definately see what i'm talking about. unfortunately, the .avi is 75mb (zipped - over 100min uncompressed), and dubbing to divx blends the texture in such a way it's useless to use as an example..

edit: i did put up a sshot where you can see the line where the texture detail changes. again, in a still shot, it's not easily noticed, but obvious when you're moving as the line moves directly in front of you (there's actually 3 lines @ 4xaa/16xaf). it's a 2.8mb bmp, but if you have broadband you can view it here.

 

[sandorski]

Cainam wrote:

-----------------
heh.. like i said, "logic".. yes, "ati lowers the quality of the filtering, but it doesn't lower the quality of the image - nvidia on the other lowers the image quality when they lower the filtering.."

that statement speaks for itself.

secondly, i am not repeating hearsay or wishful thinking as you are, rather first hand experience. why shouldn't the burder of proof be on you?

----------------------------

1) Where is the proof? It was clearly shown that NVidias attempt lowered the picture quality. Many sites went deeply into the subject. OTOH, no one has done the same indepth comparison of ATI's effect, why? Is it perhaps unnoticable?

2) The burden of Proof is on you or anyone making the claim.

 

[ZobarStyl]

I prefer to believe a guy who actually plopped down the cash for an R420...and he's right, he would need dvd-quality uncompressed movies to show this effect in motion. But in all honesty, if even one person who owns the hardware says there is a problem, then IQ is diminished, case closed. Why would he bash something he spent 400 bucks on? And only a fool would think that ATi doing bilinear won't do a thing to IQ while NV's brilinear is the devil himself in disguise. The order of quality goes bilinear<brilinear<trilinear, but somehow to you it's NV brilinear<ATi bilinear=trilinear? Wake up fanboy.

 

[sandorski]

Originally posted by: ZobarStyl
I prefer to believe a guy who actually plopped down the cash for an R420...and he's right, he would need dvd-quality uncompressed movies to show this effect in motion. But in all honesty, if even one person who owns the hardware says there is a problem, then IQ is diminished, case closed. Why would he bash something he spent 400 bucks on? And only a fool would think that ATi doing bilinear won't do a thing to IQ while NV's brilinear is the devil himself in disguise. The order of quality goes bilinear<brilinear<trilinear, but somehow to you it's NV brilinear<ATi bilinear=trilinear? Wake up fanboy.

Is NVidia AA = ATI AA? No

Nvidia AF = ATI AF? No

There is no reason that the results have to be the same. Is anyone going to even attempt to prove there is a degradation?

 

[Acanthus]

It cant be proved. Because ATi doesnt give you an option to disable the "optomization".

 

[DKlein]

That was definately noticeable, but I think the line near the top 1/4 was more noticeable.

 

[CaiNaM]

Originally posted by: sandorski

Originally posted by: ZobarStyl
I prefer to believe a guy who actually plopped down the cash for an R420...and he's right, he would need dvd-quality uncompressed movies to show this effect in motion. But in all honesty, if even one person who owns the hardware says there is a problem, then IQ is diminished, case closed. Why would he bash something he spent 400 bucks on? And only a fool would think that ATi doing bilinear won't do a thing to IQ while NV's brilinear is the devil himself in disguise. The order of quality goes bilinear<brilinear<trilinear, but somehow to you it's NV brilinear<ATi bilinear=trilinear? Wake up fanboy.

Is NVidia AA = ATI AA? No

Nvidia AF = ATI AF? No

There is no reason that the results have to be the same. Is anyone going to even attempt to prove there is a degradation?

well, perhaps you're too busy playing fanboy, but i did show and example.

as far as comparing, obviously if your only goal is to try to put ati in a better light, then your opinions make sense, however nv AA and ati AA were different, as they took entirely different approaches to reach the same goal - that is not the case with AF at this point.

AA at this point appears to be quite similar (until 6x, which nv does not provide) this generation, but i don't have an nv40 yet, so it's just second hand info at this point.

personally i preferred nv textures prior to this gen, but this gen it looks like nv lowered themselves to ati standards - nv textures have gone downhill every gen since gf4

 

[nRollo]

Hmmm. A guy goes on vacation misses all the news.

I found a lot of things interesting in this article:
1. Contrary to BFG's assertion, THG states ATI uses brilinear as well
2. In the CoD chart, the X800Xt PE had similar performance to the 6800U with its "optimizations" turned off. When you consider that, and that the 6800U is more advanced, that sort of changes the value of the X800Xt, IMO
3. ATI deserves to be roasted for this after not telling reviewers they were doing it, and asking reviewers to turn off the 6800s optimizations.
4. BFG is still saying "Nothing to see here- all is well- trust in your friends at ATI- ATI knows whats best for you and will program it's drivers to determine the optimal settings-etc."
5. I don't like to be lied to

 

[ZobarStyl]

Sandorski you are correct that Nv AA and AF are not the same as ATi AA/AF...but we are talking about the basic filtering method, which actually have clear definitions of what they entail, thus if implemented correctly each filtering method should look the same no matter who does it. Brilinear (nVidia) looks better that bilinear (both ATi and nVidia) but not as good as trilinear (both nV and ATi, up till now). Thus Nv got in trouble for brilinear because it wasn't as good as tri, but now ATi does straight Bi and you say it's no biggie?

 

[eastvillager]

Long as nvidia comes up with these 'optimizations' first, and fails, while ati implements them next and succeeds, I know whose product I am buying.


Ultimately, I don't care how they present the image, long as it looks good.

Most of this argument is meaningless, since nvidia only seems to produce paper launches lately, and not any actual consumer products.

 

[BFG10K]

there is an owner of an x800 pro saying that he notices lowered image quality during game play. and yet you still say that ati is not lowering the image quality.

Uh, no. I said the lowered quality can't be detected with screenshots but the original poster tried to use those as evidence.

I'm not arguing that Cainam can't see a difference in actual gameplay.

Contrary to BFG's assertion, THG states ATI uses brilinear as well

Tell me Rollo, what is the definition of brilinear? Not applying full trilinear an all surfaces? Not applying trilinear on all mip-maps? Both? Neither? Something else? Somethine else combined with one or more of the previous?

What?

The problem is that term is thrown around all the time but nobody ever bothers to define what it means.

4. BFG is still saying "Nothing to see here- all is well- trust in your friends at ATI- ATI knows whats best for you and will program it's drivers to determine the optimal settings-etc."

Nonsense; I'm saying screenshot cannot be used to prove there's an IQ reduction. That in no way implies that actual gameplay doesn't show a difference.

 

[nRollo]

Originally posted by: eastvillager
Long as nvidia comes up with these 'optimizations' first, and fails, while ati implements them next and succeeds, I know whose product I am buying.


Ultimately, I don't care how they present the image, long as it looks good.

Most of this argument is meaningless, since nvidia only seems to produce paper launches lately, and not any actual consumer products.

Really? People are starting to buy 6800s, how is that a "paper launch" only?
The truth is out there
I bet I can find more people who've bought 6800s than you can X800XTs?

 

[imported_Aelius]

That depends. Two suppliers already ran dry of their supply of X800XTs for less than 450(US).

Not sure on combined #s but I doubt its less than one supplier offering a massive 2 6800Us. Plus there is anywhere from 50 - 100 6800s in the market atm. That's coming from this site. See front page news.

Not a paper launch but it might as well be. Apperantly full scale production is stepping up in July if I read it correctly.

 

[nRollo]

Originally posted by: BFG10K

there is an owner of an x800 pro saying that he notices lowered image quality during game play. and yet you still say that ati is not lowering the image quality.

Uh, no. I said the lowered quality can't be detected with screenshots but the original poster tried to use those as evidence.

I'm not arguing that Cainam can't see a difference in actual gameplay.

Contrary to BFG's assertion, THG states ATI uses brilinear as well

Tell me Rollo, what is the definition of brilinear? Not applying full trilinear an all surfaces? Not applying trilinear on all mip-maps? Both? Neither? Something else? Somethine else combined with one or more of the previous?

What?

The problem is that term is thrown around all the time but nobody ever bothers to define what it means.

4. BFG is still saying "Nothing to see here- all is well- trust in your friends at ATI- ATI knows whats best for you and will program it's drivers to determine the optimal settings-etc."

Nonsense; I'm saying screenshot cannot be used to prove there's an IQ reduction. That in no way implies that actual gameplay doesn't show a difference.

It looks like you're trying to play the same "adaptive algorhyhm" spin games that ATI plays lately BFG. If they're not allowing full trilinear filtering at all, to me, that's a problem. (especially when guys like Cainam say they can see the difference)
So how about I turn the question back on you:
How is ATIs "adaptive" anisotropic and trilinear filtering, which the user can't disable and ATI never would have told you about, somehow OK when you used to flame nVidia to Hell and back for this sort of thing?

I seem to remember you saying you had to have the 9700Pro because it gave you the option of true trilinear? Are you about to buy a 6800U for the same reason?

 

[Shamrock]

round 2 for all the naysayers

http://www.ixbt.com/video2/nv40-rx800-5-p1.shtml It has comparison videos, and PLAINLY shows the differences...ATI's are DEFINTELY lowering IQ.

 

[Shamrock]

Tell me Rollo, what is the definition of brilinear? Not applying full trilinear an all surfaces? Not applying trilinear on all mip-maps? Both? Neither? Something else? Somethine else combined with one or more of the previous?

What?

The problem is that term is thrown around all the time but nobody ever bothers to define what it means.

THis has been thrown up several times now, I think you know what it is, but you ATI boys keep throwing out this question to evade the facts...again.

and furthermore, why not let links argue for me? if I just state things, FIRST thing people do is ask for links/proof...so I just learned to give the proof first. Pictures are worth a thousand words.

 

[nRollo]

Shamrock- it's "bad" brilinear if nVidia does it, but "good" brilinear if ATI does it?
Make sense?
:roll:

 

[SP33Demon]

Just buy ATI and shut up already... you don't even notice this sht in the middle of a game.

 

[nRollo]

Originally posted by: SP33Demon
Just buy ATI and shut up already... you don't even notice this sht in the middle of a game.

Why would I do that? I've had the R300 core twice already, I'm ready for a change.

What I'd REALLY like to see:
If ATIs super fantastic "not brilinear, we call it "adaptive"" filtering techniques are so great- WHY are they not implemented on the 9800XT and 9800Pro?!?!?!?

Hmmmmmm, could it be that a 9800 using these driver "optimizations" would perform a bit too close to the X800Pro and negate the reason to spend $400 on their "new" card?

They're very quick to point out how the optimizations offer no IQ hit, and we've seen how they raise performance, yet the thousands of guys who have 9700/9800s for some reason don't deserve all this free performance?

I'd be pissed if I still had my 9800Pro....

 

[GeneralGrievous]

I bet I can find more people who've bought 6800s than you can X800XTs?

The X800 XT has probably been "sold" a few thousand times over at gateway. Most of the orders seem to have been cancelled though. Zipzoomfly put up the X800XT at $440, it sold out in 45 minutes.

Nvidia has nothing against a $440 X800 XT.

 

[CaiNaM]

Originally posted by: SP33Demon
Just buy ATI and shut up already... you don't even notice this sht in the middle of a game.

actually, you do, which is the whole problem.

could care less if you had to compare sshots and zoom, but when it's obvious while playing, it's an issue.

 

[CaiNaM]

Originally posted by: GeneralGrievous

I bet I can find more people who've bought 6800s than you can X800XTs?

The X800 XT has probably been "sold" a few thousand times over at gateway. Most of the orders seem to have been cancelled though. Zipzoomfly put up the X800XT at $440, it sold out in 45 minutes.

Nvidia has nothing against a $440 X800 XT.

all 6 of em they had in stock?

still, nv clearly has a delivery problem, and the pricing is ridiculous- jakup over at nvnews paid almost $600(us) for his.

 

[Pete]

Denial is not just a river in Egypt, as sandorski seems fond of reminding us.

BTW, if Derek is right in his latest Comdex video card article, nV has only 50-100 6800Us in the retail channel. Does that sound abysmally low to anyone else? I wonder which gets closer to the truth behind that tiny figure: Dave Orton's "wafers per die" anecdote, or just a tiny supply of GDDR3?

Shamrock, I only see pics in that article. Are there videos linked in there?