PDA

View Full Version : ATI cheating with AF


Pages : 1 [2] 3 4 5

Nutty
05-16-04, 06:30 PM
No I cant see them either.. something weird is going on with him.

OWA
05-16-04, 06:31 PM
Must have been deleted.

Nutty
05-16-04, 06:32 PM
http://www.nvnews.net/vbulletin/showthread.php?t=28821

OWA
05-16-04, 06:36 PM
How funny. :) Thanks for the link.

Evildeus
05-16-04, 06:39 PM
Comparaison screenshot made by DB of B3D (2.8Mo) (http://www.beyond3d.com/reviews/ati/r420_x800/iq_compare/compresonator.zip) ;)

OWA
05-16-04, 06:44 PM
Can you provide commentary on what we're looking at with the comparison? Is it better to show more lines/dots/whatever it is in the comparison part?

Gouhan
05-16-04, 06:55 PM
Is this what it has all come down too?
So what if their doin' funny things in the drivers. If you can't see it in the game, then I guess its okay. I just think this is being anal about the whole thing.
nVidia has done worse in the past and so has ATI.

So I think people should just let this slide...

However it is interesting that people can start thousands of topics everywhere else about something as small as 2molex connectors and drag nVidia's name through the dirt. When it's something that affects you directly though (as in what your games look like), it hardly gets as much attention.
ALL the people with the attitude that ATI could do no wrong are nowhere to be found all of a sudden. I bet though is a thread started called "NVIDIA cheating with AF" there'd be 30 replies within 10mins...

Blacklash
05-16-04, 08:02 PM
This will probably be ignored.

It seems to be basically the same thing as Nvidia's forcing brilinear. The nasty little twist is that it hides itself when you are using an analyser that uses colored maps to evaluate AF. Personally I think too much is made out of the whole brilinear thing. I only have issue with hiding it and/or forcing it always when something else is requested. Both Nvidia and Ati need to be punted in the arse when doing these sort of things.

however I will defend Ati like most do Nvidia and say 'it's a bug'.

It is amusing when you read reviews that assert the X800XT 'owned' the 6800u when it's besting it's FPS by 3 or so, but when the ultra wins with 15+ it's the Ati cards "suffering more" with AA/AF on in this "particular game".

Time will tell, when one is out, I am getting it to see what's what.

Ruined
05-16-04, 08:06 PM
Heh at least you can see Nvidia's AF optimizations when present, ATI is hiding them from the user. Makes those mipmap programs 100% useless on ATI cards.

The question is, will people hold ATI up to the same standard they did with Nvidia? Not are users getting brilinear, but you were tricked into thinking you were getting trilinear. That should piss off reviewers, enthusiasts, and especially people who already bought x800 cards:

A) it effects image quality in all applications without you knowing it
B) it was a hidden optimization to make the ATI card look faster than the competition when in reality it may be the same speed or slower

This is on the level of NV's suspected clipping planes optimization (that was subsequently removed) back awhile ago in 3dmark03.

This also puts a serious dent in the "ati is faster with high res/high af" argument department

-=DVS=-
05-16-04, 08:48 PM
Actually this picture doesn't prove any cheatin X800XT got big fillrate advantage over other cards frame rate is cpy limited only pro would be suspicious but no pictures to prove bad quality......

Looked at what Evildeus linked , but i see no quality difference ingame shot from R9800 vs X800 , only 5950 slightly better on ground btw why no 6800 comparision shots :eek2:

http://www.nvnews.net/vbulletin/attachment.php?attachmentid=5890&stc=1

grimreefer
05-16-04, 08:57 PM
and my friend asked me the other day" u still gettin the nvidia card , the ati card looks much faster" Its hard to convince noobs not to listen to everything they hear.

Blacklash
05-16-04, 09:10 PM
Well we know Ati increased the clocks on their cards when Nvidia released the 6800u. We also know Nvidia came out with the '6800' extreme in response to the good showing by the X800XT. Bottom line is both card series are close that is why we have all of these thing occurring.

If Ati has it so brilinear is forced all the time and is hidden for proper trilinear when analyzers are used that is underhanded. Nvidia is forcing brilinear with the 61.XX I believe. Thing is they don't hide it from the analyzer/color maps.

gordon151
05-16-04, 09:26 PM
Well we know Ati increased the clocks on their cards when Nvidia released the 6800u

Why do you keep saying that when people who have been familiar with the card for a while say its not true?

MontoyaSG
05-16-04, 10:03 PM
hmm let's see how the fanboys reply

Nv40
05-16-04, 10:10 PM
Actually this picture doesn't prove any cheatin X800XT got big fillrate advantage over other cards frame rate is cpy limited only pro would be suspicious but no pictures to prove bad quality......

http://www.nvnews.net/vbulletin/attachment.php?attachmentid=5890&stc=1

For the record im not saying they are cheating ,it can be a BUG in AF or that reviewers are not doing well their Job .it just that im skeptic about -free Trilinear AF all the way to 16x.its simple not possible .more because they lose performance in older games.if that is true then why the performance settings exist? it is free then why there is an option at all? a drop of ~2fps (near 0%) from 0AF to 16AF trilinear ,when NVidia lose ->40%!! is wishfull thinking .you cant see "bad quality" in still shots when BI-linear is used. only in motion. the point is that no review that use ["high quality"] settings in the CP will show you the real performance of ATi hardware (when HQ AF is selected ]. and almost all reviewers use the CP HIgh quality settings ,because thats the reason for the control panel to exist. to have full control of basic QUality settings. otherwise what's the point of having a control panel at all? ..

if the Fillrate is the reason for the "stellar performance" of the R420XT with "HIGH QUALITY" AF16x ,then why the R420pro will less fillrate than the NV40 and less pipes also have a similar behaviour? R420 pro lose just only 11% of their performance. NV40ultra with much fillrate and pipes lose 40%. either ATI is extremely efficient with AF or something is else is going on here.

and R420XT does lose performance in other games with "HIgh quality" AF enable , and the FIllrate is still the same.


Interestingly, the X800 cards are unable to play their anisotropic filtering trump in the call of duty. It seems that their superior filtering performance is only available under Direct3D!?
http://www.tomshardware.com/graphic/20040504/ati-x800-18.html


THere is -no consistency- with the "stellar performance" with AF in the R420PRo and the XT . sometimes its Free all the way to 16x "HIGH QUALITY" and sometimes is not ,and it looks that depends on the game. and unfortunately it looks that it happens in the latest games that most want to see like UT2k4/FArcry ..the ones where ATI "wins" more easily .

and IF it is true that ATI is detecting software that analize what they are really doing with their real HQ settings ,and switching "off" all their "optimizations" . then i will not be surprised that they are not that fast as reviews shows ..and that indeed they are going all the way to the cheapiest possible way ,only in pre-selected games games.

micron
05-16-04, 10:38 PM
hmm let's see how the fanboys reply
You havent even made it to fanboi status yet.....just nobody status...

micron
05-16-04, 10:41 PM
I wish Hellbinder was here so I could shove this all down his throat. He made an appearance in the rage3d thread..but disapeared just as quickly...how convenient.

-=DVS=-
05-16-04, 10:46 PM
Ya NV40 im not flameing you or anything i updated my post before you wrote yours , im curious my self but without pictures showing difference its hard to base opinion on you know..... alot of people comparein 5950-9800-X800 but no 6800 pics as reference either or how should image look at highest rez, hmm i was wondering if ATI drops quality level at higest resolution but keeps it at lower maybe takeing screens from 1024x768 and 1600x1200 would be half decent comparision if AF levels changes however all these so called reviews only based they opinion on benchies :eek2:

ChrisRay
05-16-04, 11:33 PM
I wish Hellbinder was here so I could shove this all down his throat. He made an appearance in the rage3d thread..but disapeared just as quickly...how convenient.


He just posted at beyond3d. If you hurry you might catch him. :screwy:

micron
05-16-04, 11:41 PM
He just posted at beyond3d. If you hurry you might catch him. :screwy:
I did...thank you :)

micron
05-16-04, 11:43 PM
:screwy:
Why did you feel the need to use that emoticon against me?

ChrisRay
05-16-04, 11:45 PM
Why did you feel the need to use that emoticon against me?


I'm just a little bedazzled by the fact that you were curious where he was of all people ;)

micron
05-16-04, 11:49 PM
I'm just a little bedazzled by the fact that you were curious where he was of all people ;)
Well Chris...I try and keep him in my pocket....but.......... :)

Clay
05-16-04, 11:51 PM
Let's all let cooler heads prevail here folks.

micron
05-16-04, 11:56 PM
Of course:)