ATi is ch**t**g in Filtering
- Thread starter Bitpower
- Start date
As i already posted on R3D, this is not about the Texture-Stage Optimization. That one can be switched off, if you use the correct Registry Key - no questions.
BUT, there's another optimization, similar or equal to the brilinear filtering on GeForceFX - which cannot be switched off AFAIK.
BUT, there's another optimization, similar or equal to the brilinear filtering on GeForceFX - which cannot be switched off AFAIK.
Is this -
http://www.beyond3d.com/forum/viewtopic.php?t=12240
what that article is talking about ?
Mark
http://www.beyond3d.com/forum/viewtopic.php?t=12240
what that article is talking about ?
Mark
The drivers detects whether an application is using colored mipmaps. Then the driver disables brilinear filtering.
When colored mipmaps not used, you'll get brilinear all the time.
And even when you have AF set to Application Preference, the application request full trilinear, you still can't get trilinear.
________
Easy vape review
When colored mipmaps not used, you'll get brilinear all the time.
And even when you have AF set to Application Preference, the application request full trilinear, you still can't get trilinear.
________
Easy vape review
Last edited by a moderator:
jimmyjames123
Regular
Should be interesting to see what more comes out of this...
One thing that I did notice in the initial set of X800Pro/X800Xt reviews is that at times the Radeon cards had very little performance decrease when enabling AF, relative to the GeForce 6800 cards.
For instance, compare these two graphs on UT2004, the first with 4xAA and the second with 4xAA/8xAF:
http://www.firingsquad.com/hardware/ati_radeon_x800/page21.asp
http://www.firingsquad.com/hardware/ati_radeon_x800/page22.asp
As you can see, the GeForce 6800U leads with 4xAA, with the NV card 20fps faster than the X800 Pro at 1600x1200. However, when you turn on 8xAF in addition to 4xAA, the lead swings to the X8000XT, and the X800Pro is only 4fps behind the 6800U at 1600x1200. Notice the interesting behavior in frame rates on the X800 cards. The X800 Pro and X800XT only lose about 3-4fps at most when 4xAA/8xAF is enabled vs 4xAA, even at 1600x1200!
One thing that I did notice in the initial set of X800Pro/X800Xt reviews is that at times the Radeon cards had very little performance decrease when enabling AF, relative to the GeForce 6800 cards.
For instance, compare these two graphs on UT2004, the first with 4xAA and the second with 4xAA/8xAF:
http://www.firingsquad.com/hardware/ati_radeon_x800/page21.asp
http://www.firingsquad.com/hardware/ati_radeon_x800/page22.asp
As you can see, the GeForce 6800U leads with 4xAA, with the NV card 20fps faster than the X800 Pro at 1600x1200. However, when you turn on 8xAF in addition to 4xAA, the lead swings to the X8000XT, and the X800Pro is only 4fps behind the 6800U at 1600x1200. Notice the interesting behavior in frame rates on the X800 cards. The X800 Pro and X800XT only lose about 3-4fps at most when 4xAA/8xAF is enabled vs 4xAA, even at 1600x1200!
jimmyjames123 said:
Yup. Only 3-4fps lost when 8xAF enabled, even in farcry.
http://www.firingsquad.com/hardware/ati_radeon_x800/page24.asp
http://www.firingsquad.com/hardware/ati_radeon_x800/page25.asp
Quite a good way to *optimized* performance, isn't it?
________
DODGE CHARGER DAYTONA SPECIFICATIONS
Last edited by a moderator:
Checking differences with AA modes is not necessarily the best test - MSAA is a bandwidth intensive task and AF a fillrate intensive task, so in theory these should balance anyway.
jimmyjames123
Regular
Well, one way or another, it's pretty obvious that the ATI X800 cards are taking much much less of a hit in some games than the NV 6800 cards when moving from 4xAA to 4xAA/8xAF.
FWIW, some websites have also noted that NV's AF algorithm seems to be sharper/clearer than ATI's AF algorithm in some pictures, even though both are now angle dependent.
FWIW, some websites have also noted that NV's AF algorithm seems to be sharper/clearer than ATI's AF algorithm in some pictures, even though both are now angle dependent.
I've just run some images in SS:SE using the DX renderer. If I do an image compare between an 9800 XT and an X800 XT there are differences between the two, however if I do an image compare between the two with mip-map colouring still enabled then there are differences between those as well.
I'd like to know what the methodology for determining that "brilinear" was apparently on.
I'd like to know what the methodology for determining that "brilinear" was apparently on.
DaveBaumann said:
I can give you one A4-sized page of pure-AF Fillrate-Benchmarks with Villagemark in various resolutions, AF Levels and AF methods (bi, TS-opt. and "full-tri").
The result is sometimes much more prominent:
16xAF "forced full-tri" AF (real Full-tri on R360, bri on R420) in 1280x1024 resulted in 72fps for R360 and 110fps for R420 (same fillrate and bandwidth for both).
Quasar,
Does using directx or opengl make any difference ?
I mean, does the *optimisation* happen to both directx and opengl games ?
________
MORPHINE REHAB DICUSSION
Does using directx or opengl make any difference ?
I mean, does the *optimisation* happen to both directx and opengl games ?
________
MORPHINE REHAB DICUSSION
Last edited by a moderator:
There are some filtering differences between the 2 shots, and even alpha blending differences.... my conclusion can only be, there are differences (based on comparing the 2 colour images).
Quasar said:
My point being is that saying there is a different performance drop in different architectures with AA enabled, as Jimmy's example did, isn't a particularly useful example because AA can negate the fill-rate effects of AF. If we then add to that there are large architectural differences to take into account and inexcess of a 9600XT's worth of fill-rate between the two, then I don't find that example particularly useful.
However, the test you are also running doesn't necessarily point to anything because there have been changes in how the texturing is achieved from R3x0 and R420 (this was confirmed by Raja Khodri at the Technology Seminar) and you don't know how the caching and Heir-Z differences will affect things between the two.
However, my earlier point was asking whether you have done similar bit comparisons with mip-map colouring enabled. The article and 3DCenter are claiming that the driver "detects" mip colouring and turns the apparent "brilinear" filtering off - in which case if you did a bit comparison of R360 and R420 with mip-map colour on then there should be no differences, however this does not appear to be the case with the test I've just run.
Similar threads
- Locked