Chat Transcript: ATI's texture filtering algorithms

[FUDie]

Get a clue. If you expose the cap, you must pass the test.

If you said...

What's your point? Do not NVIDIA and ATI expose linear mip filtering? Do I have to spell out everything for you?

-FUDie

 

[croc_mak]

Maybe we should be benchmarking with ATI X800 cards set to all the way to performance against the 6800’s Brilinear.

NVIDIA's current settings are not very bad at all - they have considerably improved since their initial introduction.

Dave,

Have you actually checked the quality on ATI cards set all the way to performance? If so, I take it that you are implying that Nvidia's default is better than ATI's performance setting?

 

[dlo_olb]

As they say, trilinear is a compromise between "too blury" (the blending between mip levels... blending the right word?) and "too sharp". I don't see the obsession with trilinear, it's a compromise just like everything else in graphics, and if it can be improved upon, why not? Trilinear is NOT the best filtering method in EVERY situation, so why apply trilinear to everything? What is it with this love of the "one-size-fits-all" solution? People rant about relying on global CP settings, but when ATI come up with a solution for something that adapts to the situation, people complain about that.

I think people are just upset they didn't receive a personal e-mail from ATI telling them what they did.

u simply misrepresented the complaint...
people didnt complain ATI for coming up with a solution for something that adapts to the situation, they complain they wasnt informed of such "adaptive" solution and so unknowingly still assuming they are having full trilinear filtering when it is not...

btw. such "optimization" is fine if they make a note of it beforehand...

i hope u are upset not because it is ati ...

himomo!

 

[dlo_olb]

…its a cheat cuz they tried to hide it….

Because they didn’t file a report to let you know about their new Adaptive-Trilinear it’s a cheat ….??? ….. Like I say …ridiculous. Are ATI suppose to file a report with us every time they develop a new technology?

ATI hasn’t tried to deceive anyone about anything. And ATI didn’t try to hide it, it looks like they just didn’t want tell anyone about it. And I can see why now for good reason . ATI has developed a great new Adaptive-Trilinear technology that works as good or better than the old Trilinear and improves performance at the same time. Why would they want to blab about it and let their competitors know about it?. The likely reason ATI never told anyone about this new Adaptive-Trilinear technology is for competitive reasons.

thats the same way why brilinear is a cheat for nvidia... Are nvidia suppose to file a report with us every time they develop a new technology?

not necessary, they can simply be done by adding a new option saying "adaptive trilinear", dont masquerade as legacy full trilinear filtering when it is not..
they gonna let nvidia know anyway if they patent it... so thats not a reason or excuse...

is the adaptive trilinear equal or better than the leagacy full trilinear filtering (as told by the ATI engineer)???
time will tell...!!!

himomo!

 

[dlo_olb]

Nvidia hasn’t tried to deceive anyone about anything. And Nvidia didn’t try to hide it, it looks like they just didn’t want tell anyone about it. And I can see why now for good reason . Nvidia has developed a great new Brilinear technology that works as good (with significant IQ difference) as the old Trilinear and improves performance at the same time. Why would they want to blab about it and let their competitors know about it?. The likely reason Nvidia never told anyone about this new Brilinear technology is for competitive reasons.

Shall we continue this or stop here CyanBlues?

nvidia probably forget to inform us, u know we should not expect them to inform us of every technogology change.... and they should have probably announced they are patenting it when get caught...

 

[dlo_olb]

I am impressed with ATI's attitude about the whole subject. Nvidia didn't bring in senior engineers to account for optimizations in their hardware/drivers.

I can see ATI's point. Just because you use a different algorithm to eliminate unnecessary work doesn't mean you can't produce the same or better quality output.

it is nothing more than a show, regurgitate the prepared PR... dodging questions conveniently.... tho i appreciate ati's effort for trying to stage the show...

so the point is we have to prove whether this adaptive filtering equals or better than the original one.... and this is one yet to be proved...

himomo!

 

[dlo_olb]

Nvidia hasn’t tried to deceive anyone about anything. And Nvidia didn’t try to hide it, it looks like they just didn’t want tell anyone about it. And I can see why now for good reason . Nvidia has developed a great new Brilinear technology that works as good (with significant IQ difference) as the old Trilinear and improves performance at the same time. Why would they want to blab about it and let their competitors know about it?. The likely reason Nvidia never told anyone about this new Brilinear technology is for competitive reasons.

Shall we continue this or stop here CyanBlues?

significant to who? some people say they can't tell the difference, just cuz its significant to you doesnt mean it is to them. what ever happen to the argument that nvidia cheats cuz you cant disable their brilinear, and now its all okay? i dont know about you but when i get my x800xt i would still like the option to have full tri.

as is now we cant disable "adaptive" trilinear filtering either and we didnt told of this feature before... the fact ati "rectify" it later once got caught doesnt mean he wasnt cheating.... ati is still cheating unless its quality equal or better than the original...


remember:
ATI has consistently encouraged use of tools for IQ analysis that don't show the effects of ATI's adaptive method.
why the need to do that if the benefit/quality is maintained in all cases??? isnt it a good hint of cheating...?

himomo!

 

[Althornin]

remember:
ATI has consistently encouraged use of tools for IQ analysis that don't show the effects of ATI's adaptive method.
why the need to do that if the benefit/quality is maintained in all cases??? isnt it a good hint of cheating...?

himomo!

oh my, this shows you dont understand what is going on here at all!

The very fact that color mip maps DO display with full trilinear speaks very GOOD things about the adaptive nature of the algorithm.
you seem to think this is like brilinear - ie, a constant reduction in quality.
IT IS NOT. It is adaptive, and will adjust itself when nessesary - like a case where each mip map level is totally different from the one before...

 

[Kombatant]

it is nothing more than a show, regurgitate the prepared PR... dodging questions conveniently.... tho i appreciate ati's effort for trying to stage the show...

If it were a PR show, they would have sent PR folk to answer, not engineers.

 

[christoph]

Re: whql test

Read that however you want, but I know there is a trilinear test as I have seen it myself. The texture filter test runs through all combinations of exposed min/mag/mip filters with all exposed texture formats.-FUDie

could you please give me quote or test number from the specs? maybe i missed something, i couldnt find the word trilinear a single time in them.....so if this tests exist, they dont seem to be documented in the specs for me.

 

[davepermen]

some people are just braindump..

 

[tahrikmili]

I have to say I found the chat to be informative and full of reasonable answers to even flame-loaded questions at a few points. I admit I'm not very fond of the idea of companies changing filtering methods in hardware generations without announcing the change first, the answers and the evidence available on the net so far have persuaded me that their adaptive filtering algorithm does not sacrifice from IQ while it increases performance. I just hope they announce such changes prior to implementation next time to eliminate such silly fuss.

Yalaz Ozkanli

 

[Dave Baumann]

Re: whql test

could you please give me quote or test number from the specs? maybe i missed something, i couldnt find the word trilinear a single time in them.....so if this tests exist, they dont seem to be documented in the specs for me.

Did you search on "Trilinear" or "Linear"?

 

[no_way]

Because they didn’t file a report to let you know about their new Adaptive-Trilinear it’s a cheat ….??? ….. Like I say …ridiculous. Are ATI suppose to file a report with us every time they develop a new technology?

They used to. They used to do such nice things like announce new technologies, expose methods for using it in API and document the behaviour. This once common practice brought us SGIS_texture_filter4, EXT_texture_compression_s3tc, 3DFX_texture_compression_FXT1
etc etc.
Wonder why they did that .. why wouldnt the respective IHVs simply silently compressed the textures in their drivers ( maybe even analysing the textures for suitability ) and simply provided their new wonderful technologies for end users ...

 

[DoS]

it is nothing more than a show, regurgitate the prepared PR... dodging questions conveniently.... tho i appreciate ati's effort for trying to stage the show...

If it were a PR show, they would have sent PR folk to answer, not engineers.

The issue bothers mainly gfxcard-happy geeks...
Wha'ts the best PR move when you have such a audience ? sent the PR folks to answer the questions or send a couple of top-engineers who design the stuff the geeks have wet dreams about ?

nuff said

 

[davepermen]

Because they didn’t file a report to let you know about their new Adaptive-Trilinear it’s a cheat ….??? ….. Like I say …ridiculous. Are ATI suppose to file a report with us every time they develop a new technology?

They used to. They used to do such nice things like announce new technologies, expose methods for using it in API and document the behaviour. This once common practice brought us SGIS_texture_filter4, EXT_texture_compression_s3tc, 3DFX_texture_compression_FXT1
etc etc.
Wonder why they did that .. why wouldnt the respective IHVs simply silently compressed the textures in their drivers ( maybe even analysing the textures for suitability ) and simply provided their new wonderful technologies for end users ...

because texture compression in a GOOD way takes long, and thus, texture uploading would get very slow, wich would NOT be liked by anyone. the other solution would be to compress in a bad way (not analizing the image good, quantize without taking the whole data-set into account, etc), wich would result in very bad image quality.

if it could be done without hurting at all, it would've been done.

 

[suicuique]

Singer
Is this really trilinear filtering?

Andy/Raja
Yes, It's a linear function between the two mipmap levels based on the LOD.

LIE LIE LIE LIE

Their is a conditional step in their algorithim it is neither mathematically linear nor is it A*X+B*(1-X) ( which is mathematically linear ).

It is
If X < Threshold return A;
else X > 1-Threshold return B;
else return A*([X-Threshold]/[ 1 - 2*Threshold] )+B*([1-X-Threshold]/[ 1 - 2*Threshold])

Or something close to it.

In any event, it is still piecewise linear.

What is piecewise linear about "If X < Threshold return A" ?

As you may know, constant functions (though representable by a "line") are not considered linear.

So a piecewise fuction of constant and linear functions, is *not* piecewise linear in any means of the word.

/nitpicking mode off

regards, alex

 

[Kombatant]

it is nothing more than a show, regurgitate the prepared PR... dodging questions conveniently.... tho i appreciate ati's effort for trying to stage the show...

If it were a PR show, they would have sent PR folk to answer, not engineers.

The issue bothers mainly gfxcard-happy geeks...
Wha'ts the best PR move when you have such a audience ? sent the PR folks to answer the questions or send a couple of top-engineers who design the stuff the geeks have wet dreams about ?

nuff said

If you think that there are no PR people with good hardware knowledge, you are mistaken.

 

[Kombatant]

Singer
Is this really trilinear filtering?

Andy/Raja
Yes, It's a linear function between the two mipmap levels based on the LOD.

LIE LIE LIE LIE

Their is a conditional step in their algorithim it is neither mathematically linear nor is it A*X+B*(1-X) ( which is mathematically linear ).

It is
If X < Threshold return A;
else X > 1-Threshold return B;
else return A*([X-Threshold]/[ 1 - 2*Threshold] )+B*([1-X-Threshold]/[ 1 - 2*Threshold])

Or something close to it.

In any event, it is still piecewise linear.

What is piecewise linear about "If X < Threshold return A" ?

As you may know, constant functions (though representable by a "line") are not considered linear.

So a piecewise fuction of constant and linear functions, is *not* piecewise linear in any means of the word.

/nitpicking mode off

regards, alex

Ahh... mathematics Allow me to play too. Linear functions have the form of:

f(x) = a*x + b

and we call them linear because a) they are a straight line b) the 'x' is in the power of 1.

So, as you can see, if you have a=0 and b=A, you have the function:

f(x) = A

That function is a subcase of the linear functions. So it is STILL a linear function

edit: Here's one of the links that google got me, I am sure you can find a lot more : http://people.hofstra.edu/faculty/Stefan_Waner/RealWorld/tutorialsf0/frames1_3.html

 

[Evildeus]

Yeah f(x)=b is a linear fonction.