Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!
Bar81 said:
That's not the point of the article. The point is that ATI states that Trilinear Filtering is on by default, however, the truth is that it's only on when asked directly by the game to be on. Thus, Brilinear Filtering is on by default.micamica1217 said:
nice try...
the tech report has not shown that the new ATI x800 cards are not doing triliner filtering....infact, they show nothing.
nothing but talk....no wonder you show a link to it.
mica
CPL.Luke said:well it sounds like ATI uses full trilinear filtering when its needed (the game askes for it) but in many cases if you have a texture directly facing you you barely need any filtering on it according to ati. so the don't filter it. this saves performance and gives you the same image quality
xtrmeocr said:
what your talking about is ATI's "angle adaptive" anisotropic filtering. thats not what this thread is about. they have been doing that since the original r300.(maybe even r200 i'm not sure)
Pake said:
That's not the point of the article. The point is that ATI states that Trilinear Filtering is on by default, however, the truth is that it's only on when asked directly by the game to be on. Thus, Brilinear Filtering is on by default.
BTW, something weird about that sentence... kinda reminds me of someone else's words in two other threads... "nothing but talk"... hummm.
Bar81 said:So, in the end I'm right (and in case that doesn't register, conversely, you're WRONG.) I would trust John Carmack's opinion over anyone on these boards every day of the week and twice on Sunday. It's a cheat and you can see the obvious difference in quality here as well as Carmack outing ATI and the ATI Fanboys:
http://www.tomshardware.com/graphic/20040603/ati_optimized-07.html
Cowboy X said:Bar81 Please look here :
http://www.ocforums.com/showthread.php?t=302192
At the time very few people outside of ATI understood the method and thus the coloured mip map performance loss looked like a "cheat" . Since then however that has in my opinion been cleared up until proven otherwise .
I always believed that a graphics card is suppose to render a game the way the game developers made it. If he's calling it a cheat, then it's a cheat. They are the judges of quality for their product. Put aside the reviewer's comment and judge it by how the game developer wants it to be judged.This is indeed a "cheat" that both major vendors now do. Instead of always sampling the two adjacent mip map levels and doing a full blend between them, they have plateaus where only a single mip level is sampled, reducing the average samples from 8 to about 6.
It is actually a pretty sensible performance enhancement, with minimal visual issues. However, having drivers analyze mip map uploads to hide the cheat is an unfortunate consequence.
The colored mip map option in Q3 should have absolutely zero performance impact in the absence of performance options like this.
John Carmack
All comes down to the game developer to give the last word. Carmack calls it cheating on his engine, plain and simple. I call it some damn good optimizations, but when playing his game that uses his engine, it's cheating.Cowboy X said:You are missing the point , where is the cheating in the games which we all play with coloured mip maps on ????? I don't recall playing any of them . Different methods of filtering will always produce different images slightly from generation to generation and manufacturer to manufacturer . So the pixel difference screenshots in this case mean nothing if the mip map borders are being properlyy blended . Again properly blended in gameplay . I'm still waiting for it to be shown doing anything different .