Re: [info-performer] ClipTexture mipmapping algorithms?

Date view Thread view Subject view Author view

From: Paolo Farinelli (paolo++at++sgi.com)
Date: 07/15/2003 15:47:17

Hi,

By default, Performer selects the GL_LINEAR_CLIPMAP_LINEAR_SGIX min filter
for cliptextures on IR. (nearest_nearest, linear_nearest, nearest_linear
are also available).

What this effectively means is that you are tri-linear filtering the clipmap
between levels (using the fractional LOD bits).

Please take a look at the documentation in /usr/share/Performer/doc/clipmap
for more info on the 'jello' and 'wobble' bugs Allan mentioned.

Regards,
Paolo

Allan Schaffer wrote:

>Hi Curt,
>
>Curt Schroeder wrote:
>
>
>>OK, to the best of my understanding, ClipTexture mipmapping uses a
>>bilinear mipmap algorithm. Is it possible to replace the Performer
>>provided algorithm with a better one (trilinear for instance)?
>>
>>I would like to reduce some of the high frequency "crawl" artifacts that
>>are visible when flying over my high resolution (1 m/pixel)
>>ClipTexture. I'm currently using Perfomer 2.2.13 in conjunction with
>>MultiGen-Paradigm's Vega 3.6.1 on an Onyx2/IR3 platform.
>>
>>
>
>I am talking a little out of my domain but I think the "crawling" may be
>something else; it's something that's come up before, look for "jello"
>and "wobble" in the mailing list archives.
>
>To answer the question directly: the sampling algorithm is done by the
>hardware, so it can't be replaced; however it is my understanding
>(subject to correction) that it already uses a trilinear algorithm when
>sampling the levels in the clip stack.
>
>Allan
>
> 

-- 
Paolo Farinelli                                           paolo++at++sgi.com
Member of Technical Staff, OpenGL Performer              1-650-933-1808
Silicon Graphics        1600 Amphitheatre Pkwy, Mountain View, CA 94043

Date view Thread view Subject view Author view

This archive was generated by hypermail 2b29 : Tue Jul 15 2003 - 15:47:21 PDT