Please specify that a negative UNPACK_ROW_LENGTH is allowed for PixelStore. This is very useful when loading top-to-bottom saved images.
(PS: please remove selection from Pure OpenGL 2.0! And I'm a fan of SGIS_generate_mipmap, as it can be done in hardware asynchronously.)
"If you can't afford to do something right,
you'd better make sure you can afford to do it wrong!"
You mean dropping the selection buffer ? What's wrong with it ?It's saved me many times from having to write a raytracer just to find out what's at a pixel.
Yes, I mean dropping the selection buffer. It's okay as a non-pure part of OGL2.0, but I don't think it should be in the "pure" subset.
Do I have to fill out the idea assignment paperwork and fax it to 3dlabs to make sure they will consider the suggestion of making UNPACK_ROW_LENGTH allow negative values?
"If you can't afford to do something right,
you'd better make sure you can afford to do it wrong!"
The feedback form is there mostly to protect you. Basically it says to not tell us any ideas / IP you have that you consider confidential, because they might wind up in the OpenGL 2.0 spec. If you're posting those ideas in a public forum like this one you obviously don't have those concerns :-)
As for negative unpack row lengths, we'll take a look at it.
Barthold
3Dlabs
Hm, I just have this feeling that "unpure" may be synonymous with "deprecated" real soon and that would suck if the only thing in the pure part is the cutting edge stuff ?
Posting Permissions
