> I think many people would be thankful if you do. I have a hacked-up
> quantizer based on Paul Heckbert's median cut algorithm + heavy duty
> dithering (originally written by Bill Davidson, I think) that performs
> fairly well as far as visual appearance goes, but it just takes too long
> to do the job. Do you have any ideas/hunches about the relative
> performance of octree algorithm and median cut ?
It runs as well as or maybe a little better than the quantizer in Xv.
There was a comparative study done by the author(s) of the algorithm
in GG and the octree seemed to do pretty well, if I recall correctly.
I'll take a look when I get to the office on Monday.
spl
To unsubscribe, send the message "unsubscribe" to
xforms-request@bob.usuf2.usuhs.mil or see
http://bob.usuf2.usuhs.mil/mailserv/xforms.html