Re: MMX speedup for Floyd Steinberg error diffusion

Terje Mathisen <spamtrap@xxxxxxxxxx> writes:
rep_movsd wrote:
Phil, you're absolutely right, and that's why "doing it right" is even
harder than you're showing here:

There is this color quantizer/ditherer program called scolorq which
uses some highly advanced techniques with floating point precision. I

I.e. convert each pixel to fp, then to a linear color space, do all
error diffusion in this format, then use the final palette colors?

tried that out and it took an astonishing amount of time for a single
frame ( I let it run for 3 minutes before giving up in disgust ).
Turns out it has 5 deep nested loops. Some site online shows pretty
impressive results for reduction to 16 color images. Perhaps the

You might be able to do 16-color images by making the color diffusion
are significantly larger: Keep luminance info as close as possible,
but allow color to be spread across a larger area?

You might be able to just do some kind of octree algorithm.
You could even hard-code the partitions.

(Octree's also possible in 256-colours too, of course. You can
even create a slightly optimised palate by using 2 passes and
crudely bin-counting on the 3 axes.)

Phil
--
Dear aunt, let's set so double the killer delete select all.
-- Microsoft voice recognition live demonstration

.