On Mon, 5 Nov 2012 12:55:14 -0500
Monty Montgomery <[email protected]> wrote:

> Linear-space float RGB with black point at 0. and white point at 1.
> suits me fine, BTW.  It's conceptually the simplest/most correct
> option.  I'm just worried about practicality issues I'm too dumb see
> until I get there.

Any reason for not using 8/16/32 bit as a fraction (0-1)? So 8 bit would
have a resolution of 1/255, 16 bit of 1/65535 etc. All operations can
be done with 4 byte integer math.

Multiplying cannot cause overflows this way. Easy resolution change, too.
Lookup-tables can be used if necessary - a near impossibility with floats.

Possibly great speadups as integer ops can be executed in parallel,
either in the CPU or the GPU.

Just an idea.
John

_______________________________________________
Cinelerra mailing list
[email protected]
https://init.linpro.no/mailman/skolelinux.no/listinfo/cinelerra

Reply via email to