Hal V. Engel wrote:
> Just curious. Since GEGL has support for more high bit depth formats than
> just 16 bit int/channel how much more work would be needed to support a wider
> range of formats?
Support for 16 bits per channel implemented properly will give us at
least any other RGB format we want more or less for free except perhaps
64-bit floats per channel since the internal processing happens with
32-bit floats. Sure, you could easily process 64-bit float data but
since it has to pass through 32-bit float pipes it's quite pointless.
The patch was pretty thoroughly discussed yesterday on the IRC channel
and the consensus was that we want to give higher bit depths a shot for
2.8. Note that the patch is very rough and sketchy. The purpose of it
was just to show that it is pretty easy to add elementary support for
higher bit depths in GIMP now thanks to the previous of porting of core
components to GEGL.
Gimp-developer mailing list