Never mind what I just said - it does matter what order the transforms happen
in, just like in a 3D transform.
This should probably be a feature request to add a transform order control to
the Colorspace node much like an Axis node has.
-jonathan
On Jun 7, 2012, at 9:26 AM, Jonathan Egstad wrote:
> That should not be necessary - you should be able to do it in one operation
> using a single Colorspace node as the gamut remap and sRGB->XYZ will
> concatenate into the same 3x3 matrix linear transform.
>
> If you're doing something non-linear in the colorspace node like changing the
> gamma response of the data then then application order can be important.
>
> So this:
> Colorspace {
> primary_in sRGB primary_out DCI-P3
> colorspace_in sRGB colorspace_out CIE-XYZ
> }
>
> Is equivalent to:
> Colorspace {
> primary_in sRGB primary_out DCI-P3
> }
> Colorspace {
> colorspace_in sRGB colorspace_out CIE-XYZ
> }
>
>
> -jonathan
>
> On Jun 7, 2012, at 7:59 AM, John Mateer wrote:
>
>> Thanks for this. As per the other thread, I think I've found the solution --
>> use two Colorspace nodes, the first to change the gamut from sRGB to P3 and
>> the second to change the color space from Linear to CIE-XYZ. That seems to
>> do the trick.
>> _______________________________________________
>> Nuke-users mailing list
>> [email protected], http://forums.thefoundry.co.uk/
>> http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users
>
> _______________________________________________
> Nuke-users mailing list
> [email protected], http://forums.thefoundry.co.uk/
> http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users
_______________________________________________
Nuke-users mailing list
[email protected], http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users
