Randy: Ocio should help rectify this and it will be included with a future
version of Nuke. You can download it here:

http://opencolorio.org/downloads.html

I'm not sure if the one for download is precompiled with the Nuke nodes or
not.  If not it is fairly simple to compile.

-deke

On Saturday, December 31, 2011, Jonathan Egstad <[email protected]>
wrote:
> Don't overthink the Read's 'colorspace' conversion - it can only do a 1D
lut applied to RGB equally.  It's not a colorspace conversion per-se, only
a gamma response conversion.
>
> -jonathan
>
>
> On Dec 31, 2011, at 1:34 PM, Randy Little <[email protected]> wrote:
>
>> Ok I am calibrating a Camera and light meter.  shooting a grey card (
>> a good one that I know is 18%)   I take this file red it into anything
>> that reads R3d. and in every case but nuke grey is .18xxxxxxx in float
>> and 127.xxxxxxx or 128.xxxxxx in 8 bit.   huh.    in nuke its .21% ok
>> weird.  Make file in photoshop in sRGB 2.1 make middle grey 50% (cause
>> its gamma corrected sRGB)  Take that tiff file into Nuke .21 AGAIN.
>> WHAT?  ok set read node to RAW.  MORE WHAT.  It reads at .5 as I would
>> expect.  ok I studied color a little.   This sounds like color space
>> conversion matrix problem.  Yup if you use LAB 50% then all is great.
>> But None of my files are LAB and in RGB they are fine.  So it seems
>> nuke conversion going through LAB remapping RGB 50% to LAB 50% and
>> then not accounting for that.    But also if I use a Gamma node set to
>> .402  I can get in a gradient .18 to be exactly the middle of the
>> gradient as well.  This in tern doesn't distribute the values linearly
>> threw the gradient either at 10% translation I have only 7%
>> illumination change from dmax and from dmin 20% translation results in
>> 10% change.  As expected when using a gamma node on a raw sRGB since
>> srgb is a power curve.  Just nukes Power curve seems a bit off.
>> Is there a reason for this?
>>
>>
>> Randy S. Little
>> http://www.rslittle.com
>> _______________________________________________
>> 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

Reply via email to