Nuke working space

Hi everyone,
I’m in the process of setting up compositing for a feature film. Every previous ACES show I’ve worked on has used ACEScg as the internal working space but I was wondering if there are downsides to using a different working space?
Our plates are delivered as 10bit DPX Alexa LogC and when I read them in and they are converted to ACEScg working space I get some negative values for things like highly saturated blue LED lights in the scene (negative in red and green channels). When I switch to ARRI Linear ALEXA Wide Gamut for working space I get visibly the same result but no more negative values. The values are handled properly.
Since everything we comp into the plates should have similar values as what’s in the plate is there any downsides I can’t think of to using an ARRI Linear ALEXA Wide Gamut working space?

Shane Rabey

I’ve just seen that a lot is answered in the “Color primaries shifting and Negative values (OCIO) confusion” thread re the negative values.
That just leaves the question of using working spaces other than ACEScg. Is there a downside to using Alexa Wide Gamut? I see that the primaries for AWG are outside even AP0 on the CIExy diagram.
We will be delivering to client in Alexa LogC.

Comping in Alexa Linear is pretty much what we have been doing for years before ACES was around.
At the time, I personally didn’t even know about color primaries so we often ended up comping linear sRGB matte paintings and CG on Alexa Linear footage.
Nobody ever cared, and we never got a shot kicked back because of it.
Now we could convert the CG to Alexa on import without aces, but I’m not even sure it would necessarily look better.

The technical issues that can arise from working in different color gamuts are described here:

You are very unlikely to see these issues in comp. We often run shows in Alexa Wide Gamut or wider without seeing any obvious issues. Since your plates were converted into AWG you shouldn’t see any extreme negative values when working in that color space.

1 Like