We have a mov file for a shot on Alexa Mini, LOG-C, Prores 4444 XQ, EI800, We are importing it in Nuke and converting it to ACEScg as Alexa V3 LogC EI800.
When we look through a Rec.709 (ACES) viewer the image looks overexposed compared to a linear project converting the file to linear as Alexav3LogC and looking through a AlexaV3_K0S1_LogC2Video_Rec709_EE_nuke3d.cube LUT.
I understand that the Rec.709 (ACES) viewing LUT shows a higher range than the standard Rec.709 one but the picture look considerably different.
The same happens in Davinci Resolve.
Has this happened to anyone before? Is this just what it is expected?
The ACES RRT + ODT produces a slightly higher contrast image than the ARRI LUT, but if you are seeing an exposure difference, something has gone wrong in your pipeline.
Hello,
Wondering if you have the right IDT for Alexa
mov is a video file
IDT needs to be assigned manually in Nuke and DaVinci
Colors can be converted correctly
In terms of my current ACES project experience,
During the ACES process
The VFX department should not use original material for ACES by itself,
Let the DI department
Transcode video material to ACES AP0 and export EXR file
Hope to help you
Just double checking, you are converting your linearised AWG image back to LogC before applying the LUT, when using the legacy Nuke approach, aren’t you? I assume so, or it would appear completely wrong.
Thank you for the examples and the input. Also thank you, Dean.
Yes, I’m converting the linear AWG to LogC before appling the LUT.
Looking at your references I believe we are following the right process. Using the Arri reference files I get the same result as you.
The weird thing is that doing the same to our Apple Prores 4444 XD that change of contrast it’s pretty much one stop higher in the ACES script than the linear one.
Would you like to get few sample files and my script to see what I’m talking about? I would really appreciate it, but if so it would have to be after the project it’s delivered and aired.
