Linear EXR to cineon-log lut?

[Gunleik Groven]

Hi

Need to have a linear to log LUT for a bunch of EXRs that come in...
Thought that was built into Resolve, but: Oh Boy was I wrong...

How do you guys resolve this?

Cheers

Gunleik

 

[Nick Shaw]

If the EXRs contain float data above 1.0 (which is generally the reason for using EXR) it may not be possible to do the conversion with a LUT. If the data is in the 0-1 range the attached LUT should do it. Otherwise you may want to try batch converting the EXRs to Log DPXs (preferably 16 bit if Resolve supports those) using e.g. Nuke.

 

[Gunleik Groven]

Thanks a bunch!

Will try!

Cheers G

 

[jake blackstone]

try batch converting the EXRs to Log DPXs (preferably 16 bit if Resolve supports those) using e.g. Nuke.

Resolve supports 16 bit DPX.

 

[Nick Shaw]

Resolve supports 16 bit DPX.

Thanks Jake. I can't run Resolve on my system to test, as there is a conflict with my Baselight.

 

[Graeme Nattress]

I'd probably use the standard lin2log node in Nuke - I know that math is correct. You may also be able to do it in shake with it's float support and it's lin2log.

Graeme

 

[Nick Shaw]

I'd probably use the standard lin2log node in Nuke

Yes, or the the Colorspace node, with in=linear and out=Cineon. The LUT I posted is just that transform converted to a cube. But as I said, it is limited to the 0-1 range of the linear input. But then I have seen an awful lot of EXRs which were clamped to 0-1 due to people not understanding float workflows, and using inappropriate tools.

 

[Graeme Nattress]

Lut's that clamp are a pain.... But they're not hard to code for full float range if you know what you're doing, but I don't know much that works that way. Maybe in the future....

Graeme