Gamma Shift Upon Rendering?

[Margus Voll]

If you make uncompressed frame how it goes then ?

In my experience with prorez there is some sort of gamma shift.
Uncompressed is more preferable.

—

Margus

https://iconstudios.eu

[Jay Lee]

Margus good thought thank you.
An uncompressed render yields more accurate results than ProRess444 however still with noticeably ‘lifted’ blacks.
Source material is ProRes444.
Have reached out to BM directly. They have been fabulous in the past in responding to issues.

Cheers,

J

[Michael Cinquin]

I haven’t seen any gamma shift rendering to or from prores with DaVinci 7.0.1, 7.0.2 and 7.0.3

I would not judge a gamma shift on quicktime playe. It’s not unusual to find two shots that quicktime player displays differently but that are are the same in Final Cut.

Michael Cinquin

Final Cut Pro – Avid Media Composer editor
DaVinci – Color – Baselight colorist
http://www.michaelcinquin.com/tools : tools for FCP | Color | RED | subtitles | Cinema Tools | Timecode – Keycode calculator

[Jay Lee]

Agreed Michael…..we’ve been on that wild goose chase in the past.
When the resolve render QT is laid over the source QT in After Effects in ‘Difference’ mode there is no doubt a luminance shift of some sort.
The fact that different codecs are yielding different results is a red flag in its self.
See below.

Cheers,

J

Yellow for viewing purposes only.

resolvedifference.jpg

[Margus Voll]

I did today some stabilizing shots with another software and prorez.

All renders come out some what “brigther”.

I have not looked into it with Resolve but generally i’d say
prores shifts a lot.

Always go uncompressed when possible 🙂

—

Margus

https://iconstudios.eu

[Jay Lee]

Margus is there a process where by for this particular project and its looming deadline we could simply substitute the current source ProRes media (imported as one quicktime and scene-detected) for the exact same file but uncompressed with out loosing days of tracking, custom mattes and grades?

I don’t believe that the way Resolve handles ‘grouping’ that there is a simple way to add a corrective node to all grades to compensate for the luminance shift.

cheers

j

[Margus Voll]

Fast, i would think of is to reconvert initial prorez to something that is mov but has different compression.
Like uncompressed tif etc as a compressor, if you think replacing initial file is a problem there.

Make backup from your original file 🙂 so you will not overwrite it.

I have not tested it like this but generally just replace original mov with other mov in media pool or
on your drive.

Corrections etc should be the same as you only change source codec.

In the other hand if you see that the result only gets different then just render out in less “shifting”
format.

Did you get better results with uncompressed source or output. Did you make small test ?

—

Margus

https://iconstudios.eu

[Dan Moran]

I don’t believe that the way Resolve handles ‘grouping’ that there is a simple way to add a corrective node to all grades to compensate for the luminance shift.

cheers

j

If you select track mode on the top of the node graph this will add a node across every clip in your timeline easily. I use it a lot for altering gamma for outputs going to different formats i.e. tape/filmout/youtube etc…

Dan Moran
Application Specialist
Blackmagic Design EMEA

[Jay Lee]

That is a ‘gem’ Dan…..thank you.
This will get us delivered while this is sorted out.

Cheers,

j

[Blase Theodore]

Please someone correct me if I’m wrong..

As I understand Prores444 exists as both a natively RGB codec and a natively YUV codec. Basically its a wrapper for 2 different sub-codecs in different color spaces wrapped in a mysterious black box.

I’m not really sure who thought that would be a good idea, but I refuse to work with the format for the time being. I’m not convinced that anything other than direct apple products understand the codec well enough to avoid treating it incorrectly as YUV or RGB.

If someone has a better way of understanding this than trial and error, I’d welcome the insight.