Linear arriwide exr sources with resolve

Discussions - Using ACES Post (DI, Edit, Mastering)
1

Hi everybody,

I have linear arriwide exr sources, how would you manage the input in Resolve in order to get a proper aces workflow ?
With ocio, I would have the option to declare sources as linear arriwide (it would compensate arri gamut only as footage is already linear), but in resolve the only input available is alexa, which compensates arri gamut but also linearize the logC curve.
Thanks

Vanessa

2

If your files are already linear ARRI WideGamut, then one way you should be able to convert them to ACES is to use a Colorspace Transform node in the color tab. Specify the input color space as ARRI WG and Gamma as linear. Then set Output Color Spaces to ACES and Output Gamma to linear (or ACEScct if you’re managing color manually and not using the Color Management > ACES options).

Others might have other advice.

1 Like
3

Thanks Scott.

If I understand.
My projet is on ACEScc on the color management.
I don’t use IDT on the Media (I don’t understand that we can’t select arriwide Linear), the color transform will do the IDT.
I use a color space transform with input Arri_alexa Lin to ouput Aces Linear.
And if I have a lut (arriwide to rec), I must do a node before this color space transform?
Isn’t it?

Vanessa

4

I do not believe that what @sdyer describes will achieve the correct result. I just tested, and it doesn’t appear to.

The Color Space Transform will be applied in the working space, which if no Input Transform is selected will be linear AWG, converted as if it were ACES2065-1 to ACEScct. So it will have an ACEScct transfer function, and incorrect primaries. Then an AWG to AP1 matrix will be applied to that linear data.

You will get close if you apply two transforms in two node, first an OFX ACES Transform going from ACEScct to ACES, to undo what has happened on the way in, then a Color Space Transform from ARRI Alexa Linear to ACES (AP1) ACEScct. But this will still be missing the chromatic adaptation that is included in a proper IDT.

The best solution would be a DCTL IDT, consisting only of the chromatically adapted AWG to AP0 matrix.

If you are using the Resolve 17 beta, you are able to add custom IDTs using DCTL to the IDT drop-down.

2 Likes
5

Ah, thanks for testing @nick - I knew the tools are there but I can never remember when things are in the working space or not (IMO, it’s not the most intuitive and to the best of my knowledge its not documented anywhere - it’s hard to know the inner workings without testing with known values)

I had forgotten that Resolve 17 lets the user add custom IDTs, which is by far the simplest way to achieve what you are looking to do @vanessa.lafaille07

6

Hi Vanessa!

I think there could be a simpler way. But, first you need to get our of color managing in ACEScc. You’ll have to use a group instead.

  1. Put all you clips in a group.
  2. I the pre group (I think) you could simply reverse the LIN to LOGc. Add a COLORSPACE TRANSFORM OFX and go from Arri Wide linear to Arriwide Logc.
  3. Still in the pre group : add a ACES TRANSFORM OFX (after the COLORSPACE TRANSFORM). in: ALEXA out: ACES (you should stay in linear here).
  4. Add another NODE and add the FIX HIGHLIGHT DCTL.
  5. And another node and again ACES TRANSFORM in: ACES out : ACEScct
  6. in the post group : add another ACES TRANSFORM in: ACEScct out: to whatever ODT your are grading to.

So in between the pre and post you are in ACEScct and good to go for grading. BTW this ia always how I deal with ACES as it’s much a much clearer way to use it in Resolve. It takes ACES out of the regular image path of Resolve. That way transitions and titles are correct. And, within my grades I can use ACES TRANSFORM to gto from ACEScct to Linear to 709, P3 or whatever. This makes certains OFXs react properly etc…

Hopefully… This is clear.

Thanks!

7

Hi!

I tried, but the result was different bewteen the nuke and the resolve. Very strange.
The OCIO inside the nuke was correct because we had the same result with the reference from color grading.
I will explain the detail on a futur message.
Thanks

8

Is it a huge difference? Send some screenshots. It might just be the fix highlights that I told you to add.

9

You could also try this to retain ACEScc color management with IDT set to None and use 3 OFX node in some kind of Pre-Group setup like @chuckyboilo recommended.

  • OFX ACES Transform ACEScc to ACES
  • OFX ColorSpace Transform ARRI Alexa / Linear to ARRI Alexa / LogC
  • OFX ACES Transform Alexa to ACEScc

Totally agree custom IDT would be the way for Resolve 17 though, also worth mentioning the Fusion tab support OCIO but I didn’t get proper results trying something similar to the above, not quite sure why.

Edit: got thrown off by Resolve 17 now applying ToneMapping by default for the ColorSpace Transform OFX, so obviously you should disable that for this workflow.

10

Hi everybody,

sorry for opening this topic up again, but I have a few additional questions here.

To get AWG linear EXRs into ACES I was using a custom IDT (applied as DCTL to the Media Pool Clip) to just map from AWG to AP0 and set the IDT to bypass. This worked great, but as I already had applied this DCTL to the Clip, I couldn’t use the Fix Highlight LMT anymore that I was using before on problematic Alexa Clips.

I played around a little and just went for a “two-in-one” approach by putting both matrix conversions (first the Highlight Fix Matrix then the AWG to AP0 Matrix) into one DCTL and it worked pretty well. It was giving me about the same shift in color as using the LMT alone on the LogC source and removed all artefacts as planed.

I now just wanted to know if this might be a dangerous thing to do, as the input values for the AWG to AP0 matrix are changed. Did anyone try a similar approach and has some bad examples?

Thanks,
Sebastian

11

As far as I’m aware the Blue Light Fix LMT is superseded by the reference gamut compression.
I would inherit this workflow and keep the IDTs solely as IDT.

12

Nice read, thanks for the link! As we are on 17.2.2 at the moment we will look into a update beginning of the year then. I also didn’t like the approach but for some shots it was the only option to get rid of the artifacts…I will definitely give it a test on those problematic shots we had to deal with to see how the RGC performs.

13

If you are on an earlier version of Resolve which does not have a native RGC implementation, you can always try the stop-gap DCTL implementation:

14

Thanks a lot! Looking forward to play around with this DCTL as well and give it some testing.

However, the intention was to remove any fixes from the nodetree to make shure that the colorist can just copy grades between shots without the need of adding an extra node for the EXRs. As I understand by looking at this stop-gap DCTL, this needs to be applied in the nodetree as it expects cct as an input, right?
I was using the “two-in-one” DCTL because it was the only quick option that worked before the nodetree. When I tried to get the AWG_to_AP0 DCTL (matrix conversion only) into the IDT menu it gave me totally different results than when I applied it as a DCTL over the clips LUT menu. Is there anything else then adding the following command at the beginning to get a DCTL working as an IDT from the dropdown list?

#if (RESOLVE_VER_MAJOR >= 17)
DEFINE_ACES_PARAM(IS_PARAMETRIC_ACES_TRANSFORM: 0)
#endif