Working Color-Managed In A Flat File Workflow

August 10, 2021

Learn how to tag a single clip with multiple 'input color spaces' in DaVinci Resolve when executing a 'flat file'/preconformed workflow.


Series

If you’re color grading in DaVinci Resolve but your clients are editing in another software then you may be following a common workflow, often referred to as the ‘Flat File Workflow’. Here on Mixing Light, we’ve been talking about this workflow since the website was founded in 2013. If you’re unfamiliar with this approach for moving from an NLE to DaVinci Resolve, this early series on ‘preconformed’ workflows will get you up to speed.

The question I recently faced on a series I was grading:

How can I execute project-level color management in a preconformed/flat file workflow?

The crux of the problem is this: In flat file workflows, every shot references a single, common video file in Resolve’s Media Pool. A single clip can only have a single color space input transformed assigned to it. If your flat file has shots from multiple cameras recording in multiple gamma/gamut profiles, what then?

In this Insight, learn how to assign multiple color space input settings to a single video clip.

At first, this may seem nonsensical if you’ve set up a color-managed project. The key in the workflow I share in this video is to combine a typical XML export from the NLE with EDLs for each camera profile in the timeline. Then, in Resolve, you use the EDLs for each camera to ‘Split and add to Media Pool’ and the overall XML to rebuild the entire timeline.

There are a few tips in this video on how the editor should organize the timeline they send the colorist. There are also a few tips that I found useful when importing these assets to make assigning the input color space transforms painless.

Comments? Questions? Observations?

Please, use the comments below to share your thoughts on this workflow. I have a suspicion this approach may generate more questions than answers. I’ll be happy to flesh out anything in here you’d like to see explained in more detail.

Member Content

Sorry... the rest of this content is for members only. You'll need to login or sign up to continue (we hope you do!).

Membership options
Member Login

Comments

Homepage Forums Working Color-Managed In A Flat File Workflow

Viewing 16 reply threads

    • Evan A
      Guest

      Nice workflow!


    • R Neil Haugen
      Guest

      Patrick, that is the slickest Pr/Resolve workflow I’ve seen. Easy to understand the process, and allows for full and easy CM in Resolve.

      And yes, as someone who does a ton of Pr/Resolve stuff and handles queries on this sort of thing, I would love to see any follow-ons you could have. This to me is a massively updated *and upgraded* workflow.


    • Jacob F
      Guest

      It’s much simpler to just enable the “Extract reel names from EDL comments” option 🙂


    • Pat Inhofer
      Guest

      LOL. Quite true. But then, I’ve worked on plenty of run ‘n gun jobs where reel names were useless for separating out cameras. Although, adding reel names related to the cameras can work if the NLE will honor that metadata change on EDL export.

      It’s definitely a workflow variation worth testing when executing this process. Thanks for the idea.


    • Pat Inhofer
      Guest

      Neil – thanks! If you get any specific queries, definitely let me know. The ‘EDL Comments’ one is a good one worth breaking down – but that’d be a short video, so I need to collect a few more thoughts, questions, or approaches to warrant a full-blown video.


    • Jon Coy
      Guest

      Really excited to give this workflow a shot! Thanks for the ideas, Patrick.


    • Jamie Dickinson
      Guest

      This is a great Insight, exactly what I needed to explore yesterday! I went a different way though… For some reason I can’t get Split and Add to work at all, even though the same EDL works as PreConformed EDL. And it is now possible to have a single master clip but change the Input Color Space for each shot (p988 on the 17 manual). EDL comments and C-mode sort in the Lightroom work nicely and the info panel shows reel names as I step through each clip and group and colour code them. Oddly Data Burn-in Reel Name doesn’t work but %Reel Name does!


    • Pat Inhofer
      Guest

      You’re welcome. Let us know how it goes?


    • Pat Inhofer
      Guest

      Jamie, what’s happening to you on Split and Add?


    • Jon Coy
      Guest

      Will do.


    • Jamie Dickinson
      Guest

      Trying this again with another test… It doesn’t work. Either nothing is added or it just adds a couple of clips but not all the shots as I assume it should?


    • Tony S
      Guest

      Patrick, Another fine example of a workflow which makes Mixing Light a wonderful source of colour grading intel. 🙂


    • Marty
      Guest

      I think you can actually cut out your whole edl step. You can import the same clip into a bin multiple times, tag each as a different camera source then conform lock each of your timeline tracks to each of those master clips and you’re done.

      I do this a lot when I have a single camera source but with a few rec.709 stills or archive. I import the flat file twice, tag onc with the correct IDT and one with a rec.709 IDT and conform lock the timeline clips accordingly as I go along.

      The only real difference would be that you have one master clip for each camera instead of a master clip for each shot. But I don’t think that will matter since those master clips all have to share the same IDT anyway.


    • Pat Inhofer
      Guest

      For longer-form projects this workflow proved messy for me. I found it much simpler to use the track-based approach with the EDL allowing each shot to be tagged independently of all the other shots. It’s not unusual in ‘run-and-gun’ productions where a few shots were recorded differently and need to be treated differently than other shots from the same camera.


    • Troy S
      Guest

      It seems like we wouldn’t have to go through this very clever work around if Resolve could keep the reel names from the EDL after you relink the EDL to the flat file.

      My work around is to bring in the EDL and keep it offline. Duplicate the sequence and relink the duplicated sequence to my flattened .mov file. Then I cut the relinked sequence into layer 2 of the original sequence with the offline EDL. Layer 2 is what I color and layer 1 (with offline media) is there just so I can see the camera name.


    • Pat Inhofer
      Guest

      I’m not entirely sure in your workflow, Troy, if you can assign varied Input Color Spaces to your shots on V2 sourcing from the single common flattened .mov? Or are you doing node-based color management (which makes total sense for what you’re describing)?


    • Pétur R
      Participant

      A very interesting approach to a problem I’ll be facing very soon.

      Would you not consider taking the hassle of your editor and just to the whole process your self? Not taking into account if you have dissolves for example.

      Would you go about doing this exact workflow even if your editor is working in Resolve? Or would you approach this differently if that was the case?


      • Patrick Inhofer
        Keymaster

        Pétur – Here’s my thinking: No one knows the timeline I’m inheriting better than the editor. Since it’s a flat-file workflow, I don’t have access to the original camera files (that’s the whole point), so I can’t double-check camera cards, etc, if anything is unclear about a shot’s provenience.

        Plus, when we’re executing the flat-file workflow, it’s almost always related to the budget. Everyone involved would like their colorist to work off the camera originals – but a proper conform is a time-consuming process, especially since we’re moving between NLEs. On any timeline more than a couple of minutes long, there will be several hours of the colorist’s time related to matching the two timelines.

        Alternatively, the editor is going to spend an hour so prepping the timeline – but in most cases, the colorist’s hourly rate is much higher than the editor’s hourly rate.

        So when you break it all down, there’s both a monetary reason to have the editor do it and a workflow/familiarity reason for the editor to do it.

        IF the project were edited in Resolve – then I’d absolutely push to get a mirror of the hard drives and just relink within Resolve and work on the camera originals.

        Make sense?

Viewing 16 reply threads
  • You must be logged in to reply to this topic.

Hundreds of Free Tutorials

Get full access to our entire library of 900+ color tutorials for an entire week!


Start Your Free Trial
Loading...