I’m not a fan of giving over mattes. I have a grade coming up on a really big job. Heaps of roto and blue screen.
I will instinctively hold back and not offer any mattes unless requested and even then I will do it reluctantly.
There is more to these comps than just the roto and mattes. Those often come in early down the comp schematic. If the colourist uses one of my mattes and starts being aggressive with the grade, my comp will start to break very quickly.
If it was shot for real they would need to use softer patches and I would like the same on my comps.
Sometimes my shots come back from grade and I don’t recognise them anymore. I will kick back bad keys (magic mask my arse) but at least that restrains them. With one of my mattes all restraints are off.
If CG is involved, VFX (comp on flats) before color w/ show LUTs applied for postings, final grade @ the end.
If it’s all 2d bits, cleanup, screen comps, etc… pre-grade would be fine & at least color is approved; to @andy_dill’s point, you can just make the final result look good knowing no one else is going to touch it.
Short answer though? It depends… This doesn’t take into account the agency/client’s schedule on when they book the grade session, preferred colorist availability, updated shot run-thrus, tape2tape, etc etc etc.
Been burned both ways by Post-Grade breaking comps at the last minute, and Pre-Grade making CG difficult. Hosed plenty of times by giving Color mattes, only for them to break comps/edges even more painfully…
This is all from a commercials/advertising perspective, with tight schedules… Both scenarios have their pros and cons.
One thing I didn’t see mentioned is that a lot of the color first pipeline (after telecine) was also getting the cllient, not the agency, to buy off on the spot. So many of them have so little exposure to the process and aren’t in the room with you. They’re just sent a link and jump on a call with the agency to walk them through what the’re seeing. The agency is often more concerned about showing the best version of the spot sooner rather than later and color can be done in a day, early in the process. I agree, it would be great to have that be something easily integrated into our pipeline and it sounds like we’re getting close. In the commercial world, primarily we did color early in the process. All the feature and episodic work was the opposite. We often didn’t see final color until it aired or was in a theater.
Sames. The comp should hold up to everything they’re throwing at it and I feel like it should always be treated as if it were actually shot so the colorist should use what tools they have available to achieve what they’re after. That said, sometimes I lose that argument and mattes are supplied to pacify everyone.
Agreed with Richie Betts. Never give mattes to TK unless final resort.
I can’t remember the last time I did a job where it was graded first… I would never go back. Trying to track something which has had a grade window put on it messing things up. That said, I am in the enviable position of working alongside Colourists and we help each other out with issues. This is the largest benefit. It’s the opposite of working with Co3 who are, in my limited experience, a bunch of divas quick to deflect responsibility. In their defence, they are pushed from one job to the next with little investment in the project and that must be difficult: something I experienced in the old days at the mill.
I digress. Grading after the comp is completely different and really makes life easier and gets better results. However, it takes a bit of forward planning, consistency and some structure which could be difficult to achieve for single operator outfits (but not impossible).
*I only graded some spots I comped myself once. Never again. Torture to do in flame.
The grade can be set early. But the grade WILL change 99% of the time, right up to the last day. So will the comps, but it takes 10 minutes to rerun the grade on the revised comps, whereas it takes several days to redo the comps when the sources change colour.
If the comps are done well they should hold up in any grade.
I’ve only recently dipped into comping before grade – but a couple of issues have come up. When comping screens, screens will change significantly in the grade if it’s being pushed. This is a headache with brand colors, etc. Also working with volumetric light is a pain, when the grade gets pushed into day for night. Again, if the grade is being pushed significantly, it is hard to precisely account for these things in the comp.
I really do like comping before grades and see the massive benefits, but some instances haven’t worked out so great. Open to suggestions. Maybe brighter minds have already figured this out.
A over B comping with an alpha (mattes in color grading) don’t have the subtleties of comp environments. Operators like screen, add, multiply and others that might be used in a driving comp (or ______) to get edges nice get completely f__ed-up when a colorist cranks on the shot too much.
Again, it is something that can be irritating as all get out.
