[Marc Wielage]

My workaround would be to use a 3rd-party utility like Better File Renamer (or Better Finder Renamer for Mac), load the CSV into that, select the exported files, and then use the comments to augment the name of the file, either as a prefix or as a suffix.

Generally, I think it’s a bad idea to rename source file names, but as long as these are the files actually used in the edit AND in the final conform, then it won’t matter. If you relink to camera files with different names, you’re completely screwed.

[Marc Wielage]

Check the histogram. If it’s underexposed, it’s underexposed. I’ve felt there’s quite a few low-cost cameras out there that are relatively insensitive (and I would include the lower-end Blackmagic Cameras on that list). I think the Ursa and Ursa Mini are a lot better. Often, finding the “sweet spot” for exposure requires some experimentation and a lot of camera tests to figure out how hard you can push the image before it breaks. To me, one major issue is the cameras require more fill light than many people think.

[Marc Wielage]

Uh, make that MOCHA. (Typo… ahem.)

[Marc Wielage]

There is another option: you can create a tracking mask in a 3rd-party program like Bocha, which will track absolutely perfectly, then export that as an external mask and bring it back into Resolve. This will be absolutely correct, assuming the shot doesn’t change between the two versions.

[Marc Wielage]

Call me crazy: I just track it again.

You can copy over tracking information if you insist, but it requires that you have the “Apply Grades Using Keyframes Aligning Source Timecode” option checked in the right-click menu selection. I generally don’t copy the keyframes across for the simple reason that the shot contents are rarely identical. When they are, I use ColorTrace to bring the original correction over which brings everything including Keyframes (color and tracking). I don’t think this is as complex as you make it out to be.

System performance and speed depends on a whole bunch of factors, including file type, CPU speed, available RAM, GPUs, and drive I/O. There isn’t just one single factor, so it’s not an easy thing to diagnose. 64GB of RAM is not too little, particularly if you throw in a 16- or 24-core processor. Project saving/loading hinges on the complexity and content of every timeline in the project, not just the one you’re working in. Weighing it down with a massive number of shots, complex node trees, and many hundreds of Gallery stills also extracts a price in terms of time, but it’s sometimes unavoidable.

[Marc Wielage]

[Hector berrebi] “People often ask what’s conforming, the process of bringing a timeline from an NLE into a grading tool, which in most cases became a relatively simple task in recent years. “
I have a standard 2-page document I send clients that outlines how I need the edit timelines prepped for color before I get them. For something this complicated, I’d just put my foot down and say, “render a flattened ProRes 444 file and I’ll color-correct that.” The moment source files are renamed, it goes south very quickly.

Speed changes are not 100% interchangeable between Avid, FCP, FCPX, Resolve, Premiere, or anything else. They can work to a degree in some cases, provided they’re fixed speed. When Avid 8.x came out, I noticed that suddenly resizing also started to work, including dynamic resizes, which was a welcome surprise.

A lot of this stuff happens because of the limitations of XML. They probably should develop an “XML2” that includes more parameters for stuff like sizing and color and speed and all that other stuff, and do it in a way that’s 100% compatible with all software. But getting everybody to agree with those standards will be problematic.

[Marc Wielage]

And in addition to all the other advice: Workflow is a very complex subject that you can’t adequately explain in an internet forum message. It involves editing, offline, conforming, sound, VFX, color, titles, and final delivery. It’s a whole lotta stuff.

Enough, in fact, that there’s an entire book on the subject:

Modern Post: Workflows and Techniques for Digital Filmmakers
by Scott Arundale
published by Focal Press (ISBN #0415747023)
https://www.amazon.com/Modern-Post-Workflows-Techniques-Filmmakers/dp/0415747023/ref=sr_1_1?s=books&ie=UTF8&qid=1485585750&sr=1-1

[Marc Wielage]

[Joseph Owens] “For a long time, that has been the key to how a cinematographic colorist differentiates from somebody (anybody) who thinks its about making one frame look great.”
I dunno if I’d call it “cinematographic,” but I have often said it’s easy to make one shot look good. The trick is making every shot look good on its own and also when seen continuously with the other shots. I worry far more about the matching than the individual shots, but no question, everything matters.

The problems occur when the lighting doesn’t match at all, and we have to fix it. It’s particularly tough when you have shots that were done two or three weeks apart, sometimes by different people in different places, but you have to believe they all happened at the same time, same place.

But as far as the “brown look” goes, there’s all kinds of looks floating around out there, particularly the trend for very “steely” blue looks or the teal/orange look. To me, you have to start by matching it all and solving any lighting issues, then you go back and apply the overall look to the scene.

[Marc Wielage]

I did an entire feature in the 1990s that had a sepia/brown tone: Kansas City, done for Robert Altman. Very tough film, done in the pre-Resolve days. The entire look was created digitally, and the tough part was not establishing the brown look — the tough part was keeping it consistent shot to shot and scene to scene. I think that took almost 3 weeks.

[Marc Wielage]

[Glenn Sakatch] “Nobody will ever look at your final Prores output and say “it doesn’t look like you used an uncompressed intermediate””
Unless they baked in a permanent color correction and/or used an 8-bit format.