Plan for Working with 240fps footage?

Our content is focused on very fast-moving objects. Because of this, even 60 or 120fps doesn't allow us to see what we need.

We have 6 cameras. Each recording 240fps, with metadata that shows 30fps. (when viewing with generic Windows video player, the clips play in slow motion - 1 second of real time takes 8 seconds to play).

Each video is 3-10 minutes long (in real time). All cameras are started manually, with no timecode. The only time reference we have is the File Modified time.

What is the best plan for manually lining these up on a DaVinci Resolve Studio timeline, so that they are all synced (so that play head is at same instant of time for each video on the timeline)?

Use a clapper or burn a small firecracker.
Or shoot whatever makes a short sound and a related visual event.

Footage already recorded. Need plan for how to manage it in the DRS timeline.

Ideally, the timeline would display the "wall clock" time, so it will be simpler to line up the videos (to get them close...before manually adjusting to exact alignment.).

You have three options for syncing.

Timecode

Audio

Visually

The first two can be automated. The latter is manual.

This will unquestionably be a manual process. I just need some help determining how to set the Timeline frame rate to work best.

I tried just dropping the clips on the default 30fps timeline. The problem with that was the timeline frame rate made it difficult to estimate where the different video clips would go (relative to each other), since the timeline time was dramatically skewed (a 1 minute real time video clip would show as 8 minutes).

Next, I tried to create a 30fps timeline, then changed the clip speed for every clip. That worked for placing all of the video clips on the timeline, but was a major fail, because I lost 7 of 8 frames in the change clip speed process.

Any thoughts? I find it very strange that a company with cameras that do much higher frame rates has (?unnecessarily?) limited their software to 120fps.

If you're delivering 30 fps (slow motion), set your timeline up that way, and sync clips yourself by eye. The temporal resolution of the 240 fps clips should let you be quite accurate. Most people simply do not have displays capable of showing content at a higher framerate than 60p, so I imagine high framerate deliverables have not been a high priority for BMD.

Final video clips will be 30fps (showing slow motion of various angles).

I would set a first TL at 120 (or 240 ? not possible really ? never tried actually), do/check/fine-tune anything needed, then copy every track to a 30 fps TL.
It may sounds dumb, but I like to set the first TL of the project at the exact framerate of the clips for the sake of the project, actually to see clips at their speed in Resolve in the first place...
In the past I was ok with seeing em in another app like VLC, now I prefer to see them at real speed in Resolve prior to do anything else...

Since there is no 240 fps timeline and you'll want slo-mo in the end anyway, I don't think throwing the footage into a 30 fps timeline is the best solution. DR will drop 7 frames out of 8 and your visual/manual sync will be imprecise.
Go to the Clip Attributes instead and switch them all to 30 fps (you can select more than one together). Now all your clips will show as slo-mo and you can sync them precisely.

Remember, my native clip attributes are 30fps. When I drop 1 minute of real-time footage on the timeline, it is 8 minutes on the timeline in DRS.

So far, this is the only way I have found to make this even sorta work...without dumping 7 of 8 frames, which is completely unacceptable.

Well, then you are already doing the right thing.

Footage already recorded. Need plan for how to manage it in the DRS timeline.

Eyeball it using visual clues. We did this with film for decades, and somehow we survived and got all the work done. I think I've literally synced thousands of miles of film by eye in the 1980s and 1990s; while I don't miss it, it's a valuable skill I still use today with digital material. It's not that hard, though I concede it's an annoyance if you haven't prepared for it logistically with clapper slates. There is no automatic way to do it, particularly since accurate timecode won't work above 30fps.

For MOS or wild footage, particularly action sequences, film crews get a really large clapstick (like 3-4 feet long), have a guy stand in the middle of the action, they roll all the cameras and then the camera assistant whacks the sticks. This gives everybody a common starting point, which helps in editorial. The sticks were something like this:



I can recall transferring some Abbott & Costello outtakes from the 1950s where they used a big slate right before an explosion, but that's an image I can't get my hands on. You see this kind of thing occasionally in Behind the Scenes documentaries on big action films.

That would be the perfect solution, but his shots are already done.

Something that might be helpful if the moment the cameras were started differ a lot came to my mind. There are tools (like QtChange for example), which can write the file creation time into the files as TC. That will be far from precise, but can get you into the ballpark if (and that's a big IF) the cameras had at least their TOD set correctly.

From there on, it's eye and hand…

I had not heard of QtChange. That could be very interesting.

If it would at least help line up the clips close to where they should be on the timeline (based on time of day timecode), it would be a massive help in manually getting everything lined up perfectly. The major part of the task is getting the clips generally where they need to be. Getting them aligned to the exact frame is time consuming, but not that difficult.

That would be the perfect solution, but his shots are already done.

Well, I did say... you gotta do it all by hand, which is basically watching it and lining up the shots manually based on visual action. I've done miles and miles of material like this. Even in the case of sync sound, it happens: we manually synced up more than 350 hours of material for the 2008 Rolling Stones documentary Shine a Light documentary, and got zero complaints on lip-sync. No claps, very few timecode slates, all done by visual clues and guesswork. Anybody can do it... it just takes time, skill, and experience. There is no automatic way for it to work.

You can write arbitrary timecode starting values into files with ffmpeg, and you can recode them to run at any frame rate you like. If you wanted to slow your 240fps input down to say 30fps output so it ran at 1/8th the speed, you could do that by just setting the presentation frame rate.