Dennis Widmyer
Forum Replies Created
-
Very cool. Yeah, it was contacting the creator of Xto7 that got my problem initially sorted out as well. He’s amazing and got back to me very quickly. Within an hour he had all 5 reels / XMLs fixed and we were back in business.
If it ends up working out for you too, then yeah, maybe wait on upgrading to 10.1. Though I don’t know of any examples of people upgrading a project mid-way through and really having any bad issues. So up to you if you want to roll the dice.
—
Dennis Widmyer
http://www.parallacticpictures.com -
Well, two things that will help:
1. They made some big fixes to how XMLs export in FCX 10.1, so you should definitely update your software. As a safety measure, back up your project first, and make a zipped archive of the actual application file for FCX 10.0.9 (that way you can roll back versions if you’re not happy with 10.1). Also, if you update to 10.1, make sure to choose the option to Save your current projects as a backup. This will create a backup version of them that doesn’t get updated to the new Library structure.
2. After I had this issue, the owner of the Xto7 program made a big update that resolves the issue, and once he did, all our multicam clips arrived in Resolve perfectly.
The issue ended up being the way the multicam clips were created from the assistant editor. He would first sync our A and B camera clips to the audio, hence creating Synchronized Clips. Then, he’d take that new synchronized clip and make THAT into the multicam clip. So he was actually creating an extra step, and that was the problem. This would create a lot of examples where our multicam clips had audio beginning before video. And up until the last update, Xto7 couldn’t read those correctly, and hence, they weren’t surviving the XML transfer, and that was creating all the blank spots in Resolve.
In the newest version of Xto7, it addresses this hiccup and knows how to read it now. But another way to insure this doesn’t happen, is to make sure your multicam clips were created the correct way. To do that, simply highlight / select your A and B camera clips and your sound file all at the same time, and when you merge those into a multicam clip, it syncs the audio as well. In other words, do NOT synchronize the clips first. It’s a pointless step and will cause you a ton of headaches. I know from personal experience. :/
Hope this helped.
—
Dennis Widmyer
http://www.parallacticpictures.com -
That all makes sense. Wish we would’ve done it that way. Too late now. But great to know going forward.
Thanks!
—
Dennis Widmyer
http://www.parallacticpictures.com -
So you’re saying the best workflow for us would’ve been to first create the multicam clips (which comprise about 85% of our entire film project). Then, take all of those clips and, one by one, open them in the Angles Editor and line up the clapper with the audio clip?
If that’s what you’re saying needs to be done to avoid Synchronized Clip, I get it, but honestly, it sounds like a huge headache for any feature project with as many multicam clips as ours, when FCX provides a tool to sync audio with the click of a button.
Pretty annoying that you can’t tell Synchronized Clips to retain the source timecode. That would solve all of this. Yet another item to put on the future updates list.
—
Dennis Widmyer
http://www.parallacticpictures.com -
Are you certain of this, Jeremy? Larry Jordan wrote a whole article about how they do:
Yes, that will definitely be in the next release. The Mac App Store will be closed for a one week Christmas break so it may not appear until after that, depending on their review times.
FYI I was wondering why this bug hasn’t been reported before this. Is there another way to make multicam clips that would avoid this? Here’s the method my assistant editor used:
He would first grab each camera angle clip and sync it with the audio. So now he has 5 clips in the bin:
26B_1_A (camera angle 1)
26B_1_B (camera angle 2)
26B_1 (audio)
26B_1_A_Synchronized Clip (sync sound version of angle 1)
26B_1_B_Synchronized Clip (sync sound version of angle 2)Then, with those bottom two Synchronized Picture/Sound clips, he’d highlight each one and choose ‘New Multicam Clip.’ This creates a merged, synced version of the two camera angles.
The problem is, because they were synchronized first, they no longer show the source timecode.
So the issue might stem from the way my particular multicam clips were created from the get-go.
Is there a better process you can recommend for doing this? Because whatever process Larry Jordan is using, is preserving the source timecode after the clips have been multicammed, whereas mine is not.
—
Dennis Widmyer
http://www.parallacticpictures.com -
Jeremy,
Yes, this is only a problem with unsynchronized multicam clips. When I go to the unsynchronized version of the clip, I can see the source code.
All of my multicam clips start at 00:00. And opening a multicam clip in Angle Editor doesn’t solve the issue either.
—
Dennis Widmyer
http://www.parallacticpictures.com -
Yeah, it absolutely doesn’t work for me, and it’s really starting to screw me up. Ive got my co-director in NYC right now (I’m in LA) and he’s working with a colorist at Deluxe. And last night, he was texting me to fill in some gaps where shots that had speed ramping applied were missing. He needed the source timecode for these shots, and here’s how I had to figure out how to give it to him:
1. Find the multiclip on the timeline.
2. Turn on skimming. (command-option-s)
3. Skim over the clip (mouse over the actual clip). Again, for me, this doesn’t show source timecode. It shows where you are in duration of that entire raw shot, with the shot beginning at 0. (my multicam clips are ignoring the source timecode)
4. So if I skim over the first frame of the multicam clip, and the counter shows me 01:12:04, I know I’m a little more than a minute into that actual shot/take.
5. Get the clip’s shot / take number. Example: 26B_1.
6. Find it in the event browser (the solo version of it). So I’m not using Shift +F, because that would find me 26_1_Synchronized_Clip (the multi version)
7. Go to the raw shot and skim through till I get to 01:14:04 into the clip. Then make my In there.Trouble is, I’m not too good with figuring out how to precisely put an In into a clip at a specific time. I think I know how to do it from the timeline, but not from the event browser. So then I run into the headache of, looking at the source timecode on the solo clip (in the event browser) and doing math in my head so I can figure out that, if the source clip starts at 14:05:03:01, I have to add 01:14:04 to that number to figure out my In.
But then there’s yet ANOTHER added headache. In some clips, sound begins before the picture, so I have about 9 seconds of black before the clapper appears. So I then have to subtract 9 seconds from my math solution.
In short, it sucks, and trying to pull him the timecode for just one 6 frame clip he needed for the colorist, took about 20 mins.
I’ve followed all the steps; have all the settings turned on correctly; even had a long email back and forth with Larry Jordan about this: For some reason skimming isn’t showing me the source timecode of multicam clips. Could this be an issue with how they were imported?
—
Dennis Widmyer
http://www.parallacticpictures.com -
Yeah, it doesn’t like speed changes, but in this instance, it was due to the multicam shots.
—
Dennis Widmyer
http://www.parallacticpictures.com -
Agreed, there’s so many ways they could make this easier in subsequent updates:
1. Allow a way to flatten the multicam clips. (similar to breaking apart a compound clip)
2. Allow a similar tool (like Shift F) that matches the frames to the original clip, but let’s you determine WHICH angle (not just default to the synchronized multicam clip)
3. Or at least let you see the source time code of the multicam clip in the main display. Right now, it only shows me the source timecode of the single angle versions of the clip.
—
Dennis Widmyer
http://www.parallacticpictures.com -