FCPx audio drift, syncing not working

So this turns out to be hardware related. All of my video recordings for the screencasts have involved a 2018 MacBook Pro that has been connected to an OWC ThunderBolt dock on one side, and a eGPU on the other. Since this video was a comparison of a couple of docks, I was shooting the video without having the TB3 dock connected to the computer since I was showing it in the video. There was nothing I could do to get audio recorded correctly. The audio was always shorter, by about .5 to .7 seconds per minute or worse. Thought it might be related to not getting power from the eGPU, but verified it was.

This morning I hooked the OWC dock back into the system and all the tests worked. So tonight I tested it again when I got home, still worked.

So I decided to try making the video clip again, disconnected the TB3 dock so the eGPU was powering the computer, had to use a different USB connecter so the scarlet 2i2 would remain connected, and suddenly the audio is short again.

I have no clue why this is happening, I always thought with audio dropouts you just had blank spots in your audio, but seems like it would totally drop recording audio randomly throughout the test.

No idea, but not worth fixing, I’ll manually sync the file I need. ( Unless some other random thing is the cause and my theory is just coincidental to these tests and it shows up again).

Guessing this is why FCPx just didn’t even attempt to match it up.

Thanks for the response.

Not sure, I can check that when I get back to where I have this all set up. But I haven’t changed anything regarding those settings, not sure why it suddenly quit working. I guess I assumed when I imported audio into FCPx it would mange that and correct it. Maybe that’s not true or maybe there’s a setting I’m missing on import? I also thought the whole idea of audio sync was to look at the waveforms and match them up. As I said, all it’s doing is attaching the two without any modification of the audio or attempt to sync.

appreciate your help. Still puzzled by all of this since the audio out of sync appears to be quite significant, audio losing about ⅔’s second every 40 seconds or so. If I record into to FCPx or Audacity at 44.1khz or at 48khz, neither matches the audio of the camera (or is even close). If I record into my Zoom H1n set to 44.1khz, audio matches which seems strange because I thought most dslr’s type cameras recorded at 48khz. of course the entire purpose of going straight into the computer is so I can manage and process the audio with a dbx286 channel strip through a FocusRite 2i2, which means I don’t have to post process the audio or worry about getting it off the card.

This work great for 3 or 4 videos, everything synced. I need to go back to those and see if the audio matched to start with and what the difference might have been or if FCPx just managed to sync them and it’s not doing it now.

I’ve experimented with seeing if this is due to variable frame rate of the video recording, but no luck there.

Files from camera show 48khz audio when I look at the file info in QuickTime. However, in FCPx I can sync 44.1khz audio from ZoomH1n no problem, no drifting. Audio recorded into FCPx using voiceover at the same time does not match up.

I verified a video I shot last week with these exact settings matched up no problem, recorded audio directly into ScreenFlow while I recorded video on the camera, imported the video into ScreenFlow, lined them up and no issues. Now this doesn’t work.

I have no clue what has changed. I tried changing the apple audio midi setup for the scarlet 2i2 usb interface too 44.1khz, no good, 48khz, no good.

I’ll do some more testing tomorrow.

I know I’m not an engineer and no clue how this all works, but it seems the software should be smart enough to detect the rate and adjust the audio to fit the timeline of the track so 1 sec = 1 second. I can’t find any combination of bitrates that work. I’m going to test with the Sony a7r3 tomorrow to see if somehow something has changed on the camera, a video I did a couple of weeks ago I used both cameras and recorded the audio into ScreenFlow, exported the audio out and synced everything up in FCPx, worked perfectly.

Thanks for all your suggestions.

Something else going on. I ran a test this morning, recorded audio in from my scarlet 2i2 to Audacity (project set to 44.1), also to Piezo which ended recording a 44.1 m4a file. Exported from Audacity a 44.1 wav file, and a 48 wav file. Data on the camera shows audio is at 48khz. I verified all of the audio using QuickTime inspector to be sure the sample rate matched. Imported all into FCP and all audio matched perfectly with no modifications. Also pulled them into ScreenFlow and audio matched perfectly.

So I ran a quick test recording audio into ScreenFlow and video on the Sony a6500, and it matched perfectly. Exact same test that I ran when I quit testing last night before calling it a day which drifted by nearly 1 sec per minute of audio.

I have no clue what is going on, but it seems perhaps system related since it randomly works fine and doesn’t other times. Problem is I can’t tell when it is going to work and when it isn’t. I don’t think it’s related to the sampling rate though.

So any other thoughts would be appreciated.

So a little more testing this morning, this turns out to be a USB problem, apparently the TB3 -> USB adaptor I was using wasn’t compatible with my USB 3 hub, so I was only getting USB 2 speeds or some similar problem. I tried a different dongle (Apples TB3 to USB and HDMI) and everything works fine.

i use filmic when recording on an iPhone. But this is using a Sony a 6500 for video.

As I mentioned right now I’m just trying to get Sony a6500 video to sync. I rarely use the iPhone.