Peter Chamberlain wrote:Be sure to try syncing the dual system audio to track 1 on your video clip. If your video clip has the sync audio track on another channel, move it to track 1 in clip attributes to sync.
Doesn't work reliably in my latest use cases.
Material description:
Video: ProRES 422 (or HQ) 50fps, 32-bit float, dual channel "RAW" Mid-Side audio.
Audio Left: 32-bit float mono channel (for binaural use). Started manually around 20-25 sec before "video" (lasting around 15 seconds longer).
Audio Right: 32-bit float mono channel (for binaural use). Started manually around 10-15 sec before "video" (lasting around 30 second longer).
Use case:M-S microphone recorded straight into my iPhone 15 Pro (main track required). Typical duration around 120 seconds.
The two mono mics are mounted roughly 18 cm apart, with the M-S in the center on a three cold-shoe mount. They need reliable syncing to main material.
Left and right mono-channel recorders are started (Zoom F2) successively, then the video recording (on a tripod in this test case).
After starting anything up, there are three, quick hand claps just above the center M-S microphone.
Approach to be used for between 10 and 20 "snippets" (varying lengths around the stated duration above).
ResultOne test "snippet" only. Initially as described above.
Direct sync impossible. Request just ignored by my "Resolve Studio" (in this case 18.6.4 Build 6). Not even a hint, that something went wrong, and a reason why.
Retried various approaches, but the only approach that worked, was manually adjusting the three tracks (one video plus audio and two mono tracks) to very NEAR usable by hand, and then request automatic sync based on audio content. Otherwise no dice!!!
Repeated test with 24-bit material (simple conversion in outside audio editor, but otherwise not altered). Same result.
Now imagine 10-20 "snippets" requiring this manual approach EACH and EVERY time, just because Davinci Resolve Studio steadfastly rejects, what FCPX does routinely.
In this case, FCPX (10.7.1) given same original material, just selected and sync'd immediately, without large ado in form of doing 95% of the repositioning to help Davinci recognizing the three VERY clear (and LOUD) quick claps figuring prominent (without clipping) in all materials.
Earlier versions of both have similar behaviour.
Davinci Resolve Studio already has some shortcomings in other situations; just imagine 10-bit UHD ProRES 422 from iPhone replace with 12-bit 5.9k ProRES RAW from Lumix S5) and sometimes a requirement of delivery in ProRES (typically 4444 (HQ) with Alpha) - both requiring extra steps.
On top of partial manual sync of any "audio" in numerous "snippets".
In this test case, I did not need ProRES RAW video and no export to ProRES. For test purposes, even low h265 10-bit 4:2:2 UHD bitrates (100-150 megabit/sec) would have sufficed.
After a loooot of "fiddling around" (moving tracks up and down too), converting from 32-bit float to 24 bit PCM and even activating the M-S to stereo plugin (various settings for "stereo width" to see, if that had any effect). No dice.
Suffice to say, that I scrapped that project, cranked FCPX into action, and shortly after, everything was sync'd, cut and adjusted, ready for my standard compressor "simple TV test use case" export (h264 4:2:0 50, 75 megabit/sec, h265 4:2:0 50, 75, 100, 150 megabit/sec) taking care of most format read limitations of various standard "4k" TV's released to the unsuspecting public since 2020 (older models have even more "exciting" limitations in petto for the unsuspecting cutter).
Regards