BMPCC 6k and Tentacle don't sync

Hi everyone. I posted a similar question a few months ago, but I have some new equipment in the mix, so I thought I'd start a new thread. My previous set-up was a Zoom F6. Now I'm using a Track-E.

My BMPCC 6k Pro, with Sync-E attached (and synced via app) is out of sync with my Track E recorder (also synced via app). Picture lags behind audio by at least two frames. Possibly two and a half. I've come to this number by syncing via slate clap (manual).

I'm using Auto Sync via timecode. Resolve says the tracks match via the timecode overlay in the viewer. Definately not in sync.

If I Auto Sync via waveform, it's in sync.

My EOS-R syncs via Tentacle Sync-e with no issues.

I thought It may be the app and Bluetooth accounting for the inconsistencies, but support at Tentacle assures me Bluetooth Sync is accurate. EOS-R appears to confirm this.

If I need to figure out some kind of offset, I'm okay with that. It's just annoying because it negates the point of external timecode and complicates the workflow.

The problem is so constant and repeatable, I must be missing something, because I'm not hearing anyone else with this issue. Any thoughts? Thanks.

I just went through this with my cameras. From my testing it's dependent on media used in the camera for one. Every camera had a different offset and by changing media i.e. from SC to T5 SSD I would get a different offset. I used hand claps as well to calibrate the offset. You can find the offset in Fairlight but I used Cubase. Align the clips using timecode then measure the distance between claps in milliseconds. Most offsets I found were not severe enough to cause lipsync but if I tried to use two audio tracks there was a definite audible echo.

Interesting. I’ve never considered media. I’m using a T5. Sometimes and SD. I wonder how a CF Card would be.


Sent from my iPhone using Tapatalk

Interesting. I’ve never considered media. I’m using a T5. Sometimes and SD. I wonder how a CF Card would be.


Sent from my iPhone using Tapatalk

Might be pretty good. Only camera I tested with a cf was the Ursa Mini 4k (gen1) and it was the only one where it's audio was ahead of the Tascam recorder I use. It was like only 5ms off. There may be other factors as well but the media was the only consistent variable I found. There's not a lot to be found on the net regarding exactly how TC is recorded to media metadata.

At one point I tried injecting timecode via the mini XLR and updating the time code. It didn’t seem to help. But I may give it another go.


Sent from my iPhone using Tapatalk

At one point I tried injecting timecode via the mini XLR and updating the time code. It didn’t seem to help. But I may give it another go.


Sent from my iPhone using Tapatalk

Good idea. I didn't think of that. I did inject LTC into a channel on my Presonus mixer and converted that to ensure I could get it working. That worked out fine. The Tascam has a timecode input via bnc and imprints directly as metadata. I'm going to try your idea as well. Great thinking!!

Okay. Really strange. Ran Sync-e through mini xlr on BMPCC 6k. Definitely better. Off by one frame using analog slate to check sync.

But on the Fairlight page, the waveform is in sync.

I’m wondering if it’s a playback issue vs stepping frame by frame.

Regardless, it looks like best practice is to run timecode into the XLR. At the very least I can see the timecode baked on the track and know I’m not getting hung up on metadata inconsistencies.

Always happy to hear any ideas you may have.


Sent from my iPhone using Tapatalk

Okay. Really strange. Ran Sync-e through mini xlr on BMPCC 6k. Definitely better. Off by one frame using analog slate to check sync.

But on the Fairlight page, the waveform is in sync.

I’m wondering if it’s a playback issue vs stepping frame by frame.

Regardless, it looks like best practice is to run timecode into the XLR. At the very least I can see the timecode baked on the track and know I’m not getting hung up on metadata inconsistencies.

Always happy to hear any ideas you may have.


Sent from my iPhone using Tapatalk

Got me at this point. Interesting outcome.

As usual. I spoke too soon. I can’t get any repeatable results. I’m on an m1 Mac mini with 16gb of ram, maybe that’s the weak link.

I suspect it has something to do with the BRAW codec. I’ll keep digging.


Sent from my iPhone using Tapatalk

The ProRes codec appears to be syncing fine.


Sent from my iPhone using Tapatalk

The ProRes codec appears to be syncing fine.


Sent from my iPhone using Tapatalk

And the plot thickens!!! All about how the codec handles the time code from receiving it to imprinting it? Things getting buffered at different rates? Information being processed in different layers? Yikes! Super interesting for sure.

Jumping in here to note that I've had TC sync issues on multicam shoots with all Blackmagics - usually there's a sub-5 frame offset between all cameras, and that offset stays consistent throughout the shoot. We've also been shooting BRAW... maybe something with the codec? In any case, I've just been asking production to use a smart slate and match both audio and video to that.

Jumping in here to note that I've had TC sync issues on multicam shoots with all Blackmagics - usually there's a sub-5 frame offset between all cameras, and that offset stays consistent throughout the shoot.

Thanks for chiming in. I hadn’t thought of multicam issues.


Sent from my iPhone using Tapatalk

There’s always a chance of an offset somewhere.

The question is, is it always the same?

There’s great logic in recording LTC to the audio input. This should eliminate the camera clock jam process from this.

If you have a single clock you could jam the camera, immediately switch to LTC recording and try recording a few clips.

Unless a camera can Gen-lock and you use a clock that offers video or genlock sync, you’ll never be able to get sub-frame accuracy across multiple cameras. They will all be out of phase by at least a frame.

JB

I'm having similar issue with my new 6k pro when using LTC from Mixpre 3 ii via 3.5mm input (both are 23.98fps). Luckily I kept timecode audio track which saved me. Seems 6k pro will have inconsistent off from like 2 frames to 7 frames. Not sure If I did anything wrong to jam. Really don't want to buy another two Tentacle

I think I solved the problem. It seems as simple as updating the time code from the audio track. I can't believe I never tried that. I'm pretty sure I did at one point. The only thing I did differently was make sure the Sync E in the mic input was the only source of audio to the camera. No scratch audio used.

Seems to work fine.

I came upon this because I'm prepping for a Music Video and I wanted to stripe the playback audio with usable timecode for the Editor. It worked flawlessly with the exception that Resolve didn't natively recognize the audio timecode I used, so I had to manually update it. That's what made me rethink the problem I described here.

Hopefully, it's not a fluke. I can tell you that relying on the metadata of the timecode was definitely less accurate than pulling timecode from audio. Probably explains why I didn't have problems with my EOS-R.

If anyone double checks my work and gets different results, I'd love to hear about it.

Thanks for your replies.

Best,

Brendan

I recently updated to Resolve 18 and Camera 7.9.1 firmware. I ran a number of tests today and was not able to get my 6K Pro to sync clips with a Sony AX700 using Tentacle sync E generators. The Tentacle devices sync all 3 of my Sony cameras perfectly. The 6k Pro is the outlier.

I had occasional problems syncing the 6K Pro in the past but was able to get reliable results if I plugged in the Tentacle sync before the XLR audio inputs. Since I updated to Resolve 18 and camera firmware 7.9.1, I can't get it to work at all.

Here are the test details (verbose). Both cameras are set to record at 29.97 fps with drop frame tuned off. The 6k Pro is set to record Braw onto a T5 using constant bitrate and 8:1 compression. I have the camera set to stop recording if it can't write the frames. It's never stopped at 8:1.

When I plugged the Tentacle into the 3.5 mm jack of the 6k Pro, the Ext icon appeared and the clock reset to roughly the same time that was displayed on the Tentacle Bluetooth app. I hit the record button on both cameras with opposite hands at roughly the same time.

After a few minutes I loaded both clips into the Resolve Media page. I selected the Sony clip and chose "Update Timecode From Audio Track". I could see the TC Start field change to match the start time of the Braw file recorded by the 6k Pro. Then I tried to sync them on the Cut page. No luck.

I did another test with both cameras and recorded the LTC code from the Tentacle onto the Ch 1 (left) track of the 6k Pro just like I do with the Sony cameras. On the Media page I selected both clips and chose "Update Timecode From Audio Track". No luck with that either.

The Sony cameras are easy to sync in Resolve. That leads me to believe that the problem lies with the Braw file and the way Resolve is reading it. I would really like Blackmagic to look into this. The camera is much less useful if it can't be used in a multi-camera shoot.

I'll post this to the regular forum in a couple of places and see if someone has a solution that I've overlooked. If that doesn't work I'll file a bug report.

How many Sync boxes were you using?

I normally use 3 or 4. My test yesterday only used 2 but I switched to a third and repeated to make sure it wasn't a tentacle. They were all fully charged and operating normally.

When you try to update timecode from audio you said “no luck”. Does that mean the timecode is inaccurate? Or doesn’t update at all?

Also, when recording timecode to audio, are you hearing timecode on the track? Are you seeing it on the meter while recording? Are you sure your levels are high enough?

Are you on OS X? Can you try the Tentacle sync app to see if everything reads properly?


Sent from my iPhone using Tapatalk

Yes I do see the audio level on the Left channel produced by the Tentacle device on both cameras and I hear it on both clips when I play them back.

When I update the timecode on the Sony clips it does change the TC Start to be the same as the Braw clip, within a few frames because I use one hand on each camera to hit the record button. The failure occurs when I try to sync them on the Cut page.

The next method I tried was to update the Braw clip using the audio track information and that failed to sync on the Cut page as well. That should work.

The Sony clips use the LTC from the audio track anyway so it's TC Format shouldn't make any difference. Just to safe I set the TC Format to NDF (non drop frame). Same for the 6K Pro.

I tried using the Tentacle Timecode Tool to write a new file for the Sony clip using the LTC timecode instead of using Resolve "Update Timecode From Audio Track". That didn't work either.

I run the Tentacle Sync app on my Android phone and all the devices sync up fine. All the clips from my Sony cameras sync perfectly in Resolve.

I have a friend who uses a 6K and has no problem doing a Tentacle sync with other camera brands but he records at 23.97. That's not an option for me. I wonder if Blackmagic tested the 6K Pro at 29.97?

I normally use 3 or 4. My test yesterday only used 2 but I switched to a third and repeated to make sure it wasn't a tentacle. They were all fully charged and operating normally.

Were the cameras hooked to them the entire time?

No idea if this will help or not but a couple of things I picked up on with recording LTC to an audio track....

Davinci Resolve will only imbed the time code if you address the video file. It will not work on a stand alone audio file.

Tentacle routine will only work with wave broadcast files. Wave files without the broadcast addition will not work. It's the broadcast segment of the file that retains the metadata.

Yes, the cameras were hooked up to the tentacles the entire time. Yes I used "Update Timecode from Audio Track" on the video clip in the Media page of Resolve.

Factory Rest is my friend. I turned off the camera, took out the battery and plugged in AC adapter. After I powered up again, I did a factory reset from the Setup menu and reset all my parameters to the values I had used before. I ran another short test with two Tentacle Sync devices on different cameras. This time it worked!

...I did a factory reset from the Setup menu and reset all my parameters to the values I had used before. I ran another short test with two Tentacle Sync devices on different cameras. This time it worked!

Very interesting. Thanks for posting.

I'm not sure if this applies to the 6k too, as I have a 4k. If I inject timecode to the audiotrack using braw I get a delay of about 1/2 frame at 25fps. But if I connect my nanolockit to the audio in port and set input under settings to mic left/right instead, the timecode does not get recorded as an audio signal, but is 100% accurate in metadata.

For reference the sync is shown on the display (ext) if you toggle to timecode view.

I hope this helps.