Jump to content

got any "sound" advice for doc production?


Recommended Posts

Hi folks,

 

First post here in about 6 years I think!! I am a doc filmmaker but one that actually cares about good sound! I am setting up a new camera and sound setup and would really appreciate some advice and suggestions for how to do this properly. I am shooting mostly as a single operator, so I need to be able to do both picture and sound. I am shooting on a canon c300 mk3, and I have a mixpre 6 mk2 attached to the battery plate on the rear rods (recording in 32bit float 96khz). I have my trusty old 416 on the camera (I will likely update that to the cs-3e) and cos-11's going into a wisycom dual receiver.

 

i need to have record trigger working as im doing both sound and picture (sorry!!). At present I had the camera generating TC and sending it to the mix pre via the BNC to 3.5 aux in. Set to rec run. This worked fine for the trigger, but there is a strange thing happening where the timecode of the end of the clips recorded on the mix pre is then duplicated at the start of the next clip, it might be that there is some kind of delay or something happening on the mix pre end. It messes up syncing in post so that method is going in the bin. i was going to send TC from the mix pre to the C300, set to TOD, (is that right?) output from the stereo out on the top, and that works fine, but i cannot get the rec trigger to work anymore. I am talking to SD about that and hopefully there is a solution for that. 

 

I am going to be shooting the protests at COP26 in November and there will be multiple cameras, at least one other pro level camera with a TC input, so I have been looking into how to sync the cameras and what the best setups would be. What would you guys recommend? the mix pre will be always attached to the c300 and feeding it TC from its clock, so i was going to use the mix pre as the master, jam it to an Ultrasync One and attach that to the other pro camera. Does that all sound ok? am I doing anything really wrong? could the setup be better? Of course I would love a sound person, but there is no budget for one on this project.

 

thanks,

 

Chris

 

 

Link to comment
Share on other sites

Record trigger and timecode are separate things, right? I mean if you could set the tc to free run, that would (probably) make the timecode more usable. But then again i don't know if it's the timecode starting that's triggering the recording on the mixpre, or if it's a recording flag... 

 

Edit: just read the manual. You can record trigger with HDMI, or timecode. So unless there's a way for you to mute timecode when camera is stopped, you'd have to stick with rec run, and the duplicate frames. If you could go hdmi from camera to the mixpre, that would allow you to trigger the recorder separate from timecode. You'd still have to sync the camera and mixpre with Tc somehow. 

Link to comment
Share on other sites

What if you stayed with your current setup (in which the record-trigger camera > MixPre works, but with wonky TC on the MIx Pre) and split the camera TC output so it could go to both the Mix Pre TC in (for trigger) and to a spare audio track on the MixPre?  You might then be able to use the Aux TC feature of the NLE etc or a 3rd party app to post sync sound to picture?     https://www.videotoolshed.com/product/ltc-convert-auxtc/ 

 

Another question: you are going to a lot of trouble to add the MixPre to your camera (and making it kind of awkward to shoot I'd guess)...do you need that many more audio tracks than the 2 the camera has?  Do you need them all the time?  Do you have a bunch of wireless RX attached to your camera too (feeding the MixPre)?

Link to comment
Share on other sites

thanks for the suggestions guys!

 

yes the TC is also triggering the record function on the mix pre, but the TC lag or duplication or whatever it is seems like a pretty lousey bug, it basically makes the function useless. SD should be able to fix that.

 

i have an hdmi cable arriving today to see if i can trigger with that while the mixpre sends TOD TC to the camera, but its another (poor quality, lossy) cable that i dont want hanging out the back of the camera / mixer, if it was all locking connectors i wouldnt mind.

 

15 hours ago, Philip Perkins said:

What if you stayed with your current setup (in which the record-trigger camera > MixPre works, but with wonky TC on the MIx Pre) and split the camera TC output so it could go to both the Mix Pre TC in (for trigger) and to a spare audio track on the MixPre?  You might then be able to use the Aux TC feature of the NLE etc or a 3rd party app to post sync sound to picture?     https://www.videotoolshed.com/product/ltc-convert-auxtc/ 

 

Another question: you are going to a lot of trouble to add the MixPre to your camera (and making it kind of awkward to shoot I'd guess)...do you need that many more audio tracks than the 2 the camera has?  Do you need them all the time?  Do you have a bunch of wireless RX attached to your camera too (feeding the MixPre)?

 

i dont want the hassle of messing about with waveforms in post if i can avoid it (its the only work around at the moment as ive been recording a backup scratch audio on the camera itself) but im going to be shooting multi cam setups so TDO TC is the preferred way of syncing all the cameras (unless you guys have other suggestions?)

 

im using the mix pre for the 32bit float, i actually have it mounted on rods with the pots facing me so i can sort of mix levels while filming (I am a feature doc filmmaker, often working in difficult / remote / dangerous situations) but the 32bit float means i can recover anything blown out etc. i also do sometimes have 2/3 lavs plus the shotgun. also mounted on the back actually helps with the balance of the camera as i have a heavy cine zoom lens on the front that makes it very front heavy otherwise! 

 

Just 1 dual RX hot shoe mounted feeding the mixpre, or i can change the bottom and make it a slot in

 

Link to comment
Share on other sites

just to update in case it is of any use to anyone;

 

i got a HDMI cable running from the c300 to the mixpre, setting the HDMI output to 1080, I am able to get run / stop with that, with TC coming out of the mixpre into the c300 as TOD / Free run (with no frame delay) and vice versa, the frame delay is still present when using rec run from the camera to the mixpre whether i trigger with hdmi or the TC cable.

 

thanks,

 

Chris

 

Link to comment
Share on other sites

I would avoid the rec run system, the two machines cannot record the exact amount of frames I suppose. That is probably why you have this  frame delay. Also running the sound rec longer than picture is highly recommended as you might know. If you worry about missing a rec on the audio why not just record it continuously while in the middle of the action (and stick to free run or TOD TC for all units)?

Link to comment
Share on other sites

  • 2 months later...

I know it's too late, but I think I can provide a technical explanation of what's going on, along with a potential workaround.  So, for the future:

 

The way timecode works in digital files is the file gets a timestamp of the starting timecode in the file metadata.  Timecode for all later points in the file is extrapolated in software by adding the elapsed playback time to the starting timestamp.

 

The way triggering works is the recorder watches the incoming timecode signal.  It starts recording when the timecode starts advancing and stops recording when the timecode stops advancing.  That means it takes at least a one frame delay to detect when recording has started or stopped.  In reality it may be slightly more, since prudent engineering probably designs for some signal loss / dropped timecode frames.  That means, when you trigger recording, the audio file will always start at least one frame after the camera, and end one frame after it cuts.

 

Practically speaking, this makes very little difference, because a delay of one or two frames isn't going to be missed.  But it does mean that the two sources aren't perfectly aligned ... the slave device always lags slightly behind the master.  And, importantly for this use case, it means the audio take runs slightly longer than the video take.  When the the audio file is played back, the timecode is extrapolated from the starting timecode, and when it gets to the end of the file, it just keeps extrapolating timecode, even though the camera wasn't rolling (and thus wasn't advancing timecode) for that extra frame or two.

 

And, because it's record-run, when the camera starts the next roll, it uses the next timecode in camera sequence, which happens to correspond with the extra frame or two on the audio roll.

 

So, the workaround:  You need to post-process the audio files.  Before you sync, chop a frame or two off the end of the audio file (you may need some trial an error to find out how many frames you need to remove, since it may be variable).  There's lots of ways to do this.  I would probably use a command-line tool like ffmpeg to batch process them automatically (this can be set up to happen automatically whenever a new file is added to a directory), but there's lots of graphical tools out there that are easier to use.  The critical thing (and you'll need to test workflow), is you want to make sure the tool preserves the file metadata (aka BWF chunk & also some other items) when it makes the cut, or you risk losing the source timecode altogether.  I would look at Sound Devices WaveAgent, though I'm not sure whether it will let you automate batches of file trims.  Maybe someone else can chime in with their favourite batch processing tool...

Link to comment
Share on other sites

  • 1 month later...

Thank you for the replies and insights.

 

so i sent the mixpre back to Audio Ltd in the UK and they have been very helpful, they sent me a loaner while they were looking at my unit, and were able to run some tests with me to see what was going on. At the moment the mixpre series will either accept or send TC via aux input, but it cannot do both simultaneously, and it will only send a record / stop signal when set to Rec Run, which incurs this delay (the delay is random in length, usually between 5-9 frames) but they have sent a feature request to allow the unit to send and receive TC via aux in & out simultaneously, and also to allow recording in free run or TOD etc. Hopefully that will be possible and the issue will be resolved. (eg most cameras can output & input TC simultaneoulsy from any of their SDI ports, not only the TC one)

 

Also just FYI the HDMI TC works perfectly, no delays in triggering rec / stop and works with TOD / Free Run etc, the issue is the very flimsy port connection (which they replaced) and the liklihood of the cable coming loose while operating. I have a rig setup (see photos) that should prevent that.

 

On 12/24/2021 at 4:05 AM, The Documentary Sound Guy said:

You need to post-process the audio files.

thanks for the suggestion i really appreciate you taking the time to think about this, but im loathe to have to introduce time consuming and potentially damaging workarounds to fix a feature that should work (and does work with other setups). Hopefully SD engineers will be able to fix it in a firmware upgrade!

IMG_3564.jpeg

IMG_3565.jpeg

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

 Share

×
×
  • Create New...