Jump to content

Avid Workflows for Digital Cinema Cameras 


octophonic

Recommended Posts

Um… this might go on a bit, anyways. I'd like to share the new AVID whitepaper on Digital Cinema Aquistition for discussion. It arrived yesterday and got me thinking about a few questions to put to the experts on the front line.

 

My experience has been setting up workflow for films through Pre-Prod into Production and then completing Post Sound to delivery. A unique sound position, interfacing with Directors wants vs Producers needs, and all the R&D in between to streamline Camera Dept and Picture Editorial to Post Sound and Mix. This work has afforded me a huge respect for the Production Sound dept and I've been lucky to work with some of the best.

 

The entire pdf is attached below but (to shortcut all the other BS) the paragraph I'm curious to hear any comments on is this...

 

Audio Workflow Considerations

Regardless of camera type, the decision between using a single or double-system audio workflow is based on a variety of factors, including crew size and budget. Single-system workflows can be considered more convenient, with picture and sound already in sync as part of the recording, but this method does come with some drawbacks. One particular challenge is that the camera operator will need to pay attention not only to composition, focus and movement, but audio levels as well. A double-system workflow solves this issue since there is a separate recording device and a person dedicated to monitoring the audio. Double-system recording also offers the advantage of being able to capture more than the 4 audio tracks typically found on digital cameras.

Third party applications are often used as part of the dailies process and can provide the ability to sync picture to double-system sound. While many of these free applications are useful for creating “review” dailies not intended for actual editorial, they don’t usually parse or retain enough of the audio metadata, making audio conform rather problematic in post. 

 

My thoughts are...

 

With the transition from film to digital acquisition it has always been a good idea to have embedded audio (even just the mix track) in the picture media. This has obvious advantages for dailies, etc, but more importantly as a safety check for Editorial against what the Sound Mixer is recording versus the Camera Dept. Having pretty much dealt with every combination of TC/fps/pull up/down/sample rate cluster freaks over the last 20 years this is hugely helpful when "fixing it in post". Usually the problem has been created in post by clever editorial software or inexperience when sound and picture come together. Sometimes from a problem in Production. Either way, too late to fix, make it work.

 

There are a range of great products for returning the audio and TC to the camera, and some have hugely helpful metadata embedded which can be wrapped up in XML or MXF data to help out in post. For example linking camera audio to original Sound Mixer Media (TC and Userbits or Roll #) or as a safety check for TC slates to show what slate/take the mixer has used versus camera on actuation.

 

Questions. Is technology at a level where we can reliably provide an AES return of up to 4 channels (preferably wireless with TC & metadata) to camera on the shoot day, so that the suggestion of the camera operator having to be concerned with levels is not an issue? If not then can the Sound Mixer be sending an AES feed to devices like the new range of Video Devices or similar for picture and sound capture, to avoid syncing dailies? With metadata to automatically relink to wider Poly Files. Is anyone doing this already and if so how is it working out? Some may be already be supplying this to video village. Sound rolls/splits would still be delivered later with sound reports, but as a secondary delivery for the wider discreet channels. Does this make your job easier or is it complicating things unnecessarily from your POV?

 

I know how time consuming post sync of sound and picture can be with editorial trying to turn around the volume of media created with digital cameras. I also know how militant some Editorial Depts can be the first time a software link of TC and Metadata fails in post, and fair enough. How close are we to a workflow where camera audio is a reliable digital copy of the mix track (with TC/metadata) so that the camera guy can shoot, picture dept can edit, and soundpost can link to your discreet tracks with confidence in post?

 

Thanks for making it this far, please chime in so I can give AVID some feedback. We assume too much in post. Also, thanks for this site Mr JW. It is a great resource/legacy for now and the future. 

 

 

 

digital_camera_workflow_whitepaper.pdf

Link to comment
Share on other sites

The fundamental problem with single system sound is that akaik you cannot return 4 channels at the same time. You have to manually switch on camera either 1/2 or 3/4. Therefore making confidence monitoring of the 4 channels in realtime unpractical. I might be wrong though.

As long as it is like that double system will stay the way to go to get the best results when things get a little elaborate. Everytime client wants sound straight to camera i always interiorly go "meh" unless it s a one person sitdown interview. Of course my goal is always to accomodate the client within realistic workflow for the quality if his project.

For instance i COULD send 6 lavs to a 5D if they insists but you can be damn sure i ll be rolling a iso backup and making sure we use a slate.

Link to comment
Share on other sites

I'll send whatever audio the show wants to the camera that I can accommodate within the alloted budget and time+attention available during the shoot that the camera dept will allow, but I'm rolling my own rig in any case.  A point Avid conveniently leaves out is that non-cooperation by the camera dept re sound and even TC+sync is an ongoing factor, thus a soundie suggesting a ref track to camera (as well as TC) will still get blown off by camera a significant percentage of the time unless production intervenes.  (My rule of thumb: if the DP says no then no it is.  I'm not going to argue with the DP about this.)    Any audio I send to camera on anything beyond a sit-down interview shoot is generally not monitored a whole lot, and adding any new cableage between me and camera is not going to happen, esp re discrete returns.  Producers and post need to understand that as the audio requirements of jobs have increased, the sound crew size has decreased (bag world), cameras present an ever- expanding and sometimes bewildering range of connectors, gain structures, and less and less real estate to mount any external devices on while being ever more mobile and needing to move faster all the time that double system and post sync is what the drill is going to be.  There are lots of options for how to do this, get over the 1980's picture+sound VTR recording prejudices and free camera and sound to do their jobs the best they can.

 

philp

Link to comment
Share on other sites

This Avid white paper is very well thought-out, and I think in general most of what they say is correct. The only two or three cases I can think of where it makes more sense to just record sound only in the camera:

 

1) B-roll with no dialogue

 

2) news or documentary shoots under very hostile conditions (gunfire, land mines, etc.)

 

3) "live news" situations. 

 

I don't think it makes any sense for scripted narrative at all. I see only negatives and no positives. Sending a mono scratch track to camera works fine in terms of editing reference audio.

 

The best comments I can think are the specific notes from Avid that the camera operator already has too much to worry about -- composition, focus, exposure, battery life, media capacity, etc. -- that they can't worry about sound on top of that. Even if they did, riding levels is going to be nearly impossible. 

 

 

 

Questions. Is technology at a level where we can reliably provide an AES return of up to 4 channels (preferably wireless with TC & metadata) to camera on the shoot day, so that the suggestion of the camera operator having to be concerned with levels is not an issue? If not then can the Sound Mixer be sending an AES feed to devices like the new range of Video Devices or similar for picture and sound capture, to avoid syncing dailies? With metadata to automatically relink to wider Poly Files. Is anyone doing this already and if so how is it working out? Some may be already be supplying this to video village. Sound rolls/splits would still be delivered later with sound reports, but as a secondary delivery for the wider discreet channels. Does this make your job easier or is it complicating things unnecessarily from your POV?

 

Sound syncing is fast, easy, and does not even required a super-skilled person. I know: I did it for many years, to the point where I could do it by eye without timecode. Timecode is much better. As I've often said here, it usually takes :30 seconds per take. I've done it almost literally in my sleep. Even if you have 200 takes per day -- which is a fairly heavy shoot -- that's under 2 hours total for syncing. The slate also provides info for the editorial department in terms of scene name, take name, camera roll (folder) number, prod. date, plus provides a visual indication of timecode, typically time-of-day code. All these things are useful when logging shots. 

 

People make syncing dailies out to be some horrible, drudgerous, difficult, pain-staking task when truly it's not. My cure is often, "hire an assistant editor!" They need not be expensive, and a good AE will save the editor much-needed time. A good assistant can come in at 6AM and have everything synced, edit bins organized, notes taken, and coffee brewing before 9AM. 

Link to comment
Share on other sites

"Sound syncing is fast, easy, and does not even required a super-skilled person. I know: I did it for many years, to the point where I could do it by eye without timecode. Timecode is much better."

 Indeed, but these days in lieu of TC, many folks use PluralEyes which works remarkably well with a reference track.

Link to comment
Share on other sites

Indeed, but these days in lieu of TC, many folks use PluralEyes which works remarkably well with a reference track.

 

Except in cases where it doesn't. What baffles me are the nutballs who don't understand why the track won't automatically sync with Pluralize when they've used a camera mic on the camera tracks, and wireless and/or boom on the production sound track. I also seem to recall there were issues when Pluralize were used with a big set of multitracks, like 8 or 10 track poly files.

Link to comment
Share on other sites

I think the PluralEyes workflow, which has to have each clip checked and works only as well as the fidelity of the camera audio lets it (see DSLR camera mic sound) is being supplanted by a Resolve workflow.  The latter, given correct TC on both camera and sound, seems faster, more automatic and more foolproof in my observation.  Good, small, camera mountable TC generators have come way down in price now, and Resolve even allows syncing to LTC on an audio track.  Now even DSLR footage can be synced via TC.  All that is required is the free version of Resolve.

 

philp

Link to comment
Share on other sites

I have specific language about this on my rate card:

 

"As a matter of policy, if the camera body itself does not offer full size XLR inputs, then I will not do single system mixing. "

 

This guards against being presented an adapter box (juiced link, etc) for a DSLR or being expected to mix to a RED.

 

"If I press the 'record' button, whether as primary, backup, or transcription, in addition to sending a mix to camera, this configuration will be, at minimum, charged as Package B." (Package B is the basic double system package)

 

And this prevents any kind of tedious discussion about doing single system with hops and being expected to roll a backup without being paid for it. They don't pay, I don't roll. Since I began presenting it like this, I've had zero pushback about it.

 

But above all, the most helpful thing for me has been to request a set of specs from post on every new project. They list, in writing, all of their tech preferences, and I can negotiate the appropriate kit to match. No emails after the fact about sync issues.

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.

×
×
  • Create New...