Jump to content

Compression on the cart


mikeguarino

Recommended Posts

Interesting question since most of the most damaging compression is usually occurring OFF the cart --- in the wireless transmitter worn by the talent. If you are going all wireless with any of the most common types of wireless transmitters (Zaxcom being the exception) there will be limiting and compression artifacts that are often unavoidable. Utilizing additional signal processing at the cart is not a great idea since this would often further complicate the work that will need to be done in post. If you are on the sort of job, unfortunately, that has already marginalized the so-called "mix track" and you are confident that the mix is only a guide (there are jobs where post is going to go right to the isos and pretty much ignore your mix), adding a compressor or whatever isn't going to matter one way or another.

Link to comment
Share on other sites

Even if it's not going to post, the pix editor can add compression in Avid or FCP. They might not know as much as you do about fine-tuning it... but on the other hands,  1) they can monitor through speakers, which reveal compression very differently from phones,  2) they know what the other elements will be,  and

 

3) They have an UNDO button. You don't.

Link to comment
Share on other sites

While I don't believe in adding compression on location, I'm not afraid to pump up the volume on the mix track, particularly if I know the intended delivery is not going to have much of a final mix (if at all) and/or is destined for the web. I always, always run an unlimited iso at about -10dB lower, but I'll hit -7 or even -5dBfs on peaks with a limiter and not worry about it too much on the mix track for projects like this. If it were a feature or broadcast project, I'd be much more conservative about levels. 

 

I've observed far more complaints over the years from editors complaining about mix track levels being too low than too hot, assuming you're not clipping. I would be very careful with feeds to camera, however, since my experience is many cameras have input stages that tend to clip early. Ideally, a test prior to production is a good idea, so you can verify with the post crew that the levels are in a comfortable zone for them. If that's not possible, then I'd lean on the side of the mix track being a little hot rather than a little low, but not clipping and only occasionally hitting the limiter. 

Link to comment
Share on other sites

complaints over the years from editors complaining about mix track levels being too low than too hot

 

I suspect part of the problem is most editors are visually oriented and not technically trained, and NLEs display the audio waveform linearly, not as log. 

 

Even if you recorded an average of -6 dBFS it'd fill only half the waveform on their screen. 

 

-20 dBFS - what most of us consider 'nominal' for dialog - doesn't even fill a quarter of the waveform. 

 

A few NLEs have 'enlarge' functions for the track displays, but they're still displaying linearly. 

 

Drawing tracks as log isn't hard with today's computers. Heck, the freely distributed Audacity does that in realtime. But it would break the UI that most editors have gotten used to. (Same problem and 'enlarge' solution with the mainstream DAWs, too... but at least us audio ops know to look at the meters, and make our editing decisions based on the sound rather than the shape of the blob.)

Link to comment
Share on other sites

I suspect part of the problem is most editors are visually oriented and not technically trained, and NLEs display the audio waveform linearly, not as log.  Even if you recorded an average of -6 dBFS it'd fill only half the waveform on their screen. -20 dBFS - what most of us consider 'nominal' for dialog - doesn't even fill a quarter of the waveform. 

 

This has been a discussion before on the forum. What I generally saw in average mix track levels in telecine over the past 20 years was dialogue peaks that would go to maybe -8dBfs or -7dBfs or so (on a Tektronix 760 scope), averaging maybe 5-6dB below that. Just -20 for average level would probably be considered too low. In truth, I think what the picture editors were looking for on a daily basis (no pun intended) was consistency, so a scene shot on Monday would cut reasonably well with another scene shot on Thursday, or a week later or a month later. 

 

I do think iso levels should be another -5dB lower, so that even if the character were to scream or yell, it wouldn't get anywhere near 0dBfs. I usually try to keep things consistent just based on a known headphone value on my recorder, so if I monitor at the same mix level day to day, they're going to get reasonably-consistent levels that are reasonably loud.

Link to comment
Share on other sites

Every time we "ride gain" with a fader we are compressing, though inconsistently and inexactly. When riding gain on multiple microhones it's even less exact and less consistent. Yet we all do it to one extent or another. The highly acclaimed soft-knee limiters on my Sonosax mixer are just a form of compressor, and can very much enhance or reduce the quality of the production tracks.

 

Post production may claim to not like compression of production tracks, but then complain about tracks being recorded too low. With original dialog tracks, it can't be had both ways.

 

The answer is to understand the process and decide accordingly.

 

gt

Link to comment
Share on other sites

Regarding what Glen has said about "riding the gain" in our mix (even when that "mix" is only 1 microphone), I have always done a whole lot of this "manual compression" (even though it is "less exact and less consistent" I have gotten pretty good at it). I have been complimented many times by the people in post, thanking me for NOT riding the gain, not using any compressor or limiter, refraining from use of EQ, etc., even while talking about tracks where I had been doing just those things! My feeling on this issue is that even with wider and wider dynamic range available on the production track, a certain amount of manual compression (administered skillfully and with good judgement) is desirable. The rest of the massaging of the signal should be done in post where things can be un-done, repeated predictably, monitored in the proper work environment (a quiet studio with high quality monitors). 

Link to comment
Share on other sites

...In truth, I think what the picture editors were looking for on a daily basis (no pun intended) was consistency, so a scene shot on Monday would cut reasonably well with another scene shot on Thursday, or a week later or a month later. 

 

I do think iso levels should be another -5dB lower, so that even if the character were to scream or yell, it wouldn't get anywhere near 0dBfs. I usually try to keep things consistent just based on a known headphone value on my recorder, so if I monitor at the same mix level day to day, they're going to get reasonably-consistent levels that are reasonably loud.

+1 on consistency in monitoring levels, mix track delivery levels, and consistent lower levels on ISOs,

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...