Mastering Level -20 dB? what does that mean?

[TheGmiester161]

I'm doing the post production sound for a student film. I'm learning the process myself. I've done the production edit, ADR, foley, sound effects editing. Now I have to mix the project.

The guidelines I was given by the student says "THE SOUND MUST BE MIXED AT -20 dB".

What does this mean? How do I do this? I'm recording in pro tools. All I've done is made sure it doesn't go over 0 dB and that everything is at a "healthy" level.

 

[bennychico11]

ahhhh....post versus music levels

In the post world we mix/record with a reference tone in mind. You know, that annoying 1kHz tone usually recorded with color bars on top of it.

In short, what he means is that when playing back a 1kHz sine wave it will equal -20dbFS. Typically giving you 20dB of dynamic range. Now a sine wave is a constant signal. So peak levels are going to be at -20dBFS while RMS or averaging levels will ALSO be at -20dBFS. Why this is important is it means that your average mix levels (I like to follow RMS levels) are going to be around -20dBFS. Your peaks during a mix will be much higher than this (unless you have a peak limit rule set, you can be much higher that -10dBFS at times). The -20dBFS average levels (when used in conjunction with a reference tone at the beginning of the mix) equate to 0VU on the analog side. And since VU meters usually translate as averaging meters this means your average level will be around 0VU

hope that helps a bit. I know it can get kind of confusing

 

[masteringhouse]

What does this mean? How do I do this? I'm recording in pro tools. All I've done is made sure it doesn't go over 0 dB and that everything is at a "healthy" level.

Basically means that peaks should not go over 0dBFS (actually a 3 db cushion isn't bad) and "healthy" should be at -20 dBFS.

There are some digital meters that support both average (AKA RMS) and peak levels (Pro Tools meters are peak only). Essentially what the guidline states is that the average should be around -20. There are also some meters that support Bob Katz's K-System, this level equates to K-20.

 

[mshilarious]

In short, what he means is that when playing back a 1kHz sine wave it will equal -20dbFS. Typically giving you 20dB of dynamic range. Now a sine wave is a constant signal. So peak levels are going to be at -20dBFS while RMS or averaging levels will ALSO be at -20dBFS.

Ooh, that's not quite true. The RMS of a sine wave will be .707 the peak amplitude, or -3dB.

-20dbFS is the RMS reference level for movies, as promulgated in the THX standard. That means it should be the RMS level of your mix, or stated another way, that you have 20dB of headroom. That also means you should use most of that headroom for peak sounds (explosions, etc.). You should calibrate your monitors with a -20dBFS RMS test tone (like the sine wave, peak would be -17dBFS) to 85dBSPL. That would be the playback level in the theater.

 

[SouthSIDE Glen]

You should calibrate your monitors with a -20dBFS RMS test tone (like the sine wave, peak would be -17dBFS) to 85dBSPL. That would be the playback level in the theater.

That's awfully loud, isn't it? That means peaks (explosions, music swells, car dealer commercials) would be approaching 105dBSPL. That's pretty dangerous territory. I doubt that I've heard any of my local theatres go quite that loud....unless you're talking SPL a meter in front of loudspeaker, maybe.

G.

 

[bennychico11]

Ooh, that's not quite true. The RMS of a sine wave will be .707 the peak amplitude, or -3dB.

oops! you're right. there will be a slight difference. I mean to write "approx" or "almost". My meters have small type

That's awfully loud, isn't it?

yeah, I don't think I'd use a sine wave to calibrate the monitors. pink noise is more appropriate to use.... I get an ~5dB difference between the two.

 

[mshilarious]

That's awfully loud, isn't it? That means peaks (explosions, music swells, car dealer commercials) would be approaching 105dBSPL. That's pretty dangerous territory. I doubt that I've heard any of my local theatres go quite that loud....unless you're talking SPL a meter in front of loudspeaker, maybe.

G.

Well, I have, which is why I don't like to go to theaters. And it's why I run my DVD player through a compressor Actually I think a lot of theaters exceed 85dBSPL; I stuffed cotton in my ears to watch "Austin Powers", and I mean the dialog. But that is the THX standard, which is why this otherwise not-audio-aware student is asking for it.

Anyway, 105dBSPL not dangerous as a peak level. The OSHA standards are set based upon C-weighted RMS, not peak, and even then the standard for 105dBSPL is 1 hour exposure. Most loud peaks in a movie are LFE stuff anyway. And there is nothing to say that dialog has to be 85dBSPL; it should sound as it would in real life; the music would be closer to the -20dBFS reference level, probably with peaks at -6dBFS. That last 6dB of headroom wouldn't get used much.

(NIOSH thinks the exposure standard should be lower, more like 5 minutes, but in most films that's about how much peak time there would be. That would also mean that even quiet rock concerts would be banned!)

The important concept is that if a theater is calibrated to -20dBFS = 85dBSPL (which it is supposed to be), then if you don't mix to that, I think it is likely your dialog would be too loud!

That's because as the OP says, he is peaking at 0dBFS with a reasonable level. If he is used to doing rock mixes at -14dBFS RMS, and mixes the movie at the same monitor level, to a comfortable listening level in his studio, when played back in a theater it would be 6dB too loud. So I'm guessing the monitor calibration will result in the OP remixing the soundtrack to be quieter than it is right now . . .

Katz talks about this in his book.

 

[mshilarious]

yeah, I don't think I'd use a sine wave to calibrate the monitors. pink noise is more appropriate to use.... I get an ~5dB difference between the two.

OK, there will be a difference in peak level between a -20dBFS RMS 1kHz sine wave and -20dBFS RMS pink noise, but the difference in C-weighted SPL reading depends on your monitors, actually, as well as the frequency you select for the sine wave. Off the top of my head, I don't remember the THX reference signal, but it is important to know that so as not to be off.

The important concept I am trying to stress is that any calibration done for mixing music is probably too quiet for mixing film (resulting in a film mix that is too loud), unless it's a classical music studio or something.

 

[SouthSIDE Glen]

Well, I have, which is why I don't like to go to theaters. And it's why I run my DVD player through a compressor

This is something I've been wanting to try for a loooong time. You KNOW I am as pro dynamics as it gets, but with DVD video, if you are watching on anything other than a THX home entertainment system (which I rarely do myself) the dynamic range is just way too high, IMHO. You have a choice between the loud passages giving you an ear bleed so you can ear the dialog, or losing the dialog so you can live through the movie without getting tinnitus.

It's at the point where my senior citizen mother can't even watch movies at home anymore unless she rides the volume control ike a fader jockey, which she i not likely to do.

I stuffed cotton in my ears to watch "Austin Powers", and I mean the dialog.

Well, so did I but that had more to do with the actual dialog than any loudness issues. (j/k...sort of...)

And there is nothing to say that dialog has to be 85dBSPL; it should sound as it would in real life; the music would be closer to the -20dBFS reference level,

That makes more sense, I'll buy that one.

(NIOSH thinks the exposure standard should be lower, more like 5 minutes, but in most films that's about how much peak time there would be. That would also mean that even quiet rock concerts would be banned!)

Well, let's be honest here; even most "quiet" (relatively speaking) rock concerts are too loud for their own good.

I find them to be much like the alchohol-serving venues in which many of them are held; there is common abuse of the privelige. Sure, a BAL of 0.8 is the "standard" for drunkeness, and having a beer every 90 minutes will probably keep you legally sober. But out of every 100 drinkers in the audience, how any of them actually walk out of that bar at the end of the gig in full control of their facilities, whether they are below 0.8 BAL or not?

The standards for volume may say what is safe and what isn't; but I'll bet that over half of the folks that walk out of the club at the end of the show have at least temporary tinnitus and some at least small degree of cumulative long-term ear damage (assuming they are not using protection...or plugging their ears either )

G.

 

[bennychico11]

I stuffed cotton in my ears to watch "Austin Powers", and I mean the dialog.

well, despite a theater's best effort to get the speakers calibrated to a certain SPL...they don't stay that way for very long. With the many prints that move around through a theater weekly, daily or even every few hours you'll have a drama with whispering throughout one movie and Austin Powers the next show blowing your ears out because Burt the 90 year old movie goer can't hear a word they were saying in the drama and refuses to wear the assisted listening devices.
Plus you have the different sound tracks which go through different decoders. The trailers are typically one flavor (Dolby) and then for the actual print it can jump over to another one (SDDS).

I used to work in projection at a movie theater and the master levels were all over the place for every individual theater. In the end, it's the customer who determines the calibrated speaker level

 

[mshilarious]

It's at the point where my senior citizen mother can't even watch movies at home anymore unless she rides the volume control ike a fader jockey

Two words: dbx Minicomp. Find one on eBay. I love that little guy, it's perfect for this job. Set it for her, cram it behind her DVD player, and she'll never have to think about it again