>>> In a studio setting, though, how do the pruducers of music CDs determine how much compression to use? <<<
Nowadays a lot of CDs are mastered so that the song is as loud as possible. At high levels of processing distortion will kick in, but it seems like acceptable levels of distortion are getting worse as music execs are trying to make the CDs the loudest possible. This way it'll get more attention in stores and possibly on the airwaves. see:
http://www.prorec.com/prorec/articles.nsf/articles/8A133F52D0FD71AB86256C2E005DAF1C
The answer to how much compression to use is: as much compression as possible.

>>> Is there a general consensus that says something like, "average volume should be about 10db below peaks?" <<<
I don't think there is any consensus. However, a lot of CDs will be about equal volume.

>>> When I hear an orchestra on a CD, is there no compression at all? What about popular music? <<<
Classical music is usually recorded with very little compression, and pop music a lot.

>>> Finally, if I wanted to make audio recorded via Minidisc (no compression) sound like audio recorded via camcorder, what compression settings would I use? <<<
What are you trying to do? You should typically disable ALC/AGC on your mini-disc and camcorder since it creates pumping (background rushing in and out, which most people find annoying). Also if you use 2 different microphones and mic placements, both sources will sound different anyways.

To slightly add to what Glenn is saying, your final output of your project should be in the -.3 range. If there is mostly dialog, very little compression unless the vocals erupt into a screaming rage and you can't control it. If there is a lot of music/action sound/stingers...you'll want more compression. Start around 1.5:1 or so, and you'll find a nice swing in either direction. Compression for video is fairly critical if you want dialog that stays fairly out front without going over the top.

What is this term, "compression?" Are you talking about normalizing, wave hammer, or is this something completely different?

Thanks,

Ted

Normalizing is bringing an audio file up to a preset point without affecting the audio dynamics. Compression is a form of controlling the dynamics of the audio, so that quiet areas might become louder and loud areas become more quiet, or to the more common use, the louder areas of audio are brought more into line with the quiet areas, allowing for overall louder output without distortion. Used for control or used for effect, compression is fairly well necessary on most digital audio recordings, particularly with DV audio given the nature of the craft.

Spot, I guess I'm confused by your terminology. What does "-.3" mean? That the highest peaks reach -0.3db on the mixer?

And when you say "1.5:1," are you referring to the "Amount" slider in the compressor plugin?

"Yes" to both questions. Look at your meter on the master bus. Peaks (loudest points) should be hitting at about -3. Anything higher than -6 is fine, nothing over 0, ever.
I believe there is a 1.5 preset in the track compressor....but not sure.

The "rule of thumb" for compression ratio is somewhere between 1.5:1 and 3:1 these days, depending on:
a) the dynamic range of your source, and
b) the recording limitations.

To understand this, you need to realize that a live string orchestra has a GREATER dynamic range than a Metallica concert. Also, an optimized digital recording has a greater dynamic range than a cassette recording, not because it can record louder signals, but because of the level of the noise "floor." Also, modern acoustic sound reinforcement and recording capabilties are well in excess of those thirty years ago, although we've grown to know and love the warm LP recordings from that era. 2:1 was considered about the maximum usable compression ratio back then.

In spite of the cryptic explanation above, just keep in mind that if you go above 3:1 compression on raw recorded material, you should have a damn good reason for it (other than altered states of consciousness or a bad sound board op).

The amount of compression you should use depends on the medium too.

3:1 compression could be a lot depending on what your threshold is (and what your sound is like).

There are many different approaches to compression. Here's one of them:
(from Jay Rose's site, http://www.dplay.com/playroom/recmag.html)
>>> RM You keep talking about processing. Do you have a standard setup for that too?

JR Yes and no. I have a few standard things I'll do to every voice track, and some other things I do to the entire mix, but it's always tweaked to the individual announcer and spot.

For the voices, I'll start with a sharp rolloff around 90 Hz. Sounds below that are just wasting power without contributing to intelligibility, and the rolloff also acts as a pop filter. But it has to be sharp, or else you start thinning out the voice and everything sounds bad. Most console filters are too gentle, and you can forget about trying a cutoff with a graphic. Back in the analog days, I ran voices through a Crown VFX -- it's a sound reinforcement crossover, with 18dB/octave filters. I'd set it as if I were biamping at 90 Hz, but never hook up a woofer. These days I use a filter section in an Eventide DSP4000 UltraHarmonizer. The chassis says "Harmonizer", but it's also a very flexible equalizer and dynamics control.

Once the voice is filtered, I'll add a tiny amount of eq if necessary. Never more than 3 dB, just a gentle warming around 200 Hz and some extra intelligibility around 1.75 kHz. Then I'll crunch, first with a very slow look-ahead agc (it delays the signal 2/3 second and ramps the gain to accommodate peaks before they happen), then a 10:1 compressor around -15 dB with some extra de-essing in the sidechain, and then a very fast hard limiter at -2 dB. When you do all this right, you're not aware of any compression at all. There's none of that "AM Disk Jockey" sound, just an overall loudness. I've built all these processes into a standard voice processing patch that I keep in the Eventide, and I save different versions for different announcers.

This much processing requires a very clean signal path. Compression tends to make a distorted track sound even more distorted, and while a certain amount of fuzziness can be nice in music it never helps a commercial. Both the Eventide DSP4000 and the Orban workstation run at 24 bits, and I pass signals around as AES/EBU, so there aren't any extra conversions to add distortion. If I'm getting the announcer via ISDN or DAT, the signal never enters the analog domain at all (except for my monitors). This extra cleanliness means I can add more processing without a fatiguing, squashed sound.

Of course, not everyone values a clean signal. The operations manager of one of the larger Boston ad studios was once quoted in the trade press as saying "Don't waste time chasing down that last percent or two of distortion. It's a lot more important to have a neatly-typed label." I don't agree... but you can't deny they're making a lot of money. <<<
His target format I believe is radio, where you need compression to bring the volume up since the people listening have to hear over lots of background noise. For other formats you don't have to compress as much as you would for radio.

3:1 is a lot anyway...On a weak and feminine voice, this would be very mushy whereas on a Howard Stern voice, it would be pretty good if the presence was there. Too much compression, bass becomes more dominant and treble gets mooshed if standard presets are used on an average file.
Compression is so very, very tied to the audio type, there is simply no 'norm.'
As a perfect illustration, the 10:1 Rose recommends is a tremendous amount of compression, but you'll not hear it.
Think of a compressor as a person sitting at a mixer pulling a slider up and down for your mix, but doing it as or before a sound occurs. Compressors are a beautiful thing. Next to my preamps and monitors, I've got more money in compression hardware than anything else. At least 50K tied up in several different comps including 4 DBX Blues and Summit's. And an Orban and an 1176. Now, I use more WAVES and UAD than anything else. But they are HELL on the CPU when you have more than 6-8 running.

Rule of thumb? Not as much as this much :-)

Using audio compression effectively is an art. Familiarize yourself with the audio characteristics of some commercial releases that are similar in nature to what you are doing, then adjust the compression and EQ using your ears. We are all born with the ability to hear, however, the ability to listen is something that is learned over time with a great deal of effort.

John

Thanks for this thread, all contributors.

Would someone care to add a few words about the use of the Track Noise Gate - one of the three FX that SoFo put there as default, obviously for a reason.

i.e. - When you use it, how to decide which settings etc.

Thanks

Word of warning about noise gate: It will leave gaps of silence between the sounds it lets through. You have to replace those gaps with something else! Ambience, music, machinery, street noise, anything - or your sound track will sound like it was made by a complete beginner.

Not much to add to the excellent compressor advise above - BUT: do remember that when working in Vegas you can add different compression to different tracks. If you really want to dig into this field, you can even add different compression to different frequency bands of a track, by using the Multi-Band Dynamics FX.
The Multi-Band Dynamics may be a Sound Forge feature - they are put in the same bank as the Vegas ones and although they are mostly the same, SF has a few that Vegas does not. Even if it is, and you don't have it - you can approach the effect by copying the track as many times as needed, filter out certain frequency areas and apply dynamics to them one by one (track by track).

Anyway, don't get TOO hung up about rules of thumb. Have you ever seen two thumbs that were identical?
Tor

Peter,
Noise gate is used in music production to eliminate electric hum, finger noise and other unwanted sounds between the sounds that you want to hear. In the final mix the gaps will not be noticed because of other instruments, reverb etc. Also, it's easier to notice the use of noise gate in a location recording, because there is more natural beckground noise, making the gaps more conspicuous.
In a video production you CAN use noise gate to get rid of noise between the good sounds in a dialogue of FX track. But refer to the advise in my post above. After I got Sony Noise Reduction I have not used noise gate once.
Tor

Thanks Tors,
Although I do more video than music, I am working on a music multi-track right now, and am enjoying slowly learning new skills to enhance the quality.

I have a female vocal, recorded right here in my editing room, and got a really nice clean recording - noise floor was very low - near -80db, so it seems noise gate won't be necessary, but it will come in very handy with some video projects.

Regarding compression, in my explorations I had tried applying 6:1 to the vocal and it made it too "strident", but changing this down to 2:1, thanks to this thread, worked wonders - it now adds a bit of strength without over doing things.

<<adds a bit of strength without over doing things>>

=the very idea!

Tor

Fun/Funny read from a wanna-be musician who clearly doesn't understand the rules of the game. The label doesn't determine those levels, the producer and band do. His comparison to all caps is wrong and off base, but it's a good thought process. Just needs to be refined. Needless to say, his show of shaved distortion is also inaccurate, and has as much to do with his audio editor as it does anything else. He fails to take into account that early albums were all done on analog where/when it wasn't possible to approach the levels we have now. My music, much like classical music, is exceptionally dynamic. Compression is part of life, and required for anything. Otherwise, it's nearly impossible to capture every nuance. When you hear that breathy, sweet top end sound in a Whitney Houston or Enrique recording, it's due to compression being able to bring that out. But as John Cline eloquently said, it's an ART. I realize everyone wants a 'fix-all' plug that just intelligently does it. But computers can't hear, they can only interpolate. And guessing means that all of the nuances that compression was designed to enhance, will disappear.

What a great thread!

Yes, 3:1 is a lot, I would only go that high with strong miked leads and weak chorus, and might use just a bit of limiting in that situation too. I have yet to use noise gate with music, I just don't like the "holes" in the sound. If I have a noisy undermiked recording, I sometimes superimpose a very weak, compressed, inverted image to lower the floor a bit. I guess I take it for granted that everyone will use their ears and sensitivity and apply ANY processing conservatively ---- but, then I remember some of the sound butchering I did in my twenties, it didn't even qualify as bad rock . . . . . . . . . . . .

This is OT to the original question about studio compression, but I do a lot of cleanup to basic "summer vacation" video: On-board mic, filming at a distance stuff. I have to use the on-board audio compression because of the variation in levels which I cannot control. e.g., I'm doing a great job picking up the sound of the boat on the lake half a mile away, then sometime walks up to me and asks "hows it going :-?" So, on the computer, I'm faced with keeping the high-levels under control while trying to boost dialog and sounds in the -30db range. Even lower sometimes. Admittedly, this is impossible to do "satisfactorily", but I have no choice. (At least my listeners aren't expecting professional sound; they just want to hear themselves :-)

The issues are: (1) Background noise is hell. (2) The significant amount of compression required produces significant distortion. (3) You can't ride the volume manually because its too noticeable.

For the background noise, a noise gate is unacceptable because you cannot allow the sound to suddenly go to -inf. Instead, you can use an expander to reduce everything, say below -40db. Then you can add a compressor downstream. Alternatively, you can duplicate your audio track and leave one of the tracks untouched to ensure some background at all times.

For the distortion I dunno that anything can be done. I'm not even positive where it comes from. Although, given a 5:1 compression with a threshold of -50db (an extreme example), its obvious that I'm running out of bits. Still, there are times when this should not be the issue and I still get unacceptable distortion. Drives me crazy because I know how a compressor is supposed to work and I never had these problems in the analog realm.

Anyway, just an OT for the amateur hobbyists like me. If anyone has suggestions, TIA.

Consider your target listening environment. If your listeners will be in a quiet office or residential setting you can have more dynamic range and less compression. If a car or a hotel conference room with a 70dba A/C is your target environment you will need considerably more compression and less dynamic range.

Finalizers are nice too. I use less compression now that I let the Ozone finalizer take care of the little stuff that gets by the compressors.

For music my approach is to apply some mild track compression and volume envelopes to get the correct balance for the entire piece while preserving most of the dynamic range. I then apply compression and a finalizer to the master fx chain to get the desired dynamic range for the final product. This helps keep songs from getting skrunched and loosing instruments in the mix. For me, at least, it is more repeatable.

As has been said you still have to listen to make sure it sounds correct. The bypass switch is your friend.

Larryp

What is a finalyzer? That is a new term for me.

Regards, Paul

Good thread but this brings up a question.

Should a voice track be peak normalized to 0 db before applying compression? If not what level?

Thanks,

Marc