Compressing questions

[eduardo_b]

It seems that every article I read on compression starts out by saying how confusing it can be trying to understand and use it successfully. So here I go.

I understand the purpose and principles of compression, and am familiar (more or less) with general and multiband compressors. I am aware that compression is often used to make a song louder by bringing up the noise floor (or limiting the peaks). But I have questions these articles don't seem to address.

Compressors must reduce the dynamic range of music in order to achieve the goal of making the overall sound louder, correct? So it would seem there is a tradeoff between having extended dynamic range or having enhanced sound. But is this application good for any kind of music -- such as both classical and rock, for example?

It would seem that compressing classical music would make it better for certain environments -- car, portable -- where ambient noise levels make soft passages harder to hear, but in general it wouldn't seem to be a good idea. But one wouldn't compress, say, acoustic guitar, would one? To what purpose?

What about ambient or atmospheric music (my genre of choice)? A little compression or none?

I read that in the 60s and 70s, compression was used with a light hand, and that's why the recordings from those years sound different than current music. Any insights or thoughts about this?

I'm trying to decide how much effort to give to compression, and what plug-ins, if any, would make the most sense.

Finally, compression is a mastering, mixing or individual track technique -- or any of these?

[Andywanders]

... I am aware that compression is often used to make a song louder by bringing up the noise floor (or limiting the peaks). But I have questions these articles don't seem to address...

... I read that in the 60s and 70s, compression was used with a light hand, and that's why the recordings from those years sound different than current music. Any insights or thoughts about this?...

... Finally, compression is a mastering, mixing or individual track technique -- or any of these?

Personally, I feel that compression is an overused and often overated tool in mordern audio production. Try to record a song or piece of music, in any style, without using ANY compressors. If you can do that and it sounds good, then your techniques are good. Problem is these days most people believe all the hype about being 'louder' than the competition. Bollocks..!

Digital recording allows a 90dB dynamic range and most of us don't even use one tenth of it. Therefore, ultimately the music suffers.

Here's a good read on the subject...

http://www.cdmasteringservices.com/dynamicrange.htm

Cheers

[P.T.]

Compression is often used on acoustic guitar because of uneveness of playing. Same with voice. Uncompressed voice usually sounds bad because there are very few singers with good technique.

Using compressors for making a song louder is a fairly recent use and is not the main use.

[BertKoor]

In classical music compressors are used to bring up the very soft passages and bring down the very loud passages. A real symphonic orchestra has a dynamic range of about 110dB whereas a CD can only has a range of 96dB. Reducing dynamic range can be desirable, so the listener doesn't need to grab for the volume knob to hear soft passages while not wanting to annoy the neighbours on loud passages.

In rock music compressors are used all over: drums, base, vocals, and on the master. Not on overdrive guitars, since the amp already limited the dynamic range.

For your type of music: experiment. Can do wonders on some tracks, or do serious damage, depending on how it was performed and what you're after...

[Hovmod]

It's really really wrong to think of applying destructive tools in the production of intentionally extremely dynamic music like classical to accomodate those who choose to listen to it in their car or in a noisy environment.

I routinely turn off the computer when I want to listen to classical music, because of fan noise.

[aciddose]

the thing about compressors is that they do not tend to operate immediately. you can adjust the attack and release settings to make a compressor act like an automatic gain control, for example, if you want to mellow out the peaks and valleys in classical music. obviously using the instant attack and fast (50ms) release you'd do on some electronic percussion for your next gabber track isnt going to work well with classical music.

compression is a whole range of effects, infact, you should think of the whole collection of compressors, expanders, and gates in terms of "dynamic processors". all three have different uses, and there are many more sub categories of each of those as well.

one interesting thing about how people get upset when you suggest compression for dynamic music like guitar or classical is that, our brains apply a compression to everything we hear anyway. without it, it would not be possible for us to have even an 80db dynamic range. actually, i'm not really sure what the dynamic range of human hearing is, but from my experiments it seems to be around 50db. (assuming the compression is applied to the whole spectrum at once.)

find two frequencies which have equal power on the A weighted / human ear scale, and play one at a fixed amplitude. play the 2nd at equal amplitude and adjust the amplitude downward until you can no longer hear the tone. there is some difference based upon the amplitude of the first tone, meaning the compression is nonlinear just like a typical compressor implementation used in music/audio.

i wish i had the equipment to measure that.

i notice music which frequently activates the compression at extreme ranges causes irritation, while music which is already compressed so as to minimize activation is most pleasant. if i could measure human hearing accurately across multiple individuals and come up with some good numbers, i might be able to create the best compressor of all time

[jupiter8]

Compressors isn't only used on the masterbus to make everything loud as hell even though it might seem that way today. It can be used as a sound shaping on individual instruments as well. To say that music in the 60ies used less compression isn't necessarily true.
Sure they did'nt squeeze the hell out of the master as they do today.But take the Beatles for example.They used compressors on pretty much everything.

It's all in how you use it. As any tool it make make or brake any recording.

[DevonB]

I use a compressor on an individual track, not a whole song. If a track is too soft in a mix, try both the fader and a compressor. See what makes it 'fit' better.

Devon

[eduardo_b]

the thing about compressors is that they do not tend to operate immediately. you can adjust the attack and release settings to make a compressor act like an automatic gain control, for example

It's my impression from what I've read here and elsewhere that different compressor plug-ins have different characteristics with and without using their controls, and that some are better for certain applications, instruments or frequencies than others. I would think that well-designed ones would be more or less neutral in this respect.

[kritikon]

So it would seem there is a tradeoff between having extended dynamic range or having enhanced sound. But is this application good for any kind of music -- such as both classical and rock, for example?

Yes...and you have to choose how much trade off you personally want.

Both styles can benefit from compression and both styles can be impaired by it. It all depends on how it's done.

It's really really wrong to think of applying destructive tools in the production of intentionally extremely dynamic music like classical to accomodate those who choose to listen to it in their car or in a noisy environment.

Only from your own point of view. It is not wrong per se. That is a silly sweeping statement. If the majority of classical listening audience listen at home on quality HiFi setups properly set up, then maybe the majority verdict would be no compression. If the majority listen in their car or on iPods or whatever...then they are correct, not you. However, I don't profess to know the figures for classical music audience. Classical music can benefit from compression...it entirely depends on how you listen and how you perceive it.

But one wouldn't compress, say, acoustic guitar, would one? To what purpose?

No...very often accoustic guitar is compressed. It's a very dynamic instrument - which by definition means that some portions of it will be very low in volume. You might want to level it out, you might need to compress the transients so you can hear the body better. You might have to compress to even out Eq bulges, as almost always accoustic guitar in your average music is Eqed (often quite fiercely). Possibly the solo classical accoustic guitar stuff is less compressed, but accoustic guitar is used in a huge variety of modern music too.

I read that in the 60s and 70s, compression was used with a light hand, and that's why the recordings from those years sound different than current music. Any insights or thoughts about this?

Personally, that sounds very dubious to me. It does seem that master compression/limiting (certainly the limiting part of it) is more heavy handed on average nowadays. You only have to load up an older recording into an editor, and look at the RMS figures and the waveform.
But...to me, some of the older stuff sounds very heavily compressed at the channel stage. Some of the old vocals and guitar stuff sound like they have little or no dynamic range whatsoever. I've read the odd comment from 60/70s engineers that they tended to use quite heavy channel compression. And some of those old analogue units are very easy to overdo. And that makes sense to me - recording with 4 or 8 track means you don't have to have a rack full of individual compressors . Just a selct choice will do - and it wasn't unusual to record compressed material...then if it wasn't compressed running into the tape - the tape for sure was often run extremely hot at some stage in the process to get that tape saturation that everyone all of a sudden likes again. Saturation is limiting - nothing gentle or subtle about it.

I'm trying to decide how much effort to give to compression, and what plug-ins, if any, would make the most sense.

Weeell...it's one of those things that the more effort you put in, the better results you get out of it. i.e. if you apply yourself to learning what works and what doesn't with compression...your music will probably benefit. There's also the problem of greater detail/effort, the smaller the returns. i.e. you will get to a point where you really mess around with tiny little details of compression and put heaps and heaps of effort into it, and beyond a certain point you don't hear an awful lot of difference in it. Possibly even only enough difference for you personally to hear, but nobody else. All good h/w compressors each have their own character, and recently we've been getting s/w comps that have their own too...so a different comp will have a slightly different effect. You could spend half an hour trying out different compressors at pretty well the exact same settings, and you would hear subtle differences - but the differences are so subtle that they may only be of benefit to you - whereas a kid listening to your tune on his iPod likely couldn't give a shit (even if he could actually hear the difference via MP3 anyway)...
So there's a limit to the effort you need to put in...no limit to the effort you might want to put in. Depends on how much of a perfectionist you are...

Finally, compression is a mastering, mixing or individual track technique -- or any of these?

It's all of those. For modern music, usually for every tune you will ever make.

Digital recording allows a 90dB dynamic range and most of us don't even use one tenth of it. Therefore, ultimately the music suffers.

Another irrelevant sweeping statement. many of us make music that would not benefit from all that dynamic range. Music made years before the loudness race was equally heavily compressed or saturated and didn't use a fraction of the dynamic range of tape either...and it didn't necessarily suffer for it, either. Just because digital allows a greater dynamic range means nothing to some styles of music and never has. To many of us, the benefit of digital is in the sheer number of channels available, instant recall and undoable editing. Dynamic range never had anything to do with my reasons for going digital whatsoever. Same applied to everone I ever came across in the 90s who was into electronic music.
I now don't have to use a studio full of rack gear, I don't have to have miles of cables, I have almost no limit to channel number, I don't have to maintain a tape machine, I can d/l free plugins, i can buy plugins at a fraction of the cost of h/w...the list goes on. Never once did I consider the dynamic range available to me through digital media. It never once had any relevance. And still doesn't to this day. You're spouting doctrinaire dogma. And that is why we hear such nonsense amongst the truth in arguments re limiting etc.

Music develops according to the technology available at the time. And will continue to do so. Music was far more limited in terms of technology 30 years ago, so of course the methodology was completely different. That does NOT necessarily make it better though (sometimes yes, sometimes no)

[Hovmod]

It's really really wrong to think of applying destructive tools in the production of intentionally extremely dynamic music like classical to accomodate those who choose to listen to it in their car or in a noisy environment.

Only from your own point of view. It is not wrong per se. That is a silly sweeping statement.

Did you read the word "intentionally" in that sentence? If the composer wants the quiet parts to be very very much quieter than the loud parts, then it's not up to a popular vote to take it away to accomodate ipod users.

[kritikon]

I remember somebody saying once "if old recording were so good...how come some of them sound so crap?"

And some forget that little fact. We bang on and on about sometime irrelevancies such as how much limiting to use and hark back to the wonder years...do you ever sit down and really listen to some 60s/70s and definitely some 80s recordings? Some of them sound f*cking awful. The vocals are almost distorted, the saturation and tube shenannigins reduce the sonic quality to drivel occasionally, the frequency range seems to stop well below even my poor limit, and some are just downright muffled and distorted to hell. Uusally the vocals drown out every little nuance of the music. Thank god that style has changed for the better.

Even a poor recent over-limited radio play of some current music is alot clearer than some of the crap that was out in the 60s/70s/80s. Yes, some of that era music was recorded with quality and care - and much of it was shite too. But we don't hear that said, do we?

[kritikon]

Did you read the word "intentionally" in that sentence? If the composer wants the quiet parts to be very very much quieter than the loud parts, then it's not up to a popular vote to take it away to accomodate ipod users.

Another sweeping statement. So the composer has absolute say over how the music is recorded does he? Even if he's been dead 300 years and had no concept that people would even listen to his music over a HiFi system and speakers or iPod in their living room/car/pavement/office. The absolute only way he conceived of his music being listened to was in a concert hall or chamber with a live orchestra.

So the conductor and orchestra have no input into how the music is interpreted? The composer has already been put to a popular vote by different orchestras and different conductors. The composer has been interpreted this way and that thousands of times over the years. So if a record manufacturer or producer wants to compress an orchestral piece he can bloody well do whatever he wants with it - it's public music now. The public vote with their wallets and either buy it or don't.

So...actually, yes it IS up to a public vote as to how 300-year-old music is recorded.

[digitalmessiah]

i wish there's more hype on FSU plugs than compressors. compressors are so overrated and expensive. pfftttt.

[eduardo_b]

http://www.cdmasteringservices.com/dynamicrange.htm

You only have to load up an older recording into an editor, and look at the RMS figures and the waveform.

This explains the mystery of why some of the songs I've looked at in wave editors were essentially blocks of color with none of the characteristic wave amplitudes I would have expected. Given all the replies I've received to my query, it seems that a deft hand with compression can yield improved results without sacrificing significant dynamic range or audio quality. There certainly seems to be general consensus that compression is vastly overused in general.

Anyone care to comment on standard versus multiband compressors, and suggest specific plug-ins to evaluate? Would more than one or two compressors ever be needed, and would more than one be used for a song or album?