Jeff Evans

A few other compressors and limiters did not do well, especially the limiters. Very jerky response. Won't use them again after hearing them in this situation. 

Jeff, this is exactly what I would predict, and illustrates the problem with this approach.

Want "jerky" response? Here's the recipe:

1. High compression ratio
2. Hard knee
3. Peak (vs. RMS) detection
4. Fast attack
5. Fairly high threshold

Sounds like a typical peak limiter to me. It's "worse" only because it's doing something you've artificially defined to be "bad". Even though it's doing exactly what it's designed to do. Even though it might sound fine when you listen to the actual audio.


If you want to use a difference signal to find which areas to listen to more closely in the actual audio, that might be a reasonable approach. But it's not a "real" signal you're listening to here, it's a difference signal, that's all. 

Making any subjective judgments on an artificially constructed, out of context difference signal doesn't make any sense when you can just listen to the actual audio. 

I live by one rule when it comes to compression -
Compressors are meant to be seen and not heard. 

I live by one rule when it comes to compression - Compressors are meant to be seen and not heard.

I feel the same about reverb and EQ. But that's considered "old skool" these days. Nowadays the compressor is a rhythm instrument.

   Sorry but this is an idiot newbie question???

If I understand the use of compression right, is it mean to be used to shorten the width of the sound in the stereo spectrum or sound of a song? I dunno, For some reason I can not hear any difference using compression. To me it's more or less been a glorified volume increasing utility.

I guess your talking about mixing down a song and not recording it, I know much to less to nothing after 3 years or so trying to record.   

Everything I have been told, even when I paid 1800 dollars at the time 3 years or so ago to purchase a boss BR1800 was that - : Compression is the difference between HOME recordings and Radio Ready recordings.

Anyway, sorry to disrupt your thread if I did, I just thought I would ask????

Compression is a mystery to me, and I would like to understand what it really is without having to become a rocket scientist to do so.

Compression reduces the dynamic range of a sound (by attenuating the highs).  This gives you a more consistant sound.

This can be particuarly important for real sounds (like vocals ) so you can control how they sit in the mix.

Many compressors also tend to "glue" multiple sounds together so they all sound like they are part of the same song.   I'm not sure exactly why they have this effect (maybe someone that knows more about audio theory can answer).

So for example with drums, you might put gentle compression over the whole drum mix to glue them together.  Same with multiple vocal taxes etc.   And of course most people do some gentle compression over the whole mix.   

Hi Serious Noize You are certrtainly not disrupting the thread and welcome. What you are asking is more akin to what the basics are in terms of what a compressor does and how to set it up to do it. There have been numerous threads on this including one from me which gives some basic advice on how to setup parameters. Here:

http://forum.cakewalk.com/tm.aspx?m=2116921

I think firstly you need to decide if a compressor is really necessary. I like the rubber band analogy too. Things can still jump around a bit in level, but if their movement was limited by a rubber band then they would still jump around level wise but not so much. ie restricted movement. A compressor restricts the range of volume changes. The ratio is the tension of the rubber band and how it restricts movement. They can be setup in such a way that quieter material is boosted slightly too. This can be for the good in terms of mixing and keeping things present in the mix and not jumping out too much either.

But what I am going on about here is a more esoteric approach and that is we are using a technique that removes the music and only gives us sound when the compressor is actually acting upon the signal or engaging. I am putting forward the concept of listening to the difference signal and gaining information about the plugin from that.

And I still think its a good idea as well. And yes for those who mentioned it, I am not hung up on listening to the difference signal forever either. But I have noticed that if you adjust your compressor normally and you might think something is up, then go into the null mode and you might hear why. After some fine tuning of parameters in that mode, when you go back to normal it seems to sound better and more musical to me and more like what Rbh says. Seen but not heard.

Also both Stav and Bob Katz say that some digital compressors are excellent and can match their analog counterparts anyday. They use them regularly in mastering. They are not advocating that the only great compression is analog although they do seem to do it easier as such. But only the better plugins are up to the task. These better ones all seem to have something in common and that is their natural bounce and elastic movement when in the null mode.

I tested some more limiters. The poorer ones seemed to jerk around a lot and also there were some weird clicky artifacts. I tried the Waves Loudness maximizer and got a much better result. Had to delay the Souce A wave by 1.76 ms to get a near perfect null.

Also in the null mode I was really able to hear what the Release slider does. I found on the setting it was, it was clicking rather badly and releasing rather nasty on certain loud things. After slowing it down a little I was able to get a much smoother release thing going with no clicking artifacts. Back to normal mode and it sounded great. No wonder I use that plugin a lot.

The null mode is also great for showing up what things might be a little too loud in your mix. If the kick drum grabs every single elastic swoosh towards you then may be it is a bit too loud! Back on the small speaker at low volume to check.

I'm not a hard core guy either way, but I do find it quicker for whatever reason to find a nice sound with an outboard comp. Some of it is that they often have quite distinct personalities and if you know what those are you use them for the things that that would be good for. That probably starts you off a certain way there. Some of it probably is that you can easily grab knobs and adjust physically. And they do seem more forgiving sometimes.

BTW, the thing about compression working best with it doing X, Y, or Z is another one of those things that immediately runs up against the 'whatever sounds good' rule. It works best when it sounds best in the context. Sometimes, flatting something out heavily is exactly what you want. It's often done with drum room mics for instance. And sometimes you want to make it not jump out at you, and give more emphasis to the after ring or air or whatever that follows the initial attack.

And, sometimes you very much want it heard and want to abuse it. This is not at all a new phenomenon, and it goes back to the 70s at least. Even a band like CSN&Y would sometimes use hugely compressed acoustics. Lady Madonna from The Beatles has a hugely compressed piano.

I think that what really is the goal is to achieve a variety of textures that layer over each other in a nice way to create depth and space. That means that some things may get smashed and others punched up and others with tempo based emphasis, some bright, some dark, some mid-rangy, some scooped out, some dry and some ambient, depending on how you are layering things.

Some nice points in there Dean Just out of interest in the Stav article he also mentions this:

I must emphasize that when I talk about compression artifacts I am not talking about the depth of your compression, that's a matter of taste, and some production styles demand creative aggression and even punishment. This shootout reveals 'unwanted' noise. Unwanted by even the most inept producer.

The null test is also good for aggressive amounts of compression. It just takes a little adjustment to get the null point set and you tend to hear it more often with more aggressive settings. It is still the way the sound rushes toward you and goes away that is of interest. We are after that elastic bounce. With lighter amounts of compression we are in silence a lot more with just the odd music rushing towards and away. It is always interesting to note what sounds are making the ocassional bursts happen too.

Also I definately agree with Stav about artifacts. I heard some plugins making just wrong clicking sounds. Nothing to do with the music. (the music was creating it of course but the clicking sound often was not related to the tempo of the music and was random making it even easier to hear) But it does get added to your mix. In the null mode it is often possible to fine tune your plugins so the artifact goes away. Without changing the overall compression much.

This is one thread that in my opinion more people should perhaps look into and try. This is not my idea but rather Stav's but again I used this technique recently to determine how a certain compressopr plugin was behaving.

It is much harder to hear how a compressor is working and what it is doing to your individual track or mix when the music is present! By removing it, you only hear then how a compressor is behaving and very clearly too and only when it is engaging. A jerky peformance can be readjusted to become a smooth movement toward and away from you thus when you put the music back in it all seems to sound much better.

Many people have asked how to set up a compressor properly and this is certainly one way to do it. It is not too difficult to set up the track and routing etc to create the null.

Jeff, I plan on giving this a shot with some of my go-to compressors.

I would be very interested, in addition to the above, which ones were your favorites that you tried?

I am in the culling mode in my VST land, and your valuable insight just might make my job a bit easier.