Compression is meant to reduce the volume of the loud bits of a track.
The idea is to stop it being overpowering compared to another track.

That way you can still hear the guitar under the vocalist as they go "all-in".

Compression is meant to take dynamics and get rid of them while adding annoying hissing sounds. 8) The benefit of compression is that people will think your music is really cool when it's a wee bit louder than the next guy's CD. OK, so seriously, compression does help "even out" the sound, and quite frankly, using anything less than moderate compression/limiting on most mixes leaves a mix rather quiet, and most people seem to not enjoy listening to quiet music anymore. 8)

There are some things like bass guitars and kick drums that just love compression, and pretty much need SOME compression/limiting no matter how good the player is. Instruments like guitars, on the other hand, generally don't need compression and tend to sound better without it. Of course, there are exceptions to this, like if the guitarist was playing some sort of slap technique, but I think you get the idea. It seems, to my untrained ears, that most entire mix compression/limiting is applied during mastering, and most good audio software programs have special plugins for this (for example, there's a "loudness maximizer" plugin in mastering section of Adobe Audition). If you don't want to get involved in the loudness wars, I'd say don't go crazy with compression, but maybe the guys you are recording just really stink, and in that case you may need to go crazy with compression, haha. I hope this makes some sense. This is just one ameteur's experience with compression -- I went from using way too much compression to using as little compression and limiting that I could without noticably affecting any dynamics. God bless.

Ahh making important decisions about dynamics. Electronic reproduction of music is an unnatural act. Dynamics like opinions need to be wiped occasionally. And for that we use compression. Then when it gets too loud? We use limiting or the Internal Revenue Service, whichever comes first?

In my orchestral & operatic recordings, I really don't want any dynamics processing.

In my rock-and-roll, jazz, jingles, gospel, you name it, I want to crush the crap out of most of it. I want a 3 DB dynamic range. I Don't want a 100 DB dynamic range. You can't listen to that. Compression Limiting, Downward Expansion, Gating, until you run out of equipment. This makes my mixes translate very well when played at low levels on a 3 inch speaker to full tilt boogie on a PA system. You hear everything & every instrument. That's because you're providing much more consistency with levels. Nothing gets lost & nothing jumps out too far.

Jumping off too far.......
Ms. Remy Ann David

I like compression, but what is wrong with a quiet CD?

My rule of thumb is if I can hear everything in the mix when and where I want to, then I am happy. Sometime I have to compress things to make that happen, sometime I don't.

I never go into a mix assuming I will have to compress anything. But at the same time I will crush things if it sounds good.

If you want it louder can you just turn up the mixer levels for the tracks how you see fit rather than use a compressor plugin ?

If you want it louder can you just turn up the mixer levels for the tracks how you see fit rather than use a compressor plugin ?

Well, you might be able to turn up the mixer levels a bit, but the audio will distort and then clip hard if you get into the red and beyond. So, to make the audio even louder (overall), compression/limiting is used on individual instruments and/or the entire mix. I've heard about studies being done where people were played songs that were exactly identical except that one mix was a bit louder than the other, and people dominantly chose the louder mix as being the better mix. I sort of wonder if the songs being played were hip hop, but there is some truth to this, especially when it comes to music that has a history of loudness (rock, rap, etc...).

There is one downside to using too much compression. The first is the danger of pumping, but extreme levels of compression/limiting can be used without causing pumping if the user knows what he or she is doing. The real downside to using too much compression/limiting is that listening fatigue can easily set in. Listening fatigue is when, say, a CD's worth of songs have very little dynamic range. The human ear tends to get irrated by hearing everything at the exact same volume level for more than short periods of time. And lastly, SOME dynamics do sound good. Those classical composers didn't put all those dynamics into their compositions for nothing. :roll:

When it comes down to it, you can always squash a track and then manually lower the volume A LOT on sections you want to be a bit quiter. This works and I admit to being guilty. I used to want to compete in the loudness wars, but the desire to have decent sound quality prevailed, and I can still get my material loud enough by modern standards while not compromising a more extended dynamic range. But hey, whatever floats your boat. Good luck and God bless.

By the way, I play mostly speed punk, so I'm not some classical purist raving about the evils of compression/limiting.

One thing you have to understand about dynamics processing i.e. compression is it not only has the ability to affect the overall dynamic range. It also has the ability, through timing manipulation, to change what is referred to as "apparent loudness". Apparent loudness is what the compressor does uniquely. Faster release i.e. recovery times will make signals appear louder without actually changing much of the level. Making it slower will make it appear, softer, more natural, less aggressive. So we don't just use compressors & limiters to be automatic volume controls. They are so much more than that when used as a unique instrument in its own right. So when you realize that? You'll know how and which kind of dynamic processor you're unique application will require.

Dynamics processors of all sorts are generally use for some kind of dynamic manipulation. But they are also used in a sense, like an equalizer. But you really have to know how to obtain it and from which device. That takes experience my friend and plenty of it. In a sense, it changes the sonic Aperture. That is, it can improve the auditory "depth of field". You read articles where people have indicated they were only hitting the limiter about 1 DB? Well, it's not so much the limiting as it is the effect the variable attack & variable release times can impart to the sound. So yeah, an equalizer. One that works on texture instead of frequencies.

I like all flavors but can't afford all flavors
Ms. Remy Ann David

Faster release i.e. recovery times will make signals appear louder without actually changing much of the level. Making it slower will make it appear, softer, more natural, less aggressive.

I don't know how well this practice could be applied to individual instruments, like a bass or kick drum, considering that certain instruments seem to work best with certain release times. I'm not doubting it, I'm just saying that there seems to be a relatively small window for release times for some instruments. Now, for the entire mix, well I have to say that I don't own any analog equipment that can do that, and all the software I've used doesn't give you any access to fine adjustments like attack and release for entire mixes. So, I've never had to chance to play around with that. Sounds interesting, though. :D