Ok compression and Drums

[P. Rene]

Im trying to piece this together, when you guys use compression on your drums are you doing this because your drum sounds are live recorded or your using sampled drums. The reason i ask this is because many MPC heads have that velocity button which keeps all sounds at a constant volume, so unless were using live drums or live loops what do you get out of compressing pre-recorded drums.

[blayz2002]

Originally Posted by Azriel_7

Grab a couple of good compressors (or computer emulators) and start tweaking the drums through them.
You will hear why and come to understand how compressors shape sounds and envelopes of sounds.

Yep once you get to understand why your compressing drums (or anything else for that matter) then you'll understand why you might throw something through a compressor to add flavour or you want it to hit a ceratin way.

Sounds to me like do don't when you should compress something (which should basically be when it's necessary to make it fit the mix) and when you know your compressors and what they will add or take away.....Listen to some Timberland mixes, some Dre mixes..their drums would be nothing without compression, try to make your own version of one of their drum tracks and you'll see why we still compress soemthing that's alreay been compressed, i.e. synthetic drums or drum samples....it's all in the attack and release.

[Gregg Sartiano]

Compression is a process -- not just a single technique.

A long attack time actually allows the front of the envelope to "speak" before the compressor acts. Result: slow compression doesn't "squash" -- if anything, it'll make drum sounds more punchy.

This doesn't mean "some is good, more is better" -- crushing a drum sound with slow compression will just make it clicky.

Do like blayz says and experiment with a software or hardware compressor. Try looping just a snare or kick-snare beat (simplify!) and tweaking the settings.

Try compressing with a low threshold (i.e.: MORE compression) so the gain reduction is hitting 10dB or more (set the ratio 3:1 to 20:1 or more). Then adjust the attack. What does 10 ms attack time do? 30-70 ms? 100 ms? Maxing out the speed will squash it -- we all know that -- but the subtleties are in those slower settings. Push it to extremes so you can hear how the compressor acts, then take the threshold back up so it compresses less but with the same speed & response.

Appropriate use of gating and high-pass filtering can "tighten up" drum tracks too.
A gate doesn't have to max out to be effective -- even a 5 dB reduction of the mush & noise in between hits can leave space that'll help your mix sound punchy & wide. I think it's important to think SUBTRACTIVELY when mixing -- what frequencies & dynamics can I take OUT to create space, as opposed to thinking in an additive manner -- what frequencies does this mix NEED. That kind of philosophy will tend to get you a crowded mix. This goes for ambience (reverbs, etc.) as well.

I never use compression to level out kicks or snares -- I'd rather automate or gain them. Actually, my #1 way on a loop/MPC track (tracked to DAW) would be to cut/paste the offending hits onto another track to put it on another fader.

[P. Rene]

Thanks greg for the deep insight. Now so I bus the drums consisting of kick snare hihat. Now via software running comp to taste do I bounce them drums to another track or just keep the process running in realtime.

2nd thing I sample my ass off on drums, and tweak or eq to taste. Now thing is lets say I have a bass running and a few melodies sampled or just live keys and the drums are to my liking, do I compress instruments to taste also.

3rd. Say drums n instruments are comp, now the vocals are flat out where do I go from their to get them to sit well.

Reason I ask all this is because, I never actually ran any comp on my drums as far as bussing. I mixed each track I felt like mixing invidualy to taste even sometimes while tracking in to whatever medium I'm using per se.

And I've bussed vocals for little limiting or hooks when i have so many stacked to keep it from peaking. Now I'm enjoying the freeze function in software and have taken advantage of pc power.

But thanks for all replies.

[Samplecraze]

You have just described the problem that is created by so many people because of the fallacy that you must compress everything.

Once this is done, the next step is to compress the stereo mix.

Do you not see a problem with this thinking?

Many of today's ...ahem...'producers' feel that they must compress every channel so as to get the loudness they want.

What many do not realise is that a good dynamic mix with emphasis on peaks and troughs, gained only through ample headroom, affords a more fluent mix and perceived gain.

Use the compressor for 2 reasons.

1. To shape the dynamic content of the overall stereo buss mix.

2. To shape a specific sound in context with the rest of the mix.

Caning it on every channel and then again on the final stereo mix is overkill and lands you in the square wave territory.

Sampling a sample that is already compressed, and then compressing it and agin on the output stage is simply bad engineering.

Ask any ME how he would like your material presented for mastering.

[P. Rene]

I see this compressing craze all over, I dont compress every channel cause like I say, im very critical during the pre production process on which sounds i use and how soft or loud i want it before i hit tape/daw for vocal tracks. Ill post something ive been working on pretty soon to give a glimpse of what im working with.

Hey sample craze this comment -"What many do not realise is that a good dynamic mix with emphasis on peaks and troughs, gained only through ample headroom, affords a more fluent mix and perceived gain"<<<<<< is the reason why i ask this question.

Ive heard many tracks that had nice round body before A) compression on 2 buss or B) the mastering which i heard took a lot of the thickness and life out the tracks. So maybe this is the loudness war alot of mastering/recording engineers debate about?

But thanks for the info, and ill focus on the making music part, and leave the technicalities to the technical folks.

[Samplecraze]

Glad to be of help.

If you were to look at the waveform of a stereo master mastered by a good ME, then you would see fluent dynamics.

If you were to do the same with a poor ME, then you'd see a square wave L/R waveform with narrow banded dynamics.

The ear/brain combo can only distinguish gain by having a reference. A mix that is dynamic exhibits these variances, so the ear/brain perceives the gains far more distinctly than a narrow banded mix.

This is what differentiates a mix that 'tires' the listener to a mix that always ikeeps the listener interested.

BTW, we can thank the broadcasting industry for the onslaught of square wave masters.

[P. Rene]

cool god, i appreciate it.