How to compress the snare and kick drum

How to compress the snare and kick drum

Everyone knows that you should compress the snare and kick drum. But why should you do it, and how should you do it?

Audio Masterclass

An Audio Masterclass learning text.

It has become part of recording folklore that you should compress the snare and kick drum. But first, you have to know why you are doing it. If you do not know why, then you're never going to get a good result - the sound you achieve will be no more than the work of random chance.

Compressing individual drums vs. the whole drum set
There are two ways you can approach compressing the drum set. One would be to compress individual drums, the other is to compress the drum set as a whole. These will produce entirely different results. You can do both if you wish, but here we shall concentrate on compressing individual drums, principally the snare and kick, but also the toms too.

The sound of drums without compression
A while ago an experiment was carried out where a snare drum was recorded and the recording played back through a PA system. The sound of both the drum itself and the PA were fed to an audio analyzer. Apparently, to reproduce the sound of the drum accurately and maintain the transient (the initial strike) properly, it took 1000 watts of amplifier power.

The reason for this is that the transient, the very first few milliseconds, is VERY loud. The sound dies away quickly after that. So to reproduce the transient accurately, a lot of power is needed. In recording, then the level must be set so that the transient does not exceed 0 dBFS - the full scale level of the system before the red light comes on.

Why drums need compression
The problem now is that the transient is much louder than the 'body' of the sound, as the strike dies away. But the transient is short and does not fully register with the ear. So the drum is actually a lot louder than it sounds. Yes a drum played live sounds loud, but any other instrument played continuously at the level of the peak of the transient would be truly ear-splitting.

If the transient therefore can be made quieter than the body of the sound, overall the strike will sound subjectively louder. Actually, 'louder' is probably not quite the right word for the subjective experience. 'Fuller' or 'more powerful' would be better.

How to set the compressor to make the snare and kick sound fuller and more powerful
Every compressor - every decent one - has a control labeled 'attack'. This is confusing. Anyone new to compressors would think that more attack means a more attacking sound. In fact this control sets the speed at which the compressor responds to a sound. If you set a long attack time, say 100 milliseconds (a tenth of a second), then the transient of the drum would get through before the compressor had time to respond. So to lower the level of the transient, you should set a very short attack time, as low perhaps as just one millisecond.

When compressing individual drums, the attack time is the most important control. The compression ratio can be set to around 4:1 and the release time to 100 milliseconds. Naturally you should experiment with all of these settings.

Problems with a short attack time
One thing is very much for sure, you have to experiment with the attack time. Setting an attack time that is too short will result in a 'flattening' of the sound of the drum. It just doesn't sound natural any more. So you should pay a lot of attention to very small movements of the attack control because these small movements will make a lot of difference.

Differences between the snare drum and the kick drum
The main difference between the snaredrum and the kick drum is that the snare is always a very attacking sound with a sharp transient. The kick is always less attacking, but the degree of attack can vary. If a hard beater is used, then the sound will be attacking. Sometimes a piece of hard plastic is attached to the drum head to emphasize this. But if a soft beater is used, then the sound will not have such an aggressive transient. Either way, the sound can still benefit from compression. But you have to use your ears and fine-tune the settings to get the best results.

Compressing the toms
Toms can also benefit from this type of compression. However the body of the tom sound is louder compared to the transient than in the snare and kick drums. So effectively, the sound is already compressed in comparison to the snare and kick. Therefore, although this style of compression is certainly applicable, generally less compression will be used than for the snare and kick.

Summary and further considerations
What we have learned here is how to reduce the level of the transient compared to the body of the drum sound to make the overall effect fuller and subjectively louder. There are occasions, not covered here, where you might want to emphasize the transient. There is also a significant difference, not covered here, in the way you would approach compression of drums in digital and in analog recording.

Publication date: Monday August 21, 2006
Author: David Mellor



FREE 12-STAGE COURSE TOUR - Learn how Audio Masterclass can help you become a
better producer and engineer in your own home recording studio!

The Audio Masterclass 12-stage free video tour



Earlier discussion on this topic...


Best way to get great kick and snare is still running to 2 inch tape in my opinion. Minimal signal path, good mic placement and a touch of eq (just to correct any mic deficiency). I don't compress to tape only on the return. Generally I find I don't need much then either as the tape compression absorbs the transients nicely. I've never tried tape compression emulators as I have the real thing but presumably that's what they would be designed to do. Would be interested in anyone's comments on whether they've done an A/B comparrison with an emulator and a tape machine. Especially now when tape is so hard to find and costs are so high!!
Thursday January 25, 2007

Joe, Livermore, USA

... that is why many drummers nowadays "sound" just alike. Many have the same perception as a great drum sound instead of a "unique" drum sound. Every drum is unique but isn't treated that way when being recorded. If you apply this concept to all your tracks you get a more original sounding demo but still with a great sound. I suppose that's why many big time engineers have been amazed at some of my lower budget demos.
Saturday November 04, 2006

Joe, livermore, USA

So far I've seen comments on compression settings and mics but nothing about EQ. It is important to have a good mic but that's not the most important thing. I get an excellent drum sound by making sure the drums sound good to begin with. This may require tuning the drums myself and listening closely to any overtones(up close). Experimental mic placing is important and in most cases I capture a drum in 2 places. Transient is important but the level isn't the same on all parts of the drum. Much of that sound bounces off the room. So acoustics is important. Isolation solves those problems. Next is eq'ing. I learned that a sound, whether it's a drum or a vocal, gives pleasant tones and ugly tones. There's a frequency chart that tells you the ugly frequencies and you can measure this for a particular signal to see if you need to bring down the ugliness and emaphasize the character of the signal. Many digital studios today don't do this and ...
Saturday November 04, 2006

ogbaji john, makurdi, nigeria

the info i have gotten sofar has helped my production greatly.i truely appreciate.i wish i can be sent atticles to my home here in could also setup a school of sound engineering here in nigeria.i can put you through the necessery procedures.nigerian and africans would never mind schooling.i hope my reguest be granted.thanks.john+2348032852112,+2348053645566
Saturday October 07, 2006

Steve Ward, White Plains, New York, USA

In regards to: <>

The whole point of compressing the individual drums in a kit is to bring their transients down so they is closer in level to their drum's tone, not to make the transient lower in level.

Doing this will, in fact, make the overall drum sound "louder" subjectively, although the peak level will have dropped. (Louder is correct word.) With less peak level, it's common to raise the compressor's output gain, making the sustaining tone higher in level than where it started.

This technique will also slow the rate of decay, giving the drum's tone more sustain.

Rock on!
Thursday September 14, 2006

adewole solomon, lagos, nigeria

I realy appreciate all the information you ve been giveen,but,if you have a school of sound engineering i would like to enrol for 2007 session,pls send me the necessary information,either to my mail address or i would appreciate it if you can send it to my home address.thanks.
Thursday August 31, 2006

dave connor, whitehaven, england

thanks once again to Bart Nettle of sydney for recommending this book. i've looked on the site it looks really good. i'll definately be getting this book as soon as i get the cash. it's an absolute chore trying to find worthy books that don't just tell you the same old things. thanks again dave
Tuesday August 29, 2006

Bart Nettle, Sydney

Your welcome Dave but the thanks are to Stav's book

I was guessing with compressors until I got this book. "Secrets of the sound balance engineer revealed" is an apt sub caption! I thoroughly recommend anyone get this book

His book is a writeup of successfull actions more geared to engineers than a beginner. Very Valuable! Check out the web page.

You'll find a link to the "Smart Console" he invented. He also writes regular to "Audio Technology" Mag.
Monday August 28, 2006

Dave Connor, whitehaven, England

a big thanks goes to Bart Nettle, Sydney for this tutorial on accurately setting your compressor. i will be trying this as soon as i get a minute.

cheers Dave Connor
Monday August 28, 2006

Alex, Australia

I have an m-audio console which the mic pre's are clearly lacking in headroom. How would I go about creating a "big" sound where the signal can be maximised without clipping?

Would my Waves ultramaximiser or C1 be the answer when recording to hard disk?


Monday August 28, 2006

Bart Nettle, Sydney

PART TWO: Now lower the thres. This is how much the comp is being accessed by your signal or at what level your signal will be compressed.(If you flatten your sounds you will have to recreate a groove to make up for the groove you lost in the flattening. This groove can be over several bars and determined by the thres)

Lastly lower the ratio, here is how much compression is applied by the compressor when the threshold level is reached. As we have taylored the attack, release and theshold under our microscope to our sound hearing, what we wanted from each do not fiddle with these now as you won't hear them. This is how you unlock a compressor. Each tumbler dialed in the above sequence.

Now you can set a compressor acurately and not be concerned with if the release is right as you have determined it is right when you could hear it.

Saturday August 26, 2006

Bart Nettle, Sydney

PART ONE: Knowing with certainty what you are doing with a compressor every time you use one gives one predictability. How to unlock a compressor from the best recording and mixing book i've read and i've read a few:

(Not verbatum) Temporarily make these settings: Ratio to max; thres enough to make the comp work, you'll have to increase the ouput gain to hear it; release approx 50% thereabouts, not important at this stage as what you are going to listen to is the attack only. Adjusting the attack you can, with these settings hear what you are doing, like under a microscope. Adjust the thickness of the attack. Snare for eg. How thin you want it. Guitar the pick on the strings or the strum. THICKNESS how much? When you got the attack sound you want leave it: Next adjust the release, this is how long you want the comp to work for. How often the comp comes up for air so to speak. You can create a sub rythem to what you are compressing, another groove.
Saturday August 26, 2006

Nikola, Senta, Serbia

Another thing I like to use on cins are small diagraph condensers.But there is no concrete rule.Someone like to use AT4033 or Shure KSM series.On bass drum I alway put first akgd112 playing around with the mic or changing s few preamp till I don't get the sound I wan't.The main problem is that our ear can never reach there where we put the mic.We can never really hear the exact sound with our ear that the mic will record.

The bassDr is alway problematic near the 500hz,2000hz and the 200hz.Play around with the sidechComp and not with the EQ.These freq. change drasticly the sound and always give headaches to engineers.The ROOOmm shoud be a bit more reverberant than you need.Why?Becuse it caryes the energie of the whole drum trough the air that you need to catch.This is for now.Next time i'll attach some files you can hear and If you write back to me I'll explain thw way I recorded the drumz.
Wednesday August 23, 2006

Nikola, Senta, Serbia

I was reading alot of article about how this and that, and I can tell that lot of them are useless.If you want you can take a try and folow what I usually do...

First thing.Drumz:lets say minimum req.pearl export,sonor force3001...New skins and good cinbals.A great drummer is 50% of the sound, he shoul be killing the drumz if you want a rock sound-smashing.Drum tuning every segment.The SnDr mic should be a 1.2 inch away fron side of the ring and pointing to the spot where the drummer is hitting the drum for avoiding the proximiti fx and avoid reflections from the surface of the skin.The overhead nis are the MAIN thing when you record.The give you the mass the sound of the room and the drum body.I use a lot of eq but small cuts.The drastic changes I tend to make with the sidechComp on the snare.I usualy find the freq.that bugz me and just cut it out with the SCHComp.Actually It take me a whole day to find the drum sound that I was looking for.
Wednesday August 23, 2006

Nikola, Senta, Serbia

For everyone that are trying to reach a good drum sound... I have a computer and some ok mics and i still manage to get beter drumsounds than other guys that have Neumans, EV, Sennheisers and good pres.The thing is that you would have to consider the fact that the drum sound you hear in a room when its played is a compressed and limited soundmess.Your ear acxtually compresses it.Just try this.Put a guy beside a snare drum and ask your drummer to hit a big one.You will notice that the eyes of this man standing beside the drums have blinked.Why?Because the soundpresure is way over the normal so our brain forces our muscles in the ear section to resist.Actually, our ear compresses and limit the sound we hear,but the mics don't.

So, a good drummer,good drumkit,2 good ovehead mics-they give you the main sound.Always check the phase on the snaredr.I use sidechain comp on SnDr alot.I cut of the agressive mids that just make things sound unnaturale and cold...
Wednesday August 23, 2006


As with any type of effect I tend to continually ask myself "does it sound any better or worse?", and after lots of switching in and out I generally find what i'm looking for.
Tuesday August 22, 2006

dave connor, whitehaven, england

thanks for this comment to Ms. Remy Ann David. but i've read this sort of tip a number of times and thick as i may be, it doesn't seem to gain me any progress. i think because i can't make a disition on an exact setting for a given instrument. this is something i've been studying for quite a while and with the release i'm always wondering would it be better with a different setting than i've chose. don't get me wrong my mixes are almost there but after a few listens i'm a little disapointed. though friends say they sound good. i think the problem lies in compressor settings particularly in the lower bass kick drum area. (i know i'm just a being a perfectionist)

dave connor
Tuesday August 22, 2006

Ms. Remy Ann David, Virginia suburbs of Washington DC, USA

The release time control will provide the a much more audible difference when adjusted from slow to fast. When you have a slow release time, the compression/limiting will sound more natural. When you have a fast release time it will make the signal seem "apparently louder" more aggressive and more upfront sounding. This is how most radio stations get themselves louder than loud. But there is a trade-off. Too fast and things may not sound too good. So it is a careful balance between natural, louder and aggressive. This is again one of the many ways we have to color and manipulate our sound to taste. Mmmmmmmm Gooooood

Always cooking some good DB's

Ms. Remy Ann David
Tuesday August 22, 2006

dave connor, whitehaven, england

regarding using compression on drums, or other instruments for that matter. i can totally understand the attack settings but when it comes to release i find it difficult hearing any difference. can anyone offer any advice on this matter?

cheers dave
Tuesday August 22, 2006

Bart Nettle, Sydney

If you compress the drums when you reord them.

Compress them again when you mix em down.

Compress again at the mastering stage.

How much have you compressed your drums on your final CD?
Tuesday August 22, 2006

Alejandro Varela, Buenos Aires, Argentina

In my own experience, compressing drums is, maybe, the 50% of the final mix. I use a very fast compressor hardware for the work,(not software ones), like DBX 160.I have two units (one for kick and the other for SD) and today are not too expensive. I like it 'cause they are clean and faster. The ratio must set always between 3:1 to 8:1 (depends on style of music and the touch of drummer). The level of compression must be set around -6db to -10db. Then I route the whole drum mix to a Waves L2 ultramaximizer and set a level compression around -5db. Try it out. Opinions are Wellcome. Regards.
Monday August 21, 2006

Axle M., Brisbane, Australia

i know this is nothing new, but in the day of pro tools systems, and digital recording one can be forgiven for forgetting that one way to deal with transients in a nice rich musical sounding way can be summed up in two words: tape compression. although it is seen as some form of luxury to record to tape nowadays, by slightly slamming the kick and snare tracks when recording to tape it deals with the transients, whilst bringing out the body of the drums decay. I know this is nothing new, but it makes me wonder if a tape compression simulator (neve portico?) can be used for the same effect during the tracking stage? i dunno I've never tried one, but i would be interested to know if there are people out there using it to the same effect as real tape compression.
Monday August 21, 2006

Remy Ann David, Virginia suburbs of Washington DC, USA

Drums sound much better when you have a high headroom microphone preamplifier, like an API or Neve. Of course that's a luxury. Otherwise kids, don't try this at home and don't push the microphone preamplifier on the average console. I believe in running the microphone preamplifier on inexpensive desks at a lower gain setting for better headroom. S/N will not be harmed.

One of my favorite combinations on snare and bass drum is the Sennheiser MD421 with either of the above preamplifier's and a UREI 1176, followed by the old Allison KEPEX downward expander, which also removed a bit of the transient. It improved the "meat" of the drum sound, while tightening them up. The question is, should you add any equalization before or after? The answer is yes.

In the old days when I worked at the now defunct Media Sound in NYC, they had a saying I still live by. "An ounce of Punch is worth a POUND OF SOUND." IT'S TRUE!

Still punchy after all of these years.

Ms. Remy Ann David
Monday August 21, 2006

Bart Nettle, Sydney

...we perpetuated a ever so slight flattened, limited sound which made way for non drummers to substitute a synthetic or sampled sound in its place. It was not noticed! We conditioned the radio listeners into this flattened sound.

Go back to your (un authored) experiement. get your newer higher spl handling mic just right, not where you's put it to flatten it but far enough. Next, use a very good drummer, use your ears on the AB not some peice of test equipment to tell you.

Actually, go down to your local Jazz Club and have a listen to a small un mic'd band! Did you sway?

Go to a Rock Venue fully mic'd up with a 3 way out front. Make sure the sound guy is trying to sound like the record and flatteneing the drums.

"Oh yeah we sound better at rehearsal"!

It can be recorded, have a listen.

Even the earliest recording legends despite the ineffieciencies of the gear in those days, man you can feel a groove there.

Well, thats gone today! Maybe clients want it back!
Monday August 21, 2006

Bart Nettle, Sydney

Another "everybody knows" generality!

Lets say it is common practice initiated and perpetuated by sound engineers. The Why is; with our frail recording equipment of yesteryear a limit was put on in the mixdown stage to keep a loud drum sound blowing out our VUs.

The simple solution would of been to lower the program material to around 7 db and bring the drums in to hit 0db. The transient of course was greater then the 0db read on the VU but the tape magetic particles were able to absorb these transients thereby preserving the feel of the piece.

A good drummer was able to vary the attack on any drum many within the bar, some, like accents over a broader meter of several bars. the point being there is an infinite different pressure hit, damp, sustain varieties and this in regular

succession is called rythem. It makes you want to respond in sway or tap and this is the feel the groove. It makes you want to dance.

Now we in all our glory killed it because it was our gear at the time.
Monday August 21, 2006