Understanding compression: peaks and make-up gain

P

PFC9

New member
Hi, everybody.
My apologies if it sounds dumb, but I don't get something about gain staging properly using compressors.

Let's say I've a snare drum peaking around -6dBFS.

According to tutorials, my goal is to keep most of the attack "smack" using a slow attack (fast release) while getting some added sustain and body through compression.
If "slow" attack allows most of transients come out before compression kicks in, my peaks would keep staying around -6dBFS, while compression would be applied to the "body" of the sound.

Now, I should use make-up gain to restore the missing volume, but this way my peaks should be increased (???).

I'm not mentioning parallel compression to get best of both worlds (attack AND sustain), but what's wrong with my gain staging?
Slow attack = peaks go through, body get compression = no sustain gained.

Fast attack = peaks are reduced, got closer to body = more RMS.

Please help! ;)
 
Hi and welcome to HR.
Your understanding sounds correct but your approach sounds backwards.
The initial crack of the snare is usually the bit that's easy to hear already and it's the ring of the snare that gets lost,
so people use compression to tame that crack but release in time to leave the sustain alone.

Makeup gain can be useful in that case because you're taking the loudest part, the initial crack, and suppressing it.
The sustain/ring is now louder relative to the crack, but the whole thing overall is quieter.
If you used makeup gain to make the crack sound as loud as it was originally, the ring would then sound much louder than it did without compression.
 
Thank you! My missing point is: why should I use a "slow" attack if my purpose is to reduce crest factor between body and transient? And... What is a "slow" attack? My snare peaks around -6 dBFS and the body is around -22dBFS. To get a "sustained" and "fat" snare I would go for taming that huge peak with a super fast attack (near to zero) and THEN with make-up gain I would increase the overall volume. So, again... Why everybody is about "slow attack to keep transients"? How could I keep them AND increase sustain if compressor kicks in AFTER the transient (and so grabbing down the body of signal)?
 
That's what I mean - It sounds like your understanding of how it all works is fine, but the approach is the wrong way around.
You wouldn't use a forgiving attack to allow the peak through unaltered if your goal was to increase the sustain/ring.
You would use the settings you're describing if you wan't to tame the ring and make the snare more attacking + percussive sounding.


If you're wanting to increase the body/ring of the snare you'd use a fast-as-possible attack, as you say, then tweak the release to taste, such that your initial transient is compressed then the release kicks in and allows the body/ring through unaltered.
That will "keep" the transients - It's just going to tame them relative to the sustained ring, allowing you to then use makeup gain to bring the whole thing back up.
 
Depending on the compressor you're using and the amount of gain reduction you're getting, a really fast attack time can kill the initial transient, which sort of defines what a snare sounds like. Something like 12 to 18 dB gain reduction (possibly too much depending on what you're after) should work. Try it. Hear what it sounds like. The attack time question should answer itself. A slow attack somewhere around 4 to 9 ms. should allow the transient through. (or a wet/dry blend if you like parallel compression) Don't worry about the number, just play with the controls until it does what you want.

I'm not sure what this has to do with gain staging. The only times I really think about gain staging are setting the levels to peak between -18 and -12 dBfs when recording the tracks, and making sure the master buss isn't clipping before rendering.
 
You must log in or register to reply here.
Back
Top