Tips on erratic snare compression?

[SkullChris]

Right now, I am mixing a band that I just finished recording. I'm 75% done with the mixing, but I'm really not satisfied with the snare sound on 80% of what I mixed. On two of the songs (the ones I like) the drummer is playing pretty evenly and not demolishing his cymbals. But on all the ones I don't like, his snare hits are extremely uneven. This is causing me to need to compress the snare track WAY more than I want to, and I'm also losing his snare to the cymbals in the overheads. So when the tiny hits come along it just sounds like a big pop, and the rolls sound even worse. I can't seem to hit a happy dynamic range, and I seem to be losing the brightness of the snare when I have to compress it that much. Any tips? -(hris

[???????]

.

[8th_note]

There's a great article in this month's TapeOp about parallel compression. You might try that technique to see if it gets you anywhere. Other than that I'm not much help because Brad pretty well covered all the bases.

[JohnDavisNYC]

if the problem is occasional really loud hits, sometimes you can get away with a brickwall limiter (L1, etc.) on the snare that is doing nothing 98% of the track, until that one too loud hit comes along, and the limiter will 'keep it in line' with the rest.

but yeah, not much you can do about a poor drum performance.... sample replacing might be appropriate, depending on the style of music, at least to get a little more consistency out of the attack.

john

[drumsound]

Try a prarllel comp on the snare or maybe on the drum groupl. You might try to EQ after the compressor so you can put back some of the tone the comp is taking out.

But I do totally agree with Brad. IF that's how the guy played you can only get it so good. It's a recording of a band not a recording of the engineer...

[RodC]

If they are so many different changes that riddin faders or automation wont fix it...

Among the other suggestions above, I have also had some success by doubling the snare track,

For track 1 make it sound good for the loud parts (Lower stuff will prob be reall low in this track)

For track 2 make the low parts sound good, this will squash the heck out of the louder parts and you may have to work with the release times to get the loud stuff to not be so "thick"...

Then blend the 2.

OR:

Expansion can also be your friend

[joel hamilton]

But I do totally agree with Brad. IF that's how the guy played you can only get it so good. It's a recording of a band not a recording of the engineer...

Tony, I have to disagree with this. In my world, the only option is to make it sound great. I cant just throw up my hands and say, "well, that is how they played it, so F it!"

I have to figure out a way to make it work no matter what the tracks are that are presented to me. Not being there during tracking for a LOT of what I mix has led me down a path of having a working method for many, um, "adverse situations" like the one in the first post.

I have developed a bunch of compression/phase/grouping/expansion/gating tricks that get me to where I want the song to be. I may also layer in a trigger, if the drummer is not playing rolls every two seconds... I may also use a kick trigger, layered in with the original, always. I may use a very slow stereo comp on the overheads with a HP to the detector, so when the cymbals are played hard they sort of "go inside" a couple of Db, but with the attack set really late to try and "mine" the leading edge of the snare up out of all that wash. I may even use an ITB look ahead expander gate on the snare (i do this often) when I need to get the snare cracking without a ton of cymbals/hats. Depending on how much you are compressing the snare, pull back the FLOOR of the gate so it is only pulling down enough to make the cymbals not chirp when the compressor lets go... like an inverse proportion to the GR of the comp.
Also, getting the right amount of compression on the overheads and snare and kick, keep in mind that your overall kit image is crucial... if it sounds like a guy really going for it and giving an energetic take, that just happens to involve the cymbals a bunch... try getting the image to feel that way... like it may still be a bit cymbal heavy, so what? turn up the guitars a bit more and add some high end to other elements in the mix so the image feels right.. think of the WHO, where the cymbals are just bashing away the whole time. It "aint" tape that is keeping those cymbals from hurting your face, it is the other elements being balanced well to all that mayhem. Rethink your methods.. use the guitars and the vocals and the bass to make the drums feel right in the song. You got a lotta faders there, use them for goodness..

[drumsound]

But I do totally agree with Brad. IF that's how the guy played you can only get it so good. It's a recording of a band not a recording of the engineer...

Tony, I have to disagree with this. In my world, the only option is to make it sound great. I cant just throw up my hands and say, "well, that is how they played it, so F it!"

Joel,

I am in no way saying to give up on making a good sounding mix. I am saying though, that what was played is what was played and there will be certain things inherent to that performance that will never[/i] go away no matter how much you do to the signal.

You can't make a silk purse out of a sow's ear. You can however make something useful and beautiful, but it'll never be a silk purse.

[mcaff]

This is the perfect scenario for the technique in the current tape op.

Here's an backwards example.

You compress it and it sound good on certain hit, let's say you've optimied for the louder hits. the problem is the soft ones are still too soft and the don't get the extra attack because they're not loud enough to cross the threshold.

So, in from ot that compressor add parallel compression. Blend in a compression signal with the uncompressed signal.

That will make only the soft hits louder, which is basically what parallel compression does.

That may ont instantly be intuitive as to why. The reason is that when the snare hit is loud, the uncompresses sound will mask the compressed sound - you won't be able to hear the compressed signal becuase it's been evened out in volume and can't increase. however, when the signal is low, the uncompressed signal drops in volume, but the compressed signal won't, because that's what compression does.

So you're narrowing the dynamic range wihtout limiting it's maximum volume - now you compress that they way you would with a drummer who was more consistent.


To paraphrase the initia post "The dynmic range of teh snare perfeomance is so wide I can't find a compression setting that will work. What do I do?"

The answer is narrow the dynamic range to the point where a compressor can be effective.

The way to do that is to compress twice, but for the first compression use parallel compression ebcuase it's the only way to effect the low dynmaics wihtout effecting the high dynamics, so it's different than just daisy chaining two compressors. It's that difference that makes it natural.