Compression/Volume automation (My take)

[SoundSpiral]

Hi you all!

Another thread here got me thinking...

I know that you probably don't have any idea who I am and I have a low post count 'cos I like to read more than to write. So I'll give you a short background just to give a little bit of weight behind what I'm saying.

-My first touch with recording was at the age 14 (15 years ago).
-I recorded and mixed my first full length label release 10 years ago.
-I have done only AE stuff for my whole working life (of which 90% metal)

Now past the boring stuff and to the point.

People seem to have a bit of a weird "misconception" about when to compress and when to volume automate. I keep running into this so I just have to say something.

My take is that it is almost always more beneficial to volume automate rather than to compress when you are trying to control the volume of a track. You compress for sound and automate for volume. Believe me, I would be thrilled to be able to just put a compressor on an OH track with weird dynamics and be done with it. But it just never seems to work that easily. I just always end up drawing the automation on almost every track to control the dynamics. There is no substitute for a hearing and thinking human going thru the track and making decisions based on all of the available information. Compressor just compresses. I doesn't give a shit about other variables than volume.

There are of course other ways to use a compressor, like side-chaining. But it just doesn't change the fact that compressors have the musical understanding of a pineapple.

I would love to be proven wrong. That would make my life so much easier. Volume automation is one of those infuriating tasks in mixing. But then again, I'm not paid to have fun but to make records sound as good as I possibly can.

 

[~BURNY~]

I'll add saturation and clip gain control to the equation which i find more convenient than pure volume automation in some case but it's often a combination of everything that gets me where I want to be.

 

[mharwood]

Hi you all!

Another thread here got me thinking...

I know that you probably don't have any idea who I am and I have a low post count 'cos I like to read more than to write. So I'll give you a short background just to give a little bit of weight behind what I'm saying.

-My first touch with recording was at the age 14 (15 years ago).
-I recorded and mixed my first full length label release 10 years ago.
-I have done only AE stuff for my whole working life (of which 90% metal)

Now past the boring stuff and to the point.

People seem to have a bit of a weird "misconception" about when to compress and when to volume automate. I keep running into this so I just have to say something.

My take is that it is almost always more beneficial to volume automate rather than to compress when you are trying to control the volume of a track. You compress for sound and automate for volume. Believe me, I would be thrilled to be able to just put a compressor on an OH track with weird dynamics and be done with it. But it just never seems to work that easily. I just always end up drawing the automation on almost every track to control the dynamics. There is no substitute for a hearing and thinking human going thru the track and making decisions based on all of the available information. Compressor just compresses. I doesn't give a shit about other variables than volume.

There are of course other ways to use a compressor, like side-chaining. But it just doesn't change the fact that compressors have the musical understanding of a pineapple.

I would love to be proven wrong. That would make my life so much easier. Volume automation is one of those infuriating tasks in mixing. But then again, I'm not paid to have fun but to make records sound as good as I possibly can.

Completely agree. I think your point is even more valid in the modern world of DAW mixing, where volume automation has become loads easier than in pure console setups.

A midpoint between the two of sorts is some kind of envelope/volume follower VST than can record automation data. Still requires some quality control, but it can substantially reduce the workload for some things.

 

[SoundSpiral]

You have a valid point Burny.
Volume automation usually has the biggest and most controllable effect on the mix balance but of course other factors/processes impact as well.

Though my point is more pinned down to the fact that volume automation is much more efficient than compression.
More often than not, using compression to tie in the balance is not going to work in the expected/desirable way.

 

[indecizo]

I only use compression for sound shaping, all my snare samples have that to sound snappier. I never understood why people smash their bass tracks when you can just automate to get the desired levels right away. It's still a pain in the ass to get right, for sure.

 

[SoundSpiral]

With bass it's important to have a player who has his/her dynamics in check.
That way you can again compress/limit for sound and automate for balance.

You can kind of steer the dynamics to the right direction with compression but it rarely works by itself.

Vocals are also one place where you really gain some benefit in this regard with compression.

 

[Cephara]

I've only used automation when working with acoustics and vocals. I never seen a reason not to just compress with producing a metal track. Most just go for that brick-wall limit. At the same time though, my volumes are usually almost at one dynamic range, so I found compression most useful for peak control.

 

[SoundSpiral]

I'm finding this topic a bit hard to explain.
This might be one of the reasons why there is so much "false logic" around this subject.
For transients it's completely good practice to use a compressor. But when a whole part or instrument is out of whack you really should spend the time to ride the volume.

Compression (and especially the brick wall limiting) will affect the sound in unintended ways. And in mixing everything should always be intentional.

Want the volume down? Turn it down.
Want a compressed sound? Compress

 

[amarshism]

One thing about using a lot of automation is it becomes very important to mix into buss compression and limiting of some description. If you don't then all your levels will be approximate at best.

 

[SoundSpiral]

Oh, one more thing.

Have you ever thought that maybe some of the dynamics are there for a reason?
A drummer might want to have a certain part of a fill louder than the rest.
A singer might want a part softer than the other.

The limiter doesn't give a shit. Level it goes.
And all you were aiming for was to get that loud part in the chorus a tad down.

 

[SoundSpiral]

Thumbs up for automating before compression.
And you might want to automate again after.

 

[Sethis]

I won't lie, I like compression. While it does remove some of the volume changes it also brings the more subtle, quieter details up front. So yeah, in the case of a vocalist for instance it can be beneficial because you can hear every little thing the guy did that would otherwise be lost in the mix. What sounds good live does not always translate the same way recorded. There are certainly some musicians who "mix" themselves but these are very rare. Everything needs to be compressed even when the guy is singing live. In modern music soft is more about timbre than volume, the listener can still perceive the change. Plus when you need to get everything as loud as possible you can't afford to have very uncompressed sounds anyway... Ofc, I wouldn't treat a classical piece the same.

I'm not against vol automation of course, I do use it. Sometimes before compression to at least get the levels in the right direction, and sometimes after it to emphasize or fix something. I just never skip compression.

 

[Kohugaly]

The very first thing I've learned at the very beginning at the times when I was recording my band first demos. I didn't know what compression was and I learned the most coolest trick that I use even now.
Open the vocal track (currently I use it for vocal tracks only) and go phrase by phrase and hard edit volume so they are approximately the same level. I simply use my eyes to look for phases that are obviously overshot/undershot. I usually end up editing cca. 10-20% of the track and it takes about 2 minutes. In the end you have more or less consistent vocal track.
When I learned about compression I stopped using this approach cos' I thought "compression it there to do the thing". But later when I compared my older tracks and the new ones, the old tracks has so much better vocals. That was the moment I've found out that compression is not volume automation, but more of a sort of distortion.

on the other hand there are tasks where compressors do exceptionally well on volume automation. For example the famous andy sneap C4 preset to control the lower-mid range dynamics on guitars. That would be pain in the ass to automate to that point.

 

[SoundSpiral]

Sethis, don't get me wrong. I love compression and use it on almost everything with zero regrets.
You are talking about using the compressor for sound. Not for trying to "automate" the volume without the work.
Again my point is just how you can't substitute volume automation with compression.
And the other way around too. But that seems to be clear for most.

Kohugaly you have definitely grasped the point with the distortion analogy. Not really a distortion but a good way of thinking to separate it from volume automation.

With c4 though we are heading to the annoying world of nitpicking, but in my books with the sneap guitar thing you are really just trying to change the sound and not the volume balance of the track. So basically it's again compressing for sound and fits in with my thoughts on this subject nicely. Multiband compressor in this situation is kind of a smart eq. Not 100% but kind of.

 

[Kohugaly]

With c4 though we are heading to the annoying world of nitpicking, but in my books with the sneap guitar thing you are really just trying to change the sound and not the volume balance of the track. So basically it's again compressing for sound and fits in with my thoughts on this subject nicely. Multiband compressor in this situation is kind of a smart eq. Not 100% but kind of.

Yes, sorry, you're right about the "smart EQ" thing. However in this particular case I tend to think about it as automatic low-mid freq-range control. On palm mutes it damps, cos' they jump too loud and on the rest of the track it does nothing.

Seems it's a matter of point of view in some edgy cases.