Compression vs Automation

ah, what about first change your tracking technique step back from the mike 10-12 inches and or place your pop filter at a consistent distance and better control the dynamics at source ?

Actually, that's not entirely true.. I'm use to performing with SM58s and I instinctively move toward and away from a mic. One of the problems I'm having these days is the I'm not always getting a great monitor track of my voice so I may be over and/or under-compensating here and there.  Nonetheless, I have a very wide dynamic range as a vocalist and I use it.

it's a technique fab dupont recommends for vox and is more critical with good mikes which are more sensitive and may also have proximity effect

the distance reduces the dynamic range so it/signal is more constant but still allows you to set a good level for good gain staging

di ing bas is completely different as there is no mike, so then as the bass players stated on jk's other thread, they control their dynamics with their playing, ( same idea , they get it right at source) and then use compression aAso, the bass's dynamics could be s a set up and older string problem.

As players/ singers we are always the source and what we do effects the signal quality and with your bas playing perhaps instead of thinking about individual notes, feel the phrase you are playing in the context of the song structure and pay it like a breath , have you noticed how the really good bass layer's lines just seem to fit a in perfectly; thats not luck its technique and judgement wish I could play like that !

experiment with the constant distance technique for vox; it works

ok. firstly.

everything you are describing to me says not enough comp during tracking.

then not enough comp suring mixdown.

then.

fader time.

ditch the pencil. record a real fader moving about and use that as your fader automation.

I definitely push the faders up as the vocalist tails off sustained notes. i never push a fader and leave it. it needs to be constantly on the move. when i ride the fader i literally ride it up and down in time to the music. like a tremello on a wind instrument. i never just push it up in the quiet bits and pull it down in the loud bits.

sometimes i'll send all of the doubled tracks to a separate channel and use that for the fader rides.

G, Thanks..

Comp during tracking?? Why? I use to do this but stopped because I started reading that there's no reason to.
Summary of reason: dynamic range on a digital track is very large, and you can't undo what's been tracked, right?
That said, I'm finding myself using more and more compression in my old age, but it's a love/hate relationship.

Yeah - It's been a while since I've pushed a real fader.. Right now, all I've got is a mouse and an iPad app for pushing faders realtime. I do know that when I have used a real fader in the past, I still ended up going in a tweaking on the map, but it was a great place to start.

That last idea is interesting.  Just a thought, but could I do that by just copying the track and riding the fader on the second track only?  Effects issues aside, it seems I would have more subtle control. For example, let's say the tracks run at the same level.  Moving the fader 0.1 db on the automated track would have the effect of moving both faders 0.05db, right?

So this comes down to the type of aesthetic and personal work flow preference as I see it.

For me, vocals and bass are the 2 tracks I will compress during tracking. I'll also compress E Guitars sometimes as well. I'm too chicken to compress any other instruments during tracking.

Vocals will always go through the WA76 hitting 3-5 dB GR or 5-7 dB if it sounds good. The studio standard was(is) usually to follow the 1176 with an opto like an LA2A, CL1B. Obviously this varies widely but most people are using serial compression (following one comp with another) during tracking and many will do it a 2nd time during the mix. The amount of squeeze is up to you but I'm a big fan of small reduction in stages - taking a couple dB off here and there resulting in a nicely compressed vocal but without compression artifacts.

So on to automation.....I switch to automation when either the vocal is sitting nicely in the track or I'm not gaining anymore sonic benefit from compression. If you like the sonics of the vocal and it's sitting snug, start going through and just automate phrases, words, syllables in order to keep the sound that you worked so hard to attain.

Bass is a different animal and one that I'm working on right now. It is so helpful if you can play the part evenly from the start and compress it on the way in. How much comp depends but it's not uncommon to hit it pretty hard if it sounds good. Check out @noah shain's and Ward's recent posts regarding their treatment of bass in their new projects. They're hitting it hard and it sounds awesome but also that genre kinda' requires it. Bass is the foundation - I prefer it not to move but stay steady as a rock. What about the dynamics? Well, keep in mind that things can actually sound more dynamic when there are a couple of static elements like kick and bass to reference.

Experiment like crazy! Use serial compression, over compress until you hear flabby transients, artifacts, distortion and poor dynamics. It's very important to know what these things sound like. Realize them than back off and then start automating. There are so many ways to achieve what you want. I'm not the most experienced either so I hope some others chime in because this is a great topic!

You have to use a compression type that doesn't invert the dynamics, or start sounding bad when accidentally hit hard.  Some of the DBX types using the non-linear detector will make quiet parts sound louder than they should be and vice-versa.  The Langevin ELOP I have is really great up to a point, but clearly damages the sound when a big vocal transient slams the meter.  A tube vari-mu type with the time constants set incorrectly can do real damage.  

Because many singers react to the compression.  I find some will subconsciously back off their peaks a few minutes after I've set very light compression grabbing 2-5dB occasionally, and by take 2 they will hardly ever trigger the compression, or it'll start landing in the 1-2dB range.  Some singers are deaf, and notice none of this!

Layers of compression, each doing very little, usually sounds better and is more effective than hitting it all in one pass.  If you can use a very transparent and invisible compressor on the way in you'll have the first pass already done.  Just my 2 cents....

I compress a little going in, usually WA76, sometimes a little LA2 plug, then edit only one or two egregious vocal lines. Still, it's not perfect, because when I occasionally put another compressor on a finished vocal track, it rides the automated volume right back up to compensate for the now lower level. So, I guess more compression going in, and better mic technique equals little need for automation.

Thank you everyone!  So far, there's a pretty good range of responses, which has me feeling a little better about my current approach, which is to automate after I've gotten things as good as I can with compression.

There are a couple of surprises.  The first is the number of you that compress on the way in.  We all believe easier in things if they're what we want to believe in the first place, and this is no exception.  Despite what I perceive is the current prevailing argument to be found elsewhere on the net that there's no good reason to process on the way in, my gut says do it - and I don't even have the gear to do it with ..yet.  [sigh]  The other is the concept of preference for incremental compression.

Randge's post got me thinking - if most processing is pre-fader, and a little compression on a little compression is better than straight compression, then might it not be a good idea to take a track, add some compression and automation, then RENDER the track with automation - so you can smooth it out again with a little more compression? One advantage I can think of right off the bat is that you want to apply verb as late as possible, right?

It also depends on how much 'immediacy' you need in a mix too, and that varies with different genres.

In the present state of pop and new alt-rock needing everything as loud as possible, compression is critical to keeping everything up and 'in your face'. Lots of compressors too! Almost everything I track in new alt-rock gets tracked with multiple compressors and more compression in the mix, plus sub-mixes and buss compression. But not a lot each time, just 1-3 db usually. And yes, of course you ride faders in automation.

But it's all so material dependent. Jazz, swing and country don't get nearly as much compression. And there's a lot more fader riding in those genres.

In many ways your comments speak to my situation, but I realize they are generalizations, so I'm not taking them personally. Nevertheless, I feel somewhat obligated to respond.

Yes I'm not working with other musicians.. At least not face 2 face - but I suppose I could be if I really wanted to - In fact, I just had a RealGear 'guest' over to lay down some drum tracks. But you're correct, right now I'm not getting the instant feedback that can be had when playing with others.  Yes I don't have an engineer riding a fader while I sing, but actual singing represents a tiny fraction of the time I spend working in the studio.  Do engineers really ride faders while Willie Nelson tracks?  You wouldn't know if from the video I just saw in another thread that's hitting the boards today.  Correct me if I'm wrong, but my sense is that unless you know exactly what they're going to sing and how they're going to sing it, you're taking your chances when you ride a fader. Fade the wrong way and you can do irreparable damage to a take.

Luckily, I'm not serious about making records for others. I make recordings for me.  Screw the industry and the game. I'm in it as a serious hobbyist. I write and record to have fun.  With that in mind, performance and integrity of the music is more important than the quality of the recordings. I write as I record ..when I want and when I feel like it ..and if things go experimental - I can run with it and not feel pressured by the clock. Even if I had everything prepared before going into a pro studio, the pressure would be on ..and if I'm off, I'm screwed, and that's not fun. For many of the reasons described above, in some ways, the quality of what I write is better than it would be if I did it the old fashioned way. Don't get me wrong, I try to make the best recordings I possibly can, and to that end - I'm willing to 'amass' some equipment that some (including my wife) might consider overkill.  But in these days of plugs, that's not such a tall order.  Even the standard pre's that come with a mid level interface aren't that bad. As ericn intimated, all you need is the right mic, which I haven't really found yet. Actually, I'm in the market for a nice chain for my vocals - Probably a pre/mic/and maybe an EQ. Honestly, (and making your point for you in some ways) the most expensive thing in my studio right now is the treatment.. I've easily spent more money on bass traps and baffles and flooring than on any single piece of equipment in the studio.  

Yes I'd be willing to take things to the next level in terms of hiring others to put down tracks should I ever need to. I suppose if I ever really felt the need to shop a tune, I'd evaluate and consider taking a few stems to a pro studio or even re-recording everything from scratch and hire musicians to play parts, but I'm more a singer-songwriter type, so I feel that's unlikely.

Some of my favorite mics and direct boxes almost act like there is a bit of compression built in, this is why I can't emphasize mic choice enough!
A beginner has to understand this is as much an art as learning to play, contrary to what the ads say not everybody can learn how to be a recording engineer.
But remember a big part of learning is screwing up/ learning what won't work!
People pay a pro for there skill and to get the results they want not the method they want.