Beginers' Compressors

[wooden]

Thank you. I've done that, but i dont like to use the channel's own fader, and i have done this to "flattering" a bit the dynamics, and then have the channel's fader to set it in the mix, but maybe what i am doing is overkilling and more confusing and time consuming... yes, i think ill use the fader instead

[where02190]

Originally posted by wooden For "leveling" do you use the master fader as automation, or how do you do it?

 

 

I never touch the master fader, it remains at 0, where IMHO it should. PArticularly in a DAW, lowering the master fader can result in overdriving inputs to the summing amps, and, since the meters on most DAW master faders are post fader, you have no idea that it's happening until it's pointed out to you, track faders are lowered and master fader restored to 0, and you hear the life come back into your mix.

 

I pretty much take Phils approach, doign a few passes using the fader automation on our (Soundtracs Solologic) console, then if necessary fine tune with volume automation either in our Mackie HDR or Protools. For vocals, this is usually in conjunction with a light compressor, usually a TL Audio C-1 or Focusrite Green at 2:1, with jsut a touch (2-3db on peaks) reduction. It's as much for character (of the compressor) as dynamics control.

[Kiwiburger]

A lot of people are mixing in DAW's that use 32 bit floating point, so I don't think there is any worries about "overdriving inputs to summing amps".

But - when your route the digital audio to a D/A converter, that's when you need to concern yourself about the headroom of the converter./B]

With level automation - you don't always have to use the track fader. Sometimes it's nice to keep the dynamics automation seperate from the track fader, so you can 'mix' the relative track level without upsetting all your automation nodes.

You can automate a compressor output level, or eq output level, or anything else that takes your fancy. Or route it to a group channel.

[Phil O'Keefe]

A lot of people are mixing in DAW's that use 32 bit floating point, so I don't think there is any worries about "overdriving inputs to summing amps".

What does "overdrive" mean in a digital mixing situation? These are mathematical gain and addition calculations. Since you can't clip a floating point calculation, you can't "overdrive" anything.

But - when your route the digital audio to a D/A converter, that's when you need to concern yourself about the headroom of the converter.

Please support that position with a detailed explaination of why you think that is the case, and how the mechanics / mathematics of a 32 bit FP calculation prevent overloading a mix bus.

With level automation - you don't always have to use the track fader. Sometimes it's nice to keep the dynamics automation seperate from the track fader, so you can 'mix' the relative track level without upsetting all your automation nodes.

IMO, that's what the "trim" plug in is for. Sometimes I'll just adjust the output knob of a compressor I have plugged in on a track... either one of those usually is fine for when I have the track's automation pretty much where I want it overall, but the track itself "needs to come up / go down a dB or two" relative to the rest of the mix. However, if you have a preferred method that differs from that, please give us some specifics!

You can automate a compressor output level, or eq output level, or anything else that takes your fancy. Or route it to a group channel.

Is that what you were referring to? IOW, is automating a compressor's output level control your "usual" method of automating levels for a track? That seems a little convoluted and unwieldy to me, but if it works for you...

[Kiwiburger]

When I first got into automating track levels, I ran across the problem of doing all this work setting up nodes to duck certain parts, and insert silence and fades etc ...

And then I just wanted to drop the whole level of the track. What to do? It's a real pain having to carefully select all nodes, and nudge the whole lot up or down. It's easy to accidentally not select all the nodes.

I just find it's easier to keep all the busy automation (which you do in solo mode) seperate from the final mix level. That requires two faders - or some other level control. Depending on whether i've set up a group channel or not, I find it just as easy to automate a compressor output. Or whatever plugin I might have inserted pre fader.

Re: Mix Bus Overloading ... Where and Phil, you obviously have a different understanding to me - so probably i've got it wrong. Or maybe it's your years of analog training coloring your opinions?

Are we even talking about the same thing? I'm trying to understand. I appreciate a conventional analog mix bus can be overloaded - because it's analog, and has a non linear region before the point of clipping.

But - help me understand this - in the digital domain, there is no non-linear point before clipping - is there??? (Unless you code it with dsp - i'm not taking about intentional saturation effects).

I'm no rocket scientist - but I understand 32 floating bit calculations to be somthing like scientific notation on a calculator. You never run out of digits - you can't clip 32 bit floating point. I have proved this myself, but that's unnecessary because it's widely understood that you can't clip 32 bit floating.

If you don't believe me - try this: take a mix, and set all the track faders at the maximum, and the master faders at maximum. You will hear extreme clipping - from your 24 bit fixed converters. But export this as a wave, then reimport it. It will still sound clipped, but now drop the master fader and hear the clipping go away. If the wave file data was actually clipped, this would never go away. You would just see brick wall clipping, and the mix would be ruined. But with 32 bit float, the data is never clipped. You only experience clipping if you overload the D/A converter.

So this is why i'm mystified when professionals I respect talk about "overloading the mix bus" in the digital domain? I don't understand what you can possibly be talking about?

[Phil O'Keefe]

You can take the converters out of the equation... use a sine wave generator within your DAW to create a ton of tracks with -.1 dBFS waveforms.

[Kiwiburger]

I'll try that .. but i'm not expecting to see any signal degredation whether I exceed 0dBfs or not.

Things would be very different in Protools with fixed 24 bits - is that what you mean? I use Cubase which uses 32 bits floating.

I guess you mean all clones of the same sine wave - otherwise how would you view any distortion?

[Phil O'Keefe]

PT LE is 32 bit floating point, and has been for as long as I can remember.

[wooden]

I'm sorry guys, i meant channel fader not master fader...

[guitwizz]

Id have to say, the FMR has a beginners price, and sounds awesome from what all I hear, but it may not be "beginner" knob-wise. Something like the Aphex 107 Easyrider 2 ch. comp can be had for 60-100 logs on ebay. Its pretty transparent, and 3 gnobs per side. Would be a good learning tool, and sounds pretty good.

 

No mission-critical stuff can really be done on this, but its great at simplicity.

[where02190]

The FMR RNC is a fantastic unit for very short money.

[UstadKhanAli]

Originally posted by guitwizz Id have to say, the FMR has a beginners price, and sounds awesome from what all I hear, but it may not be "beginner" knob-wise.

 

 

Threshold, Ratio, Attack, Release, Gain....

 

you lookin' for anything different?

[guitwizz]

Im not lookin for anything, it was just a suggestion per the "beginner" question, which was the original topic.

 

Maybe beginner meant price-wise, instead of function wise.

Sorry if I missed the intent.