Audiobus: Use your music apps together.
What is Audiobus? — Audiobus is an award-winning music app for iPhone and iPad which lets you use your other music apps together. Chain effects on your favourite synth, run the output of apps or Audio Units into an app like GarageBand or Loopy, or select a different audio interface output for each app. Route MIDI between apps — drive a synth from a MIDI sequencer, or add an arpeggiator to your MIDI keyboard — or sync with your external MIDI gear. And control your entire setup from a MIDI controller.
Download on the App StoreAudiobus is the app that makes the rest of your setup better.
AUM Gain staging
So, i was just wondering what would be the preffered method when mixing in AUM, so i have som subgroups for FX and usally i route drums to one bus and others sound to one bus, but usually i allways get really high input levels on those busses, what should i do to keep the input levels to a good level or does it not matter ?
and good rules ?
iam thinking if i put a Limiter on each channel so it never goes above 0 before sending to subgroups, or should i go even lower ?
Comments
AUM runs as 32 bit float, so headroom isn’t an issue until you record at 16 bit, but putting a limiter before the rec out should do it.
I'm not sure I understand your routing, but it sounds like you're sending all the drums from the channel out of each to the a bus. If this is true you're routing all of your signal through the FX bus.
A more flexible way to do this is to keep the output going to the master, but insert a bus send on each channel. This gives you the ability to adjust the amount of FX you want applied to each channel, and even to send to more than one bus if you want. It rarely makes sense to apply the same amount of reverb (usually none) to a kick, as to a snare, for instance. This will also allow you to control your input levels on the channel. Even though AUM has 32 bit processing, so it won't clip, it's best practice to keep levels under control and only to apply limiting for intentional sound shaping, not compensating for uncontrolled mixing. (says the hypocrite who gets lazy and does that all the time ...)
The only thing you need to remember is to make sure any FX on the mix bus are set to 100% wet. You don't want any dry signal coming through at all or you will be doubling up on the dry part since it would be coming through both the original and the FX channel.
Thanks i will mess around with it. i usually use something like this:
CH1,2,3 Drums - > BUS A -> Master Out (Usually i use BUS H for this)
CH4,5,6 PADS,Synth -> BUS B -> Master
CH 4,5,6 - SEND TO BUS C (BUS C FX BUS)
something like that, no FX on BUS A / B, ALL FX i applied in the FX BUS (sometimes i user several BUSSES for FX)
Based on ProTools but the idea and process is the same...and in there he suggests that limiting is to add gain not remove it.
One difference is that the faders are linear in AUM, making it less important to keep the 0
Good point...It is more important to balance the level against the fader position...i.e When fader is at 0db then the level should be maxing at or around 0 db....to give you the most available fader travel.
It can be helpful, mixing, to not worry about getting maximum gain into each channel, because if you do, you get a weird looking mix, with lots of faders pulled way down, because, like, you don't need your auxiliary shaker track to be hitting 0dB, because, in the mix, it's hitting -24dB. A different mix strategy, worth trying, is getting your gains set, where your faders at up 0 (or -6, if you like having extra volume available) sounds like a rough mix.
To the original poster, i think it's good practice to avoid exceeding 0dB, internally, in your mixer, because even though AUM is 32 bit floating point, your plug in effects may not be- some may be 24 bit fixed point. You don't want to clip your processors, and it can be hard to track down why something is sounding bad, as there may not be a big red indicator that the clipping is happening.
Even if everything in the chain is 32 bit float, that still means there is no incentive to have hot levels for everything- the data system can describe small signal levels with the same excellent resolution as giant levels. Also any processors that react to dynamics, like compression or distortion, have an optimized range for the input signal, like, most compressor plug ins, you can't raise the threshold above 0dB.
I've noticed with ios instruments, there's kind of a loudness war going on with the apps, most of them are pegged with the outputs being right at 0dB, just about to clip. Apps that aren't like that, that keep some headroom, people complain that they are quiet. A lot of developers are forced to hard limit the outputs of the apps, to compete. Bringing the level down in the app, so it isn't coming in hot, at 0dB, to your mixer, means you aren't potentially hitting the app's invisible limiter (if it is equipped with one).
I'm getting serious clipping from Audiokit Synth One in AUM. Even when I pull the AUM faders nearly all the way to the bottom, Synth One is still overdriving AUM and clipping. What is the solution to this? Where do I turn down the gain in Synth One?
EDIT: Aw hell, the answer to my question is well-hidden behind Synth One's "Master Volume" knob. Let me try that....
Limiting is to reduce dynamic range, not to add or remove gain.
Unfortunately, using compression or limiting can sometimes be the only option in "DAWs" that don't have a gain insert, like Gadget for example.
At least we can choose compression with ratio 1:1 to add or remove gain
It would probably be a good strategery, at gain staging time, to simply set each of the instrument out levels at the point of leaving the instrument at about -12dB or -18dB lower than you’d normally want to hear it on its own. That’d be the equivalent of having a gain trim control on the channel input.
A very interesting topic.
I did not really mix in AUM so far, just recorded each track separately and only played around with the mixer for hearing purposes.
But I did the same as @Hansson with routing bass and drums to one mix bus, other Synths to second bus and the rest without bus.
And I saw the same effect of very high levels on the bass and drum bus.
What @wim wrote was new to me and I will eagerly try around this in the evening
But I don't understand the term FX bus.
I was doing it like so:
For simplicity ignore the 32bit floating point 'non clipping' feature of the internal sound engine.
AUM indicates clipping properly with a red master out figure.
(practically ALL converters are linear and cannot handle floating point data)
I you have a full scale channel which sends to a bus, then the master WILL clip when peaks happen to add over 0dB/fs.
With good sources (no obvious noise) there's no need to drive channels up, as mentioned above.
Loudness (impact) is much easier to control during a dedicated 'mastering stage' - instead of adjusting single channels.
Drums usually have a solid level and you'll often want them peaking with some variance as it keeps the track vivid. Some channels may be squuezed intentionally, though - if that's the sound you're after.
For acoustic sources (guitar/vocals in ballad style) it can be helpful to add some balancing (not squezing) limiting, like Pro-L does. The channel mixes much better this way.
(it avoids complicated compressor setups and a lot of automation)