Retune/Retone that Kick or Snare with a Test Osc!

[fader8]

You've been there. You love the beater sound, but you just wish the kick were tuned up or down a semitone or two. Then that snare has just the right pop for your tune, but the sizzle's just not quite right. EQ and pitch shifting aren't cutting it.

 

A couple folks asked me to write up how to enhance percussion with a test oscillator plug-in. So since today was a serious dental day for me and my tongue and jaw are in Gumbyland somewhere, I thought it would be distracting for me to write this up.

 

Here's how you do it:

 

With your clean kick track ready, assign one of its Sends to a bus. I used Bus 10 in the attached example. Option-click the Send so it's set to unity. Delete the Aux channel that gets created automatically.

 

Insert a HiPass plug on the kick channel and tweak it until you've got rid of a lot of the fundamental tone and just the beater sound is left. I used 190Hz here.

 

Next in the mixer, create an Instrument track and instanciate a Test Oscillator plug in the Instrument slot. You want a sine wave. Set it to about 70Hz and turn the gain down all the way for now.

 

Now insert a Noise Gate plug on the Test Osc track and set the Side Chain input to Bus 10. I set the Low Cut on the sidechain filter to about 180Hz to be sure I was only triggering on the higher stuff. That should tighten up the kick a bit too.

 

Now bring up the level on the Test Oscillator. Tweak the NoiseGate threshold, attack and release for the shape of the thump you want and adjust the Test Osc's level for the amount of thump. Now grab the frequency control and retune your kick!

 

In a nutshell, the kick gets its fundamental removed. The sine wave is muted by the noise gate until the kick hits and the sine wave plays based on the envelope set up in the noise gate, thus replacing the kick's fundamental tuning.

 

 

 

The attached Logic Pro 8 file also has a snare example. This is set up the same way except the snare sound is LoPassed and then the Test Osc is set to pink noise, so it replaces the decay of the snares. One of the cool things about this technique, is that you can choose to just highlight dynamics with it by adjusting your threshold where the noise kicks in.

 

The samples I put in here are a bit sucky, so hopefully your drums will sound better!

 

Have fun,

fader8

 

Uh-oh, the local is wearing off. Where did I put that bottle of codiene . . .

fader8-BD-Sn-Osc.zip

[dkgross]

mmm....codine....mmmmm..

 

thanks man! good stuff, and it make me want to maybe use a ES1 or something instead of just a sine wave....great food for creative thought!

[joshuarule]

the fun just never ends...

 

i often would add as an osc to a kick...

 

but this technique probably gets some pretty sweet sounds.

[ChicoSatis]

Thanks Much. Can't wait to try this out.

[davidpye]

Been doing this for ages, works well in certain circumstances, Metal Bands LOVE IT, unfortunately....... they come back........

[fader8]

Been doing this for ages, works well in certain circumstances, Metal Bands LOVE IT, unfortunately....... they come back........

 

Indeed, the basic technique is from the "oldest trick in the book" archives. To the point that it's usually included in 1st or 2nd term mixing classes. There's a distinction however between using it to simply layer in an external tone, common in electronic styles and yes, metal, and using it to replace part of the sound itself. If used with care and attention to detail, there's no reason it can't be used effectively for sweetening acoustic drums in any genre.

[brianm]

just play the EXS24 without loading an instrument. no need for a noise gate on an oscillator.

 

nice idea for the snare though.

[fader8]

just play the EXS24 without loading an instrument. no need for a noise gate on an oscillator.

 

That won't achieve the same effect. The audio engine runs at a much higher resolution than the MIDI engine, allowing for a lot more timing precision for this kind of operation. For example, the above session has the attack for the gate set to 9mS. By setting the side chain filters so it's triggering the gate on the sharp percussive attack of the beater, the 9mS attack delay gives the beater a chance to sound before the tone reaches full level. It's too fast to notice distinctly, but it improves the definition of the kick. It's a subtle ducking effect.

 

To achieve this kind of precision with an EXS24 under MIDI control would require quite an effort of audio to score and tweaking. While you'd still have envelope control, you would lose threshold control which is nice if you only want the effect on harder hits. Seems a lot more work for less precision and flexibility.

 

On the other hand, MIDI triggering a sustained tone with the EXS and gating that instead of a test oscillator would allow you to select an exact note and even change that note during the sequence if you wanted to give the kick a different tuning during the chorus or bridge, for example. That's definitely cool, and easy.

[brianm]

On the other hand, MIDI triggering a sustained tone with the EXS and gating that instead of a test oscillator would allow you to select an exact note and even change that note during the sequence if you wanted to give the kick a different tuning during the chorus or bridge, for example. That's definitely cool, and easy.

 

It's even easier than that, the test oscillator responds to MIDI too. In your example you've got it on an instrument track, just record some notes on it.

[fader8]

the test oscillator responds to MIDI too.

 

Who'd have thought!? Cool.

[djals]

Fader8, when you delete the Bus does this mean that it has no output?

 

Whats the reasoning behind this, just to keep a tidier work space?

 

Thanks for the tip!

[fader8]

Fader8, when you delete the Bus does this mean that it has no output?

 

Whats the reasoning behind this, just to keep a tidier work space?

 

You don't delete the bus, only the aux that get's created automatically whenever you activate a new bus send on a channel. So yes, the bus has no output, as in it has no return to your mix. The bus signal itself can still feed any side chain though.

 

As far as tidying workspace, yup, the last thing I want is to look at a bunch of auxes that aren't doing anything. The automatic aux creation thing is often more of a nuisance for me than a feature. But unfortunately, it can't be turned off.

[djals]

Ok, thats good to know. In my projects i still have all these extra Auxes for side chaining, now I can tidy up a bit.

 

Thanks.

[brianm]

Why not use the Aux? the idea is that you should now use Auxes and not bus objects. The Aux can do everything the Bus can do and more.

[fader8]

Why not use the Aux?

 

Primarily because auxes can't be a source of signal to side chain inputs. Unless of course, you send the aux signal to a bus first!

 

the idea is that you should now use Auxes and not bus objects.

 

That would be inconvenient. Note that there are two attributes to a bus, the bus path itself, and a bus object that can be created to process that path.

 

The Aux can do everything the Bus can do and more.

 

Incorrect. A bus (path) provides an audio source that can have multiple simultaneous destinations. A bus (object) provides a preprocessing point as well as a master fader control for all the bus destinations. In this respect, it's the same as a Pro Tools master fader channel, (non-VCA).

 

An aux, on the other hand, is little more than a return channel. A means to return something to your mix, like a bus/submix or an instrument channel. Or, an intermediate channel before sending off to another bus.

[David Nahmani]

The Aux can do everything the Bus can do and more.

 

Incorrect.

 

Thanks Fader8 for clarifying the difference.

 

The misconception that an Aux channel is a Bus channel with extra options (sends, inputs..) is rampant in the Logic world. Many advanced users think an Aux is simply a better version of a Bus (channel), and pass that misinformation on to newer users.

 

The way I usually attempt to explain it is:

 

• the bus is a virtual cable, allowing to "bus" the audio signal from a channel strip to another, or in that case from a channel strip to a plug-in.

• the Bus object is the master fader for that cable, and it allows you to process the signal on that cable, before it is available to any other destination (such as an Aux channel). The Bus object affects the signal on the bus.

• the Aux object has nothing to do with either the bus or the bus object, but it can tap its input from a bus, AFTER it has been processed by the Bus object (when present). The processing on the Aux object does not affect the signal on the bus, only the part of the signal that was routed to that Aux.

 

A typical application is parallel processing: let's say you want to apply NYC compression to a drum kit, or parallel compression on backup vocals. You output all your tracks to a bus, you can process and adjust the bus on the bus channel, then use various Auxes that have their input set to that bus. While the settings on the bus channel object will affect the whole bus, thus all the Aux objects, the settings on the Aux object do not affect the bus at all, and therefore do not interact with each other. That's why you can do all your "serial" processing on the bus object, all your "parallel" processing on the Aux objects.

 

Again, as Fader8 said, you can use only Aux objects and no Bus object, but then that means you now have to use 2 busses (virtual cables) for an application that would only require one when using the proper tools for the job.

 

EXAMPLE:

Let's say you have two vocal tracks, and you want to submix them, EQ the submix, then have some of that vocal mix panned left with a distortion effect, and some panned right with a phaser effect.

 

1) Set the ouptut of both vocal tracks to bus 1.

2) Insert an EQ on the Bus 1 Channel Strip.

3) Create two Aux Channel Strips, set their inputs to bus 1.

4) Insert a distortion plug-in on Aux 1 and pan it left.

5) Insert a phaser plug-in on Aux 2 and pan it right.

 

To get the same results without Bus Channel Strips you would have to use two busses (and three Aux Channel Strips).

[djals]

Yeah, this kind of made more sense to me in L7 than it does in L8.

 

What's the deal with always using an Aux channel when you want a Bus?

 

And, when I delete the Aux, where is my Bus channel strip? If I toggle the view selector to Bus, in the Mixer, it reads Nothing to Display

 

In short, where are the Busses and can I view them in the Mixer window?

 

This message worries me too.

[fader8]

Thanks Fader8 for clarifying the difference.

 

My pleasure. Between both our explanations, it should be pretty clear. But I agree this seems to be one of those things that can elude some folks.

 

Yeah, this kind of made more sense to me in L7 than it does in L8.

 

Nothing's changed really except for the auto-creation of auxes.

 

What's the deal with always using an Aux channel when you want a Bus?

 

Logic assumes you want to use the bus for an effects send, therefore it provides you a convenient return path.

 

And, when I delete the Aux, where is my Bus channel strip?

 

Logic provides bus paths but does not automatically create an associated bus channel object. That's fine as you don't always need one. When you do need one, simply open the environment audio layer and create as many as you like.

 

The Mixer view in Arrange can't show you objects that don't exist in the environment.

 

This message worries me too.

 

Here again, Logic is assuming that you need a return. If you don't, then just click "Delete Anyway".

[brianm]

Ok, I meant in Fader8's example, an Aux would function the same as a bus object. But I went through the steps again, and realized you aren't actually doing anything with either the bus or aux object. Just using the bus as a send destination for the sidechain. So in that case, sure, delete the Aux, or mute it or set it's output to zero.

 

But in most cases, the "Logic 8" way to look at things is that the bus is really only used in the background. We use the Aux to tap into that bus.

 

David, in your steps for the vocal processing, I would :

 

1) Set the ouptut of both vocal tracks to bus 1.

2) Insert an EQ on the AUX 1 Channel Strip that gets created.

3) Send the track to Bus 2 and 3.

4) Insert a distortion plug-in on Aux 2 and pan it left.

5) Insert a phaser plug-in on Aux 3 and pan it right.

 

Same number of steps, but in my mind is easier to do because Logic created all these Auxes for you. You would have to create 3 objects in your method - 1 Bus and 2 Auxes. My method has the same #, but all three are Auxes and I didn't have to make any of them.

[fader8]

Same number of steps, but in my mind is easier to do because Logic created all these Auxes for you.

 

brianm,

Yes, you could do it that way. And in simple mixes where that's the only thing you're using that bus for, that's fine. The number of steps is unimportant as those of us who use bus objects already have them in our templates and autoload.

 

The point is that as mixes become more complex, a bus can become a real octopus, having many sources and many destinations. Sometimes you don't know this starting out but as you get on with it and you're a few days down the road with the mix, you start to appreciate that you kept your signal flow simpler from the start, you have enough busses, and you have a master control of bus levels at the ready.

 

So it's really about workflow and what works best for you.

[David Nahmani]

1) Set the ouptut of both vocal tracks to bus 1.

2) Insert an EQ on the AUX 1 Channel Strip that gets created.

3) Send the track to Bus 2 and 3.

4) Insert a distortion plug-in on Aux 2 and pan it left.

5) Insert a phaser plug-in on Aux 3 and pan it right.

 

And that would give you the same sonic results, but if you have quite a bit of processing to do in a session you would quickly run out of busses and Aux objects. I used 2 Aux objects and 1 bus, you use 3 Aux objects and 3 busses. Remember you are limited to 64 busses and 256 Aux objects per project.

 

BTW if you don't want to use a Bus object, you can still save at least one bus (vs your solution).

 

1) Insert an EQ on the Aux 1 Channel Strip that gets created.

2) Set the output of that Aux to Bus 2.

3) Create Aux 2 and Aux 3, set both their inputs to Bus 2.

4) and 5) are the same.

[brianm]

I can't image more than 64 sends would be needed in most cases. Bus objects are soo 3 years ago...

 

 

Bus Channel

You can only create bus channel strip objects in the Environment. In general, you will

not need to do so, as all audio bussing (send/return routing via the Send slots of

channel strips) is handled by aux channel strips.

It is primarily included for compatibility with older Logic Pro versions, and for use with

certain audio hardware devices.

[fader8]

Bus Channel

You can only create bus channel strip objects in the Environment. In general, you will

not need to do so, as all audio bussing (send/return routing via the Send slots of

channel strips) is handled by aux channel strips.

It is primarily included for compatibility with older Logic Pro versions, and for use with

certain audio hardware devices.

 

I had a good chuckle when I read that too!

[djals]

Fader8, thanks for the breakdown, now it makes more sense to me.

 

I think i've been abusing L8's more straight forward approach, time to brush the cobwebs off that old Environment!

 

To all:- Your knowledge is priceless, I thank you sincerely for your time.