This article could well have been entitled "General Signal Flow Basics" because the principles we're about to explore are applicable to any audio situation, whether it's your home stereo, brick and mortar recording or rehearsal studio, or a computer running a DAW. But Logic is as good a place to learn as any, so let's dive right in, starting with four simple words that make up the Golden Rule of signal flow:
(And it's never the other way 'round!)
By keeping this simple concept in mind you can easily get audio signals to flow from point A to point B and beyond! This rule will also prove invaluable when tracing signal flow in an older project where you no longer remember the method of your madness, or when working on someone else's project and you have to decipher theirs!
But if this concept of "outputs go to inputs" seems a little bit too abstract, we can substitute the maxim "sources flow to destinations". Either one is fine.
We see these concepts in action all the time when playing back an audio track. Here, the signal source is an audio region on a track. Its signal is received at the input of an audio channel strip, so we can think of the region itself as an audio output (source) and the audio channel is receiving this signal at its input (destination). Now, we don't actually see an input to the channel strip because we're working in a virtual mixing world where many of the input and output connections are assumed to exist, and therefore hidden. But to truly comprehend signal flow, we would do well to consider each of our virtual audio "devices" (channel strips, effects, etc.) as if they had physical inputs and outputs to receive and send audio signals.
Without any FX inserted in the channel strip, the signal received by the channel strip flows to the channel strip's fader, and outputs from the fader (at the level you've set on the fader) to an output destination, typically the Stereo Output.
Note: clicking and holding on the output selector of a channel strip will reveal a choice of destinations for that signal. Choices include the Stereo Output, Busses, and additional output pairs (if you have a multi-output interface). There's even an "off" function (No Output)! But for our purposes we'll be exploring signal flow out of channel strips from the selection of "Stereo Output", the destination that Logic assigns by default to any newly created channel strip.
A potential source of confusion lies in the fact that the output of our channel strip's fader is routed to a device called Stereo Output, and at first glance this may seem to conflict with the rule that Outputs go to Inputs. But this would only be the case if we overlook the fact that the Stereo Output itself has an input, virtual though it may be!
Once a signal arrives at the input of the Stereo Output, if there are no effects in the Stereo Output's FX slots, the signal will route directly to the fader where you can adjust the level of the signal which is fed to the input of your audio system (this being another "behind the scenes" connection).
In turn, the audio system's output is connected '" this time by actual, physical cables '" to the input of a pair of speakers or headphones which output sound waves into the air. Those waves then arrive at the input of your brain '" more commonly referred to as your ears!
If there are plugins in the audio channel strip, as pictured below, the signal flow would follow the same path as described above except now the audio region's signal has to run the gamut of a series of effects in the audio channel, as follows:
'¢ Audio Region â†' Channel Strip input (which in this case is a direct connection to FX1's input)
'¢ FX1 output â†' FX2 input, followed by FX2 output â†' FX3 input
'¢ FX 3 output â†' Channel Strip Fader's input
'¢ The Fader's output is tied directly to the Output Destination: Stereo Output
'¢ Signal arriving at the Stereo Output's input hits its Fader input
'¢ Fader output â†' Audio System and onwards to your ears
If the Stereo Output was set up with some effects, such as an AdLimiter or EQ, the audio path would be exactly as just described with one exception: audio arriving at the input of the Stereo Output would run through those effects before feeding to fader and onward to the audio system.
The input of the Stereo Output can accommodate the outputs from multiple audio sources: channel strips of all descriptions including audio channels, instruments, and Auxes. Thus, the input on the Stereo Output can be referred to as a summing input. This accounts for our ability to send multiple tracks of audio to a single Stereo Output channel.
Now let's move on to...
In my article on Bus Channel Strips I described the relationship of Sends and Busses, something worth exploring here in terms of "outputs go to inputs".
When you create a Send on a channel strip, what you're actually doing is setting up the ability to tap the audio signal flowing through that channel and send it to a new destination; somewhere other than the channel's output. This is the method commonly used to feed a track's signal to an Aux hosting an effect plugin such as a reverb.
When you create a Send in Logic, you do so by the act of selecting a Bus: a virtual wire or pipeline on which the signal will travel on its way to an Aux. Logic will automatically create this Aux for you immediately after you select a Bus. The output of the Bus is connected to the input of the Aux, and the Aux's input selector is automatically set to receive signal from the Bus you selected on the channel.
The next step is to instantiate an effect plugin on the Aux (here, a Space Designer reverb). By turning up the Send, signal will flow along the bus to the Aux hosting the reverb.
The signal generated by the reverb plug then hits the input of the Aux's fader so you can adjust the level of reverb leaving the Aux and on to the destination chosen by the Aux's output selector (here set to Stereo Output, just like our channel strip).
When our track plays, the dry signal from the channel strip and the wet signal from the Aux are both being sent to the same destination; thus, these signals are summing at the input of our Stereo Output.
Additionally, we see here an example of parallel signal flow, where signal from two sources'"straight signal and processed signal'"are ultimately flowing to the same destination. One much-talked about situation which features parallel signal flow occurs with...
Funny, I worked in NY studios for most of my professional life and we never referred to this as NY-style compression. It's one of those things that makes me go "hmmm". Anyway...
This style of compression, also referred to as parallel compression and used to produce a particular flavor of aggressive drum sound, is created by processing a signal with a compressor (sometimes preceded by an EQ) and mixing that in with an unprocessed version of the same signal. Wow, sounds complicated, right? Truth is, the signal flow needed for this scenario isn't much different from what we explored above. Here, I'll show you...
In this scenario, we're going to play back a drum track on an audio channel, create a Send to tap the signal, and feed it to an Aux. Nothing new there. But this time our Aux is going to host two plugins: an EQ followed by a compressor. The "straight" signal comes from our channel strip directly to the Stereo Output, and the processed signal is going to get there from the Aux. Signal routing-wise, this is no different from our reverb scenario above. Our straight and processed signals are being sent to the Stereo Output simultaneously or, to put it another way...
In a future article on signal flow we'll explore a variation on this setup that will give us greater control over the setup needed to produce effective NY-style compression, as well as additional insights into Logic's signal flow.
Check out Peter Schwartz's MIDI 101: MIDI Demystified tutorial-video here!