Two of my favorite subjects in recording: signal flow and levels. Simple and often taken for granted, yet grasping them well is your safe haven in times of trouble.
Let’s start with signal flow. It’s basically a sound wave or an electric voltage (in this case music) that starts somewhere and ends up hitting our ears. It flows in a direction, with one thing outputting signal, and another receiving it as input. Its charming when someone wants to replicate the exact sound of an instrument that they’ve heard on another record. Easy, right? Well, an approximate is. But when you look at the signal flow, you begin to understand how unique every situation is, and how literally impossible it is to replicate any recorded sound truly exactly. Let’s say it’s a guitar you are recording. Following the signal flow goes something like this:
1) Fingers holding a pick pluck a string, which vibrates and creates a sound wave at a given frequency. How a person does this is as varied as the amount of humans there are.
2) This wave goes to the magnets of the guitar pickups.
3) The magnets turn this into an electric signal that goes through the wire and electronics of your guitar, out of the guitar output jack.
4) The guitar cable picks it up from there, and transfers it to the input of your pedals.
5) In and out of each pedal, until it goes out of the last pedal.
6) Another cable picks the signal up, and takes it to the input of an amp.
7) Then it runs through all of the amps circuitry, gets amplified accordingly (hopefully by tubes/valves), and then goes out of the output transformer.
8) Which goes into the speaker cable, and the speakers.
9) The speakers are another magnetic device, which receives the electronic version of the original sound wave generated by your pluck of the string, and the speaker cone moves back and forth pushing air in a way that replicates this waveform.
10) A microphone in front of the speaker captures this output with a small diaphragm that gets pushed by the air that is being pushed by the speaker, and turns this into another electronic signal.
11) This signal goes through the microphone electronics, and out into another cable.
l2) The cable takes it to a microphone preamplifier, which takes the small signal from the mic and turns it up louder.
13) Here’s where some options can occur. You might run the signal through more devices (eq or compressor), or you might go straight into your recording device. Let’s say you are recording on your computer through an interface like a Duet or something. So, the signal up to this point has been a solid waveform. Analog. But it is now converted into a digital representation of that. Which is like a connect the dots in reverse. 48,000 times a second, the converter takes a snapshot of the amplitude of the signal waveform, and that volume is a number. These numbers are output into your USB or Firewire cable into your computer. As these numbers are written to the hard disk in the form of a .wav file, a visual representation of this is displayed by your software. A connect the dots line that looks like an analog sound waveform.
14) Once in this state, the signal can be processed by plugins in the software, to reshape the waveform delightfully. And of course there’s the internal signal flow of your recording software, which can really frustrate you if you don’t understand it.
15) Back out of the computer it goes, through your USB or Firewire cable and into your interface.
16) This time, the numbers that are the dots that make up your pretend waveform are converted back into an analog, electrical signal.
17) Out of your interface it goes, into some sort of cable, connected either to your monitors, or headphones, where another speaker cone pushes air in a way that represents the signal’s amplitude and frequency.
18) Finally, this waveform as air molecules hits your eardrum, another diaphragm similar to a microphone, vibrates the bones in your ear accordingly, and turns that into an electric signal that your brain interprets as sound and music.
What could go wrong? A helluva lot, actually. Each step of the way is unique to every situation. Every cable, mic, string, converter, speaker…everything. And especially every ear, no two of which are exactly the same. No two people hear ANYthing exACTly the same.
Anyways, the point being: every step of the way is an opportunity for difference, improvement, and trouble. If you find yourself in a mess when you are recording, then it is in your best interest to understand every step your signal goes through, and what direction it’s travelling in while it goes through, so you can solve the problem. Ignorance of this may be bliss, but it is also embarrassing and costly when you lose time having to call someone who does know. Not to mention there is a unique delicacy of delight that comes from abusing the rules and practices that you know are good and right.
At every step along the way, the signal is at some sort of level. Volume, voltage, whatever. It has a measurable level. It is crucial to make sure that every point of connection between devices is matched appropriately. All ins and outs of devices are designed to operate at some optimal level. Ignorance of this is a great way to screw things up, and cause yourself grief. More specifically, I want to mention what level you actually record things to disk or tape. With tape, there was a whole set of rules, which have now been largely abandoned since tape has been largely abandoned. Things like, you couldn’t record high frequency stuff like cymbals too loudly on tape, or they would distort in a yucky way. Mid and low frequency stuff distorted nicely, though. In the digital world, its slightly different. There’s just a simple maximum volume you can record at (0dBFS), and you don’t want to go over it, or your waveform gets the rounded peaks harshly flattened in a way that sounds/feels yucky. But one step further, there are good practices that will help you maintain good sound and sanity. There are many great, searchable articles on this, but I’ll summarize with what my own methods are. I like to have my individual tracks maintain an average around -18dBFS, and hit peaks around -6dBFS. How do I measure or know these levels? My favorite plugin ever: The PSP Vintage Meter. I set the OVU reference level to -18. The needle will move a bit higher and more bass-oriented stuff, and a bit lower on more treble-oriented stuff. But it shows the average or RMS signal level in a way that I understand, even if it’s a bit old-fashioned. Reminds me of the meters on the tape machines I used to use. There are more metering plugins out there now, but this has been around for a long time, since I started recording in the computer 10 years ago, and is the one I like the most. For peaks, I just watch the metering in the software. It is usually good at showing with a little number what the peak of things like drums are hitting. Averaging that peak around -6dBFS keeps things balanced and cool, and most of all manageable. Your plugins will be more agreeable to those levels. Your mix will come together more easily, without having to keep pulling your master fader back.
For the master level, I strap another Vintage Meter on after all processing and set the OVU reference to -14. I’ll let the meter bump up to +3 a bit in loud sections. Basically, this replicates what we used to call a “hot tape level” (Which reminds me of my other favorite plugin, the one I’ve been waiting a long time for someone to make: FerricTDS from Variety Of Sound. Feels and sounds like my old MCI tape machine with Ampex 456 tape). When you mixed down to a half-inch tape machine, and you let it go that hot, it compressed and did awesome stuff. In other words, I use the old levels we used to use for tape machines and vinyl records and cassettes as my benchmarks. I tried to make super loud records back in the days when we first figured out you could, and it was fun. I mastered a demo for Jason of his band The Emergency that was the loudest record we’d ever heard at the time, which was kinda fun. But, the bummer is: it’s already loud. If you turn it up, it hurts. At least to me.
But, everyone’s ears are different.
