Music Hardware

# Music Hardware #

While you could make all of your music with just computer, you’ll probably want some tools that make your life as a musician easier and let you explore ideas faster. Like I said back in the Audio Effects Chapter, in my experience, making music depends on a really quick brain to input device feedback loop. Being able to turn a physical knob and mash physical keys will often lead to better results if only because you’re able to try things both faster and by physically interacting with something that you may not have thought of otherwise. It’s why guitars, MPE keyboards, and Eurorack are so great: they all give you a crazy amount of control and expression that is literally at your finger tips. The brain to input interface there is just so much better.

So, while I don’t think you should go crazy deep into hardware and spend your life savings, there’s some things that just make sense to get, like a good keyboard MIDI controller, some knobs, and at least one realllly expressive instrument like a guitar or MPE controller. Something that really lets you feel what you’re playing.

First things first, I also want to go over the difference between Digital and Analog in music hardware:

Digital: 1010111010010010110100101110101

Analog: ∿∿~∿~∿∿∿~∿∿~∿~∿∿∿~∿∿~∿

A quick note before you read anything further: you can only hear an analog signal. At the end of the chain, everything needs to be analog, as ultimately sound waves exist not as the 1’s and 0’s of binary, but as waves of pressure in the air- which is an analog medium.

To start, let’s look at an all-analog signal path. Here, a guitar string is plucked, the signal is picked up by the guitar’s pickup (as the metal string causes a magnet to move, generating a weak electrical signal), the signal is amplified (twice), and then the amplified signal is used to move a speaker cone, which in turn moves the air, so you can hear it.

Note, I’ve drawn a light gray cable along side all the analog signals. This is the ground cable. Audio circuits are no different from any other circuit, and require a full, well, circuit. A circuit is a full loop, with ground. Where the cables come out of the amplifier, I’ve drawn these two wires much more distinctly, with a red wire- meant to visualize the ‘live’ wire that carries the signal- and gray wire the ground. This is by convention. If you ever have to wire up an amplifier, you’ll probably see one red and one black wire, representing these in turn. In many cases (but definitely not always), things will still work if you get these backwards, but you’ll end up with audio that is out of phase, as your signal is quite literally flipped. In the case of a guitar amp with a single speaker, this probably won’t matter. But, if you have two speakers (left and right) and only flip one, this is very likely to make things sound awful.

Hopefully that made sense, now let’s look at a digital setup. Keep in mind even the most digital of paths needs to end in analog so that there’s something for you to hear.

Here, a digital keyboard is hooked up to a laptop, presumably running some sort of sound-generating software - Maybe a DAW or VCV Rack. Here I’ve shown the audio Interface (labeled DAC) as an external box hooked up over USB, but this could also just be internal to the computer. All computers or phones will have an internal audio interface, it’s just that most musicians end up attaching their own that offer better features.

Here, everything on the blue cables is digital, it’s only 1’s and 0’s. The first cable is only carrying information about what note you’re playing and knobs you’re turning, while the second one (between the laptop and the the DAC) is carrying digital information that represents the audio, but it hasn’t yet been turned into an analog signal. That’s the job of the DAC. It takes those 1’s and 0’s and turns it back into an analog signal.

Often you’ll hear the terms Audio Interface, Sound Card, and DAC used interchangeably. They’re not all the same thing, but generally it’s the part that turns the digital signal into an analog one. If you’re plugging a guitar or microphone into your computer, the terms Audio Interface and Sound Card still apply, but in that case you’re using an ADC. I’ll come back to this in a bit.

If youre confused about how digital 1’s and 0’s can represent an analog signal in the first place, refer back to samples in Chapter 2 - Sound Sources. For more information on how Digital to Analog and Analog to Digital conversion works, check out this video from xiph.org.

Some devices may have almost everything in this chain internally. This is true of a lot of digital instruments, like the Elektron Digitakt pictured here:

The Digitakt is a drum machine that uses samples saved internally as .wav files. It’s basically just a very specialized computer in a box.

Though the really confusing thing to beginners is when the device is digital, but has both analog input and outputs. Just like there’s a DAC (Digital to Analog Converter) for analog output you’ll often be working with ADCs or Analog to Digital Converters to take your analog input and use it on your computer (or internally, like in a digital guitar pedal)

For the most basic possible example, let’s look at a bit-crusher and sample rate reducer pedal. These pedals basically just hook a DAC directly into an ADC, then let you purposely mess up the conversion.

Bit crushing, as explained back in the effects chapter, is just purposefully reducing the range of values a sample of digital audio can be. So, while an 8-bit audio point can be any whole number up to

$$2^8 = 256$$ , a lower-bit sample might only be able to be represented by $$2^6 = 64$$ possible values, which will really start to sound pretty distorted. Similarly, sample rate reduction distorts the audio by intentionally limiting the rate at which new digital samples of the input audio are acquired. As long as samples are gotten at a rate at least twice that of the highest frequency in the input audio, there will be no difference, but as this goes below that, the converter simply doesn't get enough data to reproduce the input signal.

This effect is inherently digital. It’s really just purposely doing digital to analog conversion poorly. If we want it as a guitar pedal, that means the pedal has to first convert the analog signal to digital, then convert the digital signal back to analog.

note, this isn’t meant to be a real schematic. It’s heavily simplified.

Hopefully that was enough to get you up to speed on the basics of analog vs digital. Unfortunately, while some gear is obvious (A normal acoustic guitar is analog, a MIDI controller is digital) some music gear really blurs the lines, or even twists them outright. The Behringer DeepMind line, for example, is a purely analog synth with digital control’s and a big ol' screen. Meanwhile, the Modor NF-1 is a digital synth that looks and tries to emulate the sound and feel of analog equipment. Plus, there’s a ton of synths now that have analog oscillators and filters, but then do DSP (Digital Signal Processing) based effects or the other way around- with digital oscillators and analog filters. What I’m trying to say is, it’s complicated.

A few more notes:

I want to point out that there is a valid concern that your ADCs and DACs are of decent quality. Some people are crazy about this and want to spend thousands of dollars on brand name ADCs and DACs, but honestly I’ve found that it’s more just a matter of ‘is it good enough’ is really just the bar to hit, as long as it’s not causing a ludicrous amount of noise and can capture enough of the dynamic range (difference between playing softly and loud) of your instrument you’re good to go. In my experience, USB powered devices tend to have the worst time with noise, as the power going in is so ‘dirty’ that the DAC/ADC ends up putting this noise into the signal. A lot of equipment will discuss this in regard to the Signal to Noise Ratio (SNR).

Some people get crazy about sample rate. For a final recording there is absolutely no reason to exceed 24bit/48khz audio. Even that is really overkill. On the other hand, during recording there can be rather significantly audible differences in oversampled (that is 80khz+) audio, as the virtual instruments have more headroom to work with to avoid a side effect of digital audio called aliasing. You can learn more about this in

Samplerates: the higher the better, right? from FabFilter on YouTube.

Finally, I want to briefly mention that as you add more equipment, you’re very likely to run into an issue with ground loops. If you plug in a piece of gear and start hearing a hum or significantly more noise, this is likely the problem. Check out 6 simple and cheap ways to fix hum, buzz and ground loop noise from Loopop on YouTube.

Hopefully that all made sense and you feel a bit less lost about both the differences between analog and digital, and why it is probably one of the most divisive topics in music hardware. Let’s move on barreling though all the technical crap so we can get to the fun stuff later on. Let’s start with something you’re already familiar with, but dive in a little deeper: A normal ol' audio jack.

## TRS? 3.5mm? ¼ inch? #

Your normal every-day headphones that you’d plug into your phone (or did. Thanks Apple (¬､¬)) use a 3.5mm Tip Ring Sleave cable. If your headphones have an in-line mic, they’re probably TRRS. The difference, as this meme shows, is how many ‘rings’ are on the connector. Normally, for headphones, the Tip would carry the left channel, the Ring the right and the Sleave the common ground for each. If there’s an inline mic, the connections usually go Left, Right, Ground, Microphone on the TRRS pins respectively.

Don’t assume all devices will support the TRRS headsets with mics. Some devices will have a dedicated microphone in line, and others will just omit it outright.

But, because nothing can ever be simple, that’s not always the case. Enter, balanced audio.

While uncommon for 3.5mm cables like most headphones, on a lot of pro audio gear that uses the much beefier ¼ inch jacks will use a TRS cable to carry a mono signal.

[TODO] Balanced, XLR,

## MIDI #

### What is MIDI? #

Musical Instrument Digital Interface, or MIDI is what it sounds like, the primary way to get data between devices digitally. Want to tell a hardware synth to play a sequence you arranged on your computer? MIDI. Want to play notes into your computer using a keyboard? MIDI. Want to sequence drums, with varying ‘velocity’ on each hit? MIDI.

### Sounds Great? #

For the most part, yeah. Everything works together and you can make all your hardware speak the same language. Let your keyboard talk to your computer and then your computer talk to your drums so that you can play the drums with your keyboard! Send automation from your computer to a parameter on your synth to vary the sound over time, whatever.

The Catch? MIDI is outright ancient by technology standards, having come out in 1981. It’s so damn old, that it’s (mostly) a 7-bit standard. Now, ideally, a musician shouldn’t have to know all this shit and the gear should stay out of the way. Unfortunately, we’ve been sticking with this standard for so damn long that basically everything abuses it in one way or another to the extent you sorta have to know how it works. So, 7-bit, what does that mean?

Well, it’s talking about bits, so 1’s and 0’s. For each message in midi, you get 7 bit’s of data. So, when you turn a knob it can range from 0000000 to 1111111, which, works out to be 0 to 127. This means each knob, even if it feels smooth to you, only has 127 distinct levels whatever it’s talking to can receive. This is really bad. But, wait, it gets worse. This applies to *almost everything in MIDI, so how hard you hit the keys and how finely you can set the volume with a physical slider. Clearly, this blows.

And, it so happens, everyone agrees. Because of that, there’s a whole fustercluck of solutions. Some you might see include

• MIDI Non-Registered Part Number (NPRN) is one way MIDI controllers can send higher resolution signals (14bit so, 0 to 16384) by putting two, 7-bit CC’s values together such that one controls the Most Significant Byte (MSB) and the other the Least Significant Byte (LSB)
• Open Sound Control (OSC) isn’t MIDI, but rather a competing standard that’s much higher resolution and can work over ip (wifi), but it’s not universally supported like MIDI
• Mackie Control Universal (MCU) is a fustercluck of a non-standard developed by Mackie, a particular hardware manufacture, for one of their products. Originally MCU was made for Logic but eventually the control ‘standard’ wound its way into other DAWs. It mostly provide a ‘universal’ mapping for common functions like mute, solo, track select, EQ and what not. It basically just sits on top MIDI.
• Midi Polyphonic Expression (MPE) is probably the most convoluted of the workarounds, but it will require some more explanation, so I’ll come back to this in a second.

“Alright”, I hear you thinking, “so MIDI is a fucked up standard, what is it good for then?”

Well, it’s pretty much still the only one you can sequence notes in your DAW, that all instruments interface with each other and a computer, and often the only way you can control digital instruments and effects from hardware.

So, let’s poke into some of the common MIDI message types, starting with the most obvious: Notes.

### Types of Midi Messages #

#### Notes #

MIDI notes range from 0 to 127, with the highest note, 127, being a G9 @ 12543.9hz and the lowest, note 0, being a C-1 at 8.176hz. Obviously this is more than the standard full 88-key piano.

Given human hearing starts at about 20hz, the lowest notes are inaudible except for harmonics assuming no octave shifting or other quirks. As such, these lowest notes are often repurposed for control messages, though even then a lot of sound sources will only respond to a limited range of these notes anyway.

While MIDI does have an extension for supporting alternate tunings, it’s rarely directly supported. As microtonal and other non-12TET (12 True Equal Temperament) scales have become common, various tools exist to use MIDI pitch to force notes to a chose scale anyway.

General MIDI, one of various MIDI extensions (that often get ignored anyway) also defines a few specific instruments to belong to specific midi channels (rarely used unless listening to midi files directly with soundfonts) and a mapping for drums, which is used a bit more often, and is generally close to the normal mapping you’ll see in tools like Ableton’s Drum Rack. If you buy an electronic drum kit or use a drum machine, it’s likely to at least try to respect this mapping.

General MIDI Drum Map
KEY NOTE SOUND
35 B0 Acoustic Bass Drum
36 C1 Bass Drum 1
37 C#1 Side Stick
38 D1 Acoustic Snare
39 Eb1 Hand Clap
40 E1 Electric Snare
41 F1 Low Floor Tom
42 F#1 Closed Hi Hat
43 G1 High Floor Tom
44 Ab1 Pedal Hi-Hat
45 A1 Low Tom
46 Bb1 Open Hi-Hat
47 B1 Low-Mid Tom
48 C2 Hi Mid Tom
49 C#2 Crash Cymbal 1
50 D2 High Tom
51 Eb2 Ride Cymbal 1
52 E2 Chinese Cymbal
53 F2 Ride Bell
54 F#2 Tambourine
55 G2 Splash Cymbal
56 Ab2 Cowbell
57 A2 Crash Cymbal 2
KEY NOTE SOUND
58 Bb2 Vibraslap
59 B2 Ride Cymbal 2
60 C3 Hi Bongo
61 C#3 Low Bongo
62 D3 Mute Hi Conga
63 Eb3 Open Hi Conga
64 E3 Low Conga
65 F3 High Timbale
66 F#3 Low Timbale
67 G3 High Agogo
68 Ab3 Low Agogo
69 A3 Cabasa
70 Bb3 Maracas
71 B3 Short Whistle
72 C4 Long Whistle
73 C#4 Short Guiro
74 D4 Long Guiro
75 Eb4 Claves
76 E4 Hi Wood Block
77 F4 Low Wood Block
78 F#4 Mute Cuica
79 G4 Open Cuica
80 Ab4 Mute Triangle
81 A4 Open Triangle

### Types of Hardware Controls #

#### Potentiometres, Sliders, and Encoders #

+Motorized, Wheels

### Interfacing #

DIN, TRS-A, TRS-B, USB, USB host adapters, etc.

clock jitter

Note, there’s a pretty limited number of MPE capable instruments, thankfully there’s stil a decent price range between them. This list is not exhaustive, but I think gives a good sampling:

Small:

Artiphon Orba

Haken Audio ContinuuMini

Big:

Roli Seaboard (Block, Rise)

LinnStrument

Haken Audio Continuum

## Audio Interfaces #

Weird side note:

Some VOIP software (Discord, Zoom, etc.) may want you to use a Sample rate of 48Khz and a Buffer size of 192. No clue why.

[TODO openback v Closed]

[TODO Planar magnetic]

## Eurorack #

Modular Grid an online tool for dreaming about hardware setups (Eurorack, Guitar Pedals, and a few other modular formats)

## Cool Hardware #

Blokas Midihub

• r_cycle for Pure Data Library *most Launchpads

T1 Algorithmic Midi Sequencer

Sixtyfour pixels MIDI controlled relay

Sixtyfour pixels ‘Noodlebox’ sequencer

Sixtyfour pixels ‘Hack-Du-Strum’, a page feautring mods of their strummable controller, similar to an omnichord

Sleepy Circuit’s “Hypno” CV live visuals box

## Notes on Hardware NOT to buy #

• The Teenage Engineering Pocket Operators are cool, fun, etc. But they’re pretty fragile and have some issues. Generally, I don’t think they’re worth it. Here’s Some alternatives (Yes, I know there’s a huge price difference- remember, you get what you pay for.)
• Analogue Pocket
• DityWave M8
• or, if you’re feeling spendy, the OP-1 and OP-Z are options. Each have their own quirks though, so be sure to do your research. Regardless, I really don’t recomend the TE PO’s
• The non-mini Launchkey Series from Novation. They keybeds are absolute shit. I had to mod my Mk2 to make the pads work, and the mod wheel has issues too. It’s just… not good.

## Notes on how your hardware will be more expensive than you think #

[TODO] Balanced audio, group loop isolation, impedance matching, isolated power supplies, cables … so many cables

## DAWless? #

DAW-less just means not using a Digital Audio Workstation to make your music, usually though this means not using a typical computer at all. Here’s a good example of a DAWless performance:

10 Things You NEED to know before Building a Dawless Synthesizer Setup (YouTube, BoBeats)

If you would like to support my development of OpGuides, please consider supporting me on Patreon or dropping me some spare change on Venmo @vegadeftwing - every little bit helps ❤️