Lured by the siren song of modular synthesis and DIY electronics, but not sure how to navigate the piles of requisite knowledge – or uncertain what the trip down this rabbit hole might have in store?

For years, Tom Whitwell’s Music Thing was a beloved daily read, as that site and this one were among the early blog-format destinations for music tech. Tom moved on – something about a major day-gig at a paper called The Times, perhaps named after the font? – but that makes us all the more delighted to get a dispatch from him. In this guest column for CDM, he introduces one project, a brilliant FM radio sequencer, but also helps us catch up on reading on modular synthesis and electronics dating back to the origins of the technology. And he has a realistic look at what this will do to your life – all inspired by “pure enthusiasm,” as he puts it, “this is fun, you should try it.”

Hey, isn’t that what the drug dealer said in those just-say-no instructional videos we watched in the 80s? Coincidence, I’m sure. -PK

Since buying a Eurorack modular synth a year ago, I’ve spent a lot of time building DIY synth modules and reading about synths and the people who build them. (See reading list, below, if you’d like to do the same.)

The hardest part of DIY electronics is starting out. My first step was building a few guitar pedal kits and learning by reading the Beavis Audio site. Other people start with noisemaker kits like the Atari Punk Console or circuit bending. They all lead in the same direction — down a very deep rabbit hole. 

There’s a lot to buy – a kind of infrastructure you need before doing anything – soldering kit, a multimeter, and a stock of components. None of it costs much, but it’s hard and disconcerting to buy. Online megastores like Farnell or Mouser will stock 50 versions of every component. Get the part number wrong, and you accidentally order capacitors as small as grains of sand, or as large as golfballs. Smaller stores – in the UK, I use http://www.bitsbox.co.uk/ - are easier because they only stock common hobby-friendly parts. 

After making a few guitar pedals, I moved onto synth modules. They’re a great DIY platform. The infrastructure is all there, in terms of power supply, case, inputs, and outputs. Parts are cheap, there’s a healthy and helpful community, and a nice learning curve, from basic utility modules to mind-bendingly complex frequency shifters and vocoders. 

In a year, I’ve built:

• a super-simple, chiptuney oscillator
• a tiny spring reverb driver
• a stupidly-complicated and blinkenlights-covered Arduino-powered Euclidean beat sequencer
• a very useful Arduino MIDI clock
• and a simple but handy 8-step sequencer (see video, below)

For this project, I was inspired by this quote from Don Buchla, the legend of west coast synthesis: 

“My studio at that time was ten feet wide. It was so crowded in there we hauled the workbench out on the sidewalk on good days and set up my oscilloscope and worked out there. [John] Cage came by and for voltage control I had hooked up my keyboard to an FM module that I’d built, a little module that was an FM receiver and I could play stations on it because I had one of the first varactor tuned FMs. Cage, as you can imagine, was just enormously interested in the fact that I could tune each key to a station and then proceeded to play the radio” ( Source [PDF] )

Thirty years later, Don released the 272e module (see Matrixsynth on the announcement), a $1250, four-channel polyphonic FM Tuner. There’s also the ADDAC102, a very fancy stereo €270 Eurorack module [see Synthtopia, with a video]. I wanted something quick, cheap and easy that would let me follow in Don and John’s footsteps. After a lot of searching and a few dead ends, I found the wonderful video demo, below, of a battery-powered FM sequencer based on a €15 radio kit from Germany. 

Projects like this follow a predictable curve. There’s a burst of experimental excitement at the start; receiving the crucial part, building the circuit on breadboard and realizing that — YES! — it’s going to work. 

Then comes a period of frustration and tedium. Re-buying a crucial part you blew up. Fiddling with the circuit so it responds just how you want it. Transferring the breadboard layout to a piece of perfboard, or designing a PCB and waiting for it to be made in China. If you’re using an Arduino or other programmable controller, there’s a long period of writing code, battling feature creep, debugging. 

During this period, you have to really, really want the thing you’re making, dreaming of how cool it will be, how much fun you’ll have playing it and telling everyone about it. 

Tom's FM radio-sequencing module project, in all its glory.

Building music gear is more multidisciplinary than you might imagine. The interface and the feel is as important as the functionality. My Euclidean sequencer is a cool-looking thing, with a big LED matrix. It’s really useful – turning trains of pulses into Afro-Latin rhythms. But it’s fiddly and annoying to use. The FM Radio module could be 50% smaller – and size is important in any modular synth – but this time I wanted good big knobs for fine tuning the signals and control voltages. 

So, as the project continues, you’ll spend time designing a front panel, deciding how many knobs you need, removing ones you’ll never use. And along the way, you’re learning. This time round, I wanted to get the control just right – precise, stable tuning so that stations would stay locked. That meant experimentation and [asking for help on the MuffWiggler forum]. I also spent ages reading ham radio sites, trying to work out how to make a voltage-controlled Shortwave radio (I gave up). 
Eventually, the lacquer is dry on the panel, the parts are all in, debugging is complete and the module is working. The result: either elation and fun, or almost immediate maker’s remorse. It’s bad enough spending money on a piece of music gear that you never love. It’s really annoying spending time building one that you can’t then flip on eBay. 

So far, this FM module is pure fun, an injection of random audio in the heart of the system. Every time I turn it on, something else comes out – pirate dubstep stations, Turkish music, news reports and Bryan Adams. You can filter it, sequence it, use it as a noise source, or let it modulate oscillators or open filters. Listen:

Radio sequencer 2 by MusicThing

Photos of the module:

Reading List

Great online resources for learning about modular synths and the first golden age of experimental electronic music include: 

Ubuweb’s electronic music resources section 
Also at Ubuweb, several editions of Electronic Music Review, a beautifully-designed but short-lived journal boasting Robert Moog as Technical Editor. 

The Red Bull Music Academy includes long, detailed interviews with Don Buchla, Tom Oberheim, Peter Zinovieff of EMS, Robert Moog and Morton Subotnik.

Synapse magazine was a mid-70s journal of electronic music, where you’d find DIY projects from people like Serge Tcherepnin

Vasulka is a huge and rather poorly-organised archive of documents, interviews and transcripts, containing some gems.

Source Magazine was, back in California in 1967, a plush avant-garde journal. Many editions came with 10″ vinyl records, pages printed on transparencies or fur. John Cage was a guest editor, and the magazine carried experimental scores from composers like Steve Reich. Original copies sell for $500+, but the articles and scores have been collected in a book: 
Source: Music of the Avant-garde, 1966-1973 [Amazon]

What’s Next?

Tom is already on to the next build since he finished up the radio sequencer. This time, it’s a shift register sequencer. A what?

A 16-step random sequencer, something between the Wiard Noise Ring, the CGS Gated Comparator and Nav’s BITSY.

It takes random noise to fill up 4 x 4 step 4015 shift registers, shifted by a clock input. The shift registers are looped – either after 8 or 16 steps. 8 of the steps are fed into a DAC0800 analog/digital converter, which produces a 0-8 volt output.

See also the prototype:

If you liked the generative, probability-based sequencing seen earlier this week, here’s another example – and it’s free and open source, so if you do want to pick it apart and you own a copy of Max/MSP or Max for Live, you can. Co-creator Giuseppe Sorce points us to the work:

This is a simple generative music synthesizer built in Max/MSP created by Diego Caponera, Nicolò Paternoster and Giuseppe Sorce. It involves 5 FM generators which play notes randomly based on a root key and intervals defined by the user. It’s an university project made for an exam for Sound’s Science degree ( Math Department of Roma Tor Vergata ).

The software is distributed “as is”, without any warranty, under a GPL license.

Not bad for an exam project. A heck of a lot more fun than a paper (and, believe me, I say that wholeheartedly having done a bit of university teaching).

Grab it:
http://code.google.com/p/smilesynth/

First revealed last month, PreenFM is an open source hardware synth. As the name implies, it’s an FM synth, with some very serious specs: up to six-operator FM synthesis with some nine algorithms, up to 4-voice polyphony (depending on algorithm), glide, selectable LFOs, modulation matrix, and preset banks with SysEx support. It’s all usable via a display and MIDI support.

It’s also fully open source hardware; whereas early efforts often had commercial restrictions attached, PreenFM is free for use under the GPLv3 and Creative Commons. And it’s got a unique platform under the hood: the open source LeafLabs 32-bit development platform gives this some serious horsepower. It’s very much in contrast to the ultra-inexpensive 8-bit brain of our own MeeBlip synth; think of the MeeBlip as an exercise in what you can do with a little two-stroke engine versus the V8 muscle in this. (The creator says the MeeBlip helped inspire his creation – yes, synths are multiplying!)

You may have glimpsed the PreenFM making the rounds online, but I got creator Xavier Hosxe to tell us more of the gory details and share some sounds.

CDM: So this is all based on the Leaf platform?

Xavier: Yes it’s built around a LeafLab board.
I coded a first version on the Maple [development board]; then when they announced their “Maple Mini,” I realized it was going to be very easy to plug it into a PCB.
I’m not directly connected to [LeafLabs]; I participated in the forum and learnt many things from the team.

They are very friendly and helpful.

What was it like working with the Leaf?

The LeafLabs boards uses an ARM Cortex-M3 microcontroller.

It’s a 32-bit chip runing at 72Mhz that can do 32-bit multiplication in 1 clock cycle, has 128Kb of flash [memory] and 20Kb of RAM. That seems very few but it’s not, PreenFM software uses 92Kb for the moment.

LeafLabs provide a Linux/gcc toolchain that allows to develop in your IDE of choice… Eclipse in my case, which is very confortable.

They also provide a strong bootloader and some libraries that worked perferfeclty for my needs : Usart (Midi), I2C (EEPROM), LiquidCrystal (LCD).

What will you get in the PreenFM kit?

All you need to build yourself a complete synth: PCB, screws, resistors, ICs, audio/midi jack, box, 20×4 LCD, encoders, knobs, buttons… even an USB cable [for power].

You’ll also get a Maple Mini board with PreenFM soft preloaded. The Maple Mini is easily updatable, and you can experiment lots of different things with it.

PreenFM C++ source code is available on GitHub. It’s easy to read and modifiable. If you want to see your name to welcome you on the boot screen, go ahead

To build the kit, you only need a soldering iron and some solder.

There will be 2 differences with the photos you can see on the site: the final PCB will be blue (I should receive them next week). [Ed.: See the photo in blue at top; the orange one is included here for variety! -PK]

Xavier also sends along some welcome news:

Here is a sound I can get with my soon-to-be-released StepSequencer feature in PreenFM.

This is a single voice of a simple 3-oscillator voice. 1 very slow LFO + 2 * step sequencer routed to the modulation indices.

PreenFMStepSeq by cdm

Here’s a sound sample:
PreenFM 1 by cdm

http://www.preenfm.net

I took a look at Omnisphere 1.5, the synth so big it’ll make your head hurt, for Keyboard in a story out now (and readable now). As I begin the story:

Seeing its six DVDs of sound content, you might be tempted to duct-tape a key down and let Omnisphere finish your film scoring gig. While the director would probably love the results, you’d be missing out on the real fun.

In other words, what I discovered in that review was that Omnisphere, particularly with additions in the new 1.5 update, is a powerful creative sound design tool, not just a preset machine. The highlights:

• Granular section, pictured, made nicer with the ability to combine with glide and intelligent parameter control design
• Harmonia, which allows you to control each harmonic component of a sound independently using individual oscillator and synth controls
Waveshaper, which can not only add bit-crush-style effects, but work its magic on each element, polyphonically
• Individual independent arpeggiator, plus MIDI file drag-and-drop
• An Orb for exploring sonic capabilities – we’ve seen these sort of X/Y controllers before, but here you can even dynamically assign parameters in realtime, and add features like inertia and gesture recording
• iPad control via a really wonderful controller app

Being able to navigate multiple sonic parameters in real-time with touch, and combining sound-bending, far-reaching sonic tools like Harmonia and the granular features means you can really take sounds far from their original source – and sync them to tempo, if you like, with those MIDI and arp features. I need to pick up the whole tool again after the review and see what new sounds I can make; if there are any other users who wish to share, I’d love to hear what you’re making, as the possibility is really deep.

And yes, they now have done a dubstep bass tutorial, so everybody playing at home should take a shot.

Read the full, detailed review at Keyboard Magazine:

Spectrasonics Omnisphere 1.5 [Review by me for Keyboard]

How does it all work? Here are some relevant videos:

We looked at the arc controller, and interviewed creator Brian Crabtree, early this year. In a way, the design is as much conceptual, kinetic sculpture attached to a computer as it is music hardware. It’s not for everyone, but it does inspire some sound designers and composers whose work I love, giving it a secondary advantage – without owning one, I can still see people doing interesting things with it and find musical discoveries in their work.

stretta in his latest video turns the controller into synthesized sounds that reference in the title of the piece the Dharma Wheels, the objects of Buddhist reflection. A cycle of birth and rebirth seems as appropriate to the act of musical composition and sound creation as metaphor I can imagine. And so, the wheels of the arc, ringed by dancing LED lights, become sparkling meditative wheels in motion, powered by simple, FM-modulated sounds in a ringing reverb. (That reverb is one of my favorites – Eos by Audio Damage.)

The work is all Creative Commons-licensed. In that spirit, I’m going to copy streta’s thoughts below, as I think anyone who enjoys sound design will get a kick out of his thought process.

holocene by stretta

hardware: arc4
software: electric dharma wheels

I received a production-run arc4 with the final firmware on Friday. This signaled a mad scramble to update my work for that and the latest serialosc with arc support so it’ll be ready when people start receiving their units in a few days. So what do I do on Saturday? Make a new app, of course. Sure, that totally sounds like the responsible thing to do.

After receiving the arc4, I thought it might be a good idea to produce an example that demonstrates a ‘bank’ of encoder values that you can switch between. That gave birth to an application idea involving triggering modal notes from a pool of probabilities across three octaves of scale degrees. There is a separate bank of pitches depending on clockwise or counterclockwise rotation so you can shift the harmony with a simple gesture. The weighting of scale degrees is programmable and editable in real time on screen or with a MIDI controller. This allows for a more controlled structuring of compositional development over longer periods of time. The speed of the rotation determines how often a note is triggered, and can also be used as a modulation parameter for the FM synthesis engine.

Relevant synthesis parameters are also editable on the arc as the notes are triggered. The state of these parameters is overlaid on the LEDs, so interesting patterns emerge when this mode is engaged. There was a really awesome bug where switching editing modes also transposed the output modally, so I built in a score feature that allows you to advance a programmed chord progression with a button push.

A sit-the-arc-in-your-lap-and-doodle app has been on my mind a lot and I have at least three good starts in this area, but other priorities have often pushed these out of the way. The prototype arc2 I had lacked the mounting bracket for the USB cable and the logic board was floating free inside the enclosure, so I always had to use it (carefully) on a stationary, flat surface. It is really nice to have an arc that can be moved around or used in the lap. My cat disagrees.

I recorded this video, holocene, as a demonstration of this app, which I’m calling electric dharma wheels. This is the raw output from the electric dharma wheels, with some Eos reverb added after the fact.

more information about monome can be found at monome.org

more information about me can be found at stretta.com

http://stretta.blogspot.com/2011/03/holocene.html

For years, anyone with doubts that soft synths had become commodity needed only to look to Crystal. Not only is this proprietary synth completely free-as-in-beer for Windows and Mac, but it could easily be your only synth if you so desired. Crystal is a semi-modular instrument with the usual trimmings – subtractive and FM synthesis, modulation, multi-stage, graphical envelopes, tempo sync, chorus, flanger, comb filter, delay, and even a band splitter for adding effects per frequency band. If the bread-and-butter stuff doesn’t satisfy, Crystal also has Granular and Wave Density parameters for wavetable synthesis that can be used for granular-style effects and more out-there sounds.

Now, Crystal pairs the absurd value of its free desktop instrument with a mobile version that starts at 99 cents.

While it’s not quite the round-trip workflow of bleep!BOX, mentioned earlier today, Crystal’s new mobile edition for iPhone and iPad could be ideal for people who can never find enough time (cough) to do all the sound design they want. Once the currently-unreleased (and premium-priced) XT plug-in is available, you’ll be able to share patches between desktop and modular versions. You’ll also be able to use MIDI input via the iPad’s Camera Connection Kit, using a compatible USB MIDI class-compliant device. And if you’re stuck for ideas, the mobile version also uses the “Breed” feature. That means that you can do some idle sound design while on the go, so that knob-tweaking doesn’t have to distract you when you’re back at your desk and ready to actually finish a production.

Now, we have quite a few readers who aren’t terribly interested in mobile versions of software – and with good reason. While Crystal on an iPhone or iPad is portable, you lose a lot. Some of that involves sacrifices in Crystal itself. Some editing, import, and patch browsing is missing. The limited real estate of the iPhone screen knocks out editing, which is largely copied from the desktop version – meaning, as well, this isn’t necessarily designed from the ground up for mobile use. And, oh yeah, you lose polyphony. Other restrictions have to do with the iPad itself: finding compatible MIDI hardware is trickier, and the single dock connector – for now – means that you can pick one option from MIDI, external audio, and power, but not all three, depending on your hardware. (Want MIDI? Load up the battery. Want audio? Stick to the headphone jack. Update/correction: Glad to hear from substrain that hardware with class support for audio and MIDI in one device should work, and that there’s hope from third-party, non-Apple adapters! More on that soon; have to experiment.) Then there’s the fact that just sequencing an instrument is easier on a computer. For half the price of an iPad, you could get a netbook with all these features and still have something light and compact. My guess is, then, that Crystal will be best for doing sonic exploration on the go, particularly with the iPad editing interface – and that makes the availability of the free plug-in version all the more critical. On mobile only, this has questionable utility. When you have a desktop, too, it could be a killer app.

The news isn’t just for mobile users, however. While some pricey synths and effects aren’t available in 64-bit versions, an update last week also delivered 64-bit compatibility for both Mac AU and Windows VST. That gives you full 64-bit compatibility on both operating systems; no bridge needed. Cost: entirely free.

Thanks to readers who reminded me of this, especially Flplsx of In & Out fest. Full details:

http://www.greenoak.com/crystal/
http://www.greenoak.com/crystal/indexDesktop.html [Desktop version]
Crystal on the iTunes App Store (eagerly anticipating that XT version!)

Pocket Piano from Critter and Guitari on Vimeo.

Apologies to the immense powers of lumbering studio gear, but a new lifestyle may be forming around unique, mobile, small, simple synths. The latest entry comes from none other than Critter and Gitari, some of our favorite electronics designers, based in Philadelphia. This time, they’re touting a pocket synth. No MIDI, no control voltage – just wooden keys, some knobs, an audio out jack, and a speaker. But the killer feature is, it runs on batteries. That allows you to take it anywhere, including – as evidenced by the video – on the Staten Island Ferry.

The sounds are decidedly lo-fi, but varied in synthesis methods:

• Vibrato Synth
• Harmonic Sweeper
• Two-Octave Arpeggiator
• Octave Cascade
• Mono FM Synth
• FM Arpeggiator

Twist the knobs to select mode and waveform, with a colored light to give you feedback. Then play on the wooden keys, though they require a bit of what the creators describe as “a refined touch.”

Lots of additional sound samples, including some that sound like they escaped from a vintage arcade cabinet (or a really cheap alien spacecraft):
Pocket Piano

Bonus: Here’s a wonderful recorder called the Kaleidoloop from the same builders, costing $299. They’ve been documenting its many powers over the past months. It’s insanely simple – to the point that somewhere, KAOSS Pad engineers are scratching their heads — but also insanely delicious.

Kaleidoloop: Effecting a Voice Recording from Critter and Guitari on Vimeo.

So, you’ve seen lots of interesting looking iPhone apps, but most of them strike you as gimmicky. Others have interesting workflows, but limit you to working on the mobile device, not switching back to a computer. And maybe you’re perfectly happy with a phone running Windows Mobile or Palm OS.

Enter SunVox. This is not a mobile music making app for the timid. It’s a powerful suite of soundmakers and sequencers, baked together into a modular environment that lets power users tweak to their heart’s delight. It’s small, it’s fast, and it looks – and sounds – a lot like early computer music programs. It’ll run on iPhone now, but also on Palm, Windows Mobile, Mac, Windows, and Linux. It’ll run on your netbook, your MacBook, and your ThinkPad.

Incredibly, all this goodness is yours on all those platforms for ten bucks and on iPhone for $5, easily making SunVox the biggest steal in music software I think I’ve ever seen:

• Flexible architecture that adapts to slow and fast CPUs
• Synths and generators: FM, virtual analog, FFT-based “SpectraVoice”, Kicker
• Effects: Delay, distortion, filters, LFOs, reverb
• Sampler with WAV support
• WAV export when you’re done

http://warmplace.ru/soft/sunvox/

And for fans of computer music in the 90s, it’s a chance to get back to some of the no-nonsense, powerful creation of that era, without some of the distractions you may find in modern apps.

To see SunVox in action (and start to make sense of how to use it), check out the YouTube videos from Alex Zolotov (among others). They’re not the best video quality (I wish we had iPhone screencasting software, darnit), but they will give you a sense of what the app is about:
http://www.youtube.com/user/NightRadio2007

The videos made their debut on the excellent PalmSounds Google Group:
http://groups.google.com/group/palmsounds?hl=en
CDM isn’t exclusively about mobile apps, so check out PalmSounds the blog, too, if you like your music making handheld – the site has been on fire lately with updates.
http://the-palm-sound.blogspot.com/

Here’s an example of how SunVox can be used for live playing:

Here it is being used to produce lovely FM synthesis noises:

If you’re new to trackers, especially, you’ll want to brush up on the tracker workflow in SunVox:

In some of the news I’ve missed in the last couple of days are some unusual announcements. Urs Heckmann can be fairly considered one of the great soft synth designers, with accomplishments like Zebra. His latest, Bazille, like many recent soft synths, is a hybrid: FM synthesis plus phase distortion plus the obligatory subtractive synthesis. In an early teaser video (he apologizes for audio quality), he shows off its modular design. Now, modular routing is something we’ve seen in some form in other recent synths, from Maschine to Future Audio’s Circle. But for Bazille, the layout of the whole synth is clearly set up with rack-style modular routing and free-form patching in mind. There’s definitely some promise here. Oliver Chesler of the utterly brilliant wire to the ear found this first and has some other good thoughts.

The surprise news, though, is that Fairlight may be re-releasing the Fairlight CMI, their original digital sampler. The Fairlight Instruments site teases a “CMI Series 30A (Thirtieth Anniversary) Limited Edition.” Peter Vogel’s CMI, ubiquitous sound of the 80s, established many things we take for granted in computer music. Heck, it even had a light pen. So, too, will the 30A re-release. They’ll make 100 of them, you’ll get WAV import and improved sound quality, and… no, you won’t be able to afford it, though Vogel says it’ll be cheaper than the original. (In other words, it’ll be cheaper to get a new Fairlight than a new Buchla.)

Sonic State scoops the details from the man himself:
More Anniversary Fairlight Details: A little more information from Mr Vogel

Of course, I dream of a successor to the Fairlight CVI, their ground-breaking video instrument.

Alternatively… Synclavier: The Next Generation, anyone?

You probably think of social networking and messaging as being about text, about saying things like “Wow, this tuna salad sandwich I’m having for lunch is delicious!” But the next Tweet you get on Twitter could be a synthesis preset.

Say what?

Working in Max/MSP, Andrew Spitz has developed a tool called Tweet a Sound. It uses Twitter as a communications platform for “social sound design.” Instead of just saying, “Wow, I be makin’ phat basslines,” you can actually share the sound. Whip up a sound using typical FM synth parameters and Max/MSP’s sound engine, then click “send.” You’ll send a string of numbers to your Twitter account, confusing those friends not in the know. But other users will be able to grab and play with your sound.

Andrew even encourages synthesis n00bs to play without fear – grab those envelopes and mysterious-looking settings and see what comes out. So, I hope you synth geeks do share this with some friends new to synthesis, as I think they’ll have a great time.

Right now, Tweet a Sound is Mac-only; we just need someone to save a Windows standalone version. Someone has asked about a Pd port, but let’s put it this way: this is the tip of a very, very big iceberg of sharing. It’s something worth considering in anything you’re doing, not just with Twitter, but whether you can provide networked capabilities in whatever you’re happening to build.

Ableton, of course, recently added the Share functionality to Live. But with open APIs and basic networking protocols, there’s no reason you can’t explore other features. Why not build a drum machine that lets you collaborate with one of your friends on your IM list, or a sequencer that automatically posts ideas as you revise them? Just doing these things for the sake of it could be a waste of time, but on the other hand, these social features could turn Web 2.0 sites into places that actually inspire you to make and share music rather than distract you with mundane activities.

I love the idea; let us know if you have some fun with it.

Tweet A Sound { sound + software } [Andrew Spitz Blog]