identification please

Neural networks and their failure are inherently funny. I love and have, in the past, named some tracks after its generated brands of breakfast cereal, dubious recipes, Dungeons and Dragons spells, names of colors, etc. (Not so much anymore since I tend to run with album themes.)

ImageNet Roulette is a new source of laughs for me. Using the helpful GetThatPic! extension for Chrome that reveals the URLs of Instagram photos, I had it try to classify some of my recently taken photos.

Yankee, our loaner dog

INR says: darling, favorite, favourite, pet, dearie, deary, duckya special loved one

He’s cute, but no, Gretta is the favorite.

An actual ducky.

INR says: circus acrobatan acrobat who performs acrobatic feats in a circus

Gretta, after being outside in the rain.

INR says: sleeper, slumberera rester who is sleeping

Lady, our #2 ranked dog.

INR says: skin-diver, aquanautan underwater swimmer equipped with a face mask and foot fins and either a snorkel or an air cylinder

A Buchla 200e system with a Thunder controller, as seen at Knobcon.

What INR says: “trawlera fisherman who use a trawl net

To be fair, it did lure people in.

St. Louis Osuwa Taiko at the 2019 Japanese Festival at Missouri Botanical Gardens.

INR says: peddler, pedlar, packman, hawker, pitchmansomeone who travels about selling his wares (as on the streets or at carnivals)

We did buy t-shirts.

no place ̶l̶i̶k̶e̶ ̶h̶o̶m̶e̶

A couple of weeks ago I recorded a piece I called “the between.” It was between album projects, and between setups. There was interaction between elements of the patch. And I know it’s sort of a cliche, but I was thinking of liminality: a not-place, not-time between other places and times.

My first recording on the new rig was titled “No Place” — thinking of a specific not-place I visited in spiritual/imaginary travels. A sort of stopover on the way elsewhere. I can’t tell its story in words, but I can paint a picture in music. I’m sure what other people see when they hear this song is different, and I like that idea. It’s communion rather than communication.

Oh hey. Album theme. Scenes that aren’t places. Okay.

The third song refers to a story I wrote for a school assignment ages ago. I remember nothing about it other than the setting, described tersely in the first sentence. Even in the context of the story, it was arguably not a real place with a fixed location.

The Expert Sleepers ES-3/ES-6 pair is as enabling as I’d hoped. Stereo feedback loops that run through plugins and back into hardware? Check. EQ and compression anywhere in the chain I want it? Check. Using Bitwig’s phase objects for unusual clocks, and switching between quantization scales with a fader? Check and check. So good.


Last night I recorded my first “real” song with the new setup, and it went mostly smoothly. In fact… my goals during the mastering stage are -1db true peak and -14dB integrated LUFS, and look at where this one landed:

I think this is mostly coincidence, but it’s a good sign. Usually I add several stages of compression with manually drawn curves, followed by maybe Presswerk and a few rounds of adjustment. This time it was just Bitwig’s peak limiter when recording, and a very light touch of it again when rendering.

I could streamline away the extra step of recording and then rendering, but it’s an opportunity to do some of my processing and automation right there in Bitwig before moving to Sound Forge.

Sound Forge Pro 13 only lets you scan 3 locations for plugins — and that’s supposed to cover VST3, VST2 64- and 32-bit, I guess? But VST installation is not well standardized and there are a few important bits not in line, like NI Transient Master. I’m going to have to shuffle some stuff.

This was also my first recording with Mimeophon and Via Scanner. Mimeophon is an excellent delay, but as a sound source in its own right it’s amazing too. The “flip” input, which toggles the direction of delay playback, can be modulated at audio rates for some incredible textures.

Scanner is complex and unconventional, but once you grasp a couple of important things, it’s not hard to use as an audio waveshaper. (Its affect on CV signals is going to take me some time to work out with a scope.) It can sound like a traditional wavefolder, Plaits’ waveshaping model, or more like a wavetable oscillator. It can also get glitchy and weird when you start self-patching its logic outputs.

The thing about wavefolders is that their tone depends on the gain of the input — the stronger the signal, the more it folds over on itself and gains extra harmonics. So they tend to have a big gain boost at the input, controlled by a VCA. Scanner has two output VCAs (bipolar, in parallel and mixed) to allow for some other tricks, but no voltage control over the input… so it calls for a separate VCA. But I don’t need Origami anymore, and Cold Mac fits into that space… it can be my input VCA or a crossfader or compressor or a lot of other things.

And with that, I have my “final” setup for Synth Farm 3.1.


Back home from Knobcon, and back into the real world where I have to work and stuff.

On Sunday morning I dipped back into the exhibition hall and bought the Starling Via Scanner. There’s a lot of potential for reshaping both sound and modulation signals and I look forward to discovering what it can do. Then I mostly waited for Dr. John Chowning’s talk.

Dr. Chowning is 85, and his speaking voice sounds remarkably like Johnny Cash’s “Hurt” cover. Between the keynote speech at the banquet and Sunday’s talk, he described the discovery (not invention, he convinced us) of FM synthesis.

Today it seems as if “discovering” FM would simply be a matter of plugging an oscillator into another oscillator’s pitch input. But at the time, voltage controlled analog synthesizers were new and rare, and Dr. Chowning was working in the digital domain. Not in real time, but with punch cards, no buffers to speak of because memory was shockingly expensive, no digital filters, and even a DAC was an exotic piece of custom equipment.

He was a drummer who read a paper by Max Matthews about the potential of computers to make music, and decided right then that he wanted to make music “from the inside out.” He learned to code, and picked up the engineering and math mostly by osmosis and asking a lot of questions.

He was investigating the perception of sound in space — how we know the difference between a quiet close sound and a loud distant sound, how we identify individual instruments and voices in a group — and started experimenting with vibrato. He found that extreme vibrato rates shifted from a perception of pitch variation to a timbral change instead, and explored how changing only the intensity and relative frequency led to very different spectral results. FM was already in use for radios, at megahertz carrier frequencies rather than audible ones, and he found that some of the mathematical laws applied just the same, with very different implications.

In his talk he frequently showed spectral graphs of the effects of FM. FM creates sidebands around the original frequency, with particular spacing and phase characteristics, which can cancel or enhance each other when they reflect around 0Hz or the Nyquist frequency. (Band limiting in FM is just a matter of reducing the FM index as frequencies increase — something that Yamaha loved about it, because digital filters were really poor at the time.) FM can also be used in formant synthesis, which isn’t a common technique for some reason, and which definitely involves thinking in the frequency domain rather than about waveforms or vague “shaping”. It occurred to me that I almost never use spectral analysis unless working with EQ or a filter, and it’s a tool that could tell me a lot about waveshaping and FM. That was my personal moment of insight from the talk.

The concerns of FM in the purely digital domain are very different from modular. It is, by default, linear TZFM — or perhaps phase modulation as Yamaha implemented it in the famous DX7. Dr. Chowning pointed out that he didn’t work on the DX7 at all, but it was the effort of Yamaha engineers overcoming the technical challenges of the day and pushing the frontiers of digital audio.

Since I couldn’t get them at Knobcon, this morning I’ve ordered a replacement for my missing power cable as well as an Expert Sleepers ES-3/ES-6 pair. From there I’ll figure out if any other utilities would help or if they can be handed off to Bitwig.

There’s a phase of experimentation and learning to go through with this new setup, but I am thinking about the next album. Running through my head are some of the ideas presented in Music Beyond Airports: Appraising Ambient Music and the realization that my music is often about scenes. Places real, imagined, dreamed or remembered, their weather and activity at that particular moment, and the feelings associated with it. Sometimes the music creates the scenes, but sometimes I draw them from experiences. I have a few such experiences I’ve never made music for, and others that I could reinterpret. As an album theme goes, this is a little bit loose — it’s just doing what I do anyway — but the conscious awareness and intent might be enough of a guiding factor.

report from Knobtoberfest

I’m taking some time to chill out from Knobcon and let my brain catch up and feet rest (I should have worn the walking boots which are much better for standing around for long periods than my Chucks).

I’ve written this up partially on my Instagram and partially on MW, but in a nutshell:

The goody bag was middling. Stickers and postcards and flyers; a hefty Sweetwater catalog which I dumped because it’s heavy and pointless; a bag of middle-tier candy; an issue of Sound On Sound; a purple frisbee; a 2hp blank panel; a Velcro cable tie.

This year’s theme is Knobtoberfest, which was reflected in the performances at the reception being German themed “kosmiche musik” which was not only highly influential to both popular and esoteric forms of electronic music, but to pop as well. A tribute to NEU! (which I sometimes enjoyed and sometimes felt I was too sober to get it), a tribute to Kraftwerk (which was highly enjoyable), and a Berlin School duo.

A couple of years ago when I was newer, I felt like I absorbed a lot about playing modular synths by watching the performances. This year not so much, but I mostly learned that rough spots — technical issues, missed notes and cues and changes, brain farts — don’t really matter too much if you handle them well. The audience — at least an audience of musicians — accepts them as an inevitable part of being human, and gets on with enjoying the good parts.

But I also learned a little extra preparation in terms of physical setup is important. You need a big enough table, solid places so your gear doesn’t threaten to fall off and cables aren’t in your way, and if you need written notes they need to have their own space to live in instead of getting tucked under and resting on top of your instruments.

Also, sudden transitions rarely work, especially if you’re going to switch tempo. It’s like trying to change gears on a stick shift car without the clutch. An audience that’s grooving along will suddenly be thrown off.

Also, sample players are really good things to have for live playing, for many reasons.

Also if you’re a hotel sound guy, make sure the PA actually works before you tell the performers they can start whenever. You had one job…

Today was gear stuff. Workshops too, but I didn’t go to any — maybe tomorrow. The exhibition hall isn’t arena-sized but it’s still pretty large, and includes bigger companies like Hammond, Yamaha, Korg, Roland, Moog etc.; lots and lots of Eurorack manufacturers (but missing a few key ones), a few software companies, a couple of local synth-oriented music stores, and some miscellaneous weird shit. I actually did buy some Real Fake Knobs. Unfortunately our small dog guest is not ferrous, so the volume knob won’t work on him.

I won’t go through the full list of all the gear I tried. Nothing really grabbed me the way Natural Gate did a couple of years ago, but I did play some very fun things in the “I don’t need this but I kind of love it” category. The Hydrasynth has the most beautifully designed interface of any electronic instrument I have tried, in both look and feel. The Moog Grandmother was a joy to play and sounded full and gorgeous, but its price and size make it unlikely for me. I had a blast (and also created a powerful sonic blast) with the Curious Sound Objects Bitty toys.

Synth Tech’s prototype of their new effects module is 10-day-old hardware running code that was rushed in a week, so it’s got some rough edges (!) but I think it holds promise. Once the biggest bugs are worked out and the algorithms have had a couple of rounds of polish I think it will be a winner. I hope to be a beta tester for it as I was on the E352 and E370.

Make Noise Mimeophon was mimeo-fun, despite a less than stellar demo setup (note to exhibitors: we want headphones, especially to demo stereo gear!) and I went ahead and bought one instead of ordering online.

The JAMMspace room was a bust this year. Two years ago, it was a quiet, uncrowded space where every piece of gear or small modular setup had headphones, so you could really concentrate on trying the gear. This year it was set up late, most things ran through a big mixer into a loud chaotic mess through a loud PA system, and they left the outside doors open where people were taking smoke breaks (of course the smoke went directly into the room).

The Cruise-In wasn’t too successful either. Apparently nobody bought any of the gear on display. I recognized a lot of it as stuff that’s been for sale on MW. I didn’t bother to haul my 3 modules down there to try, because there’s not really a good way for potential buyers to demo them anyway. It was a nice idea, though.

The banquet and keynote speech are coming up in roughly an hour. The keynote will, I think, help me decide whether to go to Dr. Chowning’s workshop tomorrow afternoon or head home early. There are also more live performances after dinner, and I don’t know yet how much energy I’ll have for those.

I think I will likely skip Knobcon next year. Overall, I feel like the second time wasn’t as valuable a learning experience as the first. Like I thought last time, if I were more of a social creature I’d probably get more out of it. So for me it’s mostly a matter of indulging curiosity and doing something a little out of the ordinary — a mini vacation for electronic music nerds.


The reinstallation of plugins is complete. While I’ve cut down on the number of plugins considerably, they’re all set up on my categorized quick lists. I will probably weed the lists eventually — leaving access to the less commonly used plugins in the capable hands of Bitwig’s browser/search — but I wanted to put everything there for now so I can:

  • get to know all the Bitwig native plugins well.
  • revisit old 32-bit stuff that I haven’t been able to run in Maschine for a while, including my own experimental plugins
  • be aware of what is and what isn’t installed on my hard drive.

Also, deeper integration between modular and DAW, and Bitwig’s routing and containers, are going to affect my plugin usage.

A lot of stuff didn’t get reinstalled. My Native Instruments installer shows 71 items not installed, and a mere 5 installed. My software archive folder has 41 items, but my “no reinstall” archive has 147. Some of those were sample libraries, non-music utilities or other things, but some “items” are actually multiple plugins so an accurate count is going to take some doing.

Some of that material is free or not-for-resale, but a chunk of it is not. A major post-Knobcon project is going to be identifying what I can resell, the license transfer fee if any, and market prices.

Speaking of… tomorrow morning I head to Chicagoland. Not much going on the first day of Knobcon, but I do want to beat rush hour traffic. There’s registration and the reception — a live show featuring a half-dozen or so synth performers, which goes on for more hours than my endurance allowed last time. This time I will eat dinner first. Neither rum nor beer count as “dinner.”

Saturday is mostly going to be about the exhibit halls for me; there aren’t any workshops I’m invested in. In the evening there’s the banquet and keynote speech, followed by a choice of Big Room or Chillout Room for live music until whenever. It’s also highly likely I will ditch the crowd and find some quiet time at a few points.

Sunday the exhibit halls reopen, and Dr. John Chowning has a talk on FM at 1:30 I might attend (I can’t tell how “beginner” it might be or if it’ll be fairly comprehensive and give useful insights). But I want to get back home at a reasonable time, so I wouldn’t stay beyond that.

fresh step

maybe “fresh start” would have been a better phrase.

As nice as it is to have a faster, shiny (in this case literally) new computer, the first few hours are often dominated by frustration and annoyance. Stuff works differently, or doesn’t work yet or is subtly wrong because you’re still in the process of installing drivers, fixing the STUPID defaults, and installing the everyday apps and tools you took for granted. It’s like fighting a battle of identity against Microsoft. This is MY computer now, not theirs, and I should get to decide how things work.

Eventually most of the annoyances are overcome as well as they can be, and a sense of relief sets in. At least we have some control over our computers, unlike so many other things in life.

That’s where I am now, though I still need to install about half the plugins I want to keep using, plus Paint Shop Pro and Sound Forge.

It’s too early yet to say how much I’ve gained from this change, since I haven’t made serious recording efforts — but I set my ASIO buffer to 128ms with no ill effects. I’ll happily take 128 and smooth over 512 and occasionally glitchy, but I’ll try cutting it even smaller and see just what this thing can handle.


It looks like I may have chosen the right time to step away from Maschine. Credible-ish rumors and Glassdoor posts are saying that Native Instruments has been gradually letting people go for a few months and just a couple of days ago, laid off 100 people (about 50%) from its Berlin office, including the “product owners” of Maschine and Traktor and “almost all hardware teams.” Former employees wrote about management incompetence and lack of direction. Behringer has apparently offered to hire some people.

There’s also a claim that a new standalone Maschine unit was killed off one day before it was scheduled to go into production. That sounds a bit less credible to me, but who knows?

From the outside looking in, it’s felt like NI lost some of their previous vitality, in recent years. To some extent I thought that was just my perspective changing, but there’s more to it than that. Massive X was released in a not-quite-finished state and was a letdown to many people, Maschine Mikro mk3 has some weird omissions in features, and overall their focus has seemed… just not on.

There’s no official confirmation or denial of anything, but it does sound like there’s still a team working on Maschine. For all anyone outside of NI knows it’s a skeleton crew at this point though? And it’s not as if the pace of updates was rapid or the direction was in things that people were asking for.

Update: here’s some official word.

Their emphasis on “consuming and accessing creative goods and services” makes me think their primary goal is going to be selling loops and samples, and the means to “consume” them (Traktor, Maschine, Kontakt etc.). A lot of people think it could be a subscription service. Whatever. It signals to me they don’t want to be instrument builders anymore. My method of making music isn’t to “consume and access” other peoples’ sounds.

wiggling the wiggy bits

I went through a highly recommended video course for Bitwig beginners, by Thavius Beck. It walks you through creating a… frankly pretty awful tune, but it’s a good demonstration of navigating around the software and using many of its features, and by the time it wrapped up and I slapped together another couple of quick, bad songs it felt pretty comfortable to use.

I really like Bitwig’s flexibility. For audio/effects routing, there are usually multiple ways to achieve any particular goal, and you can pick the one that makes the most sense at the time. For recording and sequencing, there’s a linear arranger as well as a clip-based, loop-oriented one, and you can transfer material between them or record from one to the other. The interface itself has multiple different layout options and it’s easy to switch them as you switch tasks. Several different hardware devices can be used for controlling it (though sadly not the Maschine mk2). And I appreciate that it can run brand new VST3 plugins right alongside 32-bit VST plugins from 2003 with no fuss whatsoever.

I’m confident this will work better for me than Maschine, though I’m holding onto the hardware for a while to make sure. I’m not as confident it can also replace Sound Forge entirely, but I’ll probably do at least some of the post-processing work in Bitwig.

This weekend I briefly poked around the Grid device — modular building blocks to create instruments or effects. Rather easily, I put together exactly the kind of additive oscillator, controlled by the 16n Faderbank, that I had used on the ER-301 on Passing Through. And then I phase modulated it with the Hertz Donut for some fantastic growling tones. That was about 90% of my motive for wanting the SMR, SWN or Just Friends.

Grid also has a whole set of phase-related modules, including a sine lookup module… so I could implement a lot of the stuff I wrote about in “Sine Shaping and You” as well as phase distortion, etc. And they all apply to audio signals too; like Eurorack, Bitwig Grid doesn’t actually care whether any given signal was supposed to be audio, a pitch CV, a gate or whatever.

So the new way of things for me is going to be to keep a core of awesome modules in Eurorack, and throw open the borders between hardware and software. One of Expert Sleepers’ ADAT interfaces will be perfect for this, giving me plenty of inputs and outputs for CV and audio at Eurorack levels with minimal latency.

That gives me a new focus for checking things out at KnobCon, because a lot of situational or utility modules could be handled by Bitwig instead.

Synthesis Technology has “pre-announced” the E520 Hyperion Stereo Effects Processor, about which Paul Schreiber says “we are focusing on FFT/spectral/frequency domain transforms that have not been done 600 times in the past.” Frequency domain effects are nearly nonexistent in Eurorack and other hardware, and rare enough in VST plugins. Panharmonium was a bit of a bust for me, but this emphasis on effects processing rather than resynthesis is very promising.

“The Grid. A digital frontier.”

No, this isn’t a follow-up about TRON Legacy or its soundtrack (which was decent, but not as fascinatingly weird as 1982 TRON… the visuals were way better though). But rather, Bitwig Studio 3.

I know they look similar, but don’t be confused!

I’ve been using Maschine as my DAW ever since 2.0 was shiny and new. It began as a drum sampler/groovebox similar to Akai’s MPC, but it gained full MIDI sequencing and VST plugin support, and gradually expanded into something more full-fledged. It was an easy switch from FL Studio back in the day — cleaner and more structured, where FLS had 13 years of features bolted on and very little consistency to its UI. But its “beat making” nature is a really strange fit for the kind of music that I make now.

MCP, MPC, whatever. You know, that villain looked better in the video game than the movie, and I’m glad they didn’t bring it back for TRON Legacy. End of line.

I’ve stuck with Maschine only because I was used to it, and learning a new DAW with a completely different paradigm takes some effort. I’ve given Ableton Live and Reaper demo versions very brief attempts, as well as a newer version of FL Studio, but each time I just felt more frustrated and returned to the path of least resistance.

But I’m between albums and have a new computer on the way (*), so this is a good time. Plugins are starting to be released in VST3 format only — which Maschine doesn’t support yet — and I’ve heard a lot of praise for Bitwig’s modernity and modular orientation.

So I’ve been looking into it. I have to say the experience of learning it could have been made a lot smoother with a single official, officially endorsed, or built-in tutorial — rather than the bewildering array of unorganized, free and paid tutorial series that both the official site and a web search throw at you. It’s a bit intimidating. But I’ve found a couple of recommendations, and after a few hours I feel like I can record with it, start to finish, without getting too lost.

For what it’s worth, Thomas Foster’s “Bitwig Studio 3 Tutorial for Beginners” and Brian Bollman’s “Migrating to Bitwig Studio” series have been helpful so far.

There are deeper things to learn, such as the aforementioned Grid (a modular effect/synth builder) and various shortcuts and customizations, as well as optional hardware controller integration.

There’s also the possibility that I could use Bitwig for post-processing and mastering, without the need for Sound Forge or something else. We’ll see.

(*) I was notified yesterday that the computer was in production. I don’t expect it takes more than a couple of days for experienced builders to assemble, install Windows and drivers, and do a burn-in test, even with a customized parts list.