rolling with it

Super Tuesday (“more like Stupid Tuesday”) and the next few days were kind of a rollercoaster. Right now, I am in a place of resigned… something. Not quite acceptance, but panic and anger doesn’t really help either.

I will not be thrilled to vote for Biden if that’s who wins the nomination, but I’ll do it. I’ll be a bit mollified if Warren is his VP or he makes some overtures to the left, but I don’t actually expect either of those things.

I’m not very keen on Biden’s chances of beating trump, but I’m also not very keen on Bernie’s chances to beat Biden. I think we’re going to be playing the same bad game of Rock-Paper-Scissors as 2016: Bernie beats Trump, center-right Democrat beats Bernie, Trump beats center-right Democrat.

Logic I should dislike Biden at least as much as I did Hillary Clinton, but I don’t, quite. Maybe it’s because I haven’t really been watching debates, and didn’t see him looking smug while explaining as if to a child that “America will never have universal healthcare.” Maybe it’s because in 2016, I didn’t think Trump was actually going to get elected and that a protest vote sent a stronger message than anything else I could do, while in 2020 I’m thoroughly sick of Trump and feel just as betrayed by the Green Party as the Democrats. I just sort of feel tired about it.

I don’t like choosing a lesser evil when there’s another choice. This time I don’t feel like there is another choice.

But I’m not going to argue with people about it and keep myself tied up in knots for the next 8 months. I’m gonna put in my vote for Bernie on Tuesday and a Democrat in November and we’ll see what happens.

The castle album continues to move along, with an ash-colored cranium and a granular mostly-silicate material joining the first two. The stuff going on so far really doesn’t sound like either typical FM or typical subtractive synthesis (*), and it’s definitely not all dense chordal Akemie’s Castle drones. But it does sort of have a sound to it, a recognizable evolution of Float.

There kind of hasn’t been much subtractive synthesis, yet. Some filter FM combined with a manual range sweep. A couple of minor voices that used filter envelopes as part of the sound. But I don’t feel like I’ve quite engaged with it yet…


I went for that Stages trade, and it should be waiting for me when I get home today. I also rearranged modules one more time, in a layout that is my best compromise between ergononics, balancing the power load between the upper and lower sections, a few signal flow considerations, and not having to move things much in my planned changes.

The second Stages will go beside the first one. E520 will take over the right side of the second row from the bottom, where Supercell and the gap are. Supercell will get sold and replaced with a Typhoon that goes in the gap on the left. That leaves just 5HP empty.

I have no plans beyond that, but I acknowledge there will probably be some must-have module in the future that requires selling off one or two lower-priority modules to make some room. Contour might be one of them; the way I use and approach envelopes has changed, and my “favorite” envelope module isn’t getting as much use. Stages can do everything I like about Contour and far beyond, though with a little more cognitive load.

As I wrote on the subject elsewhere, I’m not really using a lot of envelopes overall, anymore. I used to feel like I needed at least two per voice — one for the output VCA/LPG, one for timbral modulation — but that was when I sequenced everything in MIDI. Contour’s ARSR envelope complimented LPGs with a lowish sustain level, varying gate lengths in a sequence and legato playing and note jumps, which was very much a thing I did a lot of before I got more into the “Starthief sound.” But now, most of my modular voices have their levels manually controlled, with level parameters in Bitwig serving as VCA/mixer. For my sequenced/algorithmic voices I have tended toward simple ASR, AR or R envelopes from Maths or Stages, or even generating R envelopes directly in Teletype, or just triggering R envelopes in Natural Gate, or triggering individual grains in Clouds, or purely gating a source but letting a resonator/delay/reverb provide the envelope naturally…

Stages’ envelopes can get fun, though. You’re not limited to classic envelope types, and you can loop it and hybridize it with a step sequence or other modulation sources. Why ADSR when you can A(LFO)(A/D)(S&H)(A/D)R?

I’ve decided to run with the “castle” theme for the next album, and name each song after a different fictional or mythical castle. Most likely, a few tracks that don’t feature Akemie’s Castle will get titles without “castle” in them, e.g. “Minas Tirith” or “Valaskjálf” or whatever.

Only two recordings done for it so far, but I also submitted two for the next Ambient Online compilation, did two “Feedback February” recordings (one again for a compilation), a couple other recordings I decided weren’t going anywhere, and some experimentation. It’s funny how that feels slow to me now, because sometimes I spend as much as a couple of days without recording anything.

Kraftwerk is touring this summer, on their 50th anniversary. (Hey, I’m almost 50 myself. That felt weird to write.) They’re actually playing in St. Louis, so of course I’ll go.

Kraftwerk has pretty much been a Kraftwerk cover/tribute band for the last several years. They’re down to one original member (after a falling-out and creative meltdown) and three touring musicians, playing on laptops and iPads. They don’t do analog synths anymore. But I still expect to enjoy the show, as I enjoyed the Kraftwerk covers that a dude played at Knobcon last year, only with more technical wizardry and light show stuff.

Already I feel inclined to listen to Kraftwerk stuff for the rest of my workday…

if FM overkill is wrong, I don’t want to be right

In the last few weeks I’d been thinking seriously about selling off Akemie’s Castle, having convinced myself that I have more than enough FM capability and can use Sync3 among other things for the big chords. So I was trying to decide between the Odessa or the Ensemble Oscillator as a replacement, but honestly I wasn’t totally fired up about either of those choices. They can both do cool things but I’m not fully convinced either of them is truly what I want.

Last night I was feeling pretty bleh and unmotivated, but after making a pretty crappy recording I decided to mess with Akemie’s Castle… and it brought immediate joy. The chords are huge, the FM is glorious and not quite like all my other FM synths, and it sort of does the additive thing from a different angle. I have a bond with this module, and it would make no sense at all to swap it out for something else.

With that settled, I decided there’s really no point in waiting for the E520 to sell stuff: the DSM-03 Feedback, tanh[3] and Cold Mac have been sitting in a drawer and I’ve left HD mk3 unused, and I don’t think I will miss any of those. Also the Supercell, which I’ll replace with Typhoon (unless I happen into a great trade deal for Monsoon or Microcell).

What makes much more sense is to not buy anything else, including pedals, for a while. I don’t have any particular gaps I can think of. E520 of course is going to cover a lot of effect territory. I don’t see a good argument for another oscillator, and I feel pretty happy with my set of modulation sources. Bitwig Grid and Teletype cover most things that aren’t directly present. I’m sure eventually there will be something I will want to try out, so I’ll just leave the empty space for that future eventuality.

I’ve been reading Greg Egan’s Diaspora. It starts off in a wild and weird post-human future, and only gets wilder and weirder from there. My only gripe is that while the physics might have more dimensions than we can imagine, the characters mostly tend to have one or two. It seems like the more interesting a character becomes, the more likely they will be ignored, fall into the background, or be killed off in the background in the next chapter. I’m still enjoying the scale and uniqueness of the ideas though, and the rate at which they are introduced.

thanks I hate it

What the hecking heck is this?

It’s the Roland TAIKO-1 and it is about 10% brilliant and 90% bullshit.

I was a taiko performer for a couple of years, so let me first mention the things I like about this before I tear it to shreds:

  • It looks kinda cool (if you’re unconcerned with tradition)
  • It’s light-ish and it disassembles for transport.
  • You can use it for practice with headphones, IF you want to practice katsugi okedo.
  • Built-in ji is better than a metronome… if you want to be backed up by a robot playing a horse beat, you can.
  • The shime-daiko sample set sounds okayish.
  • Could be neat for one or two performers during an interlude, or for parades or Eisa performances I guess, or for a single novelty piece by a taiko group. IF the volume and battery situation is up to the task.

And now the hatey bits:

  • This is a katsugi okedo in style. It is useless if you want to practice or perform chu-daiko, shime-daiko, or oodaiko parts or technique.
  • Each type of drum, type of stand, and style of playing requires different physical technique. Each taiko piece has its own choreography which is a vital part of the art form (some even say it’s more important than rhythm and sound). This is one drum played in one style, which happens to sound sort of like other drums. The way this is designed, I’m not even sure it can be placed on any kind of stand.
  • Would you like to watch a symphony orchestra where everyone is playing electric violins, except some of them sound more or less like cellos or oboes or French horns or timpani?
  • Would you rather go to a Japanese festival and watch a bunch of people in traditional costume playing a variety of beautiful, full-sized polished wooden drums, hand-built in the traditional manner, with fancy metal kan (the medallion handles) and proper horsehide heads tacked on in that distinctive pattern, where one of the drums is a 250-800 pound monstrosity that looms over its performer… or a bunch of electronic instruments with lightweight metal frames and fabric heads?
  • The chu-daiko sound in the video sounds synthetic to me. The shime-daiko is okay. The chappa samples are a disgrace, and there is no way all the varied techniques of chappa playing can be done with a drum.
  • Real taiko is like thunder. It’s extremely physical. You feel it throughout your body, especially in your chest cavity. In person or in a really good recording, you hear the crack of the hinoki (cypress, awesomely fragrant) bachi on the drum head as well as the deep boom. When in group rehearsals you’re supposed to wear hearing protection. I’m going to guess the sound of these is a relative disappointment in terms of physical presence.
  • Likewise, the feel can’t possibly be right for anything that’s not a katsugi part. There is a very physical feel to the movement and the way the bachi rebounds from the skin, which is different for different styles of drums and bachi, and it deeply affects the player’s expressive playing as well as dynamics and timing. It may even feel wrong compared to a real katsugi as far as that goes.

Now, I’m more of an electronic music nerd than I am a former taiko player, so I still think it’s kind of got neat aspects. And if someone gave me one, I would play it. But I don’t want to see a taiko group performing with just these things, ugh ugh ugh gah no.

an individual note

Lately there have been some threads on various forums that have tried to ascribe a little more to the design of particular modules, and the intent and motivation of people in the community, than is actually there.

As in: [these modules] are for beginners. [These modules] sound like plugins.* [These modules] are overhyped by uninformed fanboys. [These modules] are for generative ambient noodling, while [these other modules] are for techno. The real nature of modular synthesis is [such and such] and [these modules] aren’t good for that.

Or more generally, disappointment in some perceived lack in the tools: too little innovation in the technology, too little creativity in the products, no future ahead for novel synthesis methods. As if it’s the instruments, not the musicians, which are expected to be creative…

Well. I’ve been reading Daphne Oram’s An Individual Note: of Music, Sound and Electronics. She floats some esoteric ideas and metaphors extrapolated from the science behind electronic music… some of it is frankly a little crazy, but some of that is illuminating in a poetic sort of way.

What really struck me though, particularly in the context of those other discussions, was this section:

It seems as relevant to the tools of music-making as it is to a performance or recording. Perhaps even more so; music is the product (in a vague mathematical sense) of the instrument AND the musician.

And yeah, I’m saying if a module sounds bad, maybe it’s the person using it who is not good with it, either through incompatible style or lack of understanding or practice. And if it seems like it’s “for beginners,” maybe the person trying it lacks the experience to get more out of it.

(*) This of course implies that plugins have a characteristic “sound” which is deficient in some way, which is also a pretty questionable assumption in 2020.


The Strymon AA.1 arrived yesterday, after an inexplicable FedEx delay. And:

  • Yeah those cables I bought to work with it were too short and super awkward. Letting the pedal lie prone (like they normally do) atop my modular case (like they normally don’t), it barely reached.
  • It doesn’t actually attenuate hot Eurorack-level signals enough to work well with the Particle, even at Particle’s max input level tolerance setting, without another stage of attenuation. Thus, the more important half of the pedal is basically useless.
  • Stereo on the Particle is just a matter of applying the exact same effect to L and R channels individually. So all the random grain density, detuning, reverse etc. is synchronized and thus not nearly as interesting as it could be. (To be fair, Clouds is mostly like that too, except the panning of grains can be randomized or modulated.)
  • As I fought with it, I realized I don’t really love Particle all that much anyway. I switched the patch I was testing over to Clouds instead, and found I could get into a somewhat similar ballpark — and it was more fun to tweak and in some ways sounded better.

So I’m returning the AA.1 (setting a new record for shortest time any module has spent in my rack), and reselling Particle. The pedal is definitely exempt from my “don’t sell stuff yet” rule, and I think a return of something that literally does not do the thing it was made for is a fair play too.

In my rearranging of stuff I wound up pulling out Cold Mac and setting it aside. Sort of a test to see how I feel about selling it later. I can set up the crosspanning stuff and other utilities in Bitwig Grid without too much trouble. When the FM Aid arrives this evening, I expect I will do the same with the DSM03.

(* another nope: effing UPS does not follow the delivery instructions you set for them, so they did not “Leave at Front Door.” I specifically had to phone their customer service about that a few months ago to change it to that, because literally everything they delivered even if it was worth $3 was set to require a signature. But at least this time they’re not trying to charge me $5 to pick it up from one of their locations tomorrow.)

(For the record: I hate dealing with USPS, FedEx, UPS and DHL. Somebody invent that teleporter…)


I guess I didn’t mention that I had found a discounted Red Panda Particle V2 that was used as a store demo. It arrived a couple of days ago and I put it through its paces.

I’ve heard Particle described as “standalone Clouds” but I don’t think it’s an entirely accurate comparison. Clouds strikes me as a genuinely granular processor, capturing overlapping sound snippets either as an effect or as a means to synthesize new sounds. While Clouds can be choppy and rhythmic if you choose short enough grain size and sparse grain triggering, it can also do smooth (or overdo it into blurry and drowning in its own reverb), or places in between that yield interesting textures. In its original firmware algorithm, Clouds can act as a delay but that doesn’t seem to be its forte or its primary design — although now I find I’m going to want to revisit it and see how well it can pretend to be a delay without using the alternate delay algorithm.

Particle strikes me more as a delay first, with a granular twist. You can get a fairly clean-sounding delay out of it that doesn’t have any granular modulation. From there, you can lower the grain density (by randomly dropping some grains) or reverse grains (with adjustable probability of being reversed) to chop it up a bit, repitch it, randomize the delay amount or pitch for each grain, or traverse the delay buffer with an LFO which sort of “bends” the playback in an indescribable and unintuitive but sometimes interesting way. It can also freeze the buffer, either while one of the switches is held down or when the level falls below a threshold that you set.

Not all of the features are available simultaneously from the pedal’s knobs. There are hidden settings accessible only via MIDI over USB; otherwise the Mode knob determines what the Param and Delay/Pitch knobs will do, and Chop and the Freeze threshold are combined onto one knob. I don’t much like that, to be honest, although I understand that’s all the original version of the pedal allowed. Thankfully working with it through MIDI lets you get at all of those parameters simultaneously, and I can modulate them via Bitwig — for instance, increasing the particle density along with the dynamics.

Right now, I am running Particle with mono input, with Tensor immediately afterward. Particle V2 supports stereo input and output (on TRS jacks, not typical for pedals), but right now I’ve just got the ADDAC 200PI Pedal integrator to get the modular (and computer) talking to pedals, with two mono inputs and outputs. So that I can use Particle in stereo, I’ve ordered a Strymon AA.1 pedal interface and the cables I need to hook it up. (For some reason the AA.1’s “return” input is TRS, but its “send” output is a left/right pair of TS jacks.) Honestly I’m not sure how stereo is implemented with Particle — if it simply does the exact same processing to both channels, or if randomized values affect the left and right sides differently. I’m hoping for the latter, though. Even if stereo turns out to be a little lackluster, the extra pedal interface module will be helpful if/when I decide to add a couple more goodies, like the Freqout or Dweller perhaps.


Here’s a neat thing: Reverberations: an 8-bit approach to J.S. Bach

It struck me that, at least in theory, organ pipes should generate quite primitive sound waves. If so, how come a church organ doesn’t sound like a chip tune, which is also built up from simple waveforms? Well, actually it will, if you remove the church. And if you connect a Commodore 64 home computer to a loudspeaker in a large hall, it will sound like an organ.

Since the 80s, nearly every digital keyboard ever sold has had some kind of half-decent pipe organ preset, if not several — though maybe let down a little bit by faked reverb, which then starts to run into polyphony limitations. It turns out if you use a good convolution reverb, and a human performance rather than on-the-grid sequencing, even the SID chip in the Commodore 64 (or two of them anyway, for 6-note polyphony) can do a convincing job.

This is probably the least chiptunes-sounding music I’ve ever heard from the C64. I’ve often thought that there aren’t really any bad synthesizers, it all comes down to how you use them. (The SID chip was not bad for its time and price, particularly compared to the primitive beeps other computers and video game consoles had then, and is still prized/fetishized by some. It’s just a tall order for it to compete with modern synths, or other analog synths that aren’t crammed onto a single chip with 1981 technology. I’m not really into the sound of its filter compared to many other offerings, for instance. And it’s super buggy and quirky.)

a tiny bit of history

A little more than a week ago was my 17th anniversary as a member of the KvR forum. I’d meant to write something to reflect on it… but as I started the process I realized I didn’t have much to say. Neither the things that changed nor things things that stayed the same were much of a surprise to me.

Today, in a resurrected thread “Is FM synthesis your goto?,” I did find a surprise in something I wrote in mid-October 2016:

I don’t use FM as much as sampling or subtractive, but I do use it and like it. I tend toward experimental, ambient, noise types of stuff, so it’s not like I’m hooked on FM basses or something.

This was when I was planning to get my first Eurorack modules but had no idea how much that would impact my music-making methods and my style.

In the last couple of years I’ve rarely used samples anymore. Sometimes a quiet layer of a field recording in the background. I used one solitary note from NI Una Corda on No Place. I do kind of want to mess with sound-on-sound looping, which is technically sampling, but not sampling as the sound source as I meant in that post.

Likewise, subtractive synthesis has really fallen by the wayside for me. I was using my Microbrute in about 1/3 of my songs for a good while, but no more. Once in a while I use LuSH-101 or a subtractive patch on the Reface. For a lot of people subtractive synthesis pretty much IS synthesis, with the assumption that every synth has a filter and it’s the most important factor in its timbre.

I kind of got obsessed with FM from 2017 onward as I learned about “West Coast” synthesis methods, and the subtleties of exponential FM, linear and thru-zero linear FM, its relationship to phase modulation and wavefolding, etc.

Having recently heard some neat-sounding subtractive synthesis patches in product demos, and recalling how much I liked the Microbrute when used in specific ways, I’ve decided that I want to explore subtractive synthesis a bit again. So now my plan for album 13 is to do that alongside making a study of Akemie’s Castle. I’ve got a couple of great filters in my rack and can also make use of some nice software filters. Should be fun!