Author: Starthief

This week, Make Noise announced Bruxa, a module based on the prototype “Time/Filter Experiment” from the Strega. The delay is based on the PT2399 chip, a super cheap digital delay with analog feedback path, used in dozens of guitar pedals and quite a few Eurorack modules. It’s often meant to imitate an analog BBD, and it has similar characteristics — a fixed buffer length, limited fidelity — but it does use memory rather than capacitors. (Some marketing copy will call it an “analog delay” anyway, some hybrid, and some “digital”. It’s just words.)

The PT has a characteristic distortion, and when clocked too slowly (trying to extend the delay time longer/slower than it can handle) it tends to generate weird insectile/crunchy noises. In a lot of designs, the clock rate range is limited — using multiple chips in series if longer delays are needed, instead of slowing them down too much — and the signal is filtered and companded to reduce the noisiness but give pleasant, dark echoes. But in the Strega, and now the Bruxa, that noise and degradation is celebrated rather than suppressed. It does use multiple chips but also multiple parallel feedback paths to tailor the sound.

Bruxa is apparently only going to be available for preorders, and the price is a bit on the high side. Given that this one is quite close to what the Strega offers anyway — maybe some subtle differences in the feedback paths? — I don’t feel like I’d benefit all that much from also having this one, and making space for it means giving up something else. But contemplating it has got me patching up Strega in new ways, running other audio through it from the DAW or modular rather than using it as a self-contained instrument (or as a pair with the Minibrute 2S) and finding a new appreciation for not just the gritty delay, but its integrated filter and what happens when you modulate various things.

The sound of Strega’s delay fits with where this album has been naturally heading. Far from avoiding delay and reverb and going “dry” like I’d considered a few months ago, this one is celebrating noisy, distorted delay and reverb, and some similar textures achieved via other means. The new Aberrant DSP Lair reverb and Audiority Doomagorgon fuzz plugins have featured, as has Sinevibes Integer (which was inspired by BBD and PT delays). At one point so far, Noise Eng Morbus Legio adds some sub-audio noise to a distorted signal, which also has a crawly sort of texture though very different from the PT. (And again, I want to emphasize that while the gear certainly contributes to the sound and often its “natural” behavior inspires it, the biggest factor is me. This is the sound that is right to me at this time. If Lair and Doomagorgon and Bruxa hadn’t been released this month I’d still be heading in this direction, even if the exact path and destination are different.)

Anyway, I’m at about 34 minutes of material for the album and going strong. There was one rejected track, but that’s fine… a lot of the ideas that went into it (both in terms of patch design and general musical concepts) went into the next attempt, which worked. Sometimes you just have to let your brain reboot and take another stab.

I’ve got something recorded for the next album project. It’s been a while — the previous album went out on April 19. Since then:

• I got Dawesome Myth & Madrona Sumu & Harmony Bloom & Lair.
• I got Ana 2 and RYK Algo, and did a case rearrange.
• I did studies of Akemie’s Castle & Xaoc Drezno.
• Superbooth happened. Even not going, there was a lot to take in via the internet…
• Drone Day was my one actual bit of music-making.
• Guild Wars 2. I’ve been keeping up with the daily and weekly PVE Astral Acclaim rewards. Not always all of the tasks — I’ve been skipping Cantha (I haven’t been there yet, though I bought the expansion) and fractals and some of the more annoying jumping puzzles — but enough of them to always get the daily and weekly bonus rewards. To be honest, the daily rewards probably aren’t worth much effort so I’ll be taking a more relaxed attitude toward them from now on.
• Soulstone Survivors. They released another new class, and I caught up on all of the unlocks for all of the classes. It’s kind of a regular go-to game.
• Instruments of Destruction. A physics-based vehicle building and demolition game, like an erector set + Red Faction: Guerilla + a bit of Battlebots. There’s a campaign with prebuilt vehicles, a “Build & Destroy” campaign where you build your own (although most of them could be easily finished by loading an existing campaign or bonus vehicle, it’s usually more fun to make something from scratch or modify designs for the specific task at hand), and a Sandbox mode for doing your own mad engineering. In the latter, it’s very easy to lose hours just coming up with weird designs, tweaking them, and smashing buildings to smithereens in very satisfying ways.
• Visiting my parents! We took them to the farmer’s market and bought some native wildflowers, pulled weeds and cut out some invasive Japanese honeysuckle, dug through layers of rocks to plant flowers in the area in front of their front sidewalk, anchored bookcases to the walls and set up their big TV (and then the replacement because the screen got messed up during the move). And we just went for meals and to hang out, several times. I really don’t love yard work, but it’s good to be able to visit them without the round trip requiring 22+ hours of driving.

As a side note, back in January I got that big fantasy Lego set and the cases to organize things, and I honestly haven’t done anything with them since. They’re sitting there waiting for a day when I want to get away from the computer and do something more directly with my hands (aside from making music). It’s just that I don’t often feel a need to do that, I guess! It was diverting for a few hours during a period of a few weeks. If I don’t end up getting back to it, I guess I can always give them to my nephew once he’s ready for Legos.

“A Trillion Voices” (for Drone Day 2024)

1. Cellphone recording of cicadas and traffic noise, Phonolyth Velvet Machine, Mutable Instruments Beads, Bitwig Transient Split with Freakshow Industries Mishby on the transients. (I’d been hoping to get a nice binaural recording when the cicadas came to our neighborhood, but they’re a couple miles short of it still.)
   
2. Make Noise Spectraphon (chaos mode), LFOs into FM and Focus (* Audiority Xenoverb on voice 2, 3 and 5)
   
3. RYK Algo, modulation from Industrial Music Electronics Kermit mk2 and Kermit reshaped by Xaoc Drezno. (*)
   
4. Noise Engineering Cursus Iteritas Alia, Noise Engineering Lacrima Versio, Unfiltered Audio Lo-Fi-AF, Izotope RX7 Declick, Sienvibes Integer, Audiothing Fog Convolver
   
5. Spectraphon output from #2, through Xaoc Koszalin, Noise Engineering Melotus Versio (*)
   
6. Xaoc Odessa through Rossum Morpheus, Arturia Refract, Valhalla Delay

I’m done writing up my study of Xaoc Drezno. As always, I learned a few things. I love that it’s such an abstract and non-prescriptive module, a real chameleon… but as to be honest, I’m probably still mostly going to stick to old habits with it. And that’s fine, they’re habits that work for me 🙂

Drone Day is this Saturday, May 25. While I don’t plan on attending any actual events, I’ll at least see about recording something. That should get me back into the swing of music making for a while before I choose another module to write a “book” about… 🙂

It’s that time of year again, the time when a critical mass of synthesizer builders gather for Superbooth at FEZ Berlin for a massive round of show-and-tell, and synth players and sound designers all over the world inevitably rearrange their studios to make room for new goodies.

There are always teasers and early announcements, but this year there seemed to be more than usual, with a few full-on demos and product releases before the first official day (or even week) of the event. RYK Algo probably should be counted among those.

Superbooth is much more about hardware (especially Eurorack) than software, so it was probably a coincidence that Madrona Labs just released the early access version of their long-awaited Sumu.

And it is definitely a coincidence that this wasn’t all that long after the release of Dawesome Myth, the other software synth I’d been waiting for and which also is based partially (heh heh) on resynthesis.

Sumu is a lot to take in at first, and as excited as I was about it years in advance, it took me a good while with the demo (and sleeping on it) before I decided to actually buy it. In a nutshell: spectral additive resynthesis, but instead of sine waves, it uses 2-op FM (or AM) pairs. It’s semi-modular, and every “cable” can be either singular (like MIDI gate/pitch/pressure or a time index), or a bundle of 64 parallel channels, which can come from the resynthesis data, a bank of parallel envelope generators, or a bank of parallel function generators (“Pulses”) with probability skippers and rhythmic repeat/rest options. There’s also a “Spaces” block that positions each of the oscillator bank’s outputs in 3D space, with parameters for moving them around (complete with doppler shift). Like I said: it’s a lot.

One of the points where I hesitated was wanting more basic modulation — some simple envelopes and LFOs. And also, modulation of modulation depth. But I realized first that (A) Bitwig can do this for me and (B) after mastering Pulses and learning to embrace the Way of Sumu, this is a bit less of a necessity than I thought, and (C) every dial is going to gain its own simple LFO available with a right-click in an update.

A few things you can do:

• Basic spectral resynthesis. Just let the time index control the Partials Map time, patch Amp to Amp and Pitch to Pitch, use Envelopes to control the level, voila. You can still detune the partials inharmonically, and apply FM or AM if you want.
• Crank the pitch input to almost zero, so it’s just a bit of unison detuning. Lovely with FM!
• Basic 2-op FM or AM synthesis… just don’t patch the Partials Map to the oscillator at all.
• Use the Partials map as a template, but let the time stay static. Lots of timbres available there, particularly with the detuning, FM and AM.
• Control the Partials Map time with the Envelopes or Pulses, so the partials are disjointed from each other. I find that some slow motion can work really nicely here for ever-changing timbres.
• Map Partials Map pitch to FM modulator ratios instead of (or in addition to) the base frequency.
• Map Pulses to FM ratio, offsets or even base pitch for wild swoops or subtle wobbling.
• Allow wild motion to happen in Space, but map Pulses to its reset input to bring things back to a (relatively) central point where they’re audible again.

Overall I’m finding this great for drones and soundscapes. It’s quite CPU-hungry (at least in its current Early Access state) but I am finding that one voice of polyphony (about 10% CPU use) is often plenty for cases like that.

---

As for newly announced gear: there sure is a lot of it. So far, most of it is kind of only of moderate interest to me. I’m not saying it’s not good stuff — there are a few things I’d love to have if there was space, like the Soma Flux or Polybrute 12 or Udo Super 8. And more things that are probably great but I recognize I just don’t need them, either because it’s covered or not in my musical wheelhouse. But even the latest iteration of Korg Berlin’s electroacoustic gizmo seems less interesting to me this year than the previous one did last year.

Instead, an electronmagnetic… or rather, kinomagnetic (?) gizmo caught my eye: Landscape Ferrous. This lovely looking gadget rotates strong magnets at variable speeds to excite strings, kalimba tines, and so on or just to confuse magnetic pickups. Strong enough to work on bass and piano strings, it’s a ticket to drones and harmonics and other things that are right up my alley. So I put in my preorder for that.

But, yeesh, Myth and Algo and now Sumu are all new to me, and Aodyo Loom and now this are in preorder, and I’m still in a mood to explore all my older stuff too.

I got my RYK Algo, and yes, it is another 4-op FM module introduced to me by Mylar Melodies which is utterly fantastic.

There are some odd design choices:

• The output jack is TRS (like a stereo headphone jack, 3 contacts rather than 2). VERY few Eurorack modules use that. A splitter to provide separate L and R outputs is provided, but it’s inelegant.
  
  (I have a personal solution for this involving a right-angle TRS extension cable, a ventilation panel, a splitter hidden inside the case, and the two currently unused jacks on my Mazzatron Mult & Passthru, as soon as that extension cable arrives.)
  
• There is no master coarse tuning knob. There are frequency knobs for the four operators, there’s an octave button, and a fine tuning knob with +/- 2 semitone range. Also, this may be a technical issue, but my (external) FM input only accepts positive voltages (the manual says -5V to +5V) so this can’t be used to easily transpose everything.
  
  This sort of thing bothered me on Loquelic Iteritas, because I felt like it was difficult to retune the module while preserving exactly the right ratio. Here, it seems to generally be fine so far. But I’ve just been testing the module and not trying to fit it into other parts yet. If the FM thing is fixed that should be fine.
  
• I’m being picky about the knobs, as usual: it’s hard to see the pointer direction because they’re small, not super bright white dots on black knobs on a black panel, with high-contrast LEDs glaring at you. Thonk Tall Trimmer Toppers and “Pantherkill” knobs from Love My Switches will fix that.
  
• Also, the knobs follow the same diagonal orientation as their layout — so the center of the knob range isn’t “noon” but 45 degrees right. That takes some getting used to.
  
  The bigger knobs are T18 so you can reset them if you like, and trimmer toppers fix the rest.
  
• The Algo (and more rarely, Octave) buttons are used for various shift functions — free and fixed tuning options, operator panning, warp and fold styles. This can be a little awkward especially if you’re trying to do it one-handed. OTOH, everything in the interface is easy to remember with no cheat sheet required.

The sound is devastating. Gorgeous or monstrous, your choice. The sine waves and FM are a clean, blank slate. The combination of the offered algorithms (with parallel modulators feeding a carrier, and/or common modulator(s) feeding multiple carriers), combined with the ability to quantized-tune, free-tune or fixed-tune the operator ratios, opens a lot of doors. Stereo panning of carriers sounds fantastic, but of course as with most modular “stereo” you can treat it as separate signals for individual processing, or combine them with ring modulation etc. The warp and fold on their own are nothing particularly special, but can provide some extra zing or solidity to the sound at times. (Warp really doesn’t have a similar character to FM feedback and it’s not as much like Shapeshifter’s Tilt as I expected, but still.)

As it turns out, algorithm 5 isn’t a super close imitation of Shapeshifter’s Harmo3/2Tone6, but I kind of don’t care because Algo is just all-around better at the whole growly, textured business.

Algo and Akemie’s Castle complement each other well; despite the commonality of “4-operator FM controlled via knobs and CV with no menu diving,” they have different capabilities and occupy different sonic territory. One’s spicy and the other umami, perhaps.

I’ve rearranged the case — it’s no longer grouped by manufacturer, but by function. Here’s a color-coded zone map:

Red = oscillator, green = filter, yellow = VCA, blue = modulator, purple = FX, brown = controller, and gray = utilities…. at least, generally. Modules have a lot of functional overlap of course — Kermit is often used as a modulator, Blades sometimes as an oscillator, somehow I didn’t count DAPF as a filter, and I don’t know how to categorize Drezno. Also it amuses me that Ana is already purple and is somewhere between a utility and an effect.

(Also, in my pod is still Univer Inter, Auza Wave Packets, Just Friends, and Sweet Sixteen. So utility, modulation/oscillator, modulation/oscillator, and controller.)

Speaking of Drezno though, although the whole order of modules has changed, I’m going to go ahead and finish writing it up first. I’ve gotten partway through my study on it. Mostly I have audio uses covered, with one more idea to write up and create a sound example for. There’s still gate extraction, a bunch of CV uses, and generative feedback patching to write up! And I expect there are uses for Drezno I still haven’t dreamed of. And all of the rest of the Liebniz system, which I don’t have, some of which are even more abstract and non-prescriptive.

I feel like I should do a full study of Algo later, after I’ve had some time to live with it. So probably after Drezno, if I don’t launch into another album project (which I might), it’d be Zorlon Cannon. Another favorite of mine that can be used for audio, CV, and gates in many different ways…

The very day after I published my notes on Akemie’s Castle, another 4-op FM module made its grand entrance.

RYK Algo. This company definitely sticks to a certain dark minimalist 80s “banks of supercomputers deep in a nuclear bunker” visual aesthetic. I dig it.

More importantly, I dig the sound. It was shown off by Mylar Melodies, the very same gentleman whose video on Akemie’s Castle convinced me to get that one. He didn’t steer me wrong then, and he isn’t wrong this time.

No, I’m not replacing the Castle. It has that retro, grungy sound where this one seems super clean. It has the dual independent oscillator thing going on, and the chords, for massive clustery madness. I’m keeping it even though I’m also getting this one — its clean, shiny, modern younger sibling.

What’s going to step aside is almost certainly Shapeshifter. I did a whole study on it, but effectively, that shifted my habit from “use Basic1 with FM or ringmod and maybe Tilt” to “use Harmo3 or 2Tone6 with FM or ringmod and maybe Tilt.” Those two wavetables both are pairs of sines with different tuning options throughout the table.

Well… if you decipher the algorithm display in the photo above — a bit funky but it makes sense after learning it — what we have is two operators mixed and modulating two separate carriers. That is exactly what I’ve been doing with Harmo3 and 2Tone6. Except now I’ll have full control over the tunings, relative levels, and panning plus a little extra shaping. This isn’t even an algorithm that appears on any Yamaha 4-OP FM synths that I know of…!

Last year’s must-get module was Spectraphon, and Algo is this year’s. As I’ve said elsewhere, I’d like all the instrument makers of the world to get together and just chill for a while, you’ve done great work and deserve a rest. 😉

I decided to actually do a proper study and writeup of Akemie’s Castle, after going over it so many times throughout my modular journey to judge it (and always finding it worthy).

Unless I come to a point where I consider getting away from Eurorack entirely, I don’t think it’s going to be subject to that kind of trial again. It’s proven itself too many times.

I’m glad I went through the semi-formal process this time. I tried to take nothing for granted, and wound up learning some new things about this module that I’ve had for years.

I think I will proceed across the top row and hit the rest of the modules as I go. I’m not sure I have much to say about Xaoc Tallin, so I might write up a combined page with the “trivial” stuff. But next to that is Drezno, and that’s going to be an interesting one: a real chameleon that depends greatly on how you patch it. I’ll have to work out a strategy to cover it.

The long(ish) awaited Dawesome Myth was released a few days ago, conveniently after my album release. Despite a lot of words on the website about how easy and intuitive it is, it’s a synth you can go very deep with, which is reflected in its generous 90 day demo period.

Looking at the GUI and some of the hype, you’d be forgiven for thinking that the raison d’etre of the plugin is its resynthesis engine. But:

• you can leave this at a default sine wave, and using only a handful of “transformer” parameters, make several radical changes to the sound using FM and really intuitive waveshaping
• you can add modules to the oscillator, filter or FX sections of the synth which allow for multiple types of FM as well as ringmod, AM, subtractive, distortion synthesis and physical modeling
• you can entirely disable the “iris” and add an oscillator module with a choice of classic waveshapes and then use that for most of the other synthesis methods

It’s not a standard FFT analysis, additive resynthesis thing like Alchemy and many other predecessors, where it produces a similar-but-slightly-worse version of the input. It produces something that is kind of related to the input, but not always recognizable. Some people think “it’s crap” and give up right there. But as far as I can infer, the point isn’t to sound like other synths but to provide feed stock for those transformers.

On the one hand, different strokes synths for different folks. On the other, the people dismissing it because it isn’t the thing that they thought it should be are missing out.

---

I almost had talked myself out of getting Myth before digging very far into the demo myself, not because of anything about the synth itself, but because I looked over my gear usage for the last four releases and it was a bit surprising. The breakdown for voices (looking strictly at their origin, not effects/processing) is about like this:

50% Eurorack
20% software synths (VST/CLAP)
15% software modular (VCV Rack & Bitwig Grid)
5% Strega/Minibrute 2S
5% bass
5% other hardware / physical instruments

Within the softsynth category, usage was spread out between several different plugins. My absolute favorite softsynth, Aalto, got used 2.5% of the time. My second most used was, to my surprise, Manis Iteritas. I had a 5-way tie for third-most. There are plenty of excellent synths I like that got used just once or not at all… so does it make sense to get more softsynths?

After trying the demo of Myth though… yes, it does. It’s almost closer to the VCV Rack category than any other synth I have.

Also, now I don’t feel so bad about not playing the bass more. That’s just how it goes for me, and that’s OK. (But this is something I’ll come back to think about with each potential new gear purchase…)

---

Also the new Mystic Circuits Ana 2 was released. I kind of thought about skipping it after all — I don’t really need the additional utility it provides after all — but the greatly expanded potential for fun modulation made me decide to go ahead and grab it.

The new album is released… enjoy!

Notes are here. I have a bit more detail this time, showing which voices were in which individual takes, so you could see when I went back to layer something else in, which ones were constructed from lots of assembled bits (still not multitracking, more like overdubbing) and so on. If you’re curious.