STARTHIEF

It’s nice to have cooler temps finally here, after the start of October saw 89 degree weather. Though we are looking at 81 in a couple of days, after that it’s forecast to stay mostly below 70.

And with cooler weather and Halloween decor comes a reminder that the moderation policy on the ModWiggler forum is… not ideal, and “Joe.” is the poster child for destructive interference. Last year he drove Blukač Instruments away from the forum by threatening to ban them for being “political” — because they dared to mention that their production is somewhat hampered by their country being invaded by Russia and their city being actively attacked.

Today there’s a thread locked and an 11-year veteran of the site threatened with a ban because the post “plagiarizes Karl Marx.” It does not, it’s a reference to the “commodity fetishism” angle that Adorno wrote about and which Jonathan Sterne expanded on in reference to music technology in an essay… except the post is more coherent. People were not using the thread to argue about politics at all. It’s just that Joe. has to play his role of ham-fisted petty tyrant, on his transparently hypocritical political crusade to shut down even the barest hints of leftism. I have seen him kill off a couple of total disaster threads that obviously needed it, but I maintain that the site would be better without his hands on it.

A couple hours after he locked the thread, another moderator came along and demonstrated much better leadership by posting some very thoughtful additions to the conversation (which again, wasn’t going in any kind of political direction). If the rest of us were allowed to continue commenting on the thread, that post would have turned things to a more interesting angle and it could have been a fruitful discussion.

But, nope, use one of the words that Marx once used and the thread gets murdered. Okay, whatever.

I suppose while I’m grousing about forum moderation it’s only fair to point out that KvR does not do enough of it, letting a majority of threads fester as they inevitably are hijacked by tiresome shouting matches between a couple of crusty regulars. Any thread over a certain length has probably turned into garbage. Pretty much the only thing that gets immediately shutdown is piracy talk and really blatant racism/bigotry.

And Lines, though I really like the site overall and the results generally work to maintain a pleasant atmosphere, sometimes baffles me with decisions about how threads get merged, moved and categorized, and I still don’t understand some of the decisions about the Releases category.

---

I’ve got a new SSD on the way now. When I bought this computer in 2019 I picked a budget drive, a 1TB Intel 660P. SSD speed, reliability and capacity have spiraled upwards since then while prices have slid ever downward. There are sales on now but rumor has it Samsung is going to slow production to prop up their prices a bit. With game install sizes ever growing and the new WRC game (by Codemasters, makers of the Dirt Rally series) coming next month, this seemed like a good time to upgrade. So, Samsung 990 Pro, twice as big, twice as fast to read and 5x faster to write, higher reliability/durability scores, highly rated software package.

Public service message: NewEgg’s current price on the 990 Pro is the same as Amazon’s “Prime Exclusive” event price. Also, my order is shipping (free) already, within a couple of hours of ordering it. One of the discounted SSDs at Amazon wasn’t even going to be delivered until mid-November.

In some ways the After Later Alan (and expanders) aren’t quite as good as the idealized version of Turing Machine in VCV Rack. Namely, the white noise generator has some issues:

• Lower level than the knob range on Chance. The effective range of the knob is only about 10 o’clock to 2 o’clock. To be fair, someone tested their MTM Turing Machine mk2 and the range is still only about 9 to 3 on that one.
• Audible “thunk” sound on the output when flipping the Write switch.
• Noise level increases by a small amount when the Length bit is high. It’s a small change but enough to hear the difference and to skew the results. The higher noise level makes it more likely to invert the bit, so it favors writing 0 over 1. When Chance is set somewhat low, this gives it a tendency to empty out all of the bits — and sometimes it stays that way for dozens of clock steps. It doesn’t do the opposite and get “stuck” with all bits high. If Chance is closer to 50% the effect is not really noticeable since it all falls within the bell curve. Increasing chance to higher amounts makes it gravitate to an “N bits off, N bits on” cycle.

Having sussed out its behavior, I can work with it. If I want more even distribution, or even deterministic behavior, I can use CV to switch the chance.

But honestly, I’m pretty disappointed with After Later’s quality control and attention to detail, particularly after they admitted that Tilt has design flaws that weren’t caught until after they started selling them. These issues with Alan were not hard to spot, and I thought something was weird about it in the first hour of using mine.

---

That “variant” industrial beat worked out for me after all and it’ll be on the album. I may also use more drums in the future generally. I had a brief moment of regret about selling my Elektron Model:Cycles — but I think the approach I took with this one works well for me. Each drum part can be individually synthesized (with a combination of hardware and software) and includes its own FX chain, and all of that can be freely modulated. Each part can be sequenced in MIDI and/or Bitwig Grid/Eurorack with various logic applied using manual controls… it’s a good setup.

I also had a really nice patch with the HSO, combined with some pitch shifting and saturation in Beads. I’m just really pleased with how smoothly this album has been going so far (even if I wound up not working on it this weekend).

I haven’t mentioned these things in my blog, but posted about them on forums I guess.

I miss the Zorlon Cannon mk2 I used to have. To simplify the technical details, an LFSR is a device that takes a random binary number, rotates the bits, and uses the bits to generate gates and CV in looping patterns, as well as crunchy noise. It’s the same technique older video games such as Defender used, a sound that screams cyber-retro-futurism to me.

I found myself checking out the Music Thing Modular Turing Machine; after videos from DivKid and Mylar Melodies and several hours of joyfully playing with a version in VCV Rack, I bought the After Later Audio version: Alan and the expanders Morcom and Enigma. Turing Machine’s particular take is to give control over the chance of flipping the feedback bit, which effectively determines how random or repetitive the sequence is. (Mutable Instruments Modules took inspiration from that in its design, though the technical details and feature set is very different.) The expanders allow for a second, more tunable CV sequence (but related in “patterny” ways to the first), and triggers that follow the first 7 bits of the number, generating cascading rhythms and also mixing nicely into Zorlon-like noise.

To make the room for these, I’m selling off both FM Aid and jroo Loop.

I spent a little time a couple of weeks ago recording and comparing different flavors of FM from my modular. It was a flawed comparison and I accidentally deleted the recordings (they were in a Bitwig “temporary project” and it’s kind of a crapshoot whether those are allowed to live when you close the software) but I still had some findings.

• Spectraphon’s FM is much more unstable than I thought; it sounds like there’s pink noise being applied to the FM index. This is completely unnoticeable when modulating the FM index though, and can be masked by other things.
• Odessa’s FM isn’t as dirty as I once thought, it’s just susceptible to messy modulation signals. It’s got by far the deepest modulation of any of my gear, but I think also it doesn’t have much DC blocking. So, clean signal in, good ratio and a bit of restraint = clean FM out. Anything else = potential dirt and chaos… which of course can be useful!
• Shapeshifter’s internal FM is ridiculously clean, just not that strong. External FM is relatively weaker still, while the PM is somewhere in between. The DAC specs are good, but noise in the system does have a noticeable effect, and I think I do get some noise when I run stuff through Tallin.
• Akemie’s Castle is, of course, never clean. All of the operators “leak” when they should be silent, there are only 6-bit registers controlling those levels, and so on. But it sounds like discount 1990 in the best way.
• Rings in FM mode has a neat “round” character to it that nothing else does, and is worth exploring more.
• For “FM any two signals” purposes, Koszalin is a lot smoother and cleaner than FM Aid.

FM Aid has that unavoidable spike in its waveshaping, and I just couldn’t find a purpose for it where I didn’t like something else better. So even though it’s my second one and I’ve said in the past I should never let it go… it’s time.

jroo Loop is a neat thing but honestly, I haven’t been using it for its intended purpose, as a performance looper. I’ve occasionally used it as another delay, just because it’s there. For other looping purposes I’ve just continued to capture recordings in Bitwig and then loop them there. So that space can be better used to generate some cyclic mutating patterns.

Since the release of 18 Points, I had (until a couple of days ago) been recording a bunch of little snippets, putting together patches in VCV Rack or Bitwig Grid, doing some experiments — basically starting stuff and not finishing it. That aforementioned kalimba/Hypnosis feedback patch was one of them.

But that gave me some raw material which I didn’t have any trouble finding ways to put together. Two songs in a row might not really be a “trend” of course. But they’ve been assembled from these pieces, using “Mix Paste” in Sound Forge. I’ve done some of that in the past of course, especially when recording bass parts or deciding that a song needs a little something else, but rarely to this extent.

It still feels improvisational somehow, even if it’s not entirely in real time. (Fake time, then?)

The second song in particular had something serendipitous happen — a synthesized “cymbal” sound used as an accent but which was exactly the right thing for its moment, which then mutated into a processed version that bridged the gap to a later synth part — it just sort of tied everything up neatly and made the whole form work.

---

I mentioned thinking about dipping my toes into techno a bit. A forum thread had me experimenting with an industrial beat, and I took inspiration from the Vamp Acid set at Knobcon, some Prometheus Burning tracks (particularly “Battery Drain”), and Manufacturer, maybe even a bit of poorly remembered Nine Inch Nails. (In that thread, we’ve pretty much come to an agreement that NIN beats and sound design were pretty great, and the lyrics… not so much.) This was done in about half an hour before work one morning, not having done anything much with beats since at least 2017 (aside from some idle playing around with Elektron Model:Cycles a couple years ago).

Then I cooked up a slowed-down variant with heavier kicks, more distorted and “corrupted” snares, and a ride cymbal that super-emphasizes the “bell” . I decided I liked that version quite a bit, and with some changes, it might actually make it onto this album project.

This morning I was just idly patching some stuff, and experimented with Mystic Circuits Ana changing up the regular stairstep patterns from Clep Diaz. It’s a pretty nice pairing, a nice little generative melody maker. I started tweaking the voice I was using for it — Shapeshifter through Natural Gate — and decided what it really needed was a kick drum. And then maybe some hi-hats, with a bit of groove from the velocity values as well as an LFO modulating things a bit. And let’s get some accents on that Shapeshifter… oh hey I still have delay assigned to Mod B and now it’s kind of a funky flanger thing. A little bit of compression and distortion and sidechaining, and… oops, I guess I kind of made a beat. It’s not going anywhere further than this stage, but there it is.

It’s not mind-blowing stuff; that bassline is a little too generative-sounding to be properly funky, and it needs some space instead of just being constant 16th notes. But it’s not awful either in certain aspects. Maybe the Techno Curse has lifted?

(I’m still going to stick with my thing… that drone, “dark ambient,” Berlin School-ish space. But maybe with a little bit more drums.)

---

I have a little space in my rack, and every once in a while I think about a comparator and/or latch to do something a bit more with that Doepfer switch I have. Then a thread came up on MW asking about small comparators, and I started weighing pros and cons. Someone mentioned the DPW Zero-2, which intrigued me before, and I went for it.

It’s a tiny, 2hp module of the type that I don’t really favor anymore, but I can make the occasional exception when it’s just jacks and doesn’t seem overcrowded. It’s a clever gizmo: an A=B comparator which clocks a D flip-flop, whose D input is normalled to !Q. What that means is this:

• If you patch one thing into input A, the output just detects whether A is positive.
• If you patch signals into A and B, the output is high whenever A is greater than B. (A simple comparator.)
• If you patch a gate into D, the output will be synchronized by A=B. You can use this to gate a VCA without clicks (since it waits for a zero crossing), for smooth wave splicing controlled by a gate, to add jitter to a clock, and several other interesting tricks.
• There are several different fun ways to patch it with Spectraphon audio outputs. And it gets pretty chaotic if you try to splice a signal and its own filtered version…

I had another very musical weekend. Did a little “mise en place” cleaning and rerouting and relabeling stuff. Finished a track Friday night and another Saturday morning. Received the Hypnosis Saturday afternoon, set it up and explored it a bit. Made the album artwork and then mastered the entire album, which generally went extremely smoothly.

---

I realized that this album is less “dark ambient” than it is Berlin School. There’s a certain similarity to older Tangerine Dream stuff: proto-ambient with (sometimes deceptively) simple looped sequences and slow improvised parts. It’s really not that much different from what I’ve been doing for quite some time, just a continuing drift that carried me from one region into another. Maybe this time there’s a little more emphasis on sequences, a bit more percussive action, and something a little different in what I’m playing on the Seaboard vs. 0-Ctrl touchplates or Minibrute pads. Maybe some of the influence is from the creative (and technical) limitations imposed by 18edo. It’s also very possible that some of my other albums should have been categorized this way too.

My general impression is that in 70s Germany, disaffected youth who wanted to break away from staid German traditional music started playing prog rock on synthesizers with a kind of techno-utopian vision: that was Kosmische Musik aka Krautrock. As genres went it cast a pretty wide net, with no two groups that sounded alike (and some groups that never sounded like themselves). A branch of that mutated into Berlin School, getting away from the rock/pop song structure and combining drones with repetitive sequences and improvisation. That in turn was a huge influence on ambient music. I also feel like it’s a natural place to return to

The other main branch emphasized rhythm and dancing, merged somewhat with disco and became techno. That of course influenced hiphop and spawned 934 different subgenres of house, trance, electro, IDM, EBM, etc.

In the post-Knobcon glow of hearing such great live techno, I think I may actually poke a little bit into that side again, on a trial basis. I feel like there must be a way to make it work for me. If not, no big deal, I’m pretty happy with what I’ve been doing!

---

The Dreadbox Hypnosis is a very neat thing, and I’m glad I got it. Having run out of other places, I’ve set it up on the left side of my desk on an Ikea laptop wedge. Stereo routing in and out from the audio interface is convenient to work with in Bitwig: I can either add an HW FX device, or a Grid preset I’ve saved that also includes a feedback path (with a little auto-limiting).

Various stages of the device can overdrive easily, and both the Chorus/Flanger and the Delay can stack up strong resonances fast. So you’ve got to keep an ear out for clipping and back off the levels, much more so than with plugins or most Eurorack effects. Worth it though.

With the three effects, wide parameter ranges, different delay modes etc. it has a wide repertoire — it’s not just dirty spring reverb and 80s wet chorus, though it excels at those. I think if I’d gone for the reissue version instead, that would still have been useful and cool, just with a lesser repertoire and less cool. It wouldn’t have had the sound that grabbed my attention so quickly at Knobcon. And there’ll definitely be times I use that Freeze feature (as well as feedback control with an expression pedal) while playing a part. I’ve already recorded a drone with an electric kalimba and feedback…

I believe I’ve just finished recording the album. I had been very much on a roll before Knobcon, and a few days after, picked it back up again with one track recorded last night and the final drone this morning.

Since both Micro-Pitch and Entonal Studio present scales in a radial graph, the theme for the album art suggests itself.

On Monday I almost bought a used Dreadbox Hypnosis, but hesitated because of confusion over which version it was.

On Tuesday, Dreadbox announced a Hypnosis Reissue.

• Significantly cheaper. It’s an easy-to-build DIY kit with no soldering. It also removed the preset knobs, which probably saved some expense (and which I didn’t care about).
• Eurorack format (but also has a standalone case with USB power).
• Patch points to modulate delay times and LFO rates. (You couldn’t modulate the digital delay on the original.)
• Instead of attaching an LFO to the spring predelay, there’s a separate patchable LFO with slew and a random gate mode.

Now for the bad:

• The chorus/flanger’s LFO is triangle only (not a big deal since you can patch an external LFO).
• The digital delay doesn’t have tape and BBD emulation modes.
• The digital delay doesn’t have a freeze button or gate, nor CV over feedback. Freezing the delay is a very cool feature on the original Hypnosis.
• It’s mono in and out.

I feel like the reissue is a pretty nice thing if you don’t compare it to the original Hypnosis. A BBD chorus/flanger, vanilla digital delay and a spring reverb with a unique predelay feature, all for about what a plain spring reverb driver would cost? Not bad at all.

But it doesn’t have the same vibe, mojo, etc. as the original Hypnosis. That chorus needs to be stereo. The delay needs to be ping-pong stereo (normally not something I am into, but at short delay times and high feedback and with the other stages in here, it works very well). Those aren’t really negotiable.

So for me, the reissue misses the mark. They could have gone two other ways with it:

• Give it a different name than “Hypnosis”. Lean into being modular. Have individual ins and outs for the three effects, and offer CV over mix levels and feedback (rather than gates to enable/disable effects).
• Give it a different name than “Hypnosis” — maybe even “Hypnosis Junior” or “Micro Hypnosis” or “Lethe” or something. Cut it back even to just the spring reverb with predelay and LFO (maybe with a tone control/tilt EQ like many spring reverb drivers have, and maybe a feedback path that includes the predelay).

After some pondering I went ahead and bought the “real” Hypnosis.

Nobody at Knobcon had the Dreadbox Hypnosis for sale. After more research, I found that there was a revision after the original release, and the one I tried was that new version, and I should make sure that’s the one I get. Also I’m still thinking about the details of connections and placement in my rig. But it’s definitely on my mind.

I drove home, feeling pretty exhausted and not really having a lot of interest in food. Getting home gave me a bit of an energy boost though, so after telling my stories and putting stuff away, I racked up the Harmonic Shift Oscillator and gave it a go.

It sounds gorgeous on its own, with FM-like-but-not-exactly-FM tones. FM produces sidebands potentially above and below the fundamental, and increasing the modulation index brings in more sidebands — this is described by “Bessel functions of the first kind” as discovered by Dr. John Chowning. It’s complex. Whatever HSO’s technique is, cranking up the Level doesn’t bring in more bands in the same way. You can dial Stride down to very near zero, so it’s not like it can’t produce undertones, but you have more control. It never goes all the way into noise like FM does, and you can modulate Stride with an LFO without generating a whole pile of weird aliasing madness. You can get some nice bell-like tones and deep percussive oddities, and a decent variety of other timbres from it, as well as slow beating and tremolo.

The two outputs are 90 degrees out of phase with each other, and you don’t have to worry much about mono compatibility issues with it — any cancellation is temporary and limited to some of the harmonics but not the fundamental, and in fact just blending the two in mono can be useful. There are a lot of other things you can do with the two outputs — combining them in Mystic Circuits Ana is fruitful, using one to modulate a delay or filter on the other works nicely, using one to switch the other on and off to create new shapes, etc.

It also has an expo FM input and handles it beautifully, with or without combining it with Level/Stride. Modulating Level or Stride at audio rates can also work well.

It gets up to high enough audio rates to clock Drezno or Interstellar Radio — while it’s not perfectly stable up there, the whole point of using those modules in this way is to add noise and grunge anyhow. In fact, setting Level and Stride can add additional flavors to it and thus it’s more flexible at this than Synchrodyne was. It doesn’t go quite as high in frequency, but I could always gain another octave by ring-modulating the sine and cosine outputs.

So, it’s a winner! I have not yet tried it with distortion, or Koszalin to attempt to bend the partials back into harmonicity (it would have to be tuned very carefully!), or explore what Spectraphon does with it. But I will.

---

I didn’t take a day off work tomorrow for recovery and I hope I don’t regret that.

…I should have taken a day off work today for recovery.

---

Overall thoughts about this Knobcon:

1. My first time, I was a relative beginner and it was extremely valuable to see and try lots of gear and see how people were performing with it.
   
   On the second time I had a bit more experience and a lot more musical focus, enough to provide additional context and perspective from which I could learn some more through observation.
   
   This time, I have a lot more practical experience and theoretical knowledge. I know what I like and the kind of music I want to make. My rig is in a very good state where I don’t want to change much regardless of how cool other options are. There are a few things I’d like to improve, both in terms of how stuff works and how I play. Therefore Knobcon hit very differently. I had a good time, but it wasn’t as valuable or inspiring to me overall.
   
2. The gear stuff mostly confirmed to me that I want to stick mostly to what I already have. The one thing I bought was for a change I was already considering, it’s just that the opportunity happened to strike. (And ironically I researched the module a little online before going in for the live demo.) The one other thing that I’d like to pick up is in the category of stuff I already knew I liked. The thing I’d like to buy in the future when it’s available is just a refinement of a module I’ve enjoyed for a while, and I didn’t get much more info than I would have from an online announcement.
   
3. The performances were a good time for the most part. I learned from them too — I learned that I do not want to deal with the challenges they pose. I feel like recording live gives me the kind of spontaneity I want, but mitigates the risks. Also, the big thing I learned is that filling a specific time with a performance is hard when you are improvising — you either need to plan some kind of form in advance, rehearse, and keep an eye on a clock, or you need to be very nimble with transitioning to something else to keep it fresh.
   
   I do want to work on issues of precision in my improv. The very first performer Friday had dead-on precision with his violin playing that matched with the sequencing, and I’ve been thinking about that. The third one had reckless abandon and I thought that worked very well for her music, up to a point. The techno folks had flawless execution of everything they did, although the nature of their tools helped with that. For me personally, I’m thinking visible metronome, a bit more planning and practice of riffs, and probably just more practice with the Seaboard (and bass!) will be of great benefit.
   
4. I guess I need to accept that there is too much going on at Knobcon to catch it all in any case. And I’m just out of shape and oldish and there’s the long drive before and after to deal with. If I go again, I will just be more chill about it, and also give myself more time after — maybe even before. But I think I will probably skip it at least for the next few years.

The answer is two hours, I could do two hours of the Big Room performances even sitting through most of it. There were 6 performers on the slate, and two hours was half of them. I’m sad to have missed POB’s set especially, but yesterday was intense and I have a long drive today.

Space Racer was good, modular techno. She kept things varied, she had a bit of structure to it — I think there was actually a lot of improv going but it all worked. One continuous rhythmic flow, with different parts joining in and always keeping it fresh, and an interesting variety of timbres. I fully enjoyed her set and it felt short to me — she successfully left her audience wanting more.

Next up was Dub Station Zero. Obviously had a strong dub influence. Good transitions, solid and occasionally very funky rhythms, fun basslines with occasionally wild buzzing timbres. He has mastered his craft for sure. His set was longer but it felt 100% right. Good stuff. It sounded like I was listening to an album, not someone who just brought a drum machine and small Eurorack case to a hotel. I would buy that album.

And then, Vamp Acid. The dark side — driving rhythms and more than a little industrial and gothic flavor, I was totally on board with it. When it’s almost 10 o’clock and I’m that tired it takes something special to make me want to get up and dance. (I don’t dance.) She seemed to have specific songs she was doing, though not without improvisation too. Changed the tempo up a few times, including some fakeouts and surprises which worked super well. She sang, and my only complaint is that her vocals could have been louder in the mix (not that this isn’t common with goth acts). My favorite of a very good three, and I’m happy to learn she does have an album out too.

This morning I’ll go eat breakfast, check out of my room, see about that Dreadbox Hypnosis and head for home. I didn’t take a day off work tomorrow for recovery and I hope I don’t regret that. 🙂