|6:50||This is all software: u-he Hive through Unfiltered Audio SpecOps and ValhallaPlate, but "performed" and recorded in a similar process to how I record modular drones.|
|195||Oct 3||I Hear Your Bells
The length of these songs has little correlation with the time and effort spent making them...|
Voice 1: E352 Cloud+Morph mode, Z Morph modulated by LFO, Chaos by envelope. Into Sputnik LPG & Valhalla Room.
Voice 2: 0-Coast with mixed Slope+Contour controlling its Dynamics, square into mix input. TB Compressor & Permut8.
Voice 3: Rings sympathetic strings, through Freez with pitch CV into sample rate. SpecOps VST with a touch of Threshold Freeze and Smear and a lot of Noisify with the range modulated by envelope follower.
Voice 4: Tides through Sinclastic Empulatrix.
A couple of firsts in this song: the first time I used all six "main" inputs on my audio interface (the other two are dedicated to 0-Coast and Microbrute).|
My first use of Dopfer A-102 VCF9 Diode Low Pass filter, which is a Steiner-Parker design very much like the one in my Microbrute. A bit more open-ended in some ways, which makes it a little less immediate to work with in some ways, but also more flexible. All kinds of feedback options exist. But the Microbrute also has bandpass and highpass modes, and... right now I'm conflicted about whether I want to keep it, replace it with a Reface CS, or find some clever way to fit two small keyboards in this space.
I would say it was my first use of Jones O'Tool Plus, but really that was more last night as I was discovering what sort of things it could reveal and learning the filter's quirks. It's more of a learning and diagnostic tool than something that gets used in songs, except for some tuning and setup use. Very much worth its rack space though, to be able to see the input and output of a module simultaneously, or hunt down troublesome DC offsets or weird voltage ranges or the like.
Voice 1: Tides into A-102 filter. Filter CVs are from pitch, an envelope, and feedback from the filter output through Gerridae vactrol VCA and Freez. Filter output also feeds back to Tides shape, and goes out through DPLPG and a chain of multiple VST delays.
Voice 2: Hertz Donut through Sputnik LPG, with envelope severely attenuated to keep the LPG from opening very far, through ValhallaPlate and SerumFX ensemble/phaser. Layered with both Kermit channels in stereo, through Superfake chorus VST.
Voice 3: E352 in Cloud+Morph mode, with both outputs in stereo thru Zlob dual VCA. Basslane and Monomaker to partially merge them, Echobode and ValhallaRoom for a reverb with perpetually falling pitch.
My new Make Noise stuff arrived today -- Dynamix and Contour as first seen on the 0-Coast, as well as Function. The latter two are replacing my Rampage, with snappier and more flexible envelopes controlled in a manner that makes more intuitive sense to me; Dynamix is half the team that will replace my Sputnik Quad VCF/VCA.|
Since I've had WMD Mini Slew for quite some time, Function isn't too much of a surprise -- but it has a couple of different quirks. It seems to be able to cycle while also holding an input level. It triggers on the falling edge of my 2HP LFO for some reason. EOR is on during the "falling" stage and EOC is on otherwise.
Contour is pretty much exactly what I would expect from the 0-Coast, with the extra CVs and outputs. EOC is on during the "onset" stage and EON is on otherwise. Pretty simple.
Dynamix, on the other hand, is quite a step up from the 0-Coast's Dynamics section. It's dual, with a mix output and aux input. Each half has a an incoming signal level control, a CV level control, and a Dynamic input -- so you can affect just the volume, or volume and brightness together, or compress/expand/bias/waveshape in some ways. It's very flexible and I'll be exploring it for a while.
As an LPG its sound isn't quite as "round" as the Sputnik but that's on picky A/B listening tests and staring at the scope. It's still much more interesting and natural sounding than a basic VCA, I find.
Anyway, the song: named by neural network, which just seemed to fit somehow.
Voice 1: E352 in 2-Op FM mode, with Contour and Dynamix. FilterBank3 & Replika VSTs.
Voice 2: Hertz Donut (also doing FM, plus discontinuity) with Function and Dynamix. ValhallaRoom VST.
Voice 3: 0-Coast with DubStation2 VST.
Arranging modules in a rack is an art with many considerations. They're all different widths, and empty space should usually be consolidated if not reserved specifically. It's nice to keep modules from the same maker together just for aesthetics, and to keep similarly colored panels together. Some modules have displays that are easier to see at eye level than below. It can make sense to keep modules with similar functions grouped together, and there are typical flows from one type of module to another. But there could be arguments for separating functions a bit so the patch cables for each voice tangle with each other less.|
Changing the relative placement of modules has an effect on the feel of using the synth, and a subtle one on the music you make with it. Some people think it's good to just shuffle everything every so often, and occasionally leave important modules out, just to break habits and tease out new techniques.
Unlike a computer RPG, there's no auto-sort button and it's not just a matter of dragging things around. With my last few trades, I opted for the least amount of shuffling to fit everything in -- and that's resulted in a bit of chaos. It's been an interesting lesson, but I think I'm going to have to "fix" it soon.
Onto the song:
Voice 1: Three Sisters self-resonating, through Doepfer A-102 adding a little of its own resonance, through Contour-controlled Dynamix. SpecOps and ValhallaPlate VSTs.
Voice 2: MicroBrute, through BX Blue Chorus and Replika VSTs, with SpecOps and ValhallaPlate on an FX send.
Voice 3: Hertz Donut (through Zlob dual VCA controlled by Function) feeding Rings, with ValhallaVintageVerb. This is some pretty fertile ground that I'll be exploring more in the future, I expect, with a chuckle and a shake of the head at people who say Rings is a one-trick pony.
|199||Oct 8||Windows Open
The temperature's nice, time to air out the house (and play some synths).|
Voice 1: Two squares from Hertz Donut with phase offset by LFO, into Function Trigger and Hang. Through both channels of S.P.O. in serial to offset and amplify it into something closer to normal voltages ranges, then Doepfer A-102 and Dynamix (both controlled by Contour). Transient Master and SpecOps VSTs.
Voice 2: Wurlitzer EP emulation in Arturia Analog Lab 2. MTransformer, ValhallaVintageVerb.
Voice 3: A respectable synth choir on the E352 using Rom B in Cloud+Morph mode with some chaos, through DPLPG, with ValhallaVintageVerb.
Voice 4: A slow, mellow, soft pad on 0-Coast with a little sub-octave action from its Slope section.
|200||Oct 8||Gentle Correction
For this song -- my 200th so far in 2017, not counting some module demos -- I took a couple of recent patches and bent them slowly but firmly into a different shape.|
Voice 1: 0-Coast with a similar patch to the previous song, but letting the sub-octave get a little dirtier. MTransformer VST to balance the timbre of the main oscillator above the sub a bit, in a different way than the 0C's own balance control does.
Voice 2: Serum VST, using the BenjPad preset but altered a bit and filtered a lot.
Voice 3: A string-like voice from from Hertz Donut feeding Rings, in 4x polyphonic Karplusverb mode and a somewhat different configuration than the similar patch in "Uluunna." ValhallaVintageVerb.
Ring modulation and white noise are things I don't use as much in my music as some folks do, and I was thinking today I should correct that. So here's a bit of both.|
Voice 1: Tides ring-modulated against Rings, through A-102 filter. Contour provides the envelope for Tides level and shape as well as a little bit of filter cutoff; synced LFOs from Kermit modulate Tides slope and Rings brightness. ValhallaRoom VST.
Voice 2: Hertz Donut FMd by filtered crackle noise from the E352, through Function-controlled Dynamix. Layered with filtered white noise from E352, through Mini-Slew controlled Dynamix. ValhallaUbermod VST.
Voice 3: Pads from PlastiCZ VT through SpecOps and ValhallaVintageVerb.
|202||Oct 12||Blinded by Gold
An apparently angry young man opined (immediately before the post was deleted either by himself or a moderator) that I must be blind, new to modular synths, and an idiot, because I did not agree with his (ahem) "unique" interpretation of a standard modular synthesis term he found in a manual for a synth that he didn't understand how to use.|
It was more amusing and sad (for him) than anything. I've seen Randos Who Think They Know Everything attempt to take down much more knowledgeable people than myself many times. It's not a good look.
So I decided that, with my (ahem) "ignorance" of the synth in question, I should make some music that makes use of the exact thing he was asking about.
One could also reinterpret the title as a condemnation of greed. I'm down with that too.
The track has a particular technical wart, entirely unrelated to the young man's argument, which I've chosen to leave partially intact for the reasons of irony, whimsy, laziness, some sort of artistic notion of staying true to form, and experimentation. Namely, is it going to bug me when I listen to this later, or am I hyperfocused on it right now?
Voice 1: 0-Coast, with Slope cycling and feeding into audio via the Overtone knob, with a separately gated VCA allowing the squarewave osc into Slope trig to sync it. Blockfish and ValhallaVintageVerb VSTs.
Voice 2: Microbrute with CamelCrusher VST.
Noise drum: E352 clocked noise (I just like it better that most other kinds of noise, okay?) through Three Sisters and Contour-controlled Dynamix. ValhallaRoom VST.
Kick drum/owl hoot thing: A-102 filter "pinged" with gates at rhythmically appropriate lengths, with a chain of dynamics/enhancement VSTs and ValhallaVintageVerb. (This is also done wrong, but intentionally so from the start.)
|203||Oct 13||we considered ourselves to be a powerful culture
Voice 1: E352 in Morph+Fold mode, with output 1 modulating output 2's fold amount, Hertz Donut (from voice 2) modulating output 1's fold amount, and Kermit LFO modulating the morph. Through Dynamix as a compressor, then ValhallaVintageVerb.
Voice 2: Hertz Donut, with the primary osc set to LFO rates, the secondary to low audio. The XOR output FMs the primary osc, and the mod bus is set to discontinuity (orange mode). Triangle output is monitored. EchoMelt VST.
Voice 3: Tides into Rings, set to sympathetic strings mode at a low frequency. 2HP LFO stepped triangle into Position, and constant rapid strumming from Hertz Donut's modulation osc. My Rings has a loose FM jack so I intentionally wiggled the cable into it for some pitch jumps. ValhallaRoom VST.
|204||Oct 14||FLlp0oo2dr ST4rr
For what it's worth, my impression now of "Blinded by Gold" isn't "I should have fixed that thing." Instead, "doesn't fit with the kind of music I want to make now and I should have taken it in a completely different direction or dropped it as a failure." Shorter version: "meh."|
But this song is a song of instability, which seems appropriate for the present day. It's in 6/4 time with a reverse swing applied on every 1/2 note in a doggedly mechanical way. The first voice features an analog, almost organic oscillator which has many threshholds of shifting behavior, a certain unwillingness to play the exact note it's asked, and unbalanced output levels. The second voice is a synth with a filter that's jittery at times and growly at others. The third involves two very precise oscillators but also another one of those crazy filters. The fourth is an extra-quirky digital take on a design originally associated with hippies and academia, set up here in an especially atypical way.
The name is by neural network -- but curated by humans. One human made the list that fed it, set the parameters for generation, and chose which results to share. Another picked it from a list, held onto it for the right occasion, and decided this was that occasion. So in a sense, it's almost as much of a collaboration with machines as my music is.
Voice 1: Mannequins Mangrove in Constant Formant mode, with Contour controlling Air and Barrel and Function modulating Formant, running through a Dynamix channel without additional CV. Function is triggered from a division of the incoming gates, and Contour's decay is modulated by another. ValhallaVintageVerb.
Voice 2: MicroBrute FMd by its own LFO at max rate, with minimal resonance but high Brute Factor, through DubStation VST.
Voice 3: E352 in Morph+Phase mode, with phase modulated by Tides at audio rate. Output 1 modulates A-102 filter which is self-oscillating; that's mixed with Output 2 in Sputnik Quad VCF/VCA
Voice 4: Hertz Donut through Sinclastic Empulatrix. HD's primary osc is set to LFO rates and the modulator is getting through as FM, while also modulating its own phase just a little bit. A little green mode discontinuity is applied.
With modules arranged mostly to my liking and the arrival today of Mannequins Mangrove, my modular system is currently locked down until the release of Rabid Elephant Natural Gate. I've not shared a photo of it since May, so it's quite overdue. (This is not the patch for song 204, but while getting to know the Mangrove a bit.)
|205||Oct 15||Don't The Mountain
Just don't. Not even a little. Because next thing you know, you could accidentally the whole mountain. (And that's what we call a slippery slope argument. Because mountains.)|
Voice 1: Mangrove, with its square output multiplied by A-196 PLL & G8 clock divider and back into its Air CV. Through two channels of Sputnik Quad VCF/VCA (one with Freez on a delayed gate) controlled by System X Envelope. Presswerk, ValhallaVintageVerb.
Voice 2: Rings in Karplusverb mode, fed by pink noise from the E352 and also Tides, on separate envelopes (Contour and Function).
Voice 3: 0-Coast through Vapor, SpecOps and Replika VSTs.
|206||Oct 15||Command String
Voice 1: Hertz Donut being FM'd by Mangrove, through Contour/Dynamix; layered with Donut's XOR output (with the mod osc synced to Mangrove's square output) through Contour/Sputnik QVCFA. Panned slightly left and right, with BassLane keeping a solid center and IMPEACH and ET-301 delay VST.|
Voice 2: 0-Coast, with synced Slope feeding the audio input, with its time modulated by a RESIGN, TURNIP. Ratshack Reverb and ET-301 VSTs.
Voice 3: E352 in Morph mode, through Function/Sputnik QVCFA. Function's curve knob manually wiggled during recording. AMENDMENT 25 HIS SORRY ASS and ValhallaVintageVerb.
|207||Oct 17||Their Voices, Raised
I was thinking today about those who can't speak up for fear of retribution, and those who are pressured to speak up anyway. Here are four "voices" of modules not meant to make sound, processed in some cases by modules expecting sound and getting something a little different.|
It worked surprisingly (and a bit disturbingly) well and I expect I'll be doing more in this vein.
1: Function cycling at low rates, with positive output into Freez audio in, negative into Freez size, and EOR into Freez trigger. Freez output feeds back into Function's All input, and out through Dynamix (for compression) and Three Sisters highpass filter to remove bass from the clicks. Maschine Maximizer.
2: Function's EOC feeds A-196 PLL signal input. PLL square out clocks G8 clock divider in random mode; output 3 feeds back into PLL's phase comparator input. The PLL's square output goes to uFold Symmetry input while the Phase Comp output goes into its signal input. uFold's output is boosted in Gozinta. Maschine Maximizer.
3: EON in looping envelope mode. Its output feeds back into its own CV, as well as Maschine Maximizer and Transient Master (boosting attack, cutting sustain). As I turn Frames' level control for this channel, it's also creating audible tones picked up by the extreme amplification happening in software. None of these sounds should be happening.
4: Kermit ch1 at LFO rate (well below audio frequencies), with its waveshape modulated by ch2 at an even slower rate. G8 output 7 is feeding ch2 frequency control. Kermit ch1's bipolar output is boosted in Gozinta, and mixed with the unipolar output feeding Rings set to a low, fixed frequency with brightness and damping fully open. Transient Master (cutting attack, boosting sustain).
Entire mix goes through Boscomac Floodverb.
Brief notes for a quickly made and played, unsequenced patch:|
Clocks and clock divider poking at various envelopes, modulating other envelopes that modulate the first envelopes or other things. E352 and Hertz Donut are the sound sources. Eos and EchoMelt are the VST effects. Manual control mainly via Frames.
I'm kind of disproportionately pleased with this one.