Show newer

Bizarre faust problem. This is a clock quantizer: a trigger at the first input is delayed until a rising edge at the second. But it only works if the first input is irregular, if it's a steady clock or train of impulses only the first works. Add in an irregular impulse and one comes out on schedule.

quant(in, clock) = loop ~ _
with {
clock_rising = clock <: _' < _;
loop(feedback) = ba.latch(in | feedback, 1 - feedback) * clock_rising;
process = quant :, 0.001);

dried boosted

Mal was für alle Instagram-Benutzer: Schmeißt die App ASAP von euren Geräten. Marcus Mengs zeigt hier mal kurz was für Daten die App aus eurer Umgebung absaugt und an meta/facebook übermittelt.

#instagram #meta #facebook

I wanna jam with somebody
I wanna feel that beat with somebody

dried boosted

BTW I have been compiling a large list of resources for learning audio DSP, if anyone is interested:

I found a way to do the rhizoming but it's miserable because routing in Faust is like pulling teeth. Also I think this kind of combinatorial thing is the kind of thing I'd've learned to do if I actually went to school for computer science (instead, it took me quite a while). Anyway, I'll convert it to Rust, and hopefully it'll be in cv-utils sometime tomorrow.

Show thread

#faust code is pretty cryptic but here it is if you're curious 

hyst = 0.001; // 1ms rise, 1ms fall
gt = _ >= _;
new_cross = gt <: _ xor _';
process = new_cross :, hyst);

Show thread

Added intersection detector to cv-utils. It just outputs a trigger when the two input signals cross.

The algorithm is *way* better than the one in turns' quad-lfo, so I'll probably remove it from turns, but I need to find a way to rhizome-ify it (so it can compare every input to every other input for any number of inputs)

dried boosted

If we want social, environmental and economic change to happen, we also need to consider the role of electronics and digital communication.

Kudos to radioshack of yesteryear for making this little amp super easy to disassemble, clean the pots, and reassemble. Excellent $5 secondhand purchase.

more notes 

- I'm looking forward to getting to the more algorithmic part of supercollider
- tuning in to @draco's supercollider stream now to absorb some scsynth knowledge 💡

Show thread

more notes 

- sc feels a little verbose. the proliferation of brackets & parentheses has always bugged me, but creating busses for each signal redirection feels like a lot. there's probably a trick for that which I have yet to learn.
- preamp:
- spectral compressor-expander is PV_Compander from the sc3-plugins package. I'm not sure how the arguments work but the examples are useful.
- the handling noises aren't very audible in the transducer output (and it's great)

Show thread

slow developments in transducer feedback land:

- I started using a different preamp for the piezos and it's great
- I started using supercollider and it's great
- The spectral compressor-expander I was using in pd has a direct analog in sc and it's great
- I started using a plank of cedar(?) which I think will be in the next box and it's great

here are four bandpass filters roaming over resonant feedback nodes on the plank, bathed in a swath of reverb

(noisy enough?)

patch notes 

There's a lot going on, but basically: a quad resonant filter acts as a complex oscillator (each pair of filters FMing the other), sent out through a limiter, VCA, into mixing section of Cold Mac and out to the computer. Cold Mac has a feedback path through its wavefolder back to the left filter input, and the limiter is also going back to the right filter input.

I hate patch notes that are just gear shopping lists, but Cold Mac is so unique that you can't really talk around it.

Show thread

I need to stress test my analog gear. Fortunately it's . Here's a mess of feedback paths (no squeals, just noise).

I like feedback conceptually. Noisy feedback patching isn't really my sound, but it's pretty fun to play with.

Can't reproduce the clicks after a brief attempt.

Tons of line noise though, I'm assuming it's partly because I'm using the relatively dirty 5V from the Pi's GPIO. The noise increases when the SSH session to the Pi is busier (eg, with X-forwarding) -- not sure if that's RF interference or increased power draw to the Pi. Anyway, the goal is to run headless so those spikes basically don't matter, though I wouldn't mind cleaning up the overall noise floor.

Show thread

Some quick drone, the process of which is marginally more interesting: these were output from a pd patch running on a raspberry pi, into the maple box, and taken from the piezos out into my interface. The important part for me is that the whole thing -- Pi, Pi's interface hat, 20W amp, and homemade input buffer -- ran on a battery that only puts out 5V at 2A, even while I was SSH'd to the Pi.

Now to figure out those weird clicks...

dried boosted

I got zero responses last time I hit up the Fediverse offering paid creative work, but I'm not giving up!

Some #Owncast features being worked on require iconography. And while we could use an icon library or something like Fiverr it'd be nice to work with an illustrator on the Fediverse who focuses on icons to have something that is custom.

This is paid work, so please boost and tell your friends.

Feel free to message me on the fediverse or email me:

slightly rushed drone. Ringmodulated square waves through slowly modulated filters. Some nice subtle emergent artifacts.

dried boosted

What's your primary OS?

Boosts appreciated for sample size :)

Show older

SoNoMu (Sound Noise Music) is a mastodon instance for musicians, sound-artists, producers of any kind of aural noise, songwriters, bedroom producers, sonic manglers and algorave livecoders. -> more...