a lo-fi, bytebeat-based synth

[Numanoid]

just pasting junk as said, without start vals totally meaningless but maybe the forms will save some fingerlegwork. a few ppl had nice webpages and vids on this. have fun!

Intriguing

I foresee sometime in the future that the cut up technique (as deployed by Burroughs et al) will be applied to code

Who knows which plugs we will be in for then

Hopefully not just syntax error...

[bitcrusher]

OK, here's the OS X VST, in case anyone wants it / for however long it survives dropbox quota - obviously you're welcome to include that in your main archive.

https://www.dropbox.com/s/s77v8m71xinid ... 1.zip?dl=0

[xoxos]

works here (xp), left it to the last moment for the last two days and didn't feel like putting it in my ear before sleeping so figured i ought to test and report

lots of ideas for this kind of method, eg. using shorts instead of ints for longer play, textures, window selection. there are a lot of ways these equations can be written and there is weird stuff to be heard.

[xoxos]

i should add,

with all due respect given the enthusiasm and effort of other developers,

every kind of chiptune vst i've seen to date, afaik, uses antialiased squarewave oscillators instead of .. well, actual integer math.

this vst doesn't really give the end user insight into how weird and wonderful (well, for digital audio) the method is.. (probably doesn't really happen unless you are a developer). if you have ever done a lot of math or procedural stuff, you know that sometimes you ...find things.. that are kind of weird. some of the possible sounds are so exotic and diverse, i doubt too many listeners would be able to identify the synthesis method.. it's "just a range of numbers that happens to do *this* in audio" ... like every sound in the world is packed away in this process, if you can just find the right numbers..

[foosnark]

I mostly chose algorithms that were capable of relatively conventional sounds, though some start to get weird at some values (usually if B is low and not an integer). Really, to get a better understanding of bytebeat you have to just plug in formulas yourself, listen and watch an oscilloscope. I'd say the "Sierpinski harmony" is much more representative of bytebeat than the use I'm putting it to.

One of my favorite musical toys is the Olegtron 4060 -- basically just an oscillator and clock divider that you patch yourself with resistors, capacitors, diodes, transistors, etc. It's sort of the analog equivalent of bytebeat. There's definitely a logic to it, and with some experience you are more likely to be able to reach certain kinds of sounds on purpose instead of stumbling into them. But perhaps even more so than with bytebeat, it's not a friendly way to create a specific rhythm or sequence.

[Ichad.c]

Might be of interest: http://legacy.spa.aalto.fi/dafx08/papers/dafx08_13.pdf

He also has some other papers of interest too: https://mediatech.aalto.fi/~jkleimol/

[sweet_trip]

This is awesome!!!

I made a little bytebeat thingy on Java a while back and I've meant to turn it into an Android app but got beat to it. I want to learn VST dev and I thought doing a bytebeat machine would be a great way to learn, so I welcome the source code.

[foosnark]

Glad it's helpful

I'm working on another lofi project. This one was inspired by the not-yet-released Sonic Potions Malaclypse. I used Perlin noise extensively in my game development career for terrain, vegetation distribution and particle systems but had never thought to apply it to audio, so that one got my attention.

As it turns out, rather than actually using Perlin noise, the experiment morphed into table noise that mutates over time, more like an audio rate Turing Machine module. With a zero rate of change it's a pitched oscillator with a randomized wavetable, and at maximum change it produces white noise -- so of course the fun stuff lies in between. It sounds a little bit TIA/POKEY-ish.

I plan to play with some other change algorithms besides pure randomness (maybe Perlin noise will sneak back in that way, maybe not), then I'll release it as open source too.

[xoxos]

http://xoxos.net/vst/mp3/float.mp3

[foosnark]

http://xoxos.net/vst/mp3/float.mp3

Hell of a lot smoother than what I'm doing

[xoxos]

i was thinking it was about time i posted a thread on this, i'll post here since i dig your thread -

just an example of what happens when you start moving out of the sandbox. float is okay for some string pad stuff, where i like it is noisy tonalish percussion, like thrashing a thousand spears in the jungle. anyway.

add another dimension, use a quadrature osc to draw circles on it then you get that variance you mentioned by moving the circle. simple math (move center, scale radius) for stereo et c.

this industry is so deliverance

[foosnark]

Here's the noise synth. 9 algorithms, some of which don't make a lot of sense but could be musically useful to somebody weird enough.

http://starthief.net/Fnord.zip

[xoxos]

prugin no loady

[foosnark]

Weird. It's working for me in Maschine, which is a bit more likely than most hosts to barf on plugins it doesn't like.

[xoxos]

could be an XP thing (host = synthedit, pretty friendly). couldn't load module or something, needs more reports