ilovefuzz.com

UglyCasanova wrote:Of course. Really like the shitty-record-player sound as well. I still think random wave with medium depth, low mix and a tad of crackle on the Vessel's End is one of the most beautiful things I've heard. I know you had some second thoughts about that crackle knob, but personally, I've used it a lot live. Up till noon for degradation and past noon for wave peak rhythms. If you don't mind me asking; what causes that beat/pop sound when you take the parameter past noon, from a technical standpoint? So lovely and fun to play over.

random is so cool, I've started experimenting with it last night. it's a dedicated knob where noon is no random, turn it one way and it blend-in high frequency randomness on top of the existing waveform (full knob is full random), turn it the other way and it blends in low frequency random. set it back to noon for none. easy. having problems with the math, though. microcontrollers aren't the best at generating random numbers.

I like the idea of crackle/noise, I just don't like my implementation of it, or the sound of how I made it. it sounds cheesy, like a cowbell, or chiptunes. in the old vibrato, it's digitally generated, the same microcontroller that does the flo also generates noise and pulses which are just blended in.

eventually I want to build a dedicated noise/crackle/rumble synth. way more than just white noise. but that's some heavy math that I'm not ready for yet

Or lots of distortion and layers of heavy filtering on white noise. At least that's what I've been working on right now in audulus.

dunno if filtering will do what I have in mind. I have kind of these types of nosie I can imagine that's way more than filtered noise. like how thunder sounds after the initial crackle, and it fades into percussive low frequency... rumbles? it's even lower frequency than rumbling, it's more a sudden movement of air, an envelope, than an actual frequency. then I want to be able to blend that with higher frequency noise, modulate it with procedural envelopes that are pattern-ish but not repetitive. crazy stuff.

a weird reference is I don't know how to describe it. like when you tighten muscles around your ears/around your jaw, your ears pick up that super low frequency rumble that sounds like if you stuck your head under a waterfall. an uneasy unstable pitchless rumble that breaks up randomly. I want to play with that kind of noise.

cloudscapes wrote:dunno if filtering will do what I have in mind. I have kind of these types of nosie I can imagine that's way more than filtered noise. like how thunder sounds after the initial crackle, and it fades into percussive low frequency... rumbles? it's even lower frequency than rumbling, it's more a sudden movement of air, an envelope, than an actual frequency. then I want to be able to blend that with higher frequency noise, modulate it with procedural envelopes that are pattern-ish but not repetitive. crazy stuff.

a weird reference is I don't know how to describe it. like when you tighten muscles around your ears/around your jaw, your ears pick up that super low frequency rumble that sounds like if you stuck your head under a waterfall. an uneasy unstable pitchless rumble that breaks up randomly. I want to play with that kind of noise.

Take. My. Fucking. Money.

What's your problem with the maths exactly for the random/noise ?
Not sure I can help but just in case...

That sounds like the abuse of physical modeling synthesis. Do like. :O

oscillateur wrote:What's your problem with the maths exactly for the random/noise ?
Not sure I can help but just in case...

my problem is I'm just bad at math. I don't have the brain for it. I have to do it because its a means to an end.

and also doing math on microcontrollers has restrictions. they can do calculus/trigonometry stuff, but not at audible rates. stuff like sinx/x, turning data into a sine waveform, it's better to fake it on the microcontroller. and that has its own sets of challenges. because I can't lock up the microcontroller for more than a few microseconds (millionths of a second) doing calculations, or else the codec (chip that does audio conversion) shits the bed.

guess what I'm saying is it's bad enough that I'm bad at math, but I'm also bad at math and using hardware that makes it even more difficult than it should be. like you're doing a math test, but you also have to work on wet paper and a thick sharpie.

Hahah, get it .
Isn't it simpler to generate lookup tables for functions like that with this kind of platforms anyway ?

oOohh... hello! I just snagged a Vessel's End, and it makes me so very happy. I wish I could find more of your devices out there.

I read cloud's posts in the voice of a Canadian DMX fan.

oscillateur wrote:Hahah, get it .
Isn't it simpler to generate lookup tables for functions like that with this kind of platforms anyway ?

yup, that helps a lot!

sorry for the zombie bump, but I was wondering if this ended up materializing? maybe I missed the release, but it's still very intriguing

It's boxed up, though unpainted ugly grey box. Most things are working but there are weird firmware bugs that I haven't been able to fix. Mainly having to do with how some knobs are "dual function" when you hold down one or the other stomp switches, and how the firmware recovers from getting out of that, all while playing nice with the tap-tempo part of the software. a lot of head fuckery.

won't be doing modulation of the LFO as I initially wanted after all. that much "knob multifunction" management on top of my current issues would probably make the whole program implode on itself.

oh, pic of the guts

and the unfinished firmware, for those who are curious and machine-inclined

http://dronecloud.org/stuff/vibrato.html

gets injected in the big chip at the bottom of the board

doesn't show my custom adc reader (for potentiometer input) and a couple minor library functions