Something's frothing in the Sonic Crayon camp

[cloudscapes]

I'm also looking for software that'll help me create interesting waveshapes and turn it into numbers I can use

You might be able to do something quickly with processing if you don't want to go the C++ route...

Considering. Hell, I'm even considering game maker since I already own and know how to use it.

I'd make a utility that prompts you to open a picture of a wave. Then just have it scan row by row, left to right. I'd just have to make sure the waveform pictures are high contrast and clean. I won't need interpolation or fancy stuff. It's just for an lfo that will be modulating the delay time of a 2399. Accuracy is not really a concern.

With Processing it should be simple enough to make something that loads an image and does simple detection, or that enables you to draw the waveform yourself.
I haven't used that in years but the syntax is super simple and it's perfect for this kind of things...

You could even modify this existing example to draw a waveform :
http://processing.org/examples/handles.html
And then save the data somewhere :
http://processing.org/examples/loadsavetable.html

I'd rather use external software to draw waveforms though. I can use illustration software and make really cool waveforms with bezier splines and whatnot, then just export it as an image. Plus I'm black belt in photoshop-fu. I really don't want to spent more time than I need if I have to make a new tool. Especially because I won't be able to beat the flexibility of using illustrator + photoshop.

[cloudscapes]

I guess my reluctance to make a waveform tool that's too complex is because I'd only need to use it for a few minutes, once completed. I just need to get waveforms I have pictures of into number data the processor in the vibrato can interpret. Maybe 6-7 waveforms. Then t'll be done, I can forget about it, which is why I don't really want to spend more than a couple hours developing something to do that.

[Wes Mantooth]

Why in the fuck did I not study computer science?

It would be so awesome to see a documentary throughout the process of creating one of your pedals. Had no idea you could just use Illustrator and PS to make waveforms to be used by software.

[oscillateur]

I guess my reluctance to make a waveform tool that's too complex is because I'd only need to use it for a few minutes, once completed. I just need to get waveforms I have pictures of into number data the processor in the vibrato can interpret. Maybe 6-7 waveforms. Then t'll be done, I can forget about it, which is why I don't really want to spend more than a couple hours developing something to do that.

Ok, got it. Makes sense. You could still probably do it quickly in something like Processing though, especially if you have a white image with one black line for example.

If your image is already setup to only have the information you need (i.e. sized so that it contains only one cycle of your waveform), you just have to iterate over the x axis and check which pixel is not white...

[cloudscapes]

I guess my reluctance to make a waveform tool that's too complex is because I'd only need to use it for a few minutes, once completed. I just need to get waveforms I have pictures of into number data the processor in the vibrato can interpret. Maybe 6-7 waveforms. Then t'll be done, I can forget about it, which is why I don't really want to spend more than a couple hours developing something to do that.

Ok, got it. Makes sense. You could still probably do it quickly in something like Processing though, especially if you have a white image with one black line for example.

If your image is already setup to only have the information you need (i.e. sized so that it contains only one cycle of your waveform), you just have to iterate over the x axis and check which pixel is not white...

That's my thinking. I'd size the width and height of the wave image exactly as how many samples I want the wavetable data to be. So if I'm storing the wavetables as 8bit data, 256 samples "wide", then I'd scale my wave image to be 256 by 256 pixels. Then have my software scan row by row. Scan row to see if a pixel is black instead of white, if it finds one, write its vertical coordinates as data for the first sample, the move on to the next row. No need to complete the row at that point because I'll have cleaned up my images properly beforehand, to only have a single black pixel per row, representing the wave.

I might just try and do this in game maker tonight. Processing might be better, but I don't know how to use it, unlike the former. I just want to use something I already know for this.

I'd use visual basic if it wasn't broken on win 7.

Back when VB still worked, I wrote a little GUI tool that allowed me to draw patterns for a led matrix display (hollow earth) and spit out data. I wrote a "display driver" for the microcontroller-->led display that used the data. Worked really well.

[Strange Tales]

Wait how is VB broken on Windows 7? I've never heard that before.

[cloudscapes]

Wait how is VB broken on Windows 7? I've never heard that before.

older versions are apparently a headache to install and use on win7. VB6 and older anyway. there are guides I've tried a couple years ago, and it caused more problems. the new ones work, visual studio and vb net and stuff. I'm not up to date with all that.

too bad backwards compatibility is so hard to maintain.

[Strange Tales]

Damn, I didn't know that about VB6, but I never really messed with it.

FWIW, a cursory google search brought up this thread, and the first response seems to work for people: http://stackoverflow.com/questions/1049 ... windows-10

Maybe you've already found that and tried it though.

[zoooombiex]

This might be another "more-complicated-than-its-worth" idea, but in terms of generating the initial waveform images that would be scanned, i could see getting some interesting results by recording a pattern sung with your voice, then exporting an image of the pitch.

i'm was initially thinking of vibrato in hindustani vocals like alankars, but obviously could do a lot besides that. e.g., quick bend from tonic pitch down two steps, then slow bend back up to tonic, hold tonic for a time, then rapid vibrato around tonic.... etc.

[Inconuucl]

It would track deviations from initial pitch. Which is not that hard really. It would be like making a tuner that keeps records of the changes, then normalizing the changes and applying them to the vibrato circuit. The depth knob would then act as a multiplier/divider to increase/decrease the change, with middle position being what you played/sung.

[cloudscapes]

Details still need to be worked out, but fundamentally it'll just be a digital LFO with interesting blending/variations bed into a delay line. An iteration or two over my past vibrato.

There are a lot of cool ideas, but I'm trying to keep it simple for this one. I don't want to mess with admittedly cool stuff that I'll probably end up not understanding. It'll be little more than my previous vibrato. Just with better control over the random/noise lfo, and probably the hollow earth-style shaping mode.

[UglyCasanova]

I'm pretty happy about necrobumping this thread. So many juicy ideas!

[Seance]

Mechanical Sun : Needs a partial redesign. Since it's a looper delay, maybe a memory address shuffler/randomizer, so the more you crank it, the more it skips around the 'tape'. Will experiment. It's mostly in software. Also want to tweak the filter/tone control.
...
I've been fucking around with a bitcrusher/ringmod/comb-filter/bitrman killer for a couple weeks. It has 15 knobs so far, which is probably stupid. Just as an experiment for now.
Goal is to make something highly textural/droney. Turn sound into the auditory equivalent of a Pollock painting.
...
vibrato...

Exciting blueprint projections of sonic intent/experimentation!

I think the "keeping it simple" path is a great one to follow.

With my own projects I know that kitchen-sinking everything
into everything leads to a heavily cluttered "back-burner" where
a preponderance of my ideas proliferate and finally coagulate
into static avalanches of self-reproach. Unlike banks, creative
ideas can become unwieldy and "too big to succeed".

In the realm of creativity and ideas everything is related
to everything. But to get things done I need to draw lines.
Firm lines—even if arbitrarily drawn between projects—help me
to actually finish things in a relatively timely fashion so that I
(mostly) don't have time enough to get in my own way.

Of course animation and sound-making aren't the same thing as
computer programming and pedal design. A pedal has to actually
"work" whereas the viability of subjective artsy stuff is... subjective.

In any event... lots of interesting ideas with bitcuddling and LFO wrenching!

[cloudscapes]

Spent the last couple hours writing a tool to help me make wavetables! It's just a prompt for me to open a 256 x 256 PNG image of a wave I drew out or copied, shows the wave in a window and thinks for a couple seconds, and then just copies something like this to the clipboard, to be pasted in my microcontroller code:

Code: Select all

const unsigned char table_sine_test[256] = {130,132,136,139,141,144,147,150,153,156,159,162,165,168,171,173,176,179,182,184,187,190,192,195,198,200,203,205,207,210,212,214,217,219,221,223,225,227,229,231,233,235,236,238,239,241,242,244,245,246,247,248,250,251,251,252,253,254,254,255,255,255,255,255,255,255,255,255,255,254,254,253,252,251,251,250,249,248,247,246,244,243,241,240,238,237,235,233,232,230,228,226,224,222,220,217,215,213,210,208,206,203,201,199,196,193,190,188,185,183,180,177,175,172,168,166,163,160,157,154,151,149,145,142,140,137,133,130,128,125,121,118,116,113,109,106,104,101,98,95,92,89,86,83,81,78,75,72,70,68,65,62,59,57,54,52,50,47,45,43,40,38,36,34,32,30,28,26,24,22,21,19,18,16,15,13,12,11,10,9,8,7,6,5,4,3,3,3,3,2,2,2,2,2,2,3,3,3,4,4,5,6,7,7,9,10,11,12,14,15,17,18,20,21,23,25,27,28,30,32,35,37,39,41,44,46,48,51,52,55,58,60,63,66,68,71,74,76,79,82,84,87,91,93,96,99,102,105,108,111,192,193,194,195,196,197};

That's all I needed!




I kind of also want to ad skew/lean control on the lfo. But don't want to add more knobs. So maybe I'll do it how the strymon el capistan does it for additional modes. Hold down both stomp switches, and suddenly your depth knob turns into a skew!

[pd~]

I know you've already solved the problem, but are you familiar with Pure Data at all? I've started to dig into it recently - the things I've been able to accomplish with minimal effort are incredibly cool. Pd makes it pretty easy to draw/generate waveform arrays of any scale and shape:

puredataroolz.png (47.29 KiB) Viewed 538 times