Sonic Crayon FX now has a FB page. Should make updates and info easier..
http://www.facebook.com/pages/Sonic-Crayon-FX/230721366984881
Saturday, October 8, 2011
Friday, June 3, 2011
HE v2 "sampling" mode brainstorm
Is a glitchy "sampling" mode on the new HE better than no "sampling" mode at all? The intent is to have a mode that records sound input and turns it into an envilope. Here's how I'm doing it so far:
20 sub-samples are taken from the input at 5ms intervals, then averaged. the average becomes the main "sample" that lasts 100ms. this occurs 10 times a second.
so it samples at a total of 200hz. the problem is when I hit a note like low E (82-ish hz), or any note really, that the played note will get pretty close in-phase with my sampling frequency from time to time as it "drifts". because of that, for an entire 100-200 milliseconds, it'll only be sampling the "dips" in-between the sound waves and will average close to 0, even though to our ears the note rings loud and clear. as clear as it was when it was sampling at the peaks of the same waves 300ms earlier! the result is a stuttery glitchy envelope instead of smooth envelope following.
adding circuitry is a last resort (there's barely any room left on the PCB). so far I've tried:
1. adding offsets of 2.5ms every 2 samples to try and average an equal number of wave dips and peaks.
2. only keeping the very highest sample out of 20
3. ignoring sudden drops of envelope if the next one goes right back up again (loss of detail)
Increasing the sampling frequency is difficult at the moment since the whole foundation of the bare-bones design is built around 20 sub-samples per "real" sample, because of the interpolation I'm doing. I think I'll try adding a timer interrupt that samples much more/faster and independantly of the rest of the program.
20 sub-samples are taken from the input at 5ms intervals, then averaged. the average becomes the main "sample" that lasts 100ms. this occurs 10 times a second.
so it samples at a total of 200hz. the problem is when I hit a note like low E (82-ish hz), or any note really, that the played note will get pretty close in-phase with my sampling frequency from time to time as it "drifts". because of that, for an entire 100-200 milliseconds, it'll only be sampling the "dips" in-between the sound waves and will average close to 0, even though to our ears the note rings loud and clear. as clear as it was when it was sampling at the peaks of the same waves 300ms earlier! the result is a stuttery glitchy envelope instead of smooth envelope following.
adding circuitry is a last resort (there's barely any room left on the PCB). so far I've tried:
1. adding offsets of 2.5ms every 2 samples to try and average an equal number of wave dips and peaks.
2. only keeping the very highest sample out of 20
3. ignoring sudden drops of envelope if the next one goes right back up again (loss of detail)
Increasing the sampling frequency is difficult at the moment since the whole foundation of the bare-bones design is built around 20 sub-samples per "real" sample, because of the interpolation I'm doing. I think I'll try adding a timer interrupt that samples much more/faster and independantly of the rest of the program.
Thursday, May 26, 2011
Hollow Earth prototype
v2.0 prototype. It works really well! Things left to do on firmware is the sample-from-sound-and-turn-into-LFO mode, a dedicated random mode, and some tweaks to the simple random blend you can add to existing LFOs.
Thursday, April 28, 2011
Mockup
1:21 AM blog post! Just a graphic mockup of what I think the new hollow earth will probably look like.
Oh yeah, just got back from seeing Godspeed You! Black Emperor, that was pretty fantastic!
/random
Sunday, April 24, 2011
Hollow Earth redesign proto
Found the time to (mostly) finish the new prototype this weekend. What's left to be done is ironing out the few glitches and adding a couple modes in software. Specifically:
-menu to assign a function to the soft-stomp.
between tap tempo, shape record and manually triggered envelope
-a mode where it samples and makes an lfo/envelope based on
input sound. could be pretty rad for drum machine stuff!
-I should probably make an option to invert whatever shape
...oh yeah, and definitely what's my densest/most insane PCB layout. It'll take a few weeks to arrive, though.
Sunday, March 27, 2011
1165 lines of code..
..though a bunch of it was copy pasted. I figure I'm about 95% done the new hollow earth, at least programming-wise. Hardware-wise, an AVR dev board, a few rotary encoders, a trimpot, an led and a 5x7 led matrix is all I've been staring at these past few weeks.
Four knobs is probably all you'll need! They're all multi-function. Turn for one thing, hold down and turn for another, tap for another, etc.
A bit of tech info. I'm using an atmega644 uC. It's got 64KB program memory but so far I'm using only 17% of it. I needed the 2KB of eeprom it offered for preset storage. When recording custom LFO shapes, it samples at 10 a second for up to 18 seconds. And then it's interpolated times 20 to smooth it out. Speed is full adjustable while recording and after, obviously. 10sps is plenty for slow human hands, but I'll offer the user to fine tune this in a sort of "preferences" mode.
Let's see, checklist. What's done, and to-do..
(software)
-expressive display
-hook up PWM output (with freq doubling)
-do custom LFO record/playback
-x20 interpolation!
-speed control
-tap-tempo
-depth control
-random blend
-preset saving/loading
-square/triangle/ramp/sine/saw hard preset LFOs
-shape edit
-trigger mode
-multimode tap footswitch
-preferences/options
(hardware)
-LDR biasing
-power filtering
-clean tl074 buffer
-CV out
-polarity protection
Four knobs is probably all you'll need! They're all multi-function. Turn for one thing, hold down and turn for another, tap for another, etc.
A bit of tech info. I'm using an atmega644 uC. It's got 64KB program memory but so far I'm using only 17% of it. I needed the 2KB of eeprom it offered for preset storage. When recording custom LFO shapes, it samples at 10 a second for up to 18 seconds. And then it's interpolated times 20 to smooth it out. Speed is full adjustable while recording and after, obviously. 10sps is plenty for slow human hands, but I'll offer the user to fine tune this in a sort of "preferences" mode.
Let's see, checklist. What's done, and to-do..
(software)
-hook up PWM output (with freq doubling)
-do custom LFO record/playback
-x20 interpolation!
-speed control
-tap-tempo
-depth control
-random blend
-preset saving/loading
-square/triangle/ramp/sine/saw hard preset LFOs
-shape edit
-trigger mode
-multimode tap footswitch
-preferences/options
(hardware)
-LDR biasing
-power filtering
-clean tl074 buffer
-CV out
-polarity protection
Saturday, March 19, 2011
led matrix visual feedback
clicky encoders are annoying. ordered some non-clicky ones.
with the push-for-secondary-function knobs and the display now offering an expressive level of visual feedback, I've been able to simplify the next hollow earth pedal to just 4 push-knobs and one toggle at the back of the enclosure!
this was a huge priority because it means I can PCB mount everything on a single board this time!
with the push-for-secondary-function knobs and the display now offering an expressive level of visual feedback, I've been able to simplify the next hollow earth pedal to just 4 push-knobs and one toggle at the back of the enclosure!
this was a huge priority because it means I can PCB mount everything on a single board this time!
Sunday, March 13, 2011
LED matrix app
So I got myself a couple tiny 5x7 LED matrix displays to tinker around with. I thought they might be a nice replacement to 7-segment displays. On the new Hollow Earth, I'm using a microcontroller with a greater number of pins, so I think it's doable. Might make for some neat animation, too!
Then my mind caved in at the thought of manually typing binary data for each PIXEL. It's pretty daunting to do by hand, because each LED in the matrix does NOT have its own pin. Instead it's a row and column array where you have to carefully "render" the LEDs, line by line, applying positive power to cathodes where you *don't* want pixels, and so on. Arrrrghhhhhhhhhhhhhhhhhh..........
Too much, I said, so what do I do? I install something I haven't used in like ten years, Visual Basic 4.0.
An hour later, I have made an app where I can draw a number, letter or graphic just by clicking on squares, and then automagically spits out code and/or raw binary data I can simply paste in my micro!
Whatta time saver!
Writing a program that writes code for you. How zen!
Then my mind caved in at the thought of manually typing binary data for each PIXEL. It's pretty daunting to do by hand, because each LED in the matrix does NOT have its own pin. Instead it's a row and column array where you have to carefully "render" the LEDs, line by line, applying positive power to cathodes where you *don't* want pixels, and so on. Arrrrghhhhhhhhhhhhhhhhhh..........
Too much, I said, so what do I do? I install something I haven't used in like ten years, Visual Basic 4.0.
An hour later, I have made an app where I can draw a number, letter or graphic just by clicking on squares, and then automagically spits out code and/or raw binary data I can simply paste in my micro!
Whatta time saver!
Writing a program that writes code for you. How zen!
Wednesday, March 2, 2011
Redesigning a fundamentally awkward design..
so yeah..
I've started a complete Hollow Earth redesign. I really want to see it being sold again and the old design was just so difficult. Being on two PCBs, socketed ribbon cables, wired switches, hacks and tricks giving each build it's own little quirks. I've had enough of all that. The old one was designed when I was just beginning to understand microcontrollers almost two years ago and I've learned a lot since. Time to do it right, this time.
Instead:
-single PCB design
-everything PCB mounted, except jacks
-rotary encoders instead of pots, with double functions
(turn does one thing, push and turn does another, or just push, like tapping the speed knob for tap tempo)
-less parts (no "shape" stompswitch for one)
-in-program interpolation, no more smooth knob
-and so on
Probably no more vactrols either. They were a pain to bias properly so that CV-out and the actual tremolo effect played nicely together. Gonna try digital potentiometers instead.
UPDATE: I'm afraid trigger mode will probably go the way of the dodo. It never worked THAT great, and it's one of the major roadblocks in my simplifying.
I've started a complete Hollow Earth redesign. I really want to see it being sold again and the old design was just so difficult. Being on two PCBs, socketed ribbon cables, wired switches, hacks and tricks giving each build it's own little quirks. I've had enough of all that. The old one was designed when I was just beginning to understand microcontrollers almost two years ago and I've learned a lot since. Time to do it right, this time.
Instead:
-single PCB design
-everything PCB mounted, except jacks
-rotary encoders instead of pots, with double functions
(turn does one thing, push and turn does another, or just push, like tapping the speed knob for tap tempo)
-less parts (no "shape" stompswitch for one)
-in-program interpolation, no more smooth knob
-and so on
Probably no more vactrols either. They were a pain to bias properly so that CV-out and the actual tremolo effect played nicely together. Gonna try digital potentiometers instead.
UPDATE: I'm afraid trigger mode will probably go the way of the dodo. It never worked THAT great, and it's one of the major roadblocks in my simplifying.
Subscribe to:
Posts (Atom)
