schollz / pikocore

source for the pikocore drum machine

pikocore

pikocore is a hackable, open-source, lo-fi music mangler based on the Raspberry Pi Pico.

read more here: https://pikocore.com

usage

prerequisites

First install Go and then install pre-reqs:

make prereqs

This will install clang-format, cmake, the pico toolchain, gcc, python, and other useful packages.

create audio

You can use the default audio by just running

make audio

The audio is taken from the audio2h/demo folder. You can edit the Makefile to choose a different folder.

build

To build just run

make

If you are using a 2mb pico, then you should do make build2 (the default is 16mb).

Then upload the build/pikocore.uf2 to your pico.

customization

If you want to turn off the LED, change WS2812_ENABLED=1 to WS2812_ENABLED=0 in the target_compile_definitions.cmake file.

If you want to use MIDI instead of clock in (requires itty bitty midi) then set MIDI_IN_ENABLED=1 in the target_compile_definitions.cmake file.

dev

You can open a minicom terminal by running make debug after switching on DEBUG_X flags in main.cpp.

Easing functions generated with: https://editor.p5js.org/schollz/sketches/l5F_ZWjZM

pikocore's Issues

Flash save is unreliable with a slightly-old battery.

When using pikocore with a rechargeable AAA battery I often found that trying to save my pattern causes the board to hang. And it seems to be the same kind of hang I get when the battery is finally too low to power the pico. But the hang-while-saving happens much sooner in the life cycle of the battery, when everything else still seems to be functioning well. Meanwhile, I never see this happen when piko is powered by USB.

I have been experimenting with this using my bench power supply to simulate the battery. It looks like when voltage gets below 1.4v, writes to flash become a bit flaky; they start, but don't always finish. When it gets down to 1.3v, it's pretty unreliable, and can even corrupt flash to the point that I can't boot until I re-flash the firmware. The rest of piko still operates fine at those voltages.

Saving to flash appears to raise current consumption by 5-6ma. (Although maybe more than half of that is used by the RGB LED? It seems to use roughly 1-3ma, depending on color?) The problem seems triggered by low voltage more than low current, but on an AAA one will lead to the other I think.

images

stop/play only half-working, depends on button order.

The docs say that you can start & stop Pikocore by pressing the four outer buttons at once. I've been wondering why this sometimes does & sometimes doesn't work.

It appears to me that the actual behavior depends on which of the four buttons is the last to be pressed. If the upper-right or lower-left button is the last of the four to be pressed, then Pikocore stops & starts. But if it's the lower-right or the upper-left button, nothing happens.

(FWIW this is with a custom (as in custom samples) v1.3.0 firmware, generated with the tool on your site.)