If you really care about performance (maybe just in special areas of some innerloops where you need the best possible performance) you could easily localize the methods in question, shortcutting all the overhead described above and directly call them with your tables. I just want to add that the ab...

This would probably be good to report in the issue tracker: https://bitbucket.org/rude/love/issues

(Looking at the properties of the 11.2 love.exe reveals that the file version value is 11.1.0.0. I assume the LÖVE team just forgot to update the value when compiling the exe.)

To make sure one sound is played exactly after another with no interruption you need to use a queueable source. (Your example actually plays seamlessly for me but that's just luck.)

Cool! I read through it. There's one thing in step #4 I think is a bit confusing/misleading: //this is how to use the luapreprocess as a code block, remember the syntax ! to denote the start and end of the code block to be preprocessed An ! without a parenthesis simply means the rest of the line is ...

I'm not sure if it's the problem, but that image uses 16 bits per sample instead of the more common 8 bits.

joystick:getGamepadAxis() returns a number, not a boolean, which is why both prints trigger. Don't bother using love.joystickaxis() if you just want to know the current value of a joystick axis in love.update() - simply call joystick:getGamepadAxis() inside love.update(). love.joystickaxis() gets c...

If joystick:isGamepad() returns false then you'll need to bind buttons/axes using setGamepadMapping() before any of the gamepad methods will work.

Awesome! I'll have a look at that blog post.

Are you maybe drawing outside the 5000x5000 canvas? I don't see any other limits in the code.

1. You can make .apk files on any operative system.
2. Please share your .apk file so people can help you find the problem instead of blindly guessing what the issue is.

The wiki may help you, if you didn't check it already.