charliebruce.bsky.social
charliebruce.bsky.social
@charliebruce.bsky.social
I've embedded an entire .ogg file in firmware rather than stripping out the raw Opus data - not worth the extra hassle for the sake of a few hundred bytes of flash saving on the metadata/page information.

Also makes life easier for hackers/modders if they know what data they're looking at ;)
December 14, 2025 at 5:27 PM
December 14, 2025 at 4:15 AM
You also likely don't need to measure everything at the full sensor rate - measuring the wheel crank signal at 10-20 Hz would likely be totally fine. Only motor commutation needs high timing accuracy, but even then there's enough inertia that you wouldn't expect massive change cycle to cycle.
November 2, 2025 at 4:26 PM
You might be better off with a microcontroller family with a more capable timer like an STM32 instead of the ATtiny unless there's a particular reason to use them? You get an order of magnitude higher performance very cheaply. Or consider analog circuit and ADC reading the signal instead?
November 2, 2025 at 4:23 PM
This is pretty much the intended use of a timer/counter peripheral - start counting on rising edge, stop counting on falling edge, then either interrupt, or write the counter value to an address via DMA to be handled less frequently. Chip family specific, but softens timing requirements.
November 2, 2025 at 2:00 PM
Hoping that you'll just give them a free one to avoid the fuss? Using the fake email to suggest that you're going back on your word? Pretty amateur but maybe that's their plan.
September 10, 2025 at 5:14 PM