MicroPython Forum

No. You could use one of the Python routines in algorithms.py in my repo, but even the "fast" non-recursive one will be ~100 times slower than one written in assembler.

@Mike Teachman Your two design constraints refer to the maximum (T1) and average (T2) time in which "other coros yield to the I2S task". Each task actually yields to the scheduler. So when the I2S task yields, every other task will get a chance to run. So T1 is the sum of the maximum time between yi...

gwork wrote: ↑

Sun Dec 09, 2018 2:05 pm

Thanks for the suggestion. What is the difference between uasyncio and threading? Is uasyncio non-preemptive? If that's the case, it will probably work fine for my usage.

See this doc and this one.

It is not on PyPi. Nor will it run on the ESP32 as it is written in ARM assembler. The library is strictly for STM devices such as the Pyboard. I'm not aware of an index for PyPi but packages on there are generally prefixed micropython . Many user-contributed drivers are not on PyPi. The best way to...

Currently I2S on the Pyboard is unsupported.

The work I've done on enabling ESP8266 devices to work for long periods has convinced me of the following: Assuming good quality hardware it is possible to write applications that never crash or produce error messages. Such applications can survive WiFi outages without loss of program state. Brief W...

I recommend uasyncio unless you really need threading. Reasons why, and a general tutorial on its installation and use, may be found in this repository.

Interesting that sleep_ms() is consistently slower than sleep() despite the latter's use of floating point. I stand corrected on that ;) I have doubts about uasyncio coping with i2s. It depends on how the i2s code is implemented. uasyncio can handle data streams on UARTs because the UART interface u...

I'm not familiar with piezo vibration sensors but, as @Roberthh says, piezoelectric devices are extremely high impedance. They can also produce high voltages. To interface one to an ADC there needs to be a DC bias so that the mean level is in the middle of the ADC range. I would also want to see vol...

For a nominal zero delay it is quicker to issue asyncio.sleep_ms(0) and quicker still to issue yield . However whenever you issue one of these, every other coro will get scheduled before it returns. Note that the Barrier runs a (minimal) coro while a task is waiting on it. These delays soon add up. ...