It isn't slow. If all you're doing is reading, then the overhead will be pretty much the same as if you were using the Espressif SDK, because you'll find the bo

It isn't slow. If all you're doing is reading, then the overhead will be pretty much the same as if you were using the Espressif SDK, because you'll find the bottleneck is I/O, not the extras that MicroPython provides. A cold boot is slower, and takes a couple seconds, but after that most of the latency is comparable to if you were doing it with the SDK. MicroPython is heavily optimised, and doesn't supply a lot of the mutable madness that makes CPython so slow. (For example, a for loop in CPython is slower than a list expression. In MicroPython, they're the same speed.)