Hm interesting. Statistically there should be about a 1/70 chance for each BPM result, with about 80% of results skewing towards “normal” results. I wouldn’t even call myself an amature developer on a good day, so there could very well be a bug in there.
The >1000 BPM result stemmed from pushing the 300 BPM ceiling of an old electribe even further with a delay line, bringing it into tunable drone territory.