zerox~

[porres]

frequency estimate is bad... block shouldn't affetct count?

[mrufrufin]

looks like it only counts positive to negative or negative to positive (meaning > 0 to = 0 and =0 to > 0 aren't counted but < 0 to = 0 and = 0 to < 0 are). I suppose this is fine for most signals, but signals strictly >= 0 (such as [phasor~] but NOT [phasor~] * -1) would register as 0 frequency. I suppose this could be technically correct(?) since these signals don't cross zero, but it depends on what Max does. Also, this goes by block (which seems to be what Max does as well according to the documentation) so that could be another explanation for poor frequency estimate (which zero crossing isn't that great for anyways...)

[porres]

I suppose this could be technically correct(?)

I think so. For what I remember, Max has a better estimation output, this is why I should have opened this issue, but I'll double check (not a high priority right now though).