Kaspar

In below patch, how come that every time I tap a beat that is in time with the metro pulse beats, I also have an indication that this same beat is out of time??

timing-a-tap-with-a-beat

comment: I would like to also see if the person taps within 50 ms _before_ the metro beats, not just after, why I use two timers, but using two timers seems to be a problem ...

This should get you started. See the explanation inside the patch.

Hi, and thanks fpr the solution. It works :-) However, what of it was a varying metro value that you tapped in time with: a rhythm - then there would be a varying time length between the beats that you could tap in time with ... then how do I calculate the time to the next coming beat, if the time between the beats vary?

Actually, an approach to tackling this next problem was already presented in the last patch.

I just followed it through in this patch below.

Actually, an approach to tackling this next problem was already presented in the last patch.
I just followed it through in this patch below.

Hi, and thanks again. Yes I see that you can always hard code yourself out of it, but I need a more flexible solution ... For example, what if the pulse rate changes entirely? Then you cannot subdivide it into a specific set of subdivisions. I think I will try out something completely different ...

I might not have the right idea of what you're going for... but in theory, most tap tempos take in a buffer of taps and average the time across the entire buffer.

Tap 1... Tap 2 (now take [[tap1-tap2]/2], and store it in an array)...

Tap 3... (now take [[tap3-tap2]/2], and store it as the next element in the array)...

... Now calculate the averages via: (array[0] + array[1]) / 2

and you should get the average time in ms between taps, constantly varying and increasing in accuracy during each tap. Then with each following tap, replace the 1st, then 2nd elements in the array, etc.

Note that you could even change the amount of elements for more accuracy.

i.e.:
Tap 1... Tap 2 (now take [[tap1-tap2]/2], and store it in an array)...
Tap 3... (now take [[tap3-tap2]/2], and store it as the next element in the array)...
... Now calculate the averages via: (array[0] + array[1]) / 2
and you should get the average time in ms between taps, constantly varying and increasing in accuracy during each tap. Then with each following tap, replace the 1st, then 2nd elements in the array, etc.

Note that you could even change the amount of elements for more accuracy.


timing a tap with a beat