takes three arguments, aside from the first inlet, which receives the bangs sent from the metro. The first argument is the tempo for the downbeat, which is usually the same as the metro but not necessarily. The other two arguments are numbers multiplied together to create a denominator to divide the tempo by. These numbers represent the range for two random numbers. we can take one of the random numbers to determine which beats receive accenting. Bit hard to explain but the test patch will give you a pretty good idea.
There are 5 outlets:
- bang outlet for the accented beats
- bang outlet for every other beat
- float outlet for the divided tempo
- float outlet for the denominator (or how many beats in a bar)
- float outlet that counts (resets to zero when the first inlet receives the next bang from the metro)
not sure if I'm completely finished with the logic involved. Also, it's a bit fragile if the tempo is way too high. It depends on how many beats you're splitting the bar into and how fast the tempo's going. The test uses [ii]'s and [pad].
One other thing, notice there's a message [acrt $1( (short for accurate). This is used to make it so that accented beats get turned into regular beats. If you don't want any accented beats at all, set this to 1. It means the chances of the accent getting ignored is 1 to 1, 5 is the default, which means a 1 in 5 chance, so as the number gets larger, the more consistent the accenting becomes.
