I fixed Pedro's siteswap transition generator:
This was a much celebrated generator when it came out & a few people have asked for it to be fixed, but now that I've had a good look at it I have to question why.
It has an interesting method of finding transitions that I don't fully understand (the problems I fixed were all to do with external code from https://www.jonglage.net not Pedro's original code) but it doesn't find many of the possible transitions. For an n prop siteswap it is limited to finding at most n! transitions per period.
To give an idea of how limited this is using all transition length options for 801 to 504 only returns 47 different transitions (periods of the results range from 2 to 12 & throws go up to height F), whereas a quick & dirty generator I just wrote returns 130 just for periods 2 to 6 without exceeding a 9 throw1.
Does anyone ever use transitions greater than the shortest possible anyway?
I think I may put a bit more work into my generator...
1 Before Peter Bone says his generator returns 138, my one doesn't include transitions that pass through the initial state of the target pattern more than once :P
For passing siteswaps, it can be desirable to find a transition with an even or odd number number of transition throws to determine which person starts the new pattern / who does what in an even period pattern.
I hope that counts as a reason for longer-than-the-shortest-possible transitions.
PS - Hi. I haven't been around these parts for a while.
That makes sense. I've added some widgets to my transition generator to filter the results by transition length parity & also plain text/regexp
Subscribe to this forum via RSS
1 article per branch
1 article per post