Cool stuff. Seems it is always a bit short of a full tangential 180?

Seems to be some math "problems". (?)
Or script problems... Have to look some more...
I noticed a couple of times, doing a form, deleting it, and redoing it, changed from "hex" to "round"... That seemed weird.
Divide by zero?
There is some kind of limit to how much detail gets calculated by javascript or (MoI)?...
Memory or something gets filled up, or clogged up, with very large curve and point numbers?...so that re-starting the
script just doesn't work, gets "thrown out" by MoI. But all works good if MoI is restarted.
Maybe a lack of script "cleanup"?...
(Just some unskilled guesses.)

The base equations are that of circle, or ellipse, so a closed form is calculated for a full 2*pi ...?...

Edit, changed 2nd usage of i index to a k index. Put the index start point back at zero, (like in other MoI scripts).
This got rid of the hole at the south pole.

Edit: Deleted the old script, due to poor programmer understanding of when/where to do updates to factory/factories.
Moved the factories update out of the i and j loops, and added a simple factory update. Much faster performance.

Attached improved script.

Edit, revised script in post 10 below.

Seems there is a glitch! :)

I'll have a look. Probably the index needs adjusting.

- Brian

Made some modifications to this 2 year old script, which was of learning benefit back then.
The script has a lot of variables.

Roughly speaking, for 100% of a 360 degree set of curves, the new version avoids a duplicating overlap of the first and last drawn curves, by not
drawing the zero-ith curve. It will also do a percentage of a 360 degree set of curves, for example a 180 degree set of curves, including the zero-ith curve.
The default htm uses Pilou's example, with uDensity = 9, for creation of 9 curves.

Had quite a difficult time adjusting indexes.

- Brian

Excellent!

;)