Grasshopper

algorithmic modeling for Rhino

Hey
I'm doing some unroll tests right here, and there is something very basic going very wrong?
it turns out a cylinder can't be unrolled properly ...any ideas?


Views: 630

Replies to This Discussion

Hi Ivan,

unrolling is one of those operations that is only solved to within accuracy. Ok, cylinders are a special case and could be unrolled easily to a very high accuracy, but other surfaces cannot. If your document tolerance is 0.01, then 282.743 and 282.742 are the same number within tolerance.

Also note that if you increase the tolerance, all you'll get is more overlapping digits, the comparison will probably still be False. The chances of two floating point numbers being exactly identical if they are the result of two different operations are very small indeed. Almost all arithmetic operations in the computer carry some sort of fuzz in the least significant digits, and an Area computations contains a lot of arithmetic operations.

--
David Rutten
david@mcneel.com
Poprad, Slovakia
Speaking of fuzz in digits, I'm not sure if that was to blame, but I was recently getting an error when intersecting 2 circles, and getting 4 results, instead of 2. Three of them were in nearly the same place. It was easy to work around, but I figured I'd mention it, since it seems sort of buggish.

RSS

About

Translate

Search

Photos

  • Add Photos
  • View All

Videos

  • Add Videos
  • View All

© 2024   Created by Scott Davidson.   Powered by

Badges  |  Report an Issue  |  Terms of Service