Grasshopper

algorithmic modeling for Rhino

Each of the test components (Greater/Less than or equality) seem to produce erroneous results unless the values are filtered by a notes panel.

I think the result should be TRUE when zero is tested for equality against zero no matter where the zeros are coming from.

If this isn't a bug, would someone please enlighten me as to why these tests produce different results.

Grasshopper Build 0.9.0076

Views: 204

Attachments:

Reply to This

Replies to This Discussion

Hi there,

it is not realy a bug.

Numbers in computers have a limited ammount of precission. I don't want to go into detail about that -> check wikipedia for float / double...

In Grasshopper you can specify the display precission of numbers under preferences. When you connect a number to a panel it is acctualy converted into a string with that precission. e.g.:

0.000000000000000000001 will be represented as 0.0 string in the panel but of course it is not equal to 0.

But as you convert it to a String in the panel and than convert it to  a number for comparision again the string 0.0 will be converted to the number  0.0 and all of a sudden  0.000000000000000000001 equals 0.0...

So what acutally happens through the panel is a loss in precission.

The best way to test equality on doubles is to ask if the Absolute Value of |A-B| < tolerance value. If true they are equal(enough) if false they are not equal(enough) . 

Thanks for the workaround tricks and the reminder about floating point errors.

RSS

About

Translate

Search

Photos

  • Add Photos
  • View All

© 2019   Created by Scott Davidson.   Powered by

Badges  |  Report an Issue  |  Terms of Service