For the purpose of scaling, is there a component that can find the centre point of any polygon?
Please see below quote.
component (Curve > Analysis > Centre) which gives me the centre of the circles. By
connecting it to the you can see that all circles would rescale in their level without movement. - Quote from Khabasi
If it's not planar or closed, you can find it yourself by extracting the polygon vertices, separating their x,y,z values, finding the arithmetic mean for each of the 3 lists of numbers (using the mean component) and constructing a new point with the resulting 3 values.
To extract the vertices you can use the Divide Curve component set to 1 segment and the split at kinks option enabled. To separate the x,y,z values use the Decompose component.
Hi Vicente, the triangle on left of picture is the curve I am trying to make a point at centre of, the definition on right is what I have tried but does not seem to be working.?????
Permalink Reply by taz on October 25, 2009 at 1:05pm
Looks like Vicente took off...
But you should be working with points (not vectors) and you'll want to find the arithmetic mean (AM) of the x, y and z components to then construct the centroid.
Alternatively if all your polygons are triangles (I'm not sure from your screengrab...) which are always planar you can create a surface to find the centroid.
What a great strain guys. I am finding this very helpful. I am trying to do the same thing for hexagons. Could you fill me in as to what the third operation is above? What is the green line with the arrow above it?
Here's another one, it computes the average of all the corners in a polyline (or just a list of points really). Mass addition on all the points, then divide by the number of points you have. Basically the regular Average formula:
A + B + C + D + E + ... + Z
────────────────────
N
Permalink Reply by taz on November 17, 2010 at 12:40pm
When you shatter a closed curve the start and end verticies are the same. The green line with the arrow is [Shift List]. With the default shift settings (S=1, Wrap=False) the end point drops off so the same vertex isn't being counted twice.