Shatter Results Not Understood

Can someone explain my results?

It looks like the shatter component is doing what I want (shattering two curves with random lengths). But, the points (X's) on the curve don't make sense to me.

Replies to This Discussion

Permalink Reply by Vicente Soler on December 20, 2010 at 8:22am

You have to graft the curve input of the shatter component and the length component. Also, input a list of seed numbers into the random component that matches the number of curves.

What is happening in your definition is that the first number (0.77) is shattering the first curve (that's why there is a single point), while the rest of the numbers are shattering the second curve.

Permalink Reply by Chris Tietjen on December 20, 2010 at 8:40am

Any chance you can also explain why the first number coming out of the random number component doesn't match the first number coming out of the length evaluator component while all of the other numbers in the two lists do match?

Thanks.

Chris

Permalink Reply by Vicente Soler on December 20, 2010 at 8:50am

The curve parameter value is not the same as the curve length. For example, if you move a control point of the curve, the parameter values will not change but the curve length will change on a particular segment.

Permalink Reply by Vicente Soler on December 20, 2010 at 8:46am

oops, there one thing wrong in my explanation. The evaluate length component does behave like i said (it uses the first random number with the first curve, and the rest of random numbers with the second curve), but the number input of the shatter component does read a list of numbers so its shattering both curves using a list that consists of a mixture of lengths of both curves, in other words, the data structure is a mess).

Permalink Reply by Greg J on December 20, 2010 at 9:06am

Good stuff guys :)

I'm still at a loss. When I graft the inputs to the eval and shatter componets, and increase my seed input to 2, the curves shatter on both curves are the same and not random.

I agree the issue is with the data structure, but I'm not sure how to orginize it so I get the intended results (2 curves shattered randomly).

Permalink Reply by Vicente Soler on December 20, 2010 at 9:18am

I said a list of numbers that match the number of curves, not a single number which its value is the number of curves. The actual value of the numbers can be anything, as longs as they are not the same.

Permalink Reply by Greg J on December 20, 2010 at 9:58am

WOW, it works !!! Thank You Vicente and Chris.

I've been banging my head on this issue for way too long. :)

Now, back to playing.

Permalink Reply by Chris Tietjen on December 20, 2010 at 10:45am

Greg-

If you get a chance could you post the working def.?

Thanks.

Permalink Reply by Greg J on December 20, 2010 at 10:51am

There you go Chris.