Shatter component issue

Replies to This Discussion

Permalink Reply by David Rutten on March 21, 2015 at 3:34am

Hi Julien,

looks like at least one bug. Can you internalise those curves and parameters then post the file so I can check it out?

Permalink Reply by Julien Ybert on March 21, 2015 at 11:03pm

Hi David,

Thanks for your prompt reply!

I think I figured that the Shatter component doesn't seem to like extremely small values of the t parameter  (see attached screenshot)

I modified the seam of the curves and everything was working perfectly.

You'll also find attached the grasshopper definition with internalised params and curves (which kept on crashing when I tried to reopen it...)

Attachments:

• shatter bug.png, 155 KB
• shatter bug_def.gh, 18 KB

Permalink Reply by David Rutten on March 22, 2015 at 5:00am

Thanks Julien,

I'm getting a cast error (one of the curves is a Null curve). Is that supposed to be the case?

Permalink Reply by David Rutten on March 22, 2015 at 5:04am

Also your geometry is very far away from the origin and very big in comparison to your tolerance. This can cause problems and severe slowdowns and Rhino is forced to solve issues at a ridiculous level of accuracy.

Permalink Reply by Julien Ybert on March 22, 2015 at 5:24am

Not on my computer. There was only an error on the Shatter component.

Unfortunately I can't double check the definition I sent you, it keeps crashing when I try to open it..

I will fix the position of the drawing and the tolerance straight away.

Thanks a lot for your help David! 

Permalink Reply by David Rutten on March 22, 2015 at 5:54am

I found and fixed the bug in the shatter logic, it was indeed to do with extremely short segments. The fix will go out with an upcoming Rhino6 WIP.

It works now despite the tolerance issues.