Grasshopper

algorithmic modeling for Rhino

Hi All,

Following a couple of project which implement Python and Kangaroo I've been experimenting with generating Kangaroo forces within the Python scripts responsible for generating the underlying "dynamic geometry".

It seems that the new node-in-code functionality of the GHPython component is an obvious approach to this. I have just quickly messed about with it and it seems that calling forces such as "SpringsFromLine" and "Equalize" using ghpythonlib.components is substantially slower than using the corresponding Grasshopper components.

I have tried using different type-hints and access modes, but the attached version is the best result I've gotten so far:

Any ideas of what I might be doing wrong here, or, is this just maybe how node-in-code works?

Thanks,

Anders

Views: 697

Attachments:

Replies to This Discussion

Hi Anders,

It's the issue with node-in-code. Check this topic for more info. It was originally created for the purpose of parallel module issues but in the end it looks like the node-in-code was the cause of the speed drawback.

Hi Anders,

I have been working on this recently and have a prototype that shows dramatic improvements in performance. I'm meeting with David and Giulio next week and want to go over these changes with them before we push out any sort of new version of ghpython.

Thanks Djordje, I didn't even consider checking Discourse. Learned a lot from that thread. That sounds really good Steve, looking forward to seeing what you guys come up with. I hadn't really considered the potential of node-in-code, but I'm now getting appropriately excited. Thanks for the update :)

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