Grasshopper

algorithmic modeling for Rhino

Hi and Greetings to everybody

I am really enjoying your wonderful network.

I am new to grasshopper (and rhino too). After struggling with this incredible program for a month, I've found it really powerful, yet I did not find innate components for functions which I think a generative design really need them, LOOPS. ( VBscript and C#script are really hard to use for beginners and somehow time-consuming)
In addition I think a general innate "bake component" which is controllable through definition itself. (Guilio Piacentino Grasshopper tools in C# are really great)

And a question: Has anyone worked on evolutionary systems in grasshopper?

As I am an amateur. I may have missed these components. If there are such a components please guide me.

Views: 2886

Replies to This Discussion

you can't "self-reference" components yet, so that means no recursive loops using standard components. c3 and vb are quite the only thing where you could implement something like this. anyway, i haven't come across any visual programming language (i've used just vvvv and pd) that would allow you this.
i might be wrong.
What about a VB or C# script component which gets a conditional statement and a cluster of components as inputs, then execute the cluster as long as the conditional statement is true?
Do you think it is possible to write such a script?

Besides do you think there is a need for graphical loop tool in grasshopper in general?
i really don't know what to say - i think grasshopper is built with a more or less an instant updates philosophy in mind. usually evolutionary systems are much more process orientated and open ended. reaching a stable state (if this is what you are after) is nice but usually takes quite some computational time so it's unlikely this will happen in gh.

so if it's simple and easy then you can try and code it in a scripting component. louis did some funky stuff with recursively dividing up a surface: http://www.grasshopper3d.com/video/ghnet-adaptive-surface-1

there's also the timer comp and coupling that with a scripting component (in which you can remember values) should get something out of the evolutionary box.
thanks for your help and idea
in vvvv it is possible trough using 'frameDelay' node... maybe something similar could be implemented in GH?
I think looping exists in visual programming languages. ( I saw it but i can not remember which platform it was )
---> http://www.neu.uni-bayreuth.de/de/Uni_Bayreuth/Fakultaeten/1_Mathem...

What you need to do a looping component is a 100 percent break situation.
Thanks for the link. GEF and GMF both look like great frameworks.

What is your idea about questions in reply to Dimitrie?
I know I'm a little late here, but I guess I'll chime in regardless. Personally I don't really view GH as a visual programming application, but rather simply a nodal editor. Thankfully its a nodal editor on steroids because of all of the extra bells and whistles that David's added (functions, arithmetic, booleans). Also, David has structured a lot of the components to easily replicate the behavior of programming structures without having to worry about setting up those structures manually (ie creating an array of objects is not done through a double For...Loop, but rather by selecting cross reference when creating the vectors). In addition, the solving of the GH solution is done in chuncks rather than each individual object within the data stream. This means that in cases where a collection of components need to be solved as a group before the next iteration this is not possible directly, and requires setting up data tree structure to try and achieve the same behavior.

My point is that if you wanted this kind of functionality, then you would need for GH to function as a true visual programming language, and expose programming structures, such as For...Loops, Case statements, and If statements, in a manner that allows for a codelike control of the solution. This could potentially allow for proper iterative solutions (which IMHO isn't necessarily a forte of an interactive modeler), recursive functions and solutions that have much more capability to be generative or evolutionary. I think this could certainly be done, and I think it would fit best as a mode to work with in GH rather than a complete replacement of GH's current approach. There would need to be some changes with the solver itself, and also the components themselves would likely have to be "reduced" a bit.
Did you ever script looping?

I am looking for a scripted alternative to Anemone so that I can cluster it up - This requires the script approach you are talking about here.. So I wonder if you managed to solve your problem?

RSS

About

Translate

Search

Photos

  • Add Photos
  • View All

© 2024   Created by Scott Davidson.   Powered by

Badges  |  Report an Issue  |  Terms of Service