Grasshopper

algorithmic modeling for Rhino

Genoform is not necessary for creating generative models with grasshopper. Grasshopper can itself generate random values. So why Genoform ?

 

The reality is not many designers are using computational processes to explore 1000's of design possibilities. What we see now, is generative models that are driven by the designer - with the exception of Galapagos - which selects designs based on fitness functions. So we have not entered a phase where we are using computers to creatively explore design possibilities. Now, when we do that we face some real problems.

 

1) Too much designs (can be in the millions)

2) There generated designs can be too similar to each other.

3) Unquantifiable nature of the goodness of the generated designs

 

Genoform is designed to addresses these issues. These issues are generally not faced if the design exploration is driven manually. They will be faced once they are generated by the computer.

 

The current version of Genoform addresses only the generation part. The future versions will address the selection part. But advanced users may use the Excel table to develop their own selection methods - that will be required in generative schemes.

Views: 827

Replies to This Discussion

I love this idea -- I am picturing people streaming design variations to a web page to be evaluated but the community.

 

Seems like a natural sibling to Galapagos -- to solve design problems that are difficult to translated into spatial numeric values.

 

I am also interested in your link to Solidworks.  I am not a Solidworks user, but would like to export solidworks partsfrom GH with the parameters embedded in the file.  Is this possible via Genoform or through excel?

 

thanks for the interesting project!

 

 

Glad you like it !

 

Yes both Genoform and Galapagos are about using the computer for design exploration. Genetic models will one day allow designers to develop designs directly with end users. This is beginning to happen in small sectors in the manufacturing industry.

 

Our current version works with SolidWorks, the next version will work with Inventor. We want to keep the data in XL CAD independent as much as possible so that the designs can be transported into different packages as and when needed.

Thanks for giving genoform a whirl.

I like the ambition of this project, but I don't think I understand quite the advantage of Genoform over Galapagos.  

 

It seems like while Genoform gives you the possibility of letting the GA manipulate more variables in the model (where with Galapagos you might have to make a million different sliders), it seems to give you less control over the GA side of things, hiding it all within the Genoform program.  

 

Maybe I'm not understanding the potential of the program, but it seems the same thing can be accomplished with a GA script in Processing communicating with Grasshopper, or potentially with Galapagos when it develops further.

Hi Adam,

 

Both Genoform and Galapagos are about exploring design possibilities. But they take different routes. You need to familiarize your self with both to understand their strengths and weakness. That would be the best approach.

Adam, many people are asking the same question. So I have modifed the ppt on Generative Design. Hope the differences are clearer. But the best way is to try them both out.

Cheers

Sivam

RSS

About

Translate

Search

© 2024   Created by Scott Davidson.   Powered by

Badges  |  Report an Issue  |  Terms of Service