yeah nice.. tho i think there's a problem with generative design when one does not really know / care to know about the parameters involved in the creation of things. ignorance then becomes the main driving force, and merely visual playing.

nothing against the pics tho ;) beautiful indeed. must be the color.

Posted by: noci on Feb 16, 05 | 4:47 pm

I agree. This has ended up being an experiment in eye candy. My studio project is centered on materiality and less on form; Let's just say it counterbalances last semester with Hernan. I do claim only an investigative knowledge of a single generative design tool.

Interestingly, Peter makes no forceful attempt to make us use genr8. The dialog about generative design has been interesting, though.

At this point, form is not an overwhelming driver in my research.

Posted by: Steve Fuchs on Feb 16, 05 | 6:51 pm

I understand the comments about eye candy. Genr8 is focused on form and there is no notion of materiality involved in the software. Not surprisingly, this is usually one of the first comments/complaints from users. In the studios at the AA the students have explored the possibilities of analyzing the output f using other software tools. This analysis is the related to material and structure properties. The outcome of this analysis is used to change the parameters going into Genr8.

Posted by: mhemberg on Feb 17, 05 | 1:57 am

could you elaborate on what exactly the further analysis was focused on?

but i think i get your point-- you're creating kind of a evolutionary optimization scenario here, no?

mhh still- materiality or not - isn't the act of deciding on the initial set of parameters the actually interesting one? .. since it'S outside of the generative feedback loop and partially determines (does it? i'd speculate on it..) the process development itself.

Posted by: noci on Feb 17, 05 | 1:34 pm

A procedure that's been used a few times is to investigate the curvature of the surfaces (easily done in Rhino). This year's students present their work online at

[url=http://projects.csail.mit.edu/emergentDesign/genr8/AA/EmTech2004/index.html]
http://projects.csail.mit.edu/emergentDesign/genr8/AA/EmTech2004/index.html [/url]

(there should come up more presentations during the next week)

Yes, there is an optimization scenario where you as designer have some criteria (if the surface is structurally sound in some way) in mind and then (indirectly) try to express those notions throught the parameters and environment of Genr8.

Your last comment is spot on. The challenge facing the user is to be able to understand the parameters so that they can be tweaked to get the desired behavior.

Posted by: mhemberg on Feb 18, 05 | 6:36 am