top of page

Third 2: Drawings

4.022, SPRING 2019


The first project in 4.022 was to create a series of drift drawings in Rhino and Grasshopper. The catch was ​twofold. First, the volume of drawings was immense (36 per every two days, initially) and second, each one of the 3x3 matrices in which we presented the drawings had to be laser cut. Printing, plotting, or otherwise producing the drawings, was forbidden/discouraged. The iterative process should operate on the basis of operations like repetition, seriality, offset, displacement and/or slip, looking at artists like Sol LeWitt for inspiration.


(please click image to access Dropbox link and explore in high definition!)

This catalogue is a taxonomy of my drawings. Effectively, they are organised in a clockwise manner (from first to last). Because I developed Grasshopper definitions and then used the internal variables to generate a multitude of forms, the drawings are organised into sets or families. The families extend outward radially, with miscellaneous offshoots such as explorations into tessellating or fixed boundary conditions. Connections between items in different families which nevertheless possess similar properties are shown with dotted lines.


The diagram at the centre is repeated in the corner as well; this was created before I began working on specific definitions and seriously experimenting with Grasshopper, and is more CAD/Rhino-based.

Screenshot 2019-09-13 at 10.40.16 PM.png


I missed the first week of 4.022, so to say that I had a steep learning curve would be an understatement. There was a good weekend where I spent ~10 hours each day doing nothing but try to figure out how Grasshopper worked. It also, perhaps as a consequence, took a while for me to figure out where I was trying to go with my project. This was made particularly difficult by the fact that my interests pointed in a different direction to most of my friends'. Shortly after I began exploring Grasshopper's functions I realised that I wanted to be surprised by what came out of the algorithm, to work entirely parametrically, and eschew compositional manipulation in Rhino. 

Perhaps this desire arose out of the training I had received at Cornell, and the lessons the previous semester's projects had taught me — when I am doing coursework or architectural work, the less I allow my own sense of aesthetics to affect the process, the more successful the outcome is. (This juxtaposes my approach towards drawing in a rather interesting way.) Either way, I wanted to lay down a set of rules that would produce something dynamic that I could see evolve in real time according to the inputted parameters. Consequently, rather than taking the more popular approach of creating objects and then manually manipulating them, most of my time was spent developing the rules themselves. Each code then produced a family of objects, which I ultimately assembled into a taxonomic tree.

A further difference was that my objects were self-contained. My boundary conditions were imposed by the base geometries. This was an important constraint for me because of the layers of processing in each Grasshopper file. I wanted to create something that did not immediately appear to have arisen out of relatively simple shapes, yet was profoundly affected by the information encoded in the said polygon (number of sides, number of repetitions, etc.) 


Pinup. Unfortunately, the lighting was terrible and the corridor very narrow...

bottom of page