Computational design in today’s architecture is an essential method that allows calculating through Software, an enormous amount of data and complexity on a single element, using Grasshopper + Rhino, the digital assembly become in a way an immersive generation.
Starting from a primitive design of a landscape, the physical composition of a future shape, it is not a general fabric obtained using anchor points, but instead a fascinating result of millions of simulation between existing and proposed. Essentially there is not just one final prototype. We can obtain different architectures that are able to respond to our landscape conditions.
Primitively, separate studies were produced on a series of shapes, trying to explore an alternative generation, introducing the Raindrop command, stress analysis, live physics engine for interactive simulation, optimization and at the end drainage paths.
The result represents a scenario that focuses on developing elementary prototypes, where the critical position of the user is to select and compare final issues, based on the assembly of primitive raw data and research.
The argument becomes much more interesting if we start adding other parameters on the initial formula, where different aspects can influence the composition of our piece of architecture. Machine learning methods can analyze for instance forces that act in parallel with the landscape, movements of people or environmental condition, spatial configurations where problems become a potential solution for the final object.
In this case, the last one, had an enormous influence on the scale of the design as well as on patterns of social interactions, where the capacity for direct adaptive responsiveness enable an approach based on the fragmentation and research in different parts, from a natural structure to dynamic techniques.