Contemporary advanced simulation software allows, on the one hand, a more accurate un- derstanding of material behaviour at an architectural scale, and one the other, as a form-finding methodology. The potential of continual structural evaluation in form-finding allows for the morphology of an architectural system to be informed by physical laws instead of mathematical definitions in real-time, enabling the evalua- tion of multiple iterations of the same system to happen simultaneously.
However, despite the increasing tendency to utilise these softwares in architectural design research methodologies as a means of solving multiple variations of a (structural, material) system, there has not been much attempt to rethink the more generic digital tools that architects utilise today. This is an opportunity to begin to question and interrogate the way we utilise these tools, with a particular interest in how to incorporate open-source tools into or against the more generic softwares that are now increasingly used daily by architects. Thus far, the development of open source tools has remained largely within the problem-solving linked to specific architectural projects or problems. SoftModelling on the other hand, is an open source Java application developed to address not only a specific project, but also can cover the basic function of a digital design software as its code is open source and easy to manipulate in order to create multiple versions by using Processing as a framework. It is multi-scalar in its application, as it connects together two of the most-used design tools; poly-modelling and physical simulation, the application functions differently than other software packages available. To give an example, when modelling in Maya the designer has to convert the model into a physical simulation, but cannot operate directly on the model when it is in the physical simulation. Kangaroo for Grasshopper works similarly, as you have to model the architectural element and then run the simulation, etcetera. SoftModelling deals with this by seamlessly integrating between modelling and physics.
– Processing by Casey Reas and Ben Fry (MIT)
– HEMesh by Frederik Vanhoutte
– Toxiclibs library by Karsten Schmidt
– ControlP5 by Andreas Schlegel
– PeasyCam by Jonathan Feinberg
Code available at: