The process by which architectural forms are imagined generally involves deeply imbedded relationships between the mind, the hands and a repertoire of preferred tools that are highly resistant to reexamination. These relationships, more than anything else, define the parameters within which we allow forms to be created.

The profession of architecture has slowly, but finally, acknowledged the importance of computers in the office. The machine is accepted as a tool that creates enormous efficiencies in calculations as well as in the repetitive tasks involved in the production of construction documents. The implications that the presence of a brain with such highly expanded mathematical capabilities could have at the earliest stages of the design process – at that point where form is first conceived – have been only minimally understood or taken advantage of.

What if architects understood more clearly the underlying assumptions behind program languages and could demand of them a tool to rethink the standardized constructions most computer-savvy architects tend to employ? Further, what if architects surrendered their own prejudices to the highly advanced capabilities of the computer to understand mathematically complex relationships that are simply too time consuming for the human mind to work out – forms, in fact, that the limits of our own language have prohibited us from imagining?

This work by Peter Samarin explores a realm beyond the sketch or the material model, in which the tools of the computer program – in this case Alias, a program developed for industrial design and animation purposes – provide the possibilities, set the parameters and determine the constraints of a methodology for the construction of architectural form.

Samarin conceived of the topography of the project, sited at the eastern terminus of the Century freeway, as a recording of the events that have occurred on the site. Against a grid of the standard house lot size, “interruptions” were recorded – the carving of the 605 freeway, the channelization of the San Gabriel River and the future (at the time of this work) Century freeway – by assigning curvatures and elevations to each.

This process of building the site in the computer established a working vocabulary of curves that Samarin could then experiment with in the construction of form. Four such curves, joined into a perimeter line, outlined the edge of a proposed shape. By interpolating between these four curves, the computer was able to construct a wireframe structure of the surface in three-dimensional space.

Samarin utilized four specific construction operations in the creation of his series of surfaces. In bi-rail construction two opposite edges are defined that differ from one another – although the second might be a result of manipulations and deformations applied to the first – and the other pair of edges are defined as identical. The structure is an interpolation between the two different curves that is extruded along the lines of the two identical curves. In contrast, swept surface construction is able to interpolate a surface between four distinct curved edges. In square surface construction two faces of a cube are interpolated and the side edges are then defined in order to create a space-enclosing framework.

Finally, there is an extrusion process in which one shape is extruded along the path of another, resulting in a genuine hybrid of two forms. This process is probably the most mathematically intense of the four operations, and the one able to generate complex forms beyond the ability of the designer to visualize.

In each of these operations, there are a number of opportunities for the designer to intervene and manipulate the form in order to alter the outcome. As the computer interpolates between the edges of the form a single curve, or a portion of that curve, can be favored over others to alter the resulting surface structure. Once the surface is defined, it is infinitely malleable, as it would be in other modeling programs.

In some ways, Samarin explains, the process was a little like putting your hands to work inside of a black box. The wireframe structure, at this point, was still a mathematical abstraction: the creation was in fact a two-dimensional projection of an object that had only just been constructed and not yet seen. With the application of a surface layer and lighting effects, views of the form could be generated using Alias’s animation capabilities and the constructed surface could then be studied as a three-dimensional object. A similar sequence was created from the topographic information, allowing for a series of views through the constructed site.

In a project such as this, the uniqueness of the computer has begun to be recognized: it is a tool that carries with it its own intelligence, along with its own methodologies and sets of constraints. The result of setting aside more familiar and more static instruments and opening up the design process to the computer’s potential is this compelling set of images.

-Chava Danielson

Peter Samarin is trained as an architect and is currently working in interactive multimedia.





Back to May 1995  Newsletter