Archive for the 'Analog Generative System' Category

03NovAnalog Generative System (pt4)

I have spent the past two and a half weeks producing drawings and diagrams of the aforementioned physical model.  Beyond simply representing the model, I wanted these drawings to express two core ideas: the formal language of tangency that emerged withing the ordering system of my physical model; and the idea of differential scalar perception (mentioned briefly in the previous post).

As an extension of the physical model, I produced a three-dimensional pointcloud to represent a field significantly larger than was physically practical.  The points are derived from surfaces which closely resemble the individual formal bodies of each vertical cluster (see boards below for further graphic explanation).  The intended result is to subvert the perception of the system as an object, which seemed to be unavoidable within the 18″x18″x18″ limit.  As such, the field is imbued with a new reading as a landscape and  identifyable clusters aquire a geological dimension relating to the aggrigation of the unit. (See Image Below)

Returning to the idea of differential scalar perception, the views represented above are analogous to certain views of the physical model in terms of their formal reading.  More specifically, the first image is similar to one of the macro shots of the model, where connections and individual units are highly legible, but the overall system is quite abstract.  Conversely, in the image above, the unit (and perhaps even the cluster) are less important and the overall system is no longer understood as individual units, but rather as a unified surface or landscape.  Intermittent views expose various aspects of the system at both local and global levels.

The full presentation boards for the project are included below (click to enlarge).

13OctIteration One (Analog Generative System pt3)

I have completed a physical model of one iteration of my system.  My favorite part about the piece is that a viewers understanding changes dramatically based on distance.  From 10′ away, ones perceptual resolution is relatively low, thus the object appears as a chaotic mass.  From 5′ away, directional variations and a density patterns emerge.  From 2′ away, one can identify continuous lines tracing through the geometry and order becomes increasingly apparent.

In the coming weeks I will be producing a set of analytical drawings to descibe the organizational system and the properties of the unit used in its development.

View a full set of model photographs HERE.

12OctEmergence (Analog Generative System pt2)

In my research of ordering systems, I found the concept of emergent structures to be a relevant model for my own work.  I will be posting images of my final model shortly.

Emergence (n) In philosophy, systems theory and the sciences, emergence is the way complex systems and patterns arise out of a multiplicity of relatively simple interactions. -Wikipedia

First Order Structures

At a local level, connections rely upon material elasticity in bending and torsion. The confluence of opposing forces yields stability at the unit-unit level. Therefore, connections with opposing bending and torque vectors will be highly stable (fig.1). This prerequisite for structural stability serves as the primary regulator of potential unit-unit connection types. Agglomerations of semi-autonomous bodies, or clusters, result from these conditions.

Second Order Structures

Clusters are best understood as ‘seeds’ with specific capacities to spawn further growth in the system. At this level, growth is regulated by input nodes and proximity to other clusters in addition to the aforementioned qualifiers for unit connections.

Third Order Structures

The system as a whole is also subject to performance constraints. Mass is accommodated by a density gradient, moving from the most dense at the bottom, to the least at the top. Directionality results from transfer of system-wide forces. Emergence, in this case, is a product both of the physical parameters of the unit itself and a system of behavioral tectonics that yield myriad growth patterns. The resulting complexity is characteristic of emergent systems and can be seen in natural phenomena such as termite mounds (fig. 2).

Photo taken by Brian Voon Yee Yap. Cathedral Termite Mounds in the Northern Territory.

06OctAnalog Generative System

The inaugural project of UCLA’s MArchI program explores the development of generative systems though tectonic assemblies of modular components.  Paramount to this process is a differentiation between top-down and bottom up systems.  More specifically, emergent formal typologies result from the physical properties of the module and connection variants as opposed to a ‘master plan’ envisioning the result.

Tactically, systems may be considered either derivative or generative.  In each case, a set of rules renders a specific formal grammar or syntax.  Derivative systems in architecture are often used to subdivide complex geometries into developable units.  Surfaces may be panelized.  Volumes may be divided into cellular components.  In many cases, this process of subdivision creates a type of ornament, articulating texture, structure, and scale.  Generative systems aggregate units into novel compositions based on fixed relationships.  The formal result may respond to inputs, but it will display behavioral variation.  Processes of branching, cellular automata, and others allow for highly differentiated outcomes.

This project is generative.  It is a study of potentiality.  This is necessitated by syntax: produce a unit, study its limits, develop a formal grammar, and produce a desirable outcome.  Ambiguity arises in the final step(produce a desirable outcome) due to an analog means of execution.  In theory, a complete set of rules would allow for the computational generation of an infinite number of solutions.  Subjectivity is introduced by the choice of a particular iteration.  In this case, subjectivity is injected prior to determining an outcome.  As such, rules are deployed loosely to produce a desirable outcome - grammar is reverse-engineered.  This promotes visual composition while somewhat subverting the nature of generative systems.  Conversely, a looser rule-set leads to the emergence of more complex relationships and accommodates anomaly.  Pedagogically speaking, it would seem that the exercise is meant to foster a consciousness about formal relationships and to encourage a systematic methodology in future designs.

The core unit in my proposal is a triad with six distinct connection points.  Each connection point may accommodate one or two inputs, resulting in 13 unique connection types.  Ordered clusters exploit the tectonic qualities of 1/64” Bristol board, recognizing that structural stability results from opposing curvature between at least two units.  Primary and secondary cluster types categorize the degree to which a cluster may be joined to other clusters; primary clusters are more readily joined (or offer a greater number of potential connections) than secondary clusters.

Clusters may be joined into higher ordered systems of super-clusters.  Super-clusters are the building blocks of the final composition.  The transitions and connections between super clusters generate a dynamic network system suggesting direction, formal volumes, and figural bodies.  Three initial ordering typologies are discussed below.

TYPOLOGY 001

This composition creates a regular two-dimensional matrix.  Structurally, it is extremely stable and could be expanded to large sizes.  Variations in curvature and direction may be achieved by physical alterations to the core unit type, i.e. elongation of ‘legs’ or changes in relative leg angles.

TYPOLOGY 002

This composition is regulated by an effort to maintain continuity of curvature between unit clusters.  The piece exhibits bilateral symmetry and a sense of completeness or closure by joining loose legs at the periphery.  Expansion of the system is limited only by the physical strength of the material.  It is unlikely that compositions larger than 18”x18”x18” could be self-supporting.

TYPOLOGY 003

This composition exhibits 3-part radial symmetry in response to the core unit geometry.  Variations in unit density produce visual interest and degrees of stability.  It is unlikely that the system could expand vertically beyond 18”.  Lateral expansion may be infinite.  As in Typology 001, a sense of completeness or closure is produced by joining loose legs at the periphery of the composition.


flickr feed

White on White on Brass on GlassWhite on White on Brass on GlassWhite on White on Brass on GlassWhite on White on Brass on GlassWhite on White on Brass on GlassWhite on White on Brass on GlassWhite on White on Brass on Glass