Author Archives: Dallas Puckett

Design Development




FP1FP2 Interior


During the design development weeks, I worked mostly on interior details and plans. I had to figure out exactly how all of the first floor space would be utilized, as well as how the folding glass doors on the second floor would work.

Final Board


P5 Site

FullSizeRender[1]FullSizeRender IMG_0200


P9 Building Skin

South Facade

Bridge View Entry View


With the west end of the restaurant/bar being connected to the Purple People Bridge, I wanted to carry over the structural integrity into the form of the building. The adjacent bridge employs a truss system that hangs below the road. I used a similar truss to support the roof and both floors, and are open to being experience at various points below.


truss3 truss2 truss

I used a top down approach to design the shell of the building. The roof was developed first, and offset points being projected to the ground to form angled columns. The diagonal is a often used element in the design, again calling to the structure of a bridge. The entire truss system for both floors weave back and forth in a diagonal motion.



A couple moves were made because of the circumstances surrounding the site. The most obvious is that the building is lifted about 18ft off the ground. This is partially to avoid flooding, which occurred earlier semester. The other is a 10ft roof cantilever on all sides of the building. The program will be especially heavy during the summer, so providing shade and comfort is important. Both floors are tucked away under the floor or roof above, so outdoor dining will be possible even during the hottest summer days.

IMG_0125 IMG_0126


I wanted to express that the long ramp was a separate system from the building structure. I used a proximity 3D component to model columns off of the large diagonal columns to the bottom of the ramp. The same tree-like columns were created in the long span of the ramp.

Ramp Columns Ramp Columns2

The pros of this design method has led to a fairly cohesive structure that appears buildable. The structure is independent from the existing bridge and concession stand, and provides a pleasant contrast to the historic red brick wall of the bridge. However, because the design is spawned from roof, any changes in the roof or large columns would lead to the need to change most elements of the building. There was no true parametric data-driven analysis ran because the walls are glass on the second floor and stud walls on the first floor.


Moving forward, I want to look at creating a partially 3D printed model. I would love to have a model for DAAPWorks, but also don’t want to spend hundreds of dollars on it. I will most likely begin renderings next week and leave the model for last.

P8 Midterm

plansSite Render 4Render 1 Render 2 Render 3  Roof_Artistic



Concept Mass





P4-Frame Optimization

Grasshopper’s Millipede plugin can be very useful is analysis and creation of structural members. As parametricism struggles to sway traditional architecture theorists, Millipede provides a solution to real world problems. The program analyzes and suggests the most efficient system in a given framework, which could save clients’ money and architects and engineers headaches.

The area of Millipede that I chose to study is the basic analysis and simple framework. If a frame system is created, Millipede can provide the required thicknesses of beams at different moments. Maybe a beam needs to be thicker at one end, but thinner at the other end, for example. The script began with identifying the type of boundary condition. I chose a point load so the frame would have stress in the center due to gravity. The next step was to identify the material and beam type of the frame system. Thickness and inner radius of the hollow circular beam can be adjusted. This will be relative in the end product, however. The beam type was thin plugged into the Frame Curves component, which assigns the frame system to a given set of curves. The boundary condition and Frame Curves components are then plugged into the FEA System, which produces all the analysis of the frame (geometry, properties, etc.).


This component needs to be connected to a solver in order to extract the information needed. In this case, the FE Solver is used to determine optimum frame and shell thickness. The final step is to plug the solver into the Frame Visualization component. This is the only component in the script that will allow the user to preview the work.


For the second part of the assignment, I feel that a joint member would be best to demonstrate the optimization that Millipede produces. I ran the script twice on a set of simple frames with an attached cantilevered system. This is to create additional stress and create the most difference in the size of the members. You can see in the image below how the center of the frame becomes thicker where the most stress occurs.


I wanted to increase the amount of diversity among the members so in the second iteration I randomized the angles and length of the beams. The result was drastically different sizes and shapes at the center point.


Moving forward, I’d like to advance the joint connections to possibly 3D print.


Script: FrameOptimize

Rhino: 000_simpleframes_eigenmodes


P3-Dallas Puckett

Drawing1Drawing1 Drawing2_PSDrawing4_b

P2-Dallas Puckett

panel_1 panel_2 panel_3 panel_4