Click here to return to Home/Introduction

Making computers a part of the Process of Design is a special interest of Joe Calvin. He has both explored much of the design-oriented software available to the architectural profession, as well as tinkered with creating his own. Design is a difficult process in and of itself, requiring the full attention and resources of the architect. Joe believes that in order to be successful, design-oriented software, above all else, has to be extremely intuitive and easy to use so as not to distract the user from the design process. The process of entering information into the computer has to be at least as easy as using a pencil on sketch paper. To that end, Joe has continuously been designing, coding, and demonstrating various interface concepts since 1992. Below are two examples of this work.


ArchSketchPad
ArchSketchPad is a software application, designed for architects, that Joe has been working on and reworking since 1992. The concept is simple---the user uses a pencil tool to draw freehand and the computer interprets the sketching as a three dimensional (3D) for a building.

The computer makes no assumptions about the point of view or the scale of the drawing as sketching begins. These are inferred from the 2D relationship of the lines being drawn. Once the computer "catches on" to the design, scroll bars are displayed which allow the model to be rotated to any view for additional sketching. Just like a real world pencil, the user does not have to pick up the point at the start and finish of each drawn stroke; the mouse can be dragged continuously and the straight line and arced geometry are inferred. As lines close to form shapes, opaque faces are formed automatically in 3D space and act as canvasses for the user to draw additional geometry on. In this method an entire building can be designed.

Once the initial design is quickly sketched out, the user can select strokes and faces to make dimensional adjustments. A "recursive change engine" in the program helps propagate the indicated changes throughout the model. A new model will react to changes like Jell-O; an older more refined model will react like mashed potatoes.

In ArchShetchPad, the "camera" is automatically positioned to always show the model (or selected portion) comfortably scaled in the center of the screen. This not only saves the user from having to work "view" controls, but also prevents "lost in space" syndrome, common to users new to 3D software.

As the user draws, ArchSketchPad infers the use of different materials according to color. This materials list is automatically generated and self-organizing. Manipulating the list, the user can make changes to any material as a whole, temporarily hide any material, or adjust the translucence of the material showing the interior details of a model.


Etch-3D
Joe had two objectives in writing Etch-3D. The first objective was to learn "OpenGL/QuickDraw3D." OpenGL and QuickDraw3D are extensions to computer operating systems that allow programmers to display high speed 3D graphics. While typically used for arcade games, OpenGL and QuickDraw3D are finding there way into design software. The above screen shot shows solid rendered objects. These are not the result of a rendering command issued by the user. The objects are instantly and continuously redrawn on the screen as solid objects even during real-time rotation of the scene.

The second objective in writing Etch-3D was to adapt the simplicity of a child's Etch-a-Sketch toy for 3D modeling. Instead of turning the "little white wheels", they are dragged horizontally along the edges of the frame to draw horizontal lines. The frame itself can be dragged up and down to draw vertical lines. Whenever the lines form a closed shape (such as a square), a solid opaque face is formed.

Features of Etch-3D include:
• Real-time, solid, perspective view
• Simple Etch-a-Sketch style Interface
• Automatic face recognition
• Visual feedback for easy precise alignment (snapping)
• Scroll bar real-time rotation of scene
• Pen-Up and Connect-the-Dots button for diagonal lines
• "Fog" Depth perception cueing
• A 3D grid for dimensional accuracy
• The red frame interface can be hidden for presentation.