Eureka!

199 0

When it comes to conceptual design, there’s a vacuum in the digital market
This month I have been thinking about how traditional CAD software can sometimes inhibit rather than help and why we need a new generation of digital conceptual tools. Of course any new technology needs a suitably snappy name and while my proposal of Computer Aided Conception may fit the bill, it might not be the best name when you consider the acronym.

In researching what digital tools engineers and architects use to experiment and capture their formative designs, the one thing that became clear was the general lack of solutions.

Current tools range from the traditional: pen, paper, whiteboard, cardboard, to physical 3D formats like Lego, clay and foam. While some digital tools are popular these tended to be Photoshop and Google SketchUp, or perhaps advanced as McNeel Rhino but very few conceptualise in a solid modelling system.

Obviously there’s something about the traditional parametric 3D CAD tools that doesn’t make them particularly useful when being über-creative but there are a small number of digital conceptual design systems that have had some success, particularly at the high end, in automotive and product design. Autodesk’s Alias software suite is a case in point.

Alias has a unique interface and allows sketching through to modelling but years after acquiring the technology, Autodesk has only just now started to feed the technology through to its core design products, Inventor and AutoCAD. This represents a significant widening of scope for its customers, Alias Sketch for AutoCAD is just £500.

I talked with Carl Bass, CEO of Autodesk, to find out why traditional CAD didn’t work for conceptual design. Bass explained that, “There are many types of conceptual design: aesthetic, functional, behavioural but most common of all is to simply experiment with potential geometric shapes.

Traditional parametric CAD is not well suited for exploring ‘rough’ geometric ideas, the technology is too constraining, so designers have to rely on a potpourri of physical and digital applications and methods to explore form.

Advertisement
Advertisement

In the Autodesk portfolio the four key products that have been used for conceptual design are Maya, 3ds Max, AutoCAD and the Alias product line. The reasons these have been popular is that they don’t need the user to excessively commit to too much detail early on.”

So the issue boils down to freedom and the need to express ideas without constraints. This is something that traditional solid modelling CAD metaphors and systems do not cope well with. In fact, they actually hinder experimentation.

CAD is about documentation or assembling virtual prototypes to analyse, simulate and fabricate. To do this ideas must be robust and well thought through – you can’t model an engine and not yet know what the back of it will look like. Modelling is explicit while concepts lack definition.

Conceptual design tools need to be able to capture a number of basic ideas quickly and easily, be able to communicate the vision and then if possible, within a digital process, feed the engineering process with data to start fleshing out the idea.

So the issue boils down to freedom and the need to express ideas without constraints

Bass added that while it was early days, the digitisation of the conceptual process was starting to be addressed by CAD vendors.

With a new breed of conceptual software, combined with all the traditional craft skills and methods, designers will soon have an ever-expanding range of tools. There are also a few emerging technologies that I’d like to mention. These include:

Digital sketching: The advent of touch surfaces and digital ‘paint’, the core artistic drawing skill has a computer-based equivalent. Sketching and painting can now start on a computer. There have been some stunning works created on the various iPad paint tools.

Scan/photo-to-model: It’s magic to my eyes but photographs, laser scans or video of any object will soon be able to be used to reconstruct manipulable 3D models. Imagine being able to start any design from any form found in nature.

Research is currently underway to grab and digitise reality, the tricky part is to understand the captured surfaces to make them intelligently editable. This will give designers access to pretty much every physical object, man-made or natural.

Generative/computational design: The idea here is that a computational framework is created, geometric relationships can be built and designers can ‘play’ with the the design to find the best configuration. Led by Bentley Systems (MicroStation), McNeel Associates (Rhino with Grasshopper) and Autodesk are working on this too (AutoCAD).

Unfortunately this requires a fairly complete idea, as the original framework will have to be defined and geometry limits set. However, complex geometry can be generated very quickly from equations and analysis performed and structures optimised.

The fact that designers like to mix and match their tools and media means that a broad range of techniques need to be captured, simulated and implemented in any conceptual design software. To date there is not one tool to cover this wide gamut of creative styles but we are now seeing multiple attempts to digitise key artistic disciplines.

Traditionally expensive conceptual tools are also being brought down for the masses. With the traditional CAD markets saturated, I feel as if the conceptual market will now get increased attention and interest.

Bass admitted that he didn’t think conceptual design will ever be totally replaced with digital tools but would still continue to be expressed through a collection of methods. However these will move, more and more, into a digital workflow as the technologies and their adoption evolve.

I too am sure that with advances in laser scanning/photo capture there will be a blurring of physical and digital methods, although I think I may have trouble getting my CAC market segment name to catch on!


Martyn Day invents Computer Aided Conception