Arnhem Masterplan
Foto divulgação [UNStudio]

GC: Tobias, please tell me about your education. You studied architecture at TU Berlin, then you did your masters at Stuttgart University, and then another masters at Columbia University. Which part of your education was more influential to the kind of work that you are doing here and your enthusiasm about new technologies?

TW: Let me start from the beginning. At TU Berlin in the first years I had the integrated projects. We worked in a team of four people for almost two years. I learned to work in a group from urban design all the way to detailing, and also learning how to make decisions like how to design a house in a very short period of time. Coming in the morning and handling the design a few hours later. That's a type of education that allows you to work in a very flexible way. Focusing on something in depth, but also learning to do it very quickly, and most of all learning to understand the connections and how each decision relates to other decisions. Designing is making decisions and the most important thing is to have a framework on how to structure your decision. That was mostly designing by hand. When I graduated I was still using AutoCAD 11 or 12, but I only used it for very complicated geometry, for generating a simple model that I could plot in perspective and use it as a base for tracing drawings by hand. I graduated in 1995. When I moved to Columbia I decided not to do anything by hand anymore. I took two paperless studios with Jesse Reiser, and that's when I had this idea of using the computer, not so much to do something faster but as something to think about architecture in a new way. The thing I was most interested in was the perception of space, to develop something that was much more complex than what you could have imagined in other ways.

GC: Did you learn coding back then?

TW: As a child my first computer was a Commodore 64 and I started programming it, writing simple codes in Assembler, so I got an understanding on how to program. But the work I did in Columbia was more about using Maya and Alias, and animations and time-based models, and not so much coding. That was 1996-97. Then I moved to UNStudio, where I worked for 10 years, I started working on a large interchange project in Arnhem. A project that we will finish construction this year, after 17, 18 years, it's a really long process. That was a project that had started [by hand] at UNSudio and then and took it over working with Maya and Alias Wavefront. It's all free form, all Nurbs, it was very much about fluid design. We realized very soon that it was very complicated to translate so we defined a translation [from those programs] to AutoCAD, because everybody else in the studio was using AutoCAD to develop the drawings. It was by DXF, very time consuming. And then in the competition for the Mercedes museum it became clear that we needed something with which we could handle internal communication much faster. The Mercedes museum was the first project in a row that copy/move/rotate, so it had to search a system that had a modular center, which does not appear to be modular, but there is a clear logic behind it that organizes the different parts. And from that logic it became very clear when we won the competition that it needed to be programmed and we needed to set up the logic. If you have the logic it's easier to program and to test a lot of different ways of doing it. For that we got in connection with Arnold Walz who started to program. I explained to him how the design came together. It's mainly about the hierarchy of decisions, which is something that influences your programming very much. If you change your hierarchy of decisions everything is different.

GC: How about digital fabrication? When did you start working with it and does that influence the way you design, do you keep in mind that something is possible to do by digital fabrication?

TW: It also came by with the Mercedes museum. We started to realize that if you have any sort of non-repetitive or non-standard structures, then people would see it as a big problem and that would increase the cost enormously. So one thing is explaining to [the client] that this is not totally free form like the Arnhem Interchange was in the beginning, but there is a logic behind it, and this logic has to do with the possibility of assembling or building things. If you have a twisted surface it's good to know what are the material properties, how far is it possible to bend that material. Or, in the case of concrete, you have to cast it how you make the formwork, and that started with the Mercedes museum when we went into research about how to create the formwork for the double curved surfaces for the twisted elements. One way to do it would be steel, which you can use several times, but the end it was a wood element and because we wanted a very smooth surface we used plywood; very thin layers of plywood and we tested how much you can torsion it, and then we had to rework the surfaces, to make it possible within this range. The interesting thing about digital fabrication is that you can get feedback because design is becoming faster and faster, and you can take parameters that come from a one to one prototype and you can readapt your design. So it's not a linear process anymore when you just have an idea and a shape and you find someone to make that shape possible. Now you can go and see how is it possible to make that shape and maybe if you change the shape a little bit perhaps you can make it much easier and then perhaps you can manage the budget.