Interview with Benjamin Black of Visarc Inc.

Interview with Benjamin Black of Visarc Inc.

Visarc works in collaboration with architects, builders, and building owners to create high quality computer simulations of architectural environments prior to construction. Visarc provides value to our clients by reducing their risk through the ability to accurately test, evaluate, and communicate design intention.

CGA: Could you introduce yourself?

I am Benjamin Black, co-founder and Executive Vice President of Visarc, Inc.

CGA: Tell us about your background and how you originally became involved with computer graphics and the architectural industry?

I grew up in a community supported by a federal government scientific research laboratory. As a result, I think I was influenced a lot by science as a way of knowing the world. Also, I feel like my brain is wired for architecture since many of my early childhood memories are filled with architectural environments. I went to the University of Washington determined to study architecture, although I continued to work as a research assistant in science labs over three summers. As a student, one of the things that bothered me a lot about architectural design was that it was extremely non-rigorous. While I do not expect architectural design to be modeled by scientific research, I think that architecture could have more credibility and be taken more seriously in our culture if the profession could adopt design methods to make itself more accountable. With conventional design technology, while it is certainly possible to create a plethora of design ideas, it is totally impractical to thoroughly test and evaluate even one of them. With science, experimentation, testing and evaluation is the creative process that leads to discovery. I got my first chance to combine science and architecture while pursuing a Master of Architecture degree at MIT. While there, using physically based computer simulation techniques, it was finally possible to realistically test and evaluate comprehensive design concepts from an experiential point of view in a practical manner.

CGA: Can you tell us a little bit about Visarc Inc. and how it was started?

Visarc was established in 1996 by Jack de Valpine, a colleague at MIT, and myself, to help building industry professionals leverage emerging computer technology enabling architectural design exploration and evaluation, communication of design intent, and management of risk. While pursuing graduate studies at MIT, we developed methods to integrate physically based architectural simulation technology with an exploratory and experiential design process. These methods enabled us to more effectively realize design goals, identify elusive design problems and resolve them in greater detail. Most importantly, these methods helped us understand what we were doing as designers, opening new doors toward inventiveness and creativity. It was obvious to us that these methods had to somehow be transferred to the profession, and we launched the company immediately after graduating.

CGA: Visarc has taken a somewhat unique approach in that you do more realistic simulations, rather than final presentation graphics. What led you to take that route when you started Visarc?

All of our work ends up as presentation graphics. People need to communicate design intent, and pictures (computer renderings or otherwise) are good ways to do this. Our presentation graphics tend to emphasize realism rather than artistic impression. When we started Visarc, our intention was to aid the design process so that designers could predict the (at least the visual) outcome of a design proposal from an experiential point of view. For this reason, it was essential to provide a deliverable that could help designers realistically evaluate their design intentions and drive design decisions. Our physically based simulation method is an ideal design driver because it can be predictive at many different scales and at many levels of detail.

	Visarc Click to enlarge	Visarc Click to enlarge	Visarc Click to enlarge

CGA: Have you developed specific processes or proprietary software to help you in achieving your realistic simulations, or are you able to so this with of the shelf software?

We have a specific set of processes that are proprietary to the company. We also have proprietary software that is designed to manage complex projects with variable staffing configurations. It feeds data into the Radiance rendering engine developed by the Department of Energy. We do use commercial software tools like Autocad for 3D modeling, but those are also slightly modified with api's.

	Visarc Click to enlarge	Visarc Click to enlarge	Visarc Click to enlarge

CGA: Can you explain a bit about the process that you go through with clients to develop their simulation?

In general, our process is set up to be iterative in order to allow change in the architectural design process. We work with architects and building owners at all stages of the design process from schematic design through construction documentation. With initial design documentation (sketches, plans, elevations, sections, photos, 3D model, etc.), we build preliminary images for client review. They make comments and provide additional support documentation, and we make a new set of images. Depending on the design objectives, this can occur once or twice over the course of a few days or it can go on for a long time.

CGA: What is Visarc doing that separates them from their competition?

We have a unique philosophy about how simulation tools can be used in a creative design practice and a proven set of custom made tools to carry out this mission. On the technology side, we have reliable, robust rendering power that eliminates the need for post process painting and compositing, thus keeping our labor requirements low on the post production side. We routinely render highly complex scenes with accuracy - we have several scenes in excess of 10 million geometric entities, and others with more than 1500 light sources using manufacturers photometric data. Our process is set up to manage modeling changes rapidly while handling simultaneous design options for comparison and evaluation purposes. All of this is being done with physically based global illumination techniques, of course.

CGA: Your work has been cited by SIGGRAPH for technical achievement, can you tell us about the project that won you that citation?

In the past, our work has been peer reviewed and selected to be part of the technical slide set included in the conference proceedings. In 1996, my business partner submitted a simulation of the MIT architecture studios that were about to be built. This was an early example of a detailed and complex architectural lighting simulation with global illumination. It was created while we were still students. Our work for a daylighting animation was cited in 1997. Although less complex, the project was remarkable in that it showed accurate daylighting performance throughout the day. More recently, some of our staff including myself, have had work featured that was part of graduate school research.

	Visarc Click to enlarge	Visarc Click to enlarge	Visarc SIGGRAPH Click to enlarge

CGA: The limits of photorealistic simulation seem to be pushed forward on an almost daily basis. Do you find that your client’s demands for absolute realism have increased or are most still impressed by the "wow’ factor alone?

Different clients are impressed by a lot of different things. Some of the things they are impressed with do not have anything to do with realism, or realistic simulation. There will always be people interested in non-realistic images, and people will always be "wowed" by paintings from masters like Vermir or Van Gogh. I think the demand for realism will continue to increase for business reasons, not for artistic or emotional reasons. Building owners will want to know more precisely what they are paying for, and realistic simulation offers that capability better than any other method. From that standpoint, realistic simulation has a huge opportunity to transform the way design services are procured.

CGA: What do you find most challenging about working on realistic simulations?

Helping people understand the difference between a conventional computer rendering or visualization and a physically based realistic simulation.

CGA: What type of deadlines are you typically given to complete a project?

We work on projects with time frames anywhere from a couple of days to over a year. For the longer projects, we organize goals on a weekly or bi-weekly basis based on the current objectives of the design team. There are times when we make significant changes and provide fresh images in the same day or next day in order for the design team to have visual feedback that they can use to advance the design.

CGA: What project are you most proud of and Why?

I am proud of a lot of projects we have done. One that sticks out as a favorite was a feasibility study for a major university. Working with an architect, the assignment was to show a university committee a few options for a building addition in an important campus quadrangle. There were many different opinions about what should be done, including whether anything should be built at all. A good quality, believable simulation was an excellent way to demonstrate performance of various options. We worked from two 8½ x 11 pages of sketches, annotations and diagrams created by a design principal, and a few site photos. In one week, we created a 3D model of the quad and three design options while the designer went on a previously scheduled vacation. The work was presented to the university and promptly settled the disagreements. I am proud of this project because we provided excellent value to our client by delivering great results quickly and working closely with high level design talent.

	Visarc Feasibility Study Click to enlarge	Visarc Feasibility Study Click to enlarge	Visarc Feasibility Study Click to enlarge

CGA: Are their any projects that you have not yet tackled, but would like to try?

There are many large scale civic projects that citizens argue about constantly. "Its too ugly, too big, too tall, too uninviting, too unfriendly" are common phrases tossed around when many projects are debated. Totally non-rigorous because no one has any idea what these things will really look like. People have a hard time understanding conventional architectural documentation, and artistic renderings are too biased to provide honest interpretation. Consequently, common citizens are at a serious disadvantage when given the opportunity to review a design proposal. For the benefit of the developer, designer, and citizen, it just seems silly to not have a realistic simulation of a proposal. I think it may be the only way an honest conversation can occur.

	Visarc Feasibility Study Click to enlarge	Visarc Click to enlarge	Visarc Click to enlarge

 

CGA: Where do you see the future of architectural simulation heading?

Increased access to accurate light distribution algorithms may help general quality improvement (i.e. looking more like plausible space rather than surfaces of glowing chalk) of architectural renderings. This trend may also make tools more complicated to use, and therefore force the cultivation of individuals with specific skills. I think we will also see significant evolution and advances in design methods because of simulation technology, and ultimately physically based simulation methods could transform architectural design in a way no living practitioner has ever experienced.

CGA: What do you feel is the most important aspect of rendering process that contributes the most to a successful simulation?

To simulate an architectural subject, it is necessary to have an accurate light distribution algorithm and pay attention to detail in geometry and materials. Light is everything. It is the basis of vision and a powerful component of our perception of the world. Without accurate lighting, it is not really possible to simulate an architectural environment.

CGA: What is in the future for Visarc Inc.?

We plan to continue to develop solid relationships with forward thinking firms and organizations, working with them to provide simulations that facilitate the meeting of their design objectives.

CGA: What tip(s) can you give our readers to improve their architectural renderings?

If you want to make accurate architectural renderings, it is absolutely necessary to use global illumination methods. Obviously, we have our preferences, but there are other solutions out there. Regardless of the tool, pay attention to detail and observe the real world to see how light behaves physically so you can be critical of your own work.

CGA: What do you not like to see in computer generated architectural work?

Oh man, where should I start? Poor texture mapping, grossly inaccurate lighting, and camera angles that make a building look like it is falling over backwards are among the things that bother me. I especially do not like work that the author claims to be "cutting edge" when obviously it is not.

CGA: What is your favorite link to visit on the web? (not necessarily CG related)

I used to like mlb.com to catch webcast ball games, but they changed the rules and I have not managed to pony up the $10.00 for unlimited access. So I have reverted to the radio. I also think that ebay is fascinating.

CGA: Which/What web based resources that you have found the most informative? (provide link)

Right now, I like websites with guts, especially sites that provide something that is difficult or impossible to access in a conventional manner. The IRS has done an outstanding job with their site (irs.gov). Fidelity (fidelity.com) and NASDAQ (nasdaq.com) are great as well. Like the IRS, many municipal and state government sites are extremely useful to the average citizen. I am hoping that the web as a resource will enable our governing infrastructure to become more efficient. Probably a boring response, but that is how I feel right now.