Light Matters: Can Light “Cheat” In Simulations?

Oslo Central Station. Architecture: Space Group, Credit: Luxigon,

In recent years the use of CAD and programs has resulted in a new understanding of light in architecture. The drawing board and its lamp have given way to the self-illuminating monitor. The result is that concepts in architecture are now made of light from the very first mouse click.  In the visualisation process, luminous space now predominates.

However, this begs the question: has the luminous impression (part and parcel of the perfect, rendered setting) become more important than the engineering or architectural concept itself? With the improved interplay of shades, contrast, and brilliance, can lighting actually obscure the point of a realistic simulation?

More Light Matters, after the break…

Foyer concept and previz for ERCO. Credit: Axel Groß, Electric Gobo.

While shades of sunlight can be illustrated in sketches with relative ease, imagining the interaction of multiple light sources in a space quickly becomes difficult. Thus, before the new millennium, architectural plans used to be displayed without indirect light due to a lack of hardware resources.

However, modern rendering engines can now clearly display daylight as well as incorporate indirect illumination interacting with surfaces, generating images with photographic realism. Furthermore, virtual luminaires (lighting fixtures) with photometric data offer the option to calculate illumination with physical accuracy, which allows for precise quantitative analysis in addition to aesthetic decisions. But even with the most sophisticated rendering engines, the qualitative results have their limits.

IwamotoScott Architecture with proces2, Jellyfish House, 2005–2006, model; Collection SFMOMA, Accessions Committee Fund purchase and Gift of IwamotoScott Architecture; © IwamotoScott Architecture

Although the more recent High Dynamic Range (32-bit/HDR) format offers a greater scope of contrast than previous processes with 8-bit and 255 gradations and theoretically covers the full luminance range of Nature (in which the sun is 10,0000 times brighter than a shaded zone), the perception of a perfect HDR rendering in print or on screen still does not fully compare to the atmospheric impression of a real, bright space, because image media currently cannot transmit factors such as glare or adaptation.

Courtesy of ERCO / Electric Gobo, Axel Groß

Thus, the advent of digital media has not only increased the speed of image production but has also given rise to new, though initially unconsidered, understandings of and ways of using light – as dictated by the characteristics (and limitations) of hardware and software. In particular, two distinct approaches towards visualization have emerged: on the one hand, scientific engineers who are mainly interested in quantitative data analysis in order to ensure a technically correct construction; on the other, Visionary Designers, who seek a subtle or impressive atmosphere with light. They develop a quality of light without any technical restrictions and ”cheat” with details in order to convincingly communicate an idea to a client. Even if this approach is less committed to reality, it can nevertheless stimulate the imagination and become a valuable tool in innovative designs.

White paper on mental ray. Concept and visualization for Autodesk. Credit: Axel Groß, Electric Gobo.

For Jeremy Birn, a lighting technical director at Pixar Animation Studios, “cheating” with light occurs on a daily basis: “Cheating is performed, to some extent, on almost every project done in 3D. … Light on a character that appears to come from a lamp may actually come from a position far away from the lamp if it lights a character better.” For Jeremy, lighting and cinematography are arts and not sciences. No matter the lighting style, it only matters that the light be believable to the viewer.  Thus, the most important requirement is that the image be internally consistent, e.g. showing a beam of direct sunlight brighter than the light of a task light or forming a shadow correctly.

Of course, the challenge for architecture starts when a client is keen on the imaginative visualisation and then asks the architect to make the concept a reality.

YouTube Preview Image

However, even still, perhaps the “cheat” is better than the alternative. For example, video visualizations have become more and more relevant in architecture, and yet the technology often wields unsatisfying results, as rendered films with a high degree of realism (in terms of indirect light and reflection) are difficult to produce; the spatial dynamics for precise lighting simulations require considerable computer power.

In the case of the Graz Art Museum by realities:united, an abstract project which utilizes a complex video mapping technique on the facade, the visualization was severely limited to a flat treatment of the pixel façade that excluded indirect lighting and reflections. The GreenPIX simulator by Simone Giostra & Partners and ARUP similarly reduced realistic lighting details for the benefit of a faster rendering process. The NIX project study by realities:united, on the other hand, although not particularly naturalistic in all lighting details, at least allowed a dynamic impression and different perspectives that vividly demonstrated the interaction between the interior lighting and the effect on the urban environment.

And so, perhaps “cheating” is a preferable alternative, one that more accurately conveys the intent of the project, rather than letting the project fall flat. What is your opinion? Is it OK if you cheat with the lighting in architectural renderings? Do lighting simulations need to be accurate from the early concept up to working drawings? Or is some artistic and atmospheric leeway preferred? Please share your view in the comments below.

For further reading:

- Birn, Jeremy: Lighting & Rendering. New Riders, Berkeley. 2006.

- Ochoa, C.E., Aries, M.B.C. & Hensen, J.L.M.: State of the art in lighting simulation for building science: a literature review. Journal of Building Performance Simulation, 5(4), Pp. 209–233. 2012.

- Groß, Axel: Animation and digital lighting. PLD 2nd Global Lighting Design Convention, Berlin. 2010.

, a monthly column on light and space, is written by Thomas Schielke. Based in Germany, he is fascinated by architectural lighting, has published numerous articles and co-authored the book „Light Perspectives“. For more information check or follow him @arcspaces

Cite: Thomas Schielke. "Light Matters: Can Light “Cheat” In Simulations?" 15 Jul 2013. ArchDaily. Accessed 28 May 2015. <>
  • solidsignal

    I was trained as a cinematographer in the era just before CGI and related image-modification tools became practical. One of the key things we learned was not to take light too literally. In other words, the aesthetics of lighting as perceived by the viewer (and created by the cinematographer) in any given situation may or may not correspond to the literal direction & reflection of light as dictated by physics. Amateur cinematographers tend to get hung up on ideas about recreating (simulating, as it were) natural behavior of light (i.e ‘realism’), as they understand it, often at the expense of the aesthetics of an image, scene or scene-to-scene continuity. Soon, a good cinematographer learns that “good lighting” is all about cheating. And, then, another revelation eventually occurs: Light in the wild does some very unusual things (e.g. combinations of reflection, glare, filtration, interactions with apparatus, etc etc) that a cinematographer would never think to if he were going by a strict interpretation of the physics of light and illumination. Yet, look around at the world, and there they are: these outlandish, “unrealistic” instance of light doing it’s own beautiful thing. Often it’s those beautiful tricks of the light that a cinematographer’s “cheats” reproduce, and they are the distinguishing feature of great vs utilitarian cinematography. I think the analogies with architectural lighting are obvious. In my mind, they all point up the fact that architecture, and architectural renderings of light, have both a technical and aesthetic component, and “cheating,” beyond the technical bounds of calculation & simulation, is always going to be essential to the proper discovery and presentation of the aesthetic component.

  • Sven-Erik Åkerman

    It depends of the purpose of the “cheating”.

    It’s important to distinguish between selling your project to be realized and to create an aesthetic representation of something. If visualization deviate too much from the final result, I believe that the public may be disappointed by the final product. Then it could appear that the architects or lighting designers are visionaries who haven’t managed to recreate the impression they have given in visualizations.

  • ganeadesign

    En las representaciones arquitectónicas, el principal objetivo es vender el proyecto y, haciendo trampas con la iluminación resulta más atractivo para el cliente.

  • Mark Li

    I would say that it depends what architect wants from visualization? I own my own visualization studio and I can tell that some architects want to win a competition at any price and they asking for “beautiful” picture! Others ask for physically accurate rendering, an image is usually not very pretty but close to real life.

  • stremeor

    Looking at history, “cheating” was actually necessary to achieve a certain look with reasonable efforts, in film production but also in architectural visualization. Simply because physically accurate simulation of light distribution was considered not affordable, nor was it even available in rendering solutions. Today’s advancements in software and hardware allow to produce perfectly realistic results in little time, which can easily be combined with stylized or artistic effects for visually desired “cheats”. The time spent in hand-”cheating” real-world lighting effects continues to shrink I think.

  • r b

    No, light does not cheat in simulations. It is the simulation itself that cheats.