IAEEL newsletter 1/95

Lighting Design Software: Not Yet Good Enough

Computer software for use in designing lighting systems and making associated calculations and visualization can provide valuable help in all phases of the design and planning process. However, non of the PC-based lighting design software programs available today meet all the needs of demanding users, but the future looks promising.

Everyone who attempts to define "good" lighting conditions must take the following factors into account: light levels on both the horizontal and vertical planes, reduction of glare, luminance distribution, contrast, reduction of flicker, and the spectral qualities of the light. Consequently, any demanding lighting designer would require that lighting design software be able to calculate these factors. In addition, lighting design software should take the direct and indirect energy impacts of a lighting installation into account, be user-friendly, provide highly accurate output, and have the capability to visualize the outcome of a lighting installation. On top of this, the program should be designed for tomorrow's powerful PCs.

Today, it would be asking too much to demand all these things, but based on a market survey done for NUTEK (Swedish National Board for Industrial and Technical Development) we found that two programs (or program packages) have great potential.

Most of the PC-based lighting design software programs in use today have severe limitations:

Although many are easy to use, they were developed for old hardware.
They evaluate too few lighting quality parameters; most often only horizontal illuminance is calculated.
They can only deal with simple, box-shaped rooms with no interior objects.
They do not include daylight, nor do they make energy-flow analyses.

Most of today's PC-based lighting design software is marketed by luminaire manufacturers, and such programs typically only operate with the companies' specific products.

There are some effective and highly qualified software for the lighting design process on the market. However, they are made for workstations and are too expensive and too complicated for the typical lighting designer to handle. In addition, they generally do not include other integrated functions, such as the capability to calculate energy flows.

Pure visualization programs are another category of software used in the lighting design process. These programs are impressive in their ability to make a realistic rendering of the outcome of a given design, but they do not have the capability to calculate and quantify the outcome.

TODAY'S NEEDS, TOMORROW'S PROGRAM
Optimally, lighting design software should help the designer and planner to come up with good solutions to lighting problems in a short time. Moreover, the software should facilitate and speed up the communication process between all involved parties in the design, planning, and construction process, i.e., consultants, lighting designers, architects, installers, and their clients. The requirements listed below are based on insight into the planning process as well as on what we believe are prerequisites for creating a high-quality visual environment.

It should be noted that the specifications outlined below would exceed the demands of many users. This was our intention, since we envisage a software application meeting user needs but with a number of "user levels" from which the user can choose, depending on the task that has to be solved.

The demands below do not reflect a mere "wish-list", but what we regard as realistic targets for a PC-based program:

It must be able to calculate daylight in addition to artificial lighting. Daylight is a necessary part of the human environment, and it also represents an important potential source of energy savings.
It must include and analyze the energy flows of a building.

It would be a definite advantage if the program could include the necessary algorithms for calculating energy flows throughout the building, including those caused by daylight entering the building. At the very least, the program should produce output data in a form that can easily be used by other programs.
It must include effective algorithms that produce results with small errors in a short time.

Most PC-based lighting design programs today are very fast because the calculation square net is wide and fixed. However, as a result, the accuracy of the calculations is compromised. Programs would perform better if they utilized varying square patterns, where the squares would vary according to the variation of the illuminance and luminance: The greater the variation, the smaller the squares, and vice versa. Calculations with such algorithms are much more accurate and less time consuming.

It should also be possible to introduce time-saving algorithms for calculating the inter-reflections and the "shadowing" effect of the luminaires. (In principle, each inter-reflection could be calculated an indefinite number of times. Algorithms that simplify these calculations while still providing realistic and accurate results are possible to define.)
It must handle rooms of any shape with furniture and other objects. A furniture library could be created to save time.
It must be able to make use of all kinds of luminaries on the market. The ways in which data on luminaires and their luminous intensity distribution are represented vary depending on the country and company of origin. Facilities should be included in the program that make it possible to accept all major existing representation systems as input data.
It must provide the option to calculate the illuminance in arbitrary directions, luminance on all surfaces, contrast reduction factors in user-defined directions and on user-defined surfaces, and glare indices in arbitrary directions and positions.
The input formats should be easy to handle, including a CAD (Computer Aided Design)-module.

If it is complicated and time-consuming for the operator to enter data, the cost in work time will be high, thereby reducing the attractiveness of the program. By using DXF-format (a general format for transferring files between different CAD-programs) and other kinds of standard formats it should be possible to use whole computer-created drawings as input to the program.
Output formats of the program are especially important for end-users and clients.

These formats should include figures, tables, and graphical displays in color in such a way that consultants, users, and clients can easily comprehend the results. Today's computer graphics (even on PCs) used with modern plotters and printers can produce excellent paper copies in color and of high quality.

TWO INTERESTING PROGRAMS
In our market survey only two programs (or program packages) were found to have the potential to become very useful in the next few years, based on the above criteria.

One of them is Lumen Micro from the Colorado-based firm Lighting Technologies Inc. This PC-based program has been on the market for a short time and partly meets the requirements of the above list.

The other program is Adeline, which is actually a package of four inter-linked programs. This package is the result of international teamwork at the German Fraunhofer Institute under the auspices of the International Energy Agency (IEA).

In the Scribe program module the room and building properties are worked out. In the Superlite program only daylighting effects are calculated. Programmers plan to include artificial lighting effects in the next version. The Superlink module evaluates the lighting-associated energy flows in the room, providing output in a format that can be utilized by other energy analysis software. The fourth part of Adeline is Radiance, probably the most advanced program in the world for lighting design purposes. With this application photo-realistic color renderings of the result can be printed. Several of the possibilities here go far beyond the demands discussed above. For example, all other PC-programs for lighting design make use of the Lambert model to calculate reflected radiation. The Lambert model treats the surface as being totally diffuse and thus radiating the same amount of flux in all directions. But the Radiance program can evaluate the real reflectance properties, i.e., blended reflectance as well as specular and total diffuse reflectance.

The technical specifications outlined here ultimately represent the need for good and energy-efficient lighting, as expressed by end-users and specifiers in the NUTEKlarge-scale office lighting demonstration project. It is our hope that the articulation of these requirements will contribute to and help speed up the ongoing development of useful lighting design software.

For a discussion, see letters 3-4/95.

Adeline has a home page: radsite.lbl.gov/adeline/HOME.html

Nils Svendenius
Peter Pertola

Nils Svendenius is an assistant professor of Atomic Spectro- scopy, Department of Physics, University of Lund, Sölvegatan 14, S-223 62 Lund, Sweden.
Tel: +46 46 10 77 32
Fax: +46 46 10 47 09

Peter Pertola is a lighting designer at Pelk AB, Drottninggatan 71 C, S-111 36 Stockholm, Sweden.
Tel: +46 8 20 72 32
Fax: +46 8 24 21 44

Top of page