Abstract
This report evaluates the daylight simulation capabilities of Autodesk 3ds Max Design 2009 and Daysim 3.0 by comparing their results to real indoor illuminance measurements in a sidelit space. The study involved five different fenestration and glazing configurations under various sky conditions. Both programs used external irradiances as input to predict indoor illuminances. The results showed that 3ds Max Design and Daysim provided comparable reliability, although 3ds Max Design underestimated light flux through translucent panels, and Daysim was less accurate with venetian blinds.

1. Introduction
The introduction discusses the Radiance backward raytracer, a highly regarded lighting simulation program, and its validation through various studies. The paper highlights the lack of similar validation studies for other programs and introduces a new data set from the NRC Daylighting Laboratory to address this gap.

2. Methodology
2.1 Daylighting Test Cases
The test cases were conducted in the NRC Daylighting Laboratory's East room, featuring different configurations such as a basic sidelit space, lightshelf, translucent panels, and venetian blinds. Measurements were taken using calibrated Licor illuminance sensors and outdoor irradiances were collected using a Yankee rotating shadowband radiometer.

2.2 Daysim Simulations
Details on the simulation procedures for Daysim, which uses the Perez sky model and Radiance backward raytracer.

2.3 Autodesk 3ds Max Design Simulations
Details on the simulation procedures for Autodesk 3ds Max Design, which uses Exposure technology and mental ray software raytracer.

3. Results
3.1 Façade Illuminances
Comparison of simulated and measured illuminances for different test cases.
3.2 Base Case (TC1) and Lightshelf (TC2)
Analysis of results for the base case and lightshelf configurations.
3.3 Translucent Glazing (TC3)
Evaluation of simulation accuracy for translucent glazing.
3.4 Venetian Blinds (TC4 and TC5)
Assessment of simulation performance with venetian blinds.
3.5 Error Analysis
Discussion of mean bias errors and root mean square errors.

4. Discussion
4.1 Practical Considerations
Considerations for practical application of simulation results.
4.2 Modeling Movable Shading Devices
Challenges in simulating movable shading devices.
4.3 Autodesk 3ds Max Design and Daysim/Radiance
Comparison of the two programs.
4.4 Other Lighting Programs
Brief mention of other lighting simulation programs.

5. Conclusion and Outlook
The study concludes that both programs are sufficiently accurate for typical daylighting design investigations, with some limitations noted in specific configurations.

Material Properties Overview
The document provides detailed descriptions and measurements of various materials used in the NRC Daylighting Laboratory. Each material's reflectance properties are measured using a Minolta CM2500d spectrophotometer, and the results are presented in terms of diffuse and specular reflectance. Key materials include:

• Interior Back Wall: 77% diffuse, 0.4% specular reflectance.
• Interior Ceiling: 88% diffuse, 0.1% specular reflectance.
• Interior Floor: 12% diffuse reflectance, no specular reflectance.
• Exterior Parking Lot: 11% diffuse reflectance, no specular reflectance.
• Double Clear Glazing: 66.1% visual transmittance, 72.0% transmissivity.

Sky Conditions and Measurements
Measurements were taken under various sky conditions, categorized into sunny and cloudy skies. The document provides a table summarizing the number of sky conditions for each test case, with a total of 2314 conditions for TC 1 (No Shading) and 3107 for TC 3 (Translucent Panel).

Daysim Simulations
Daysim, a Radiance-based daylighting analysis tool, is used to simulate daylight conditions. It employs a daylight coefficient approach to efficiently calculate illuminance or luminance time series under varying sky conditions. The document notes that Daysim results are similar to Radiance Classic, especially under overcast conditions.

Autodesk 3ds Max Design Simulations
3ds Max Design uses Exposure technology for lighting calculations, incorporating the Perez Sky Model. The document details the use of mental ray raytracer for global illumination, employing both forward and backward raytracing techniques. Simulation times for indoor illuminances under a single sky condition range from 6 to 12 seconds.

Simulation Parameters
Tables in the document list the simulation parameters used for both Daysim and 3ds Max Design. These parameters were optimized for accuracy and simulation time, ensuring consistency across all test cases.

Specifications and Parameters
The document discusses the simulation parameters used in 3ds Max Design, focusing on the number of rays per FG Point and Diffuse Bounces, which are crucial for achieving accurate lighting simulations. A sensitivity analysis indicated that beyond 2000 rays per sample, the accuracy was acceptable. Light sensors in 3ds Max Design cast eight times more rays than set in the Render Setup Dialog.

Simulation Parameters
Key parameters include enabling Raytracing, setting Max Trace Depth, Reflections, and Refractions to 10, and enabling Final Gather with specific settings for Rays per FG Point and Diffuse Bounces. Caustics and Global Illumination were disabled.

Results and Comparisons
The document compares simulation results from 3ds Max Design and Daysim with measured illuminances. Both programs showed similar results under sunny and partly cloudy conditions, with some divergence under rapidly changing cloud cover.

Case Studies

• Base Case (TC1) and Lightshelf (TC2): Simulations accurately modeled the effects of window mullions and lightshelves, with some differences in peak illuminance predictions.
• Translucent Glazing (TC3): Daysim closely followed measurements, while 3ds Max Design showed a constant offset, likely due to incorrect diffuse transmittance settings.
• Venetian Blinds (TC4 and TC5): Both programs modeled complex fenestration systems with some challenges, particularly in accurately simulating sunlight reflection off blinds.

Error Analysis
The document provides a statistical analysis of simulation errors, using Mean Bias Error (MBE) and Root Mean Square Error (RMSE) to compare simulations with measurements. It highlights significant errors in some test cases, particularly due to time shifts in peak illuminance predictions.

Conclusion
The document concludes that while both simulation programs capture the general lighting conditions, there are areas for improvement, particularly in modeling complex systems like venetian blinds and accurately predicting peak illuminances.

Overview
The document discusses the performance of two lighting simulation programs, 3ds Max Design and Daysim, in comparison to measured data across various test cases involving different shading devices. It highlights the accuracy, limitations, and practical implications of using these tools for daylighting analysis.

Key Sections

• Simulation Accuracy: The document presents Mean Bias Errors (MBEs) and Root Mean Square Errors (RMSEs) for different test cases, indicating that large errors are often due to specific conditions like the 'shift effect' and underestimation in simulations.
• Practical Considerations: It emphasizes the importance of user expertise in achieving reliable simulation results and discusses the generalizability of the findings to other buildings. The document suggests that both programs can approximate interior lighting levels effectively under various conditions.
• Daylight Performance Metrics: The document explores the use of climate-based metrics such as Daylight Autonomy (DA) and Useful Daylight Illuminance (UDI) to evaluate lighting performance. It concludes that both programs are capable of supporting design decisions based on these metrics.
• Modeling Challenges: The difficulty in accurately modeling complex fenestration systems like venetian blinds is discussed, highlighting the impact of occupant behavior on simulation accuracy.
• Comparison of Programs: The document compares the capabilities of 3ds Max Design and Daysim, noting that both are based on similar models but have different strengths in terms of simulation speed and accuracy under various conditions.
• Future Directions: The document suggests improvements such as implementing a daylight coefficient approach in 3ds Max Design to reduce simulation times for annual daylight simulations.

Conclusion
The document concludes that both 3ds Max Design and Daysim are viable tools for daylighting design decisions, with each having specific advantages depending on the complexity of the scene and the type of analysis required.

Introduction
This document discusses the use of 3ds Max Design and Daysim for daylighting simulations, emphasizing the reliability of NRC daylighting test cases. It suggests improvements for future test cases, such as including high dynamic range (HDR) images to better distinguish between sky model and global illumination errors.

Daylighting Test Cases
The authors found the five NRC daylighting test cases reliable but noted the need for additional parameters. Future versions should include HDR images of sky luminance distributions and inside views to improve error separation and quantify typical error ranges for glare evaluation.

Radiance and Daylight Coefficient Approach
The new rtcontrib routine within Radiance offers an alternative daylight coefficient approach, which can process sky conditions efficiently.

Conclusion and Outlook
The study concludes that 3ds Max Design and Daysim are effective for daylighting design decisions in complex scenes. The authors anticipate more simulation programs will undergo validation using NRC test cases, increasing the number of validated programs. Challenges remain in identifying suitable daylight performance metrics, refining occupant behavior models, and developing databases for complex fenestration systems.

Acknowledgement
This work was funded by Autodesk Media & Entertainment and the National Research Council Canada. The authors thank various contributors for their support in preparing the NRC Daylighting Laboratory and providing valuable feedback.

References
The document cites various studies and standards related to daylighting systems, simulation validation, and performance metrics, including ASHRAE Standard 140 and research by Mardaljevic, Reinhart, and others.

Appendix
The appendix includes base geometry and descriptions of test cases involving lightshelves, translucent panels, and Venetian blinds, along with a frequency distribution of relative errors for these cases.