Blog Menu

Is OpenRP Another Attempt to Replace STL?

June 26, 2004

Filed under: additive manufacturing — Terry Wohlers @ 10:53

The OpenRP website (http://openrp.com) says “No.” OpenRP is a non-profit initiative aimed at offering the rapid prototyping industry an alternative method of data exchange while maintaining compatibility with the STL file format. Seems like a good idea to me. The OpenRP initiative is also offering free software that reads and writes the files. According to the website, all software will remain free and none of the applications, plug-ins, or libraries will expire. The proposed format was announced in early June 2004. 

Floating Point Solutions Private Limited, based in Goa, India, is behind the effort. The company develops software for CAD, CAM, reverse engineering, and RP, including plug-ins for AutoCAD, SolidWorks, Rhino, and Surfcam. Its website is clear, crisp, and professionally written in English, and it states that the company has 650 customers in 38 countries.

For more than a decade, various industry groups, government agencies, such as the National Institute of Standards and Technology (NIST), and individuals have debated the strengths and limitations of the STL file format. It was created by Dave Albert of the Albert Consulting Group for 3D Systems in the 1987 time frame. Since then, STL has worked well to transport CAD data to RP systems. Using STL, it is possible to export nearly any CAD model (surface or solid) to this triangular faceted format through a basic method called tessellation.

The STL format, however, has been plagued with limitations such as an inefficient method of storing coordinate points due to redundancy. Also, it stores little information about the model, including the unit of measure, such as inches or millimeters.

Can Floating Point Solutions persuade an established industry to consider its new file format? The two problems that the OpenRP initiative is addressing are large file size and the lack of security of the STL format. Neither, however, have been big enough issues to prevent STL from growing to the popularity that it enjoys today. For the OpenRP initiative to succeed, it will need to address more important issues, such as the need for a header that stores information such as the designer’s name, date, notes, and the unit of measure. Also, it would be helpful to store color information from the CAD model, as well as some way of describing surface texture. And finally, it would be helpful to carry both faceted and smooth math data (e.g., NURBS), giving the the recipient of the file the opportunity to choose. Unless the initiative can consider these and other features, OpenRP is unlikely to advance data exchange for users of RP.