Re: Why STL, nearly closing question

From: Bob Yancey (
Date: Tue Jul 29 1997 - 15:38:36 EEST

On 28 July 1997, Leonel Graca wrote:

>Thanks, Kevan and Bob
>So STL files are facets files. Output from CAD to RP or from reverse
>engineering to CAD or CAM.
>But on doing reverse engineering we usually import clouds of points for
>surface reconstruction. I confirmed with our partners who also do
>reverse engineering, and they argue that this far, STL files result
>worse on surface reconstruction then the clouds of points.
>If I could explain the reason why we and our partners work better
>directly over poins then STL to do build CAM or CAD files, opposite to
>you I wold be near to close my doubts.
>Software we use is suposed to be worlds leader cam software, STRIM, from
>Matra Datavision. We also use Euclid 3 and newborn QUANTUM.
>Best Regards

It really shouldn't matter if you are working from a point cloud or an STL
file for surface reconstruction as long as you have the same density of
points as you do triangles. In our work with computed tomography, we
prefer to work with the STL file since it also contains surface normal
information. Therefore, if you have thin walls, you know which points
belong with which wall.

Bob Yancey

Advanced Research and Appplications Corporation (ARACOR)
514 E. Dayton-Yellow Springs Road
Fairborn, OH 45324

(937) 879-4200
(937) 879-3934 (FAX)

This archive was generated by hypermail 2.1.2 : Tue Jun 05 2001 - 22:39:58 EEST