Re: STL for ever?

From: Stephen Rock (rocks@rpi.edu)
Date: Thu Dec 16 1999 - 19:17:33 EET


Dear All:

Some have listed positive attributes of the STL data exchange format:

> "STL- Format has also big advantages:
> 1. Public
> 2. Available in nearly all Systems
> 3. Easy to use
> 4. Flexible in use

<snip...>

> 5. Robust in transfers
>
> My experience with STL in foundry work suggests that it is a robust format
> for file transfers in getting fabber business done. If you need better
> detail, you simply go to a larger file (more bytes), and today's computers
> and data transfer systems handle them with ease.

It's great to hear that, at least for some users, problems dealing with STL
files have become a thing of the past. However, even today this is not
universally the case. Missing facets, internal facets, and problems due to
numerical rounding inconsistencies can still cause difficulty and there is
nothing offered by the STL format that makes these problems difficult to create
or even easy to test. Any appearance of being robust is likely owed more to the
credit of the software responsible for converting (tessellating) your
geometrically precise CAD representation into an approximate representation
using triangular facets). When we say "3. Easy to use", it might be more accurate
to say STL is easy to GENERATE -- at least to generate poorly. Downstream CAD
model processing using a set of facets devoid of any relationship information
requires much more effort to validate, possibly correct, and slice.

If an alternative format was defined that required less than
half the size of an STL file and also included topological information (that makes
validation and slicing much more efficient), would any RP software developers
be interested in supporting it? If low-cost or free conversion utilities were
offered to convert between STL->NewFormat and NewFormat->STL, would any end users
be interested in using the format to reduce storage requirements (despite the
seemingly exponential increase in storage availability)?

Any feedback appreciated,
=======================================
Stephen J. Rock
CII8015
Center for Automation Technologies
Rensselaer Polytechnic Institute
Troy, NY 12180 USA
(518)276-8652 Fax -4897
http://www.rpi.edu/~rocks
=======================================

For more information about the rp-ml, see http://ltk.hut.fi/rp-ml/



This archive was generated by hypermail 2.1.2 : Tue Jun 05 2001 - 22:53:47 EEST