Re: FW: Future of RP (STL Files vs. Alternatives)

From: Stephen Rock (rocks@rpi.edu)
Date: Sun Dec 06 1998 - 03:45:24 EET


9410561 a d bhatt pg me wrote:

> STL is acceptable to all rp machines does not mean that we should not think
> beyond it. The hardware capabilities of some of the machines is to produce
> an exact slice where as in STL we get only a polygon contour. We can not make
> a part better than the software model. Thus I strongly feel that being too
> much dependent on STL is hindering the progress of RP.

There is plenty of opportunity for improving the STL format, and facet
representations of solids may be less elegant than representations using
higher-order geometry (e.g. NURBS). But is the use of STL files really
what is hindering RP progress?

Computer storage and memory are plentiful. Compute power keeps
growing exponentially while prices fall. The precision attainable using a
facet-based representation (I'm not a proponent of the STL format in particular,
see http://www.rpi.edu/~rocks/PUBS/SFF91-FlexFormat-Abstract.html ) can
be increased by tessellating the original geometry more finely with a
corresponding increase in file size. Additionally, facet-based representations
may be obtained from sources other than CAD databases such as from
CT scanners, range sensors, radar data, ...

Meanwhile, layer-based RP hardware continues to build using discrete
layers, with limited mechanical tolerance in the build plane, and
often with a one-size-fits-all "paint-brush" (e.g. the spot of a laser beam,
the droplet size from a jet, or the diameter of an extruded fillament) which
naturally limit process throughput absent parallel operation.

For facet-based geometric representations to be limiting the progress of RP,
it seems we would have to see improvements in RP process hardware
(layer thickness reductions, material deposition precision and throughput
increases, and cost reduction) outstripping improvements in computer
hardware. With no disrespect to the improvements made in RP since
its introduction, it seems the economies of scale driving information
technology have made possible more rapid price-performance
improvements.

There are certainly potential benefits of using alternatives to STL and
facet-based representations, but there are also drawbacks. What do
others think about this subject? It is interesting that this debate has
been ongoing for years, recurring from time to time in the rp-ml
forum, but appears to remain an open issue.

Regards,
Steve
=======================================
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 - 22:47:32 EEST