RE: File Format for RP

From: Ian Gibson (
Date: Tue Jun 02 1998 - 04:12:45 EEST

Dear rp-ml

I agree with you Justin. It is an interesting thread. I would add that
there are problems related to how each CAD system generates (and
regenerates after model changes) stl files. It is also somewhat cumbersome
to work with stl data should you require last minute changes.

I think also on this list, people have extolled the virtues of stl (its
simplicity, its relative lack of ambiguity, etc.). I guess if we were to go
to a new standard, we must beware of throwing the baby away with the

This looks like a good subject for further debate. I still havent got a
volunteer to address the software problems associated with RP in my 'future
of RP' internet conference (although the subject of stl may come under the
'standards' debate. Would anyone care to take this on? Its still not too late.


At 01:42 PM 6/1/98 -0400, you wrote:
>I think this is an interesting thread, because so many people have
>problems with STL, but it seems that the RP companies don't seem to care
>too much because they have a lot invested in sticking with STL. Some of
>the problems mentioned so far this week:
>- too many triangles are needed for high resolution models
>- since STL is redundantly redundant ;) these files get TOO BIG!
>- no topology information included
>- no support for color (this may be important someday soon)
>- ...
>I'm interested in knowing what else people think is bad about STL, either
>in terms of specific features (support for "X") or in terms of more vague,
>"I wish an STL file was/had..." type answers. I will try to compile the
>wish list and post back with a comprehensive list. Someone mentioned a
>while ago something about the need for a "3-D Postscript" to replace
>STL...let's try to find out all the things it should do! And remember,
>look to both the present and future of RP for features to include...
>On Mon, 1 Jun 1998, Denton, Karl R. wrote:
>> Al,
>> Agreed! I now spend more in file space (files with several million
>> tri's) in order to get our stl files to a resolution high enough that
>> the triangulation will not show through on our Sanders equipment. Truly
>> a more efficient file would help Williams International.
>> Karl
>> -- [ From: Al Hastbacka * EMC.Ver #2.5.3 ] --
>> Someone Who Is Looking for a New RP File Format,
>> The present STL format is enough in accuracy for the present RP
>> processes with dimension forming accuracy of 0.25mm. STL is still simple
>> , effective and easy to process.
>> As a rebuttal to this comment, the following is my opposing position.
>> In some of our advanced development efforts, the hardware is more
>> accurate than the STL file. Therefor, it would appear that a new file
>> structure or type is needed.
>> Regards,
>> Al Hastbacka
>Justin Kidder, Research Asst. | Automation and Robotics Laboratory
> | University of Pittsburgh
> Home page:
> There are two rules for ultimate success in life:
> 1. Never tell everything you know.
>For more information about the rp-ml, see
Dr. Ian Gibson
Associate Professor
Department of Mechanical Engineering
The University of Hong Kong
tel: (852) 28597901
fax: (852) 28585415

I'd rather have a free bottle in front of me than a pre-frontal lobotomy.

For more information about the rp-ml, see

This archive was generated by hypermail 2.1.2 : Tue Jun 05 2001 - 22:45:47 EEST