On 6/7/23 16:13, neri-engineering via Discuss wrote: And thunderbird does not recognize your .eml extension so does not quote it. As for stl's that contain multiple objects, I don't care for how cura re-arranges its multiple copies so close together that the printer may weld them. So I have developed the habit of designing a single piece, then wrapping that up in a loop that makes copies with more separation between the copies. And I've yet to have an error in cura or the printer from feeding it 50 copies of a wiring clip as one stl file. Or multiple copies of a half nut it takes the latest AppImage 7 minutes to render. What am I doing wrong? ;o)> Cheers, Gene Heskett. -- "There are four boxes to be used in defense of liberty: soap, ballot, jury, and ammo. Please use in that order." -Ed Howdershelt (Author, 1940) If we desire respect for the law, we must first make the law respectable. - Louis D. Brandeis Genes Web page <http://geneslinuxbox.net:6309/>

Greetings, I am making further progress at looking into OpenSCAD/CGAL issues of sub-optimal polyhedron combination. This email is not meant to be a "hot-potato" nor is it a destructive criticism; in fact if it were a hot potato then you could say that I'm playing hot-potato with myself, by investigating behaviors of OpenSCAD and assigning further work in this area to myself. The potato is too hot to handle and so I'm letting it cool off a little by throwing it up in the air and then catching it. The best way to improve software is by realizing and admitting its shortcomings, even when those shortcomings are due to shortcomings in software that it's using at a lower level (e.g. CGAL and/or OpenCSG). In the same way, the best way for self-improvement and the best way for an increase in intellectual capacity is by admitting one's own mistakes, and/or realizing bottlenecks that have been created at other levels of society. Sometimes this entails admitting some uncomfortable Truths; for example, when the sum of known World History has so many self-contradictory anomalies that make it clear to any logical thinker that the history we've learned is false, then oftentimes the correct course of action is to begin to realize actual Truth instead of believing history ("his story", "her story"), regardless of whether or not admitting such Truths will cause you to end up imprisoned, due to the fact that said Truth was critical of a group of people known to be controlling our world in every single aspect, including its history. No, I'm not playing hot potato. I have delved into this very difficult task of computational geometry, and am delving further and further. I may end up writing my own version of CGAL in a different language, in a completely different sort of way. This sort of work is right up my alley and I've been craving for such a challenging project. In the general case, to achieve overlaps in polyhedron (in OpenSCAD) the only reasonable solution is to create small "pyramids" at the join faces, as demonstrated in the image below. The right side of image shows the addition of the "pyramid face" on the bottom. (Attachment.) While adding such "pyramid join faces" may seem like a little bit of extra work, even with such tricks, CGAL will still create many undesirable artifacts, in the form of small near-zero-area facets. I have done a study of a specific case of this, please see https://sourceforge.net/p/stl-ops/code/HEAD/tree/cases/openscad/screw-fragment/ which has a README file explaining the matter in great detail. The sample code which exposes the problem is concise and simple to understand, by design. I am beginning to suspect more and more that many related problems could be alleviated if more care were taken in casting between high-precision floating point numbers and low-precision floating point numbers. I seem to recall that STL calls for 32 bit floating point numbers in its format, but nothing would prevent someone from thinking a little bit outside of the box and making one's own rules for self-government. You see, it's this sort of out-of-the-box thinking that will eventually lead us out of slavery. It's this sort of out-of-the-box thinking that they don't want us to participate in. They would rather have you take the vaccine without question, and they would rather have you praise those who forced you into that sort of decision. I am still looking into this problem, to see what low-hanging fruit can be improved upon, if any. I don't even know if OpenSCAD's use of CGAL can be tweaked to improve this matter. I don't even know if CGAL is the culprit, and/or if it can easily be fixed. Still looking into it. Just sharing my insights with others if they want to participate in my understanding of Truth. By this point I have come to a definite conclusion, which is that the way STL is generated in OpenSCAD (through CGAL of course) isn't optimal. When many little artifacts are created like this, the creation of additional artifacts with more successive operations will intensify. The complexity of the problem increases very much, as does the time required to make the computations. We need to nip these sorts of problem in the bud, problems in larger society included. Thanks. -Nerius Anthony Landys Sent with Proton Mail secure email. ------- Original Message ------- On Thursday, June 8th, 2023 at 12:14 PM, gene heskett <gheskett@shentel.net> wrote: > On 6/7/23 16:13, neri-engineering via Discuss wrote: > And thunderbird does not recognize your .eml extension so does not quote it. > > As for stl's that contain multiple objects, I don't care for how cura > re-arranges its multiple copies so close together that the printer may > weld them. So I have developed the habit of designing a single piece, > then wrapping that up in a loop that makes copies with more separation > between the copies. And I've yet to have an error in cura or the printer > from feeding it 50 copies of a wiring clip as one stl file. Or multiple > copies of a half nut it takes the latest AppImage 7 minutes to render. > > What am I doing wrong? ;o)> > > > Cheers, Gene Heskett. > -- > "There are four boxes to be used in defense of liberty: > soap, ballot, jury, and ammo. Please use in that order." > -Ed Howdershelt (Author, 1940) > If we desire respect for the law, we must first make the law respectable. > - Louis D. Brandeis > Genes Web page http://geneslinuxbox.net:6309/ > > _______________________________________________ > OpenSCAD mailing list > To unsubscribe send an email to discuss-leave@lists.openscad.org