|
Syntopia
|
 |
« Reply #150 on: March 23, 2014, 09:25:15 PM » |
|
Hi bib, it is a separate program as of now. I'm experimenting a bit with Java, because there are some nice mesh libraries (in particular Hemesh), that I want to explore in order to improve the mesh quality.
|
|
|
|
|
Logged
|
|
|
|
|
taurus
|
|
« Reply #151 on: March 23, 2014, 11:04:41 PM » |
|
Nice to see, that there is still motion in preparation of printable meshes. I remember my expierience with mystic fractal programs, which export a native mesh, without the "cubified" voxel stack approach. My problems with voxel stacks are not the quality of the initial mesh. It is the lack of possibilities when it comes to optimisation. I described it here in detail. http://www.fractalforums.com/format-printing-and-post-production/shapeways-for-3d-printed-fractals/msg55054/#msg55054So my question is to you Syntopia. Does your Program export triangulated point meshes, or are they block-meshes like voxel stacks are? |
|
|
|
|
Logged
|
when life offers you a lemon, get yourself some salt and tequila!
|
|
|
thargor6
Fractal Molossus
Posts: 789
|
|
« Reply #152 on: March 24, 2014, 12:38:13 AM » |
|
Thanks for the hint about Hemesh, as I still have mesh-export for my software in mind. Flame-fractals as real objects, a thing I'm really fascinated of, all the time :-) And thanks to all of you for investigating this 3D-printing-stuff, it is overall very fascinating, but I had no time for it, yet
Andreas
|
|
|
|
|
Logged
|
|
|
|
|
Syntopia
|
|
« Reply #153 on: March 24, 2014, 09:39:11 PM » |
|
Shapeways actually has a quite impressive tool for working with voxels, called ShapeJS: http://shapejs.shapeways.com/It sends small JavaScript-programs to a Shapeways-server which evalutes them and creates a polygon representation. Interestingly, it uses a Java OpenSource API for doing the voxel computations on the serverside, called AbFab3D: http://abfab3d.com/, which might be useful to the ones doing Java-stuff here at the forums. It seems to be doing a good job on e.g. the Gyroid: http://shapejs.shapeways.com/examples, try comparing with my own Marching Cubes export: https://twitter.com/SyntopiaDK/statuses/434336584452685825Nice to see, that there is still motion in preparation of printable meshes. I remember my experience with mystic fractal programs, which export a native mesh, without the "cubified" voxel stack approach. My problems with voxel stacks are not the quality of the initial mesh. It is the lack of possibilities when it comes to optimisation. I described it here in detail. http://www.fractalforums.com/format-printing-and-post-production/shapeways-for-3d-printed-fractals/msg55054/#msg55054So my question is to you Syntopia. Does your Program export triangulated point meshes, or are they block-meshes like voxel stacks are? My software creates a triangle mesh based on uniform grid sampling of a distance function - done using the Marching Cubes algorithm, which is perhaps the most wellknown meshing algorithm. So it is not terribly novel. I'm hoping to be able to optimize and reduce these triangles in a better way that Meshlab allows, but I am beginning to doubt that there are much better methods then the ones implemented in Meshlab - I tried a few other academic remeshers (found on the net) and these did not produce better results. |
|
|
|
|
Logged
|
|
|
|
|
taurus
|
|
« Reply #154 on: March 25, 2014, 04:03:46 PM » |
|
My software creates a triangle mesh based on uniform grid sampling of a distance function - done using the Marching Cubes algorithm, which is perhaps the most wellknown meshing algorithm. So it is not terribly novel. I'm hoping to be able to optimize and reduce these triangles in a better way that Meshlab allows, but I am beginning to doubt that there are much better methods then the ones implemented in Meshlab - I tried a few other academic remeshers (found on the net) and these did not produce better results.
Thanks so far for the info Syntopia.  I guess the raw marching cube algorithm creates simillar problems in optimization as voxel stacks. I read some about simplifications and improvements of marching cubes, but my knowledge is far too little, to take any advantage that. In the end I guess, I need to contact Terry Gintz and ask him, what kind of wizzardery he is using, to create his meshes... >>edit. hm, the more I read the more interresting this marching cubes stuff is getting. maybe I judged too early. Looking forward to see future developments... |
|
|
|
« Last Edit: March 25, 2014, 04:24:38 PM by taurus »
|
Logged
|
when life offers you a lemon, get yourself some salt and tequila!
|
|
|
tomot
Iterator
Posts: 179
|
|
« Reply #155 on: March 25, 2014, 06:43:33 PM » |
|
Shapeways actually has a quite impressive tool for working with voxels, called ShapeJS: http://shapejs.shapeways.com/It sends small JavaScript-programs to a Shapeways-server which evalutes them and creates a polygon representation. Interestingly, it uses a Java OpenSource API for doing the voxel computations on the serverside, called AbFab3D: http://abfab3d.com/, which might be useful to the ones doing Java-stuff here at the forums. It seems to be doing a good job on e.g. the Gyroid: http://shapejs.shapeways.com/examples, I'm curious, what use can be made with the output.x3db file format? It seems strange to me develop yet another 3d file format. Don't we have enough of them already?  |
|
|
|
|
Logged
|
|
|
|
|
|
|
|
|
taurus
|
|
« Reply #158 on: May 16, 2014, 07:33:33 PM » |
|
Very nice! Seems, that incendia provides much sharper models, than the fuzzy voxel stacks of m3d. Really looking good...
|
|
|
|
|
Logged
|
when life offers you a lemon, get yourself some salt and tequila!
|
|
|
|
bib
|
|
« Reply #159 on: May 16, 2014, 11:02:17 PM » |
|
Yes, Incendia exports directly in STL with a choice of algorithms to produce an optimized mesh. If you draw a parallel with traditional rendering, it's like M3D+Fiji produces an aliased image, which is not a problem if you render at high resolution and then smooth and scale down, whereas Incendia produces an already "anti-aliased" mesh, which somehow simplifies the workflow. But Incendia has some drawbacks too...
Thanks!
|
|
|
|
|
Logged
|
Between order and disorder reigns a delicious moment. (Paul Valéry)
|
|
|
|
bib
|
|
« Reply #160 on: June 07, 2014, 07:39:23 PM » |
|
<![CDATA[<a href="http://www.youtube.com/v/Zd6ceu17kHM&rel=1&fs=1&hd=1" target="_blank">http://www.youtube.com/v/Zd6ceu17kHM&rel=1&fs=1&hd=1</a>]]>Model made with Incendia Size: 24.5 cm / 9.6 in. Wire thickness: 0.8-1mm Density: 0.22%    https://www.shapeways.com/model/1980200/apollonian-tetrahedron-xxl.html
|
|
|
|
« Last Edit: June 07, 2014, 07:40:57 PM by bib »
|
Logged
|
Between order and disorder reigns a delicious moment. (Paul Valéry)
|
|
|
LMarkoya
Strange Attractor
Posts: 282
|
|
« Reply #161 on: June 08, 2014, 02:12:34 AM » |
|
Beautiful stuff Bib....and so delicate....
Nice job
|
|
|
|
|
Logged
|
|
|
|
|
mclarekin
|
|
« Reply #162 on: June 09, 2014, 03:00:00 AM » |
|
|
|
|
|
|
Logged
|
|
|
|
|
bib
|
|
« Reply #163 on: July 04, 2014, 02:09:50 PM » |
|
If you understand French, here's an interesting article which explains that fractals (the Menger cube in particular) have been at the inception of the invention of stereolithography (3D printing) back in 1984. http://i3dactu.fr/on-a-retrouve-les-vrais-inventeurs-de-l-imprimante-3d/ A couple of months ago I had the chance to meet Professor André, the true inventor of stereolithography, who told me the research was now focusing on volumetric 3D printing (as opposed to layer by layer as it works today)... |
|
|
|
« Last Edit: July 04, 2014, 02:14:06 PM by bib »
|
Logged
|
Between order and disorder reigns a delicious moment. (Paul Valéry)
|
|
|
LMarkoya
Strange Attractor
Posts: 282
|
|
« Reply #164 on: July 05, 2014, 05:40:50 PM » |
|
All great stuff, don't quite understand volumetric printing though....will you start with a volume and remove mass? not sure how that would work....but great stuff
|
|
|
|
|
Logged
|
|
|
|
|
October 24, 2014, 12:53:48 PM
Dilber MC Theme by HarzeM