More efficient supports

[montressor]

Would love the option to use more efficient support structure creation, perhaps as a plugin, to cut down on material usage and print time

http://www.autodeskresearch.com/pdf/Sup ... esSG14.pdf
http://hpcg.purdue.edu/bbenes/papers/Vanek14SGP.pdf

[jimc]

meshmixer style support. it looks cool but i dont know anyone who has used it for more than 2 weeks, myself included. when meshmixer first came out it was like a fad and everyone was using it. it didnt take too long for everyone to figure out it wasnt good and revert back to their original slicer. with the dense support layers in s3d it really doesnt get much more efficient than that while still being reliable. set it to 10-15% infill with 3-4 layers on top that are 50%. i highly doubt you will ever see it in s3d because to entire support system would have to be reinvented. its just not how s3 works. of course this is all my opinion and to each his own but if you like meshmixer style support then by all means go for it.

[tenaja]

Would love the option to use more efficient support structure creation, perhaps as a plugin, to cut down on material usage and print time

http://www.autodeskresearch.com/pdf/Sup ... esSG14.pdf
http://hpcg.purdue.edu/bbenes/papers/Vanek14SGP.pdf

+1 this request!

meshmixer style support. it looks cool but i dont know anyone who has used it for more than 2 weeks, myself included. when meshmixer first came out it was like a fad and everyone was using it. it didnt take too long for everyone to figure out it wasnt good and revert back to their original slicer.

Poor implementation does not invalidate a valid method! If you read the Perdue article, they even compare their approach to Meshmixer, and discuss its failures.

BTW, how long has it been since you have tried Meshmixer? I have not tried it, but I most certainly have excellent experience with the concept, on a weekly basis!

FormLabs (maker of the Form1) upgraded to the Meshmixer support method, and I am certain they will never go back! The supports are much easier to remove, with efficiency gains shown in the article. PreForm is most certainly an example of a successful implementation!

Perhaps Meshmixer has improved since its implementation.

[montressor]

[quote="tenaja"]Poor implementation does not invalidate a valid method!/quote]

Well said, and I wholeheartedly agree.

[jimc]

Poor implementation does not invalidate a valid method!/quote]

Well said, and I wholeheartedly agree.

sure thing, i totally agree with that. i didnt read the article and what it was proposing. just that meshmixer is the only software i am aware of that has those style supports and it was finicky and unreliable every time i used it and everyone i know who did use it abandoned it a long time ago as well. s3d's current support imo is far superior as well as most of the other slicers out there. they are just to fragile and generally print poorly because of that which makes them quite inaccurate. i know form 1 does this but that a whole different ballgame with resin. your not shaking a hot end around a bed knocking off little twigs that arent where they should be because of plastic shrinking and warping. that style of support could work but it just needs to be more substantial and rigid. this however would defeat the purpose that i think montressor was shooting for with using less material. now to put it all in perspective and none of these are hard figures or anything. just approximation for discussion purposes. lets say you have 2 of the same models and one you do with s3d support, 15% with a 50% top 3 layer. the other model support is generated through meshmixer. let say the meshmixer support takes 2 meters of filament to print the support material. now say the s3d support take 3 times as much material to do the same job and clocks in at 6 meters. a 4 meter difference with current abs prices comes out to about 25 cents. i dont know your actual financial situation but im not losing sleep over 25 cents and 4 meters of filament. just one failed print from little twigs getting knocked over on a tall print will far exceed those losses anyway. guys dont take any of this the wrong way. i am not trying to make any kind of argument here or tell you guys you are wrong or anything like that. just having a interesting discussion on support types. everyone has their own preference and thing that works for them. i dont have anything to do with s3d so im not any kind of authority but from a feature request perspective and from what i know of the software i wouldnt hold my breath on such a drastic rework of the s3d support system.

[montressor]

For me, the cost isn't a huge issue. Waste is: it's bad enough we're printing with plastic (at least with ABS), but wasting a bunch of that plastic kinda bugs me. More important is the time: on some prints the support drastically increases the print time. Seems a more efficient way of creating supports (assuming it works) is a win all around. I don't think S3D's current supports are bad, but a new approach that cuts everyone's consumable usage - some of which is petroleum-based products - and improves print times is worth considering.

[Bassna]

Meshmixer's support system is actually really good for some thing's. You just need to adjust some setting's in the support settings. Here is a great article explaining it, http://www.extrudable.me/2013/12/28/mes ... ting-role/ . Making the base thicker, more connecting points, thicker tips, etc. I have not really used it since I purchased S3D, mainly because I want to get my money's worth out of using S3D I guess. But from what I did use Meshmixer with customized support's, it worked great.