search

nick-parker / Bread

An experimental slicer for FFF 3D Printers
GNU General Public License v3.0
66 stars 10 forks source link

Fix printing E moves twice for each line #9

Closed michaelnew closed 6 years ago

michaelnew commented 6 years ago

E moves were being printed twice on each line, i.e. G1 X111.742 Y115.362 Z0.225 E2.35796 E2.35796 and now it looks like this: G1 X111.742 Y115.362 Z0.225 E2.35796 I had a couple hangs trying when trying to print and this change fixed it (I think).

On a separate note, do you have any plans to work on this project again? I know there hasn't been updates since 2016, but none of the current popular slicers can do 3d layers like this, and considering the way they're built they probably never will. I wouldn't mind contributing some time into the project either if it helps. It seems like there's a lot of potential for this kind of approach and I'd love to see it develop further.

nick-parker commented 6 years ago

I don't have any current plans to expand Bread, no.

I'm doing similar work now at https://praxismfg.com but that's not much use to you until we expand to wherever you're located.

I don't think 3D layers have much of a future on hobbyist machines, at least in the near term. First of all, they don't work very well on 3 axis machines because you really need to be able to tilt your nozzle. Second and more importantly, there's a big unsolved UX problem.

You either need to write wicked smart heuristics/ML/whatever to do 3D layers automatically, or you're asking users to put more work into configuring a 3D print than they would a typical milling job, and that defies the point of hobbyist 3D printers.

michaelnew commented 6 years ago

Hey thanks for the response and for the input. The need for a 5-axis machine to really utilize the 3D layers makes sense, although I wonder if a thinner extrusion width and a more optimized nozzle tip could alleviate that somewhat. And there are some 5 axis machines out there, so it's in the realm of possibility.

asking users to put more work into configuring a 3D print than they would a typical milling job

I actually don't see this as a problem at all. Compared to a milling job, yes, it's more work. But when you're building 3d assets for, say, a game, you're building a mesh, normal maps, texture maps, an armature, weight-mapping the armature to the mesh, and so on. For most anyone that does 3d modeling the notion of an "object" goes well beyond just a mesh, and defining the curvature of your layers isn't really a huge ask in that context. And for any good 3d print you're putting effort into optimizing wall thickness for the extrusion width, minimizing overhangs, bridges, etc., so a good designer is thinking about pathing and optimization to begin with.

Anyway, I think you've done some really impressive work here and I'd love to play around with it. Not sure yet how much time I'll invest, but at the very least I'd like to appease my curiosity and explore some ideas. I think I'll just work on my own fork, and if you'd like me to submit PRs and get things upstreamed I'll be more than happy to. But if you've moved on and don't really want to spend your time on it that's totally fine too.

Thanks again for the great work and for sharing it.