Been thinking whether now that we have cheap SLA (resin) 3d printers, f-rep (SDF) modeling might make more sense?

In f-rep modeling with e.g. libfive many modeling operations are much simpler to implement (e.g. boolean ops), but for traditional FDM printing you still have to go to the slicer with a STL file or similar, so you have to go through the whole marching cubes or whatever thing and have a surface mesh which is kind of a bummer

Follow

with SLA you could just sample the SDF to get a bitmap for each layer which is really straightforward. No polygons involved!

Otoh I don't really know what goes on in an FDM slicer. Do they even use the polygons? Or did they just inherit the most common representation? After all they also bork out when an input model isn't watertight etc...

@s_ol what would be interesting is implementing the slicer directly onto the sdf representation. distance to surface is an actually relevant measure here.

@phooky @s_ol years ago before i knew about any of this 3D printing stuff, i used povray to generate a bunch of bitmap outline slices of an object that i 2d printed and carefully cut out layer for layer.
i got bored real quick but it did totally work.

critically it’s actually easier to slice literally anything other than a mesh- the csg of povray, and SDF have clearly defined interior solid area.

meshes are like fragile baloons. spring leaks if you look at em wrong.

@zens @phooky
I guess you'd still have to polygonize the outlines at some point since the FDM input is vectorial, but the Slicer could be a lot easier probably. Writing a new slicer sounds like a lot of work though - otoh for a SLA printer I'm quite confident I could hack something together if I find a documented protocol for getting the frames into the machine

@zens @phooky
oh wow, guess I figured out the protocol for getting the image data into it 😅

I did *not* expect that... well good I figured that out now, guess I need to take my USB VGA card with me next time I'm in the lab...

@s_ol @phooky sometimes the simplest solutions are the best solutions

@s_ol @zens The tricky parts of an FDM slicer are infill and shells, which you don't have to deal with in SLA. From what I understand the tricky bit with SLA is working out the support material and getting adhesion to the bed but not the tank glass. (FWIW, I've only written FDM slicers, and I've been out of the 3d printing world for a while.)

@s_ol they ordinarily slice the triangles at the cutting plane, take the bag of resulting segments,, and stitch them back into closed polygons. That's why holes in the mesh are so troubling; if you don't have a closed curve you can't determine outside vs. inside. There are sneaky tricks you can use to guess, but most slicers just put the onus on the user.

Sign in to participate in the conversation
Merveilles

Revel in the marvels of the universe. We are a collective of forward-thinking individuals who strive to better ourselves and our surroundings through constant creation. We express ourselves through music, art, games, and writing. We also put great value in play. A warm welcome to any like-minded people who feel these ideals resonate with them.