Bug in 2.3.0
I have an STL file [I downloaded it from thingiverse] that looks ok, but when I try to slice it Prusaslicer goes into some sort of loop [eating up all 8 of my win10 system's CPUs] and it never completes [at the least , not while i was waiting: it stops with the green progress bar about 1/4 of the way across. I've attached the STL file.
RE: Bug in 2.3.0
NO attachment. You may want to zip it before attaching, or include a link to TV.
RE: Bug in 2.3.0
You can’t attach a stl file directly. It’s not an allowed file type. You need to zip them up into an archive before attaching them.
RE: Bug in 2.3.0
Sorry either i missed it or it didn't tell me that it was ignoring my .stl file. Now attached as a ,zip
RE: Bug in 2.3.0
Hmm.. I still don't see the file attachment... trying again... It definitely says that the .zip file is attached. I must be doing something wrong. ??? Ah, I see -- the file is 10.2 megs and the max allowed is 10 mb. It'd be nice if the forum software told you that a file wasn't OK. I've tried a 7zip file -- 7z gets it down to 6 megs... sorry for the incompetence
RE: Bug in 2.3.0
It's currently slicing in my copy of Prusaslicer, and yes, it's sitting at 25% with 100% cpu usage.
But I think it's going to finish—I hope — at some point. This model has almost one million facets, it's huge. I think it's just a LOT to analyze. Or it may well be too complex for Prusaslicer to handle.
For the fun of it, I just started a slice job in Cura. So far I'm having the same experience — it's just sitting there slicing....
So I'm pretty sure it's not a bug but a model of such complexity that neither PS or Cura can handle it.
I have other things to do, but if I were you I would just start slicing in the evening to see if it finishes over night...
RE: Bug in 2.3.0
I should perhaps say that it's not so much the number of facets, I guess, but the complexity of the tool path. I just sliced a model with 11 million facets and while it took a couple of minutes, it finished okay.
RE: Bug in 2.3.0
I agree with Fuchsr, that model is very very complex. It uses a massive amount of memory when slicing. I reduced it to 25% of the facets with a decimate modifier in Blender and while it slices at that reduction its still more than it needs so I could reduce the facet count further. It has way more detail than needed for fdm printing to start with.
The other issue with it is that a lot of it wont slice anyway with a standard 0.4mm nozzle profile as the parts are too thin. Even when I scaled it to 185% alot of it was missing or not connected to other parts. Even with detect thin walls enabled which normally I don't have on.
I had to use a 0.1mm xy compensation to get it to slice all connected. So thats 185% scale AND a 0.1mm xy (I turned thin walls off as I was using xy comp).
If you want it at the original scale then I had to use a 0.2mm x/y compensation to get all the perimeters to slice. There's a lot of gap fill too. Not sure how nice it would actually print. Its only 3h23m though lol.
Could possibly use some of Bobs techniques with perimeter widths to get that better but I'm half afraid that bumping up the extrusion widths will make half of it disappear again. Using the variable layer height and perimeters I got it down to 2h40m ish.
If you want the simplified version just lmk and I'll attach it.
RE: Bug in 2.3.0
Thanks for all the info. I don't need this particular one, I was just surprised that it went into a black hole. Thanks for digging into it. Attached is a pretty one that is a lot less complicated and prints nicely