2.7.2 Error encountered didn't before this version.
 
Notifications
Clear all

2.7.2 Error encountered didn't before this version.  

  RSS
Hondaman74
(@hondaman74)
Member
2.7.2 Error encountered didn't before this version.

Good Morning Everyone.  I upgraded to the 2.7.2 this morning when I got the notification update in the app. and now I am getting this error trying to slice a model that I have used before many times with no issues but now getting this error

Voronoi cell doesn't contain source point!

I am probably going to go back to previous version but it's with any printer and any profile that I select.  And it's only with one of the models in the image.  But why now and I getting this error?

Thanks

Posted : 02/03/2024 5:14 pm
FoxRun3D
(@foxrun3d)
Famed Member
RE: 2.7.2 Error encountered didn't before this version.

Why is probably explained by this in the release notes:

In PrusaSlicer, we use Voronoi diagrams as part of several features such as Arachne, multi-material segmentation, gap-fill, and thin-walls. We use the Voronoi diagram implementation from the Boost library because it is both fast and numerically stable. Unfortunately, it rarely produces a non-valid Voronoi diagram for some input polygons (the graph is not planar, missing vertices, etc.), which could cause a crash of PrusaSlicer, artifacts on external perimeters with Arachne or spilled layers with multi-material segmentation.

In 2.5.0, we implemented several mechanisms to detect a non-valid Voronoi diagram, and by manipulating the input, we could ensure that the Voronoi diagram would be valid. These mechanisms were originally implemented only for Arachne, and they were heavily tight with Arachne's data structures.

In this release, we have generalized these mechanisms to be used anywhere in PrusaSlicer. This itself solved many of the spilled layer issues with multi-material segmentation and also one crash during thin-wall generation (#10632).

Why this particular model generates a non-valid Voronoi diagram, who knows. I would open a ticket on the Prusaslicer github site so Prusa can look into it.

Formerly known on this forum as @fuchsr -- until all hell broke loose with the forum software...

Posted : 02/03/2024 11:41 pm
Share: