Slicing issue with textures


I am running into the following issue when trying to slice this model when using a texture. These random white artifacts are getting placed near the top of the model and are in the sliced gcode.

towersymmetric.stl (490.0 KB)


Hi @pyroclasmx, are you using the very last update of Voxelizer (Voxelizer Experimental release notes 2019-01-18)? The algorithm has been completely rewritten and these kind of errors should not happen anymore.


I am running V2.1.0-2019-01-18-all


Ok then could you save and share with me the Voxelizer project? I will have a look a it.


Tower.vox (6.6 MB)
Here you go, I noticed your screenshot doesn’t look quite right either with a large white area and not bricked


Thanks. All right, the random lines are definitely a bug so as the white parts, I will work on that to fix it.
Anyway in general I can tell that for this model the path width you chose is too big and the lines in most of the places don’t have the space to cross each other. You should choose a thinner one or use a model with thicker walls (or you could try the thickening filter on the same one, 1mm will be enough).


Thanks, I would love to be able to print this when you fix the bug, for the path widths I probably just need to adjust the layer height, at this point I think it just went with the last print I did which I did print higher than normal. With my .5 nozzle I usually print at .2 - .25 layer height. I think this model is set to print at .4 height which increases the path width in the software.


The issue with the random lines should be solved. Looking at the code, the white parts were not completly a bug, but related with a decision we made some time ago not to map the inside parts of an object. On Monday we will go through this again. Now it looks like this:


That looks great! I can’t wait to give it a shot,

I would argue that while technically it would appear enclosed at the sliced level it is an outward face as a whole. Maybe would be worth checking if infill would potentially be applied or not in the empty space to see if it should be textured?

Thanks for looking into this bug!


My pleasure;)

Yes we look at the sliced level to know if an outline is inside or outside. The decision we took was due to the fact that if you want to map a logo on, say, a cup, you probably don’t want it also on the inside but this is obviously not always true, like in your case. Probably the best thing will be to add a new parameter for the material and let the user decide where he want the mapping to happen.


Awesome! I’d also be interested in being able to adjust the uv mapping of the texture and cube mapping but I will add that in the suggestions area.


In the next release there will be an option, off by default, to choose whether to map only the outside of the object or not.

You didn’t write about this yet, what do you mean? :slight_smile:


Sorry I got distracted with some coding work of my own :slight_smile:

Right now you have cylindrical and 2 other maps. Those work great but a cube map would be good for box like objects like the tower. I think with a box mapping the ramps inside the print wouldn’t get warped lines. Moving UV would allow me to position the texture how I want it on the model. I saw the ability to scale the texture but not adjust it on the model.



Ok we will consider it.

Aren’t the three rotations OK for this?


I posted a youtube video addressing why the rotation of the projection does not work well in this specific case.


Ok I see, will try to do something, this or next week.


Awesome, I hope that was a helpful explanation.


I installed the latest version and there are no more white lines cutting through the dice tower with the brick texture. Looks great!


Cool, looking forward to see the printed model :slight_smile:


I’m having an issue that may be related. When slicing a model with 3 color texture, the final slice includes 4 colors, one of which the model did not even contain. The colors I chose are on the outside, while the color I did not select appears to be on the top infill and possibly all of the infill.