KISSlicer and Slic3r gcode observations

[mhackney]

I've just started looking at KISSlicer but I've been a Slic3r user since it's inception. I've heard good things about KISSlicer so I wanted to see for myself. I have not printed anything yet but I have sliced the same object in both slicers in order to compare gcode. What I can say is that KISS cuts up segments into very short lengths. This creates larger gcode files. However, it also results in very precise extruder control. It is remarkably shorter than an equivalent Slic3r output. Extraction moves are typically on the order of .01mm in KISS and .1mm in Slic3r - an order of magnitude larger. Here are 2 screenshots scaled to the same with the code selected so you can see the difference:

KISSlicer:

Slic3R:

For one thing, these short segments give much smoother arcs than the longer ones generated by Slic3r, resulting in better print quality. I also think (and intend to do some tests) that they also provide greater predictability over how much material is deposited and where.


Anyway, I thought this was an interesting observation that I wanted to share.

[Flateric]

This is really great info, thanks for taking the time to do it and upload the pics as well.

I would assume, perhaps in correctly, that the extruder would also see the same level finer command level?

Regardless, good stuff, I wonder if the segment length is an adjustable setting such as in many other gcode creators (artcam, cambam, mach3 internal config)

[mhackney]

I know the segment length is not adjustable in Slic3r and does not appear to be in KISSlicer. I've now created 2 very simple objects:

1) a 2cm square .2mm high
2) a 2cm diameter circle .2mm high

I configured both slicers as similarly as I could and created the gcode. For these very simple objects there is very little difference in file size or # of segments for the perimeters (on the circle). However, it doesn't appear that KISS allows you to specify the orientation of the fill layer? I'm using the latest beta 1.1.0. Also, the Slic3r version of the circle has horrible infilling at opposite sides as you can see here:

required 48.0 mm of filament

KISS did a better job:

required 33.8mm of filament

But this is a very simple test. I'll work up to more complicated parts to see how they differ.