3D Printing 7 min read

What a Slicer Actually Does

A slicer converts a shape into instructions. Knowing the order it works in explains why certain settings interact, and why some problems can only be fixed at one particular stage.

Vextrude Team

Updated Aug 21, 2026

How a 3D printing slicer works — Vextrude

A slicer converts a shape into instructions. Knowing the order it works in explains why certain settings interact, and why some problems can only be fixed at one particular stage.

Step 1: Intersect the Mesh With Horizontal Planes

The slicer takes your triangle mesh and cuts it with a stack of horizontal planes, one per layer. Each intersection produces a set of closed 2D outlines — the cross-section of your object at that height.

This is where layer height enters. At 0.2 mm, a 40 mm model becomes 200 slices. Halving the layer height doubles the slice count and roughly doubles print time, which is the entire quality-versus-speed trade.

It is also where mesh errors become fatal. If a plane cuts through a hole in the surface, the resulting outline does not close, and the slicer cannot tell inside from outside. Flipped normals produce the same ambiguity. This is why a model that looks perfect on screen can fail here — the renderer does not care about closed volumes, and the slicer cannot work without them.

Curved surfaces also reveal themselves at this step. Stair-stepping on a shallow slope is not a defect; it is the unavoidable consequence of describing a curve with discrete horizontal layers.

Step 2: Walk Perimeters Inward

For each 2D outline, the slicer insets a path by half the extrusion width and traces the boundary. That is your outer perimeter. It repeats inward for each additional wall.

Extrusion width, not nozzle diameter, is what governs this. A 0.4 mm nozzle typically lays a line around 0.45 mm wide, because the plastic spreads slightly as it is pressed onto the previous layer.

Two perimeters is the common default and gives roughly 0.9 mm of solid wall. This is where part strength largely comes from — walls contribute far more than infill, which is why increasing perimeters is usually the right answer to a part that broke, and increasing infill usually is not.

Thin features expose a limit here. If a wall in your model is narrower than two extrusion widths, there is no room for both perimeters, and the slicer either thins the lines, prints a single line, or omits the feature entirely.

Step 3: Fill the Middle

Whatever remains inside the innermost perimeter gets filled with a repeating pattern at your chosen density.

Infill exists mostly to support the layers above it, not to make the part strong. At 20% density the interior is largely air, and the pattern's job is to give the top solid layers something to bridge across.

Pattern choice matters less than people assume for ordinary parts. Gyroid is isotropic and prints without crossing itself; grid is fast; cubic is a reasonable middle. The larger effect comes from density, and even that has diminishing returns — going from 20% to 50% adds significant time and weight for a modest strength gain, while adding a perimeter is cheaper and more effective.

Top and bottom layers:

Solid layers at the top and bottom are counted separately from infill. Too few, and the top surface sags into the infill voids — the pillowing effect. Four to five solid layers at 0.2 mm is a reliable starting point.

Step 4: Detect Overhangs

The slicer measures the angle of every surface against vertical and flags anything beyond the overhang threshold, typically 45 to 55 degrees.

The 45-degree rule of thumb comes from geometry: at that angle, each new layer is supported by roughly half the width of the one below it, which is generally enough for the plastic to stay put. Beyond it, an increasing fraction of the line is extruded into open air.

Bridges are treated separately. A horizontal span between two supported points can often print without support because the extruded filament is stretched taut between them, and slicers detect these and adjust speed and cooling accordingly.

Support is the setting most worth avoiding through design. Rotating a part 45 degrees on the bed, splitting it in two, or adding a small chamfer where an overhang begins will frequently eliminate supports entirely — and supports always leave marks on the surface they touch.

Step 5: Turn Paths Into G-code

Finally the slicer converts every path into machine instructions, and this is where the numbers come from.

For each move it calculates how much filament to push. The volume of plastic needed is the path length multiplied by extrusion width and layer height; dividing by the filament's cross-section gives the E value in the G-code. Get this wrong — through a bad flow rate or an incorrect filament diameter — and every other setting is fighting it.

It also plans travel. Between printing moves the nozzle has to reposition without leaving strings, so the slicer retracts filament, chooses a route that avoids crossing open perimeters where possible, and decides where to place the seam on each layer.

The output is a plain text file, and it is worth opening once. You will find your printer's start sequence, then thousands of lines of coordinates, and a clear view of the fact that the printer itself knows nothing about your model — only where to move and how much to extrude.


Slice something now

A free browser-based slicer — adjust settings, preview layers and export G-code.

Open Print Control