Same results with PETG, 0.8mm.

Also ran the e-steps calibration with PETG. First 100mm was under by 2.5mm. Adjusted on the LCD and ran again then I was over by 3.5mm. So I put it back to the original value before running the calibration print.

When I reran the e-steps on PLA it was spot on. Am I making a mistake expecting pretty good consistency on the 100mm extrusion, or does that indicate something might be slipping?

The other thing I found a bit odd was that no matter what layer height I set the extrusion width stayed at 0.4mm in the gcode. Is that a quirk of MatterSlicer? Does slic3r let you set the width and then it controls the extrusion rate to get it?