Same file, two machines, big gap in print times - which one is lying to me?

I ran into something similar a while back on my Dual Overture setup, which isn't quite the same chassis but the print enviornment is comparable. I was running a batch of those small hinge brackets, maybe six or seven at a time, and I noticed one machine consistently finished about ten percent slower than the other when printing the same geometry. The top layer quality was identical on both, which immediately told me it wasn't an extrusion issue in terms of physical failure, but rather something affecting the motion profile itself.

My hunch was always about the actual flow rate coming out of the nozzle, not the requested speed the slicer sends. Sometimes one extruder might just be slightly undersized internally, or maybe the drive gears are a millimeter out of alignment, causing the motor to spin at the right RPM but the plastic flow to be a bit hesitant. It's subtle, but over a long job, that hesitation adds up significantly to the overall time. I ended up messing with the flow rate compensation setting in my slicers, adding a little buffer to the slow machine untill the timings aligned properly, which smoothed things out for my production run. I never looked at the firmware since it was fine, but adjusting the flow through the slicers was a solid, practical fix that didn't require any weird hardware teardowns. If you want to chase this down yourself, just monitor the actual flow rate coming from the hotend versus what the slicer says it should be providing.

I had a similar hiccup when I was running a job for some custom parts last year on a diferent setup of mine. The difference wasn't in the print speed settings, because I monitored the feedrate output and it was exactly what the slicer told it to be. Instead, the variance seemed tied to how flat the print bed was settling during the middle of the print. My original issue was with a small desk surface tilt. When one of my machines sat on a slightly uneven wooden workbench, the printer would start with one end slightly lower. By the time the print was halfway through, the pressure distribution changed, which led to slower extrusion rates even though the motors were spinning at full speed. It was like the slicer was demanding a speed that the physical resistance of the plastic against the slightly uneven surface was fighting back against. I ended up using shims under the feet on the corner that was too low. After correcting the bed level and ensuring everything was perfectly level, the runtime stabilized across both machines and the quality held up because the gap between the nozzle and the bed was consistent. For me, it was all about the physical foundation and ensuring zero play in the extrusions.

Don't just assume the internal electronics are the fault, because often the problem is purely mechanical on the bench itself.

Queue the batch to the fast one and stop losing sleep, that's what I'd do. I ran a pair of machines side by side for years and one was always the slow kid no matter what I tried. Turned out the slow one's rails were bone dry and the carriage dragged on the quick moves, so it just never hit full speed. A wipe down and a drop of oil on the rods bought back most of the time. Back in the day we just called that character and moved on, but it's worth a look before you assume it's some hidden menu. If it still lags after that, let it be the backup machine.

Check the minimum layer time setting before you write the machine off. Small brackets like that spend most of the job on short layers, and if one printer has that floor set higher it crawls through the thin sections and pads the whole ETA. I had a batch of little camera mounts where one unit took almost 90 minutes longer per plate and it turned out to be exactly that, a leftover value from an old profile baked into the machine. Same file, same slicer output, completly different wall clock time. Look at where the slow one loses the minutes in the preview versus the actual print log if you can pull one.