When placed horizontally, the Sequenced Gearshift controls normally (more importantly, consistently) when placed and rotated (using the Mechanical Bearing).

However, when facing up/down, the Sequenced Gearshift seems to act differently depending on the orientation. When facing North or East, the instructions are in one direction. When facing South or West, the directions for mechanical piston movements flip directions.

The instructions in the Sequenced Gearshift don't actually change, the Mechanical Piston just behaves differently.
Visually, the shaft between the Sequenced Gearshift and the Mechanical Piston moves consistently regardless of the direction it's facing; the issue is the actual movement of the piston.

https://filebin.net/08rthm3wfdx6thxz This is a link to a schematic file wherein I have made an example contraption, with both a vertical and horizontal-y aligned piston/gearshift arrangement.

This issue has been marked as stale because it has been inactive for 3 weeks. It will be closed if it remains inactive for another 3 weeks.

This issue has been closed since it has been inactive for 3 weeks since it was marked as stale.

I believe hand cranks also display this behavior

This might just be an issue with the mechanical piston and it's orientation.

This is an issue with the mechanical piston, not the sequenced gearshift. It is, arguably, not a solvable problem -- because the piston is powered by a perpendicular shaft, it has to pick an arbitrary correspondence between clockwise/counterclockwise and extend/retract. This means that oiston directions can also reverse if you have a horizontal piston and rotate it 180 degrees around the axis it points along.

This is why I think the piston and piston pole should be replaced by the "rack mount" and "rack" -- a rack being cog-driven from the side has an unambiguous direction.

I definitely disagree with @stellHex. I don't see any reason why the directionality can't be consistent. When the piston is aligned with the shaft running vertically, any input to the top (or bottom) could theoretically be coded to be constant. Say, when aligned vertically and powered with a clockwise shaft in the top, it extends. There's nothing about rotating the mechanical piston (on the axis of the shaft such that the piston moves between cardinal directions) that makes this non-deterministic. It can just extend when powered with clockwise motion (or however it should be defined).

Sorry if there was some confusion among the direction of the piston. Maybe a new issue should be made concerning just the directionality of the piston, and this issue merged with that.

But then if you rotate the entire contraption 180 degrees (around the axis of the piston extension), the piston will reverse direction.

Well yes, if I understand you it would reverse direction simply by the act of rotating it. I.e. if you rotate it 180° from extending out North, it would extend South instead, thereby you could say it has flipped (because in a sense it is retracting in the north direction, albeit the piston is flipped)

I guess its a matter of implementation/prioritization, maybe theres a contraption that utilizes the “flipping” behavior of the instructions (in the current implementation). However, I do think it is more user-friendly to have it act consistently regardless of orientation, as I have described above.

Ah, this is what I get for not making an example picture like I should have, this stuff doesn't translate well into text.

The top and bottom pistons, and left and right pistons, have reversed extensions, even though from a symmetry perspective, they're exactly the same, and if the whole contraption were rotated 180 degrees around the radial chassis then they extended pistons would retract and vice versa.