Duke3D: Regression with SE32 (and possibly other effectors)
Started with commit fbf4ab59
Sectors that contain a SE32 (angle pointing upwards) which contain both a Masterswitch as well as an Activator now misbehave.
In the attached test map, you will find three SE32 sectors, each being a lowering ceiling. The leftmost sector contains both a masterswitch and activator, the middle contains only a masterswitch, and the rightmost contains only an activator.
Prior to fbf4ab59, the ceiling of the leftmost sector would lower, like the rightmost sector, and raise again upon a second switch activation. The middle sector will lower once and then not move again. After this commit, the leftmost sector will no longer lower on first activation, but it will react properly on subsequent activations. Most likely, both the masterswitch and activator are triggering the ceiling movement on first activation, cancelling out its motion.