It’s not using the Baro from the Transponder. There are three (kind of) independent local Barometric Pressure “Indicated Altitude” registers. The AP uses the Barometric pressure stored in the “Indicated Altitude:2” register. The transponder gets its barometric pressure from register “Indicated Altitude:3” (instead of “Pressure Altititude” like it should be), and the Altimeter from register “Indicated Altitude:1”. These registers are independent of all gauges. They are just memory storage locations.
I say kind of independent because they are all changed by the B key. The B key updates all three registers to the local barometric pressure. Changing the Altimeter Kollsman changes the “Indicated Altitude:1” register, but leaves the other two alone. Hence, they all can get out of sync. The KAP140 does have an option so you can independently of the Altimeter change the “Indicated Altitude:2” register if you turn/press the right combination of buttons. Changing the Altimeter Kohlsman value has no effect on either “Indicated Altitude:2” or “Indicated Altitude:3” registers.
I have a modified Transponder that ignores the local Barometric pressure and just reports Standard pressure, as all Transponders are supposed to, if anyone is interested.
I agree that local Barometric pressure should not change as if you flew through a wall. It can change very quickly, it could be quite different over a distance of a few miles, but, it won’t just suddenly jump.
I suppose the issue is, at what point do you start a change in Barometric pressure, since reporting is likely by points as opposed to being able to look at say, an infrared satellite area map? How do you figure out where the changes occur, and how do you approximate an appropriate ramp? A change could happen anywhere between two points. I suppose you could just come up with a standard distance over which to make changes, maybe as a percentage of the distance between 2 points, and place the inflection point halfway between reporting points.