What is the issue?
When using a toolchanger and swapping either from an AFC lane or to an AFC lane, the time required is much greater than toolhead-toolhead swaps on standalone toolheads. Also the swap time seems to increase over print duration. While swaps to lanes are expected to be longer than normal, swaps from a lane should not take more time. See below test results:
The setup is e0, e1, e2 are standalone. E3 is connected to BTT ViViD and only has lane3 loaded.
Graph colors are swaps to a given lane, shapes are swaps from that lane.
The red lines at the top show that swaps to lane3 take more time and that time increased over the print duration. The middle lines are swaps from lane3 and they also slightly increased. The bottom sets are extruder-to-extruder swaps and remained constant. The overall toolchange duration went up as extruder-to-extruder swaps were replaced by extruder-lane swaps.

What is the issue?
When using a toolchanger and swapping either from an AFC lane or to an AFC lane, the time required is much greater than toolhead-toolhead swaps on standalone toolheads. Also the swap time seems to increase over print duration. While swaps to lanes are expected to be longer than normal, swaps from a lane should not take more time. See below test results:
The setup is e0, e1, e2 are standalone. E3 is connected to BTT ViViD and only has lane3 loaded.

Graph colors are swaps to a given lane, shapes are swaps from that lane.
The red lines at the top show that swaps to lane3 take more time and that time increased over the print duration. The middle lines are swaps from lane3 and they also slightly increased. The bottom sets are extruder-to-extruder swaps and remained constant. The overall toolchange duration went up as extruder-to-extruder swaps were replaced by extruder-lane swaps.