It seems that with large trigger runtime values (e.g., 50 min) the first parameter of the trigger command will be negative. The resulting FPS is wrong/weird. This seems to crash the STM and it needs to be rebooted. This state is a bit difficult to catch while running an experiment as superficially everything seems to work.

E.g., experimenting with different values (GUI) I got these:

Sending hardware trigger command:
Sending hardware trigger command:
Sending hardware trigger command:

Sending the command via a terminal (e.g., CoolTerm) works as expected. So I think that this issue is related to the GUI.