I think the way the current limit is being used is not as efficient as it could be because I would prefer to perform the flip at the latest point rather than at the earliest point.
For instance, if I have a 20m window and my exposures were only 60 seconds then I could keep imaging later, before needing to perform a flip. The software should be able to effortlessly work this out, but it currently lacks sufficient information.
The simplest way to envisage what I am thinking is to have two settings
1) I can perform a flip after this point
2) I must perform a flip before this point
This will allow the software to know that under some circumstances it needs to suspend capture, because it would finish after point 2) but wait until after point 1) to perform the flip.
I have to admit it has taken me a lot of thinking to understand how all this works.