Happened to be in the neighborhood, so dropping by this thread again.
To Simon-lee: I think it's an excellent idea for you to approach this problem in two different ways. You're gonna learn a lot from the process.
I gave suggestions based on sticking with the general structure you started from (with 3 parallel loops). While doing it, I gotta say that the 3 separate loops felt like they were making the solution more complicated than necessary.
I haven't dug into any of mcduff's state machine approach, but when he first suggested a 1-loop solution, it sounded like good advice to me. Don't believe him when he calls himself a hack.
As you approach things in parallel, you might want to rethink the way you partition your 3-loop code and see if you could do it with 2 loops instead. It might still be handy to have that 2nd loop because sometimes with only 1 loop certain processing gets a little "stuck" waiting in line for its turn. A 2nd loop can defer the low priority things and keep them out of the way of the high priority things.
Still, I'd venture that you'll find the 1 loop state machine approach to be more straightforward to implement and debug. I'd focus on that one and get it working correctly first before going after the parallel loop approach. What you learn by getting the 1 loop version debugged will help you decide how to partition the 2 (or 3) loop version.
-Kevin
ALERT! LabVIEW's subscription-only policy came to an end (finally!). Unfortunately, pricing favors the captured and committed over new adopters -- so tread carefully.
