Performance issues using control references in Analog control loop?

Question

My main vi of a tensile tester control application calls a number of sub-vi's, including a analog control loop which controls the test. The control loop must update some boolean and digital indicators and respond to user input on the front panel of the main vi during a test.
To simplify my main vi, I moved the control loop code into a sub-vi, and used control references to access the controls and indicators on the front panel. However, this has dramatically affected my loop performance, and the loop can no longer keep up with the acquisition speed.
Do control references always cause such a slowdown? Is there anything that I can do besides moving the code back into the main vi?
Thank You,
David Creech

Answer 1

I have had the same problem. I have discovered other funny things about references also; some kind of memory management (or mismamagement?) is taking place behind the scenes.
Regardless, you can often do away with the references by passing the initial state of a control or indicator into the subvi, changing it inside, and passing the altered state back to the caller. Once back in the caller you update the front panel control or indicator by using a local variable.
One thing to watch out for if using this scenario is the dreaded race condition; this can be avoided using a state machine. Check out
http://www.advmeas.com/goodies/statemachine.html
for a good example. It is a shame that references behave in this way; it limits thier usefulness.
Perhaps someo
ne else will point out a way to utilize them more sucessfully?