State Machine Basics

[Navarre]

Hi,

I have done a fair amount of LabView development but have rarely used (or run into) the state machine construct. From what little I have read it seems that almost anything can be modeled as a state machine. Are there situations where a state machine wouldn't be effective? Are there real-world examples of things that are not state machines?

Thanks,

Nina

[V_T_S]

Does your code need to respond to user interface?

Does your code has series of steps/defined sequence?

Then "YES", state machines are the way to go.

[PaulL]

Are there real-world examples of things that are not state machines?

Yes. For instance, a transformational system (a data processing algorithm not requiring user interaction) is not a reactive system (that responds to events). State machines are appropriate for designing reactive systems.

Here is a link I have found helpful when contemplating the essence of state machines: http://xlinux.nist.gov/dads/HTML/statemachine.html.

[crelf]

For reference (and open to ppl improving): http://labviewwiki.org/State_Machine

[Oakromulo]

Even if some code doesn't actually respond to user interface it could still be useful to model it using a state machine pattern. Useful in a sense that it could actually improve both code readability and performance.

For example: imagine a VI with a "latch when released" switch and a boolean indicator. For every push in the switch you toggle an automatic blinking pattern to the output. I have implemented this solution both with and without state machines. Take a look below. Which one do you prefer?

without.vi

withSM.vi

Edited December 21, 2012 by Oakromulo

[ShaunR]

Without