Multifunction DAQ

cancel
Showing results for 
Search instead for 
Did you mean: 

State data conduits - a good idea?

If you're using a state machine, you've got state data that you pass to successive iterations. The typical approach is to cluster together like terms, then pass them through shift registers. Generally, the amount of state data is proportional to the complexity of the program, and inversely proportional to the neatness of your diagram.

 

I would like to hear some feedback on the following approach: creating a top level cluster and calling it our state data conduit, which is simply a container for all the shift register data.

 

 State Data Conduits

 

One advantage with this approach is it offers much neater calculations. A typical calculation may include a constant, some volatile data, and some non-volatile data. I have defined volatile data that stays in memory only as long as the VI stays in memory, and nonvolatile data that is written to disk once the VI is exited, then repopulated on the next call of the VI. Also, you can store references nicer as well, especially references that may dynamically change, such as ones used for dynamic event registration. Finally, for states that don't access certain clusters, there is no need to wire a short all the way across the case for each shift register on individual state clusters.

 

Can someone provide some disadvantages for this approach, or offer comments?

 

Best regards,

Jack

Message Edited by mechelecengr on 10-16-2008 02:51 PM
0 Kudos
Message 1 of 3
(2,599 Views)

Please ignore this thread, and instead refer to the one on the LabVIEW discussion forum. While previewing the post and returning to the message creation page, the forum to post to defaulted back to the board I was previously on (Multifunction DAQ). If a moderator sees this thread, please delete it from this forum since it is duplicated on the correct LabVIEW forum.

0 Kudos
Message 2 of 3
(2,591 Views)
0 Kudos
Message 3 of 3
(2,578 Views)