I have a question about code architecture and getting multiple devices controlled and sychronized for one experiment. This is an "architecture"-type inquiry, so I am hoping for some suggestions on how to proceed.
I run an experiment in which I control 2 NI PCI-6733. I am soon to add a Tektronix AFG 3022B, and have long been putting off an opportunity to rewrite my labview code from the ground up. I inherited it from an earlier research, and while functional, I would like to make it easier to modify and break up into subVi's and such. Link to the current program (labview 8.6.1) is here if you would like to see the code that is currently used, and the subvi's are in a zip file. The current version of the experiment consists of just one stage, all the writing of the data to the PCI cards is saved for the end. I need to change this in my new setup though.
The new experiment consists of 2 stages. The first will run (looping an output array to the PCI cards) until it hears a "true" from another computer (connected via TCP). At that point, it should switch to stage 2 and run a sequence (usually 10^5 timeunits in length, where the time unit is 0.1 ms) that outputs to the two PCI cards, the AFG 3022B, and with the flexibility to add more devices in the future.
Most appreciated would be structural advice. How to arrange the VIs, if it's good to use a "master" VI that would control the two subVi's of stage 1 and 2, etc... Feel free to ask for more details if it would help clarify my question. Thanks!
