03-12-2010 04:25 PM
DianeS wrote:
I knew about the method for setting all front panel controls to default, but obviously I didn't know about the one for individual controls. Thanks! That might come in handy for me some day...
I think We are ALL getting a lot of good ideas out of this thread. WOW! great OP and the OP is working hard to master concepts and style. I could stick with threads like this all day!
03-12-2010 04:33 PM
Agreed on all points! There's nothing more fun than helping someone who genuinely wants to learn and who works hard in order to do so -- and there's the added bonus that the rest of us get to learn right along with them!
Go, OP!! ![]()
03-12-2010 05:35 PM
Haha,
The OP also feels like a bonehead for missing that one. I'm chalking it up to a long week, and it being Friday.
Speaking of which, Happy Friday to everyone, and thanks for the compliments!
Hoping to finish cleaning it up this weekend, and post Monday.
03-14-2010 03:50 PM
Is there a way to enque two elements in a given order in one operation?
I've condensed my state machine quite a bit (thanks Jeff for that suggestion) but I still have instances where I want one case to follow another for a given event. I could probably combine states further, but it does help keep the code clean by breaking out a few of those states.
Currently I wrote it (incorrectly) by adding an element to the queue from within the state machine.
03-14-2010 03:57 PM
Also,
Is there any caveat to using action engines? I find myself prefering them over using shift registers on the main vi, as they result in cleaner looking code. Do they use any more or less memory, processor, etc?
03-14-2010 04:06 PM
03-14-2010 04:44 PM - edited 03-14-2010 04:46 PM
03-15-2010 02:03 AM
03-16-2010 09:43 AM
Ok,
So not as clean as I would like it. Long Monday. Anyway, here is the code, its a little laggy for the solution to the forced response (probably all the sines and cosines). Not sure if I added multiple tasks to the queue in the best way either.
Is there a way to disable the FP during certain solve type actions? It would be nice to have a light or something at least to indicate solve time.
I think I still end up stacking up multiple solves under certain cases. I'm considering moving the solve to another state, and activating it differently.
I think most of the code could be improved, but since I am my own customer, this can be a work in progress for a long time.
03-16-2010 10:05 AM
And I'm very unhappy with my fourier series. It works most of the time, but if the number of number of points in the evaluation isn't high enough for the peak amplitude, the error value eventually skyrockets, causing an infinite loop.
If the FFT will do what I need, perhaps I should look harder into that.