I tried to give a kudo but the software seemed to burp. I'll try again later, because you deserve it.
Why the kudo? Because you took just that little bit more of your time to show where I could find what the manual assumed I knew and didn't say (lest I become bored, I guess.) ...yes, I checked the manual, first IOW.
Now my experience:
Not a large VI, really. But in the subVIs eventually a memory intensive VI is encountered. The main VI contains only a couple of subVIs. The reason for the intense memory usage is the FFT is working with a 2.5MB dataset. The f(t) itself is stored in a monster (from my perspective) text file, read in, analyzed, and then I display the magnitude and phase information in a couple of X,Y plots. I'm using the two-sided -- I want to see those complex numbers do their thing. (I must admit some of the reason I'm saying this is perhaps another reader out there can explain a better way to work with such a large dataset.)
The whole operation takes around 10 seconds. That wasn't the problem. I don't mind that, especially if I'm getting correct information.
My numbers were this:
Prior to run, about 65MB memory.
After one run, jumps to 950MB.
During one run, jumps to 1.3GB, max.
After further runs memory usage stays around 950MB.
Moving controls and then reverting, memory usage drops to 650MB and then returns to 950MB after another run.
CPU usage never exceeds 20%.
Moving the slider in question to 1 from the default 5 caused a 4-5 second plus annoyance when working with the editor to drop to a reasonable 1 second and a normal control click speed. (That was the worst. Click a control and wondering whether you did. That was so annoying.)
Well there you go. Perhaps the above will be of assistance to others in some way. (Also, yes, about the large dataset thing, I'm open to advice.)
