Rapid subpanel execution

[Madmanguruman]

Yes. I get what you're saying.

I think my best way out is to embed the data collection VIs into a VI just for the subpanel (which can be left to run and idle as needed) - that way I don't need to worry about re-architecting and re-qualifying a ton of other programs.

Appreciate the support everyone.

[Bob_Schor]

---

@Ben wrote:

If you keep track of the reference of each of the target sub-VI you can remove them (while they are still running) from the sub-panel and insert another.

---

Yes, this is precisely what I do with my multiple Stations.  I start them all, they all run for about 2 hours, and at any time I select one (or none) to "view" and implement this by wiring their RefNums into the "Insert VI" Node associated with the sub-Panel.  Note that I don't need to use "Remove VI" to get rid of the existing Asynchronous Clone whose Front Panel is currently being displayed -- the only time I use this Node is if I'm turning "off" the Display so that none of the Clones are being monitored.

Bob Schor

[Ben]

---

@Bob_Schor wrote:

---

@Ben wrote:

If you keep track of the reference of each of the target sub-VI you can remove them (while they are still running) from the sub-panel and insert another.

---

Yes, this is precisely what I do with my multiple Stations.  I start them all, they all run for about 2 hours, and at any time I select one (or none) to "view" and implement this by wiring their RefNums into the "Insert VI" Node associated with the sub-Panel.  Note that I don't need to use "Remove VI" to get rid of the existing Asynchronous Clone whose Front Panel is currently being displayed -- the only time I use this Node is if I'm turning "off" the Display so that none of the Clones are being monitored.

Bob Schor

---

Well that is interesting!

It never occurred to me that I could just slap a new one into a sub-panel that already had a sub-VI running in it.

That is the result of just mucking about with things and not really knowing what I am doing.

I will have to try that the NEXT time (like tomorrow maybe?) I am stuffing a sub-VI in a sub-panel.

Ben

[BertMcMahan]

---

@JÞB wrote:

---

---

The front panel is for a user too observe.  20 frames per second is all the human eyeball can see..... and that is faster than the brain can assimilate the data!

---

Apologies for sidetracking the main discussion, but I'd like to point out that 20 FPS is the *minimum* for human eyes to see motion, not the max. For LabVIEW it's perfectly fine, but if you're expanding into anything high speed it's very different. For example, a football thrown on a 20 FPS display will look blurry, but not on a 60 FPS display. You can easily tell the difference between 60 FPS and 90 FPS if you're playing a video game, and VR goggles running below 80-90 FPS will get you nauseated pretty quick. For further reading:

https://www.testufo.com/# (view some different framerates)

https://gfycat.com/merryspiritedbass-pcmasterrace (a 3D scene rendered at 30 Hz vs 60 Hz)

/sidetrack over, carry on please 🙂

[Madmanguruman]

So - I have followed the suggestions of this thread: I wrote a message-handling "wrapper" sub-VI that is visible in the subpanel. This VI sits idle when data collection is off, and periodically calls the pesky sub-VIs when data collection is on, communicating with the top-level VI via queues.

Front panel -> subpanel window -> wrapper VI -> multiple data-collection VIs

This approach works - no more mouse flinching. Now I just have a subtle issue - everything works well the first time I run the VI, but if I re-run it after everything stops, the producer loop in the wrapper VI never starts. I need to close the project and re-load it for the loop to work - could this perhaps because I am using a static reference to the VI to load it into the subpanel, and somehow it's preserving state between subsequent VI runs? 

[BertMcMahan]

Stopping your top level VI won't automatically stop other VI's inside subpanels. Make sure you send the subVI a message saying "stop" when your main VI closes.

[Madmanguruman]

The wrapper VI does stop - I am sending a shutdown message to it to explicitly force it to stop processing data and to shut down all its loops, which is working.

Replacing the static VI reference with a call to Open Reference (and a call to Close Reference during shutdown) seemed to fixed the restart issue. From what I read, the static reference is preserved until the top-level VI is dismissed, which mabye explains why I had to close the project to get it to restart correctly. So all issues are now resolved.

This was an insightful thread, thanks for the opinions and guidance.