LabVIEW

cancel
Showing results for 
Search instead for 
Did you mean: 

Graph doesn't update when moved to a new sub-pane

This bug seems to be dependent on a multitude of things that are difficult to explain.  To summarize:

I had a graph on the FP that was updating based on the timing from data being sent across a network-based Shared Variable.

When I made the FP multi-pane and moved the graph to the newly created sub-pane, it quit updating properly.  There were numerous things that affected this behavior (complicated), so I have attached a demo.  The demo files use a network shared variable and probably need to be re-bound correctly for your particular computer once you save the project somewhere.
0 Kudos
Message 1 of 10
(4,109 Views)

Message Edited by m3nth on 06-02-2006 01:44 PM

0 Kudos
Message 2 of 10
(4,110 Views)
Sorry...had problems uploading.

Message Edited by m3nth on 06-02-2006 08:37 PM

0 Kudos
Message 3 of 10
(4,090 Views)
Well, from what I can see, you have two different shared variables:
  1. Acquired Data Send
  2. Acquired Data Rcv

The sender VI writes to the first shared variable, which is never read anywhere.
The receiver VI reads the second shared variable, which never receives any data from anywhere.

This just does not make a lot of sense. Don't you want to read and write using the same shared variable? 🙂 OK, let's use a single variable in both locations.

---> I agree there is something wrong with the chart update if that weird "status subVI" is inside the receiver loop. Very strange. I agree this must be a bug... (I don't quite see the purpose of that subVI, though).

Did you notice that deleting that brown empty cluster and its useless shift register in the receiver ALSO fixes the chart update problem? 🙂 (even with status subVI remaining in the loop).

0 Kudos
Message 4 of 10
(4,083 Views)
more thoughts:
  • the bug files are a stripped down version which explains that...
  • ...the original only has one network shared variable, but i couldn't get the stripped down version running on the same pc to work properly with one... easier to write to one and bind the second one to it.
  • ...the status subVI just reduplicates that i am calling a status subVI. in the real thing, it updates a front panel indicator with status that the variable got updated or something along those lines.
it doesn't surprise me that deleting other random objects makes things work or not work.  it is very intermittant, which points to the fact that something's not quite working right.  i'm not sure what is the root cause of the failure though--if there's a way to mitigate it every time.

0 Kudos
Message 5 of 10
(4,042 Views)
Well, I discovered that using a property node (value write) will always work, whereas writing to an indicator or a local variable won't.
0 Kudos
Message 6 of 10
(4,037 Views)
I have to say I can't reproduce this. I bound the receive variable to the send variable and ran the two VIs. The graph seemed to update just fine. If I moved the status VI out of the loop, I didn't see any difference.

Are either of you running LabVIEW 8.0.1? It seems to work fine in 8.0, but again, this could be an intermittent problem.
Jarrod S.
National Instruments
0 Kudos
Message 7 of 10
(4,017 Views)
I'm running v8.0.1
0 Kudos
Message 8 of 10
(4,014 Views)
It does indeed seem to be limited to 8.0.1. I tried it on my neighbor's machine and voila: the graph doesn't update until you move the splitter bar.
Jarrod S.
National Instruments
0 Kudos
Message 9 of 10
(4,008 Views)
Yup, 8.0.1 here... 😉
0 Kudos
Message 10 of 10
(4,003 Views)