Chart and Waterfall Indicators Lagging Application

[James.M]

Hello,

I have an application that is pulling in three sets of accelerometer data. This data is already analyzed on my cRIO for RMS and Peak data as well as Waterfall plots for frequency information. Once the data gets to my application running on the host computer, it writes the data directly to indicators (other than a small self-made waterfall history length).

I have attached a snippet of the code where the indicators are updated. This happens in parallel with all other processes and happens only at about 10Hz. The For loop in there is always a single iteration with the way the data flow is set up currently, but it's still there to handle the possible case of sped up data acquisition. The case structures I put around the indicators are there to test the lag that I am observing. If I disable one, the other, or both, the lag lessens. The lag is gone when I disable both case structures. This lag shows itself in the form of observable jumps/lags in the other indicators on the screen which are also being updated at 10Hz.

So my question is, is there some way to sperate these indicators from everything else as their own thread? or maybe a smarter way to update the indicators? 10Hz is pretty slow, so I was surprised when disabling the charts sped everything up.

Thanks you the help,

[JÞB · Accepted solution]

Just a guess (but a good one) have you done any benchmarking on the sub-vi?  Assuming debugging is disabled UI 3 may not execute with Enable Waterfalls = False since the outputs go to unreachable code.

Do any of the charts overlap total bounds?

Can you save that snipette and sub-vi back to 2013? (preferably with an array constant of typical buffered analyzed Data)

[James.M]

"Assuming debugging is disabled UI 3"

Well here I am not even checking the little things... This solved the issue (mostly). I still get a lag when I set the waterfall age to be longer than a second, but I think I just need to run an average on the data to get fewer data points on the plot. Trying to display 4000 samples on the waterfall plot might be too much.

Thanks!