Performance interactive plot GUI

Hello together,

I've build a GUI the last weeks, for displaying recorded data from Logs of my companies systems.

The goal was to show up to 16 plots (chosen from up to 250 variables) with a linked x-axis. Usually there are not more than 1000 samples per plot. Finally I'm using stackedplot and it works fine...nearly.

The performance for interactive actions in the plot(s) is not acceptable as user experience for my team. Generally it seems to be an issue with the performance of linked axes - no matter which plot techinque I'm using in App Designer.

I'm questioning, if I'm using the right tool for my GUI. As I read, matplotlib for python is even worse in performace for interactive plots.

Does someone have a hint, where to go next with my approach?

Thanks in advance!

Christof

Answer 1

[Harald]

Hi,

it would be great if you can share some of your code, possibly with dummy data.

Instead of interactive zooming with linked x-axes, consider using the new Slider (Range) control or, in R2020b, two coupled sliders for the minimum and maximum x-value to let the user control the range of x-values. This might perform better than linked axes.

Additionally if the data is modified, it is helpful to not regenerate the plot, but just replace the x- or y-values as needed.

I hope that helps. If not, please consider giving us an example so that we can see the issue.

Best wishes,

Harald

P.S.: I notice that you are using release R2020b. Some aspects of performance have improved in newer releases. We would need to find out if this also improves performance in your particular case.