Saving persistence data

[Kuz]

The persistence window produces a nice color image that summarizes many traces of the captured waveform.  Each pixel represents the number of waveforms that crossed that voltage level at that time relative to the trigger.  Is there a way to get a matrix of that data?  It is obviously stored somewhere to produce the image, so this should be low hanging fruit.  I tried to use the data logger to capture this data, but all it gives is a header with general information.

On a related note, when using the logger to capture a series of traces, it stores the full trace, not just the part that is displayed.  Since my data is localized to a small time slice, the huge files are a waste since only a small portion of them contains the information I need.  Is there a setting to capture just the data displayed in the scope window?  This is related to the above question because I only need the persistence data in the time range of the displayed image.

I am new to this, so apologies for my naivete.

Edited November 3, 2023 by Kuz
correct grammar

[attila]

Hi @Kuz

Internally in the application for each channel a persistence 2d array is computed, from this rgb images generated and drawn/overlaid in the Persistence view.
Exporting the image is not suitable for you ?

[Kuz]

5 hours ago, attila said:

Hi @Kuz

Internally in the application for each channel a persistence 2d array is computed, from this rgb images generated and drawn/overlaid in the Persistence view.
Exporting the image is not suitable for you ?

Thanks, Attila, for your response.  I need the actual number of times the traces have crossed each pixel for analysis.  An image file would work if I could use it to back out this information.  However, there are two issues with this.  First, I don't know the mapping between the RGB color values and the number of traces that crossed that pixel.  I assume that the colors are chosen for the sake of visual clarity, not linearity.  Secondly, I would want just the color plot with a black background so that the labels and markings don't get mistaken for a crossed pixel.  And of course, I would need to know the time and voltage coordinate for each pixel.  Since the plot here is generated from the internal array that you mention, and is exactly what I want, the simplest solution is to get that internal array as an export feature.

[attila]

Hi @Kuz

I am working on it

[attila]

Hi @Kuz

[Kuz]

Hi @attila This is fantastic. I never expected this being in a beta release so quickly. It seems to work perfectly. I’ll now run it with our experiment to see how it performs in our application.  Btw any response to the second part of my original question?  

[attila]

Hi @Kuz

The next version will add width/height options, horizontal/vertical header for csv/txt exports, 1bit mono, 8/16 grayscale or 24bit color image options.
Am I missing something ?

[Kuz]

Hi @attila This is fantastic. I never expected this being in a beta release so quickly. It seems to work perfectly. I’ll now run it with our experiment to see how it performs in our application.  Btw any response to the second part of my original question?  

[Kuz]

Btw I’m not sure if my last post had an issue.  I posted last night and this morning it showed as not posted so I reposted it again. To summarize it appears that this feature does exactly what I need. The only question I have is on the interpretation of the numbers. The important info is the number of times each (time,voltage) point is activated. Maybe a better way of saying this is the number of traces that crossed that time/voltage. The persistence plot is a visual representation of it, but depends on the color scale from the right bar but it’s the raw data that generates it that is of interest. Again, I really appreciate your quick implementation and believe the ad3 is a great value.  

Edited November 10, 2023 by Kuz
Fix error

[attila]

Hi @Kuz

 

[Kuz]

Thanks @attila  Works great!!!