GCBallesteros / trattoria

The fastest streaming algorithms for your TTTR data.
6 stars 0 forks source link

Time trace problem #6

Open fbbarnes opened 6 months ago

fbbarnes commented 6 months ago

Hi, I am having trouble using the timetrace function on data I am generating with the HydraHarp GUI.

As a test, I am using the timetrace.py from the Examples in this repository. The timetrace is calculated, however the tt_res.t seems to be calculated incorrectly. For example, I collected data in a .ptu file over 60s with a count rate displayed by the GUI being roughly 1-2kcps. The TTRResult_NumRecords shows ~80k records as expected. However, the timetrace only calculates the max value of tt_res.t to be about 1.4s. This also affects the normalisation of counts.

I have tried this on the latetst version of trattoria with python 3.11 (ina clean conda environment) and with older versions of trattoria.

I will attach the .ptu file, .py script, and plot.

Thanks!

Aside: Previously, this has not been an issue and I suspect it has something to do with recenetly using snAPI (although I have no clue what exactly).

fbbarnes commented 6 months ago

trattoria-tt-issue-files.zip

fbbarnes commented 6 months ago

Update: Using trattoria to calculate time traces from files taken in T3 mode with snAPI seems to work fine.

GCBallesteros commented 6 months ago

That's far from surprising as the code has never been tested with T3 data, so it's highly experimental. I don't know when I will have time to work on this as it is really hectic at the moment for me.

fbbarnes commented 6 months ago

Hi! Just to clarify it seems like T3 is working and T2 is not working.