Change DUT support parameter with -o

How do I modify the parameters of the DUT, for example, the support thickness or position, from the options when running the executable, it is not clear how fromt he documentation.

Cheers,

Mathieu

Hi Mathieu,

to change the geometry configuration on the command line, use the -g option:

allpix -c config.conf -g dut.sensor_thickness=300um

It’s not in the published manual because this is so far only available in master but not in v1.3.2.

Cheers,
Simon

Thanks ! but if it is for the support ? now I have a file defining a diode+ collimator :

type = “monolithic”

number_of_pixels = 1 1
pixel_size = 1cm 1cm

sensor_thickness = 100um
sensor_excess = 250um

[support]
thickness = 0.1mm
size = 10mm 10mm
material = “paper”
location = “absolute”
offset = 0 0 -50mm
hole_size = 1 1mm
hole_type = “cylinder”

and a file where I position it :

[mydetector]
type = “diode_collimator”
position = 0um 0um 0um
orientation_mode = “xyz”
orientation = 0deg 180deg 0deg

so it would be something like :
allpix -c config.conf -g dut.support.thickness=300um

?

Hi Mathieu,

Since your detector is named mydetector the invocation should be like:
allpix -c config.conf -g mydetector.support.thickness=300um

Hi Mohamed,

I’m not 100% sure this works as expected - maybe we’re not treating the dot for subsections properly here. Have you tried if it actually changed anything when changing the setting?

The part of the code to look at would be here:


plus the part where we read the configuration from the command line and pass it to the ConfigManager, here:

Cheers,
Simon

Hi Simon,

Yes you are right nested settings doesn’t work. I also noticed that there aren’t any information logged -trace or debug or warning- regarding setting parameters from the command line and there is no indication to tell the user if their settings were put on place or not which can lead to confusion.

Mohamed

Good point - successfully overwritten parameters should certainly be worth a INFO-level message.

Ticket opened: https://gitlab.cern.ch/allpix-squared/allpix-squared/issues/160

1 Like