The errors introduced by compressing the triggers is having a more adverse affect than anticipated so it is planned to scale the trigger to range 0 - 255. In this plan a new scheme a compression scheme 0,0,7. Then based on the requested detector mask the triggers can be rescaled/unscaled back to their original values.
Where for example in TM(21, 6, 21) Nint ~ NIX00405 and Nacc ~ NIX00407 (converted to trigger groups)
An example from the AX log 202310152214-terminal.log gives n_int=5, n_acc=1, No=1000 and Ns=6 but evaluating that gives 1000/(30*5*1) = 6.66 so need to be careful about how it is handle on board floor, ceil or round.
Confirmed onboard FSW also essentially floors the value so to reconstruct need to add 0.5 to scaled value with an error of +/- 0.5 before un-scaling in the above example 6 -> 6.5 -> 6.5 * 30 * 5 * 1 = 975 +/-75 = 6 * 30 * 5 * 1 900 to 7 * 30 * 5 * 1 = 1050
[ ] Get scaling parameters
Will become parameter in FSW but fixed for no at 30
The errors introduced by compressing the triggers is having a more adverse affect than anticipated so it is planned to scale the trigger to range 0 - 255. In this plan a new scheme a compression scheme
0,0,7
. Then based on the requested detector mask the triggers can be rescaled/unscaled back to their original values.Meeting notes with code https://v000658.fhnw.ch/index.php/apps/dashboard/
Ns = No / (30 * Nacc * Nint)
Nint
~ NIX00405 andNacc
~ NIX00407 (converted to trigger groups)n_int=5
,n_acc=1
,No=1000
andNs=6
but evaluating that gives1000/(30*5*1) = 6.66
so need to be careful about how it is handle on board floor, ceil or round.6
->6.5
->6.5 * 30 * 5 * 1 = 975 +/-75
=6 * 30 * 5 * 1 900
to7 * 30 * 5 * 1 = 1050
30