Calculate L1phase when statistics reached value#12833
Calculate L1phase when statistics reached value#12833shahor02 merged 1 commit intoAliceO2Group:devfrom
Conversation
|
REQUEST FOR PRODUCTION RELEASES: This will add The following labels are available |
| // If statistics reached N*thresholds try to re-calculate L1phases | ||
| // if they changed wrt previous, replace ccdb object, otherwise do nothing | ||
| if (mNprocessed % minStatisticsForCalib == 0) { | ||
| minStatisticsForCalib *= 2; |
There was a problem hiding this comment.
Note that you update the calibration not at N*trhesholds but at 2^(N-1) * thresholds. Why do you need to minStatisticsForCalib *= 2; after each update?
There was a problem hiding this comment.
L1 object is unique for whole run. Normally it should be calculated from minimal threshold. If in the run we have noisy channels, I think it is better to collect considerably more statistics before next try. This should reduce the number of overlapped objects in ccdb.
There was a problem hiding this comment.
ok, thanks, though then the comment about N*thresholds is somewhat misleading...
| if (mNprocessed % minStatisticsForCalib == 0) { | ||
| minStatisticsForCalib *= 2; | ||
| mCalibrator->checkSlotsToFinalize(o2::calibration::INFINITE_TF); | ||
| mCalibrator->endOfStream(); |
There was a problem hiding this comment.
just to be sure: mCalibrator->endOfStream() does not really mean endOfStream, but just to finalize the calibration for the next portion of TFs, right?
There was a problem hiding this comment.
Yes, mCalibrator::endOfStream() calculates calibrations but does not reset anything.
| // If statistics reached N*thresholds try to re-calculate L1phases | ||
| // if they changed wrt previous, replace ccdb object, otherwise do nothing | ||
| if (mNprocessed % minStatisticsForCalib == 0) { | ||
| minStatisticsForCalib *= 2; |
There was a problem hiding this comment.
ok, thanks, though then the comment about N*thresholds is somewhat misleading...
2 participants
No description provided.