Draft: UTDecoder ADC weighted algorithm

The RawBank now decoded and UTHit reconstructed as the ADC weighted value. The offset from the center of strip is also taken into account in UTHit x value estimation.

Run with LHCb!3670 (closed), Moore!1647 (closed)

assigned to @mrudolph, @xuyuan, and @tskwarni

added RTA label

mentioned in merge request Moore!1647 (closed)

changed the description

mentioned in merge request LHCb!3670 (closed)

approved this merge request

• [2022-07-06 17:57] Validation started with lhcb-master-mr#4961
• [2022-07-08 14:34] Validation started with lhcb-master-mr#4974
• [2022-07-12 17:06] Validation started with lhcb-master-mr#5008
• [2022-07-28 22:37] Validation started with lhcb-master-mr#5254
• [2022-09-23 11:09] Validation started with lhcb-master-mr#5733
• [2022-09-23 15:28] Validation started with lhcb-master-mr#5739
• [2023-06-14 00:36] Automatic merge failed in [lhcb-master-mr#8232](https://lhcb-nightlies.web.cern.ch/nightly/lhcb-master-mr/8232/Rec/checkout

mentioned in issue Moore#456 (closed)

added 159 commits

• aa49d4e8...842bce81 - 158 commits from branch master
• 8e7f4d22 - adcwgt

Compare with previous version

mentioned in issue Moore#465 (closed)

mentioned in issue Moore#467 (closed)

added Decoding UT labels

marked this merge request as draft

marked this merge request as ready

added 157 commits

• 8e7f4d22...7e2ecf80 - 156 commits from branch master
• 9f02130e - adcwgt

Compare with previous version

Hi @xuyuan!
I ran some tests to check if the drop in efficiency is related to tracking or something else.
As the tests that @decianm pointed us to are not using the latest decoding version, I could reproduce the efficiency drop by reverting this line here (and here: LHCb!3670 (diffs)).
The fix should be implemented as something like if(UTDAQ::version::v4) offset /= 4.;, but I don't know exactly where this should happen.

@xuyuan I'll mark this draft for now, please mark it ready once you've addressed this

Hello @xuyuan

There is something else I realized: The weight is not updated according to the different sizes of the clusters, but I would assume the size of the uncertainty depends on the size of the cluster, and this should be taken into account for the chi2-fits. Could you make the decoder return the cluster size, such that one can actually use that information?

marked this merge request as draft

mentioned in merge request Moore!1703 (merged)

marked this merge request as ready

mentioned in issue Moore#598 (closed)

mentioned in issue Moore#599 (closed)

This MR cannot be tested due to merge conflict. @xuyuan can you please have a look?

marked this merge request as draft

Dear reviewers. We are trying to sort out the situation with changes in !3005 (closed), LHCb!3670 (closed) and Moore!1647 (closed). Since then, we moved this MR to draft.

Any news on this ? LHCb!3670 (closed) has been closed recently. Is this MR still relevant ?

mentioned in issue Moore#623 (closed)

This MR is not relevant. The changes here were absorbed to other MR, and many of them are already in master.

closed