Sphenix-production-l mailing list

[Alex Lebedev <lebedev AT iastate.edu>]

This has been discussed in various mattermost channels, but if

someone has missed it, here are the details.

The calorimeter reconstruction has been split into two consecutive steps. Instead of a single step reconstruction which produced DSTs from

packets and was written to

/sphenix/lustre01/sphnxpro/physics/slurp/caloy2test

we now have two steps.

The first step does waveform fitting (which is database independent

and slow) and writes to

/sphenix/lustre01/sphnxpro/physics/slurp/caloy2fitting.

It is assumed that this step will be done only once (or if event building

code changes).

The second step uses the results of waveform fitting to produce

DSTs and writes to

/sphenix/lustre01/sphnxpro/physics/slurp/caloy2calib.

This step is quick and can be done multiple times if the

database is updated.

So, what was previously in caloy2test goes now to caloy2calib.

Sasha.