Core Minutes 10/31/2006

Present: Joanne Bogart, Anders Borgland, Eric Charles, James Chiang, Seth Digel, Richard Dubois, Warren Focke, Berrie Giebels, Navid Golpayegani, Tom Glanzman, Heather Kelly, Michael Kuss, Bryson Lee, Julie McEnery, Chuck Patterson, Igor Pavlin, Leon Rochester, Tracy Usher

celebrating early

• Rerun of All-Gammas: A bug concerning computing ACD Active Distance had crept in to Gleam v9r16, so we needed to redo a 60-million event run. (Eric) The results with v9r17 look like v9r15, as they should. The new release also fixes another bug which had caused too many Active Distances (both upwards-going and downwards) to be computed. (Richard) Tom turned off production of ROOT trees both because of space concerns and because there was a rumor that not all ACD tuple quantities could be computed from the ROOT file. (Heather) The bug has been fixed; in any case the quantity which had been omitted was not one anyone cared about. [news from Richard: the v9r16 run was killed and deleted.  The v9r17 run is underway.]
• Preparation for one-day Gleam run:
  • Interleave (Richard, Heather) Goal is to upgrade interleaving such that
    • may have multiple background sources
    • an XML file specifies event selection parameters, among which are the locations of the ROOT files containing the event.
    Dan has designed a suitable format for the XML file. Heather is pinch-hitting for Dan by writing the code in the Interleave package which will consume it.
  • Status of ObSim (Julie) Can't say yet whether it will be ready in time.
  • Simulating level 0 data (Heather) See the familiar page. Take special note of the Halloween update concerning lsep_write (we're in the clear) and the section Demonstrating it all works as advertised (ongoing).
  • Running mulitple filters (Tracy) "I know nothing." He is attempting to find out, but hasn't yet gotten a response from Flight Software (very busy folks) or other knowledgeable people who are difficult to reach.
• Move to TClonesArray We have been using a "temporary" work-around (which doesn't always work) for the performance problems caused by TOBjArrays for over three years! Due to ROOT improvements, there is no longer a reason not to switch to using TClonesArrays everywhere. See Heather's comprehensive description of the situation and our options. The consensus was to go ahead with a backwards-compatible solution. Heather thinks December is a reasonable target date.
• Multiple calibrations: (Richard) For CAL the plan is to use calibrations from different eras and perturb one. Is this possible for Tracker? (Leon) See his discussion of how one might deal with different kinds of Tracker calibrations. (Anders) Hiro might know of calibrations done during thermal vac. (Leon) Michael might have alternate alignments. (Michael) does have something already for intra-tower. He will find some time to work on inter-tower.
• Miscellany:
  • Macs (Jim) Is there a Mac he can use for debugging Mac-only Science Tools problems reported by GSSC? (Richard) You might be able to use the Mac cluster at SLAC.
  • Beamtest (Richard) Will there be new ancillary data in the GSI runs? (Michael) He doesn't know. However a tile has been moved. [update from Luca L: no, there isn't any ancillary data.]
  • Builds (Richard) The proposal for old builds is to just keep libraries needed for ROOT trees.
• Saving the best for last:

  Heather: I guess you can also slide into the minutes that I'm staying on

previous

minutes index

next

J. Bogart, Last Modified: 01-Jun-2010 15:48:11 -0700