Core Minutes 11/23/2010

• L1: (Richard) We made the switch last week to using the new Dataclean event class.

• ScienceTools: (Jim) No new tags. The exposure issues from last week are still outstanding. He will write a new app to replace gtexpcube, probably this week.

• Pass8: (Leon) He and Tracy have been working on incorporating information from truncated planes* into the new tree-based pattern recognition. It is now close to ready to commit and a new GR tag is likely to follow shortly thereafter.

  *The term truncated plane refers to half-planes whose readout may have been truncated. In order to avoid the possibility of saturating a cable, no more than 14 hits are read out from any half-plane.

• Documentation: (Chuck) sends the following:

  Progress on the likelihood tutorials was slower than expected, primarily due to my errors while testing the unbinned likelihood. I've since filled in several gaps in the setup documentation and have finished tweaking the procedures. Tasks remaining include testing the binned likelihood, updating and testing the navigation elements, then checking everything into CVS so that the subject matter experts can review everything before it's posted to the web. Thanksgiving and SME review will be the gating factors. In the meantime, I plan to continue working on updates to the remaining Science Tools and the last newsletter of the year.

• ADASS Recap: (Richard) was presented at last Friday's SCA meeting.

• Core talk:

  • New Gaudi and vc90 (Joanne) is waiting for a vc90 debug build. Heather provided these details:

    Gaudi Debug build on windows is progressing. All Gaudi externals are now built in debug mode. Nonetheless, there remains a dependency on vsvcrt90.dll, which should be vsvcrt90D.dll. Searching through the cmt requirements files to fix that up...while I continually locate new places where /MD is set rather than /MDd, the problem is not yet solved.

  • More notes from Heather:

    • GR branch tag (Heather) is putting out a GR v15r47p12gr16 with a patch to ldfReader to properly initialize all AcdDigi members. This avoids the potential for garbage values in the PHA and range in the case where there is a "veto", a bit in the hit map for a particular PMT, but the associated accept map bit is not set.

    • Concerning the problem associated with Johann's new class in reconRootData: the class is not being written properly to the output ROOT files.

    • Heather will be out of office Wednesday through next Monday but will have internet

    • No SCons meeting this week.

previous

minutes index

next

J. Bogart, Last Modified: