Minutes for E895 Meeting at LBNL on August 19 Attending: Hans-Georg Ritter (HGR), Heng Liu (HL), Nathan Stone (NS), Randy Wells (RW), Mike Heffner (MH), Bill Caskey (BC), Mike Lisa (ML), Chris Pinkenburg (CP) HL - have run-wise gain calculation code ready. NS - have pack/unpack modules ready, but not in CVS - need to upgrade the TRKS table, and want to get consensus on that - will include "confidence levels" from HL - need to resolve range switches ML - for momentum switches, use 10X beam momentum for Maximum Pz - suggest we _copy_ helix parameters and covariance matrix instead of "packing" them NS - will check w/dieter about covariance matrix NS - any other XDF failures? BC - XDF will not compile on HP NS - strange... it compiles on PDSF HP. ML - Pentium dry run on KSTAR4.LBL.GOV - no exceptional problems - did not try it with XDF CP - I have development code that people can download from the Web ML - I don't like that. Can we do something else (more secure)? NS - I will create a write accessible "incoming" directory in CVS for dissemination of code. All "deposits" to CVS should go through this directory CP - what about am_vertex? NS - it isn't in CVS? CP - no. NS - I got it from Bill, and I thought I put it in. I'll make sure it's there. (editor's note: it was there) BC - I emailed Chris about the 150cm DCA values. Did you see that? CP - yes. but what is a track with DCA > 150cm? BC - a short track at the end of the TPC. (a track start) CP - okay, but that is somehow unsatisfying... CP - BTF is not filling the momenta and/or coordinates in the TRKS table correctly, as evidenced by the fact that we can't reconstruct V0's with BTF (but we can with JTF). - BADPAD table and gain decal/recal are available. NS - has anyone shown that running with gain/decal/recal works? CP - yes, but not with 2DH. ML - I will add the gain code to 2DH, but I want to know whether I have to apply two tables or one... CP - I will divide them and create one "absolute gain" table for you. CP - When pass1 is resolved (JTF/BTF) I will also generate T0's - will make one for JTF now. CP - I still have a slight "hole" in my reaction plane distribution HGR - you may just have to use the NA49/E877 methods to flatten the plane ML - but flatness is not the only point... HGR - but the point is that these methods are consistent ML - could drift velocity be causing this? CP - the problem is not up/down oriented, but left/right HL - I have drift V for these energies. I will put it in CVS BC - you do realize that acceptance effects can be left/right? CP - yes, but the peculiarity is that I am not losing Px/Py=0 or large Px, but small Px (Py=0) HL - but low Pt particles don't contribute to reaction plane anyway CP - yes, but high-Pt particles are extremely sensitive to the TPC angle, and if we cut out low Pt, that is a _lot_ of particles RW - looked at track splitting/merging - at 2 GeV (6 GeV) I saw 2.1 (0.6) split tracks per event RW - working on diagnostics Ntuples, and considering changing to tables instead SP - I should show you the E877 monitoring pages consensus - diagnostics should be stored (on disk) in metafile GIF and RZ formats. Only the most useful stuff should go to GIF, while other "moderately" useful things can be stored in RZ files. ML - what about distortion tables? MH - well, don't use VSCOPE data, nor EXB files. The ones called MPS6 are close, but not final. HGR - no meeting next week. ML - so can we start DST production with JTF, just to keep up momentum? HGR - you need to talk with Gulshan about that. ML - see the 2DH homepage for new/interesting details... CP - what about "low # good hits" error from am_dedx ML - you can turn up the values in the switch tables if you want to eliminate this, but it will cost you CPU. CP - everyone did get my email about purchasing DEC-alphas instead of PCs ? HGR - yes, but we have no VMS infrastructure to allow such a move SP - what about software for alpha? OS would cost LBNL $2K... CP - we'll have to check into that...