Release notes for CJC test version
====================================

1) 30.12.99

The cosmic track fit has been improved. Until now a weighted average of the two
track parameter sets has been put into the CJCS bank. Before the averaging
systematic errors have been added to both covariance matrices (Kappa:0.25E-4,
Phi: 0.15E-2, Dca: 0.25E-1). Now a fit is performed with the hits from both
tracks allowing for a track t0 correction and the correct flight time correction
for the upper track. With the results of this fit the RPhi parameters and the
RPhi covariance matrix elements (no systematic errors anymore !) of the CJCS
bank are updated. The cosmic t0 correction is put into column 6 of the CJCC
bank (and error in 14). Therefore the t0 of the cosmic is now the sum of columns
2 and 6 of CJCC.

In addition the new subroutine CJSWPG(IG1,IG2,NWSWP) allows to swap the wire
groups IG1 and IG1 in the CJC hit bank CRJE. At least for 1998 and 1999 the
groups 190 and 192 have to be swapped. For 2000 it has been fixed. CJSWPG has
to be called ONCE in the main program before CJCREC. It can't remember how often
its has been called. If corrected data is written out after the next call it
will be bad again !

2) 09.02.00

This version is using calibration correction banks (CSF8,CSF0) to update the
standard CJC calibration from NDB (C0J8,C1J8,CJCV). The correction banks have
to be read in before runstart. For this they to be put in the steering file:
 - directly
 - indirectly via DBCARD calls:
$NDB $TEXT
   DBCARD '/h1/h1rec9/csf8_ep99_test.car'
   DBCARD '/h1/h1rec9/csf0_ep99_test.car'
END$

New calibration 'degrees of freedom':
 - seperate Lorentz angle for CJC1 and 2
 - seperate drift vel. correction per layer SIDE (TLFC)

3) 03.08.00

This version is the result of the calibration with cosmics '99, for ep data most
probably some corrections (angle track-drift) are missing. New corrections
implemented:
  - variation of magnetic field (difference of expected drifttime for B=B(dca)
    and local field is added to measured drifttime per hit), energy loss in CJC
    not yet implemented
  - wire positions from survey (89/90)
  - anode/cathode field corrections (D. Duenckelmann) per layer
  - C1 cathode distortions per cell
  - drifttime corrections depending on pulse shape (slope)
  - hit resolution/weight function of layer, drifttime and pulse shape, not
    simply amplitude anymore
  - flight time corrections now ~1/sin(theta), only for theta=90 absorbed in
    wire t0
  - improved staggering, sagging parameters (staggering now deviating from
    construction)

Calibration banks needed: CSF8, CSF0, CJPC, CJFC, CSFC
To be read in with DBCARD:

$NDB $TEXT
   DBCARD '/h1/h1rec9/csf8_c99_test2.car'
   DBCARD '/h1/h1rec9/csf0_ep99_test.car'
   DBCARD '/h1/h1rec9/cjc_pos_corr.car'
   DBCARD '/h1/h1rec9/cjfc_c99_nf.car'
   DBCARD '/h1/h1rec9/csfc_c99_all7.car'
END$

4) 16.11.00

Differences to last version:

  - CJSWPG has NOT to be called anymore to swap groups in CRJE, instead the
    intermediate JRJE bank is corrected.
  - Online millepede has been implemented (al,vd,t0 per chamber), active for
    H1ENVI flag=3.
  - For single sided hits the Z dependence of the propagation time on the wire
    is corrected.
  - Effective isochrone radius implemented (0.25 cm)
  - Sin(theta) dependence of fight time correction implemented too in:
       CJROAD: used for event t0 determination
       CJUPSW: (old) wire t0 calibration with CJCLOOK

Calibration banks needed: CSF8, CSF0, CJPC, CJFC, CSFC
To be read in with DBCARD:

$NDB $TEXT
   PRINT=2
   DBCARD '/h1/h1rec9/csf8_c99_test4.car'
   DBCARD '/h1/h1rec9/csf0_ep99_test4.car'
   DBCARD '/h1/h1rec9/cjc_pos_corr.car'
   DBCARD '/h1/h1rec9/cjfc_c99_nf.car'
   DBCARD '/h1/h1rec9/csfc_c99_all7.car'
END$

5) 21.02.01

Differences to last version:

  - Usage of new calibration banks (CJRR,CJBG,CJFC,CJCF,CJLV,C2J8,CJPC)
    prepared.

  - Worsen RPhi resolution for hits from FADC with clock problems (CJBG).

  - Hits from crates with wrong event number (-> DAS4) killed.

6) 01.08.01

The results of the ZS/dEdx studies have been implemented:

  - Correction of FADC response function ('722' -> '755') with column 4 of C2J8

  - New ZS/dEdx calibration (-> C2J8) with superfit (inherited from V.B.),
    DEDXMOD will not fit anymore to new charge calibration !

  - Resolution in ZS now function of charge and layer (-> CJZR)

  - check for overflows (reduce weight in ZS fit)

  - Hit charge corrected for sin(beta), cos(theta), low P effects (-> CJQC)