Pythia file, probably c-> mu, 1000 events Left: # planes hit, usually 4. n = 769 Right: first, last plane hit radiogram. n = 1324
Using hits in planes 1 and 4, point back to the vertex and plot z of the intercept. This is using the pisa hits directly. Left: intercept on large scale. Why are there entries in the non-zero bins? (scales are in cm). n = 662 Right: zero-bin expanded. Only xxx entries. Sigma 195 um. n = 528
Same thing, with hits 'integerized' into 50 um strips. n = 662 n = 528

This is WITH the barrel Now remove the barrel (SI01,2,3,4, SICT, SISP, SISR:

Run 104: Now remove the barrel (SI01,2,3,4, SICT, SISP, SISR:

root [2] .x resolution_1.C
 FCN=30.0216 FROM MIGRAD    STATUS=CONVERGED      69 CALLS          70 TOTAL
                     EDM=1.62199e-09    STRATEGY= 1      ERROR MATRIX ACCURATE
  EXT PARAMETER                                   STEP         FIRST
  NO.   NAME      VALUE            ERROR          SIZE      DERIVATIVE
   1  Constant     4.34746e+01   2.54666e+00   5.41944e-03  -3.98455e-06
   2  Mean        -1.55981e-03   1.01074e-03   2.74243e-06  -7.06657e-03
   3  Sigma        2.18635e-02   8.07205e-04   2.49604e-05  -5.34256e-03
 FCN=67.0223 FROM MIGRAD    STATUS=CONVERGED      75 CALLS          76 TOTAL
                     EDM=2.94698e-11    STRATEGY= 1      ERROR MATRIX ACCURATE
  EXT PARAMETER                                   STEP         FIRST
  NO.   NAME      VALUE            ERROR          SIZE      DERIVATIVE
   1  Constant     3.51972e+01   2.30582e+00   6.75759e-03   3.33097e-06
   2  Mean        -1.88074e-03   1.26277e-03   5.08510e-06  -1.00883e-03
   3  Sigma        2.53874e-02   1.24092e-03   4.73387e-05   6.41738e-04
root [3] .q

Run 106, barrel in place. Psi->mumu file, special format 1000 events. This file has a realistic vertex distribution (see panel 5). Left: # planes hit, usually 4. n = 1472 Right: first, last plane hit radiogram. n = 2944
Using hits in planes 1 and 4, point back to the vertex and plot z of the intercept. This is using the pisa hits directly. Left: intercept on large scale. Why are there entries in the non-zero bins? (scales are in cm). n = 1472 Right: zero-bin expanded. Only xxx entries. Sigma 203 um. n = 596
Event vertex distribution n = 1472 n = 597, sigma 242 um

Run 107, barrel removed. Psi->mumu file, special format 1000 events. This file has a realistic vertex distribution (see panel 5). Left: # planes hit, usually 4. n = 1472 Right: first, last plane hit radiogram. n = 2944
Using hits in planes 1 and 4, point back to the vertex and plot z of the intercept. This is using the pisa hits directly. Left: intercept on large scale. Why are there entries in the non-zero bins? (scales are in cm). n = 1472 Right: zero-bin expanded. Only xxx entries. Sigma 203 um. n = 596
Event vertex distribution n = 1472 n = 597, sigma 242 um

From runs 106, 107 are with the special psi->mumu file:

	without barrel	with barrel
perfect tracking	152	203
50 um 'strips'	192	242

Assuming these are gaussian quantities, we can estimate the 'error' introduced by the barrel scatter, and the error caused by the 50um resolution:
error due to barrel scatter: ~140 um
error due to strip resolution: ~125 um

Run 109: Barrel in place, but Multiple Scattering turned off (in pisa.kumac, MULS=0).