The problems associated with the 
measurement are similar to
those of the multiplicity measurement. The major additional
complication is the determination of the vertex to allow
calculation of the pseudo-rapidity
(
).
measurements are generally averages for
many events. Statistical fluctuations
in the average over many events are therefore less
important than the event-by-event fluctuations in the total multiplicity.
Consequently, measurements should be possible for p+p, p+Au, and
Au+Au.
First, the vertex
position must be found, this is discussed in the following section.
Next, the
range of 
occupied by each chip is calculated. Because the
chips with strips parallel to the beam give more reliable information on the
number of hits, only those
are used to calculate the number of hits (Nhit)
in each range of . Corrections
for the efficiency, noise, double hits, and charge sharing are made.
The "measured" value is the average of
over many events.
Fig. 3 compares the "real" and "measured" distributions for
p+p, p+Au, and Au+Au events. The shapes of the "measured" distributions are
always close to the "real" distributions. Since the distributions are
calculated using the vertex found by the pseudo-tracking algorithm, which does
not always find the correct vertex, some of the differences may be from events
with an incorrect vertex position used in the calculation.
As with the multiplicity measurements, noise complicates the
measurements. The number of particles which hit each chip is much smaller near
the edges of the detector, but noise causes a constant
fraction of the strips to be "on". This means
that the signal/noise ratio is much worse (factor of 
10) at the
largest 
values. For p+Au, even with the optimistic assumption
that 
, the signal and noise will be comparable around
, resulting in larger statistical uncertainties on the points
around these values. This noise problem will be worse for p+p collisions
where the multiplicity is lower, but unimportant for central
Au+Au events.