-Used StFgtAVEfficiencyMaker (see CVS)

-Use Disk 1 and 6 as seed (only first 2 quads in disk1)

-Look for estimated postion of hit in other disk

-if within 1cm (first plots) or 0.5 cm (x, y separately) success

-only accept events in which 3 or 4, respectively hits overall.

The efficiencies are pretty high, maybe there is still a bug in the code?

There is probably still a lot of noise, here I take all clusters, even those with a very 'weak' seed in which the seed strip is only required to have one high strip.

Run is this one:

drupal.star.bnl.gov/STAR/blog/avossen/2012/mar/02/radio-plots-with-new-peds-13061013

One indication of the noise is the ratio between r and phi strips hit (ideally 0) in each disk.

Not the case here:

Also the number of hits per plane is pretty high (overall I have ~4k events here) but mostly below 10 clusters in each plane after corrections for noise etc.

Here with 1cm cut and 3 hits req:

 1 cm and 4 hits required :

3 hits, 0.5cm:

4 hits, 0.5cm: