Hello, there are good news from the v3 libBlast candidate good enough that
we should crunch data only this way. Already at this stage (when final
geometry, eloss, final wire.cal are not implemented) data reconstructed
from v3 shows a considerable improvement in resolution. On a few recent
ABS runs - and before kinematical corrections - chris and chi could prove
that the new code provides
d_p 27 MeV (from 65 MeV)
d_theta 0.77 deg (from 1.1)
d_z 1.4 cm (from 2.5)
coplanarity 1.0 deg
as benchmarked from wc_res.C. A few points you ought to know for your
analysis,
- The analysis is still done by hand, the command is the following
cd ~/pro2003/analysis/CRUNCH/v3
lrn_v3 ###
- The compiled code is in cvs_v3/BlastLib2. It compiles with "make lrn_v3"
_ libBlast-v3 is not backward compatible with a number of our runs
At this stage is not even tagged, this is the famous 2.96 version...
_ By definition libBlast-v3 only stays in cvs_v3 and is not in the LD path
- While chi and chris have put some effort is done to provide bugfixes
(when possible) to libBlast.2.20 this is not a priority. However since
really only the crunch job requires v3, libBlast.2.20 is the default
library in ~/lib. In this way we don't have to update all our macros, at
least until v3 is more stable.
_ Keep working off cvs_v2 for changes, they'll be merged at a later stage
-- ________________________________________________________________________________ Tancredi Botto, phone: +1-617-253-9204 mobile: +1-978-490-4124 research scientist MIT/Bates, 21 Manning Av Middleton MA, 01949 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
This archive was generated by hypermail 2.1.2 : Mon Feb 24 2014 - 14:07:29 EST