I think the problem this time is in BlastLib.
I am having problem with reconing monte-carlo using newest MC and
BlastLib(both v3) too.
My problem is even worse in that nsed/lrn skipped right to the end of the
data file. seems some problem with TBLRaw.
However, 2.96 keeps working!!!!!
if I use b2_96 for both blastmc and BlastLib2, MC+recon work as
expected.
if I use lastest blastmc and b2_96 BlastLib2, it works. But one has to
specify -m option to use electronics map for MC, with which one has to use
linear time to distance. So one must check the last few lines of fort.99
to make sure 1 is set in WTDC 1.0 1100.0 6000.0, not 2. update to latest
fort.99 in Blast_Param v3, and Aaron has comments on this I could not
imagine to be any more detailed but transparant.
Otherwise, misreconstruction and losing of track occur due to the use of
incompatible t2d/d2t.
Unfortunately, I have not been able to make latest BlastLib and
blastmc(both v3) work together.
Good we have not decided to migrate to v3 just yet.
So the solution for the moment could be:
blastmc: use latest but set the WCTDC switch properly,
use b2_96 which does not have the swtich at all
BlastLib2: use b2_96.
It is probably a little inconvenient for Vitaliy as he has to patch two
blast.geom files to correctly position NC in both blastmc and BlastLib2.
but it is better than nothing at all.
since v3 BlastLib is reconstructing real data, I am sure it is just some
minor gliche somewhere we still have to smooth out. before declaring a
fully functional new version.
Chi
This archive was generated by hypermail 2.1.2 : Mon Feb 24 2014 - 14:07:30 EST