[Frost] CHI-SQ table has huge number in the super-layer1 and super-layer2

[Franz Klein]

Sungkyun,
trk_mon_lite makes column based ntuples, which are used in dc3. Such sort 
of ntuples are known to be very limited. For the number of parameters in 
these ntuples you cannot collect more than about 64000 events (I've to 
check the dc3 code but I guess that it uses the last cycle, i.e. the 
option N100000 will end up as ~36000 events). I would not play with number 
of events in trk_mon_lite (dc3 won't read in more events anyway). Taking a 
small number of events does not make any sense for DC calibration (the 
parameters are already so correlated that you might get crazy 
solutions!).

Thanks for doing this tedious job
Franz


On Tue, 22 Feb 2011, Sungkyun Park wrote:

> Hi Eugene,
>
> I had some problems when I used trk_mon_lite. So I have added the option to set a limit on the events in trk_mon_lite.
> In the most of runs, the error was not happened when events are 60000.
> After using runs with N6000, I have one problem, which is that CHI-Sq values in super-layer 1 and super-layer 2 had very big.
> They had sometimes more than 20.
>
> The following is one example. run number is run 63407.A05.B00
>
>        CHI-SQ Summary
>          sec1    sec2    sec3    sec4    sec5    sec6
> sl1 :     24.59    0.35    10.34    0.10    11.30    0.90
> sl2 :     9.54    0.20    0.12    0.16    0.09    0.15
> sl3 :     0.25    0.13    0.13    0.16    0.18    0.12
> sl4 :     0.27    0.14    0.13    0.18    0.13    0.14
> sl5 :     0.19    0.19    0.22    0.29    0.18    0.15
> sl6 :     0.18    0.17    0.17    0.20    0.23    0.21
>
> There are huge chisq numbers in sl1-sec1, sl1-sec3, sl1-sec5, and sl2-sec1
> I guessed the reason is the limit of the total events in trk_mon_lite.
>
> I can make hbook file with N100000 in run 63465 (for DC alignment).
> I had one test. that is, I had DC Calibration in the exact same process using runs with N100000 and N60000.
> The DC3_N100000.jpg has the CHI-SQ table and fitting function (sl1-sec1) made from run with event 100000.
> The DC3_N60000.jpg has the CHI-SQ table and fitting function (sl1-sec1) made from run with event 60000.
>
> In the first jpg file, there are not huge CHI-SQ. In the second file, there are.
> I think that I can not use the option for the event limit any more in trk_mon_lite.
>
> My error was the following:
>
>   RZOUT: current RZ file cannot support > 64K records
>
>          or individual directories > 64K
>   RZOUT: previous cycle(s) for this key (      30)  deleted
>   RZOUT: please consult ZEBRA manual for further details
>
>
> I have followed the solution suggested in the part, "How to create very large ntuples (RWN or CWN)" of http://www.jlab.org/~brads/Manuals/pawfaq/. However, I can not fixed my problem.
> Do you have any idea about this problem?
>
> Sung
> Florida State University
>

===============================================================
                   Franz J. Klein, Associate Professor
                   CUA, Department of Physics
                   Washington, DC 20064
   office: Hannan Hall 206          phone: 202-319-6190
   or: Jefferson Lab,CC F-243       phone: 757-269-6672
---------------------------------------------------------------
-------------- next part --------------
A non-text attachment was scrubbed...
Name: sp06k.vcf
Type: text/x-vcard
Size: 267 bytes
Desc: Card for Sungkyun Park<sp06k at fsu.edu>
Url : https://mailman.jlab.org/pipermail/frost/attachments/20110222/6b4a4678/attachment.vcf 
-------------- next part --------------
_______________________________________________
Frost mailing list
Frost at jlab.org
https://mailman.jlab.org/mailman/listinfo/frost