Back to the CSG main page
 
So What Data am I Supposed to Use Again ???


Monte Carlo

For Monte Carlo you should use the new smeared Monte Carlo as described on the CSG p17 page


My analysis is not affected/doesn't care/can deal with the wrong hadronic calibration

You can find links to all the data on the normal CSG p17 page. This includes skims of the fixed data and unaffected data from p17.09.00. That page also contains links to the corresponding CAF trees.

My analysis is affected by the wrong hadronic calibration and I want my results ready by Moriond

This assumes that you can work with the selections described on the refix page. Only a subset of all skims has been fixed. You can find there both thumbnail and CAF tree definitions (if available).

You should be aware that you have to take special measures for your luminosity calculation in this case.

In addition you can use the unaffected p17.09.00 data.

My analysis is affected by the wrong hadronic calibration but I'm aiming for Summer only

You can use the unaffected p17.09.00 data now (this is about 100pb-1). Check if you can do any of your studies with either the buggy dataset or the selected refixed sample if you need larger statistics.

We are in the process of refixing the whole affected dataset, but this has only started of January 14. You fill find instructions on how to get unskimmed and skimmed data here. Skimming will start soon after the refixing, so you should looking at the new data coming out of this as soon as you can.


Anything else I could do ?

If you are affected by the calibration bug, but still need a higher statistics sample for some study, you can try to use those runs from our initial dataset that were actually unaffected. Jan has put together a web page and a run list.

For practical purposes, here are two dataset definitions which allow you to run over the good runs in this list. Since these are merged files, you will still have to apply a filter on the good run number on the analysis level. Otherwise you might pick up events from a bad run merged together with a good one.

Here is how to define such a dataset (in bash, from Herb):
  setup sam
  skim="aCSGSkimName"
  def="CSskim-${skim}-PASS2-p17fix"
  newdef="MYDEF-GoodRuns-${def}"
  sam create definition --defName=$newdef --group=dzero \
  --dim="__set__ $def and ( \
  (run_number>=195165 and run_number<=195175) or \
  (run_number>=205253 and run_number<=205685) or \
  (run_number>=205961 and run_number<=206406) or \
  (run_number>=206567 and run_number<=209069) or \
  (run_number>=209148 and run_number<=209164) or \
  (run_number>=209831 and run_number<=209990) or \
  (run_number>=210291 and run_number<=210405) or \
  (run_number>=210515) )"

Comments, 20 January 2006