next up previous
Next: ana10 program Up: slim2 program Previous: Developing your own version

Running slim2

Run program with command

slim2 -c control_file -r slim_root [-n] [-new/old]}

Control file should contain a list of reconstructed ROOT files with h10 tree you want to slim. Note, full path from current location (from which slim2 will be started) should be provided. Also it can contain other slimmed files. So one can add new portion of data or MC to existing slim, or maid reduction in several steps. Note that only raw data, or similar MC simulations cna be slimmed together. Also, do not slim 1e and 1e+ reconstructions of the same MC together (slim would not know that those are the same MC split into two, and would double the number of MC events simulated).

slim_root is a name for output file with preselected events.

Other two options are not necessary. E.g option -n 100 means to skip first 100 events. It is useful for test purposes, i.e. if slim2 crash at some event and you want quickly catch its number.

New/old option tells what type of ROOT files are slimmed. Due to small modifications in nemor preselecting, recently reconstructed files are slightly different. Basically, they contain number of generated events for MC, time for rawdata and run number information. If you do not specify this option, old type is supposed by default. If one use new files with old option, there is no harm, slim still will work OK. Just this additional information will be lost in slimed ROOT h10 tree.

Slim2 prints out processed file names, results array, and calculates number of MC events generated/total time (for MC/rawdata) for normalisation purposes in further analysis.

It is a good style to produce one slim for every MC component. Hopefully there is no difference between reconstructed and slimmed h10 trees. So one is able to provide slimmed ROOT as input for slim2 to make further preselecting. But be careful with number of MC event/time calculated at the end. Though all efforts was done to make it correct in some tricky combination of slim and unslimmed data it can be wrong. So it is worth to analyse program output screen.


next up previous
Next: ana10 program Up: slim2 program Previous: Developing your own version
Vladimir Vasiliev 2008-12-02