Where is my Filterbank Data
Unfortunately at this stage fining out what tape your filterbank data
is on is a non-trivial exercise. If you are reprocessing old data then
there are two ways to check which tape it is on
The first involves looking through the log books whcich live in
Dick's office. Not only is this method laborious it is often not very
accurate especially now that the observing is mostly controlled from
the Sun Workstation.
The second method will be more accurate but not necessarily any
less laborious. The first place to check for old data is in the
obsarch.log and obspsr.log files in
/psr3/data/log. These contain an extensive list of all timing
observations up to May 1995 (It is hoped to keep these files as up to
date as possible). The obsarch.log file lists all
observations by archive name, that is UT order, while
obspsr.log lists by pulsar order. The basic format of these files is:
psrname; UT; tape number; recorded file number; estimated file number
Note that there may be information after these initial four entries
and these will correspond to things like observing frequency and
sampling rates. There are a couple of other points to note, only the
number of the tape is recorded and there have been three different
letters used. The Q tapes are the most recent and any observation
after January 14 1995 will probably be on a Q tape. The P tapes
preceeded the Q tapes and also contain mainly timing data however
their numbering convention is not quite so simple. They go from 0-65
then from 866-999. The S tapes correspond to the 70cm survey tapes and
these contain some timing data also and they number from 0 to
approximately 650. So you see that there is cause for confusion
between the early P tapes and S tapes.
So the suggested course of action is to firstly go to obspsr.log
and find the tape number then if it lies in the zone of confusion go
to obsarch.log and look for the archive you want, if there are a whole
lot of archives on the same tape then chances are it is a P tape, if
not then it is probably a S tape.
Another difficulty is that sometimes the file number is recorded
incorrectly so the estimated true file number tries to
correct for that. However this may still be incorrect so it is best to
extract a number of files either side of the actual file you want as a
precautionary measure.
If your observation has not yet been added to the .log files then
it will be contained in a file on the vax, Bunyip, at Parkes. In order
to access it you will need to ask someone who knows the username and
password for this account. Once you are logged in type usur
followed by dir summ* which will list a number of summary
files. The most recent of these will be called summary.wk and it will
contain a list of all the observations made using the filterbank since
a file called Summary.MonthYear was made. To look at any of these files
type type/page filename. The current format is
Pulsar name; Date; Tape Number; File number; ...
there is a lot of other information stored there, but these are the
important entries in this case.