Usage of sisrmg v1.0

["Koji Mukai, US ASCA GOF, (301) 286-9447" <MUKAI@LHEAVX.GSFC.NASA.GOV>]

Dear ASCA FTOOLS users,

Several subtle interface issues regarding sisrmg v1.0, recently released
as part of FTOOLS v3.5, have been brought to our attention.  These may lead
to users generating SIS response matrices that are not appropriate for their
data.

(1) In most cases, the second parameter (Name of ARF input file) should
be "NONE", which is the default, since typically the rmf file is made
before the arf file is created (using ascaarf).

(2) In general, this version of sisrmg uses information found in the
input PHA file.  However, the parameter file contains many hidden
parameters that can be used to _overide_ whatever is in the PHA file.
This should not be necessary for most users (see the exception below).
If you ever use the override particularly by editing the parameter file,
it is adviseable to the local version of sisrmg.par so as to return
to the system defaults.

(3) The exception is the grades parameter (default value UNKNOWN).
However, the current PHA files do not record any grade selections
that may have been applied during Goddard processing, in ascascreen,
or in XSELECT, in a machine-useable form.  To ensure that matrices
of correct grade combination be generated, please specify the grades
on the command line (usually grades=0234; grades=02346 if grade 6
events have been included).

This is not an interface issue, however:

* Dispte some statements in the release note to the contrary, sisrmg
v1.0 does run on Sun machines for fast mode data.  However, the release
note was not completely wrong: sisrmg v1.0 does have a machine-dependent
bug for fast mode data.

We will be presenting this and other issues regarding ASCA FTOOLS status
at the HEAD meeting next week; we will update the ASCA GOF home page
shortly thereafter for those of you who will not be attending the meeting.

				Koji Mukai for the ASCA GOF