new isis-devel (isis-0.9.7)

From: John Houck <houck>
Date: Mon, 15 Jan 2001 01:14:47 -0500

I've installed isis-0.9.7 as isis-devel.  No setup scripts or
changes to your environment are necessary to use the new version
aside from setting a path to the binary.  For that, try:

  alias isis /nfs/wiwaxia/d4/ASC/src/isis-0.9.7/bin/${ARCH}/isis

Those not reading this in isis-users may be interested to know
that there is a mailing list for isis.  For instructions on
how to join, see the (recently redesigned) ISIS web page:
The isis-users mailing list is being archived locally in the
same area as acsds_help so, if you don't want to join,
you can still check periodically to see what's been posted.
There have been many recent changes, a few of which break

The main back-incompatible changes are:

(1) Instead of plasma ("aped") one now types 
          isis> plasma(aped);
    Actually the plasma() and atoms() functions now take 
    a structure argument instead of a string.  
    As an example of local site-configuration, 'aped' is
    a function which generates the necessary Struct_Type.
    See $ISIS_SRCDIR/ (which, if it exists is 
    automatically loaded at startup and used for local
    btw: I still don't have the SPEX database in this format --
    if anybody needs it, I can probably get it.
(2) The data-access functions 
    now return structures.  Instead of this:
        (lo, hi, value, err) = get_data_counts (1);
    one would use this:
         d = get_data_counts (1);    % d is Struct_Type
(3) Use of the mode-dependent functions
        get_data, put_data, get_model, put_model
	plot_data, plot_model
	fit, eval_fun
    is discouraged.
    The replacements are get_data_counts,
     Although they require more typing, these mode-independent
     functions should work better in scripts.  If typing is too
     much of a pain, remember that you can make aliases for
     common functions -- e.g.
         isis> alias ("plot_data_counts", "pdc");
Several new features have also been added:

(1) This might be old news, but John Davis's FITS module 
    and pileup kernel are now built-in as are the low-level 
    PGPLOT intrinsic functions.
(2) I've added an xspec-like 'renorm' function which should
    simplify setting the initial normalization of a fit.
    I also added a group_data() function to simplify
    grouping data in groups of N bins instead of grouping 
    to get a given minimum number of counts per bin.

(3) Isis now handles the atomic-data file layout in the ATOMDB
    directory included with ciao.  However, because the ciao
    distribution is missing a couple of files, for now I'm
    pointing the isis/aped database at my own atomdb directory and
    not the ciao one.
(4) I finally updated the entire manual so (almost) everything is 
    documented in some way.  Unfortunately, due to problems with
    latex2html, I have not yet been able to generate an HTML
    version so the new version is available only in postscript
    or in the ascii help files which are searchable at run-time.

The manual still needs work and the recent changes need more
testing. If you're interested in testing out the new version, the
best place to look for bugs is probably the model fitting code.

If you find any bugs or would like clarification of the recent
changes, please let me know.

I'm sure I've left out some stuff, but thats more than enough
for now.


You received this message because you are
subscribed to the isis-users list.
To unsubscribe, send a message to
isis-users-request_at_email.domain.hiddenwith the first line of the message as:
Received on Mon Jan 15 2001 - 01:14:57 EST

This archive was generated by hypermail 2.2.0 : Thu Mar 15 2007 - 08:45:50 EDT