Uppsala Software News @ 961126
                ******************************

* RAVE
******
Almost all RAVE programs have (finally) been changed to allocate memory
for maps and masks dynamically.  This means that you can change the
sizes of maps and masks that these programs can handle on the fly,
either through the environment variables MAPSIZE and MASKSIZE (must
be all capitals), or through command-line arguments.  Similarly,
you can define the maximum number of maps that MAPMAN can handle
with NUMMAPS (max. 10), and the number of masks for MAMA can be
set with NUMMASKS (max. 10).  A similar mechanism has been implemented
for DATAMAN (see below).

Examples:
- put "setenv MAPSIZE 10000000" etc. in your .cshrc file for your
  run-of-the-mill work
- use "run mapman mapsize 20000000 nummaps 1" to run MAPMAN with
  just one big map
- use aliases: "alias fatmapman 'run mapman mapsize 30000000 nummaps 1'"

Notes:
- you cannot allocate more memory than can be addressed by your
  machine (or than you have physical + virtual memory)
- see the program manuals for details (point your Web-browser to:
  http://xray.bmc.uu.se/~gerard/manuals/gerard_manuals.html)
- the "4dbig", "6dbig" and "virus" versions previously available from
  the ftp-server have been removed, since they are no longer needed
- I am working on a brand-new version of ESSENS, so the currently
  distributed version does not have dynamic memory allocation.
  However, the new version will have this, as well as some cute
  new features.  Also, "ye olde mappage" program has not been
  updated since you shouldn't be using that in the first place
  (use MAPMAN instead)
- the programs all seem to work on SGIs, ESVs and ALPHAs, but I
  haven't done any extensive testing.  Therefore, if you run into
  any sudden coredumps, it's probably my fault.  In that case,
  please send an E-mail to: gerard@xray.bmc.uu.se explaining
  what you did when the coredump happened.  Thanks.

The programs which use dynamic memory allocation are:
    - MAPMAN  (version 5.0; NUMMAPS + 1 maps  )
    - DATAMAN (version 5.0; see below         )
    - MAMA    (version 4.0; NUMMASKS + 2 masks)
    - AVE     (version 4.0; 2 maps, 1 mask    )
    - MAVE    (version 3.0; 2 maps, 1 mask    )
    - COMAP   (version 2.0; 3 maps            )
    - COMDEM  (version 2.0; 3 maps            )
    - IMP     (version 2.0; 2 maps, 1 mask    )
    - NCS6D   (version 2.0; 1 map             )
    - MAPFIX  (version 2.0; 1 map             )

Thanks to Morten for pushing me to finally get this done ;-)


* DATAMAN 5.0
*********
Minor change so that MTZDUMP files are read again.  The program now
also reads the cell constants from an MTZDUMP file, and copies all
information in the header to the screen.
  Slightly improved CIF output files.
  The program now uses dynamic memory allocation.  The maximum number
of reflection data sets is controlled by NUMSETS (max. 10), and the
maximum number of reflections per data set by SETSIZE.  As with the
other RAVE programs, these parameters can be set as environment
variables and/or as command-line arguments.  DATAMAN needs approx.
(4 + 8 * NUMSETS) * SETSIZE words for its major arrays.


* AVE 4.0
*********
Uses dynamic memory allocation.  Bug fix (wrong interpolation
routine was called, although the numeric differences are minute;-().


* NCS6D 2.0
*******
Properly initialise the array of the 100 best RT operators prior
to the search (thanks to David Borhani).  Note that only solutions
with a correlation coefficient greater than zero will be listed.
  Uses dynamic memory allocation.


* MAPMAN 5.0
********
Implemented PIck HIgh option (e.g., to pick high values in a rotation
function map, rather than just the peaks).
  Implemented dynamic memory allocation, allowing the user to define
both the maximum number of maps, and the maximum size of maps that
the program can handle.


* MOLEMAN2 1.1.1
**********
Changed X-PLOR "generate" input files produced by the SPlit command
so as to delete hydrogens (and atoms with unknown coordinates), in
accordance with Axel's new practice.


* XPLO2D 2.2
********
Changed X-PLOR "generate" input files produced by the AUTO command
(no hydrogens etc.).


* SEAMAN 1.2
********
Fixed a nasty bug which *sometimes* occurred for commands like ALAN,
GLYC, MINI, BFAC etc. (thanks to Margareta Ingelman).


* CELLO 0.9.4
*******
Minor bug fix.


* MOLEMAN 7.2.4
*********
Minor bug fix.


* PACMAN 0.6
********
Minor bug fix.


* AUTODEP Hints
***************
Those of you who have been struggling with bounced PDB files while
using the PDB's new automatic deposition facility may be interested
to know that MOLEMAN2 can fix some problems that may occur:
- the PDB doesn't like O's chemical element number (6 for carbons,
  etc.); simply reading and writing the file with MOLEMAN2 will get
  rid of these numbers
- the PDB complains if you call your hetero-atoms ATOMs instead of
  HETATMs.  Use the MOLEMAN2 command PDb HEtero Deduce to automatically
  sort this out
- non-standard atom and residue names can be changed globally with
  the PDb NAme command
- PDb Chemical will add the appropriate chemical element type
- the CHain commands can be used to generate chain and segment names


--cd

CD's Word of the Month: Uxorious