Bob,
Before you release a Linux version, we should chat about compilers.
Redhat 7.0 is probably not a good choice. I have to run but I'll try to
give you a call tomorrow.
Brian
Bob Von Dreele wrote:
>
> Hi everyone,
> I have made a new "flavor" of GSAS for MS Windows. You'll find it in the
> directory "windows" at ftp://ftp.lanl.gov/public/gsas. It is a new
> compilation with the g77 compiler for Windows so that the code is identical
> to that used for the Linux version of GSAS. The biggest change is that the
> graphics is now "true" Windows graphics (i.e. not DOS graphics). The
> executables are also "true" 32 bit (i.e. not DOE extended 32 bit) and run
> in a console window. This version will work in any of Win95/98/NT4/2000.
> The graphics is produced in a separate window which has a multitude of
> viewing options; a given run of a GSAS graphics program will produce
> multiple windows each with a plot. These can be horizontally or vertically
> tiled or laid in a cascade. Each plot can be copied for pasting elsewhere
> (Word or PowerPoint). The plots have extensive zoom and scrolling
> capability for getting a close look and each plot can be dynamically
> resized. The two choices for the plot style are either black on white or
> visa versa.
> This version must be considered a "beta test" as there a few features &
> glitches:
> 1) Each GSAS program creates a zero length file with the same name as the
> program with no extension. These will clutter up a directory and may be
> deleted any time. I have no idea where these files come from. Any advice
> would be welcome.
> 2) The cursor/keyboard behaves differently when a plotting program has an
> active cursor. The keyboard gives no real response in Win2000 (maybe WinNT
> too) but works fine in Win98(95?). The mouse always seems to work. There
> maybe some machine dependence with this behavior.
> 3) The experiment and raw powder data files are fixed length direct access.
> This is as it always been. For this new version each record is 82
> characters long and have a CR/LF as the last two characters on each record.
> This makes the files appear to be like "text" files for DOS/Windows so they
> may be examined & edited without conversion. A routine CNVFILE has been
> included to convert unix style direct access files to this file structure.
> Moreover, EXPEDT, etc. will convert these files "on the fly". This is very
> similar to what the DOS & Linux versions have done recently. I have
> prepared a new Linux version that will behave in exactly this same way but
> am not quite prepared to release it yet.
> 4) The graphics can be stunningly slow if run at the same time as a big
> GENLES refinement is run. This is probably a function of processor speed &
> memory. I saw this on my 633MHz/128Mb laptop.
> 5) The distribution kit does not include PC-GSAS.EXE as it fails badly on
> WinNT/2000. The QBASIC routine gsas.bas & gsas.bat work fine. Don't tell
> Bill G. but I included a old copy of qbasic.
> 6) Pay special attention to the README as the installation instructions are
> slightly different from the DOS version of GSAS.
> Have fun with it and let me know of any bugs that show up.
> Bob Von Dreele
--
********************************************************************
Brian H. Toby, Ph.D. Leader, Crystallography Team
[EMAIL PROTECTED] NIST Center for Neutron Research, Stop 8562
voice: 301-975-4297 National Institute of Standards & Technology
FAX: 301-921-9847 Gaithersburg, MD 20899-8562
http://www.ncnr.nist.gov/xtal
********************************************************************
