I'm on the scrounge for a bit of help, although it will cost you
nothing. It involves testing some software. It should take < 15
minutes on reasonably modern hardware.
Some time back I wrote some software called 'atlc' for the analysis of
two-conductor transmission lines. This was published in QEX nearly 7
years ago - see http://atlc.sourceforge.net/qex-december-1996/atlc.pdf
if interested.
Current versions of atlc handle *any* cross section (even a bit of
u-shaped material if you wanted). It also handles couplers too and can
do synthesis too (design you a coupler for example).
Over the years I have developed 'atlc', although I'm no longer
bothering to worry about Windoze - only UNIX or Linux systems. The
code can be downloaded from:
http://atlc.sourceforge.net/
(try only the latest version).
If anyone has access to a UNIX or Linux box, I'd be keen to know if
this works on their system.
Extract the .tar.gz in the usual way, then configure with:
% configure --with-threads --enable-hardware-info
% make
% make check // most important this stage.
% make install // not that important, so don't bother unless you want
to.
I'd like to know:
a) The hardware you use - Pentium IV PC, Sun SPARC, IBM RS/6000 etc.
b) The operating system - Redhat Linux, Debian Linux, Solaris, AIX,
HP-UX etc.
c) The version of the operating system you use.
d) The compiler you use (gcc, Sun's cc etc) and it's version.
e) Any test failures that occur during 'make check'
f) What (if any) information the program prints about your system
(manufacturer, operating system, cpu type, number of CPUs you system
could take, number of CPUs actually present etc etc). Also if that
information is correct!
g) The results reported about multi-processor performance. Even if
your system only has one CPU, this information is useful to me,
although 'atlc 'should NOT be configured (except for these test
purposes) with the --with-threads option if you only have one CPU.
Doing so will just slow atlc down. On systems with more than one CPU,
it should speed things up.
If any problems occur, try configuring *without* the --with-threads
option and/or the --enable-hardware-info options. Both these options
are a bit less portable, so might cause compilation problems or test
failures.
The program seems quite portable, but I'm limited by the number of
hardware/software systems I can check it on.
If anyone has problems building/running etc. atlc, I would not mind a
bit of help resolving them but that is not important. If I'm made
aware of problems, I might be able to obtain access to the hardware
that is causing them and resolve them myself.
PLEASE DON'T CLUTTER THE MAILING LISTS/NEWSGROUPS WITH RESULTS - SEND
THEM BY EMAIL TO ME.
--
Dr. David Kirkby,
Senior Research Fellow,
Department of Medical Physics,
University College London,
11-20 Capper St, London, WC1E 6JA.
Tel: 020 7679 6408 Fax: 020 7679 6269
e-mail davek@medphys.ucl.ac.uk web:
http://www.medphys.ucl.ac.uk/~davek
|