Dotcards.c


To ng-spice-devel@ieee.ing.uniroma1.it
From Alan Gillespie <alan.gillespie@analog.com>
Date Mon, 11 Sep 2000 16:32:11 -0400
Delivered-To mailing list ng-spice-devel@ieee.ing.uniroma1.it
Mailing-List contact ng-spice-devel-help@ieee.ing.uniroma1.it; run by ezmlm
References <20000911220516.A1952@cgmd80028.chello.nl >
Reply-To ng-spice-devel@ieee.ing.uniroma1.it
Sender agillesp@epc.co.uk

Hi Folks,

about that dotcards.c file. I haven't looked at it yet, but as far as
I remember, all the fixes in there were either cosmetic or to do
with getting data out from Spice when you run it in batch mode
as opposed to compiling bspice, the batch version. I don't think
it would do any harm to the real fixes that I've implemented if
you just remove all the changes I made in there.

By the way, the outitf.c file that I gave you is a major hack. I don't
know if you tried to incorporate that or not, but most of the
changes there are to do with producing PSPICE format output
data. I don't think you want to put any of those changes in yet.

The main useful (to me anyway) change I put in outitf.c was
to stop all the internal device node voltages being saved in
the rawfile. I introduced an extra "save" flag which, if set,
causes all the internal node names to be changed to terminal
current names, and saved to the rawfile. This is done by -

.save allv alli

allv is the same as all, and alli saves all terminal currents of those
devices that have internal nodes. It was a crude hack which I'm
not proud of ;-) but it served it's purpose at the time. That's
definitely an area that could do with cleaning up.

Anyway, I suspect that if there's any problem with that file also,
it should work if you just remove all my changes. The model fixes
should still work, and the automatic gmin stepping and source
stepping algorithms will still work.

By the way, you get both of those by specifying gminsteps = 1
or srcsteps = 1. Those are set to be the defaults, so if you don't
specify either of them, you'll get the auto version. Setting either
to 0 disables the stepping.

Apart from those problems, has anyone noticed any improvements ?
Or have I definitely broken it :-)

Cheers,

Alan


Partial thread listing: