Re: gEDA-user: one schmatic ==> two netlist

[sdb@cloud9.net (Stuart Brorson)]

Hi --

> The problems are comeing up, when I have transistors and digital gates in=
>  the=20
> same schematic. I have to check out what your spice netlister is doing wi=
> th=20
> my digital symbols. I have to find a way to get them as subcirciuts in th=
> e=20
> netlist.

Offhand, I think that you will be slightly dissapointed with
spice-sdb.  It outputs SPICE files with SPICE3f5 devics, and won't
recognize the digital parts natively 'cause they're not SPICE3f5.
Therefore, it won't just "do the right thing" without some tweaking.
And in any event, it sound like what you want to do is slightly more
sophistocated than what spice-sdb does.  Correct me if I am wrong.

But anyway, if you have a .subckt description of each digital part,
you can do the following:  Make the refdes of each digital part start
with X (e.g. Xor1, Xand1, etc.).  Place the name of the digital model
or subckt (i.e. the name on the .subckt line) in the digital part's
"model-name" attribute.  Then place the full path to the part's .subckt
file into the "file" attribute of the digital part.  Then, when you
netlist the netlister will stick the contents of the .subckt file into
your netlist.  This netlist will be flat, of course, but it might
help.

(Read the help spew which spice-sdb emits; you can use the -I flag to
just insert a .include into the main netlist if you prefer.)

IMPORTANT NOTE: (sorry for shouting)  The netlister has a bug, in that
it outputs the port pins of the .subckt in reverse order when it
writes the file's contents into the main netlist file!  If you want
to try the netlister, I will post a patch to fix this bug.  I have
been holding off on this because I didn't want to break legacy
schematics.  However, I can't imagine that there are many legacy
schematics out there using my stuff, so now is the time to fix it, I
guess.  (I was gonna do it before Ales announced the next rev of gEDA,
in any event. . . . .)  Please say the word if you want to try
spice-sdb & I will fix it.

> I will try ....

Good luck,

Stuart