www.delorie.com/archives/browse.cgi   search  
Mail Archives: geda-user/2013/09/03/16:56:21

X-Authentication-Warning: delorie.com: mail set sender to geda-user-bounces using -f
X-Recipient: geda-user AT delorie DOT com
X-TCPREMOTEIP: 207.224.51.38
X-Authenticated-UID: jpd AT noqsi DOT com
Mime-Version: 1.0 (Apple Message framework v1085)
Subject: Re: [geda-user] Announcement: gnet-spice-noqsi
From: John Doty <jpd AT noqsi DOT com>
In-Reply-To: <B81CA3CB-4A4F-46B3-8604-3E9A5216937E@noqsi.com>
Date: Tue, 3 Sep 2013 14:56:06 -0600
Message-Id: <741CA36B-13AD-4040-B267-3E79669E6B64@noqsi.com>
References: <B81CA3CB-4A4F-46B3-8604-3E9A5216937E AT noqsi DOT com>
To: geda-user AT delorie DOT com
X-Mailer: Apple Mail (2.1085)
X-MIME-Autoconverted: from quoted-printable to 8bit by delorie.com id r83Ku3M1005314
Reply-To: geda-user AT delorie DOT com
Errors-To: nobody AT delorie DOT com
X-Mailing-List: geda-user AT delorie DOT com
X-Unsubscribes-To: listserv AT delorie DOT com

The installer should work on more versions now.

It doesn't seem to work on 1.9: gnetlistrc doesn't manage to disable hierarchy traversal. Is the gnetlistrc mechanism broken?

Thank you for your reports, both here and by private mail.

On Sep 2, 2013, at 1:42 PM, John Doty wrote:

> The spice-noqsi back end for gnetlist has reached a state where I believe it is ready for ordinary gEDA users. My goal has been to make it easier to construct schematics that can be input to both simulation and layout.  Past SPICE netlisters for gEDA have tended to force the user to maintain separate simulation and layout schematics for several reasons:
> 
> 1. The use of pinseq attributes for sequencing connections is incompatible with the use of pinseq for slotting.
> 
> 2. Component reference designator prefixes in SPICE are sometimes incompatible with printed circuit conventions.
> 
> 3. SPICE primitive models aren't always compatible with actual discrete components. The most common example is the MOSFET, where discretes rarely have a separate substrate pin, but SPICE needs one.
> 
> 4. Discrete device models may be subcircuits, even for "elementary" components.
> 
> 5. Previous SPICE netlisters have expressed hierarchy in a way that's incompatible with gnetlist's hierarchy traversal.
> 
> Previous SPICE netlisters have attempted to cope with these problems by identifying and fixing special cases. That works most of the time, but when it doesn't work the problems are intractable.
> 
> My new spice-noqsi netlister avoids these problems by recognizing some new spice-specific attributes. The most important is spice-prototype. Attached to a symbol, it describes how a symbol instance's attributes and connections should be expressed to SPICE. You don't have to attach these to most symbols: defaults handle most cases compatibly with previous SPICE netlisters. However, when you need to, you can take control.
> 
> To try it out, you'll need ngspice installed (and gaf, of course). Then:
> 
> git clone https://github.com/noqsi/gnet-spice-noqsi.git
> cd gnet-spice-noqsi/
> 
> (for the next step you may need to be superuser)
> ./install.sh
> 
> cd BBamp/
> make
> 
> You should obtain a plot of the frequency response of an amplifier.
> 
> Please try it out. Have a look at the README files, the schematics, and the Makefile to understand how it works. Send me reports, suggestions, complaints, ...
> 
> John Doty              Noqsi Aerospace, Ltd.
> http://www.noqsi.com/
> jpd AT noqsi DOT com
> 
> 
> 
> 

John Doty              Noqsi Aerospace, Ltd.
http://www.noqsi.com/
jpd AT noqsi DOT com



- Raw text -


  webmaster     delorie software   privacy  
  Copyright © 2019   by DJ Delorie     Updated Jul 2019