[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Help-gnucap] Gnu cap src help
From: |
_griv |
Subject: |
Re: [Help-gnucap] Gnu cap src help |
Date: |
Fri, 24 Aug 2012 03:59:24 -0700 (PDT) |
Felix Salfelder wrote:
>
> On Fri, Aug 24, 2012 at 01:13:36AM -0700, _griv wrote:
>>
>> Does anybody know where in the source code of Gnucap i can find the
>> algorithm
>> used to convert the netlist inserted into nodes in order to take place
>> the
>> nodal analysis and LU decomposition for analog simulation.
>
> Hi Chris.
>
> it roughly looks like this:
>
> while the netlist ist parsed, devices are stashed into a CARDLIST, see
> u_lang.cc, LANGUAGE::new__instance(...). that cardlist reappears in
> SIM_DATA::init() (u_sim_data.cc), here "expand", "map_nodes" and
> "*iwant_matrix" are called on all devices. after this, all devices are
> present and know where to write their matrix contributions. then look at
> s__solve.cc, solve(). here, evaluate_models() tells the devices to add
> their contribution to the matrix & right hand side, solve_equations
> finally does the LU decomposition and calculates the preimage _sim->_v0.
>
> there are more things happening, of course. feel free to ask for
> details. what are you trying to do?
>
> regards
> felix
>
> --
> () ascii ribbon campaign - against html e-mail
> /\ www.asciiribbon.org - against proprietary attachments
>
> _______________________________________________
> Help-gnucap mailing list
> address@hidden
> https://lists.gnu.org/mailman/listinfo/help-gnucap
>
>
Hello Felix,
thanks for your quick response
I will try to follow the path you suggested.
What i really want to do is understand the
way the information of the netlist is
stored into matrices and how the modeling of
resistor networks is done, in order to be able to apply
an algorithm which aims to the reduction of large resistor networks.
--
View this message in context:
http://old.nabble.com/Gnu-cap-src-help-tp34343039p34343610.html
Sent from the Gnucap - Help mailing list archive at Nabble.com.