[SI-LIST] REPOST: Optimization in Hspice

From: "Ingraham, Andrew" <a.ingraham@xxxxxxxx>
To: <si-list@xxxxxxxxxxxxx>
Date: Tue, 23 Dec 2003 10:53:00 -0500
I'm sending this again because it never made it to the list server.


----- Original Message -----
From: "Ingraham, Andrew" <a.ingraham@xxxxxxxx>
To: <si-list@xxxxxxxxxxxxx>
Sent: Sunday, December 21, 2003 09:50 AM
Subject: Re: [SI-LIST] [Fwd: Optimization in Hspice]

> Can anyone suggest that in hspice optimization, for improving accuracy
> which options we have to use. Means in a spice file I was given 1ns as
> goal but the result is not getting exactly 1ns. For getting exactly 1ns
> which options
> I have to use.

Can you be more specific?

"1ns as a goal" doesn't mean anything without more information.  Where is
this 1ns?  Is it a delay of some device in the circuit being simulated?  Or
is something supposed to happen at 1ns into the simulation?  Or do you need
SPICE's internal timestep to be 1ns?

If the delay of some element is supposed to be 1ns, and your boss has told
you to adjust it until you get exactly 1ns, then this is a circuit
optimization problem and your choices are unique to your circuit.  (i.e.,
adjust the elements within your circuit.)

If you just need something to happen at precisely 1ns into the simulation,
SPICE's independent voltage and current sources are under your control, and
you can tweak their initial delay times (or phases) to get the desired
result.

If you need SPICE to use 1ns timesteps, good luck.  SPICE (including HSPICE)
chooses its own internal simulation timepoints.  You can influence them in
various ways, but you cannot dictate what they will be.

One consequence of SPICE's choice of internal timepoints, is that SPICE
usually does not provide a solution at precisely 1ns, or at any other time
that you may request.  If you want to see the results at precisely 1ns, you
need to interpolate between the two actual simulated timepoints that
straddle it.  This can sometimes result in slight "errors" (or differences
in results) if you are trying to measure with sub-ps accuracy.

As for SPICE's accuracy as a simulation engine, take a look at the *TOL
options and start from there.  The HSPICE documentation should provide
clues about the other options that interact with the *TOL options.  Changing
these options will have secondary effects on things like element delay
times, but not necessarily in a predictable way ... use trial-and-error to
see what affects what with your particular circuit.

Regards,
Andy


------------------------------------------------------------------
To unsubscribe from si-list:
si-list-request@xxxxxxxxxxxxx with 'unsubscribe' in the Subject field

or to administer your membership from a web page, go to:
//www.freelists.org/webpage/si-list

For help:
si-list-request@xxxxxxxxxxxxx with 'help' in the Subject field

List technical documents are available at:
                http://www.si-list.org

List archives are viewable at:     
                //www.freelists.org/archives/si-list
or at our remote archives:
                http://groups.yahoo.com/group/si-list/messages
Old (prior to June 6, 2001) list archives are viewable at:
                http://www.qsl.net/wb6tpu
[SI-LIST] REPOST: Optimization in Hspice

Other related posts:
