[SI-LIST] [Re: Re: Spice simulators - HSPICE vs PSPICE]
- From: Peter LaFlamme <plaflamm@xxxxxxxx>
- To: si-list@xxxxxxxxxxxxx
- Date: Fri, 12 Oct 2001 09:08:56 -0400
-- Attached file included as plaintext by Listar --
X-Mozilla-Status2: 00000000
Message-ID: <3BC5F77A.92BA00BF@xxxxxxxx>
Date: Thu, 11 Oct 2001 15:48:10 -0400
From: Peter LaFlamme <plaflamm@xxxxxxxx>
X-Mailer: Mozilla 4.7 [en] (WinNT; U)
X-Accept-Language: en
MIME-Version: 1.0
To: "Sanchez, Louis" <louis.sanchez@xxxxxxxxx>
Subject: Re: [SI-LIST] Re: Spice simulators - HSPICE vs PSPICE
References: <46DB8150F9B3D311AC4700902761048402959618@xxxxxxxxxxxxxxxxxxxxx>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Hi Louis,
Here are the directions sent to me by an applications engineer at
Cadence. I have also included at the end of the directions all of the
options. I hope that this helps you and any others interested.
You can extract a Spice netlist from a SPECCTRAQuest topology by using
the spc2spc command. spc2spc creates a file named "main_gen.spc" which
is the Spice netlist of the interconnect in your Allegro board file.
The procedure is as follows.
>From SPECCTRAQuest:
* Initiate the Signal Probe command (Analyze -> SI/EMI Sim ->
Probe)
* Select the net and run your simulation, be sure to turn on the
"Save circuit files" toggle.
* In a UNIX terminal or MS-DOS command prompt, cd to the
appropriate signoise/case#/sim directory (where the *.spc files
are located), and enter "spc2spc"
>From SigXplorer:
* Open the topology file
* Run the simulation
* In a UNIX terminal or MS-DOS command prompt, cd to the
appropriate signoise/case#/sim directory (where the *.spc files
are located), and enter "spc2spc"
***************************************************************************
OPTIONS and Usage for SPC2SPC
***************************************************************************
usage: spc2spc [options] infile
Options specifying output file(s):
-mapOut=<file> Filename for connectivity map output
-spectreOut=<file> Filename for spectre language output
-spiceOut=<file> Filename for spice language output
Options modifying processing:
-commentInfo Include comments from source file before each
statement
-dir=<directory> Write spice output file(s) to <directory>
-fileInfo Include commented info about source file name and
line number
-flatten Flatten the circuit so that no subckts remain
-h Print this help message
-includes Add 'include' statements where needed.
-nelement Retain transmission line 'N' elements
-rename=<node,elem,subckt> Rename nodes, elems & subckts
Format codes are:
alphanum change to 8-char alpha followed by
numbers
asis do not change the name
clean change non-alphanumerics to '_'
numeric change to numbers
-sourceInfo Include commented source file text
-version Print software version information
-welement Use W element for all transmission line models
-wminlen=<len> Set minimum length for W elements e.g. 10.0mil
(if no units specified: assumes meters)
Options for tuning ladder network generation
-s Convert all single-line transmission lines to T
models
-t=<RF_time> Minimum rise/fall time in ps
-x=<seg_delay> RLC segment delay in ps
Default operation:
If "rename" option not specified
node names will be converted to numbers
If neither "welement" or "nelement" is specified
-spectreOut=<file> Filename for spectre language output
-spiceOut=<file> Filename for spice language output
Options modifying processing:
-commentInfo Include comments from source file before each
statement
-dir=<directory> Write spice output file(s) to <directory>
-fileInfo Include commented info about source file name and
line number
-flatten Flatten the circuit so that no subckts remain
-h Print this help message
-includes Add 'include' statements where needed.
-nelement Retain transmission line 'N' elements
-rename=<node,elem,subckt> Rename nodes, elems & subckts
Format codes are:
alphanum change to 8-char alpha followed by
numbers
asis do not change the name
clean change non-alphanumerics to '_'
numeric change to numbers
-sourceInfo Include commented source file text
-version Print software version information
-welement Use W element for all transmission line models
-wminlen=<len> Set minimum length for W elements e.g. 10.0mil
(if no units specified: assumes meters)
Options for tuning ladder network generation
-s Convert all single-line transmission lines to T
models
-t=<RF_time> Minimum rise/fall time in ps
-x=<seg_delay> RLC segment delay in ps
Default operation:
If "rename" option not specified
node names will be converted to numbers
If neither "welement" or "nelement" is specified
transmission line 'N' elements will be replaced by ladder networks
If no output file is specified
Spice language output will be written to "main_gen.spc"
***************************************************************************
***************************************************************************
Regards,
Peter
-----
Peter LaFlamme
Applied Micro Circuits Corp.
Staff System Applications Engineer
200 Minuteman Rd, 3rd Floor
Andover, MA 01810
978-247-8470 phone
978-623-0055 Fax
"Sanchez, Louis" wrote:
>
> Hello Peter...........
>
> I picked this message from the SI list, to which I subscribe. I'm an
> engineer with Intel in Sacramento. I noticed your reference to Spectraquest
> and the spc2spc utility. Our group recently purchased Spectraquest, and I'm
> in the process of ramping up on how to use it.
>
> Can you describe the process whereby you extract a spice model of the
> netlist, and designate it to be either W-lines or T-lines? Appreciate
> anything you send my way. Maybe I can help you in some other area. I'm an
> Analog design engineer, and I have been tasked to do the SI work for our
> group.
>
> Best Regards,
>
> Louis A. Sanchez
> Staff Systems Integration Engineer
> (916) 854-5609
> louis.sanchez@xxxxxxxxx
>
> Intel Sacramento
> 9750 Goethe Road
> Sacramento, CA 95827
>
> -----Original Message-----
> From: Peter LaFlamme [mailto:plaflamm@xxxxxxxx]
> Sent: Thursday, October 11, 2001 5:59 AM
> To: chris.h.simon@xxxxxxxxx
> Cc: si-list@xxxxxxxxxxxxx
> Subject: [SI-LIST] Re: Spice simulators - HSPICE vs PSPICE
>
> In response to Chris, I use the Innoveda Viewdraw schematic capture with
> Spicelink to create HSPICE compatible databases. There is some initial
> setup and a learning curve. You need to create your own symbols and link
> models to those symbols, but I think that the PSPICE schematic tool
> requires that also (if the device is not included in the library). I am
> not sure of the price for Viewdraw and the Spicelink add-on but it is
> not super expensive (I believe less than $10K, not sure though). You can
> also look into Hyperlynx which has "spice writer" that allows you to
> extract your net topology into a spice file for simulation.
>
> Also Cadence Spectraquest has spc2spc utility which allows you to
> extract a spice model of your netlist. It allows you to designate
> whether you want W-lines , or T-lines. The sigexplorer has a 2D field
> solver which creates RLGC files for w-lines that you extract. There is
> also a learning curve associated, but it is also pretty easy to use.
>
> I have used all of these tools and think that they are all pretty good.
> I suggest Ashok look into these tools if you are thinking of using
> HSPICE or doing other SI work.
>
> Regards,
> Peter
>
> --
> Peter LaFlamme
>
> Applied Micro Circuits Corp.
> Staff System Applications Engineer
> 200 Minuteman Rd, 3rd Floor
> Andover, MA 01810
>
> 978-247-8470 phone
> 978-623-0055 Fax
>
>
> chris.h.simon@xxxxxxxxx wrote:
> >
> > Ashok,
> >
> > I don't know the current price of HSPICE, but I believe the PSPICE
> > price, if you include the schematic entry package, is more than $4500.
> > One advantage of PSPICE over HSPICE is that PSPICE does have a
> > schematic entry interface, whereas HSPICE requires that you use a text
> > editor to enter the SPICE model lines into an ASCII input file. There
> > are people who ARE using schematic entry to enter HSPICE simulations,
> > but I don't believe that theses schematic editors are commercially
> > supported for that purpose.
> >
> > As far as modelling multi-GHz signals, the transmission line model
> > built into PSPICE is completely inadequate. Or at least it were a
> > couple of years ago when I did a comparison of model vs. measurement.
> > The w-element model in HSPICE is as accurate as anything I've come
> > across in a commercial tool. (That's not to say that it's perfect.)
> > The w-element is not supported by PSPICE. Given this I would say that
> > even if the cost of HSPICE is more than the cost of PSPICE, it's worth
> > the money if accuracy in transmission line models is important.
> >
> > As far as simulating integrated circuits, I believe that HSPICE is
> > used very widely and supports BSIM MOSFET models. We use it for all
> > of our bulk CMOS IC simulations.
> >
> > Chris
> >
> >
> > "Ashok Babu K"
> > <k.ashokbabu@gdate To:
> <ariazi@xxxxxxxxxxxxxxx>, <si-list@xxxxxxxxxxxxx>
> > ch.co.in> cc:
> > Sent by: Subject: [SI-LIST] Re:
> Spice simulators
> > si-list-bounce@fre
> > elists.org
> >
> >
> > 10/11/01 01:46 AM
> > Please respond to
> > k.ashokbabu
> >
> >
> >
> > Hi All,
> >
> > I have gone through the discussions on SPICE tool on this list. Main
> > contenders are HSPICE and PSPICE. What about other SPICE tools like
> > ISPICE
> > from Intusoft, or SPICE tools from Ansoft? Also some free SPICE tools
> > like
> > WinSPICE are avalable. What about their limitations? What about price
> > comparisons of PSPICE and HSPICE?
> >
> > Actually we would like to purchase Spice tool for our simulations at
> > OC48
> > and OC192 systems. We considered PSPICE and HSPICE. The PSPICE pricing
> > is
> > given as about 4500USD in their website, and no idea about HSPICE. Can
> > anyone give us an idea about HSPICE pricing and its updates etc? We
> > are
> > awaiting response from Avanti, but we hope the response very fast in
> > this
> > list.
> >
> > At high frequencies, frequency dependent lossy transmission line modes
> > W-Element transmssion line models are used extensively. Is this model
> > supported only in HSPICE? Can PSPICE support this? Does PSPICE has its
> > own
> > model? Some SPICE tools support only RLC models. I feel that they
> > maynot be
> > sufficient at frequencies of 2.5GHz. Can you please shed some light on
> > this?
> >
> > Also we would like to use HSPICE for our ASIC backend services? Will
> > it be
> > feasible?
> >
> > Thanks and Regards,
> > Ashok.
> > ----- Original Message -----
> > From: "abe riazi" <ariazi@xxxxxxxxxxxxxxx>
> > To: <si-list@xxxxxxxxxxxxx>
> > Sent: Tuesday, September 04, 2001 10:44 PM
> > Subject: [SI-LIST] Re: Spice simulators
> >
> > >
> > > Dear Luca:
> > >
> > > Thanks for your good question. =20
> > >
> > > First let me mention that I really like PSpice due to several
> > reasons =
> > > including:
> > >
> > > i. It allows graphical schematic entry of the design.
> > > ii. It is quite easy to learn (and work with).
> > > iii. A free version (however, with a limitation on maximum number of
> > =
> > > components) is available.
> > >
> > >
> > > On the other hand, PSpice has some important drawbacks in comparison
> > to =
> > > HSPICE. For example, IBIS models cannot be directly used by PSpice
> > =
> > > (and it is often necessary to approximate an IBIS models with VPULSE
> > or =
> > > VPWL); whereas, HSPICE allows direct utilization of IBIS models.=20
> > >
> > > HSPICE also offers a wider choice for transmission line modeling,
> > such =
> > > as lossless (T-element), and lossy (W-element); plus a field solver
> > for =
> > > generating RLGC matrice.=20
> > >
> > > Additionally, there are certain SPICE models which can be directly =
> > > simulated only with HSPICE. =20
> > >
> > > There are many HSPICE experts on the si-list and hopefully they will
> > =
> > > explain why HSPICE is their preferred choice for SI simulations.
> > >
> > > Best Regards,
> > >
> > > Abe Riazi
> > > ServerWorks
> > >
> > > -----Original Message-----
> > > From: Luca Giacotto [SMTP:lgiacott@xxxxxxxxx]
> > > Sent: Tuesday, September 04, 2001 12:49 AM
> > > To: si-list@xxxxxxxxxxxxx
> > > Subject: [SI-LIST] Re: Spice simulators
> > >
> > >
> > >
> > > On Sat, 1 Sep 2001 08:46:58 -0700
> > > "Abe Riazi" <ARIAZI@xxxxxxxxxxx> wrote:
> > >
> > > >=20
> > > > PSpice and HSPICE are among popular SPICE simulators.
> > > > Each program possesses important advantages and disadvantages;
> > > > however, HSPICE offers greater capabilities.
> > > >=20
> > > > The Avant! Star-Hspice Manual contains detailed sections on
> > > > transmission line (both lossless and lossy) modeling, and a
> > chapter
> > > > devoted to use of IBIS models.
> > > >=20
> > >
> > > About spice simulators, I always wondered why SI-people is
> > especially
> > > devoted to the HSpice. At least, this is my feeling following the
> > > discussions on this list.
> > >
> > > Why is PSpice less interesting from an SI point of view? Does it
> > support
> > > IBIS models?
> > >
> > > Best Regards,
> > > Luca Giacotto
> > >
> > >
> > >
> > >
> > > ------------------------------------------------------------------
> > > To unsubscribe from si-list:
> > > si-list-request@xxxxxxxxxxxxx with 'unsubscribe' in the Subject
> > field
> > > For help:
> > > si-list-request@xxxxxxxxxxxxx with 'help' in the Subject field
> > >
> > > 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
> > >
> > >
> > >
> >
> > ------------------------------------------------------------------
> > 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 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
> >
> > ------------------------------------------------------------------
> > 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 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
> >
> ------------------------------------------------------------------
> 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 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
>
--
------------------------------------------------------------------
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 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
Other related posts:
- » [SI-LIST] [Re: Re: Spice simulators - HSPICE vs PSPICE]
