Re: Re[2]: Model availability for DOS Hspice..

Hello Arpad and Stephen,
The availability of IBIS models greatly depends on the number of times your device
vendor was requested to supply I/O models, and IBIS models in particular.
Most of the vendors supply those models as a service to their customers and would like
to put as little effort as possible into it. It is much easier for them to dig the
Spice deck for a device's I/O buffer from an engineer's drawer than to assign
an engineer to generate the IBIS model.
The IBIS concept is fairly new to many IC vendors and too often I see IBIS models
that are incorrect or simply don't have sufficient data in them.
If the vendor is knowledgeable about IBIS in general and understand what you are asking
for, then I ususally request (and also get) the model in IBIS format. Otherwise, I
prefer to get the Spice deck and generate the model (IBIS or vendor specific) by
myself. However, as you already know, thing are changing very rapidly in our industry
and expect more and more vendors to supply correct IBIS models as times goes by.

David Fogel

>
>
> Text item:
>
> Many times that is the only thing we can get because the vendor doesn't support
> or make IBIS models yet. I wish they would make IBIS models, though. I would
> not have to convert them then.
>
> Arpad
>
> =============================================================================
>
> Hello Don, Arpad:
>
> Im curious, what needs/requirements drive you to use SPICE models?
> Can the vendor supply you with IBIS models instead? (Not that I'm
> biased towards
> IBIS or anything <grin>.)
>
> Regards,
> Stephen Peters
> Intel Corp (and member of the IBIS open forum)
>
>
>
> ----------------------
>
> Text item:
>
> Don,
>
> Most of the models I have ran across were written with level 3, 13
> or 28 MOSFET
> models. The bigger issue in my mind is whether the vendors will give you
> (usable) transistor level models at all...
>
> Just now, I am working with several vendors (3) on some models, and
> so far, none
> of them are running. I found syntax errors, which I don't know how
> to correct
> (not because I don't know HSPICE, but because I don't know what is
> in the chip);
> I received netlists without process files, and similar issues which
> caused lots
> of phone calls, waiting, and headaches... I have to admit, one of them came
> with level 37, but that was also written in a company proprietary SPICE
> language, which I could not get to give me correct results on HSPICE yet, not
> even on the worksation version.
>
> Some of your questions could be best answered by Meta Software, I
> would suggest
> to give them a call.
>
> Good luck...
>
> Arpad Muranyi
> Intel Corporation
> ==============================================================================
> Hello!
>
> Perhaps you folks could give me some information.
>
> I am reading from the HSPICE User's Manual 1995 Volume 2 "Elements and
> Models", page 6-2. It says that the PC (DOS-based) version of Hspice
> has built-in support for MOSFET model levels 1 - 8, 13, and 28.
>
> The Unix version supports MOSFET model levels 1 - 15, 17 - 28, and 30 - 39.
>
> My use for Hspice is simple transmission line analysis using an actual
> spice model instead of a behavioral model. I am using a plethora of
> Signal Integrity (SI) tools to supplement my work.
>
> My question is, can I get the models I need from vendors for the PC version?
>
> I bought the PC-version of Hspice for the following reasons:
>
> (1) Cheaper
>
> The software is $5K and so is a fast Pentium PC with 32M of RAM. Total
> solution cost is ~$10K. The Unix version is ~$21K and a Sparc5 (32M of
> memory) is ~$7.5K. Total solution cost is $28.5K. (I think the PC is
> even faster than the SparcStation).
>
> (2) I forget
>
> I don't want to hold up this e-mail while I try to remember the other
> reason (just another example of why you should write in outline
> (brain-dump) form first and wordsmith it later). :)
>
> The sales literature, form 73302/4.95/1M, mentions very little about
> proprietary model support, and nothing about the PC version supporting
> less than the Unix version.
>
> Do I have to spend more money, (i.e. get the Unix version), to get
> the model support? And if the answer is "yes", should I go ahead and
> take a closer look at some of the other more full-featured signal
> integrity tools?
>
> Any and all input and suggestions will be appreciated.
>
>
> Thank you
> Don Abernathey
>
> Text item: External Message Header
>
> The following mail header is for administrative use
> and may be ignored unless there are problems.
>
> ***IF THERE ARE PROBLEMS SAVE THESE HEADERS***.
>
> Content-Type: text/plain; charset=us-ascii
> Mime-Version: 1.0
> Cc: dla@pyrodactyl.or.pyramid.com
> Subject: Model availability for DOS Hspice..
> To: si-list@silab.Eng.Sun.COM
> X-Mailer: Z-Mail (3.2.0 06sep94)
> Date: Wed, 11 Oct 1995 12:04:15 -0700
> Message-Id: <9510111204.ZM17811@pyrodactyl.or.pyramid.com>
> From: "Don Abernathey" <dla@pyramid.com>
> Received: by pyrodactyl.or.pyramid.com (5.67/Pyramid_Internal_Configuration)
> id AA17813; Wed, 11 Oct 95 19:04:16 GMT
> Received: from pyrodactyl.or.pyramid.com
> by gossip.pyramid.com (5.61/OSx5.1a Pyramid-Internet-Gateway)
> id AA27229; Wed, 11 Oct 95 12:04:19 -0700
> Received: from gossip.pyramid.com by mercury.Sun.COM (Sun.COM)
> id MAA06982; Wed, 11 Oct 1995 12:04:22 -0700
> Received: from mercury.Sun.COM (mercury.EBay.Sun.COM) by Eng.Sun.COM
> (5.x/SMI-5.
> 3)
> id AA08751; Wed, 11 Oct 1995 12:04:23 -0700
> Received: from Eng.Sun.COM (engmail1) by silab.Eng.Sun.COM (4.1/SMI-4.1)
> id AA03453; Wed, 11 Oct 95 12:04:41 PDT
> Received: from silab.Eng.Sun.COM (silab-188.Eng.Sun.COM) by Eng.Sun.COM
> (5.x/SMI
> -5.3)
> id AA23045; Wed, 11 Oct 1995 12:08:08 -0700
> Received: from Eng.Sun.COM by mercury.Sun.COM (Sun.COM)
> id MAA08119; Wed, 11 Oct 1995 12:08:26 -0700
> Received: from mercury.Sun.COM by aurora.intel.com (5.65/10.0i); Wed,
> 11 Oct 95
> 12:25:51 -0700
> Received: from aurora.intel.com by fmmail.fm.intel.com with smtp
> (Smail3.1.28.1 #2) id m0t36lM-000rJpC; Wed, 11 Oct 95 12:24 PDT
>
> Text item: External Message Header
>
> The following mail header is for administrative use
> and may be ignored unless there are problems.
>
> ***IF THERE ARE PROBLEMS SAVE THESE HEADERS***.
>
> From: Stephen Peters <speters@ichips.intel.com>
> Date: Wed, 11 Oct 1995 15:52:35 -0700
> Subject: Re: Model availability for DOS Hspice..
> Cc: Arpad_Muranyi@ccm.fm.intel.com
> To: dla@pyramid.com, si-list@silab.Eng.Sun.COM, ibis@vhdl.org
> Message-Id: <9510112252.AA14562@xtg801.intel.com>
> Received: from localhost by xtg801.intel.com (4.1/10.0i); Wed, 11 Oct 95 15:52:3
> 6 PDT
> Received: from xtg801 by ichips.intel.com (5.64+/10.0i); Wed, 11 Oct 95 15:52:37
> -0700
> Received: from ichips.intel.com by hermes.intel.com (5.65/10.0i); Wed, 11 Oct 95
> 15:52:47 -0700
> Received: from hermes.intel.com by fmmail.fm.intel.com with smtp
> (Smail3.1.28.1 #2) id m0t3A14-000rHfC; Wed, 11 Oct 95 15:52 PDT
>
Received on Wed Oct 11 20:49:27 1995