开云体育

I am attempting to create a historical timeline of the history
of SPICE as it has grown in function, use and popularity in the
engineering community. This is to be in bullet point format and
I intended to include only those forms of SPICE that were most
ubiquitous in their time, i.e. the various Berkeley SPICEs, then
PSpice, and, of course, LTspice.

What I've got so far I will put at the end of this message. It has
errors and is not complete, especially the part about LTspice. I
am looking for corrections and input as to the major additions and
events regarding LTspice. (What additions over the years seem
especially noteworthy to you?).

I will fill in the historical dates from the Change Log and from
the group message archive. When the history is complete, I will
add it to group files as a PDF and also add it as a new section
over at the LTwiki (so there is no need to copy it just yet).
________________________________________________

THE HISTORY OF SPICE

1969 beginnings of CANCER (Computer Analysis of Nonlinear Circuits, Excluding Radiation)
? CANCER began as a derivative of a program that was the class project of a series of courses taught by Ron Rohrer with the approval and encouragement of Professor Donald O. Pederson
? Larry Nagel wrote the netlist parser and the analysis core and was student group leader
? Lynn Weber developed a noise analysis feature that utilized adjoint network techniques
? Bob Berry wrote the sparse matrix LU decomposition package
? CANCER project's key features:
. o Was the first circuit simulator to utilize sparse matrix techniques
. o Used Newton-Raphson iteration method heuristically modified for bipolar circuits
. o Utilized implicit integration to accommodate widely spread time constants of an IC
. o Integrated DC operating point analysis, small-signal AC analysis and transient analysis
? Project presented by Ron Rohrer at the 1971 ISSCC , but the code was considered partially proprietary and was never publicly released

1971 SPICE 1 (Simulation Program with IC Emphasis) direct outgrowth of CANCER
? Ron Rohrer leaves UC Berkeley and further development of CANCER (renamed SPICE) became Larry Nagel's Masters project with Don Pederson taking over as faculty advisor
? KEY EVENT: Don Pederson insisted that all further work be releasable to the public domain
? SPICE 1 release's key features:
. o Models for bipolar transistors were changed to Gummel-Poon equations
. o JFET and Shichman-Hodges MOSFET devices added (for Dave Hodges' MOSFET design class)
. o Fixed time step and strict Nodal Analysis (true voltage sources and inductors not supported)
. o DC, AC, Transient, Noise, and Sensitivity Analyses in the same program
. o Built-in models for diodes, bipolar transistors, MOSFETs, and JFETs
? Was about 6k lines of FORTRAN at first informal limited public release in late 1971
? Official public release was May 1972 with first formal paper presented by Don Pederson at the 16th Midwest Symposium on Circuit Theory, April 12, 1973
? SPICE 1 becomes industry standard simulation tool running on large mainframe computers

1972 SPICE 2 begins
? First version of SPICE 2 was Larry Nagel's Ph.D. project under Don Pederson
? Modified Nodal Analysis (MNA) added, enabling voltage sources and inductors for the first time
? Ellis Cohen added dynamic memory allocation
? Adjustable time-step control added, greatly speeding most simulations
? MOSFET and bipolar models overhauled and extended
? Was about 8k lines of FORTRAN when first released to the public domain in late 1974
? Larry Nagel departs for Bell Labs and his thesis becomes the SPICE 2 Users Guide

1975 journey to SPICE 2G6 (the pinnacle FORTRAN version)
? Ellis Cohen becomes primary contributor with later help from Andrei Vladimirescu
? First of a series of public revision releases after Nagel's version 2B begin in 1978
? Along the way, sub circuits, poly sources and transmission lines are added
? Version 2G6 ends up implementing three MOSFET models:
. o MOS 1 is a simplistic model described purely by ideal square-law I-V characteristics
. o MOS 2 is an analytical model, MOS 3 is a semi-empirical model and both include second-order effects such as channel length modulation, sub threshold conduction, scattering limited velocity saturation, small-size effects, and charge-controlled capacitances
? 2G6 released to public domain in April 1983 (and is still available today from UC Berkeley)
? Many commercial simulators today are based on SPICE 2G6

1983 SPICE 3 begins
? Tom Quarles begins work, writing first version in RATFOR, a C-like preprocessor for FORTRAN
? Was fully converted to C in 1985 with first early versions released in March of that year
? Added models: MESFET, lossy transmission line and non-ideal switch
? Arbitrary behavioral voltage and current sources added
? Includes polynomial capacitors, inductors and voltage controlled sources
? Allowed the use of alphabetical node labels rather than only numbers
? Features a graphical interface for viewing results
? New version eliminates many convergence problems
? Added noise, distortion and pole-zero analysis, temperature sweeping, Monte Carlo and Fourier analysis
? Not fully compatible with SPICE 2G6
? Was about 135,000 lines of C code at first public release in 1989
? Final version at Berkeley, SPICE 3F5, released to public in 1993
? XSPICE was developed at Georgia Tech as an extension to the SPICE language to allow behavioral modeling of components
. o Drastically improve the speeds of mixed-mode and digital simulations

1984 PSpice (micro Processor SPICE)
? Developed by MicroSim to run on the first IBM PC, initially released in January 1984
? Was the first commercial offspring of Berkeley SPICE to run directly on the PC platform
? Was the first SPICE program to gain wide acceptance in both industry and academia
? KEY EVENT: A zero cost (but node-limited) student version is introduced in 1988 – for the first time, SPICE becomes ubiquitous in the electrical engineering community
? Evolved from Berkeley SPICE 2G, but added many proprietary enhancements
? Probe, a waveform viewer module, was added when PC VGA graphics became available
? Schematics, a graphical front end, was added much later sometime in the early 1990s

1999 LTspice/SwitcherCAD III first released to public
? 1981 Linear Technology Corporation founded
? 1991 DOS SwitcherCAD available (equation based)
? 1996 μPower SwitcherCAD available(simulation based)
? 2008 LTspice IV

Some possible noteworthy events/additions:

Ver 2 Jan03: graphical symbol editor hierarchical schematics
Apr04: Chan inductor, undocumented behavioral inductor revealed

Your suggestions?

I would think HSPICE should be mentioned. (That is, unless you want to
trace only LTspice's ancestry.) In the semiconductor business community, I
believe it is one of the more significant and most used versions of SPICE.
Also one of the most expensive.

Regarding LTspice, I am not entirely certain whether it is considered in
the SPICE family. It's my understanding that it differs greatly from
Berkeley SPICE, and is not derived from either SPICE 2 or SPICE 3. Now
whether that means that so much of the Berkeley code was re-written or
replaced that it's a whole new program right from the start ... or if it
means Mike started from scratch but then borrowed from SPICE, and made it
extremely compatible with SPICE ... I can't say. I have seen it stated
that it is not an evolutionary step from SPICE, the way PSPICE was from
SPICE 2.

Andy

--- In LTspice@..., "analogspiceman" <analogspiceman@...> wrote:

I am attempting to create a historical timeline of the history
of SPICE as it has grown in function, use and popularity in the
engineering community. This is to be in bullet point format and
I intended to include only those forms of SPICE that were most
ubiquitous in their time, i.e. the various Berkeley SPICEs, then
PSpice, and, of course, LTspice.
1999 LTspice/SwitcherCAD III first released to public
? 1981 Linear Technology Corporation founded
? 1991 DOS SwitcherCAD available (equation based)
? 1996 ìPower SwitcherCAD available(simulation based)
? 2008 LTspice IV

Some possible noteworthy events/additions:

Ver 2 Jan03: graphical symbol editor hierarchical schematics
Apr04: Chan inductor, undocumented behavioral inductor revealed

Your suggestions?

analogspiceman,

I assume you saw the recent article in Electronic Design that contained interesting information on how Mike made LTspice so fast. If not, here's the link.


I think a noteworthy and unique feature of LTspice is the VDMOS model. As you know the use of this model avoids the horrible MOSFET subcircuits some vendors provide.

Another interesting and useful feature is the Chan nonlinear inductor model. Actually, it should be called the "Engelhardt model", because Mike has a patent on it, and the patent says there is a flaw in the Chan model on how minor loops can cause trouble. I wonder if that problems exists in the PSPICE implementation of the Chan model.


Rick

That is pretty much my understanding from Mike. He says that he rewrote the core, from scratch, to optimize it away from the batch FORTRAN mindset into the desktop machine environment. I believe that he also added the alternate solver. But, I do believe that the core was Berkeley SPICE before the rewrite. I am pretty sure that he uses the term "rewrite" rather than "redo from scratch".

FWIW, I used SPICE, as a punch card deck running on a CDC6400 at Tektronix in the mid 1970s. That was then one of the biggest and fastest commercially available machines. And turn-around times were on the order of an hour. Though, much of that was in the printout using reams of fanfold paper.

Jim Wagner
Oregon Research Electronics

On Jul 12, 2013, at 10:05 AM, Andy wrote:

I would think HSPICE should be mentioned. (That is, unless you want to
trace only LTspice's ancestry.) In the semiconductor business community, I
believe it is one of the more significant and most used versions of SPICE.
Also one of the most expensive.

Regarding LTspice, I am not entirely certain whether it is considered in
the SPICE family. It's my understanding that it differs greatly from
Berkeley SPICE, and is not derived from either SPICE 2 or SPICE 3. Now
whether that means that so much of the Berkeley code was re-written or
replaced that it's a whole new program right from the start ... or if it
means Mike started from scratch but then borrowed from SPICE, and made it
extremely compatible with SPICE ... I can't say. I have seen it stated
that it is not an evolutionary step from SPICE, the way PSPICE was from
SPICE 2.

Andy

[Non-text portions of this message have been removed]

[Non-text portions of this message have been removed]

Hello analogspiceman,

You will find the history of LTspice in the Slides.ppt
included in the World-Tour's zip-file.



Best regards,
Helmut

--- In LTspice@..., "Helmut" <helmutsennewald@...> wrote:

You will find the history of LTspice in the Slides.ppt
included in the World-Tour's zip-file.

That is where I got what little information that I already have. Do you have
more?

Regarding the LTspice code, yes it is all Mike's, but it is still very much
based on the SPICE methods in general.

On 7/12/2013 11:24 AM, analogspiceman wrote:

I am attempting to create a historical timeline of the history
of SPICE as it has grown in function, use and popularity in the
engineering community. This is to be in bullet point format and
I intended to include only those forms of SPICE that were most
ubiquitous in their time, i.e. the various Berkeley SPICEs, then
PSpice, and, of course, LTspice.

What I've got so far I will put at the end of this message. It has
errors and is not complete, especially the part about LTspice. I
am looking for corrections and input as to the major additions and
events regarding LTspice. (What additions over the years seem
especially noteworthy to you?).

I will fill in the historical dates from the Change Log and from
the group message archive. When the history is complete, I will
add it to group files as a PDF and also add it as a new section
over at the LTwiki (so there is no need to copy it just yet).
________________________________________________

THE HISTORY OF SPICE

1969 beginnings of CANCER (Computer Analysis of Nonlinear Circuits,
Excluding Radiation)
. CANCER began as a derivative of a program that was the class project
of a series of courses taught by Ron Rohrer with the approval and
encouragement of Professor Donald O. Pederson
. Larry Nagel wrote the netlist parser and the analysis core and was
student group leader
. Lynn Weber developed a noise analysis feature that utilized adjoint
network techniques
. Bob Berry wrote the sparse matrix LU decomposition package
. CANCER project's key features:
. o Was the first circuit simulator to utilize sparse matrix techniques
. o Used Newton-Raphson iteration method heuristically modified for
bipolar circuits
. o Utilized implicit integration to accommodate widely spread time
constants of an IC
. o Integrated DC operating point analysis, small-signal AC analysis
and transient analysis
. Project presented by Ron Rohrer at the 1971 ISSCC , but the code was
considered partially proprietary and was never publicly released

1971 SPICE 1 (Simulation Program with IC Emphasis) direct outgrowth of
CANCER
. Ron Rohrer leaves UC Berkeley and further development of CANCER
(renamed SPICE) became Larry Nagel's Masters project with Don Pederson
taking over as faculty advisor
. KEY EVENT: Don Pederson insisted that all further work be releasable
to the public domain
. SPICE 1 release's key features:
. o Models for bipolar transistors were changed to Gummel-Poon equations
. o JFET and Shichman-Hodges MOSFET devices added (for Dave Hodges'
MOSFET design class)
. o Fixed time step and strict Nodal Analysis (true voltage sources
and inductors not supported)
. o DC, AC, Transient, Noise, and Sensitivity Analyses in the same program
. o Built-in models for diodes, bipolar transistors, MOSFETs, and JFETs
. Was about 6k lines of FORTRAN at first informal limited public
release in late 1971
. Official public release was May 1972 with first formal paper
presented by Don Pederson at the 16th Midwest Symposium on Circuit
Theory, April 12, 1973
. SPICE 1 becomes industry standard simulation tool running on large
mainframe computers

1972 SPICE 2 begins
. First version of SPICE 2 was Larry Nagel's Ph.D. project under Don
Pederson
. Modified Nodal Analysis (MNA) added, enabling voltage sources and
inductors for the first time
. Ellis Cohen added dynamic memory allocation
. Adjustable time-step control added, greatly speeding most simulations
. MOSFET and bipolar models overhauled and extended
. Was about 8k lines of FORTRAN when first released to the public
domain in late 1974
. Larry Nagel departs for Bell Labs and his thesis becomes the SPICE 2
Users Guide

1975 journey to SPICE 2G6 (the pinnacle FORTRAN version)
. Ellis Cohen becomes primary contributor with later help from Andrei
Vladimirescu
. First of a series of public revision releases after Nagel's version
2B begin in 1978
. Along the way, sub circuits, poly sources and transmission lines are
added
. Version 2G6 ends up implementing three MOSFET models:
. o MOS 1 is a simplistic model described purely by ideal square-law
I-V characteristics
. o MOS 2 is an analytical model, MOS 3 is a semi-empirical model and
both include second-order effects such as channel length modulation,
sub threshold conduction, scattering limited velocity saturation,
small-size effects, and charge-controlled capacitances
. 2G6 released to public domain in April 1983 (and is still available
today from UC Berkeley)
. Many commercial simulators today are based on SPICE 2G6

1983 SPICE 3 begins
. Tom Quarles begins work, writing first version in RATFOR, a C-like
preprocessor for FORTRAN
. Was fully converted to C in 1985 with first early versions released
in March of that year
. Added models: MESFET, lossy transmission line and non-ideal switch
. Arbitrary behavioral voltage and current sources added
. Includes polynomial capacitors, inductors and voltage controlled sources
. Allowed the use of alphabetical node labels rather than only numbers
. Features a graphical interface for viewing results
. New version eliminates many convergence problems
. Added noise, distortion and pole-zero analysis, temperature
sweeping, Monte Carlo and Fourier analysis
. Not fully compatible with SPICE 2G6
. Was about 135,000 lines of C code at first public release in 1989
. Final version at Berkeley, SPICE 3F5, released to public in 1993
. XSPICE was developed at Georgia Tech as an extension to the SPICE
language to allow behavioral modeling of components
. o Drastically improve the speeds of mixed-mode and digital simulations

1984 PSpice (micro Processor SPICE)
. Developed by MicroSim to run on the first IBM PC, initially released
in January 1984
. Was the first commercial offspring of Berkeley SPICE to run directly
on the PC platform
. Was the first SPICE program to gain wide acceptance in both industry
and academia
. KEY EVENT: A zero cost (but node-limited) student version is
introduced in 1988 -- for the first time, SPICE becomes ubiquitous in
the electrical engineering community
. Evolved from Berkeley SPICE 2G, but added many proprietary enhancements
. Probe, a waveform viewer module, was added when PC VGA graphics
became available
. Schematics, a graphical front end, was added much later sometime in
the early 1990s

1999 LTspice/SwitcherCAD III first released to public
. 1981 Linear Technology Corporation founded
. 1991 DOS SwitcherCAD available (equation based)
. 1996 ?Power SwitcherCAD available(simulation based)
. 2008 LTspice IV

Some possible noteworthy events/additions:

Ver 2 Jan03: graphical symbol editor hierarchical schematics
Apr04: Chan inductor, undocumented behavioral inductor revealed

Your suggestions?

__

During the 1980s time frame when Micosim's PSPICE became popular
Intusoft Spice program was competitive with Microsim and Electronics
Workbench was a low cost Microsim competitor. I believe Intusoft is
worth a mention because during the 1980s and 1990s time frame Intusoft
published a large variety of simulation examples in their Newsletter.
All of the the old Newsletters are still available on Intusoft's Web site.

Howard

And some of those intusoft newsletters still have some of the best model ideas for electromechanical and thermal problems.


Jim Wagner
Oregon Research Electronics

--- In LTspice@..., "analogspiceman" <analogspiceman@...> wrote:

--- In LTspice@..., "Helmut" <helmutsennewald@> wrote:

You will find the history of LTspice in the Slides.ppt
included in the World-Tour's zip-file.

That is where I got what little information that I already have. Do you have
more?

Regarding the LTspice code, yes it is all Mike's, but it is still very much
based on the SPICE methods in general.

Hello analogspiceman,

One of the links is about HSPICE.







Best regards,
Helmut

Thanks for posting the links. All were very interesting.

Howard

Hello Howard,



I like especially this part of the story below.
... And she said well why don't you call Fab 3 over at Intel and
ask them what the temperature is. And so I got on the phone and
called down to Fab 3 over at Intel and I said what temperature
are the ovens at and they said 830 degrees. And I said thank you.

Best regards,
Helmut

In message <krpsf9+e62p@...>, dated Fri, 12 Jul 2013, Helmut <helmutsennewald@...> writes:

And so I got on the phone and called down to Fab 3 over at Intel and I said what temperature are the ovens at and they said 830 degrees.

I guess that's Fahrenheit - Gas mark 23 (not in Germany, which has bigger marks!).
--
OOO - Own Opinions Only. See www.jmwa.demon.co.uk
Why is the stapler always empty just when you want it?

John Woodgate, J M Woodgate and Associates, Rayleigh, Essex UK

Hello John,

The point isn't whether it's Fahrenheit or degree. It's about
stealing knowhow.

Best regards,
Helmut

In message <krpvc7+8b3g@...>, dated Fri, 12 Jul 2013, Helmut <helmutsennewald@...> writes:

The point isn't whether it's Fahrenheit or degree. It's about stealing knowhow.

I know: I heard the story before, in a company seminar on NOT giving away know-how.

Translating 830 degrees to the cooking oven scale was just another of my weak jokes.
--
OOO - Own Opinions Only. See www.jmwa.demon.co.uk
Why is the stapler always empty just when you want it?

John Woodgate, J M Woodgate and Associates, Rayleigh, Essex UK

A couple of programs that were nodal analysis (net list only entry) on mainframe computers that were precursors to SPICE. Namely ICAP (IBM Circuit Analysis Program) and then PCAP (Princeton Circuit Analysis Program).

Regards, (9V1MI, WN8P) Larry