what can Pspice better than LTspice?

[alivdel]

Hello,
im wor?ing on a project about simulation with Pspice and LTspice. in a part of my project i need to write about differences between this two softwares.
As we know LTspice is free but we schuld buy Pspice too expensive. i want to know what is the advantage of Pspice compare to other one that companies spend a lot of money to buy it while they can use LTspice for free.

Thanks
Ali

[Pjotr]

LTspice == Pspice

Pspice was originally developed at Berkeley University as just plain text based. Thereof came many commercial based implementations with a graphic shell and a schematic entry shell. One of the first did come from Intusoft for the Mac and MicroCap under DOS.

LTspice is just a give away implementation from Linear Technology. The more expensive commercial packages differ mostly in ease of use, advanced graphics and sophisticated convergence algorithms. My favourite is the current version of MicroCap, but that is personal. It all depends of your use. For hobby LTspice is fine but for professional use the paid versions are more productive. So if time is money, although expensive, the paid versions will pay back.

[EI6JA]

Hi,
Is LTspice limited to Linear Technology active components? If it is that is one of your differences.
Regards,
Stephen

[Siwastaja]

It's not limited to Linear parts per se, but it does lack support for some higher-end simulation functions, and therefore, not all SPICE models (downloadable from part manufacturers) work at all. Some require minor tweaking.

Last time I checked, Orcad Pspice was completely broken and totally unusable, being hit by the enterprise bloat virus. The older Pspice was practically identical to LTSpice from the user interface viewpoint; and it's very easy to get started.

[Wobbegong]

LTspice == Pspice

Pspice was originally developed at Berkeley University as just plain text based. Thereof came many commercial based implementations with a graphic shell and a schematic entry shell. One of the first did come from Intusoft for the Mac and MicroCap under DOS.

LTspice is just a give away implementation from Linear Technology. The more expensive commercial packages differ mostly in ease of use, advanced graphics and sophisticated convergence algorithms. My favourite is the current version of MicroCap, but that is personal. It all depends of your use. For hobby LTspice is fine but for professional use the paid versions are more productive. So if time is money, although expensive, the paid versions will pay back.

This is absolutely not the case. LTSpice is a unique implementation of spice with proprietary charge models and solvers invented by Mike Engelhardt at LT. There are some cases where LTSpice is the highest performance circuit simulator out there but as others have pointed out, they don't go out of their way to support 3rd party models so it is not the best choice for simulating non-LT parts.

There was a great Amp Hour interview with Mike a while back:
http://www.theamphour.com/196-an-interview-with-mike-englehardt-spice-simulator-synteresis/

[Siwastaja]

The LTSpice UI is so close to the "old" PSpice that I understand how people make the mistake that they are the same.

Pspice and LTSpice are both trademarks of different companies. SPICE is the more generic name. Maybe Pjotr meant "spice" instead of Pspice.

It seems to me that about 1/3 of IC models work out of box, about 1/3 require some minor tweaking, and 1/3 are practically impossible to make work due to incompatibilities. Almost all downloadable spice models are done and tested on PSpice which is, sadly IMO, kind of industry standard.

LT parts of course all work.

[Pjotr]

The LTSpice UI is so close to the "old" PSpice that I understand how people make the mistake that they are the same.

You're right about that. Pspice was Orcad's implementation of spice I think. And yes, all the vendors have their own implementation.

But LTspice is the best? I can't agree. A few years ago I had to develop a 2KW soft switching H-bridge converter with a wide output range (charging batteries). This thing had some nasties with diode recovery of the build-in diodes of the switching fets. Investigating this was hopeless with LTspice, if it managed to converge. While MicroCap did the job pretty well.

[diyaudio]

Hello,
im wor?ing on a project about simulation with Pspice and LTspice. in a part of my project i need to write about differences between this two softwares.
As we know LTspice is free but we schuld buy Pspice too expensive. i want to know what is the advantage of Pspice compare to other one that companies spend a lot of money to buy it while they can use LTspice for free.

Thanks
Ali

Simulating using abstract models is the most practical way to realize a circuit, LTpsice and other simulation packages (TINA, Top SPICE ect.. they all good for the job) users may be more critical on the UI aspects, example Ltspice has a poor UI, but hey its FREE!, Pspice has some really useful and interesting tools that sits on top of the simulator like the PSpice Optimizer, and Graphing Math Functions and many more interesting stuff to get the job done. However most people (non-engineering thinking) don't use this level of design to realize a circuit. They only looking for equivalent aspects trying to get as close to reality as possible. which is actually useless, as there are way to many dynamics at play typical (Simulation vs Real World debate)



 
As a side note I own some pretty impressive books, I would say some of the best Circuit Simulation literature that exist is written for Pspice 

PSpice for Analog Communications Engineering
PSpice for Circuit Theory and Electronic Devices
PSpice for Digital Communications Engineering
PSpice for Digital Signal Processing
PSpice for Filters and Transmission Lines

These are just some of the good stuff written for Pspice, there are lots more.

[joeqsmith]

Actually PSPICE was a product by MicroSim Corporation.    The first version came on floppies that were laser burned for copy protection.  You had to have the master disk installed to run the program.  They had supplied it as a packed EXE and the debugger would not handle it.  To add anything you had to use the Microsoft Fortran to rebuild it.  So someone could rebuild it as an unpacked EXE then debug it to place a jump over the laser coding without too much trouble.   Next they started using printer port dongles to protect it.  However, these early dongles were not much more than a counter.   They made better dongles later when Orcad took it on but really, not more than a counter with some decoding.   That's the history of PSPICE protection from what I remember...

Earlier we had SPICE running on a VAX 11-780? and later on a Daisy system.    Much later Saber came along.  What a change that was.   

The best SPICE story I have is a company I was working for at the time was looking for a new tool.  A group of us went to one of the major suppliers and they were showing us everything they could do.  It looked pretty impressive.   I asked them about PSPICE and how they would stack up.   Now PSPICE was an ORCAD product by then and really, we are talking about a tool costing thousands vs many 10s of thousands.  They had a chuckle over this.   Later after they had demo'ed their tool, they asked if we had any questions or wanted to try anything.   I reach into my pocket and pull out a small schematic and asked if they could simulate it for me.  There were only a few parts and they said it would be simple enough.   So I told them the way I wanted it ran and they kicked off the run.   Back then, PCs were still fairly poor so most of the real software ran on workstations.  It was running along just fine for a few minutes, then Core Dump!!!   They were frantically trying to recover but the system was dead.   They ask me if I had simulated that circuit before and I explained I had..... with PSPICE!!   That ended the presentation.

What was the circuit that brought down their system?  A switching supply running in transient mode.   

I took some pictures of the early manuals....

[diyaudio]

Actually PSPICE was a product by MicroSim Corporation.    The first version came on floppies that were laser burned for copy protection.  You had to have the master disk installed to run the program.  They had supplied it as a packed EXE and the debugger would not handle it.  To add anything you had to use the Microsoft Fortran to rebuild it.  So someone could rebuild it as an unpacked EXE then debug it to place a jump over the laser coding without too much trouble.   Next they started using printer port dongles to protect it.  However, these early dongles were not much more than a counter.   They made better dongles later when Orcad took it on but really, not more than a counter with some decoding.   That's the history of PSPICE protection from what I remember...

Earlier we had SPICE running on a VAX 11-780? and later on a Daisy system.    Much later Saber came along.  What a change that was.   

The best SPICE story I have is a company I was working for at the time was looking for a new tool.  A group of us went to one of the major suppliers and they were showing us everything they could do.  It looked pretty impressive.   I asked them about PSPICE and how they would stack up.   Now PSPICE was an ORCAD product by then and really, we are talking about a tool costing thousands vs many 10s of thousands.  They had a chuckle over this.   Later after they had demo'ed their tool, they asked if we had any questions or wanted to try anything.   I reach into my pocket and pull out a small schematic and asked if they could simulate it for me.  There were only a few parts and they said it would be simple enough.   So I told them the way I wanted it ran and they kicked off the run.   Back then, PCs were still fairly poor so most of the real software ran on workstations.  It was running along just fine for a few minutes, then Core Dump!!!   They were frantically trying to recover but the system was dead.   They ask me if I had simulated that circuit before and I explained I had..... with PSPICE!!   That ended the presentation.

What was the circuit that brought down their system?  A switching supply running in transient mode.   

I took some pictures of the early manuals....

Good story.

[amspire]

Actually PSPICE was a product by MicroSim Corporation.    The first version came on floppies that were laser burned for copy protection.  You had to have the master disk installed to run the program.

I probably have that version of PSpice somewhere - definitely have the manual. I remember scratching the surface of a 5 1/4" floppy disk to exactly match the location of the burnt sector on the master disk. Took several goes to get the position exactly right. We had a license, but I didn't like the idea of relying on the single master disk to always work.

[Pjotr]

Yes, they asked really huge amounts of money those days. In the early '90-ties, I asked for a licence for a SUN workstation for the university group I worked those days The representative came along and gave a demo of a whole day. When it came to price, I discussed that as a university we were tight on budged and 20K was the max I had to spend on it. Nevertheless they offered me a quotation of 200K + a yearly 20K for single node locked licence. No way.

[tggzzz]

Actually PSPICE was a product by MicroSim Corporation.    The first version came on floppies that were laser burned for copy protection.  You had to have the master disk installed to run the program.

I probably have that version of PSpice somewhere - definitely have the manual. I remember scratching the surface of a 5 1/4" floppy disk to exactly match the location of the burnt sector on the master disk. Took several goes to get the position exactly right. We had a license, but I didn't like the idea of relying on the single master disk to always work.

Somewhere I have copies of Pspice50, 54 and 60. I don't recall them needing floppies; they were probably eval versions.

[amspire]

Somewhere I have copies of Pspice50, 54 and 60. I don't recall them needing floppies; they were probably eval versions.

The 5 1/4" floppy version was probably one of the first versions from the early 80s. They hopefully would have at least gone to 3 1/2" floppies by Version 5.0.

PSpice did have educational versions - I think they were free - that only allowed a small number of parts.

[Pjotr]

PSpice did have educational versions - I think they were free - that only allowed a small number of parts.

Maybe, they didn't offer me that. Instead they phoned me back regularly a couple of times. It didn't help. End  of the '90-ties MicroCap became grown up and was way more affordable under WinNT.

[amspire]

The amazing thing about Spice is it was written in the 60's when there wasn't much computing power around. Fell in love with it immediately at University, but my first Spice assignment was a bit of a disaster.

We had this really bad lecturer who would lecture by facing the blackboard writing in very small letters while talking very fast and very quitely. He set an assignment to design a 3GHz 9th order ellyptical filter that used discrete capacitors and inductors. We had to verify our design using Spice via punched cards. To meet his very demanding specifications for this assignment, it turned out that all the inductors and capacitors had to have a 0.00000001% tolerance. Luckily, we didn't have to assemble and test the filters.

I did actually get a copy of the Berkley Spice code and compile it for myself. It was originally in Fortran, but this was Spice source rewritten in C. No graphics though - just numbers and some crude graphs done with letters and punctuation symbols.

[Pjotr]

Those were the days. When I did my study we had spice running on a IBM360. I remember we didn't even had automatic card punchers. We had to punch the characters one by one by hand with a kind of "Dymo" like punch apparatus    If you made a mistake you had to do the whole card again by hand from the beginning on. Taping off the wrong holes was not allowed. On the other hand it learned you careful programming, sloppy doing was punished immediately

[amspire]

And every now and then someone would drop a huge stack of cards. They did have sorting machines to correct the card order, but not for undergrad students.

[tggzzz]

And every now and then someone would drop a huge stack of cards. They did have sorting machines to correct the card order, but not for undergrad students.

The traditional technique was to have a large X drawn with a felt pen on the edge of the deck. No idea how that coped with editing, though!

But I do know how editing was done on a 5 channel (later 8 channel) paper tape.