--
| James E.Thompson | mens |
| Analog Innovations | et |
| Analog/Mixed-Signal ASIC's and Discrete Systems | manus |
| STV, Queen Creek, AZ 85142 Skype: skypeanalog | |
| Voice:(480)460-2350 Fax: Available upon request | Brass Rat |
| E-mail Icon at http://www.analog-innovations.com | 1962 |
I'm looking for work... see my website.
Thinking outside the box...producing elegant & economic solutions.
--
| James E.Thompson | mens |
| Analog Innovations | et |
| Analog/Mixed-Signal ASIC's and Discrete Systems | manus |
| STV, Queen Creek, AZ 85142 Skype: skypeanalog | |
| Voice:(480)460-2350 Fax: Available upon request | Brass Rat |
| E-mail Icon at http://www.analog-innovations.com | 1962 |
I'm looking for work... see my website.
Thinking outside the box...producing elegant & economic solutions.
It doesn't choke, it just runs slower when you add time constants that are faster than most of the rest of the circuit. A simulator with adaptive time steps will run faster when things are changing slowly, and slower when some things move fast. That makes a lot of sense.
Does pspice use adaptive time stepping, or do you always set it yourself?
My leakage inductance drives two schottky diodes in series, so there's fast nonlinear ringing going on. I only care about the leakage L for a small part of the sim, so it's reasonable to disappear it when it doesn't matter. Actually, it doesn't matter.
formatting link
LT spice does go for speed at the cost of accuracy. A simple ringing LC will be a few per cent off frequency with the default settings. It's trivial to tweak the max time step when things like that matter. Every simulator that I've used or written benefits from intelligent choice of time step.
The thing about LT Spice is that it's free, it's supported, it's fun to use, and it makes me money. I don't understand your hostility.
--
John Larkin Highland Technology, Inc
lunatic fringe electronics
Any Berkeley-Spice-compliant simulator runs wide-open until it sees a voltage change greater than vntol or a current change that exceeds abstol, then backs up, finds the correct data-point, then flies forward again. You can force a step ceiling (max time-step), or leave it blank, and risk the simulator missing an important transition.
I would suggest you set Solver=Alternate if you're interested in accuracy.
It's fine if you know how to use it properly... by setting tolerance variables correctly... which means you lose the speed advantage.
My "hostility" derives from having to waste my time proving to clients that LTspice will lie to you unless you are very, very careful.
(And clients who have no clue how their own product actually works. I recently was put in the position by a client to fudge my Spice modeling to fit their misconception... so I forced a disclaimer in the model file :-)
Plus the LTspice GUI is the crappiest of all crappy GUI's... even worse than Cadence Virtuoso ;-)
I keep inviting anyone to come by and see an easy-to-use simulator (PSpice) in action... no one has taken me up on the offer. ...Jim Thompson
--
| James E.Thompson | mens |
| Analog Innovations | et |
| Analog/Mixed-Signal ASIC's and Discrete Systems | manus |
| STV, Queen Creek, AZ 85142 Skype: skypeanalog | |
| Voice:(480)460-2350 Fax: Available upon request | Brass Rat |
| E-mail Icon at http://www.analog-innovations.com | 1962 |
I'm looking for work... see my website.
Thinking outside the box...producing elegant & economic solutions.
If I were building a precision oscillator, which sometimes I do, I'd give up accuracy for speed. Of course, that tends to be a losing game: just try simulating a crystal oscillator in time domain. If a switcher makes some linear approximations of the exact rising-waveform, it doesn't matter.
All you need to do is reduce the time step until it doesn't change anything important. Then compare that to the default. Usually the defaults are good enough.
If a switcher sim if off by 1%, it doesn't matter. A crystal oscillator has to be PPM or PPB accurate, so no time domain sim is practical.
Of course you have to use it properly.
Any simulator can lie. You should always be careful.
But you actually did it? I wouldn't.
Works fine for me. I found it easy to learn, and it sets up and runs things fast. It makes me money.
Me neither.
--
John Larkin Highland Technology, Inc
lunatic fringe electronics
I don't know. I stopped paying 'maintenance' when Cadence began cobbling it. Fortunately I have an 'in perpetuity' license since I was very instrumental in the early days wringing out beta versions... back in MicroSim times. ...Jim Thompson
-- | James E.Thompson | mens | | Analog Innovations | et | | Analog/Mixed-Signal ASIC's and Discrete Systems | manus | | STV, Queen Creek, AZ 85142 Skype: skypeanalog | | | Voice:(480)460-2350 Fax: Available upon request | Brass Rat | | E-mail Icon at
formatting link
| 1962 |
I'm looking for work... see my website.
Thinking outside the box...producing elegant & economic solutions.
Aw! Can't you just thumb a ride ?>:-} ...Jim Thompson
-- | James E.Thompson | mens | | Analog Innovations | et | | Analog/Mixed-Signal ASIC's and Discrete Systems | manus | | STV, Queen Creek, AZ 85142 Skype: skypeanalog | | | Voice:(480)460-2350 Fax: Available upon request | Brass Rat | | E-mail Icon at
formatting link
| 1962 |
I'm looking for work... see my website.
Thinking outside the box...producing elegant & economic solutions.
Is the version included with Orcad any hobbled in any way? I think I've got a license for v16 or something but the license server won't run on anything but XP so it's apparently a security risk.
Cadence chopped away a lot of the netlisting features which I find handy for 'talking' to my layout guy. Fortunately that feature control was all in an .INI file when I bailed, so I was able to keep all the features.
Now that OrCAD Crapture is forced on you I don't know what features have been cut (I still use the original-crispy-flavor MicroSim PSpice Schematics).
Hey! It's actually raining here! First time in months! ...Jim Thompson
--
| James E.Thompson | mens |
| Analog Innovations | et |
| Analog/Mixed-Signal ASIC's and Discrete Systems | manus |
| STV, Queen Creek, AZ 85142 Skype: skypeanalog | |
| Voice:(480)460-2350 Fax: Available upon request | Brass Rat |
| E-mail Icon at http://www.analog-innovations.com | 1962 |
I'm looking for work... see my website.
Thinking outside the box...producing elegant & economic solutions.
That is what VirtualBox is for. Just create a small VDI and load XP. You can link to a folder on Windows to transfer files back and forth. Delete any drivers that are not needed, such as LAN, USB, etc. Copy the VDI to another folder for backup.
The risk of infection is very small. I have been running XP for many years and have never had a problem. However, if one does occur, it cannot go far since you have blocked any access to the host. Just copy the backup VDI over the infected one and continue running as if nothing has happened.
This arrangement is a little bit like Qubes, but is simpler and more secure. Instead of running in a sandbox which can leak, you have a complete functioning operating system with rapid backup in the case of a finger fumble or unwanted software download such as Adobe that also installs McAfee. But it's best to keep the apps at a minimum. You don't need Flash, email, MS Word, and all the other bloatware that clogs windows. Just focus on the minimum that is needed to run your system.
VBox offers fixed and dynamic VDI files. Dynamic is supposed to be slower, but I have both and I cannot tell the difference. The advantage of dynamic VDI is you can keep the actual file size o just exactly what is needed, With fixed VDI, you have to prdict what you may need, the file then is huge and most of it is wasted, and you cannot easily increase the size later if it is needed.
Stay away from VMware. These people will hound you to death to purchase their upgrade version. If you want to deal with them, use a disposable email address for when you want to quit.
MS vm is typical MS crap. Don't waste your time. VBox works, it is free, and easy to learn and set up.
No LAN, no email, no browser, no flash, no pdf, no USB, no unwanted links to the web.
Any web links are behind a separate NAT provided by VBox. Very hard for malware to bypass.
Easy backups by copying the VDI file to a backup folder.
Only the minimum files are needed. My VDI file is 3 GB, and I still have
1.6 GB free disk space. This makes backups fast and easy, so you are more likely to perform them regularly. This helps protect against fumble fingers and unwanted downloads. You can also copy the VDI fle to a USB drive or a separate external drive. This provides protection against HDD failure.
Qubes doesn't do these things. You have one VM, and that is the complete operating system. It is wide open to attack, fumble fingers, and HDD failure. I don't know how you would back the system up.
Qubes has already experienced at least one leakage to the main system. As the documentation states, once malware gets to the main host, it is game over.
If I thought Qubes was safer, I'd use it. But VBox has so many more significant advantages, there is no contest.
Not with the huge attack surface presented by the host OS. The hypervisor helps against attacks from inside the VM, but can't do much about the outside.
They're had to deal with some Xen bugs, but no specifically Qubes Dom0 problems. The ITL folks are super smart and very careful.
As
Dom0 in Qubes is a far harder nut to crack than whatever fully-upholstered OS you're running VirtualBox on.
I have no quarrel with your using whatever system you like, but calling VirtualBox on a gigantic OS "more secure than Qubes" is ridiculous.
Cheers
Phil Hobbs
--
Dr Philip C D Hobbs
Principal Consultant
ElectroOptical Innovations LLC
Optics, Electro-optics, Photonics, Analog Electronics
160 North State Road #203
Briarcliff Manor NY 10510
hobbs at electrooptical dot net
http://electrooptical.net
I was musing last night over dinner about how many different simulators I've used over the years, so, this morning, I decided to list them...
(No particular order)
Berkeley Spice 2G6 thru 3F3 on a VAX780 Sspice Silvaco SmartSpice MultiSim TINA IBIS Development Studio HSpice IsSpice4 TopSpice8 Microcap Cadence (Virtuoso front-end) LTspice IV LTspice XVII PSpice v3 thru v15.7
Then I pondered why all the hoopla over LTspice, and I thought of an automotive analogy...
Suppose you've lived a sheltered life, never owned or leased a car, in fact never ever seen a car...
2CV, FREE no less ;-)
In your wonderful amazement you will swear it's the best car EVER!
Nevermind that you don't know that Cadillac's, Mercedes, Audi's, Porsche's and Infiniti's even exist.
So this LTspice hoopla is all about exposure... LTspice users have no exposure... most have never even used, let alone seen, any other kind of simulator. And wouldn't know how to run one if they tried.
Thus my apologies for intruding myself on this thread.
Enjoy your 2CV >:-} {*}
{*} Watch Joerg jump in here and extol the benefits of a 2CV ;-)
...Jim Thompson
--
| James E.Thompson | mens |
| Analog Innovations | et |
| Analog/Mixed-Signal ASIC's and Discrete Systems | manus |
| STV, Queen Creek, AZ 85142 Skype: skypeanalog | |
| Voice:(480)460-2350 Fax: Available upon request | Brass Rat |
| E-mail Icon at http://www.analog-innovations.com | 1962 |
I'm looking for work... see my website.
Thinking outside the box...producing elegant & economic solutions.
I still have a copy of Microsim, but I never use it for anything. LTspice works fine, for values of "fine" accessible to board-level designs.
I've had pretty good luck with LTspice for discrete designs, but the op amp models are all crap. That bargain-basement 100-ps sampler came out pretty nearly bang-on.
Cheers
Phil "Not the biggest SPICE fan to begin with" Hobbs
--
Dr Philip C D Hobbs
Principal Consultant
ElectroOptical Innovations LLC
Optics, Electro-optics, Photonics, Analog Electronics
160 North State Road #203
Briarcliff Manor NY 10510
hobbs at electrooptical dot net
http://electrooptical.net
--
Dr Philip C D Hobbs
Principal Consultant
ElectroOptical Innovations LLC
Optics, Electro-optics, Photonics, Analog Electronics
160 North State Road #203
Briarcliff Manor NY 10510
hobbs at electrooptical dot net
http://electrooptical.net
I used to write my own simulators, for circuits and control systems, when I had a PDP-11 system. The first sim that I wrote (and the first control system I ever designed, still an undergrad) was for a 32,000 SHP steam turbine on the LASH ships. Worked first time.
My first PC-AT simulator was ECA, which was really nice and very well documented. Manual typed netlists.
I bought EWB/Multisim but it was so bad I sent it back.
--
John Larkin Highland Technology, Inc
picosecond timing precision measurement
jlarkin att highlandtechnology dott com
http://www.highlandtechnology.com
ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.