Hacker News new | past | comments | ask | show | jobs | submit login
An open-source, free circuit simulator (hackaday.com)
157 points by jerryjerryjerry on Aug 2, 2023 | hide | past | favorite | 37 comments



> Qucs is a GPL circuit simulator. And if you want the GUI option, you might want to try out QucsStudio, which uses Qucs under the hood, and is free to use, but binary-only.

Huh? I used Qucs many moons ago and it definitely was complete with GUI.

https://qucs.sourceforge.net/screenshots.html

Unfortunately Qucs is unmaintained, probably doesn't compile anymore on newer distributions and its binaries were therefore removed from many repositories. Another problem: Qucs was released as GPL while QucsStudio, which uses its sources, has been kept closed by the authors; I smell a GPL violation here.


> I smell a GPL violation here.

Unless they have been specifically granted a licence other than the GPL, which the original maintainer(s) would have the right to do (assuming other contributors, if any, agreed or signed over the relevant rights when contributing).

Edit: in fact both qucs and qucstudio are listed as authored by Michael Margraf which might make that relicensing/cross-licensing process simpler.


Not only that, but Qucs also has a SPICE version which sees a lot of activity, and unlike QucsStudio, is free software: https://ra3xdh.github.io

It's only missing a solid components library like LTSpice has. Hopefully its library will see more contributions as it becomes more popular.


Might depend on whether or not Qucs was compiled in statically. E.g. they could release an OSS wrapper around it that exposes it over ZeroMQ and then talking to it over that protocol, maybe?


GPL ≠ LGPL


They're talking about exposing it over IPC, not through linking.


It's a shame that circuit simulators, especially those derived from SPICE, have tended toward closed-source. It's nice that there are several free-as-in-beer options available, but a good lesson in the consequences of permissive open-source licenses. SPICE predates copyleft licensing, so software that was once open has disappeared into proprietary versions.

LTspice (sponsored by Analog Devices) is closed-source. The recently-released QSpice (sponsored by Qorvo) also closed-source. QucsStudio was taken closed-source by Qucs' original author, and he's not commented on his intentions in this decision.

Big companies typically open-source software when it's not part of their strategic business value, and offers community engagement benefits and good press. LTspice is kind of an advertisement for AD's chips. AD and Qorvo aren't remotely in the software and EDA business, so their decision to keep them closed is even more puzzling, but I guess the instinct of hardware companies is guard everything.


I think the thing is that no one in the industry gives much of a crap if it's open source or not. Most of the EDA users out there are on Windows so there is no reason to ship source for that platform either. The end users only care about the price which is why LTspice is popular.

Also worth noting that the SPICE engines have been modified heavily over the years in the open source side of things. Inside they look nothing like the original open source SPICE distributions.

When I'm, rarely, doing EDA work it's LTspice for simulation, KiCad for capture+PCBs and Excel for any calculation glue. All are chosen for price/performance rather than open idealism. One just happens to be open source.


there is little to no incentive for these companies to simply open source these applications. customers don’t care if the tool is open source or not, they just want a good simulation.

there is strong disincentive to open source these, because any competitive advantage your tool has with simulation is no longer a competitive advantage.


Last statement makes no sense since AD and Qorvo are in the business of selling silicon. These tools are marketing for their product lines.

It's like saying Google and Facebook are at a business (dis)advantage over the merits of Tensorflow and PyTorch, when their actual business is something completely different.


At least the author of QSpice has committed to release the source of all models (including his tweaks for convergence). So you can take the transistors to a different SPICE engine with a simple netlist import.


The touted benefits of QSpsice include better numerical fundamentals, so who knows what happens when you take a tuned model and plug it into a different engine? The way it is described, the improvements to the SPICE engine could be valuable to anyone interested in numerical code and scientific computing, all the more reason why keeping it closed is a shame.


On windows there is LTSpice which is free, has a lot of models, but not open source and ngspice which is open source but it does not have a GUI (although i heard that it can be used with KiCAD).

So i really don't underestand, what would be the advantage of using Qucs ?


> has a lot of models

That being the important part. Good models, ones someone has validated by comparing measured data on real parts with the model, are not cheap to generate. If the models are not close to reality, when you build what you simulated, it won't work right.

LTSpice has models of, unsurprisingly, most of the parts Linear Technologies makes.


(now Analog Devices)


QucsStudio now "competes" with ADS. It includes HB simulation, EM simulation and models for RF components.


As an amateur I always default to CircuitJS, it's a great learning resource

https://www.falstad.com/circuit/circuitjs.html


And I do professionally. I haven't seen a good interface on any of the other products. They're stuck with clunky 90's interfaces. Circuitjs is far ahead.

Circuitjs is superb for exploring electronic concepts. The quality of the simulation itself is subpar, though mostly good enough. The ability to save a circuit to a url is another thing that sets it apart, and is invaluable for communicating concepts.


You should consider using qucs-s https://ra3xdh.github.io/ a fully open-source fork of qucs still currently maintained and improved.


When talking about open source tools, here are some:

KiCad may be used as a schematic entry tool for ngspice, especially for discrete or PCB-based electronics. You may watch several simulation examples (including oscillators) here: https://forum.kicad.info/t/simulation-examples-for-kicad-ees.... KiCad integrates ngspice internally, but may also provide external ngspice with netlists for simulation.

Another GUI to ngspice is XSCHEM, especially useful for IC design work (see https://xschem.sourceforge.io/stefan/index.html). Device models are available by the Open Source PDKs from Google/Skywater, Google/GF, or IHP. A growing community is supporting digital, analog or mixed-signal design flows.

QUCS-S (https://ra3xdh.github.io/) is a GUI for ngspice or XyCE.

Indeed device models have to be added manually to the devices in the circuit schematic, when invoking ngspice via KiCad or QUCS-S, except for some basic devices with integrated models. Models are provided by device makers, distributors or web sites like this one: https://ngspice.sourceforge.io/modelparams.html .


Can it simulate oscillators? Many of the lower end or cheaper systems fall apart when it comes to modelling real world radios. Probably because in reality everything is a bit capacitive/inductive and there is a noise floor at any given frequency.


Yes. That is the key point of QucsStudio. It has support for professional RF design.


Wokwi is free open source and neat with an MIT license https://github.com/wokwi/avr8js https://wokwi.com


From "Show HN: PicoVGA Library – VGA/TV Display on Raspberry Pi Pico" https://news.ycombinator.com/item?id=35120757 ; wokwi/rp2040js

> Wokwi > New Pi Pico + MicroPython project: https://wokwi.com/projects/new/micropython-pi-pico

And from https://news.ycombinator.com/item?id=36408561 :

> "Learn PWM signal using Wokwi Logic Analyzer" https://blog.wokwi.com/explore-pwm-with-logic-analyzer/

> Wokwi > New (Pi Pico w/ Micropython) LED project: https://wokwi.com/projects/300504213470839309

jupyter-micropython-upydevice: https://pypi.org/project/jupyter-micropython-upydevice/


Wokwi is a completely different tool. You would never be able to use it for radio frequency simulation, or much outside of "virtual breadboard" (although I've tried!)

Its strengths are around simulating microcontrollers.


surprising to see nobody's mentioned ngspice, wr-spice, or omedit (the openmodelica editor), all of which are open-source free software capable of simulating circuits

(though ngspice has had some licensing issues in the past i think those are fixed now)

you can get spice models for just about any part on the market and use them in ngspice, wr-spice, or the original berkeley spice, though not falstad's circuit.js or omedit

relevant to this week's tantalizing superconductor news, wr-spice is the only one with a decent josephson junction model


> omedit (the openmodelica editor)

No Apple Silicon version? They recommend installing it in a Linux VM...


presumably you can compile it for apple silicon



[flagged]


You have an inaccurate picture of the HN audience.


I thought the HN audience liked innovative stuff. Not the same old Windows-only proprietary circuit simulator


The HN audience isn’t homogeneous, and they don’t exclusively like “innovative stuff”. They also like learning about simply useful and interesting software. And there’s a significant portion of Windows users here.

Furthermore, submissions like this may inspire someone to create a non-proprietary front-end.


"HN is not one person" and "this might inspire someone" can justify basically anything. It's not really a good argument


The fact is that enough of the HN audience finds the submission interesting that it has been upvoted to the front page. And you are being downvoted. Which means that a significant enough portion of the audience doesn’t share your assessment.

So the rational thing to do would be to adjust your expectations regarding the spectrum of what is popular on HN.

Maybe also consider that probably most HN members each individually only find a fraction of the front-page submissions interesting.


> Why

Maybe to celebrate the page 2nd-level linked in the submitted, "30 Free Circuit Simulators Lightly Reviewed" / "Best free analog circuit simulators", where the third and last substantive paragraph is all about "This will not suit everyone but look"

https://www.designworldonline.com/best-free-analog-circuit-s...


This article seems to be writtten by someone who doesn't use circuit simulators in real life. As someone has mentioned, you need good models, but when your circuit simulator cannot simulate current sources or transistors is useless.


Maybe it's news that hackers and other technical software dev type people would like?




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: