I hate to say it, but I am not sure if KiCad is ready for this at the moment. KiCad has a long history, starting in 1992 KiCad - Wikipedia, but most of that time it has been slumbering in obscurity. In about the last 8 to 10 years (and especially the last 4) it has become much more capable. I am quite confident to say it has surpassed the needs of most hobbyists, and it’s also suited for small companies. But when it gets to bigger companies… It probably depends on the sort of PCB’s you develop in your company. KiCad is still underdeveloped in some areas, but I do not know much about those, because I do not have a need of those “advanced” functions myself.
On the plus side. KiCad’s development is accelerating. A lot of small functions that boost productivity are getting added, and also big things such as database libraries have been added in KiCad V7 (Introduced earlier this year), so this is still a quite new feature for KiCad.
KiCad is also still a relatively small player with a limited budget. If I assume your company has some 100 seats for EDA design, and your current (or other candidates) are budgeted at EUR 5000 a seat per year, then that is EUR 500k For that kind of money you can put a few full time developers on KiCad to implement exactly the features you want. You can Fork KiCad and do this in house, or you can set up some contract via the KiCad Services Corporation. But either way will get you a lot of attention for your personal needs. This is quite in contrast with what I have heard from other companies. Things like bugs not getting fixed in years, and anti features such as being forced into yearly subscriptions instead of perpetual licenses, what ever more software companies are trending towards.
If you want to evaluate KiCad properly, then regarding it’s potential should be a part of it, and long term costs and needs. It’s quite possible that you are just a few years too early with this evaluation, and this makes it even more difficult for you.
Last weekend, there was KiCon, (The KiCad Conference) in Spain, and all the lead developers of KiCad were there, so I assume they have been busy with preparations before that. Quite a shame you missed it, as you could have talked to them directly.
Yes this is exacly what need to be adressed in a eye to eye conversation with a professional with management present. I’m not in a position where I can make such promises / prospects by myself, I’ve also never been involved in KiCad development.
I really have no idea about the budget available and how it can be spent. For instance it makes a big difference if it is just spent in some products, or for some service. Hiring someone for a internal position (for development of the tool) is much harder even if it would be at the end way more cost effective.
I don’t know, maybe it’s not for us after all, but I really want to find out first, if effectively possible within the time budget given.
I agree with this sentence.
When in 2017 we (very small company - only me designing PCBs) decided to move from Protel 3 (we used since 1997) to KiCad it had some important advantages over Protel and I also missed some important features. KiCad was V4.0.7 those time. Now V7.0.7 has everything I need, but I suppose big company have needs I don’t imagine. Your rating on 7.0.7 can probably be like mine was on 4.0.7.
The invaluable value of KiCad is that if you find a serious bug, it is usually fixed within a few days.
It happened to me once that when we were discussing an error on the forum, when I, an hour later, filled out bug-report, it was immediately closed (the bug has already been fixed) - that is, one of the developers saw our discussion and corrected it before I could report it.
Oh I do believe you . . . users rarely know what they need, they often know what they want though, these two things are usually not the same.
From my perspective the biggest issue for me going forward, coming from a Solidworks PCB (Altium based) system will be the lack of variants in KiCad. I’m sure there is a way to work around that limitation but I don’t know what it is right now and I don’t have time to a pressing need to find out just yet.
to add to what people have already said, @Seth_h and @stambaughw (AFAIK the people running KiPro) are probably at the moment quite busy with the KiCAD Conference this year, there is even a thread were you can post some questions if some come to mind, they will be asked at the conference “later” (sorry, not sure at what time is the QA of the conference).
EDIT: The thread for the QA of “this afternoon” (11.09.2023)
This is indeed a “must have” requirement at the moment. I’ll check how capable and future-proof external plugins (KiVar or similar) are. Maybe the evaluation will be over much quicker than I expected
I have been working professionally with KiCad since 2005 for SMEs.
Some of them can be called big companies, but the team running pcb design had a few members, so in fact they worked as a SME.
That said, KiCad is excellent to design a pcb from a schematic. Everything needed to design a pcb can be done with it as well as with other packages.
I do not know how things are made with other EDAs since I do not use them.
In the early years we had to use a lot of workarounds to get what we wanted. In the last years the amount of new functions has increased very fast so routing and finishing a design is painless.
Do you need database based libraries (I presume you do)? Maybe this part is not mature enough, though its development has improved in the last months.
Do you need electrical simulation? Kicad has a built in spice tool.
Do you need thermal simulation? Kicad files can be exported to use external tools.
Do you need integration with mechanical tools? It can be done with step files.
In a nutshell, any process to design a working can be achieved with Kicad though some of them require some effort.
Unfortunately I cannot compare how much effort related to other EDA packages.
If it was my choice I would not hesitate to choose Kicad from the quality point of view. Price is secondary. In a big company the price should not be a drawback.
Feel free to contact me when you have more specific questions.
I can confirm that Seth and Wayne at KSC are the right folks to address this topic, and they are both traveling today so may not be back to you until tomorrow at the earliest.
I do not think it is years too early for any size company to evaluate KiCad and consider it alongside other options.
Personally, I don’t see KiCAD as enterprise-ready. The bigger players have complete solutions and can integrate with other company systems (eg, procurement, production, planning…)
But I think your focus should be somewhere else: preserving the value of your company’s future work as well as the existing.
By this I mean custom libraries, design files, BOM databases and, and, and. This is tens of thousands of hours of work that no one can replace.
So the strategic criterium is: are your file formats open?
What happens if you go broke tomorrow? How do we preserve our work? A closed file format would kill us.
How can I transform your file format to another?
Etc., etc.,
(All Kicad files are text/ASCII based btw.)
These are killer questions for an EDA sales guy, and you’ll see two responses; either an open “yes, all file formats are open and documented to you, the customer.” or “Umm, eh, uh, well, we have conversion tools to other formats, but they cost…”
Take the decision further from there. It will narrow the field quickly.
I would argue that’s still a funny subject. Even 10 years ago, nobody really had insane integration with everything. It just became a sales pitch because they had nothing else to sell you after they implemented most of the CAD functionality. Companies survived just fine before and can survive just fine without it.
I am no expert on EDA, but my big question is multi-page schematics that are not hierarchal. In 2014 I was not yet aware of KiCad and I made a home project using ExpresssPCB. As primitive as that software is, it handles multipage easily. I THINK that project would have been difficult without multi-page capability. The last that I know, KiCad does not do that. Am I missing something?
So I am not sure entirely what you are talking about?
To throw in a dime. I personally think that Kicad… it crashes too often as is now. Sure that may be related to plugins, it may be related to using Dropbox. Whatevers causes it, it is really annoying. I cannot even use the window manager to swap between projects. I have windows explorer open at all times as workaround.
When these crashes are gone (which is a matter of time) I would recommend the entire world to use Kicad. Besides the little annoying crashes, the program is really great. I think I make pritty artworks with it
Object orientated approach with which you can use one schematic to make more than one objects with.
The ‘replicate layout’ plugin with which you have to route just one of those objects and replicate that routing to all objects.
The JLCPCB plugin to assist fabricate production files
good 3D viewer.
abiltity to reuse schematics and board files so you don’t have to redesign the same DC-DC converter over and over again.
The custom libraries. I even made special symbols for different values, footprints and vendor codes to speed up my workflow.
So yeah after some intens labouring I can poop out schematics and boards at lightning speed.
The only thing I truly miss, is a better graphical system in both schematic and board (things like drawing lines and such). I like to use Kicad to make connection schematics for people. Only Kicad wasn’t designed for this and is this not great in it. But… with some workaround → workable
Regarding crashes, Kicad may crash when you use a third-party application. The plugin developer is responsible for the plugins… and Kikad cannot guarantee the support of these plugins…you don’t have separate sheets, but a main sheet and subsheets… this is clearly visible in comparison with Altium
why do you want to change - which annoyances/limiations do you want to overcome?
which other new benefits you hope to get?
Only after answering these questions one can decide if kicad (and the other CAD-tools) would fulfill the requirements.
Some other unsorted thoughts:
I think the size of the company is not the best indicator to decide for a CAD-package, it’s better to base the decision on the complexity and nature of your typical designed pcb (or take your most complex design as worst case).
it also depends on your usual customers: do you design pcb for your own products or do you have excessive data-exchange with other companies? (The automotive-area for example often uses Altium - some see it as the de-facto standard in that area).
personally I would second marek: Kicad is ready for 90% of medium complex boards. There will always be a missing feature - but that’s true also for other tools. I have seen plenty some middle sized companies (>100 employees) in the past which worked successful with eagle - and that level is also provided by kicad.
Last remark and a little warning: It’s also not a good idea to buy and use a oversized tool. If you don’t use that big tool regularly enough (or the users are unwilling to learn all features of that tool) the complexity of that tool slows down the schematic/pcb design instead of increasing the efficiency!
if you have more detailed requirements (or if you could show a typical pcb-board) then we would give more exact answers
I don’ know if i’m allowed to say how old the existing system is. Let’s say it runs in a virtual machine maybe you get an idea.
We want to start over with everything. A big factor is collaborative and version controlled workflows. We want to move from (local) file based storage to a VCS with all the features that you expect. Also automatic creation of manufacturing data is a big issue.
A big question is, since KiCad is clearly not as powerful as the expensive tools, if we nevertheless can get an overall better experience. I guess it’s not about saving money but making a better investment. Using KiCad without professional support will not happen. I’m really curious how much there is possible. Can we request (order) features that we need? Has this been done before?
You are right, the actual size of the company is not the correct indicator since we are a small team within the company. We only design own products that are expected to have very long lifetimes (30+ years of continous operation) and high reliability. Having the possibility to open and debug such old projects is also a requirement. So at least the basic functionality should be available as lifetime licenses. Thats the reason why we haven’t chosen altium yet since the collaborative features are rent only, even if you host an on premise server.
KiCad,on the one side, has a definitive advatange since it is open source and we therefore we have this aspect in our own hands. On the other side, a lot of the high level functionality seems to come from external plugins and there is a big risk that they will stop working or get superseded by an internal feature.
@ML9104
I do understand the need of open file formats. However there are other approaches that help in a similar way. Altium, for instance, lets you open files saved by a recent version with program versions that are way older. You’ll get a detailed report of missing features, but you can work with them. It helps ensure editability in the future. We do not make customer projects, so there is no need for data exchange.
@pedro
Is it ok if I reach out to you with a private message in the next few days?
Sorry for the off-topic
This is what I do too: my designs are divided in blocks, each block in its sheet even if there is not any hierarchy and all the sheets are at the same leve but the top level sheet. For example, SMPS has individual sheets for Input, pwm driver, transformer, output and feedback. Usually I place the feedback optocouplers on the top level sheet just for clarity.
KiPro brings additional developer resources to the KiCad project. Our time is driven by the features most requested by our active subscriber base. A KiPro subscription adds your needs and priorities to our list.
We don’t direct the KiCad development. We make it happen faster.
All KiCad users get lots of new features for free each year because of all the wonderful volunteers working on KiCad, but when you pay for support, you also get priority development. And I guess that if you want special features implemented, you can negotiate of the price to get it a high priority. But (very likely) your feature has to fit within KiCad’s philosophy, but you really have to talk to KiPro for such things.
I have also seen several bug reports on gitlab go confidential, because companies uploaded a proprietary design t help with the bug fixing. After that happens, the whole topic disappears for regular folks such as me.
From what I read, they’ll also implement special features just for you if these don’t fit into the main KiCad application, or at least they’ve done so historically, and then send you a private plug-in or a patched version. But this might be more expensive than just implementing it as a mainline feature.
