These layout automations they show in the landing page are the first thing they’ve showcased that I’m intrigued enough to want to try for myself.

The thing I wonder about with all of these EDA companies is whether any of them understand that they’re not attacking the biggest problem of circuit design.

Capturing the SCH/PCB is a relatively small part of the process of getting a working design ready to sell. Less than ten percent of the time, I’d guess. The bulk goes in to design documentation, testing, certification, compliance, and DFM/DFA.

A part of me sort of hates to see these companies do their thing, because it’s making the fun part of the job shorter and faster.

I like your perspective on tackling the biggest problem. We can show a lot more of how users have been doing automation for firmware, docs, testing, etc ...

An underrated part of an autorouter is that you can enforce rules for DFM (and SI) automatically, so some of that tougher stuff is addressable with the right approach. Including a lot of compliance.

Certainly not trying to shrink the fun part - tried to make layout more of an optimization tool than a drafting tool. Although "I should not be doing this" is a mentality a lot of EEs bring to layout.

I’d be happy to generate some additional perspectives on your tool if there’s a way I can try it without spending $10k or going back to college for an .edu email address XD

An underrated part of an autorouter is that you can enforce rules for DFM (and SI) automatically, so some of that tougher stuff is addressable with the right approach. Including a lot of compliance.

You must be referring to an autorouter tool I’ve never tried. (Maybe your own?) All the ones I’ve ever seen produce trace fanouts that look like regurgitated spaghetti.

"I should not be doing this" is a mentality a lot of EEs bring to layout.

Probably explains why so many EEs struggle to get their designs to pass radiated emissions compliance tests. Layout is still circuit design: you just don’t know any of the component values you’re dealing with. Gotta learn to make good choices once you understand how board parasitics work.

Unrelated - I’m pretty sure I spoke with you or one of your cofounders about your product in comments when you first launched. I asked you about emissions compliance and I think DDR trace matching and routing in a comment thread. Glad to see you got at least one of those problems licked.

1) the value a good EE design brings to an organization, or

2) the amount of money selling electronics is worth to a company

You have to feed yourselves and your employees, and all the companies who really need this tool are ones that have no problem spending 10k per year to make their engineers more effective. (Thinking of Lockheed Martin here as they’re one of your flagship customers.)

The average Kicad hobbiest wants this, but doesn’t need it. At least, not in the same way someone with a boss, a program manager, a deadline, and four projects to support does.

Turns out saving that day is incredibly hard, because the professional workflows are intense.

Value of the product ends up being non-linear: either you're valuable enough to be in an EE's regular workflow saving time, or you're not. Worth a whole lot once you are, worth next to nothing if you are not.

Yeah, but they're also never going to pay for it (or not anything close to this price) in a personal capacity.

So give it away free like for current students (would I get away with an alumni^ address I wondered) and generate hype, Youtube content, a pipeline of prospective employees who encourage their employer to adopt it; raving at work about how good it is for their hobby projects and would solve all the problems their having, etc.

^that's the genitive singular, lest you think it erroneously plural ;)

If you are new on the scene, give it to .edu for free and charge everyone else $100 for an annual license. If your product is legitimately good, in ten years you will be the entrenched and established one.

Tbh, that's true for lots of things.

Personally, I spend about 2% of a projects time on kicad related stuff, and the rest on software development for the board.

Anything complex enough to require real knowledge and experience I'd rather hire a board and circuit designer.

For sure! PCB layout is one of the best and most enjoyable parts of the job.

My major concern with these products is always the quality of the schematics they produce. It's not enough for a schematic to have all the nets connected correctly - a schematic is a piece of documentation that, when drawn well, should convey a lot more information than simply what pin is connected to which others.

As an outsider, my assumption is that the schematic would become less of a design document and more of a simple visualization.

What am I missing?

For analog stuff though, a schematic is absolutely fundamental to understanding how the circuit works and how to troubleshoot/fix it. Handing an engineer a code description of an analog circuit will basically have them just trying to mentally build the schematic in their head while reading it.

Imagine a model sharing 3D printing website that got rid of all the 3D model representations and just put the model code in their place. It would be borderline useless (despite your 3D printer having no issue still printing the stuff). Similar situation.

Technically, yes, but 99% of EEs would not approve of code in some weird new DSL as documentation for a circuit. While it may contain the same information as a drawn schematic, it is much more inscrutable and introduces mental overhead when you actually need to read it.

There's also a huge difference between a poorly laid out schematic and an excellent one; the latter makes it easier to follow the flow of signals and power through relevant parts of the circuit and allows you to easily pick out the different parts/modules and how they play their part in the circuit as a whole, all with a quick glance at a drawing.

I guess a (poor) software analogy would be that you could surely implement a function as a bunch of GOTOs and/or BRANCH commands, but just making a nested for loop would make the logic much clearer.

The major issue is that there are a lot more stake-holders in a schematic design than just the person who draws it.

Often the person who draws the schematic is not the person who lays out the PCB, so the schematic needs to encode information about how to lay the PCB out. For example, you may have several components in parallel in a filter circuit; the order that they are placed on the schematic can be used to communicate which order they should be placed on the board (but does not affect the electrical correctness of the schematic at all), and this in turn helps to betray the function of these components.

Even if the whole PCB design is being done by one engineer, other engineers need to be able to review it. If a schematic is electrically correct but otherwise a mess, it makes reviewing it a lot more difficult.

Then there will be third-parties who need to read the schematic and understand the design - for hobbyist products this could be the end-users, and in consumer products it might be service engineers. A well drawn schematic makes a PCB easier to understand and therefore easier to debug, repair, and modify.

In the analogy of software vs schematics, schematics are both code and documentation at the same time.

You are right that the schematic becomes visualization/documentation here - it's not necessary to generate, because the only information it technically provides is a netlist which we already have from code.

But it's needed documentation. In practice it's helpful to use as you debug your design code.

But maybe I'm not the target audience here... pricing seems high for anyone who is not using this commercially (and at scale). $1000 per month (or $60 a day, which I don't really understand the utility of).

Not to say we won't have an open free/freemium offering soon. Hobbyists needed the new layout tool we just shipped last Tuesday, so it's much closer now.

- Paid, which supports private designs.

- Free, where every design is open-licensed and discoverable.

There are approximately zero companies which will want their IP open. On the other hand, this makes this available to approximately 100% of the university and hobbyist market.

Don't cripple the free version in any way beyond that.

The one lesson I learned watching CAD and other professional markets for decades: You want your tool used in university classes. Heck, if there weren't conflicts-of-interest, it'd be worth paying universities to adopt your tool.

Hobbyists are less important (although there have been many examples of hobbyist projects turning into major companies).

Definitely agree with a free version though - we want to help those just starting out as well! But they needed the support from the router we just shipped so earlier offerings would have fallen a bit flat. In the cards now though...

- Marketing / visibility: You're a repository of free designs

- Network effects: If people want to reuse designs, they will standardize on your tool

- Training data. Want to know a pin on a given chip is SCK or 3.3V? Or which pins, by default, to connect between a USB chip to program a microcontroller? That's in your library.

- Component library. If I make a footprint, Spice model, and 3d model for some part, you can get it for free for use by paying customers.

- ...

Yes, it can be circumvented. But the truth is 75% of businesses won't break your license anyway. Some will (and there's an obsession with dongles and what not), but those are mostly tiny fly-by-night operations which aren't likely to buy a $12k license anyway.

It is a lot of work, and I'm not advocating for where this should fit on your priority list, or how much is needed in an MVP. Just that you should consider this as perhaps the right long-term model.

It seems like a smart — if long term — strategy, since today’s young hobbyists are likely going to be a significant number of tomorrow’s founders, choosing the software their companies use.

It is admittedly a little different. Hobbyists getting into 3D printing are surely more common than hobbyists designing and ordering PCBs. But the latter is more accessible to low volume customers than it’s ever been.

The same argument is made for student licenses.

My life experience is different: people who love to learn about topics where such expensive software is commonly used very commonly don't get a job in the respective industry, or at least don't get a position where they can choose the software that the company uses. To put it somewhat polemically: people who love to learn all the time are often nerds; these people typically don't end in management positions where they can buy software for the company.

Also, highly smart people often learn about very different things (e.g. software for 3D graphics and PCB design), but there is hardly any industry/departmentment where software from very different areas is used.

In other words: the whole argument "today’s hobbyists/students will in the future choosing the software their companies use" does in my opinion not hold. Better respect that hobbyists/students form a very passionate user base that is typically rather disjoint with the (less passionate) commercial customers.

Students become professionals and DO make choices. I've seen more than one CAD company succeed primarily by courting universities. Indeed, Apple for a long time dominated graphic design mostly due to its education programs.

Within hobbyists / nerds, there are also two general categories:

- Ones who do this on the side

- Ones who do this as part of a job, but in a different industry

For example, a biologist might tinker in writing medical software or designing some kind of lab instrumentation. Those often do turn into commercial products. The PCB is often almost incidental, but if it's in a particular tool, it's very unlikely to ever be ported out.

Of course, that leads to O(1) license, whereas an EE shop will have O(n) licenses with the number of people they employ. But it does lead to sales.

Joel had a nice article on pricing:

https://www.joelonsoftware.com/2004/12/15/camels-and-rubber-...

The basic upside is giving your stuff away to students and hobbyists costs O(nothing), if you can differentiate in a way which doesn't let paid users not pay.

Well, I'm reasonably smart (math PhD, FAANG experience, etc), and I interviewed with you a couple of years ago.

Went through all technical rounds, didn't get the job.

I've spent some time with Meta since then, and made enough money that pay isn't a priority for me now. Especially for a project like this that I both believe in, and see as a great fit for my background.

I'm looking for the next move now, and I still would love to be with JITX.

Pay me less. Let me help you being your vision to reality.

Today JITX is for professional EEs mainly, and we don't have a lot of resources to help people learn electronics from scratch. (Still want to help the students though - this company was inspired by my grad school robotics struggles).

However a code-based approach boosts reusability enough make open-source hardware a whole lot more meaningful. So we selectively grant access to people able to use the tool effectively today who are also going to contribute to open source.

Overall, we're opening up more and more as the tech moves from special purpose to general purpose.

It's ok, I did play with it a bit but the tooling was a bit of a pain to get set up. Not going to be as polished but it's not $1000/mo either.

(It looks like JITX is handling layout and PB is dealing with the netlist side, but it's in the same space)

To be honest, I don't really understand how these tools are differentiated from one another. I personally know the Quilter guys, so if I had to pick I'd use their product, but I don't think the market crowding here is great for anybody.

Our mission is to take the “hard” out of hardware

We have 240k users on the platform…hobbyists, students, small/mid sized teams to Fortune50 companies. We host millions of projects!

It’s a huge ecosystem not just of users but also semiconductor manufacturers, fab houses, etc.

Think GitHub on steroids!

And it’s the only AI first end-to-end design tool in the market…you can go from idea to research to architecture to BOM to schematic to PCB layout and then to manufacturing. Workflows that usually take months in other tools can be done in a few hours to minutes in flux.

It’s still early days though but we are moving insanely fast! So if we don’t met your teams bar today I suggest checking back in next month.

I think it’s good for more players to enter the market…EDA tooling is stuck somewhere in the 80s and we need to fix that.

It’s too expensive and risky to make hardware…and we deserve better!

In the end, the users will win from this crowding, I suppose.

I'm not so sure about that. A while ago I was really excited about Altium introducing CircuitStudio (the first iteration, I think they have a new product with the same name that's completely different). CircuitStudio was pitched as Altium-lite. I paid, I believe, $3000 for it (with some kind of maintenance license?) and absolutely loved it. A year or so later they reached out to me to let me know that they were changing their licensing costs to $900/year or something like that and that they wanted to reward me as an early adopter by giving me a lifetime license.

But... the product didn't, as far as I can tell, ever get enough traction to actually become a long-term-viable product. Eagle, at the time, was I think $1000 for a professional license (way before Autodesk bought Eagle). CircuitStudio was, to me, a dramatically better product than Eagle. But where I've ended up with this... I have a perpetual license for a product that stopped getting updates years ago. No one I work with has software that's compatible with it and they can't just buy a license. It's a total albatross that I keep around essentially so that I can open my old designs.

New designs? They're in KiCAD now. While I really don't love KiCAD, no one's going to pull the rug out from under me. I can share designs with clients & colleagues and they can open them and work with them. 10 years from now I'll be able to, worst case scenario, run Ubuntu 20.04 in a VM with KiCAD 7 and open the designs (that's assuming KiCAD 37 in 2034 doesn't still have backwards compatibility support :)).

Cloud-based ECAD scares me even more. A year or two ago I needed to debug an issue with a super old design for a client. The schematic and board layouts were done in Protel 99SE (Altium Designer before it became Altium). I installed it, opened the design no problem, and extracted the info I needed. Will JITX still be around in 25 years? If they go out of business... what happens to all of my designs? Exporting to PDF and Gerbers is an OK step but doesn't help me at all if I need to actually interact with the design.

A crowded market is great for downward pressure on price, but it's really not great for companies to survive long-term in a relatively small niche market.

I took a different path, but we have arrived in the same place.

I have a coupe kicad gripes, but there is no doubt it is getting better all the time…. And for free.

Unfortunately for them, though, the mainstream market, along with all of the incumbents, is where the $$ is that would fund non-mainstream support. If the prime rate was still ~0% you might see a strategy out of JITX wherein they'd first take some spectacular swings for the fences in smaller markets. Those kind of successes would make quite the splash here.

But now they've got to go fight it out for the enterprise seats, and, for a half-dozen reasons, post-haste. If they can manage that successfully, I expect they'll soon also be able to do great things for hobbyists, on hobbyist budgets.

Good automation can dissolve false dichotomies.

Ok, disagree though.

Being able to specify a generic part requirement instead of hunting for a specific part is nice sometimes, but any company that makes more than a couple boards already has this covered with a bom management system.

Adding parts via text is nice and fast, but also glosses over many aspects of the part. Say I use a Diodes Inc buck regulator. It has a valid input voltage range it will accept. It has multiple ways it can be wired depending on the application. Wire for buck, septic, etc. PFM on/auto/off, etc. I don't see control over details like that.

Your About Us page is longer and more extensive than your actual product example page.

I see 4-5 very basic designs that I could bang out in Altium in under a day each. Are you selling to people unable to make PCBs? I look for ways to save time because I wear many hats in my job, only 1 of which is doing a board. However, I would not be able to save any time using this tool, because it would produce an inferior result. Additionally, after only 8 months of paying for this product, someone can already afford a full Altium license.

I want to save time on stuff like breaking out and pin swapping IO on a large FPGA. Handle DDR3 routing for me. These are things that actually take time, because you need to understand the device and read through tons of PDFs. However, I think that might also be the most difficult part to add to your product.

Finally, how does it handle physical constraints like non- square board outlines, mounting hole placement, and 3d STEP integration?

Comments/answers to questions in order:

1. Yes, finding a single resistor is easy. Part solving is more about making parametric circuits possible. (like a buck regulator that can configure itself on SS, output voltage). There you need to calculate values from parameters and then go find a part you can actually buy. Doing that manually is less magical of an experience.

2. Yes - some example code here (have not documented yet, apologies) https://github.com/JITx-Inc/jitx-cookbook/blob/main/regulato...

3. Our customers are professional EEs doing really complex work (e.g. many 30Ghz SERDES), and our website is drastically out of date. Will update design on site soon - this post was some guy submitting our landing page to HN. I'm trying to keep up!

4. Yes, this works already. Was very hard to do on the product side, design code was relatively simple. DDR3 example here - https://youtu.be/bw4KxhV-d8g

5. Physical constraints are imported or generated - works quite well to automatically sync electromechanicals. Mutliple 3D models including STEP can be added to parts, again I need to update the website.

(it is a freestanding tool)

The integrations that the other reply mentioned also threw me. I was excited by the “Kicad integration” checkmark. I don’t like the term “power user” but I have done a lot more with Kicad than most folks I know, and I don’t really want to go thru another EECAD learning curve.

I think the main thing is that there is no screenshot of the entire UI or window of the program on your website. It looks like there's some VSCode and as others mention, other CAD software styles, leading to the impression that this is a plugin that you use alongside other things.

The nature of the tool itself is another reason. It claims compatibility on the pricing page with various tools and says something along the lines of you can use your existing tools to finalize placement and routing, leading the user to think it's a plugin that automates some of it.

I would suggest putting a more prominent link to your FAQ (https://www.jitx.com/blog/faq) along with updating it with more information in it.

Separately to this, I think it would be nice to have a more complex (more parts, some power and mixed signal elements) design on the website that users can easily browse and evaluate (view layout, schematic, animation/timelapse of design process), so that users can more easily visualize how it might work for an actual project. It seems like many of the examples are fairly simple currently but the power supply TPS chip is fine.

If this was a fully featured freestanding CAD tool, you wouldn't need to do any finalizing of placement or routing in real CAD software, the final step would be directly exporting gerbers for manufacturing. Exporting to other CAD software would be only necessary for migrating design away from JITX.

With how prominently the description mentions exporting to other CAD software, makes it sound like JITX lacks some essential parts of ECAD software, meaning that most users will want to export and do parts of work in "real" ECAD software.

Can you fully complete most projects at professional level without using other ECAD software? Using other software for some specific tasks like simulation or testing is fine. I am talking only about parts which you expect to be included in most PCB design software.

Keep up the excellent work, seeing the autorouter come to fruition and demoed on the homepage is a massive accomplishment and mind blowing.

Once the library of snippets gets beefy, and people can build boards with their Raspberry Pis and Zeros on them, it'll be a critical mass.

Good luck!

[1] for example it has a built-in field solver for diff-pair impedance, not the approximations that vary from online tool to online tool

But the reality is I unsubscribed from those creators, because I never want to see someone design a board in Altium, I want to see how they build a board using kicad.

From the software side, I really like the license model of the all products pack from JetBrains. And it goes with me wherever I go.

I would say it doesn't compare.

Is there software that would tell me what I need to buy to make this [Arduino QT Py] work with [this specific set of 200 IC LEDs]? Alternatively, how can I check if a specific combination of components would work together? I spent a few hours with Flux and EasyEDA, but didn't get far.

Just wanted to echo some more resources to get you oriented. The second video shows actual examples of how to leverage copilot in your exact scenario. Flux End to End tutorial: https://youtu.be/Nz-XvvlozK4 AI Architecture Exploration: https://youtu.be/iGJOzVf0o7o

The challenge for me is that the parts I'd like to use are not in the Flux library (or I can't find them). If I understand correctly, I would need to add them manually. To make sure I can use your simulator to check if all parts would work together, I would have to write some code, which is, sadly, way above my skill level. (I am fine with writing code, I just don't know enough about this domain.) I do like the Copilot feature, but, based on your video, it seems that I would need to add some properties that I don't much about. Just to be clear, Flux looks great, but it is perhaps for more advanced folks than me.

Regarding my specific project, I would like to use:

1. Adafruit QT Py ESP32 Pico - WiFi Dev Board with STEMMA QT - 8MB Flash 2MB PSRAM (ESP32-PICO-V3-02): https://www.adafruit.com/product/5395

2. BTF-LIGHTING WS2812B IC RGB 5050SMD Pure Gold Individual Addressable LED Strip (one or two strips): https://www.amazon.com/gp/product/B01CDTECSG

I also figured out that I should probably use:

3. Adafruit NeoDriver - I2C to NeoPixel Driver Board - Stemma QT: https://www.adafruit.com/product/5766

I haven't found these components in Flux.

Flux.ai does that…have a look at this tutorial https://www.youtube.com/watch?v=Y56AqPOjp0k&t=366s

Flux Copilot will happily assist you going from an idea to selecting the right components all the way to manufactured board!

Or to save a click: https://github.com/atopile/atopile

I'm a hobbyist PCB designer, and a hobbyist (and generally happy) openSCAD user, with the BOSL2 library, and a hobbyist ESPHome user and I would really love some code alternative to KiCad.

It also sounds like maybe I have too many hobbies.