Two years ago, AI could give you circuit design suggestions, but now AI has already begun to actually "do circuit design"
Can AI design circuit boards yet?
161–170 of 208 posts
Re: Can AI design circuit boards yet?
#162I found LLMs circa 4 months ago to be decent at textbook-style electronics, but they choke at anything nontrivial. They approach things like someone that knows the undergrad and grad textbooks, but has absolutely zero experience. No ability to answer questions such as "is this adequate", "do I need a LPF here", "how do I kill resonance without increasing impedance", etc. I'm guessing that this is due to a lack of RL…
That pig can hardly dance...
Re: Can AI design circuit boards yet?
#163[flagged]
Re: Can AI design circuit boards yet?
#164Earlier quoted context omitted.
They do, https://jlcpcb.com/parts
What if I need .. just one?
You can do that just fine. You can order a minimum of five PCBs and choose to have only two assembled. Then you'll get three unpopulated PCBs and two assembled ones. That's a way to get finished PCBs very cheaply.
Be aware that every component that you use that is not in their basic and preferred extended parts library incurs a one-time three-dollar fee. This hurts disproportionately if your entire design is low count and low cost. So if you always go for the latest and greatest ICs as advertised by TI, instead of the Chinese jellybean clone, you'll add up hidden fees quickly. However, when I choose components, I start out on their basic parts page and only go for a non-basic part if it's not there and I really need it.
Re: Can AI design circuit boards yet?
#165> This has worked much better for us than asking a model to draw lines in a GUI. It also means the benchmark can spend less time testing computer use and more time testing electronics. this is key I think - I use it for graphics tasks - it sucks at graphical environments - always use some text based representation, not really surprising I suppose. chatgpt even suggested I'll use the text representation, claude said h…
Everything that can be objectively tested and has already tools to do so in place will work better in an automated process, regardless of the underlying technology.
PCB design with circuit simulation fits that bill perfectly.
Re: Can AI design circuit boards yet?
#166Earlier quoted context omitted.
> Yet, it's all the most likely token and my brain hurts trying to understand how that can be. You and everyone else. That's the great mystery of transformer architectures as applied to language. To be clear though, they're only good at schematic capture , which is very much a textual representation. Most of the data basically boils down to netlists, which are a text based format mapping connections between abstract…
Where LLMs completely break down is the next step, PCB routing. No. Take a look at https://www.eevblog.com/forum/eda/claude-code-for-pcb-design... . Fable did that by working directly on an EAGLE .brd file (well, "directly" by writing a Python program to do it, but still.)
That is indeed impressive, but at least the excerpt given from the layout seems very easy to route, due to high regularity and an ample routing channel.
Even so, there remained some "dozens" of unrouted traces, which are likely to be much more difficult to route, after the easy traces have already occupied the space.
Many decades ago, I have written a PCB routing program, which would have routed the example shown at that link at least as well, while using many orders of magnitude less resources, i.e. while running on a single-core 233 MHz Pentium MMX.
Obviously that program would have had great difficulties to complete a real high-density PCB design, including many irregular parts and analog circuits with special requirements. I doubt that Fable would fare better.
Re: Can AI design circuit boards yet?
#167Earlier quoted context omitted.
I find it's (Sol at least, probably also Fable) great at writing Skidl schematics, evaluating PCB layout work and operating lab equipment while testing board (scope, psu, load, debugger)
For reviewing schematic/boards I recently developed something for internal tests: 1. A frontier model (either Opensource or closed source) gets a load of text to review a Kicad project systematically 2. The frontier model can extremely "ask" a cheap Gemini vision model to give it data from datasheets or specs. Main problem: the quality dramatically hits the shitter done once it falls back to "pdf2latext" due to compl…
> Main problem: the quality dramatically hits the shitter done once it falls back to "pdf2latext" due to complex tables.
Could try screenshotting the PDF and passing that to gemini
Re: Can AI design circuit boards yet?
#168I do believe PCB routing is something that could be performed successfully by A.I. but new designs from scratch? Only if it was in the training material.
Re: Can AI design circuit boards yet?
#169A personal data point: I had Claude Opus 4.8 design a fairly textbook circuit that outputs a monochrome image burned in an EEPROM over standard 640x480 VGA using only 74 series logic and GALs. It designed the circuit and GAL code, and I did the routing, and got it made through JLC for $6. After it came back, there was one error that was not caught, which I could blue-wire, and it works just fine otherwise. I was fair…
Col! Can you share it?
Re: Can AI design circuit boards yet?
#170I have 15+ years of PCB design experience. Mostly hobby stuff but a fair amount of processional work. Kilowatt range brushless motor controllers, basic RF stuff, lots of microcontroller stuff. I had Fable design an LED earring. Rechargeable coin cell, RP2350 cpu, IMU, 45 addressable LEDs. It made two mistakes - missed the through holes on the coin cell holder footprint and made the center pad too small. I was able to…
It is very hard work and with multilayer boards and surface mount components it is not really a hobby for me anymore, maybe with AI it can be again?