Rendered at 06:41:34 GMT+0000 (Coordinated Universal Time) with Cloudflare Workers.
alangibson 20 hours ago [-]
Token cost is conspicuously missing here.
I've had descent results getting Astra to do PCB layout in Kicad. I'm using the kicad-happy skills. I forced it to set up design rules first then use autorouter plugin as much as possible.
I'm about to start assembling the board so wish me luck.
andyfilms1 1 days ago [-]
Maybe I'm missing it, but I don't see how long it took to create these, or how much they cost (in tokens.)
PCB layout is a weird problem. As the article says, you can get halfway through before realizing you need to re-do everything, and there's no way to predict that from the outset. It's an inherently self-bottlenecked process.
There is also a lot of judgement that goes into it, that I'm not sure could ever be fully covered with AI. You can never hit 100% perfection on every metric (crosstalk, parasitics, DFM, BOM optimization, trace length matching, etc.) so determining where and by how much these things can fall short of optimal is very much a part of the process, and seems like it would be an overfitting nightmare for an AI.
There are also simply decisions that have no empirical right or wrong answer, but are still critical to the system functioning. What frequency do you want to run your data busses at? How much power can it draw? How long does it need to last in standby? How hot can it get?
1 days ago [-]
jakescustomshop 1 days ago [-]
There is a comparison plot that had model Electronics benchmark scores vs $ cost. All tested models were between $2000 and $5000.
I’ve been messing with GPT 6.1 sol PCB routing, but it always comes out really ugly and inefficient. It mostly just runs an autorouter.
Sol IS really good at part selection given loose parameters and JLCPCB’s inventory.
hard_times 1 days ago [-]
We need a good modeling language for general purpose electronics, some sort of UML for hardware.
skyberrys 12 hours ago [-]
When I clicked through the links it turns out Anthropics collaborator here, Diode Computers, is building a DSL for hardware. https://github.com/diodeinc/pcb/blob/main/docs/pages/spec.md... it might align with the what you are asking for. This is first time I heard of it.
andyfilms1 8 hours ago [-]
This...just looks like a file format?
andyfilms1 1 days ago [-]
I mean, that's what the schematic is for. But a logical diagram doesn't tell you the correct trace width and spacing for a particular impedance for a particular substrate, or the experience and context to know if that even matters for this project or not.
I've had descent results getting Astra to do PCB layout in Kicad. I'm using the kicad-happy skills. I forced it to set up design rules first then use autorouter plugin as much as possible.
I'm about to start assembling the board so wish me luck.
PCB layout is a weird problem. As the article says, you can get halfway through before realizing you need to re-do everything, and there's no way to predict that from the outset. It's an inherently self-bottlenecked process.
There is also a lot of judgement that goes into it, that I'm not sure could ever be fully covered with AI. You can never hit 100% perfection on every metric (crosstalk, parasitics, DFM, BOM optimization, trace length matching, etc.) so determining where and by how much these things can fall short of optimal is very much a part of the process, and seems like it would be an overfitting nightmare for an AI.
There are also simply decisions that have no empirical right or wrong answer, but are still critical to the system functioning. What frequency do you want to run your data busses at? How much power can it draw? How long does it need to last in standby? How hot can it get?
I’ve been messing with GPT 6.1 sol PCB routing, but it always comes out really ugly and inefficient. It mostly just runs an autorouter.
Sol IS really good at part selection given loose parameters and JLCPCB’s inventory.