Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

ChatGPT for Hardware Design

docs.flux.ai

Post reply on HN
Print view View original

31–40 of 165 posts

Re: ChatGPT for Hardware Design

#31

Post by rock_hard »

hammyhavoc wrote: post #12
rock_hard wrote: post #10

Earlier quoted context omitted.

LLMs do indeed hallucinate. What’s been working well for us is to give it access to factual data sources such as calculators, component data sheets, etc. It does really well from our tests in the past few weeks, so well that we decided to invite everyone to explore the boundaries with a community preview Just give it a shot and report back

Can you elaborate on "really well"? Because a few weeks of testing and "really well" versus the claim of reducing costly mistakes on the marketing blurb whilst simultaneously acknowledging hallucination doesn't inspire confidence. I'd be interested to know a lot more about testing methodology too. A hallucination could be a much more expensive problem than someone simply making a basic mistake that someone might othe…

To give you an example of the typical response quality we encountered:

@copilot can you design a low pass filter for me with 3db point of 300khz?

copilot: Sure, I can help you design a low-pass filter with a 3dB point of 300 kHz. A simple passive RC (resistor-capacitor) low-pass filter can be used for this purpose. To calculate the values of the resistor (R) and capacitor (C), you can use the following formula:

f_c = 1 / (2 pi R * C)

Where f_c is the 3dB cutoff frequency (300 kHz), R is the resistor value, and C is the capacitor value. Let's choose a standard resistor value, for example, 1 kΩ. Then, you can calculate the capacitor value as follows:

C = 1 / (2 pi R f_c) = 1 / (2 pi 1000 300000) ≈ 530.5 pF

Since 530.5 pF is not a standard capacitor value, you can use the closest standard value, which is 560 pF. With these values, the actual 3dB cutoff frequency will be approximately:

f_c = 1 / (2 pi 1000 * 560e-12) ≈ 283.5 kHz

The low-pass filter can be assembled by connecting the 1 kΩ resistor in series with the input signal, followed by the 560 pF capacitor connected to the ground. The filtered output can be taken across the capacitor. This design will give you a first-order RC low-pass filter with an actual 3dB point of approximately 283.5 kHz.

Re: ChatGPT for Hardware Design

#32

Post by arcticbull »

greg wrote: post #4

Engineer on the team here. Some of the interesting challenges of doing ChatGPT for circuit design were: 1) How to communicate the circuit connections to ChatGPT in the text prompt. We ended up just using the standard netlist format. Apparently it knows about netlists from its training data. It also knows about datasheets. 2) How to get the needed accuracy and precision for questions like "calculate the resistance nee…

Out of curiosity, I noticed in the video that it recommended for a voltage regulator either an LM7805 or an LM1117 - either a super inefficient 5V regulator that has no business in new designs, or a 1.2V low-dropout 800mA regulator that is probably not appropriate to the other components it recommended (i.e. an LM7805 -> LM1117 would be 36% efficient at best).

It then also recommends a LiPo battery with a nominal 3.7V which obviously wouldn't be compatible with its LM7805 recommendation.

It also recommends an ESP8266 which is a 3.3V part so neither of the recommended regulators, nor the battery alone, would be appropriate.

Also, it recommends a 168MHz STM ARM micro - but it already has an ESP8266 with GPIO.

I'm not sure what it's suggesting I build but it doesn't really sound like a working environmental radiation logger ;)

How should I think about this?

Re: ChatGPT for Hardware Design

#33

Post by numpad0 »

mhb wrote: post #5

I recently used ChatGPT to find a part. Maybe this was an anomalous result, but it was wrong. Don't know if this will be an issue with your project. Me: Is there a digital to analog converter with an 8V analog range and serial input? CGPT: Yes, there are several digital-to-analog converters (DACs) available with an 8V analog range and serial input. One example is the AD5541A DAC from Analog Devices. It has a 16-bit r…

ChatGPT talks all from its memory, they're not suitable for hard data. Bing Chat type of architecture is more suited, though it still couldn't find a part matching that criteria.

Re: ChatGPT for Hardware Design

#34

Post by aloer »

The timing could not be more perfect. Yesterday I realized that I need to create my very first PCB to repair and smartify a device that I broke. Today I thought I should use ChatGPT to better understand how things work. And now this.

There are a lot of space constraints so first thing is figuring out if it is even possible to build as small as needed. Usually I would probably be better off starting to build big first and then scale down. But I feel confident that I can already get quite far with the amazing resources out there.

I spent hours watching videos about soldering, PCB design, some of the components I think I would need (very flat flexible cables, small ESP32 SMD, no space for USB) and then how to start actually doing the work. My last exposure to electronics was in grade 8 and that was very heavy on cables.

This 3h video here was very helpful to understand the process. It's an overview of building your own ESP32 dev board with EasyEDA from start to finish: https://youtu.be/S_p0YV-JlfU

I like these top-down big picture kind of things that go over everything in just enough detail to know what to search for. Does something like this but with flux.ai exist for complete beginners? The tutorials seem more focused on users switching over from other PCB design tools.

Every step of the way I am just amazed how today I can get physical things custom made and shipped within days.

And then there's fun videos like this here: Small scale fully automated PCB production in what is basically a garage -> https://www.youtube.com/watch?v=ZH6ljHUx_IM&t=800s

Re: ChatGPT for Hardware Design

#35

Post by hammyhavoc »

rock_hard wrote: post #31
hammyhavoc wrote: post #12

Earlier quoted context omitted.

Can you elaborate on "really well"? Because a few weeks of testing and "really well" versus the claim of reducing costly mistakes on the marketing blurb whilst simultaneously acknowledging hallucination doesn't inspire confidence. I'd be interested to know a lot more about testing methodology too. A hallucination could be a much more expensive problem than someone simply making a basic mistake that someone might othe…

To give you an example of the typical response quality we encountered: @copilot can you design a low pass filter for me with 3db point of 300khz? copilot: Sure, I can help you design a low-pass filter with a 3dB point of 300 kHz. A simple passive RC (resistor-capacitor) low-pass filter can be used for this purpose. To calculate the values of the resistor (R) and capacitor (C), you can use the following formula: f_c =…

And how about the negative responses involving hallucination? Ideally giving an example of the worst hallucination encountered thus far and least worst thus far.

As a potential customer, I'm more interested in knowing where the potential known failure points or points of friction are than knowing what's working well.

Yes, when it works, it's certainly compelling in its responses—but as it never knows when it is wrong, it is equally compelling at a cursory glance when it hallucinates.

For most folks interested in this, their Dunning-Kruger complex will prevent them from picking up on when it's wrong because it's so convincing and sure of itself.

Re: ChatGPT for Hardware Design

#37

Post by lnsru »

Nice stuff. But hard to find good application in my environment for it. I mostly re-work old designs or use reference designs with particular chips (that the company has in stock at the moment). Maybe somebody could greatly benefit without the limitations I have at work.

Re: ChatGPT for Hardware Design

#38

Post by heyzk »

rock_hard wrote: post #24
steve_adams_86 wrote: post #17

Earlier quoted context omitted.

If this works well, are you interested in training a model specifically around the domain you work in so you could get deeper integration with your product? I have a personal project similar in concept to flux, but rather than working at the circuit level, it's at the device, firmware, and infrastructure level. I've been trying to use GPT to intuit how to network devices using different protocols, such that it might…

Yea def looking into this We already use a multi model approach and mix of non deterministic and deterministic tooling in the pipeline to optimize for a given task Lots of more work to be done here

The flux docs indicate that you're using a custom-trained LLM and in another comment it was mentioned you're using LangChain (to integrate non-deterministic tooling perhaps). Just curious, but are you using some combination of your own model, GPT3/4, and calculators / datasheet readers / etc?

Re: ChatGPT for Hardware Design

#39

Post by nunuvit »

rock_hard wrote: post #2

Founder here: Incredibly stoked to announce this today! This is a dream a long time in the making. When we first thought of rebuilding the hardware design tooling stack from scratch, AI driven design seemed like the dream at the end of the tunnel...only 3+ years later we are here Given the open ended nature of AI Expert agents we are launching a community preview. We are committed to keep building http://flux.ai in t…

Have you reevaluated the decision not to implement much by way of basic interface features like symbol and footprint editors for such a long time? It's a real barrier to even playing around with the software. I want to see you succeed, and my experience in the EDA industry and as an electrical engineer is making me wonder about that approach.

It makes sense if you're just shopping it around to the EDA incumbents since they'd have you redesign those parts anyway, but the fact that you currently offer a paid subscription to users makes me think that either you've underestimated or I've overestimated the importance of having those features by now. I guess that could also make sense if you want a more manageable user base for now.

Re: ChatGPT for Hardware Design

#40

Post by rock_hard »

heyzk wrote: post #38
rock_hard wrote: post #24

Earlier quoted context omitted.

Yea def looking into this We already use a multi model approach and mix of non deterministic and deterministic tooling in the pipeline to optimize for a given task Lots of more work to be done here

The flux docs indicate that you're using a custom-trained LLM and in another comment it was mentioned you're using LangChain (to integrate non-deterministic tooling perhaps). Just curious, but are you using some combination of your own model, GPT3/4, and calculators / datasheet readers / etc?

The answer is: all of it!

Not everything is in prod yet but we def combine all the things you can imagine to pull this off

Post reply on HN