Pure Silicon Demo Coding: No Cpu, No Memory, Just 4k Gates
Key topics
The fascinating world of hardware design is abuzz with a demo coding achievement that showcases a tiny tapeout with just 4k gates, sparking a lively debate about the parallels and divergences between hardware and software development. Commenters weighed in on the tradeoffs between the two, with some noting that while hardware offers cheaper parallelism, mistakes are far more costly, and others pointing out that the skillsets required for each are distinct, with hardware taking decades to master. As one commenter mused, the internet has enabled niche communities like this one to thrive, even as it contributes to the homogenization of culture. Amidst the discussion, a nostalgic thread emerged, with some reminiscing about the pre-algorithmic era of shared cultural experiences.
Snapshot generated from the HN discussion
Discussion Activity
Very active discussionFirst comment
2h
Peak period
37
0-6h
Avg / period
8.2
Based on 74 loaded comments
Key moments
- 01Story posted
Dec 20, 2025 at 11:45 AM EST
13 days ago
Step 01 - 02First comment
Dec 20, 2025 at 1:17 PM EST
2h after posting
Step 02 - 03Peak activity
37 comments in 0-6h
Hottest window of the conversation
Step 03 - 04Latest activity
Dec 23, 2025 at 7:23 PM EST
10 days ago
Step 04
Generating AI Summary...
Analyzing up to 500 comments to identify key contributors and discussion patterns
Want the full context?
Jump to the original sources
Read the primary article or dive into the live Hacker News thread when you're ready.
Software takes 1 year under someone smart in a production environment.
People that conflate the two... longer or more likely never.. =3
That's very funny.
During a career role most have no idea "why" chips were designed and built a certain way, nor require this information to work within abstract domains.
In many ways, vibe-coders are the absurd optimization of a naive trajectory toward zero workmanship standards. =3
https://en.wikipedia.org/wiki/Five_stages_of_grief
https://en.wikipedia.org/wiki/Metastability
https://en.wikipedia.org/wiki/Clock_domain_crossing
For example, you probably want Verilator if you're working on a CPU but you probably want to use Icarus if you're building an IC2 interface since the protocol requires you release the bus line (i.e. drive Z), which is not something you can directly verify with Verilator.
For example, I've taken code optimized for Xilinx, ran it for another vendor, and resource count ballooned because stuff that was built-in/free on one wasn't on the other. It's a lot of work to truly make generic code and usually just means switching out modules per vendor.
If you have a ROM, it's not "no memory".
Needlessly pedantic!
I thought this was pretty cool but the first video didn't play. All this write up and I really just want to see the damn demo in action first! (Edit: reloaded the page and it worked. I still would like to see it on rela hardware!)
https://youtu.be/7xPS-0nydms
Bucket-brigade delay lines?
But I'm not really familiar with what that is.
Imagine a pair of MOSFETs connected to a pair of capacitors, and a bunch of those joined together in a chain. All the gates of each one of the pair of MOSFETS are connected together, giving you a "left" and "right" clock input.
When you put a signal in if you pulse the "left" and "right" inputs, it'll store the signal voltage in one capacitor, then pass it off to the next capacitor in turn, like old-timey firefighter handing buckets of water down a line of people.
They used to use this for delaying audio signals before digital memory and analogue to digital conversion was cheap enough to use.
I have no qualms saying a stateful device can have no memory in the addressable memory sense.
if you have pedantry, it's also not "no memory"
There are many bad things about LLMs, but a benign shift in popular language usage isn't one of them.
but without the craft of a good advertising slogan. So worse!
Culture contaminates.
Organic shifts in language are fine. What is not fine is Big Money (which most forms of AI are) manipulating society at large - and that's not just the AI companies' doing. Think of Tiktok leading people to say "unalive" instead of the various clear words before (e.g. kill, murder, executed, run over by car, mauled to death by animal).
x=CPU y=Memory Z=4k gates
The project has a narrow scope of use-cases. =3
Analog i/o pins: definitely limited, even if you purchase the highest option available (6).
ok, but silicon is doped so it's slightly impure, and CPUs are also silicon and memory is also silicon.
you actually meant "4K gates, no clock, no synchronization, no timing" and maybe a little "not exactly sure when the output is rea... is rea... is ready"
The HAKMEM sine/cosine generator is such an elegant choice - it's numerically stable in fixed-point and requires only adds and bit-shifts. Perfect for hardware. I used a similar approach once for generating test patterns in an FPGA.
The fact that you can iterate on this in simulation, then deploy to actual silicon via Tiny Tapeout for $150 is honestly mind-blowing. We're living in the future.
https://www.youtube.com/watch?v=K9mu3getxhU&t=780s
CORDIC: - Iterative algorithm (needs multiple clock cycles) - Accuracy improves with more iterations - Generates both magnitude and phase - Typical hardware implementation: 12-16 iterations for decent precision
HAKMEM (Item 149): - Single-cycle computation (just two adds per step) - Uses the recurrence: x' = x - εy, y' = y + εx - Accuracy depends on word width and epsilon choice - Numerically stable in exact arithmetic if ε² < 2
It's really cool but it doesn't seem practical at all. They aren't setting up print runs, just one-offs (https://tinytapeout.com/faq/#how-many-chips-will-i-receive-c...) and $150 could get you... many orders of magnitude more power than that.
... For that matter, apparently the microcontroller in the dev kit is a https://en.wikipedia.org/wiki/RP2040 , which seems like a beast in comparison. And it's still available for less than $1 USD on PiShop.
Remind me to participate in the next one!
> 1024x32 Commercial SRAM > CF_SRAM_1024x32 > Commercial SRAM: 1024 words x > 32 bits (4KB) with Wishbone Bus interface > Area: 0.17mm² > GPIOs: 0 > License: Commercial - $2500 per project
Oh shit, this prompted me to check and turns out TinyTapeout is back to life! https://tinytapeout.com/