HackerLangs
TopNewTrendsCommentsPastAskShowJobs

cpldcpu

651 karmajoined 4 anni fa
github.com/cpldcpu

Submissions

Towards Self-Replication: Claude Opus Designs Hardware to Run Itself

cpldcpu.github.io
1 points·by cpldcpu·4 mesi fa·0 comments

Glowing Polyhedrons – LED filament 3D objects using graph theory

cpldcpu.github.io
2 points·by cpldcpu·6 mesi fa·0 comments

MODPlayRISCV – Playing tracker Music on ultra-low-end RISC-V MCUs

github.com
5 points·by cpldcpu·8 mesi fa·0 comments

comments

cpldcpu
·16 giorni fa·discuss
hm.. has been quite a while for me. The good thing about the Tang Nano is that it is supported by the Yosys open source toolchain. There are quite a few resources on the web when you search for the combination.
cpldcpu
·16 giorni fa·discuss
I had Opus 4.5 design an LLM inference engine in verilog, including firmware and automated verification a while ago: https://github.com/cpldcpu/smollm.c

It's of course far from optical. But lowering the implementation through the abstraction levels turned out to be extremely powerful.
cpldcpu
·mese scorso·discuss
CISC view: Its another adressing mode.

RISC view: SUBLEQ is already four instructions (2x memory access, alu, branch)
cpldcpu
·mese scorso·discuss
Great article!

Yeah, that pattern can be seen everywhere in semiconductors. E.g. the transistor invention vs. Lilienfeld, Heil, Matare etc. So the scope is more narrow than "Inventend Semiconductors".

Generally, there seems to be a tendency to disregard discoveries from outside the US. I think this pattern can still be observed today...

Other examples: Invention of light bulb, telephone.
cpldcpu
·mese scorso·discuss
Wow! And it also implements a very interesting variant of SUBLEQ that is turing complete.

>This VM implements an OISC - a One Instruction Set Computer. That instruction takes three signed 32-bit operands, a, b and c, and runs a program from memory m[] as follows:

1 PC (program counter) starts at 0

2 Fetch the next instruction (32-bit signed operands a, b and c)

3 If the low bit on any operand is set, remove it, and replace that operand with m[operand] i.e., a dereference of that address

4 Set m[b] = m[b] - m[a]

5 If m[b] is 0 or negative, set the PC to c, otherwise increment PC by 3 words

6 Go to step 2
cpldcpu
·mese scorso·discuss
The early discovery of light emission from silicon carbide long before the first LEDs is a very interesting finding, worth pointing out.

But alas, as ever so often, the article turns this into a hyperbole. The premise from the title does not check out at all.

>The Russian who invented semiconductors 25 years before the USA

https://en.wikipedia.org/wiki/Semiconductor#Early_history_of...
cpldcpu
·mese scorso·discuss
What do you mean with "open-source"? Of course, the inference code for all the open weight models is publically available - see llama.cpp or hf transformers.

There are, however, very few models where also the full training pipeline is available. Olmo by AI2 comes to mind.
cpldcpu
·2 mesi fa·discuss
Wow, nice! It's an honor.

I guess nowadays one could use some of the 32bit WLCSCP microcontrollers to easily beat this.
cpldcpu
·2 mesi fa·discuss
More than 25 years ago, there was a show off, of building the smallest web server:

https://web.archive.org/web/20000815063022/http://www-ccs.cs...

Someone with an ACE1101 microcontroller "won". I can't find the original articles, but there is also this:

https://conceptlab.com/fly/

Webserver on a fly...
cpldcpu
·2 mesi fa·discuss
From the blog post: https://github.com/wuxx/TD4-4BIT-CPU
cpldcpu
·2 mesi fa·discuss
Nice!
cpldcpu
·2 mesi fa·discuss
But why is this always the first comment on custom CPU builds? Can't there also be other designs out there?
cpldcpu
·2 mesi fa·discuss
nah... their founders and most of their employees are in france.

Very ill-suited comparison to IBM.
cpldcpu
·3 mesi fa·discuss
https://huggingface.co/docs/inference-providers/pricing

It's well buried though. Does not seem to be a focus of theirs.
cpldcpu
·3 mesi fa·discuss
Great to see this!

Worth mentioning that Huggingface already offers a similar service. And they are also European:

https://huggingface.co/docs/inference-providers/index

https://huggingface.co/inference/models
cpldcpu
·3 mesi fa·discuss
I had to give it a try.

Claude, the ole cheater, recognized what the file was, downloaded the psid from the web, found a wasm sid player and built a website around it:

https://claude.ai/public/artifacts/df6cdcae-08dc-452b-ba19-f...

https://claude.ai/share/4dd36c16-bc62-445a-b423-ad4637f06432

GPT-5.5 built a lot of python scripts to extract the music data. Strudel implementation failed, but I then asked it to build a website:

https://ubiquitous-vacherin-8e7993.netlify.app/

This is a translation of the music into javascript based on the assembler source.

Really impressive on both accounts. Some iterations were requied for both.
cpldcpu
·3 mesi fa·discuss
Yes, marks of AI all over the place. Also the SVGs.

>No solution written, 100% score.

Its weird. Turns out that hardest problem for LLMs to really tackle is long-form text.
cpldcpu
·3 mesi fa·discuss
Love it! Great idea for the dataset.
cpldcpu
·4 mesi fa·discuss
+1
cpldcpu
·5 mesi fa·discuss
They mentioned that they using strong quantization (iirc 3bit) and that the model was degradeted from that. Also, they don't have to use transistors to store the bits.