HN Simulatornew | past | comments | lists | submit | jitl's commentslogin

I’ve also spent the last couple of weeks vibe coding a component-based UI framework for C. My target is embedded hardware with 2 * 300mhz ARM 32 CPU, and a Nintendo 3DS like OpenGL ES 1.1 GPU from DMP: Fujifilm cameras from 2013-2016 like the X70, X-E2, X-T1, X100T. It’s a bit faster than iPhone 2007 on CPU side, and slower on the GPU side due to needing to copy from GPU back to main memory for scan-out. It’s funny how these things seem to come up and then everyone’s doing it.

I didn’t see the code here because I couldn’t find an easy link to its Github. My UI framework i’m calling FocalUI. We do layout and basic geometry with a fork Clay where we added “layer” and “focusable” markers to Clay elements. Instead of drawing immediate mode, we store a list of draw commands and layer geometry in a CoreAnimation-like tree. The renderer composits the layer tree and renders to OpenGL, and we use damage tracking to copy just the changed pixels from the GPU back to the camera’s framebuffer for scan-out.

It’s been a huge effort to add touch inertial scrolling, spring animation, navigation effects etc at 55hz to this device that before only did d-pad nav at 25 Hz, but things are starting to shape up for a release this week :)


Mentioning vibe coding and “It’s been a huge effort to add …” in the same phrase what a joke

life’s more fun if u do funny stuff

I absolutely love sketchup, and seeing it passed between various developers and never improved while gradually getting less and less free was very sad. it was awesome when owned by google and totally free. super happy to see a spiritual successor!

Classic case of high value to end users but low revenue, because ease of use doesnt bring money.

Gained a huge user base quickly then sold to Google who wanted it apparently to help people toy around in Google earth, then they did little with it and eventually sold to Trimble, mostly a surveying company, who similarly had hopes it would help expand their users and market reach, which it hasnt really its mostly still evades commercial revenues and continues to be used by the same kind of users as originally


no TLS 1.3, and last released in 2018?

True, but its in active development, see https://bearssl.org/gitweb/?p=BearSSL;a=summary.

The thing is very carefully designed.


i use opus 5.5 to delegate to sol and luna via codex-thread-bridge (https://github.com/saidelike/codex-thread-bridge). these days i’m experimenting with opus using luna as “hands” and it’s largely resolved the tokens per output issue with my Anthropic subscription.

¯\_(ツ)_/¯ it’s not efficient but yeah it’s the best


tried the claude code delegating and codex doing workflow with sol 6.1 xhigh and even that messes up my codebase still find lots of problems and half baked assumptions codex makes. wonder how you are doing that while integrating luna too..


problem with agent? use another agent to fix problem ._. i have the architect agent review the code and do a project report & refactor plan every 3 hours.


i am 100% sure i need opus 5.5 model to build stuff like this because i try lesser models because i want my tokens to go farther and lesser models simply cannot: https://fujimod.com


Not sure if this is possible on the Fuji, but back in the old days of the Canon PowerShot being very relevant, there was CHDK firmware which ran off the SD card on boot. No physical impact to the device and fully reversible by reformatting the SD card. This would be the way to go to reduce bricking chances.


yeah i currently only distribute sd card dual boot media where the the install process is “turn it on holding REC and pressing rear command dial” and uninstall process is “turn it off”


> All of this is unreviewed vibe-coded slop that could be hazardous to your camera’s mental, emotional, or digital health. If you're uncomfortable with risk to your camera, please don't use it.

You don't care anyways. Actual engineering works "100%" fine with small models now. You should try it!


I've tried smaller models and the best open models for a specialized UAV focused CAD + CFD + DFM optimization application I'm building.

The best open weights models still had a lot of defects and very egregious mistakes to fix even after I provided a very well detailed Requirement/Design/ Technical Implementation plans.

Opus and Gemini 3.8 (and now Argon) did a significantly better job. So I've baseline using these two models to build this application. I now use Opus for design/review and Flash 3.8 for the code gen/implementation work.

I'm sure small models and even open big models are probably fine for many common small apps and such, but they aren't really able to perform the same across all use cases.


as i said, i did try it. smaller models are not up to the task of deep reverse engineering work.


$200/mo @ 1 year = $2400 2400 gets me about 1/2 of a 128gb NVIDIA DGX Spark, so i could do a year of nonstop inference of (checks notes) quantized Qwen 3.8-27B, a model i would never choose to use. yay, such savings.


And that Spark isn't running a top end frontier model. Yes, the open weights smaller models that can fit on Spark are getting better, but they are still quite a way off in performance depending on use case.


> PDF is an old format from a more innocent era

if apple parsed pdf with (js, swift, rust) none of this would be possible. the problem is memory-unsafe code.


> if apple parsed pdf with (js, swift, rust) none of this would be possible

Interpreters and compilers can have bugs. Bugs can cause memory safety issues.


Since when was javascript memory safe


Since it was created.


Why is it we didn't rewrite coreutils in js a decade ago instead of waiting for rust


Because it’s a high-level interpreted language. Still memory-safe!

Rust is interesting because it’s a low-level language with strong memory safety. That makes it a good fit for projects like coreutils. I’d argue it makes it a bad fit for many other projects that get rewritten in it, though. Not everything actually needs to be in a low-level language with compile-time memory management.


claude is also online-only so i doubt they mind


Until the server goes down. Subversion going down at work stopped everything. At least with git you can still find work to continue on with even if you can't push.


idk, over the past 2 weekends i have a repo for hacking on fujifilm camera firmware that's gone from zero to tens of gigabytes: i've got an emulator for older arm cpus like the one in the X70 (2016), a Hypervisor.framework VM for running their newest AArch64 software for X100VI (2024), a complete C re-implementation of the x70 source, many tools for injecting code and executing on live camera RAM over USB. the emulator has gone through 3 rewrites: first it was python and naive, then it was python with hot parts in rust with a more mature device interface, and now it's rust with some cold or experimental devices still python.

i really don't want to brick my camera, so the deterministic emulator stack lets me do integration tests end-to-end. if i was doing this by hand with no ai, i'd probably have to run stuff on the camera just to test it. but because ai, i get deterministic simulation testing, at the cost of needing to burn $100 of subscription on it.

ai code is more like clay you gotta continuously keep shaping over time, but you can also get very very far, very quickly if you decide the code itself doesn't matter and all you care about is results.


Cool project. Empty repo? https://github.com/justjake/fujimod Not much interested in the code, more in the development process (prompts, skills).


Agreed, a buddy and I wrote a simple mod that shimed into Warhammer 40K: Rogue Trader and were able to pull all of the art and description assets for that game by walking the object graph. Took about 4 hours total (including the site).

Similar to your camera, we would never have been arsed with doing that much work were it not for LLM vibbing. It allowed us to get something working (sufficiently) very quickly.

At what point of maintaining this code would it be cheaper for me to actually find underlying patterns and program this myself with some actual thought, rather than continuously "shaping the clay" of my LLM slop? These questions keep me awake at night (not really), and probably predict the future of software engineering!


I'm curious, was there a specific reason you went with that solution for grabbing the games assets?

Because asset-ripping is a pretty widespread, well documented practice in gaming - it's how a lot of modding communities figure out and expand on the games they mod.

A quick search shows WH40K: Rogue Trader is built on Unity too, so any number of existing Unity-specific asset-ripping tools should have worked for your purposes. Though I should say I've never actually dabbled much with modding Rogue Trader, so maybe I'm wrong?


It was January I think and I don't remember our thought process. I think we mostly just remember looking at the fandom.com or whatever crap was there and thinking how out of date that stuff was.

We also just wanted to try some modestly sized project with an LLM. Neither of us had any real experience doing any of the things we ended up doing for that project (other than my making the website and him understanding the graphics pipeline).

So, unfortunately, no, there was no specific reason I suspect.

https://stephencagle.dev/posts-output/2026-01-30-warhammer-4... is my best recollection of the thought process.


Are you planning to release any of this? Fuji cameras are so nice but need some opening up


hopefully! i’ve got lots of good stuff for the EXR2 processor generation.


what do you think they train their ai on?


Guidelines | FAQ | Lists | API | Security | DMCA | Apply to YC | Contact

Search: