https://github.com/greggman/sedon

It was just an experiment and I probably won't work on it more but still, I was blown away with how far we got. There's a quite a bit we worked through even though it was only part time of those 4 weeks.

If you're an AI skeptic still, maybe seeing more concretely what's possible today might convince you otherwise? I'm not making any claim the code is good, though I'm also not claiming is bad. I had fun giving it a try. I was pretty blown away how far we got (We being the LLM and myself). Like many projects I also realized, it's too much work to finish and likely too few people would use it. Maybe it would inspire someone more motivated to do something along similar lines though.

Comments URL: https://news.ycombinator.com/item?id=48371199

Points: 5

# Comments: 0

It was used extensively in outdoor games like Jak and Daxter.

Browsers generally only allow a fixed number of WebGL contexts per page. So a generic element effect library has the issue that too many elements some will start losing contexts. The workaround is to just make one large screen size canvas and then figure out where the elements are you need to draw an effect for. now you only have one context drawing all the elements. But, you can’t know where to draw until the browser scrolls and renders so you’re always one frame behind.

https://webgl2fundamentals.org/webgl/lessons/webgl-multiple-...

WebGPU doesn’t have this issue. You can use the same device with multiple canvases

https://webgpufundamentals.org/webgpu/lessons/webgpu-multipl...

I get all the concerns, and I review all AI code at work and most AI code for personal projects. This one in particular though, not so much. I get that's frowned on but this is a small, limited scope, personal project. Not that I didn't pay attention, Claude did do some things in strange ways and I asked it to fix them quite often. But, conversely, I have zero rust experience, zero OpenXR experience, zero wgpu expericence, next to zero relevant Windows experience.

I'm guessing I spent about ~30 hours in total prompting Claude for each step. I started with "make a windows app that opens a window". Then I had it add wgpu and draw hello triangles. Then I had it add OpenXR and draw those triangles in VR. That actually took it some time as it tried to figure out how to connect a wgpu texture to the surface being drawn in OpenXR. It figured it out though, far far faster than I would have. I'd have tried to find a working example or given up.

I then sat on that for about a month and finally got back to it this weekend and zoomed through getting Claude to make it work. The only parts I did was make some programmer art icons.

I can post the prompts in the repo if anyone is interested, and assming I can find them.

Also in the last 2 weeks I've resurrected an old project that bit-rot. Claude got it all up to date, and fixed a bunch of bugs, and checked off a bunch of features I'd always wanted to add. I also had Claude write 2 libraries, a zip library, an rar decompression library, as well as refactor an existing zip decompression library to use some modern features. It's been really fun! For those I read the code much more than I did for this one. Still, "what I time to be alive"!

Comments URL: https://news.ycombinator.com/item?id=47749295

Points: 14

# Comments: 1

I went to gemini, picked "cavnas". used this prompt

> There's a famous CG scene in the movie Hackers where they "Hack the Gibson". It shows a bunch of translucnt cubes with glowing edges. The textures on the cubes are live computer text. The camera slowly flyies between the cubes tilting gracfully and it searches for the main one.

> Reproduce this scene in Javascript. Be sure to include each of those features

> 1. live computer text which you can simulate by drawing to a canvas offscreen and uploading to a texture, adding more output as it goes. You can even use "function.toString()" of the code you write as input

> 2. a post processing step so we get a glow

> You can probably use three.js for this

Here's the result.

https://codepen.io/greggman/pen/XJKPBZW

No, it's not as good as the site linked above and it's unlikely it would be. On other hand, it got this far on the first try. Maybe a few more iterations and it could get the stuff you want.

Non-ascii, I tried various subsets of Unicode. There’s the geometric shape area, CJK, dingbats, lots of others

Different fonts - there are lots of different monospace fonts. I even tried non-monospaced fonts tho still drawn in grid

ANSI color style https://16colo.rs/

My results weren’t nearly as good as the ones in this article but just suggesting more ways of exploration

https://greggman.github.io/doodles/textme10.html

Note: options are buried in the menu. Best to pick a scene other than the default

Most of the music I like is loops pasted together in some DAW. Sure, it requires taste to make a good song but if AI figure out how to replicate that taste can crank out catchy tunes I wouldn't have a problem with it. I can only guess though that too much of a good thing will lead to be getting bored with it ... maybe.

It's not like most pop music isn't formulaic. I enjoy the currently popular songs from K-Pop Demon Hunters but they're so cliche, if they turned out to be AI generated I wouldn't be surprised :P

Like a simple example, I expect if have a object selected and press Cmd/Ctrl-C, Cmd/Ctrl-V I get a duplicate.

These interactions seem like basic table stakes

https://post-processing.tresjs.org/guide/pmndrs/ascii

https://forum.babylonjs.com/t/ascii-shader-using-glsl-postpr...

https://threejs.org/examples/?q=ascii#webgl_effects_ascii

https://fwdapps.net/l/asci/

https://codesandbox.io/p/sandbox/ascii-postprocessing-n628p8...

https://www.youtube.com/watch?v=NxeRcnLr0ko

https://www.shadertoy.com/results?query=crt

I made a Pico-8 post processing script using a few

https://greggman.github.io/pico-8-post-processing/nano-villa...

also an article on starting one on WebGPU here

https://webgpufundamentals.org/webgpu/lessons/webgpu-post-pr...

If you click save you get the option to use a URL.

The problem with a URL every edit is a new URL. So you send the URL to a friend, then fix a typo, they need a new URL.

The other problem is of course the space limit.

https://github.com/google/toucan

There are ~40 million in the PST time-zone. Some percent have smartphones (80%+), ~50% of those are iPhones (16 million). Of those, the majority are set it English (80%+), and are divided into screen sizes. But basically, if you have an iPhone, you have the same fingerprint has at least a million other other people in the PST time size. You are at best, 1 of 100, not 1 of x,xxx,xxx,xxx.

You might be x,xxx,xxx,xxx of people who visited that unpopular site but no one needs tracking on an unpopular site. On a popular site you will not have a unique finger print.

https://webgpureport.org/

But, they are bucketed

https://www.w3.org/TR/webgpu/#privacy-considerations

It's not zero pieces of info but it's also not close to as bad as it looks. Effectively, everyone who has, say an NVidia GPU, will likely have the same list of features and limits.

As a more general example: The number is just a flat out wrong

> Unique to 1 in 2,147,483,648+ devices.

No, I have an iPhone Pro and am in the PST time zone, set to English. It has the exact same finger print as millions of other devices among the 40 million people in the PST time zone. In general, The only things different between 2 iPhones of the same model are time-zone, laguange setting, and font size.

Please STOP EXAGGERATING!

I tried this in 2014 with happyfuntimes

https://docs.happyfuntimes.net/

My conclusion was, past a certain number of people no one wants to game

Around the same time was AirConsole and still available

https://www.airconsole.com/