HN user
I love AI side projects, it allows for quick iteration and finishing on ideas that weren't feasible due to time constraints before. I just made one yesterday/today. I used Magit before and during the making of this I found out GitUI exists too but I think it's pretty powerful if the tool doesn't do exactly what you want and are opionated you can tailor your tools the way you want to.
A1: I don't think so now, but good idea. I haven't thought of it because well it depends entirely on how you interact with things. I tried many alternatives in the past from Magit to shell scripted vimdiff setups and now I usually just use VSCode or git from the CLI. I didn't want to open up another slow VSCode instance for every directory so this is mainly my motivation for making this. I just add an alias t='path/to/tuide' to .bash_aliases and type 't' in the repository I want to see.
A2: No it doesn't support that now. I did think of just having it be like a opionated visual git diff and using it in that way at first.
A3: Not sure. Couldn't you already do this with git reset and branches? I do have some features in mind to make some things easier like selecting multiple commits and then combining them for a rebase, but wait ah if that fails or something goes wrong the above snapshot feature might come in handy.
Thanks for the suggestions.
Some developers working at the NOS (Dutch Broadcasting Foundation) as hobby project made it possible recently to view teletext through SSH.
ssh teletekst.nl
I didn't expect to see "Toeslagenaffaire" on HN. Does this even belong on HN?
54 requests
57.54 MB / 11.83 MB transferred
There's only 11.83 MB transferred, but that doesn't excuse it.
There's nothing stopping someone from never buying another game on Steam, and moving to another marketplace on PC, unlike the store monopolies on consoles and mobile devices.From one of his other blogposts. "Guidelines for computing sizes and subscripts"
Never mix unsigned and signed operands. Prefer signed. If you need to convert an operand, see (2).
If AMD and Intel can't match NVIDIA then how can Huawei?
I don't know how well this makes you understand your dependencies. As for C/C++ a lot of people probably depend on stb single header files libraries. There's stb_truetype but it specifically mentions not to use it on any untrusted/outside .ttf files which I do like but you have to keep in mind to bake to bitmaps or only use your own .ttf provided files, thus I would put this dependency in another place like tooling. Is there a way to do this in other languages like JS and NPM? Maybe carefully choosing which dependencies you include is better?
https://github.com/nothings/stb/blob/master/stb_truetype.h#L...
Your game trailer links to a short called "Fly me to the moon - 19 September 2024". I think this is a mistake?
I wonder whether it'll be possible to compress enough of the game to make (almost) every possible scenario that you could encounter in the game be playable. Same issue that the previous AI experiment for Minecraft and others had is that objects and enemies seem to pop in and out of nowhere. Could the "learned" probability be high enough for this never to be an issue? You ever think you're seeing something in real life but it's just an optical illusion, it kinda feels like that to me. Obviously this still requires an entire game to be made before you can train on it, but could maybe open up other development and testing of games.
what
Whenever a new AI model gets released and is available for the public. From the last few I've tried they were always NVIDIA only because I assume that's what the researchers had at their disposal.
If I had to describe Quaternions to someone I would first try to explain a Plane (Ax + By + Cz + D = 0) to them. ABC being a (normal) direction that the Plane is pointed towards and D being the distance from the origin.
A Quaternion from what I believe is just the same but instead of distance it just encodes the rotation around that direction as a fixed axis. (Instead the angle stored is half etc).
Feel free to correct me if I'm wrong, I'm not a math-heavy person.
"If you steal from one author, it's plagiarism; if you steal from many, it's research." - Wilson Mizner
After briefly skimming through their talk they mention they're a third party consultancy company making recommendations on safety for NVIDIA, so "Nvidia Security Team" take it how you will. Also I think these drastic changes moving towards a completely different language always bug me, especially when the syntax is also completely different.
Then there's like everything libraries wise and CUDA is all C/C++ if I'm not mistaken. In every large project I'm sure you're eventually going to find some exploit attack vector. Especially if the focus from what I believe for GPU/CUDA until recently wasn't mostly focused on security rather than performance and those are always trade-offs.
Stochastic processes
I tried to switch to Linux a few times over a couple of years but had a hard time switching from Windows to Linux until I found a replacement for Visual Studio (non-VSCode), which I replaced with VSCode on Linux. Aside from a few games that wouldn't work for any reason (Kernel level anticheat, WINE bugs, manually setting export env variables) it's been great. Being a tinkerer/developer I wonder whether this made it harder or easier to switch, you have more knowledge of how to fix things but the same amount of time as everyone else and if you don't depend on specific programs / workflow like using a browser you may be able make the jump to Linux easier.
Reminds me of Quake patchDef (Biquadratic Bézier surface patches) https://quark.sourceforge.io/infobase/maped.curves.html
“One of my most productive days was throwing away 1000 lines of code” ― Ken Thompson
Didn't most first person games have this automatically because of using the Quake engine due to their PVS system? It was used to reduce drawing overhead but also beneficial in not sending more data than needed and preserve bandwidth serversided.
I went down a rabbithole and wow.
Found a comment from the author of https://github.com/stclib/STC apparently and then came across this example:
https://stackoverflow.com/a/76887723
int coro_a(struct a* g)
{
cco_routine (g) {
printf("entering a\n");
for (g->i = 0; g->i < 3; g->i++) {
printf("A step %d\n", g->i);
cco_yield();
}
cco_final:
printf("returning from a\n");
}
return 0; // done
}
After running it through a preprocessor, it gives me this.
int coro_a(struct a* g)
{
for (int* _state = &(g)->cco_state; *_state != CCO_STATE_DONE; *_state = CCO_STATE_DONE) _resume: switch (*_state) case 0: {
printf("entering a\n");
for (g->i = 0; g->i < 3; g->i++) {
printf("A step %d\n", g->i);
do { *_state = 14; return CCO_YIELD; goto _resume; case 14:; } while (0);
}
*_state = CCO_STATE_FINAL; case CCO_STATE_FINAL:
printf("returning from a\n");
}
return 0;
}Suggestion to add to the list: https://github.com/ioquake/ioq3
From the title at first I thought this was a Quake browser port [0]
Just a matter of time before Chrome pushes another JS API like File System API and you can actually defrag your drive from the browser.
Is this is it or did they already lock down "their" hardware so you wouldn't be able to install alongside or solely another operating system? It seems that it's something we shouldn't take for granted as of now that the separation of OS and hardware allows you to install Linux or any other homebrew operating system on "your" computer.
You could try https://polyhaven.com/
What are atlases used for? https://www.sharetextures.com/atlases
Apple Vision Pro personas competition
Is this similar to binary search or could it be a binary search where you're just excluding all possibilities that you know can't be the probable result that you're looking for?