HN user
I think there's a market for a home AI server that can run an LLM or video gen model behind a web frontend. Not literally a raspberry pi strapped to an H100, but something with lopsided enough specs that people joke it is.
And precisely because it's such a huge headache to do yourself, I think a small company could make a nice business wrapping up used datacenter cards in that sort of server.
There's a really noticeable difference in time frame covered in your examples (80s and 90s) and the one in the comment you're replying to (2010s and 2020s).
Is that just two people with different go-to examples? Or is there something going on here?
(I don't mean this as a leading question to some conclusion in my back pocket, I genuinely have no clue.)
Gosh that Polish language scores graph is good. Clean bell curve, slight skew, classic little bump at 100 because the measure is truncated, and then this giant mess at 30 that looks more like something out of a heart rate monitor than a normal distribution.
There's a whole scene around making bots for Starcraft: Broodwar, using an API (BWAPI) that doesn't allow cheating. They're quite good now, better than most humans. But the top bots still can't beat a pro, or even a high ranked ladder player.
It looks like some of these terms aren't indexed (or the site is just too hug of deathed right now), but I'd like to see the graph of like, social media, iot, cryptocurrency, ai.
The transition between crypto and ai on the graphs is already pretty funny. https://hackernewstrends.com/?q=crypto&q=chatgpt
26 emojis? You could do the entire bestiary from the original Rogue with that!
There was a similar wave of IPOs with Uber and AirB&B, not tied to a bubble popping.
(I mean, I think this looks incredibly like a bubble too, but for completeness sake, that's the counterexample I can think of.)
Yeah, Love2d is a great option for gamedev. It doesn't have the same built-in tools as Godot so you'll need something else for putting together maps (use Tiled [1]), and you'll need to write your own main/render loops (these are just two for loops, nothing fancy).
I watch things from unknown-to-me creators in a private window, then copy the URL over to logged in window if it's any good. Same idea, might be an easier workflow.
Absurd that we have these sorts of workarounds, but of course the view numbers are better if it keeps fishing for just the right kind of clickbait trash that you'll wolf it down endlessly.
When I was a kid, we'd host the pics we want to post on forums on geocities and rename the file extensions to .txt to get past its "no hotlinking images" policy. So it's not like much has changed.
There are a lot of barriers between kids and better solutions, one of which is that anything needs a domain and a server, and that means a credit card.
Modern AAA games take tons of people because of ballooning scope and graphical fidelity expectations. Games like Super Mario World have went from highly technical team efforts to something a person with no training can accomplish solo. (However, 3D tools have lagged behind dramatically. Solo dev Mario 64 is possible but needs way more specialized knowledge.)
Lovely page. Reminds me of the venerable Think Labyrinth (https://www.astrolog.org/labyrnth/algrithm.htm) page, but the live demos add a lot.
My favorite maze algorithm is a variant of the growing tree algorithm - each time you carve a cell, add it to a random one of N lists. When choosing a cell to visit, pop the last cell off the first non-empty list. It's considerably faster than the standard tree algorithm, but more importantly, changing N has a dramatic impact on the texture of the maze (compare 1 2 4 8 etc on a decently large maze).
Dang, it's been cool watching gaussian splats go from tech demo to real workflow.
Is it possible to replicate the digital->film transition with tone mapping? (I assume the answer is yes, but what is the actual mapping?)
People are working on recovering PBR properties, rigging, and editing. I think those are all solveable over time. I wouldn't start a big project with it today, but maybe in a couple years.
If you want a real cursed problem for Gaussian splats though: global illumination. People have decomposed splat models into separate global and PBR colors, but I have no clue how you'd figure out where that global illumination came from, let alone recompute it for a new lighting situation.
If you want that 2000s feel for your own sites: Gradient background, drop shadows, and side bar nav.
From the article:
The kernel is still adding support for some 32-bit boards, he said, but at least ten new 64-bit boards gain support for each 32-bit one.
And
To summarize, he said, the kernel will have to retain support for armv7 systems for at least another ten years. Boards are still being produced with these CPUs, so even ten years may be optimistic for removal. Everything else, he said, will probably fade away sooner than that.
So, no, he does not think that at all.
Punctuation characters tend to be pretty consistent; there's a few lineages of roguelikes that copy them from each other, and only a handful of things a given mark can mean. ! means potion
Letters are almost always monsters, but which monsters are in which game is anyone's guess.
Caves Of Qud uses Markov chain generated text to great effect in some places. I think something light that's still more competent than Markov chains has a lot of potential.
I don't think either reach the real wtfs of their language. id() does some weird things with string interring, javascript coerces cross-type arithmetic into weird nonsense instead of throwing a type error. Whatever, neither of these come up in the real world.
Builds so messy that you need to resort to venv, every second DOM API being someFunc(thirdMostImportantArgument, { mostImportantArgument: val }), these are the real wtfs.
The stories I've heard have been mostly about scraper bots finding APIs like "get all posts in date range" and then hammering that with every combo of start/end date.
Gaussian splatting models the scene as a bunch of normal distributions (fuzzy squished spheres) instead of triangles, then renders those with billboarded triangles. It has advantages (simpler representation, easy to automatically capture from a scan) and disadvantages (not what the hardware is designed for, not watertight). The biggest disadvantage is that most graphics techniques need to be reinvented for it, and it's not clear what the full list of advantages and disadvantages will be until people have done all of those. But that big disadvantage is also a great reason to make tons of papers.
I prefer "ess queue ell" these days, but the first DBA I ever worked with pronounced it "squirrel".
I don't really think this is true. Gaussian splats certainly came from a context where an opaque representation is expected and normal, but they ended up being an entirely comprehensible format. They're not as simple to operate on as an SDF or voxel representation, but I think they're on par with triangle mesh geometry. A transformed fuzzy sphere is about as complex as a triangle, and spherical harmonic colors, while more conceptually difficult than textures, have fewer moving parts.
I agree; the point of a struct type would be to allow a compact memory representation, and you're not going to get it if your constructor can do if(someArg) { a = 1; } else { a = 1; b = 2; }.
You don't strictly need known/consistent types, but it sure helps, since otherwise everything needs to be 8 bytes.
I don't think a way to read into and out of ArrayBuffers is possible, since these can have pointers in them. I think it needs a StructArray class instead, so there's a way to actually make a compact memory array out of all of this.
I can't say whether Java as a whole will ever become the next COBOL, but Java 8 already is well on the way there.
That's called bricktext[0], it used to be a thing you'd occasionally see on usenet and places like GameFAQs.
[0] http://www.catb.org/esr/jargon/html/B/bricktext.html (Also wow Google has completely erased this concept from its search results.)
Coloring a graph is trivial, just give every node its own color.
The NP-complete problems are for finding the optimal number of colors, or determining if the graph can be colored using a set small number of colors.
Erin's solution aims for a small number of colors, but doesn't try to be the optimal number.
Going as far as adding lines support to Blender gltf export is quite a flex! And it looks really good on the Ferris wheel.
Downtown Ballard had a lot of power outages over the last two years. Apparently they were using a different transformer than everywhere else, all of which are getting old, and some procurement mistake lead to the spare parts being back-ordered by a year.
(This is second hand through a neighbor who actually went and bothered them about all the outages, so there are probably mistakes.)