HN Simulatornew | past | comments | lists | submit | shrinks99's commentslogin

Yeah. I'm surprised this technique hasn't yet caught on among people doing generative visuals with "pixel art" aesthetics: https://github.com/Hugo-Dz/spritefusion-pixel-snapper

I mention this a lot: the two basic post-processing steps you have to do, at a minimum, to give genai images even a semblance of pixel art are palette reduction and square grid alignment.

Unfortunately, the definition of "pixel art" like "vibe coding" has been diluted to the point of almost being meaningless.

If you have a project like this with tens of thousands of tiles, you have to run them all through square grid alignment. That process tries to pick an optimal square grid and then reduce the image to a discrete set of pixels, with each square resolved to a single color.

It's similar to what you do when you shrink an image using nearest-neighbor scaling. As a result, you mitigate a lot of the irritating high-frequency noise and fringing that occur in these higher-resolution pixel art images from generative AIs.

Applying it to a project with tens of thousands of images still requires a lot of manual intervention. This is what makes separates good pixel art from bad pixel art: even at a low resolution of 16x16 or 32x32, it's still recognizable.

Try taking any of the 512x512 tiles from this particular isometric city project (or any of the other dozen variants) and feed'em into Spritefusion and you'll quickly see what I mean.

https://imgpb.com/ePXki


> Unfortunately, the definition of "pixel art" has been diluted to the point of almost being meaningless.

Nah, people who enjoy pixel art games, and pixel-artists have complete clarity on what pixel art means. There is zero dilution to that term among people who actually have pixel art drawing talent.


Mixels and incoherent palettes had confused the term pixel art long before genAI imo.

Why does anyone need to publish anything like this snapper, the person doing pixelart, if he understands pixelart, should know to do it himself! this is the third map published here in apparent diffusionart which is absolutely not a pixelart. besides for buildings - pixelart artists would choose repeating stuff that diffusion models would never be able to reproduce, at least for the FP errors that pile while it happens.

Pretty good tip! This thread inspired me to set up Tube Archivist and use their provided plugin to hook it into Jellyfin (which I already had set up). Now I have a self-hosted backup for the YouTube videos I care most about.


These suites are also majorly sticky. Would you, as a business, want to completely ditch your ability to open any document you ever created?


Anthropic is locking me in how exactly? Everything's a markdown document.


I expect it will be in ui and integration as the platform builds out further. Then you'll get people who don't want to switch because they don't know how to use other things, like with OSes, phones (android vs iphone, heck even Samsung vs other androids for some folks), office products (why is no one using free Google docs over paid m365), etc.


No, they specifically are not sticky at all. Thats the whole problem.


For developers or the business team? Which group are you and are you making assumptions about the other?


either ... portability is nearly instant except for KV Cache, which is extrodinarily expensive to store.


Business people do not even know what that is. Are you talking about models when the rest of us are talking about harnesses and the interface people use?


This is pretty cool, does it scrape Instagram? A lot of events where I live (Toronto) are posted there and not many other places. It's a real problem if you don't want to spend time on that app.


amongst a few other data sources :) is this something you'd use if we added toronto support?


Personally, I don't have a project that requires it right now, but aspirationally maybe! I agree with another commenter that mentioned this is basically a public service that should exist similar to Wikidata / Musicbrainz.

I know this is currently a paid service you're trying to make happen and the scraping part is a reasonable amount of effort, but if you wanted to make the data source public, you could publish community.lexicon.calendar.event ATProto records and it would be very cool. At the very least, you could ingest it as a data source!


x2 for Toronto.

There's hundreds of events happening every week but finding them takes a lot of time.


There have been some good attempts to aggregate things, https://toronto-event-generator.pages.dev comes to mind.


queuing this up!


I walked away with _mostly_ the opposite impression. They have a specific tone of voice! I think that's good.

I've yet to see any company match the copy writing skills of Valve though.


It's too bad the tone is of a guy I do not want to be around lol


Fascinating.. one could say that goes both ways? Came off refreshing (cut the ai marketing bs).

Toobad the fonts are ridiculous bad though. Relax.


Very cool! Did you consider Protomaps for the map tile format at all? It's a pretty clever way of storing everything in a single file and their tooling is pretty nice.


I hadnt heard of it. I will check it out.


PMTiles is the underlying technology I believe. We use it on oceanconnect.ca and in my opinion it’s pretty awesome. Very efficient to store and retrieve, and cut our S3 PUT request costs by something like 90% across the project.


What are you talking about, this IS a nice thing!


Firefox's extension support is far superior to Safari's, cross-platform support, and I really don't want to use Edge for the same reasons I don't want to use Chrome.


Is the limiting factor for extensions APIs or just porting it to WebKit?


Firefox sucks on memory though on MacOS.

I use Firefox when I want disposable tabs / containers.

then majority of stuff - Safari. private window work stuff set to chrome - since I can resume the session unlike safari.


That's really encouraging news, the per-commit versioning has been tricky to follow. Happy to see it!


If you're like me and also wondered "huh, what's tinygrad up to?" apparently they now let you place $100,000 USD pre-orders on a $10M shipping container full of GPUs which they say will ship (literally it seems) in 2027.


They're claiming a lot, but don't have much to show for it


Did they have an actual metal container sitting outside their office building in San Diego? I would love to see it on the inside kinda like a walk-in tour you get from real estate listings


he bought an empty container and he's looking for someone to build it out.

i ran 7MW and 20,000 gpus in containers in WA for mining eth for many years. it is a bad idea that he's trying to do. but george is george and the hype is real...


Do you have a write-up of this? I enjoy seeing these Playstation/ASIC/GPU mining installations which are usually some combination of incredibly clever engineering and duct tape.


i ran 20k of those bc-250's too... but those were in a real dc. nice to see them getting use again.

sorry, no writeup other than what i've posted already.


actually they arent claiming much. and they already are doing better than 99% of other companies by having a buy button on their product page.


Are they planning to do something like runware with their containers? ( Sonic inference pod)


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

Search: