There's something incredibly satisfying about this

If this runs on a PI... can it run on webgpu ?

The core of it uses rust with no_std and on the pico we use embassy-rs with no os. At the same time, the core can be build for webassembly but it doesn't use webgpu.

Unlikely, Raspberry Pi Pico 2 W is a microcontroller, not a computer. It runs either bare metal or an RTOS.

"4" ... f!cking "4" !!!!

Here i am with fully random 32 chracater passwords for my fu!king gmail and these a$$holes are using "4" for an international CA key

The thing is they could probably have easily had the pvp and economy and soverignty of eve, and that would have been amazing

I don't play PVP and am not interested in a heavy economy simulation game. I like NMS the way it is, like Minecraft, a sandbox.

There really are a lot of things that they could do it with it, but they've already done a lot of things over the years; you can't say they haven't tried things, though I can't say they've always delivered a fun experience with the various experiments. They did pets and pet battles (boring to me), various building ideas (freighters and corvettes, bases, moderately more interesting), trading (I don't care for the implementation, feels detached), salvage, colonization (you become the governor of existing towns and make them run more efficiently and build stuff), etc. And, the worlds and critters got more varied several times over the years since release.

In short, I found it a very fun game until it wasn't, and then I had no reason to log back in. There are things you can do to make your own fun outside the story line; grind for a better freighter, find an build on interesting/valuable planets, find the coolest ships.

Once the story ran out, I was pretty much done, but even if you just play the story and then quit, it's still an excellent value! I got many hours of enjoyment out of a game I paid maybe $20 for. It's a hell of an entertainment deal. I recommend it to anyone who hasn't played it all the way through in recent-ish times. Even if you only play the story out one time and don't find any of the other stuff fun, it's still a very enjoyable game with hours of fun play at a good price that runs well on any decent system.

If I were a designer on the game, I'd find more reasons for human players to be in the same places at the same time and have their actions have a lasting impact in the universe. Riding in someone else's ship (which is what corvettes added) aint it, though. Projects to colonize specific worlds where anyone can stake a claim in a landrush type thing and maybe you setup competitive trade networks or battle for terrain or strategic resources or whatever, would be super fun, I think.

And, I wish space battles were more skill-based, as well. I've never played Eve, but I assume the space battles require skill and strategy and not just the biggest/baddest ship? So, if NMS got really good space battle mechanics, they could implement crazy cool space wars. But, they've got to shrink the game world for that to work (not literally delete the billions of planets, just give players a reason to all fight over one planet or system). Maybe competing for unique resources, like the spice on Dune.

I'll definitely try to find time to log back in and check out the new stuff. Whether it keeps me coming back after I try out all the new stuff and run whatever new missions/stories they've introduced remains to be seen. But, I don't think anyone plays it for the "billions of worlds" thing, because they're all the same after you see the first couple dozen.

To me it feels like what i actually want is Eve's economy, PVP/PVE strapped onto the noman gameplay, at least personaly

This i did like 15 images just today, mostly just editing and fixing one of my wifes photos that she wanted the lighting and stuff leaned up with so i decided to test out 2.5... it did a great job, but that was easily 15 images 1 of which was the end result but all 15 count toward that figure, i'd honestly guess its more 1:100-1:500 that actually makes it outside the chatgpt servers

This is the most normal thing, and this is the best example

AI is: 'my wife wanted better lighting in our reno pic thingy'

This is healthy and way in the range of normal, not dystopian.

I assume the comment I am responding to was made in sarcasm, so I'm basing my response on that register, so apologies in advance if I have got this wrong.

I don't understand what is dystopian about this: With the exception of mildly elevated resource consumption to achieve the end goal (a nicer photo), is this not the sort of things that humans have been wanting and doing for as long as photographs have been a thing?

It was previously mostly unachievable for regular people to have their snapshots processed in such a way, but it feels like it's just making what used to be special mildly trivial. Anyone who has had their wedding photographed for the past 20 years has had the photos touched up on the way through. Any school portrait for just as long.

100 years ago people would dodge and burn photographic prints to improve them as much as the technology allowed. This is just the forward evolution of that old tale.

No sarcasm.

Maybe slightly odd is how anyone could possibly think I was being sarcastic. :)

I'm not attacking you here or anything :)

I'm just a bit shocked at how cynical people can be. The thought would not cross my mind ... and I'm a skeptic of many things!

There's a video somewhere of some people in Wyoming, riding horses.

All of the comments are young people shock of these 'boomers' committing 'animal abuse'.

I mean ... just ranchers riding horses.

People think that 'Ward Cleaver' form 'Leave It To Beaver' is some 'creepy twilight zone figure' ... when he's just literally a nice guy. Or rather the character is.

We're getting way to jaded!

So - I really mean it: AI is going to be mostly very pedantic stuff.

I witnessed the birth of computers and the internet - and while AI is definitely a big deal ... I think it's just going to feel like 'more computing stuff' in the end.

Computers started out for Science - then it became about video games, socialization, design, commerce, and of course 'gambling and porn' and all the 'same vices'.

Every step of the way there were people thinking it was being 'debauched' etc..

AI is going to be about the most mundane things, just as most technology is today.

It will be boring and in the background, and probably weird like a poorly designed, slow web-page full of cruft, but it will 'do the thing'.

Shipyards shtting down not IPFS, the title is editorialied

All these weird partial language thought patterns im surprised none of the teams have taught the models to think in something like court stenography or some very dense pattern (i mean they even tried caveman language at one point)

Keep in mind that the model is thinking in a token space, itself a compressive representation of language.

(Note: there's still a huge grammar penalty, so, ugh do think small.)

The real breakthrough is going to be thinking in latent space.

Arguably this is already happening: the whole state of the model gets fed through from token to token, and even just shoving a bunch of dashes in between the input tokens and the model's output can improve performance (thinking tokens from the model help a little bit more, but the difference is not as large as you mught expect).

It selects tokens but they expand to embedding vectors which are huge, also in memory and attention requirements, I think?

It's not obvious if that would help. Some tests have shown that what exactly the thinking tokens are only makes a small difference to the performance of the model, and that the contents of them are sometimes only tenuously related to to what the model actually does after them. It seems like it could be they are more like a kind of "mumbling" and that the underlying mechanism by which they actually help performance is just that it makes more computation available to the model by just giving more passes on the earlier input tokens through the network.

One of the main goals at the moment is to keep thinking human legible. You can imagine how much harder it’d be to do root cause analysis on the recent OpenAI event if we couldn’t even tell what they’re thinking.

Sily question... why not all 3

All 3 is in the app! It's a much harder challenge to deal with the parallax and other artefacts that come up from combining 3 lenses, so that particular feature is launching in beta.