validate against what? The “inner monologue” is the llm itself. It won’t be any better than itself.

it’s federated. It’s the only way it can work. Everything still on that ist must suffer from the same thing. Federation means handing stuff to someone else. Once that’s done, it’s out of your hands forever.

but not all of it is converted back to heat. The energy used to do actual work is used for that. Some of it comes back out as heat

this is just horrible advice

it’s the second language that comes to mind when I think of the word “footgun”, right after old c++

all json is valid yaml and can be parsed with a yaml parser. Yaml is literally a superset of json. In what world are they not comparable?

I’m not talking about drivers for stuff that is not the cpu itself. But the processor itself usually contains a bunch of peripherals that need their own stuff.

one of the most popular languages, used in one of the most popular game engines, has no learning resources?

Press X to doubt

Quite bad actually, since most of this stuff is not specific to c#, and are just basic programming concepts. This leads me to believe that your python experience is “coby and paste stuff in until it looks like it works”, and you never took the time to understand what the code does.

it’s very possible to get linux to run on a processor without having implemented al functionality. You can just not support some onboard peripherals yet and have to do some things inefficiently in software. You don’t need good power management to simply be “running”, etc.

Getting linux to run is the first step, not the last. It’s the barest minimum you could do to have a product to sell. Running well, taking advantage of all hardware features properly is a whole different game.

do they come with all the necessary drivers? Or are they hoping they magically appear in the linux kernel after they’ve sold a bunch?

nah, this is just copium. Apple don’t release dev-kits to the general public. It was a real product, and it was a dud

i would like to filter out all “massive multiplyer online arena shooter” There are way too much of them

Any processor can run llms. The only issue is how fast, and how much ram it has access to. And you can trade the latter for disk space if you’re willing to sacrifice even more speed.

If it can add, it can run any model

raytracing is insanely expensive. If you saw what current cards can render in real time, you would see a very very noisy, incomplete image that looks like shit. Without ai denoising and a lot of temporal shit (which only looks good in screenshots). It is very very very far from being able to render an actual frame with decent performance.

it’s actually unknown. It looks like it, but it is not proven

“the next generation cloud-native”

that’s as far as I got. Cloud native is an immediate, non-negotiable red flag for me