How do you know Ohio is real? Have you been there yourself? Have you seen it with your own two eyes? Or do you just trust all the people who claim to live there?
You see, believing in the existence of Ohio is exactly the same as believing that my dad works for Nintendo and I got the play their next game early. It was awesome btw.
That's not just food though, that's literally everything.
A person whose spent thousands of hours making and consuming music is going to notice and value different things than someone who just listens to whatever is on the radio.
A person whose driven hundreds of cars probably has a way better idea about what they like and what they don't like in a car than someone whose driven 3.
That's what having a "developed taste" means. Yeah it's all opinions, but you can't even know what your opinion is of something unless you try it, and you can't develop an in-depth opinion of something unless you've tried a whole bunch of similar somethings.
That doesn't mean you need to do that with everything of course. Not everyone needs to cultivate an appreciation for everything, I don't know or care what good beef Wellington is like for example, but I'm also not going to get pissy with people purely because they have different interests than me.
Being able to summarize and answer questions about a specific corpus of text was a use case I was excited for even knowing that LLMs can't really answer general questions or logically reason.
But if Google search summaries are any indication they can't even do that. And I'm not just talking about the screenshots people post, this is my own experience with it.
Maybe if you could run the LLM in an entirely different way such that you could enter a question and then it tells you which part of the source text statistically correlates the most with the words you typed; instead of trying to generate new text. That way in a worse case scenario it just points you to a part of the source text that's irrelevant instead of giving you answers that are subtly wrong or misleading.
Even then I'm not sure the huge computational requirements make it worth it over ctrl-f or a slightly more sophisticated search algorithm.
Makes me wonder to what degree the longer life offsets the carbon savings from bike commuting.
There are several games already that have some amount of mandatory ray tracing (with no other lighting technology to fall back to to replace it).
IMO the comment above yours is kinda like being in 1996, looking at Quake, and going "it doesn't even look that much better than Duke3D for the massive performance hit polygonal rendering incurs, and you can do room-over-room in Duke's engine anyway, I don't see us switching away from 2.5D games any time in the near future".
I can tell you right now developers are not going to keep doing all the crazy bullshit they need to do to fake light bouncing around now that we can just simulate it. Just like how devs in the 90s didn't want to keep doing all the crazy bullshit they needed to fake the space being 3D when they could work with an actual 3D engine.
Developers are especially not going to want to keep working with two versions of lightning that work completely differently from each other.
Bigotry is as old as time.
Bragging about how vintage your racism is isn't impressive.
It's not a secret, Nvidia publishes white papers about what their technologies are and how they work:
https://research.nvidia.com/labs/rtr/neural_appearance_models/
It seems like everyone in this thread thinks it's like that AI generated Minecraft demo. Though I can't blame them too much since the article is complete shit as well.
Large proteins begin passing through the nephrons en masse and damaging them because they’re too big.
Just in case this isn't enough warning, note that this can permanently reduce your kidney function and harm your body in other ways as well.
I know it's very unlikely for anyone here to do this, but if this happens you aren't just having a "crazy workout" you are giving yourself an injury and possibly permanently affecting your future health. So please don't exercise until you piss blood.
The Xbox 360 had 512 MB of RAM that it shared between its CPU and GPU. I have 128x that amount of RAM in my PC right now. That's the same multiple as the difference between the 360 and the N64.
Imagine calling Crysis “retro”.
This is a video that came out back in 2007. He is using 2x of the highest end GPU you could buy at the time in SLI to run Crysis at 720p with an average of 27 FPS:
https://m.youtube.com/watch?v=PSI9nvIXaF4
Meanwhile here is a demo using the highest end GPU you can buy right now to render a forest at 4K resolution and 60+ FPS (16x more pixels and more than 2x the fps, if we're keeping track):
https://www.youtube.com/watch?v=A7tp4eg0ax8
Most of the maps in Crysis were a few hundred feet across. The forest map in the video above is 4 square kilometers.
Crysis is retro my dude. It is as old now as Super Mario World was when it released.
That's what Google was trying to do, yeah, but IMO they weren't doing a very good job of it (really old Google search was good if you knew how to structure your queries, but then they tried to make it so you could ask plain English questions instead of having to think about what keywords you were using and that ruined it IMO). And you also weren't able to run it against your own documents.
LLMs on the other hand are so good at statistical correlation that they're able to pass the Turing test. They know what words mean in context (in as much they "know" anything) instead of just matching keywords and a short list of synonyms. So there's reason to believe that if you were able to see which parts of the source text the LLM considered to be the most similar to a query that could be pretty good.
There is also the possibility of running one locally to search your own notes and documents. But like I said I'm not sure I want to max out my GPU to do a document search.