Instalar Steam
iniciar sesión
|
idioma
简体中文 (Chino simplificado)
繁體中文 (Chino tradicional)
日本語 (Japonés)
한국어 (Coreano)
ไทย (Tailandés)
български (Búlgaro)
Čeština (Checo)
Dansk (Danés)
Deutsch (Alemán)
English (Inglés)
Español - España
Ελληνικά (Griego)
Français (Francés)
Italiano
Bahasa Indonesia (indonesio)
Magyar (Húngaro)
Nederlands (Holandés)
Norsk (Noruego)
Polski (Polaco)
Português (Portugués de Portugal)
Português - Brasil (Portugués - Brasil)
Română (Rumano)
Русский (Ruso)
Suomi (Finés)
Svenska (Sueco)
Türkçe (Turco)
Tiếng Việt (Vietnamita)
Українська (Ucraniano)
Informar de un error de traducción
That rules out singleplayer, unless the game runs on a server, online-only, but people don't like that, and what happens when the server shuts down? Multiplayer games already cost a lot of money to upkeep, and integrating one of these AI's would cost even more on top of the fees.
Multiplayer has other players, I don't really see the need.
Yeah, it can. The same way Stable Diffusion and DLAA/DLSS work locally on hardware: have a supercomputer create (train) the mathematical framework, the 'model' file, and then use the model file locally. Then the CUDA cores or whatever AMD has do their mathematical associations against the trained model file.
It's the training that takes a supercomputer, as long as what you want out of the end result is narrow enough the end processing most definitely can be done locally.
https://www.youtube.com/watch?v=R0wBMDoFkP0
However, there is only so much of The Binding of Isaac you can play before you start to realize you are simply replaying all of the same rooms, just mixed up in a different arrangement and with a different set of items made available to you on each play through, and I expect its application in other areas to be similar.
Albeit it was ages since I was experimenting with AI, I remember that chat models require a moderately capable machine to run on, and I don't believe they're exactly learning anything. They require better hardware than those image generators, if I recall. You couldn't exactly run a good model on a laptop.
animations
https://m.youtube.com/watch?v=0A2-Af5JEWU
They're not learning anything. They work simply by mathematically predicting what word should come next. That's all. Maybe that says something about human language, but that's how AI does it.
Of course I think the math involves 12,000+ association dimensions per word.
But the Internet is ubiquitous, and most game require an always-on connection these days, anyway.
Better at commercials:
https://youtu.be/I35sW7TtQ_w
https://youtu.be/iCR1OmL-8Ys
https://youtu.be/54kTQfrNi7g
Of course Mr. Puzzles is an instant classic:
https://youtu.be/WZzbxNoMjGM
thats crazy man scary to think what could happen in the future, with the way ai is now and how quickly people are predicting it could improve, and with enhancements in robotic engineering how long before we get a realistic robot or android XD