this post was submitted on 03 Oct 2024
86 points (98.9% liked)

Games

16742 readers
794 users here now

Video game news oriented community. No NanoUFO is not a bot :)

Posts.

  1. News oriented content (general reviews, previews or retrospectives allowed).
  2. Broad discussion posts (preferably not only about a specific game).
  3. No humor/memes etc..
  4. No affiliate links
  5. No advertising.
  6. No clickbait, editorialized, sensational titles. State the game in question in the title. No all caps.
  7. No self promotion.
  8. No duplicate posts, newer post will be deleted unless there is more discussion in one of the posts.
  9. No politics.

Comments.

  1. No personal attacks.
  2. Obey instance rules.
  3. No low effort comments(one or two words, emoji etc..)
  4. Please use spoiler tags for spoilers.

My goal is just to have a community where people can go and see what new game news is out for the day and comment on it.

Other communities:

Beehaw.org gaming

Lemmy.ml gaming

lemmy.ca pcgaming

founded 1 year ago
MODERATORS
nanoUFO@sh.itjust.works
86
DOOM ported to run almost entirely on AMD GPUs (www.gamingonlinux.com)
submitted 1 month ago by RmDebArc_5@sh.itjust.works to c/games@sh.itjust.works
9 comments fedilink hide all child comments
 

cross-posted from: https://sh.itjust.works/post/26097055

you are viewing a single comment's thread
view the rest of the comments
[โ€“] Martineski@lemmy.dbzer0.com 1 points 1 month ago (1 children)

Nope, a neural network:

https://youtu.be/0Xn8xGV_w9w

https://arxiv.org/abs/2408.14837 "Diffusion Models Are Real-Time Game Engines"

https://gamengen.github.io/

We present GameNGen, the first game engine powered entirely by a neural model that enables real-time interaction with a complex environment over long trajectories at high quality. GameNGen can interactively simulate the classic game DOOM at over 20 frames per second on a single TPU. Next frame prediction achieves a PSNR of 29.4, comparable to lossy JPEG compression. Human raters are only slightly better than random chance at distinguishing short clips of the game from clips of the simulation. GameNGen is trained in two phases: (1) an RL-agent learns to play the game and the training sessions are recorded, and (2) a diffusion model is trained to produce the next frame, conditioned on the sequence of past frames and actions. Conditioning augmentations enable stable auto-regressive generation over long trajectories.

[โ€“] ggppjj@lemmy.world 2 points 1 month ago

Wild. Neat!