• 80 Posts
  • 79 Comments
Joined 1 year ago
cake
Cake day: September 7th, 2023

help-circle









  • 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.





























  • I’m not sure how accurate this info from the modder is but this is how he describes it under his mod:

    Experienced players can predict that the pipes will always malfunction 2 or 3 times, first time around 35%, second time around 70%, and maybe a third time after 90%. Generally, it’s every 50-70 seconds (average of 60 seconds).

    This simple mod shakes things up by randomizing when the pipes will break - while still maintaining the same average time between breakdowns. Now pipes can jam anywhere from 5 to 115 seconds between breakdowns (average of 60 seconds). So you never know what you’re gonna get!

    Mod: https://mod.io/g/drg/m/random-refinery-breakdowns

    This is the only instance of it being described that I’m aware of because I was looking up refinery mods in the past. I tried looking up this info on the web to answer your question with more credible sources but I didn’t find any and didn’t bother spending too much time on that.