-
Notifications
You must be signed in to change notification settings - Fork 148
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Expected performance on a consumer GPU? #6
Comments
Yeah about the same performance on my RTX 3070Ti laptop |
around 14s/it on an rtx 3060 12GB |
I really wonder what kind of server GPU they have to generate the game so fast |
Well 4090 doesn't seem like enough lol, maybe 6090? Maybe then someone can make UI like on the website to control it in real time. But I'm kinda confused why it runs so slow, Stable Diffusion is bigger and faster |
H100 or the brand new "ASIC" (Etched’s upcoming Sohu chip.). |
Even for 30fps you have to generate a frame in 33ms, which is like 200x faster than OP's 4090... |
Made a fork, and a playable pygame, at 16 steps half precision, compiled it gets up to 9fps on a 4090. |
Trying this on an NVIDIA RTX 4090 giving me around 6~7s/it, and generating 32 frames takes around 3 minutes. Is this the expected performance?
The text was updated successfully, but these errors were encountered: