Derg

Don’t forget the value of actually knowing your codebase if you wrote it… or the value lost if you just have an LLM do the whole thing.

Cross products! If you take the cross product of two edges of a face, you get a perpendicular vector for that face. It shouldn’t be too bad to do 20 of those per frame to find if a face is level with the ground (cross product points straight up and/or down depending on how you’re doing them)… now I don’t know about the velocity/acceleration numbers (including rotation)…

If you use something like four bytes instead of one, your error drops dramatically... If you're generating numbers from 0-100, You'd have ~50 "bad" cases (just imagine the remainder is the same) out of 4 billion instead of ~50 out of 256. Unless you're running a casino or something (and even then...) it's probably "fine enough" but of course you have to know about it!

I think you’re coming from the right place but I was trying to make a metaphor there. I do appreciate the sentiment. And if you caught it… I really really do wish I had the guts to leave but it seems we’re gonna be together for a while yet… not exactly the ideal situation for a bad relationship

Yeah I’m kinda with an abusive girlfriend but when I talk about the vile shit she’s done people always say that she’s living rent free in my head. Am I wrong to be thinking about how much she’s fucking me over? Should I just ignore it? We’re still together.

You can, as of this very moment, download a 16 GB LLM and run it on your local computer. Or any bigger version, if you happen to have a beefy computer. LLMs are software. Good luck banning software. This is the internet.

ALSO, even if you do happen to be able to ban AI training facilities and such in the U.S., and those models become outdated, you can even go download some Chinese models too! They are not going away.