Cool I didn’t know llamacpp could do gpu acceleration at all. I’m going to look into that.

I think I might be on to something that contributes to the problem. The built-in “KoboldGPT chat” option puts some example queries in its context memory. They aren’t very long responses so I think it’s just seeing that and using it as a guideline for what to say which results in shorter answers.

If I use the “new chat” option instead of “KoboldGPT chat”, it makes it so that nothing is in the context. No prompt and no memory. This way when I tell it to write 500 words of crap, it doesn’t quite write that much but it’s a lot better than before. Pressing enter to make it generate more text works more often this way too.

I don’t write documentation because all my shit is always so broken that the source code and error messages are the documentation.

I just commit and push the entire contents of my home folder and let people figure it out for themselves

Coqui xtts is broken right now, at least on Debian. I did not try xtts2. Yet turns out I already tried that one. After my computer is done being frozen from my latest hacking attempt to make a borked tts python maybe work, I’ll add whatever the error is.

The docker installation for the first one is broken just FYI but I guess I should be trying to use whichever one is the most up to date for higher chances of success.

I kind of don’t really care anymore. Fuck all this broken shit. I’m sure whichever leading github repository that does this same thing a couple years from now will be easier to install so I’ll just wait until then to mess with it.

The real question is: did it work?

I use ipv6 when possible but it’s rarely possible. I’ve never had home internet that was ipv6 ready enough for my wan address when googling “what’s my ip” to be something besides an ipv4 number.

Could I get ipv6 over otherwise non ipv6 compatible hardware using a vpn?

Another part of it is the gpu bios. The gpu bios contains x86 opcodes that it expects the host system to run for gpu-specific functions like video mode switching and probably lots of other stuff. I know that Vesa bios extensions mode switching requires a pointer to the functions in the gpu bios which the cpu runs. I tried to make a platform independent Vesa driver one time and couldn’t figure out how to circumvent using the gpu bios for it since the functions you’re supposed to call are compiled for x86. Even the well-refined projects like Seabios still rely on the VBE pointers for non-legacy video modes.

Legacy vga does also has a bios but it’s relatively not that difficult to circumvent using the bios on legacy vga cards, only issue is that legacy vga modes are mostly useless.

I think there’s a newish way of doing this stuff that doesn’t involve Vesa or legacy vga but I don’t know what it is. This I’m sure is only one of the many problems that have to be overcome if someone wanted to hack a 1080ti onto a raspberry pi or something.

I don’t want arm or risc v to replace x86 unless we still get to have motherboards with upgradable cpu sockets. And ram slots. And gpu compatible pcie slots. None of the current players are going to give us anything like that.

I’d be happy if they even sold hobbiests loose chips so we could figure it out ourself though. The arm and risc microprocessors on mouser are like the bottom of the barrel leftovers after the corpos have had their pick.

What would be the possibility of there someday existing arm or risc v boards with pci-e slots that can take graphics cards? I don’t know that steam even lets developers make a arm linux version of their game or not, but in theory you’d be able to play open source compilable stuff from github.

I installed void linux which I guess defaults to wayland and everything ran like crap until I switched to x11. Have fun with that.

sjajvxuwjdofgwu

AjsgGhS77bndugxg

gehshagfahcdvwjdvwjd

AjsgGhS77bndugxg (2)

New file

New file (2)

New file (3)

New file (4)

2020 has become the decade of reading books. Search results these days are so bad.

inb4 they wait until his last day then roll back the changes because functional code/unauthorized changes are against company policy and actually they need that bug to slow down the user so they don’t click so fast the database crashes.