I backed up all my files.
I made USB boot disks of various #BSDs.
I made a clonezilla clone of the entire disk.
I booted into the #FreeBSD disk and was about to wipe out #OpenBSD.
At the last minute, I decided to go through pt.4 of @rootbsd's OpenBSD setup video to see if I could squeeze any more performance out of it.
No better, as far as I could tell.
I went through OpenBSD Guy's (YT) OpenBSD performance video.
Still 25+ seconds to open up Firefox. :'(
But...

...

...

I just can't wipe it.
I like #OpenBSD too much, even if it runs like maple syrup candy.

...

Show thread

...

I don't want to abandon a BSD who's devs *actually* freaking use it on their home machines, and who don't do presentations on "unix bad, bsd irrelevant, linux good, systemd great" from their freaking macbooks (yes, I'm summarizing/lampooning one particular FreeBSD dev's presentations. I won't say who. They have a right to their opinion, and they are an intelligent, decent human being, as far as I know, and I wish them the best)

...

Show thread

...

I don't want to leave a BSD that has excellent manpages, freaking SENSIBLE config files, and a focus on elegance.

I'm just going to pretend that my Core 2 Duo Thinkpad X200 is a Pentium III T23 until I can get a faster box to play with #OpenBaSeD on. ;)

Show thread
Follow

@RL_Dane Have you ever considered trying your hand at making your own OS ;)

@golemwire

Ermagherd, I haven't even set up Arch, Gentoo, or LFS, lol.

Actually, yes. I would like to.

But in #uxn, not native. Assuming I scrape together enough time to actually learn the assembly language. But I know I will.

@RL_Dane I've dabbled in UXN. I was thinking about making an OS for UXN, but 64KiB and other limitations of it have turned me against the idea of implementing my dream OS in it. I've had ideas for a while....

On and off I work on my own instruction set architecture. UXN was the final boost that got me to get it to a semi-reasonable place.

Perhaps I'll finish up that ISA's design and implement a VM for it. Then, I could make a little OS for it.

@RL_Dane And taking it further, I could implement it as an EFI executable and run it on bare metal. Sure, the actual OS would be interpreted but it would be really neat.

(At "implement it" I was talking about the VM)

@RL_Dane Another thing. I wonder what it would be like to run UXN (or other homebrewn small ISAs) code on an FPGA?

@golemwire

I'm not sure you'd get much out of dedicated hardware. The #uxn core is EXTREMELY tight.
The main loop is like 100 lines of C. It's insane.

It was designed to be easy to implement first, and fast second.

@RL_Dane I saw that - 100 lines; unbelievable.
Still though... something about running something as small, old-school and implementable as the UXN CPU on bare hardware... is so perfect

@golemwire

I'm curious if anyone's made a vm for the 6502.

The parallels between the 6502 and #uxn are pretty stark. Probably on purpose ;)

@RL_Dane
Probably. I think the CHIP8 can run on a 6502. Or if you meant emulating the 6502, there are loads of emulators for 8bit micros that used the 6502...
@golemwire@social.librem.one

@dheadshot

No, definitely running *on* the 6502.

I'm amazed that the 6502 can emulate anything.

Heck, I'm amazed that the 6502 can *do* anything! Not being mean, that's just such a limited chip.

I "grew up" in the 16-bit era, and while I used the 8-bit machines for many years before that, I didn't really understand what they were doing at the time.

@golemwire

An EFI OS makes me think of the MBR games that the 8-bit guy (YT) showed off a few months ago.

As far as "intepreted" goes, dynamically-transpiling emulators is a relatively recent invention. No shame in being interpreted ^__^

@golemwire

That's super cool!

The way around the #uxn/#varvara 64KiB limit is to dynamically load modules. I don't know if you can load a file into a *part* of memory or if has to overwrite from $0, but IIRC that's how @neauoire gets around the limitation for their website.

...

@golemwire

...

I'm also thinking that at some point I'll write .TGA files as a way to have hi-color output from a #uxn program, similar to how really early Macs (supposedly) used video adapter boxes that "printed" color images onto a color monitor via the serial port.

It would be cool to have a #uxn program that approximated output in four colors, but wrote hi-color or 8-bit greyscale files, kind of like how "Digital Darkroom" let you edit 7-bit grey images on a 1-bit screen.

@RL_Dane @golemwire Potato is a pretty well rounded OS, and it's 20kb.

wiki.xxiivv.com/site/potato.ht

The application API is relatively simple and applications are at most 5kb so far, even the whole window manager is just 6kb.

Oquonie.rom is 700kb, but it makes use of the expansion loader to bring data in and out of addressable range.

@golemwire

A couple years ago, I wanted to buy a "The C64" and bootstrap my own toolchain and eventually OS in assembly starting with nothing more than Commodore Basic (POKEing hand-assembled opcodes into RAM from basic).

#UXN is scratching that itch, and I don't have to suffer with 40-column text. XD

@RL_Dane I get the feeling. Even though I'm a, uh, young whippersnapper, I've messed around with old computers and I miss them. Just raw computer, neither backdoored nor "smart".

@golemwire

Old souls that love old computers are excellent peeps, of any age. ;)

Sign in to participate in the conversation
Librem Social

Librem Social is an opt-in public network. Messages are shared under Creative Commons BY-SA 4.0 license terms. Policy.

Stay safe. Please abide by our code of conduct.

(Source code)

image/svg+xml Librem Chat image/svg+xml