I'm still constantly baffled by just how absolutely beyond shit modern computers are
Discussion
Loading...
Post
@OpenComputeDesign x86 was a mistake? :)
@OpenComputeDesign
16 bit was a mistake?
transistors were a mistake?
how modern are we talking? :)
16-bit/early-32-bit was my favorite era. (Basically, the #68k era ;)
Computers were just becoming capable, but not too big for their britches.
@OpenComputeDesign @rl_dane I had a good time with my first amd athlon 64 but sure, simpler times :)
I think computers were honestly better when they were limited to absolutely no more than 1GB RAM, no more than 256 colors, and no more than 1024x768 screen resolution.
1GB RAM: no LLMs
256 colors: no horrid low-contrast soupy interfaces
XGA Resolution: no horrid empty spaces and bloated interfaces
I keep wanting to make that as an OS 😄
(If only I had the skillz)
1+ more replies (not shown)
Yeah, and the pushback I get from statements like that is insane to me.
"But we don't want to go back to Windows 95."
I don't either, it was a crap OS, but the interface was better than the crap interfaces they're shipping today, so ?!?!????!?
I'd rather w95 with its software suite and interface than w11 with its.
W11 is a worse OS than w95 was.
@pixx @OpenComputeDesign @kabel42
It does have memory protection, though. That was Windows 95's most glaring weakness.
Edit: I meant to say that it doesn't. derp.
Edit2: No, I was saying that W11 has memory protection. lol
@rl_dane
Meh. Memory protection means i need preemptive scheduling instead of cooperative.
I'd rather a cohesive system with cooperative scheduling (with maaaaybe overrides for audio but, really, I'd rather require a 2 core minimum and use one as a hard real time processor)
@pixx @OpenComputeDesign @kabel42
Whaaaat, why would you want cooperative scheduling? That means one application crash takes down the whole OS, because it never returns control.
@rl_dane
Because i want applications that don't SUCK
And i want a design that requires competency.
@pixx @OpenComputeDesign @kabel42
Brofam, I lived through the cooperative years.
Wild horses couldn't drag me back there.
Having your entire OS go belly-up because StuffIt Expander stuffed itself was not fun, and stuff like that happened a lot.
Just imagine what a modern web browser could do to a cooperatively-multitasked OS, YE FLIPPING GODS!
#FreeBSD can barely handle heavy sites on Firefox without hiccups as it is!
@OpenComputeDesign @pixx @kabel42
I didn't use MacOS 9 much (6.0.4-7.0.1, Leopard to High Sierra), but to my memory, classic MacOS was not graceful when you ran out of RAM. I don't even know that it had anything approaching an OOM-killer.
Well, I guess you really missed out, then. Because I've actually had pretty good experience with 9. _Much_ better than I had using much newer macs running, idk whatever version of 10 was out in the late 2010s. Was my first experience with mac, and it really turned me off the platform.
Although, I've recently gotten an iBook with a very early version of 10, and it seems like a pretty sweet OS, too.
@OpenComputeDesign @pixx @kabel42
My impression was that MacOS X got better and better until Lion. The early versions had performance and stability issues.
OS 9 was still a 1980s OS at its core, and was showing its age, badly.
It had some advanced features like preemptive multitasking, but only within the finder. The APIs weren't significantly changed since 1984 and it could still be quite unstable if you ran poorly-written software.
@OpenComputeDesign @pixx @kabel42
NO.
tmp $ cat ocds-fubar.c
#include<stdio.h>
int main() {
unsigned long int x=0;
while ( 1 ) {
x++;
printf( "%lu\n", x );
}
return( 1 );
}
tmp $ gcc -Wall -O ocds-fubar.c -o ocds-fubar
tmp $ wc -c ocd*
15960 ocds-fubar
145 ocds-fubar.c
16105 total
tmp $ time ./ocds-fubar
1
2
3
4
5
(...etc...)
Meanwhile:
tmp $ uptime
10:10:43 up 6 days, 22:49, 1 user, load average: 0.16, 0.36, 0.47
tmp $ uptime
10:10:45 up 6 days, 22:49, 1 user, load average: 0.39, 0.41, 0.48
tmp $ uptime
10:10:46 up 6 days, 22:49, 1 user, load average: 0.39, 0.41, 0.48
tmp $ uptime
10:10:47 up 6 days, 22:49, 1 user, load average: 0.39, 0.41, 0.48
tmp $ uptime
10:10:49 up 6 days, 22:49, 1 user, load average: 0.39, 0.41, 0.48
tmp $ uptime
10:10:50 up 6 days, 22:49, 1 user, load average: 0.68, 0.47, 0.50
tmp $ uptime
10:10:52 up 6 days, 22:49, 1 user, load average: 0.68, 0.47, 0.50
tmp $ uptime
10:10:53 up 6 days, 22:49, 1 user, load average: 0.68, 0.47, 0.50
tmp $ uptime
10:10:54 up 6 days, 22:49, 1 user, load average: 0.94, 0.52, 0.52
tmp $ uptime
10:10:56 up 6 days, 22:49, 1 user, load average: 0.94, 0.52, 0.52
tmp $ uptime
10:10:57 up 6 days, 22:49, 1 user, load average: 0.94, 0.52, 0.52
tmp $
@OpenComputeDesign @pixx @kabel42
The code you specified could not possibly eat up RAM, unless it was implemented in an exceptionally crappy language.
@OpenComputeDesign @pixx @kabel42
If you were having a memory leak in C, it would've been your own darn fault. 😆
tmp $ ps aux |(head -1; grep "[o]cds-fubar")
USER PID %CPU %MEM VSZ RSS TTY STAT START TIME COMMAND
rld 3129287 99.9 0.0 2560 1408 pts/5 R+ 10:21 3:44 ./ocds-fubar
tmp $ time ./ocds-fubar |wc
^C
real 3m56.722s
user 5m44.331s
sys 0m30.909s
tmp $
@OpenComputeDesign @rl_dane @pixx last time i tried to create a 10TiB array in python it just killed that python instance 🤷
@OpenComputeDesign @rl_dane @pixx never had problems with that. And i've crashed win9x machines by being in the same room
Computers should have been magically capped at 1GB RAM.
That would've been utopian.
(What do you mean, "'utopic' isn't a word?!?" It should absolutely be a word!)