| Time |
Nick |
Message |
| 00:23 |
|
rubenwardy joined #luanti |
| 00:53 |
|
Tirili joined #luanti |
| 00:55 |
|
mrkubax10 joined #luanti |
| 01:37 |
|
luxalopex joined #luanti |
| 02:14 |
|
anemofilia joined #luanti |
| 04:00 |
|
MTDiscord joined #luanti |
| 04:49 |
|
FeXoR joined #luanti |
| 05:05 |
|
sugarbeet joined #luanti |
| 05:31 |
|
YuGiOhJCJ joined #luanti |
| 05:34 |
|
dnp1 joined #luanti |
| 05:46 |
|
YuGiOhJCJ joined #luanti |
| 07:55 |
[MatrxMT] |
<Blockhead256> core.getmodpath(modname) vs core.global_exists(modname) - any really difference? do ye prefer one or the other? |
| 07:55 |
[MatrxMT] |
<Blockhead256> *get_modpath |
| 08:07 |
sfan5 |
you could argue that checking the global should be preferred in case there's another mod that provides a compatible API |
| 08:08 |
[MatrxMT] |
<Blockhead256> that's probably right, with the small edge case being mods that don't have a global table, and you would know about that.. |
| 08:08 |
sfan5 |
yeah if you're depending on items or so it still makes sense |
| 08:08 |
sfan5 |
could directly check the presence of items tho |
| 08:09 |
[MatrxMT] |
<Blockhead256> always sensible to make more granular checks if it matters |
| 08:11 |
[MatrxMT] |
<Blockhead256> if I wanted to discriminate, say, a mod replacement vs one provided by the base game, I could snoop in the path. A better pattern would probably be to embed a flag in the table for that mod though |
| 08:11 |
[MatrxMT] |
<Blockhead256> can't guarantee that's going to get upstreamed, but it's a pattern for e.g. `farming.fork` |
| 08:12 |
|
cow321 joined #luanti |
| 08:24 |
|
antitheus joined #luanti |
| 08:35 |
|
SoruCoder joined #luanti |
| 08:36 |
|
xor_ghost874 joined #luanti |
| 09:00 |
|
cow321 joined #luanti |
| 09:09 |
|
chloetax joined #luanti |
| 09:23 |
|
YuGiOhJCJ joined #luanti |
| 09:23 |
|
fling joined #luanti |
| 09:23 |
|
est31 joined #luanti |
| 09:49 |
|
pgimeno joined #luanti |
| 09:58 |
|
dv^_^ joined #luanti |
| 11:00 |
|
___nick___ joined #luanti |
| 11:01 |
|
mrkubax10 joined #luanti |
| 11:03 |
|
___nick___ joined #luanti |
| 11:08 |
|
calculon joined #luanti |
| 11:24 |
|
germ_ joined #luanti |
| 12:33 |
|
Tux342 joined #luanti |
| 12:36 |
[MatrxMT] |
<the4spaceconstants> global_exists |
| 12:36 |
[MatrxMT] |
<the4spaceconstants> it's a native boolean and it's a direct check of availability for accessing libraries |
| 12:43 |
|
turtleman joined #luanti |
| 12:58 |
|
Guest38 joined #luanti |
| 12:58 |
|
Tux341 joined #luanti |
| 12:59 |
|
GNUHacker joined #luanti |
| 13:11 |
|
PoochInquisitor joined #luanti |
| 14:28 |
|
Panquesito7 joined #luanti |
| 14:50 |
|
MisterE1230 joined #luanti |
| 15:05 |
|
kamdard joined #luanti |
| 15:44 |
user333_ |
cheapie: look what i did https://user333.sweeux.org/tmp/Screencast_20260824_124006.webm |
| 17:38 |
|
HuguesRoss joined #luanti |
| 18:28 |
|
Mantar joined #luanti |
| 19:09 |
|
crazylad joined #luanti |
| 19:23 |
|
dudz joined #luanti |
| 20:15 |
|
___nick___ joined #luanti |
| 20:15 |
|
antitheus joined #luanti |
| 20:38 |
|
kaylie joined #luanti |
| 21:17 |
|
mrkubax10 joined #luanti |
| 21:34 |
|
BuckarooBanzai3 joined #luanti |
| 21:43 |
|
lemonzest joined #luanti |
| 21:52 |
|
BuckarooBanzai0 joined #luanti |
| 22:00 |
|
BuckarooBanzai joined #luanti |
| 22:03 |
|
BuckarooBanzai joined #luanti |
| 22:06 |
|
pinapelz joined #luanti |
| 22:33 |
|
panwolfram joined #luanti |
| 22:59 |
cheapie |
user333_: of course you did |
| 23:00 |
cheapie |
Is that just hooked up to a separate QEMU instance? |
| 23:12 |
cheapie |
In other news, on a SiS 550 Luanti is.... slow (slower than your RPi). With plain MTG it's not unplayable but the mapgen sure takes its time |
| 23:13 |
technomancy |
minetest the gathering??? |
| 23:13 |
technomancy |
(jk) |
| 23:16 |
|
book` joined #luanti |
| 23:20 |
user333_ |
<cheapie> Is that just hooked up to a separate QEMU instance? | no, it's a generic RISC-V emulator in lua slightly modified for luanti |
| 23:20 |
cheapie |
ah |
| 23:20 |
cheapie |
also https://cheapiesystems.com/media/2026-08-24%2018-14-37.webm |
| 23:20 |
user333_ |
i had to do a few workarounds for it to work, but hopefully i can get it in a formspec |
| 23:21 |
cheapie |
Not too bad IMO for Luanti running on what effectively amounts to a 25-year-old DVD player control chip |
| 23:21 |
user333_ |
o.o |
| 23:22 |
[MatrxMT] |
<Blockhead256> could probably run https://www.youtube.com/watch?v=p-k5MPhBSjk on some seriously low end stuff |
| 23:22 |
cheapie |
Although at about 2:00 you can see something it tends to do on single-core CPUs like this, where everything runs fine in already-generated mapblocks but generating new ones stalls for a while |
| 23:22 |
technomancy |
is this the endgame of sandboxing; just give every mod their own entire operating system? |
| 23:23 |
user333_ |
i was JUST about to link that Blockhead |
| 23:23 |
cheapie |
It would probably help if I could use LuaJIT, but that seems to emit later (686?) instructions that aren't supported by this CPU |
| 23:24 |
[MatrxMT] |
<Blockhead256> :P |
| 23:24 |
cheapie |
This is only a 586, it's based on the core from the Rise mP6 :D |
| 23:25 |
user333_ |
older than anything i have.. |
| 23:25 |
* cheapie |
wonders what's next... 486? |
| 23:25 |
user333_ |
i wonder if i could run luantiserver from my RPi Pico |
| 23:26 |
cheapie |
Unfortunately I don't think I can do it on my 386 (not enough RAM, and also the OS I'm using for these shenanigans needs a 486 or higher), but I do have a 486SX-25... |
| 23:28 |
user333_ |
crosscompile! |
| 23:28 |
cheapie |
I mean the 386 box doesn't even support enough RAM for the OS to /boot/ |
| 23:28 |
user333_ |
oof |
| 23:29 |
user333_ |
try a really old (2.6x) linux perhaps |
| 23:29 |
cheapie |
Maybe, but getting Luanti to compile on that sounds... painful at best |
| 23:30 |
cheapie |
Speaking of which, with the final settings I ended up with, Luanti (server only) took about 12 hours to compile on the SiS 550 |
| 23:30 |
user333_ |
for once my RPi1 wins by twice as much |
| 23:31 |
cheapie |
I think I mentioned before, but the OS in use here was NetBSD |
| 23:31 |
user333_ |
i hear netbsd has good support for old hardware |
| 23:31 |
cheapie |
Most Linux distros don't ship a 32-bit version anymore and most of the ones that do need 686+, but NetBSD's system requirements seem to still be along the lines of "computer" |
| 23:32 |
cheapie |
Supposedly you can still install it on a VAX-11/780 from 1977, as long as you have enough RAM |
| 23:33 |
user333_ |
now THAT is impressive |
| 23:33 |
cheapie |
It needs about 32MiB of RAM though, and my 386 box (which is actually not supported anyway, it needs a 486) is a 386SX so it can only address 16MiB |
| 23:36 |
* cheapie |
briefly wonders what sort of lag Dreambuilder on a 486 would end up with |
| 23:36 |
[MatrxMT] |
<Blockhead256> dreambuilder? not enough RAM |
| 23:37 |
cheapie |
something something swap something |
| 23:37 |
user333_ |
apparently my main PC can compile luantiserver in 20 seconds, so.. 108000% faster than my RPi1 if i did the math right |
| 23:37 |
crazylad |
<user333_> ...but hopefully i can get it in a formspec | iirc that would require a lot of formspec updates |
| 23:37 |
[MatrxMT] |
<Blockhead256> client timeouts I think... |
| 23:37 |
user333_ |
yes, a LOT |
| 23:37 |
user333_ |
i had to do some hacks to get it to look right in chat |
| 23:37 |
cheapie |
Besides, 486es can theoretically address up to 4GiB of RAM, although I don't think any boards go quite that high :P |
| 23:38 |
user333_ |
should i run a server from my linux phone for fun |
| 23:38 |
technomancy |
I ran a social media server on my phone once |
| 23:38 |
technomancy |
well |
| 23:38 |
technomancy |
twice |
| 23:38 |
cheapie |
Someone should get that Dmitry guy to get Luanti running on a 4004 |
| 23:39 |
Sheriff_U3 |
user333_: yes!!! that way we can try to break the mod sandbox and send text messages with your phone!!! lol |
| 23:39 |
user333_ |
good luck :D |
| 23:40 |
user333_ |
first i have to do some routing to actually make the server accessible from the internet |
| 23:43 |
cheapie |
I might have to see if I can find some way to make this one accessible |
| 23:53 |
|
wallabra joined #luanti |
