| Time |
Nick |
Message |
| 00:16 |
|
proller joined #minetest |
| 00:27 |
|
Noisytoot joined #minetest |
| 00:54 |
|
Hawk777 joined #minetest |
| 01:12 |
|
mazes_83 joined #minetest |
| 02:11 |
|
Verticen joined #minetest |
| 02:36 |
|
mmuller joined #minetest |
| 02:45 |
|
illwieckz joined #minetest |
| 02:53 |
|
YuGiOhJCJ joined #minetest |
| 02:58 |
|
specing_ joined #minetest |
| 03:09 |
MTDiscord |
<josiah_wi> IceDragon, did you get the CMake part behaving nicely? |
| 03:10 |
IceDragon |
Kinda |
| 03:10 |
IceDragon |
I got it to compile on my system, but I had to bump the c++ requirement up to c++17, so that ended up breaking older compilers |
| 03:11 |
MTDiscord |
<josiah_wi> I volunteer a lot of CMake assistance, so if there are any questions I enjoy that stuff. |
| 03:11 |
MTDiscord |
<josiah_wi> Is the C++17 a LuaU requirement? |
| 03:12 |
IceDragon |
It's required for some of the templates they used string_view and optional |
| 03:13 |
IceDragon |
Which is required for the parser |
| 03:13 |
IceDragon |
Which is required for the compiler, so you can actually get lua code into the vm |
| 03:13 |
MTDiscord |
<josiah_wi> If by some miracle none of the headers require C++17 you don't need it for Minetest. Keyword miracle. |
| 03:13 |
MTDiscord |
<josiah_wi> afaik |
| 03:13 |
IceDragon |
C++ has been fresh out of miracles for years I think |
| 03:14 |
IceDragon |
I fear the standard now, because I didn't even know they got to 17 |
| 03:14 |
IceDragon |
last time I seriously used C++ was in the early stages of 11 |
| 03:15 |
IceDragon |
when the std flag was cxx11 |
| 03:15 |
MTDiscord |
<josiah_wi> They're at C++20 now... but I learned C++11. |
| 03:15 |
IceDragon |
Anyway I'm off on a tangent, here is my branch https://github.com/IceDragon200/minetest/tree/luau-support |
| 03:15 |
IceDragon |
It will probably compile for you, but don't expect it to actually start |
| 03:16 |
IceDragon |
It needs a "dofile" implementation to load other lua files |
| 03:17 |
MTDiscord |
<Jonathon> does it have loadfile? |
| 03:17 |
MTDiscord |
<Jonathon> if so dofile is easy |
| 03:19 |
IceDragon |
It had zero file loading out of the box |
| 03:19 |
MTDiscord |
<Jonathon> seems not https://luau-lang.org/compatibility |
| 03:19 |
MTDiscord |
<Jonathon> >removed for sandboxing, no direct file access this reason is crap/lazy |
| 03:20 |
IceDragon |
I hacked one in https://github.com/IceDragon200/minetest/blob/luau-support/src/mt_lua.cpp#L84-L102 |
| 03:20 |
MTDiscord |
<josiah_wi> I don't know why I'm trying to read this on GitHub's mobile page. I will investigate this tomorrow, and maybe I can PR a LuaL target. |
| 03:20 |
IceDragon |
So maybe you could use that |
| 03:20 |
MTDiscord |
<josiah_wi> Unless LuaU has a target in which case not using it is missing out. |
| 03:21 |
MTDiscord |
<josiah_wi> For example that would've (unless they did it wrong) automatically fixed the C++17 thing |
| 03:23 |
IceDragon |
If they did, I wouldn't have to load the compiler/parser headers in yep |
| 03:23 |
IceDragon |
But they didn't so... suffering for me |
| 03:24 |
MTDiscord |
<josiah_wi> I'll consider whether I have time and maybe open an issue on their repo and PR the target myself lol. |
| 03:25 |
IceDragon |
You can't open issues on their repo |
| 03:25 |
IceDragon |
I tried today to report the conflicting luaL_error implementation |
| 03:26 |
IceDragon |
You can only open one if you're a contributer |
| 03:27 |
MTDiscord |
<josiah_wi> Bummer. Time to fork. (/s) |
| 03:29 |
|
queria joined #minetest |
| 03:33 |
|
queria joined #minetest |
| 03:51 |
|
grouinos joined #minetest |
| 04:00 |
|
MTDiscord joined #minetest |
| 04:57 |
|
IceDragon joined #minetest |
| 04:58 |
|
Extex joined #minetest |
| 05:01 |
|
riff-IRC joined #minetest |
| 05:09 |
|
jonadab joined #minetest |
| 05:19 |
|
YuGiOhJCJ joined #minetest |
| 06:12 |
|
Sven_vB joined #minetest |
| 06:54 |
|
MTDiscord1 joined #minetest |
| 06:55 |
|
YuGiOhJCJ joined #minetest |
| 07:00 |
|
valhalla joined #minetest |
| 07:52 |
|
TomTom joined #minetest |
| 08:05 |
|
hendursa1 joined #minetest |
| 08:14 |
|
CWz joined #minetest |
| 08:19 |
|
valhalla joined #minetest |
| 08:19 |
|
hendursa1 joined #minetest |
| 08:51 |
|
search_social joined #minetest |
| 09:18 |
|
specing_ joined #minetest |
| 10:11 |
|
grouinos joined #minetest |
| 10:21 |
MinetestBot |
[git] sfan5 -> minetest/minetest: Fix typo in buildbot scripts 6db9147 https://git.io/JPA9I (2021-11-05T10:12:56Z) |
| 10:22 |
sfan5 |
IceDragon: MT has its own dofile implementation for sandboxing btw |
| 10:22 |
sfan5 |
but it could be that the mainmenu or such relies on the default implementation |
| 10:53 |
MTDiscord |
<exe_virus> Don't worry, C++ is improving steadily, once we all understand templates, operator overloads, and they fix the dang dependency system I think it'll feel like a brand new swiss army knife |
| 10:53 |
|
AristotIe joined #minetest |
| 11:23 |
|
olliy1or joined #minetest |
| 11:49 |
|
Sven_vB_ joined #minetest |
| 12:01 |
|
dzho joined #minetest |
| 12:06 |
|
calcul0n_ joined #minetest |
| 12:33 |
|
tech_exorcist joined #minetest |
| 12:37 |
|
longerstaff13 joined #minetest |
| 13:01 |
|
proller joined #minetest |
| 13:12 |
|
Sven_vB joined #minetest |
| 13:28 |
IceDragon |
sfan5: So the mainmenu operates outside the sandbox? |
| 13:37 |
|
specing joined #minetest |
| 13:38 |
rubenwardy |
Correct, the mainmenu isn't sandboxed currently |
| 13:39 |
MTDiscord |
<josiah_wi> IceDragon, luau not being found if enabled should be a fatal error, no? |
| 13:40 |
IceDragon |
It should be, if it was required, otherwise it should fallback to the shipped lua (minetest has a copy) |
| 13:40 |
IceDragon |
if you want to force it you would enable REQUIRE_LUAU=1 |
| 13:40 |
|
tech_exorcist joined #minetest |
| 13:40 |
MTDiscord |
<josiah_wi> Would it be alright if I PR some CMake quality changes? |
| 13:46 |
|
tech_exorcist joined #minetest |
| 13:48 |
IceDragon |
Sure thing, I don't mind |
| 13:49 |
IceDragon |
Though you could probably submit them upstream, my little branch I don't think I'll persue it further unless I rekindle my interest in luau |
| 13:49 |
IceDragon |
Seeing as I'd have to replace everything io at this point |
| 13:55 |
|
est31 joined #minetest |
| 14:01 |
|
absurb joined #minetest |
| 14:01 |
|
Fixer joined #minetest |
| 14:04 |
MTDiscord |
<josiah_wi> Maybe someone will pick it up where you left off, who knows. |
| 14:11 |
|
olliy joined #minetest |
| 14:28 |
MTDiscord |
<josiah_wi> LuaJIT headers are found instead of Luau headers. This is a minor issue. |
| 14:31 |
IceDragon |
That was a weird one as well, even with the prefix set, it still picked up luajit headers instead of Luau, I find clearing the build resolves the issue |
| 14:34 |
|
Lunatrius joined #minetest |
| 15:25 |
|
erle joined #minetest |
| 15:52 |
|
Extex joined #minetest |
| 16:01 |
|
c2a1 joined #minetest |
| 16:31 |
|
tech_exorcist joined #minetest |
| 16:32 |
|
garywhite joined #minetest |
| 16:32 |
|
garywhite joined #minetest |
| 16:41 |
|
garywhite joined #minetest |
| 16:41 |
|
garywhite joined #minetest |
| 16:44 |
|
Talkless joined #minetest |
| 16:48 |
|
kamdard joined #minetest |
| 16:55 |
|
Verticen joined #minetest |
| 17:03 |
|
fluxionary joined #minetest |
| 17:10 |
|
grouinos joined #minetest |
| 17:11 |
|
Sven_vB_ joined #minetest |
| 17:17 |
|
tech_exorcist joined #minetest |
| 17:35 |
|
___nick___ joined #minetest |
| 17:35 |
|
tech_exorcist joined #minetest |
| 17:36 |
c2a1 |
Rip |
| 17:42 |
|
riff_IRC joined #minetest |
| 17:42 |
|
flok joined #minetest |
| 17:43 |
flok |
what is the latest software to import java minecraft server maps into minetest? |
| 17:45 |
sfan5 |
mcimport is your best bet I guess |
| 17:45 |
flok |
that one failed with KeyError: 'Blocks' |
| 17:50 |
|
GNUHacker joined #minetest |
| 17:52 |
|
tech_exorcist joined #minetest |
| 18:00 |
|
tech_exorcist joined #minetest |
| 18:13 |
|
___nick___ joined #minetest |
| 18:15 |
|
___nick___ joined #minetest |
| 18:15 |
|
kamdard joined #minetest |
| 18:26 |
|
definitelya joined #minetest |
| 18:26 |
erlehmann |
luatic fun brain teaser, can you figure out why my patch to use 16bpp textures produces slightly more reddish browns but leaves other colors visually indistinguishable? https://git.minetest.land/MineClone5/MineClone5/issues/39#issuecomment-28082 |
| 18:34 |
|
moongaia joined #minetest |
| 18:54 |
|
GNUHacker joined #minetest |
| 19:02 |
|
GNUHacker joined #minetest |
| 19:06 |
|
tech_exorcist_ joined #minetest |
| 19:20 |
|
Sven_vB joined #minetest |
| 20:45 |
|
mdk left #minetest |
| 21:02 |
|
grouinos joined #minetest |
| 21:10 |
|
Sven_vB_ joined #minetest |
| 21:16 |
|
clavi joined #minetest |
| 21:16 |
|
clavi joined #minetest |
| 21:23 |
|
Fixer_ joined #minetest |
| 21:26 |
|
jfindlay_ joined #minetest |
| 21:31 |
|
plate[m] joined #minetest |
| 21:39 |
|
proller joined #minetest |
| 21:40 |
|
longerstaff13 joined #minetest |
| 21:43 |
|
iliekprogrammar[ joined #minetest |
| 22:08 |
MTDiscord |
<luatic> A lossy reduction from bit depth 8 to 5 should be paired with proper rounding (you're currently doing "bit shifting" by dividing and flooring afterwards) and dithering (see https://github.com/appgurueu/voxelizer/blob/master/dithering.lua for a reference implementation). It looks like you aren't seriously pursuing a bit depth of 5 though? Regarding the tone change of browns, I see three possible explanations: (1) Irrlicht messes up while |
| 22:08 |
MTDiscord |
upscaling bit depth, as you have indicated (though usually - if simply bit shifting instead of replicating the leftmost bits - it would be the other way around, meaning less reddish browns); (2) Some weird color space properties (I'd expect green to be most significant though, as it is concerning luminance); or (3) A bug in your encoder. |
| 22:12 |
|
Verticen joined #minetest |
| 22:14 |
erlehmann |
luatic the encoder needs to be fast, therefore it has to be single-pass |
| 22:14 |
erlehmann |
luatic actually, i'll just do 24 bit palette encoding, but i still don't see why the colors are off. pro the rounding. |
| 22:14 |
erlehmann |
prob |
| 22:15 |
erlehmann |
luatic fleck already noticed that all the colors get slightly darker |
| 22:15 |
erlehmann |
> It looks like you aren't seriously pursuing a bit depth of 5 though? |
| 22:15 |
erlehmann |
if it is visually indistinguiashable, i'd like to, hehe |
| 22:17 |
|
Sven_vB joined #minetest |
| 22:19 |
MTDiscord |
<luatic> See the PNG spec for a better formula (which is the one my PNG encoder uses when rescaling 16 bit to 8 bit): https://www.w3.org/TR/2003/REC-PNG-20031110/#13Sample-depth-rescaling |
| 22:29 |
|
Extex joined #minetest |
| 22:30 |
erlehmann |
luatic thx |
| 22:55 |
|
calcul0n__ joined #minetest |
| 22:56 |
|
c2a1 joined #minetest |
| 23:00 |
|
Sven_vB joined #minetest |
| 23:16 |
|
illwieckz joined #minetest |
| 23:20 |
|
Verticen joined #minetest |
| 23:21 |
|
proller joined #minetest |
| 23:34 |
|
illwieckz joined #minetest |
| 23:58 |
|
fluxionary joined #minetest |
