Time |
Nick |
Message |
00:12 |
|
pgimeno joined #luanti-dev |
00:19 |
|
SFENCE joined #luanti-dev |
00:49 |
MTDiscord |
<nathan4220776> I'm pretty sure that a possible way to deal with this is to /emerge twice. |
00:49 |
MTDiscord |
<nathan4220776> The second emerge operation will probably go much faster... |
00:50 |
MTDiscord |
<nathan4220776> But yes, this would be a wonderful thing for... someone else to fix. ;) |
00:51 |
|
SFENCE joined #luanti-dev |
01:00 |
MTDiscord |
<exe_virus> On the mapgen thing: why do we need overgeneration, is it to make stuff within the mapchunk correct? If so, why not overgenerate, but not write outside the chunk bounds? |
01:23 |
MTDiscord |
<nathan4220776> I don't know about the true reason, and this probably depends on the implementation, but maybe the idea is that if an area has been emerged, then the adjacent area is probably going to get emerged soon too. So the server might as well pre-emerge that area too so that the adjacent area appears to be merged faster by the player. |
01:28 |
MTDiscord |
<exe_virus> unlikely, or we would just disable it entirely haha |
01:29 |
|
SFENCE joined #luanti-dev |
01:50 |
MTDiscord |
<exe_virus> Me: Alright let's get luanti built here on windows and fix this bug 1 hour later: vcpkg still building dependencies To be continued. |
01:53 |
MTDiscord |
<nathan4220776> Setting up a vm would probably be faster. |
01:54 |
MTDiscord |
<nathan4220776> Yeah, I know that feeling. On an older computer, visual studio and the various sdks took about 2 hours to install. |
01:55 |
MTDiscord |
<exe_virus> I'm a big believer in keeping open source working on windows fully natively, otherwise normies never adopt. But fair |
02:00 |
MTDiscord |
<nathan4220776> Maybe you could have msys2 running in wine in a linux vm on windows. |
02:01 |
MTDiscord |
<nathan4220776> ( ͡° ͜ʖ ͡°) |
02:01 |
MTDiscord |
<exe_virus> And honestly, this has been very straightforward. One hour and I now have VsCode with automatic cmake build going (using MSVC community for the msbuild instance) |
02:01 |
MTDiscord |
<nathan4220776> Excellent. |
02:01 |
MTDiscord |
<exe_virus> (though I still prefer VsCodium) |
02:01 |
MTDiscord |
<exe_virus> much better than even 2 years ago. Msys2 build was much harder back then |
02:02 |
|
SFENCE joined #luanti-dev |
02:02 |
MTDiscord |
<nathan4220776> I haven't done much windows-specific compilation, but is it absolutely a requirement to use microsoft visual c++ for compiling minetest on windows? |
02:02 |
MTDiscord |
<nathan4220776> Are mingw and the related tools not good enough? |
02:04 |
MTDiscord |
<exe_virus> they are good enough, for sure! |
02:05 |
MTDiscord |
<exe_virus> Just not what I used in this case, I have a whole video dedicated to doing everything from IDE, debugging, intellisense, packages, etc. from pure mingw/gcc/msys2 |
02:05 |
MTDiscord |
<exe_virus> But this time - I'm going for speed, and hard to beat vcpkg and their maintenance crews |
02:30 |
|
SFENCE joined #luanti-dev |
02:36 |
MTDiscord |
<exe_virus> Okay initial reactions after seeing the issue and looking at the code, is it's just synchronization issues related to what is the top and bottom and when in this system. Will study it further to get a good grasp on the system. Gotta admit this is pretty complex from all the things tacked onto mapgen in here. Also my trick to look back as far as 4.17 did not help, sadly. |
02:37 |
MTDiscord |
<exe_virus> (which means it's not due to a commit change, but rather just not correct logic) |
02:38 |
|
aliasalreadytake joined #luanti-dev |
02:38 |
MTDiscord |
<exe_virus> Maybe it'll make sense if I can put in debugging breakpoint logic and catch it in the act of making the wrong nodes ... |
03:04 |
|
SFENCE joined #luanti-dev |
04:00 |
|
MTDiscord joined #luanti-dev |
04:22 |
|
SFENCE joined #luanti-dev |
04:47 |
|
SFENCE joined #luanti-dev |
05:48 |
|
SFENCE joined #luanti-dev |
06:50 |
|
TenPlus1 joined #luanti-dev |
06:50 |
TenPlus1 |
Hi folks... |
06:51 |
TenPlus1 |
To whom it may concern :P today's 5.12 dev release is not letting me connect to any servers, you see the connecting message then after a time goes straight to timed out. Note: the webassembly release of luanti/minetest works fine connecting to servers... |
06:52 |
|
SFENCE joined #luanti-dev |
06:52 |
|
Sharpman joined #luanti-dev |
07:08 |
sfan5 |
there were no changes in that part |
07:09 |
sfan5 |
@exe_virus this bug is not supposed to affect lua mapgens |
11:29 |
MTDiscord |
<exe_virus> Makes sense it wouldn't, sorry if I implied so |
11:32 |
sfan5 |
I mean if you find out it does, open a bug |
12:18 |
sfan5 |
reviews needed on #16095 |
12:18 |
ShadowBot |
https://github.com/luanti-org/luanti/issues/16095 -- Remove PrefersNonDefaultGPU from desktop file by Jan200101 |
12:52 |
|
Sharpman joined #luanti-dev |
12:55 |
|
MTDiscord1 joined #luanti-dev |
13:43 |
|
Noisytoot joined #luanti-dev |
15:15 |
|
Noisytoot joined #luanti-dev |
15:36 |
|
wsor4035 joined #luanti-dev |
15:37 |
|
luk3yx joined #luanti-dev |
15:38 |
|
luk3yx joined #luanti-dev |
16:49 |
|
Noisytoot joined #luanti-dev |
16:49 |
|
SFENCE joined #luanti-dev |
17:19 |
|
fluxionary joined #luanti-dev |
17:56 |
|
Noisytoot joined #luanti-dev |
18:17 |
|
Noisytoot joined #luanti-dev |
19:34 |
|
SFENCE joined #luanti-dev |
20:27 |
|
Noisytoot joined #luanti-dev |
20:35 |
|
SFENCE joined #luanti-dev |
20:37 |
|
fluxionary joined #luanti-dev |
20:46 |
|
Noisytoot joined #luanti-dev |
21:28 |
|
SFENCE joined #luanti-dev |
22:33 |
|
panwolfram joined #luanti-dev |
23:05 |
|
Eragon joined #luanti-dev |
23:25 |
|
SFENCE joined #luanti-dev |