| Time |
Nick |
Message |
| 00:18 |
|
SaKeL joined #minetest-dev |
| 00:20 |
bleak_fire_ |
i now have this problem after installing github version |
| 00:20 |
bleak_fire_ |
2015-12-16 00:19:39: ERROR[Main]: Couldn't find a locale directory! |
| 00:21 |
bleak_fire_ |
[Main]: Automatically selecting world at [/home/(myusername)/.minetest/worlds/world] |
| 00:21 |
bleak_fire_ |
ERROR[Main]: Subgame [] could not be found. |
| 00:21 |
bleak_fire_ |
ERROR[Main]: ServerError: Supplied invalid gamespec |
| 00:21 |
bleak_fire_ |
is there a new flag i'm supposed to pass? |
| 00:23 |
RealBadAngel |
do you have a game in your /games folder? |
| 00:23 |
bleak_fire_ |
yes |
| 00:24 |
bleak_fire_ |
it was previously in /usr/share/minetest/games |
| 00:24 |
bleak_fire_ |
under the slackware package |
| 00:24 |
bleak_fire_ |
symlink? |
| 00:24 |
RealBadAngel |
could be, can you create new world? |
| 00:25 |
bleak_fire_ |
no i mean should i symlink |
| 00:26 |
SaKeL |
hi guys i am getting recently this error: |
| 00:26 |
SaKeL |
ERROR[CurlFetchThread]: servers.minetest.net/announce not found (Timeout was reached) (response code 0) |
| 00:26 |
SaKeL |
and my server is not showing on the server list :( |
| 00:26 |
SaKeL |
maybe someone can help ? |
| 00:27 |
RealBadAngel |
bleak_fire_, i can suggest while using dev one to build in place, so all the folders are in build directory |
| 00:27 |
RealBadAngel |
so you wont mess your global installation |
| 00:28 |
RealBadAngel |
just compile with cmake . -DRUN_IN_PLACE=1 |
| 00:28 |
RealBadAngel |
this way you will get "portable" one |
| 00:29 |
bleak_fire_ |
when i remove .minetest/world and debug.txt i get similar but "Subgame specified in default game [minetest]" |
| 00:30 |
bleak_fire_ |
default_game |
| 00:31 |
bleak_fire_ |
i'll try a symlink |
| 00:35 |
bleak_fire_ |
no still getting the "no locale" and "subgame[]" errors |
| 00:35 |
bleak_fire_ |
"subgame [] not found" |
| 00:36 |
bleak_fire_ |
what is a subgame? |
| 00:37 |
RealBadAngel |
minetest_game is a subgame |
| 00:37 |
bleak_fire_ |
--world list correctly lists the world |
| 00:38 |
RealBadAngel |
or minimal, or any other one put in /games |
| 00:38 |
bleak_fire_ |
were there changes in the config file formats? i know the minetest.conf.example is smaller |
| 00:38 |
RealBadAngel |
lately we have chaged setting menu |
| 00:38 |
RealBadAngel |
you dont need config file to edit any setting |
| 00:38 |
bleak_fire_ |
name =, etc, all missing there's only a few |
| 00:39 |
RealBadAngel |
theyre all aviable in main menu |
| 00:39 |
bleak_fire_ |
how do i change it through a vps? |
| 00:39 |
bleak_fire_ |
when there's no x |
| 00:39 |
RealBadAngel |
for that you will still need to edit minetest.conf |
| 00:39 |
RealBadAngel |
i was talking about client side one |
| 00:40 |
bleak_fire_ |
well the minetest.conf i have in the ~/.minetest directory has these changed: |
| 00:40 |
bleak_fire_ |
name= |
| 00:40 |
bleak_fire_ |
name=redblade7 |
| 00:40 |
bleak_fire_ |
sorry |
| 00:40 |
bleak_fire_ |
bind_address |
| 00:40 |
bleak_fire_ |
server_name |
| 00:40 |
bleak_fire_ |
server_description |
| 00:41 |
bleak_fire_ |
server_announce1= |
| 00:41 |
bleak_fire_ |
server_announce=1 |
| 00:41 |
bleak_fire_ |
default_game = minetest_game |
| 00:41 |
bleak_fire_ |
motd |
| 00:41 |
bleak_fire_ |
max_users |
| 00:41 |
bleak_fire_ |
creative_mode=true |
| 00:41 |
bleak_fire_ |
enable_damage=false |
| 00:41 |
bleak_fire_ |
give_initial_stuff=true |
| 00:41 |
bleak_fire_ |
default_privs |
| 00:41 |
bleak_fire_ |
enable_pvp=false |
| 00:42 |
RealBadAngel |
if you are tryin dev build, i said its safer to build it in place, so it will use own settings from the build folder |
| 00:42 |
bleak_fire_ |
static_spawnpoint |
| 00:42 |
bleak_fire_ |
ok |
| 00:42 |
RealBadAngel |
and wont be messed with outdated one you had before |
| 00:42 |
bleak_fire_ |
so what do i have to move? |
| 00:42 |
bleak_fire_ |
so the entire server is outdated? |
| 00:42 |
RealBadAngel |
dev means there has lotsa things changed since last stable |
| 00:44 |
bleak_fire_ |
no make uninstall. nasty mess to clean up |
| 00:44 |
RealBadAngel |
just make separare folder for it |
| 00:44 |
RealBadAngel |
put there sources and compile |
| 00:45 |
bleak_fire_ |
ok wait |
| 00:45 |
bleak_fire_ |
make install lists the files to delete before i can install it your way |
| 00:46 |
bleak_fire_ |
ok i got it |
| 01:58 |
bleak_fire_ |
hi |
| 01:58 |
bleak_fire_ |
thanks everyone for helping me set up the git version |
| 01:59 |
bleak_fire_ |
RUN_IN_PLACE |
| 01:59 |
bleak_fire_ |
but now that i'm compiling from source |
| 01:59 |
bleak_fire_ |
i see that there's an option to increase speed by using LuaJIT and LevelDB |
| 01:59 |
bleak_fire_ |
If I install those packages, are those options buggy? |
| 02:00 |
VanessaE |
they work fine |
| 02:00 |
VanessaE |
leveldb is stable, and luajit works for like 99.99% of the stuff you could throw at it (one or two things may not work as expected) |
| 02:02 |
bleak_fire_ |
ok |
| 02:03 |
bleak_fire_ |
luajit is being used i think |
| 02:03 |
bleak_fire_ |
but as for leveldb |
| 02:03 |
bleak_fire_ |
would that being a different database, would i have to start over with e.g. xban2 |
| 02:04 |
VanessaE |
nah |
| 02:04 |
VanessaE |
that's something you'd use for map storage. |
| 02:04 |
VanessaE |
and the engine has migrate commands. |
| 02:05 |
bleak_fire_ |
and would the existing world i have work if i compiled with leveldb? |
| 02:05 |
bleak_fire_ |
which i need to install |
| 02:05 |
VanessaE |
yes |
| 02:05 |
bleak_fire_ |
ok |
| 02:05 |
VanessaE |
because sqlite would still be compiled in anyway |
| 02:05 |
VanessaE |
you could just migrate the world to leveldb if you wanted to. |
| 02:06 |
bleak_fire_ |
is that hard? |
| 02:06 |
VanessaE |
nah |
| 02:06 |
bleak_fire_ |
or does it do it automatically |
| 02:06 |
VanessaE |
you do it by a command |
| 02:07 |
VanessaE |
something like minetestserver --migrate leveldb --other --switches --here --worldname FOoBar |
| 02:07 |
VanessaE |
(of course that's an example, not a working command) |
| 02:07 |
VanessaE |
once migrated, just move the old sqlite world out of minetest's reach, and start. |
| 02:22 |
bleak_fire_ |
what about gettext, it's disabled by default |
| 02:23 |
bleak_fire_ |
i know i have a lot of spanish and some french speakers on my server, dont know if that will help them |
| 02:25 |
VanessaE |
enable it. |
| 02:25 |
VanessaE |
you'll want that. |
| 02:26 |
VanessaE |
install, from your system repositories, whatever libs it needs of course |
| 02:27 |
bleak_fire_ |
i'm sure you can ask vitaminx about this, since i think he has linode too, but if you ever use slackware with linode,there are tons of missing packages |
| 02:27 |
bleak_fire_ |
it doesnt install them all |
| 02:27 |
bleak_fire_ |
so i had gettext-tools but not gettext |
| 02:27 |
bleak_fire_ |
lol |
| 02:28 |
bleak_fire_ |
will gettext enable everything to be automatically translated for them, or do they have that with the client? |
| 02:29 |
bleak_fire_ |
freetype too was missing lol |
| 02:29 |
VanessaE |
it does not translate anything |
| 02:30 |
VanessaE |
rather, the client has pre-translated messages that it can use |
| 02:30 |
VanessaE |
depending on the client user's locale. |
| 02:31 |
bleak_fire_ |
ok |
| 02:32 |
bleak_fire_ |
well i'm sure they'll be a lot happier now |
| 02:32 |
bleak_fire_ |
not too many english speakers for some reason |
| 02:36 |
bleak_fire_ |
migrated |
| 02:40 |
bleak_fire_ |
lol i have all the height_min/max & y_min/max deprecation warnings showing up |
| 02:40 |
bleak_fire_ |
and a bunch of others like noise_threshhold/noise_threshold |
| 06:35 |
|
Hunterz joined #minetest-dev |
| 06:37 |
sofar |
ok, I suck at making the uvmaps work again :( |
| 06:44 |
sofar |
is there a tool to do this? |
| 06:45 |
sofar |
I hate to fire up blender, takes me 2 hours to open a file |
| 07:06 |
|
DFeniks joined #minetest-dev |
| 07:12 |
sofar |
fehhhhhhhhh |
| 07:14 |
sofar |
who is this Perttu Ahola and can he help me redo the uvmaps for NDT_FENCELIKE? |
| 07:14 |
sofar |
(joking of course, I know who it is) |
| 07:15 |
sofar |
but seriously I can't figure out why remapping the UV's with proper corners just doesn't fix the visual artifacts |
| 07:16 |
sofar |
it just always looks like I'm looking *into* the nodebox |
| 07:16 |
sofar |
glitchy |
| 07:17 |
sofar |
I've got all my x's and y'z and z's in order |
| 07:19 |
RealBadAngel |
can you make a screenshot? |
| 07:19 |
sofar |
of the code? |
| 07:19 |
sofar |
or the effect? |
| 07:19 |
RealBadAngel |
of how it looks like |
| 07:20 |
sofar |
same point I was yesterday |
| 07:20 |
sofar |
http://i.imgur.com/7DYxc4a.png |
| 07:20 |
sofar |
note that it works fine for the bars to +X |
| 07:20 |
sofar |
just not to -X |
| 07:20 |
RealBadAngel |
and the code? |
| 07:20 |
sofar |
I fixed +X textures, they're all decent |
| 07:21 |
sofar |
https://gist.github.com/sofar/e33af1e0666e4299c8cc |
| 07:21 |
sofar |
that's the one I'm working on now |
| 07:21 |
|
DogePony joined #minetest-dev |
| 07:22 |
sofar |
if I'm getting the xzxz -y annotation at the end, then *shrug* it's what makeCuboid says it should be |
| 07:23 |
sofar |
is there an issue with normals? |
| 07:23 |
RealBadAngel |
can i see your whole code? |
| 07:23 |
sofar |
totally |
| 07:24 |
RealBadAngel |
normals are recalculated at the end of making mapblock mesh, so they shouldnt be a problem |
| 07:24 |
sofar |
https://gist.github.com/sofar/88701507cadbe7199f10 |
| 07:24 |
sofar |
I've commented out the GOOD +x section |
| 07:24 |
sofar |
and the +/- Z ones |
| 07:25 |
sofar |
so I can focus on figuring out why I can't get the -X working |
| 07:25 |
sofar |
the +x works good, like I said |
| 07:26 |
RealBadAngel |
gimme a few minutes to compile it |
| 07:26 |
sofar |
shouldn't be too bad, doesn't touch headers |
| 07:26 |
sofar |
unless you need to branch from something |
| 07:27 |
sofar |
I mean, the concept is dainbread misple |
| 07:27 |
sofar |
I'm making it connect to GLASSLIKE_FRAMED so I can look at the textures at each end |
| 07:27 |
sofar |
just to verify that I'm looking at everything |
| 07:27 |
sofar |
should probably include more nodetypes, but, later. |
| 07:28 |
sofar |
I didn't get to the +z one yet since I can't follow the // comments from celeron55 there... those numbers are weirdly sized imo |
| 07:31 |
sofar |
RealBadAngel: I don't even understand why I can't use the +x uvmap for -x?!?! |
| 07:31 |
sofar |
it's the exact same shape and orientation |
| 07:37 |
sofar |
RealBadAngel: wow we need to get you a better computer :) |
| 07:37 |
RealBadAngel |
readin your code atm |
| 07:37 |
sofar |
it's ... pretty trivial |
| 07:38 |
sofar |
I just shortened the fence bars and paint them at all 4 sides of the pole |
| 07:40 |
RealBadAngel |
sofar, please do change nodebox vertices |
| 07:41 |
sofar |
wrong ordering? |
| 07:41 |
RealBadAngel |
yeah |
| 07:41 |
sofar |
for -X? |
| 07:41 |
sofar |
ohh |
| 07:41 |
sofar |
-BS/2 is smaller than -post_rad |
| 07:42 |
RealBadAngel |
theyre not automagically sorted, you have to take care of it |
| 07:42 |
sofar |
goddamn, I thought I checked that ! |
| 07:42 |
sofar |
thanks man, that totally works |
| 07:42 |
RealBadAngel |
np |
| 07:42 |
sofar |
now I know I'm not crazy, and understand how it works :) |
| 07:44 |
sofar |
ahhh that was it, now they all work properly |
| 07:44 |
sofar |
no glitches |
| 07:45 |
|
DFeniks joined #minetest-dev |
| 07:46 |
sofar |
RealBadAngel: http://i.imgur.com/1QYTVqE.png |
| 07:46 |
RealBadAngel |
indeed, way better :) |
| 07:47 |
VanessaE |
sofar: no fancy end-to-end joints? :) |
| 07:47 |
sofar |
yeah, and I'm reasonably happy with the uv choice working out |
| 07:47 |
VanessaE |
come on, dovetail those suckers :D |
| 07:47 |
sofar |
so I'll probably not change it further |
| 07:47 |
VanessaE |
(jk) |
| 07:47 |
sofar |
VanessaE: buy me a woodworking shop |
| 07:47 |
VanessaE |
sofar: you already have one. it's called blender :D |
| 07:47 |
VanessaE |
but really that does look good. |
| 07:48 |
sofar |
so question now is, what nodetypes should I have it connect to |
| 07:48 |
sofar |
I have NORMAL and GLASSLIKE_FRAMED |
| 07:48 |
VanessaE |
"normal" and anything that has an appropriate node def property |
| 07:48 |
sofar |
how about |
| 07:48 |
VanessaE |
if you have GLASSLIKE_FRAMED, you should also use regular glasslike. |
| 07:48 |
sofar |
3 patches |
| 07:48 |
sofar |
1) change painting to this method |
| 07:49 |
sofar |
2) add normal |
| 07:49 |
sofar |
3) add API for more types? |
| 07:49 |
VanessaE |
sounds fair to me |
| 07:51 |
DFeniks |
can i choose them to not connect ? i realized that actually while i sorta like this , i also liked old behavior of fence , like on my LOTT server buildings |
| 07:51 |
RealBadAngel |
if talkin bout api, maybe add nodebox def for both post and bars? |
| 07:51 |
sofar |
DFeniks: maybe, but would also be an API change |
| 07:51 |
VanessaE |
could be useful, but complicated |
| 07:51 |
RealBadAngel |
this would allow custom fences |
| 07:52 |
sofar |
I talked with est about that |
| 07:52 |
DFeniks |
althought lott probably will not update anyway , who knows |
| 07:52 |
sofar |
I'm thinking of adding a new NDT_CONNECTING type that is more flexible |
| 07:52 |
RealBadAngel |
safar, have you saw my concrete posts mod? |
| 07:52 |
sofar |
you showed the video |
| 07:53 |
sofar |
NDT_CONNECTING can also handle xpanes, more glass stuff, maybe horizontal glass even |
| 07:53 |
sofar |
if we design it right |
| 07:53 |
VanessaE |
sounds like it would handle your "walllike" idea too |
| 07:53 |
RealBadAngel |
if nodedef could carry custom nodeboxes, makin such mods will be damn easier |
| 07:53 |
sofar |
it would obsolete fences and walls of any implementation |
| 07:53 |
sofar |
right, that's the trick |
| 07:53 |
RealBadAngel |
atm such mod needs shitload of nodedefs instead of just single one |
| 07:54 |
sofar |
exactly |
| 07:54 |
sofar |
it's like 20 for walls, to do a decent job |
| 07:54 |
sofar |
more if you get crazy |
| 07:54 |
DFeniks |
sorry if i only half pay attention to this topic , i might not understand all , because i dont pay attention |
| 07:54 |
sofar |
could all just be rendered client side |
| 07:55 |
RealBadAngel |
you just need to pass nodeboxes from lua to the renderer |
| 07:55 |
sofar |
yup |
| 07:55 |
RealBadAngel |
and adding any kind of fence will be damn easy |
| 07:55 |
sofar |
and uvmap it |
| 07:55 |
sofar |
that's likely the hard part |
| 07:56 |
sofar |
projecting it should be doable, though |
| 07:56 |
sofar |
just mapping the in-node position to the uvmap for each face |
| 07:56 |
RealBadAngel |
same way as NDT_NODEBOX does |
| 07:56 |
sofar |
fence, wall, glass pane, metal bars, even vines |
| 07:57 |
sofar |
let me start PR-ing this |
| 07:57 |
sofar |
so we get some feedback :) |
| 08:00 |
RealBadAngel |
sofar, btw, you can just specify one single bar, lets say x+ |
| 08:01 |
RealBadAngel |
UV map it |
| 08:01 |
RealBadAngel |
and then just rotate it |
| 08:01 |
sofar |
and then copy/move yes |
| 08:01 |
sofar |
it's not a bad idea to uvmap it nicely for each direction |
| 08:01 |
RealBadAngel |
i mean rotate it with irrlicht |
| 08:01 |
sofar |
right |
| 08:01 |
RealBadAngel |
you wont have to care about UV then |
| 08:01 |
sofar |
just like how plantlike rotates |
| 08:02 |
RealBadAngel |
or mesh cache |
| 08:02 |
RealBadAngel |
easier to handle |
| 08:03 |
sofar |
submitting this then bed |
| 08:06 |
|
Darcidride joined #minetest-dev |
| 08:09 |
sofar |
DFeniks: making "old style" fences is already trivial in mods, btw |
| 08:12 |
sofar |
https://github.com/minetest/minetest/pull/3459 |
| 08:12 |
sofar |
happy hump day, I'm off to bed |
| 08:34 |
|
nrzkt joined #minetest-dev |
| 08:45 |
|
SaKeL joined #minetest-dev |
| 09:20 |
|
Amaz joined #minetest-dev |
| 09:29 |
|
jin_xi joined #minetest-dev |
| 10:03 |
|
Darcidride joined #minetest-dev |
| 10:05 |
|
Fixer joined #minetest-dev |
| 11:22 |
|
proller joined #minetest-dev |
| 11:28 |
|
dzho joined #minetest-dev |
| 11:53 |
|
proller joined #minetest-dev |
| 12:01 |
|
behalebabo joined #minetest-dev |
| 12:03 |
|
proller joined #minetest-dev |
| 12:23 |
|
zat joined #minetest-dev |
| 12:27 |
|
Darcidride joined #minetest-dev |
| 12:32 |
|
Calinou joined #minetest-dev |
| 12:55 |
|
turtleman_ joined #minetest-dev |
| 12:57 |
|
Fixer joined #minetest-dev |
| 12:58 |
|
nrzkt joined #minetest-dev |
| 12:58 |
|
Obani joined #minetest-dev |
| 13:01 |
|
Player2 joined #minetest-dev |
| 13:18 |
|
SaKeL joined #minetest-dev |
| 13:40 |
|
dfelinto joined #minetest-dev |
| 14:13 |
|
ElectronLibre left #minetest-dev |
| 14:25 |
|
ElectronLibre joined #minetest-dev |
| 14:38 |
|
leat joined #minetest-dev |
| 14:40 |
|
DFeniks joined #minetest-dev |
| 14:43 |
dfelinto |
sofar: hi |
| 14:44 |
dfelinto |
sofar: you were right yesterday. there is no need to change irrlicht for what I need |
| 14:45 |
dfelinto |
I just sent a pull request with an initial feature relative to what I will need - https://github.com/minetest/minetest/pull/3460 |
| 14:45 |
dfelinto |
this is useful on its own, and set the groundwork for my Mesa-3D support |
| 14:58 |
|
H-H-H joined #minetest-dev |
| 15:25 |
nrzkt |
thanks sfan5 for the code guidelines guide :) |
| 15:25 |
nrzkt |
link* |
| 15:25 |
sfan5 |
:) |
| 15:31 |
|
CraigyDavi joined #minetest-dev |
| 15:41 |
|
proller joined #minetest-dev |
| 16:11 |
|
Megaf joined #minetest-dev |
| 16:24 |
|
proller joined #minetest-dev |
| 16:26 |
|
proller joined #minetest-dev |
| 16:30 |
dfelinto_afk |
sfan5: thanks :) |
| 16:35 |
dfelinto |
sofar: I still dont know what is wrong with the patch though. unless the anaglyph code is also incorrect |
| 16:49 |
|
Megaf joined #minetest-dev |
| 16:58 |
|
alket joined #minetest-dev |
| 17:03 |
|
Drangue joined #minetest-dev |
| 17:09 |
|
Hunterz joined #minetest-dev |
| 17:15 |
|
jin_xi joined #minetest-dev |
| 17:22 |
|
Krock joined #minetest-dev |
| 17:37 |
|
nrzkt joined #minetest-dev |
| 18:02 |
|
zupoman joined #minetest-dev |
| 18:02 |
|
zupoman joined #minetest-dev |
| 18:05 |
|
hmmmm joined #minetest-dev |
| 18:18 |
|
Darcidride joined #minetest-dev |
| 18:27 |
|
GhostDoge joined #minetest-dev |
| 18:27 |
|
Robert_Zenz joined #minetest-dev |
| 18:38 |
|
rubenwardy joined #minetest-dev |
| 18:59 |
|
Ritchie joined #minetest-dev |
| 19:01 |
|
sapier joined #minetest-dev |
| 19:06 |
|
turtleman_ joined #minetest-dev |
| 19:12 |
sapier |
our coding style is missing a clear definition where to place braces |
| 19:42 |
Krock |
The Linux kernel style guidelines describe some cases |
| 19:43 |
sapier |
as our code doesn't follow those on many many locations we should clarify this in our own guidelines |
| 19:44 |
sapier |
to me it's strange to add a single patch in "correct style" within a whole file of different coding style |
| 19:45 |
|
younishd joined #minetest-dev |
| 19:47 |
Krock |
I was told that the code is replaced a bit with every new pull, trying to follow the Linux kernel guidelines. |
| 19:47 |
Krock |
Style corrections are just unneccessary |
| 19:47 |
sapier |
no they aren't |
| 19:47 |
Krock |
I mean, you would have to update all files |
| 19:48 |
Krock |
Braking all other pull requests |
| 19:48 |
sofar |
I often end up copying style around the place that I'm editing only to have people commented that the style I used (copied) is bad |
| 19:48 |
Krock |
:/ |
| 19:48 |
sapier |
everyone does copy old code and noone does at first fix the style of that code thus you'll always have a lot of "but coding style is wrong there and there and there" comments |
| 19:49 |
Krock |
Yes, I see the problem there |
| 19:49 |
sapier |
considering the amount of time lost due those comments it's more then worth do fix files once and forever ;-) |
| 19:50 |
sapier |
but you can't use code style formater as or coding guideline is somewhat underspecified |
| 19:50 |
Krock |
But as I said above. A cleanup for the complete source would break almost every pull requests. Depends if they're gonna accept that or not |
| 19:50 |
sapier |
even if you did find a code style formater capable of doing it that way |
| 19:50 |
sapier |
krock file by file would be a start |
| 19:50 |
nrzkt |
sapier, we tend to go to Linux kernel guidelines |
| 19:50 |
nrzkt |
at each commit we fix the current style |
| 19:51 |
nrzkt |
on the modified portion of code |
| 19:51 |
Krock |
* mostly |
| 19:51 |
sapier |
I suggest we add a rule top 10 coding style complainers have do to 1 file per month ;-) |
| 19:51 |
sapier |
nrzkt on that base you'll not have it done till 2200 ;-) |
| 19:52 |
nrzkt |
sapier, maybe :p |
| 19:52 |
sofar |
just writing a statement at the top of the style guide that says "you must fix style issues, copying style issues is not allowed" |
| 19:52 |
sofar |
that... would help |
| 19:52 |
Krock |
You're good in panning, sapier. Let's see if we get ready until then |
| 19:52 |
Krock |
*planning |
| 19:52 |
sofar |
sapier: +1 on that idea ;) |
| 19:52 |
* sofar |
looks around |
| 19:52 |
sapier |
see the facdir limiting pull if I did fix it for those 2 if's I add it's almost the onle location in that file beeing correct |
| 19:53 |
* Krock |
does sapier++ too |
| 19:54 |
sapier |
btw imho for braces kernel coding style is stupid |
| 19:54 |
sapier |
if you place { on following line you don't have to add an empty line below if |
| 20:04 |
sapier |
grr guess I'll have to create a new forum and wiki account neither password email resets do work |
| 20:05 |
sapier |
at least not for me |
| 20:09 |
|
Megaf joined #minetest-dev |
| 20:15 |
|
leat joined #minetest-dev |
| 20:17 |
|
leat joined #minetest-dev |
| 20:48 |
|
est31 joined #minetest-dev |
| 20:49 |
est31 |
about fixing the code style of whole files, I generally agree with it |
| 20:49 |
est31 |
but yeah it destroys git blame a bit |
| 20:51 |
sapier |
what's more important save developer time or find someone guilty? |
| 20:52 |
|
Megaf joined #minetest-dev |
| 20:53 |
sofar |
I would recommend writing a recommended 'indent' usage statement |
| 20:53 |
sofar |
so that people can at least have a way to see what is wrongly indented right now |
| 20:54 |
sofar |
and how a fixed version would roughly look |
| 21:09 |
Fixer |
not an expert, but someone said that loading meshes to GPU causes stutter, and often there are 10-30 meshes generated creating a big stutter (like 16ms->30ms->16ms), is it possible to queue generated meshes and upload 1 mesh per rendering frame to avoid stutter? so it will be 16ms->17ms->16ms. Is that possible? |
| 21:09 |
|
est31 joined #minetest-dev |
| 21:11 |
est31 |
Fixer, generated meshes are already queued |
| 21:12 |
est31 |
the only thing to change now is that the queue is emptied only on a limited basis |
| 21:13 |
Fixer |
somehow smooth the process so you will have not a BIG spike, but very slight but longer and not observed by the eye that much |
| 21:33 |
Fixer |
!tell est31 yes, emptied on a limited basis, spending just a ms or so per rendering frame to avoid stutter, that will be cool |
| 21:33 |
ShadowBot |
Fixer: O.K. |
| 21:37 |
sapier |
everything above 42ms will be noticeable |
| 21:37 |
sapier |
some ppl will even notice below |
| 21:43 |
Fixer |
sapier, say another way, on any descent videocard with limited view you will hit 65fps limit (or so), that is 15ms per frame, I often see events that cause it to go up to 40-120ms, so it is 15->15->40->15->15->15, that is very noticable |
| 21:43 |
sapier |
possible |
| 21:44 |
Fixer |
sapier, or worst part is this -> 15->60->15->60->15->15->80->15, step like stutter (like in forests), that is just bad |
| 21:44 |
Fixer |
it is sadistic on your eyes |
| 21:45 |
Fixer |
even on flatgen mod without anything (just grass block), when it generates meshes, my frame time can go from 15ms to 40ms and back... on a flatgen, Karl |
| 21:53 |
|
proller joined #minetest-dev |
| 21:54 |
|
proller joined #minetest-dev |
| 21:57 |
|
Terusthebird joined #minetest-dev |
| 22:36 |
|
Fixer joined #minetest-dev |
| 22:59 |
|
celeron55 joined #minetest-dev |
| 23:06 |
|
sapier left #minetest-dev |
| 23:10 |
|
celeron55 joined #minetest-dev |
| 23:45 |
|
celeron55 joined #minetest-dev |
