| Time |
Nick |
Message |
| 00:43 |
|
behalebabo joined #minetest-dev |
| 01:00 |
|
Lone_Wolf joined #minetest-dev |
| 01:00 |
Lone_Wolf |
Anyone have any thoughts on this? https://github.com/minetest/minetest/pull/10013#issuecomment-640965161 |
| 01:08 |
hecks |
Lone_Wolf maybe it's the stupid back-and-forth camera offset math destroying any precision it was meant to provide |
| 01:11 |
Lone_Wolf |
The nametag pos is broken too though, it's in the same place |
| 01:12 |
|
ANAND joined #minetest-dev |
| 01:12 |
hecks |
so this is the camera's position + nametag offset without the entity position being added? |
| 01:14 |
hecks |
you'll need to tell me more |
| 01:14 |
hecks |
like what's the global position of the camera, the object with the nametag |
| 01:14 |
hecks |
and whether this is happening all the time or if it just blinks for one frame |
| 01:15 |
hecks |
I might be able to solve this |
| 01:20 |
Lone_Wolf |
Code's right there in the PR. cam_pos is the camera pos in world coordinates and nametag_pos is the coordinates of the object the nametag belongs to AFAIK |
| 01:21 |
Lone_Wolf |
It consistently happens more than 8 nodes away, and is fine near 0, 0, 0 from what I can tell |
| 01:24 |
Lone_Wolf |
I can reproduce near 0, 0, 0 and the camera/nametag coordinates are pretty much the same actually |
| 01:25 |
hecks |
>cam_pos is the camera pos in world coordinates |
| 01:25 |
hecks |
getAbsolutePosition is not in world coordinates |
| 01:25 |
hecks |
it's in the client's stupid coordinates |
| 01:26 |
Lone_Wolf |
Ah |
| 01:27 |
hecks |
I don't know which space client side raycasts use but you need to add the camera offset to node positions to get the true world pos |
| 01:27 |
ANAND |
Updated docs for #7587 |
| 01:27 |
hecks |
or subtract it, I don't remember, see GenericCAO::getPosition or just use that function if you can |
| 01:27 |
ShadowBot |
https://github.com/minetest/minetest/issues/7587 -- Implement on_wielditem_change callback registration by ClobberXD |
| 01:46 |
skyliner_369 |
what's the best way to update the minetest version with git and rebuild? just run the git command and run build again with a clobber option? |
| 02:11 |
skyliner_369 |
in other words, do I have to run git clone, then cmake, then make? or can I like run git, (what command is basically just a git update?) and then make? |
| 02:12 |
hecks |
have you looked at what the buildbot does? |
| 02:12 |
hecks |
there's a script in util/buildbot that gets the latest version with dependencies and builds it |
| 02:13 |
Lone_Wolf |
I do 'git pull -r minetest_repo_remote && make -j $(nproc)' |
| 02:14 |
|
ANAND joined #minetest-dev |
| 02:22 |
skyliner_369 |
might make a simple bash script to build from a git pull or something |
| 02:22 |
skyliner_369 |
I actually don't know my way around git and such that well |
| 02:23 |
skyliner_369 |
like, does cmake have an update build option? same for make |
| 02:24 |
hecks |
if you don't delete the build directory, cmake will do a quick rebuild if it can |
| 02:24 |
skyliner_369 |
ahh makes sense |
| 02:24 |
skyliner_369 |
I'm thinking about making a sudo bash to put the game system-wide |
| 02:24 |
hecks |
but every now and then it can just decide to rebuild everything, so don't be alarmed by that |
| 02:25 |
skyliner_369 |
Alrighty |
| 02:25 |
hecks |
I mean, the build instructions in readme are clear enough for linux |
| 02:26 |
skyliner_369 |
it's not building I'm thinking about, I'm thinking more about updating the build without having to uproot everything |
| 02:26 |
hecks |
though it might be worth it to peek at the windows buildbot anyway just to know the possible cmake incantations |
| 02:27 |
hecks |
well then: pull, cmake, make with run in place enabled |
| 02:27 |
skyliner_369 |
buildbot builds windows builds whenever the commits update? |
| 02:27 |
hecks |
pretty much |
| 02:28 |
hecks |
anyway following the readme.md instructions should give you a working build in the same folder, runnable from ./bin/ |
| 02:30 |
skyliner_369 |
actually, I'll be doing it as super with a bash script I can run on the calendar thingy I forget the name of. once a day and tada. set run-in-place to false |
| 02:31 |
|
Miner_48er joined #minetest-dev |
| 02:32 |
skyliner_369 |
does buildbot post DL links in this channel? |
| 02:33 |
hecks |
I don't know, never seen it, there's a forum thread with links though |
| 02:33 |
skyliner_369 |
well I mean the bot that's got the edgy sonic character name |
| 02:34 |
hecks |
this bot posts git links |
| 02:38 |
skyliner_369 |
I think I saw it post a version link too before? I could be going mad though |
| 02:50 |
hecks |
I don't sit here all day, so maybe you did |
| 02:54 |
rubenwardy |
skyliner_369: -> #minetest |
| 02:55 |
rubenwardy |
this is a channel for Minetest engine development, not for user questions and dicussions |
| 03:15 |
|
Ritchie joined #minetest-dev |
| 04:56 |
|
Wuzzy joined #minetest-dev |
| 05:58 |
|
calcul0n joined #minetest-dev |
| 06:02 |
|
ANAND joined #minetest-dev |
| 06:35 |
|
oil_boi joined #minetest-dev |
| 06:42 |
|
mizux joined #minetest-dev |
| 07:08 |
|
NetherEran joined #minetest-dev |
| 07:25 |
|
Ritchie joined #minetest-dev |
| 08:00 |
|
ShadowNinja joined #minetest-dev |
| 08:35 |
|
Beton joined #minetest-dev |
| 09:41 |
|
pmp-p joined #minetest-dev |
| 10:16 |
|
NetherEran joined #minetest-dev |
| 10:31 |
|
proller joined #minetest-dev |
| 10:37 |
|
proller joined #minetest-dev |
| 10:48 |
|
Fixer joined #minetest-dev |
| 11:25 |
|
NetherEran joined #minetest-dev |
| 11:31 |
|
Ritchie joined #minetest-dev |
| 12:34 |
|
erlehmann joined #minetest-dev |
| 12:55 |
|
calcul0n_ joined #minetest-dev |
| 13:09 |
|
Fixer joined #minetest-dev |
| 13:35 |
|
lisac joined #minetest-dev |
| 13:45 |
|
NetherEran joined #minetest-dev |
| 14:34 |
|
gorbachev_pizza joined #minetest-dev |
| 14:41 |
|
Lone_Wolf joined #minetest-dev |
| 15:11 |
ANAND |
Have anyone tested #7924 or #7587 yet? :) |
| 15:11 |
ShadowBot |
https://github.com/minetest/minetest/issues/7924 -- Allow binding dig, place actions to keys; remove LMB/RMB hardcoding by ClobberXD |
| 15:11 |
ShadowBot |
https://github.com/minetest/minetest/issues/7587 -- Implement on_wielditem_change callback registration by ClobberXD |
| 15:11 |
ANAND |
Has* |
| 15:43 |
|
lisac joined #minetest-dev |
| 16:07 |
|
lisac joined #minetest-dev |
| 16:13 |
|
Taoki joined #minetest-dev |
| 16:21 |
|
oiaohm joined #minetest-dev |
| 17:20 |
Krock |
will merge #9690 #10008 and #8785 in 15 minutes |
| 17:20 |
ShadowBot |
https://github.com/minetest/minetest/issues/9690 -- [NO SQUASH!] TouchScreen control bug fixes and code cleanup by MoNTE48 |
| 17:20 |
ShadowBot |
https://github.com/minetest/minetest/issues/10008 -- Fix player-to-object attachment teleport bug by hecktest |
| 17:20 |
ShadowBot |
https://github.com/minetest/minetest/issues/8785 -- LuaItemStack: Add __tostring metamethod by pauloue |
| 17:21 |
Krock |
#9572 too |
| 17:21 |
ShadowBot |
https://github.com/minetest/minetest/issues/9572 -- Some vector functions useful for working with rotations by NetherEran |
| 17:21 |
|
LoneWolfHT joined #minetest-dev |
| 17:25 |
Krock |
sfan5: mind proposing a PR for those network analytics code? |
| 17:25 |
Krock |
it would be really interesting to have that for future network issues |
| 17:26 |
sfan5 |
sure |
| 17:26 |
sfan5 |
will have to clean it up to be less hacky though |
| 17:27 |
Krock |
the enum/string struct mainly used in the script API might help to clean it up |
| 17:35 |
Krock |
merging... |
| 17:38 |
Krock |
Done |
| 17:43 |
NetherEran |
thanks :D |
| 17:44 |
sfan5 |
u32 needed_count = 2; // require at least two solid blocks |
| 17:44 |
sfan5 |
if (count >= needed_count) { return true; } |
| 17:44 |
sfan5 |
spot the mistake |
| 17:44 |
Krock |
= |
| 17:45 |
Krock |
I miss the context |
| 17:45 |
sfan5 |
right more context is needed to spot it |
| 17:46 |
sfan5 |
https://github.com/minetest/minetest/blob/master/src/map.cpp#L1135-L1137 |
| 17:47 |
Krock |
it can hit multiple times per node |
| 17:47 |
sfan5 |
good point but that wasn't it |
| 17:47 |
Krock |
so there are 2 bugs |
| 17:47 |
sfan5 |
since count starts at 0, setting needed_count = 2 in fact requires *three* solid nodes before the check returns true |
| 17:48 |
Krock |
how come? increment happens before |
| 17:49 |
Krock |
it's like if (++count == needed_count) |
| 17:49 |
sfan5 |
hm |
| 17:49 |
sfan5 |
you're right |
| 17:49 |
sfan5 |
but this doesn't match my debug output, wtf |
| 17:50 |
Krock |
wrong profiling or multiple hits per node |
| 17:50 |
sfan5 |
(111,11,-148) main:grass (111,10,-147) main:dirt (112,10,-147) main:dirt (112,9,-146) main:stone occluded! |
| 17:50 |
sfan5 |
who knows what crafter has set as light_propagates for those nodes... |
| 17:51 |
sfan5 |
anyway the check running multiple times for the same node is unlikely since the occlusion starts at 1.2 node-lengths per step |
| 17:52 |
Krock |
main:grass is likely to be light_propagates = true to not cause light artefacts |
| 17:52 |
Krock |
FWIW: light_propagates is also set to true when paramtype = "light" |
| 17:52 |
Krock |
s/also// |
| 17:57 |
Krock |
rubenwardy: I think you could be interested in this #8475 |
| 17:57 |
ShadowBot |
https://github.com/minetest/minetest/issues/8475 -- Add cube map and equirectangular rendering mode by srifqi |
| 17:57 |
rubenwardy |
What is it? |
| 17:57 |
rubenwardy |
I don't understand the use of the rendering mode |
| 17:57 |
rubenwardy |
For YouTube 3d? |
| 17:58 |
Krock |
for 3D-recording, yes. |
| 17:58 |
Krock |
might also be used by VR glasses, not sure though. |
| 17:59 |
rubenwardy |
Don't vr headsets use side by side? |
| 17:59 |
rubenwardy |
I guess some curvature makes sense actually |
| 18:22 |
|
hecks joined #minetest-dev |
| 18:38 |
hecks |
sfan5: rebased |
| 18:52 |
|
gorbachev_pizza joined #minetest-dev |
| 19:14 |
|
Lone_Wolf joined #minetest-dev |
| 19:15 |
hecks |
#10015 looking for reviewers |
| 19:15 |
ShadowBot |
https://github.com/minetest/minetest/issues/10015 -- Fix bone-attached entities by hecktest |
| 19:19 |
Krock |
404 reviewers not found |
| 19:22 |
hecks |
if I bribed or blackmailed celeron into making me core, would my bugs need one less approval? =] |
| 19:23 |
Krock |
bugs need no approvals, but PRs would only need one, yes - unless someone is opposed |
| 19:23 |
hecks |
well, I'm talking about dead bugs |
| 19:25 |
Krock |
will merge #9969 in 10 minutes |
| 19:25 |
ShadowBot |
https://github.com/minetest/minetest/issues/9969 -- Fix broken coloring of wielditems by dcbrwn |
| 19:25 |
Krock |
nice |
| 19:26 |
Krock |
hecks: sorry what? dead bugs are resolved bugs |
| 19:27 |
hecks |
oh never mind, please review my fix if you have time |
| 19:27 |
Krock |
other PRs are waiting for months |
| 19:29 |
hecks |
and I just got a fix merged in 3 days, so I'm optimistic about this one too =] |
| 19:29 |
hecks |
these are all trivial |
| 19:36 |
Krock |
mergign |
| 19:37 |
Krock |
you're happy to have somebody who's familiar with the code and got the right test code to check all known bug sources |
| 19:37 |
Krock |
this is by far not the case for other PRs.. |
| 19:40 |
sfan5 |
#9299 could use a second review too |
| 19:41 |
ShadowBot |
https://github.com/minetest/minetest/issues/9299 -- Improve default inventory+wield images of node drawtypes by Wuzzy2 |
| 20:02 |
hecks |
I've done a bit of self-review to streamline the process further, hope that helps |
| 20:16 |
|
appguru joined #minetest-dev |
| 20:25 |
|
NetherEran joined #minetest-dev |
| 20:45 |
|
mntmn joined #minetest-dev |
| 20:46 |
|
Lunatrius` joined #minetest-dev |
| 22:49 |
|
proller joined #minetest-dev |
| 23:17 |
|
Thomas-S joined #minetest-dev |
| 23:17 |
|
Thomas-S joined #minetest-dev |
