| Time |
Nick |
Message |
| 00:01 |
|
AndChat|88724 joined #minetest-dev |
| 00:18 |
|
halt_ joined #minetest-dev |
| 00:20 |
|
halt_ joined #minetest-dev |
| 00:27 |
|
octacian_ joined #minetest-dev |
| 00:47 |
|
Grandolf joined #minetest-dev |
| 01:04 |
|
Grandolf joined #minetest-dev |
| 01:14 |
|
Grandolf joined #minetest-dev |
| 01:20 |
|
Grandolf_ joined #minetest-dev |
| 01:26 |
|
paramat joined #minetest-dev |
| 01:30 |
paramat |
+1 for #4524 ,tested. sofar perhaps you could check the updated PR? |
| 01:30 |
ShadowBot |
https://github.com/minetest/minetest/issues/4524 -- Clouds api by bendeutsch |
| 01:41 |
|
ssieb joined #minetest-dev |
| 01:47 |
|
octacian joined #minetest-dev |
| 02:11 |
|
twoelk left #minetest-dev |
| 04:11 |
|
destt_ joined #minetest-dev |
| 04:55 |
|
nerzhul joined #minetest-dev |
| 05:04 |
nerzhul |
i think i will refactor client event ugly union at a point... to use inheritance, this will permit to reduce memory usage on events |
| 05:10 |
|
Hunterz joined #minetest-dev |
| 05:11 |
nerzhul |
and after release, prepare a new session model, RemotePlayer is really a strange object, mixing Session, Player, SAO parts, it's not clear :( |
| 05:12 |
|
paramat joined #minetest-dev |
| 05:14 |
nerzhul |
sfan5, are you fine with #5658 ? |
| 05:14 |
ShadowBot |
https://github.com/minetest/minetest/issues/5658 -- Fix #5655 by red-001 |
| 05:16 |
nerzhul |
merging #5659 in 5 min |
| 05:16 |
ShadowBot |
https://github.com/minetest/minetest/issues/5659 -- Tiny documentation fix. by BluebirdGreycoat |
| 05:24 |
nerzhul |
merged |
| 05:53 |
sofar |
I've got one major beef with that cloud API PR |
| 05:53 |
sofar |
see the PR thread |
| 05:53 |
sofar |
#4524 |
| 05:53 |
ShadowBot |
https://github.com/minetest/minetest/issues/4524 -- Clouds api by bendeutsch |
| 05:53 |
sofar |
videos posted of what I think it should do |
| 06:07 |
sofar |
https://www.youtube.com/watch?v=aiGSZR-q6AM fwiw |
| 06:12 |
sofar |
honestly I don't know noise2d_perlin() well enough to understand what the constant seed factor should be here |
| 06:12 |
sofar |
paramat: if you have time, please check |
| 06:13 |
nerzhul |
seed is used in every perlin noise calculator to generate noise |
| 06:14 |
sofar |
well yeah, but why does the constant m_seed value (it doesn't change) not do anything? |
| 06:14 |
sofar |
if I seed it with x,z derivative values it works as expected, just watch the video |
| 06:15 |
sofar |
well, it sometimes jumps (probably because a value rolls over once in a while) |
| 06:15 |
sofar |
without my patches, every time I call set_clouds() the entire sky is wiped and reseeded |
| 06:16 |
sofar |
with my patch, the clouds are behaving as I would expect them to |
| 06:16 |
sofar |
anyway, too tired to think, I'm off to bed |
| 06:36 |
|
Hunterz joined #minetest-dev |
| 06:42 |
|
Hunterz joined #minetest-dev |
| 06:51 |
|
Hunterz joined #minetest-dev |
| 06:56 |
|
Hunterz joined #minetest-dev |
| 06:58 |
|
Hunterz1 joined #minetest-dev |
| 07:00 |
|
Hunterz2 joined #minetest-dev |
| 07:10 |
|
kilbith joined #minetest-dev |
| 07:10 |
|
nerzhul joined #minetest-dev |
| 07:21 |
|
Hunterz joined #minetest-dev |
| 07:22 |
|
nerzhul joined #minetest-dev |
| 08:26 |
|
nerzhul joined #minetest-dev |
| 08:38 |
|
lisac joined #minetest-dev |
| 08:39 |
|
Sockbat joined #minetest-dev |
| 08:42 |
|
proller joined #minetest-dev |
| 08:59 |
|
troller joined #minetest-dev |
| 09:00 |
|
kilbith joined #minetest-dev |
| 09:09 |
|
troller joined #minetest-dev |
| 09:17 |
|
troller joined #minetest-dev |
| 09:42 |
|
proller joined #minetest-dev |
| 09:45 |
|
proller__ joined #minetest-dev |
| 09:55 |
|
Fixer joined #minetest-dev |
| 09:58 |
|
YuGiOhJCJ joined #minetest-dev |
| 10:47 |
|
DI3HARD139 joined #minetest-dev |
| 11:11 |
|
DI3HARD139 joined #minetest-dev |
| 11:16 |
|
proller joined #minetest-dev |
| 11:18 |
|
Fritigern joined #minetest-dev |
| 11:43 |
|
Megaf joined #minetest-dev |
| 11:45 |
|
kilbith joined #minetest-dev |
| 12:37 |
|
DS-minetest joined #minetest-dev |
| 12:59 |
|
twoelk joined #minetest-dev |
| 13:23 |
|
lisac joined #minetest-dev |
| 13:28 |
|
octacian joined #minetest-dev |
| 13:29 |
|
Hunterz joined #minetest-dev |
| 13:40 |
|
RobbieF joined #minetest-dev |
| 14:02 |
|
RobbieF left #minetest-dev |
| 14:11 |
|
XtremeHacker joined #minetest-dev |
| 14:14 |
|
benrob0329 joined #minetest-dev |
| 14:27 |
|
DS-minetest joined #minetest-dev |
| 14:30 |
|
Megaf joined #minetest-dev |
| 14:38 |
|
RobbieF joined #minetest-dev |
| 14:39 |
|
RobbieF left #minetest-dev |
| 15:10 |
|
destt_ joined #minetest-dev |
| 15:32 |
|
YuGiOhJCJ joined #minetest-dev |
| 15:36 |
|
Darcidride joined #minetest-dev |
| 15:45 |
|
Hunterz joined #minetest-dev |
| 16:39 |
|
RobbieF joined #minetest-dev |
| 16:39 |
|
RobbieF left #minetest-dev |
| 16:39 |
|
troller joined #minetest-dev |
| 16:45 |
|
Fritigern joined #minetest-dev |
| 16:48 |
|
Krock joined #minetest-dev |
| 16:48 |
|
Krock joined #minetest-dev |
| 17:00 |
|
Grandolf joined #minetest-dev |
| 17:00 |
|
Grandolf_ joined #minetest-dev |
| 17:01 |
|
Grandolf joined #minetest-dev |
| 17:01 |
|
halt_ joined #minetest-dev |
| 17:02 |
|
halt_ joined #minetest-dev |
| 17:03 |
|
nerzhul joined #minetest-dev |
| 17:06 |
|
RobbieF joined #minetest-dev |
| 17:06 |
|
RobbieF left #minetest-dev |
| 17:14 |
|
kilbith joined #minetest-dev |
| 17:50 |
Krock |
Does any dev agree on closing #3515 ? He's inactive on Github and the pull is, as paramat already said, partially replaced. |
| 17:50 |
ShadowBot |
https://github.com/minetest/minetest/issues/3515 -- Improve waving definitions by BlockMen |
| 18:06 |
|
troller joined #minetest-dev |
| 18:18 |
sofar |
Krock: fine with me, obviously |
| 18:19 |
nerzhul |
Krock, okay for me |
| 18:20 |
Krock |
good, thanks. |
| 18:21 |
nerzhul |
i will merge #5658 in 1h |
| 18:21 |
ShadowBot |
https://github.com/minetest/minetest/issues/5658 -- Fix`TOSERVER_RECEIVED_MEDIA` handling. by red-001 |
| 19:10 |
nerzhul |
merging #5658 |
| 19:10 |
ShadowBot |
https://github.com/minetest/minetest/issues/5658 -- Fix`TOSERVER_RECEIVED_MEDIA` handling. by red-001 |
| 19:10 |
nerzhul |
done |
| 19:11 |
nerzhul |
i also push Android.mk fix, ShadowNinja broke it when renaming files |
| 19:24 |
|
RobbieF joined #minetest-dev |
| 19:24 |
|
RobbieF left #minetest-dev |
| 19:27 |
|
pandaro joined #minetest-dev |
| 19:27 |
|
pandaro left #minetest-dev |
| 20:08 |
|
paramat joined #minetest-dev |
| 20:12 |
|
Megaf joined #minetest-dev |
| 20:13 |
Hijiri |
#5612 |
| 20:13 |
ShadowBot |
https://github.com/minetest/minetest/issues/5612 -- Allow overriding tool capabilities through itemstack metadata by raymoo |
| 20:48 |
|
ssieb joined #minetest-dev |
| 20:59 |
|
rubenwardy joined #minetest-dev |
| 21:16 |
|
benrob03291 joined #minetest-dev |
| 21:34 |
paramat |
game#1712 seems pointless? |
| 21:34 |
ShadowBot |
https://github.com/minetest/minetest_game/issues/1712 -- Ownable Doors are useless, i can just destroy the blocks around it |
| 21:36 |
sfan5 |
yes |
| 21:38 |
benrob0329 |
Perhaps seperating steel doors from.locked doors would help (aka adding a lock tool) |
| 21:38 |
benrob0329 |
Then you don't have to have a steel door be a locked door |
| 21:39 |
paramat |
nice idea, separate issue though |
| 21:39 |
benrob0329 |
Yeah, realised that after I posted it |
| 21:39 |
paramat |
:] |
| 21:43 |
|
Taoki joined #minetest-dev |
| 21:44 |
Fixer |
nerzhul: i confirm log writing can slow down or even freeze entire game |
| 21:45 |
Fixer |
if thousands of errors are written |
| 21:46 |
red-002 |
^ |
| 21:46 |
|
benrob03291 joined #minetest-dev |
| 21:46 |
Fixer |
1712 is almost pointless in singleplayer, but you can mitigate it by using walls out of locked trapdoors or chests |
| 21:46 |
nerzhul |
Fixer, oh you read my issue :p |
| 21:46 |
Fixer |
yes |
| 21:47 |
Fixer |
there was a bug in one PR i tested that printed like few thousands message per minute, and it was _very noticable_ |
| 21:47 |
Fixer |
horrendous stutter |
| 21:47 |
nerzhul |
maybe i should talk about it on saturday, and decide if we do it for 0.4.16 or not, not sure if it's critical |
| 21:47 |
nerzhul |
(in regular cases i mean) |
| 21:48 |
|
benrob0329 joined #minetest-dev |
| 21:49 |
|
troller joined #minetest-dev |
| 21:54 |
|
benrob0329 joined #minetest-dev |
| 21:55 |
|
benrob03291 joined #minetest-dev |
| 21:56 |
|
Megaf_ joined #minetest-dev |
| 22:06 |
|
benrob0329 joined #minetest-dev |
| 22:08 |
Fixer |
please save my middle scroll wheal from breaking, introduce faster crafting of whole stack |
| 22:10 |
|
Taoki joined #minetest-dev |
| 22:19 |
|
XtremeHacker joined #minetest-dev |
| 22:20 |
|
Sokomine joined #minetest-dev |
| 22:20 |
|
destt_ joined #minetest-dev |
| 22:22 |
|
kilbith joined #minetest-dev |
| 22:26 |
Taoki |
paramat: Please see my last comment. Both #5666 and #3496 are now closed. This is only shutting down the discussion again. Could you keep either of the two open please? |
| 22:26 |
ShadowBot |
https://github.com/minetest/minetest/issues/5666 -- Extruded drawtypes |
| 22:26 |
ShadowBot |
https://github.com/minetest/minetest/issues/3496 -- Add 3d_plantlike optional drawtype by RealBadAngel |
| 22:26 |
Taoki |
You closed my issue saying it's a duplicate of another issue which is also closed. Fair enough there... but that makes both issues closed. |
| 22:26 |
paramat |
oh hmm .. |
| 22:27 |
paramat |
will reopen |
| 22:27 |
Taoki |
Thanks. Just yeah, think it should at least be rediscussed... hard for that to happen if everything stays closed. |
| 22:27 |
paramat |
it just seems the discussion is ongoing in 3496 so didn't see a good reason to duplicate it |
| 22:27 |
|
destt joined #minetest-dev |
| 22:27 |
paramat |
but ok |
| 22:28 |
Taoki |
#3496 has been dead for over an year, so doubt it can still count |
| 22:28 |
ShadowBot |
https://github.com/minetest/minetest/issues/3496 -- Add 3d_plantlike optional drawtype by RealBadAngel |
| 22:29 |
Taoki |
Also posted another comment in mine, explaining why I definitely think this is good as a separate feature and using mesh won't do (way way way less flexible and more limited) |
| 22:29 |
Taoki |
Thanks |
| 22:29 |
Taoki |
(saw it's open now) |
| 22:36 |
kilbith |
no definitely not this |
| 22:36 |
kilbith |
if you have an image of 512px, this will a shitload of vertices to render |
| 22:37 |
kilbith |
*will be |
| 22:37 |
paramat |
yes see my comment, RBA's implementation is unusable |
| 22:38 |
paramat |
and nodeboxes and meshes will soon wave |
| 22:38 |
paramat |
even with 16px textures all those pixel cubes will be too much |
| 22:39 |
|
Fixer_ joined #minetest-dev |
| 22:40 |
kilbith |
I agree |
| 22:52 |
|
proller joined #minetest-dev |
| 22:54 |
|
proller__ joined #minetest-dev |
| 23:03 |
Fixer |
I have very serious question |
| 23:04 |
Fixer |
I'm trying to backtrace a crash, so I'm playing via gdb and debug build |
| 23:04 |
Fixer |
catched something amazing |
| 23:04 |
Fixer |
when someone very far away explodes large TNT claster, I see this https://i.imgur.com/AY6tQ84.png |
| 23:05 |
Fixer |
both server and client are at 0.4.15-dev |
| 23:05 |
Fixer |
this is very strange |
| 23:05 |
|
RoaryTiger joined #minetest-dev |
| 23:05 |
Fixer |
explosion was _very_ far away and my v_range was 100 at most |
| 23:06 |
paramat |
mainloop dtime spike > 1s |
| 23:06 |
Fixer |
paramat: during very far huge TNT explosion by other person |
| 23:06 |
paramat |
mainloop_other near 1s |
| 23:06 |
Fixer |
paramat: this is debug though, it was at 80fps during test |
| 23:06 |
Fixer |
but when they exploded TNT it went crazy |
| 23:06 |
paramat |
interesting |
| 23:06 |
Fixer |
WTFFF???????? |
| 23:07 |
Fixer |
I receive block updates for their explosions??? |
| 23:07 |
Fixer |
i see network packets |
| 23:07 |
Fixer |
157... |
| 23:07 |
Fixer |
explosion was at -700 |
| 23:07 |
Fixer |
deep |
| 23:07 |
Fixer |
far from me |
| 23:07 |
Fixer |
notice how other people say Lag |
| 23:07 |
Fixer |
this is not coincidence and I seen this many times on it |
| 23:08 |
paramat |
'fell off my cloud' hehe |
| 23:08 |
Taoki |
<kilbith> if you have an image of 512px, this will a shitload of vertices to render |
| 23:08 |
Taoki |
Yes, this shouldn't be abused for overly high-res textures. Then again, the same can be done for wielditems already. Perhaps in addition, a customizable texture size limitation for extrusion wold make sense. |
| 23:08 |
paramat |
worth an issue |
| 23:08 |
Fixer |
yes |
| 23:08 |
Fixer |
very very strange |
| 23:08 |
Fixer |
and serious |
| 23:08 |
paramat |
Taoki even 16px is a problem |
| 23:09 |
Fixer |
debug.txt was clear |
| 23:09 |
Taoki |
Anyway, automated extrusion is the only way to have this feature for multiple textures, without making it a nightmare. Manually modeling in the cubes is difficult and requires changing the model with each update. I'd say this is the most important reason why it shouldn't be shot down but considered. |
| 23:09 |
paramat |
and think about how a few thousand nodeboxes affect performance |
| 23:09 |
Taoki |
I can see that... obviously it's a setting, doesn't even need to be on by default. |
| 23:10 |
|
benrob0329 joined #minetest-dev |
| 23:10 |
Taoki |
Technically normal cube voxels affect performance too. Though actually, nodeboxes don't merge with one another AFAIK, so they're worse. |
| 23:10 |
paramat |
almost no-one's system would cope even with 16px |
| 23:10 |
Taoki |
Would be awesome if overlapping flat faces in neighboring nodeboxes could be detected, to remove unneeded surfaces and optimize them. |
| 23:11 |
Taoki |
I'd be willing to try it :P Depends on how much grass the biome spawns also. |
| 23:11 |
paramat |
so anyway the individual pixel cubes are not doable, but a continuous waving mesh is |
| 23:11 |
paramat |
looks better too |
| 23:12 |
paramat |
best would be to automatically extrude a texture and convert it into the existing meshnode format, as nodeboxes are |
| 23:13 |
paramat |
so a completely new implementation is needed |
| 23:14 |
Fixer |
holy shit |
| 23:14 |
Fixer |
confirmed on nondebug |
| 23:15 |
Fixer |
you won't believe how nice it looks |
| 23:15 |
Fixer |
https://i.imgur.com/ychuKNX.png nondebug |
| 23:18 |
|
troller joined #minetest-dev |
| 23:19 |
Taoki |
Individual pixel cubes can go, that's fine by me. I'm more interested in simple extrusion like for the wielditem. |
| 23:22 |
paramat |
does it depend on whether you are facing the explosion? |
| 23:23 |
Fixer |
paramat: explosion was like 1000 nodes away |
| 23:23 |
Fixer |
paramat: i had vrange at 100 and 160 facing in different directions, and i was still spammed with net packets during explosion |
| 23:23 |
Fixer |
this is worrying |
| 23:23 |
Fixer |
paramat: amazing part is other people confirmed this |
| 23:24 |
Fixer |
paramat: back then I tested that strange fps drop bug, and when it suddenly kicked it - i observed sudden and persistent fps drop and ... other person saying "lag" "low fps" in chat |
| 23:24 |
paramat |
did the first test freeze your FPS for a second? do these affect FPS? |
| 23:25 |
Fixer |
paramat: yes |
| 23:25 |
paramat |
weird |
| 23:25 |
Fixer |
paramat: in debug build it went from 80 to 8 |
| 23:25 |
Fixer |
and back |
| 23:25 |
Fixer |
non debug build it went from 10ms to 59ms, Carl |
| 23:25 |
Fixer |
mainloop_dtime spike |
| 23:26 |
Fixer |
this is very strange, since explosion was at -770 deep, and i was at different area on sea level |
| 23:26 |
paramat |
so dtime determines FPS? |
| 23:27 |
paramat |
it's almost like you are seeing stuff you should not |
| 23:27 |
Fixer |
mainloop_draw |
| 23:27 |
Fixer |
is for drawing iirc |
| 23:27 |
paramat |
ok |
| 23:27 |
Fixer |
need to read that explanation again |
| 23:27 |
Fixer |
paramat: mainloop_dtime is draw + other lags |
| 23:30 |
paramat |
in your 1st test 'other' was relatively huge, 2nd, relatively small, but peaks align |
| 23:30 |
Fixer |
yes, that was on debug build, that is slower |
| 23:30 |
paramat |
i see |
| 23:31 |
Fixer |
it went from 80 to 8 fps... game almost freezed |
| 23:31 |
Fixer |
the question is why? i'm just a client, wth I receive some packets from some very far event |
| 23:37 |
Fixer |
https://github.com/minetest/minetest/issues/5667 |
| 23:37 |
paramat |
affecting FPS too. c55 will be interested in this |
| 23:39 |
|
YuGiOhJCJ joined #minetest-dev |
| 23:39 |
paramat |
celeron55 this is interesting https://github.com/minetest/minetest/issues/5667 |
| 23:39 |
paramat |
(for when he awakes) |
| 23:46 |
Fixer |
paramat: guy did >30 tnt explosion -10000 deep, result https://i.imgur.com/n6Try2g.png |
| 23:46 |
Fixer |
almost 100ms spike |
| 23:46 |
Fixer |
wtfuck |
| 23:47 |
VanessaE |
Fixer: did you by chance visit that "1000 nodes away" location recently, before the explosion test? |
| 23:47 |
Fixer |
VanessaE: no |
| 23:47 |
Fixer |
NO |
| 23:47 |
VanessaE |
like went there, came back to there you are now, then told the other guy to set off the TNT? |
| 23:47 |
Fixer |
VanessaE: _NO_ |
| 23:47 |
VanessaE |
ok ok |
| 23:48 |
Fixer |
i logined to my home, put minimum vrange, he blows up TNT >10 000 nodes away from me, huge spike appears |
| 23:48 |
VanessaE |
ok |
| 23:50 |
Fixer |
i have some data from verbose debug |
| 23:51 |
Fixer |
2017-04-27 02:44:46: INFO[Main]: cmd 70 count 85 |
| 23:51 |
Fixer |
2017-04-27 02:44:46: INFO[Main]: cmd 71 count 275 |
| 23:53 |
|
twoelk|2 joined #minetest-dev |
| 23:54 |
Fixer |
updated |
| 23:58 |
Fixer |
good night @ have fun figuring this out since this seems like having a big implications |
