Minetest logo

IRC log for #minetest-dev, 2017-05-09

| Channels | #minetest-dev index | Today | | Google Search | Plaintext

All times shown according to UTC.

Time Nick Message
00:25 paramat ok i'm going to revert those collisionbox and stepheight commits, the proper implementation is not easy and will take time to work on
01:09 paramat there it is #5733
01:09 ShadowBot https://github.com/minetest/minetest/issues/5733 -- Revert custom player collision box and step height commits by paramat
01:11 Warr1024 joined #minetest-dev
01:19 octacian_ joined #minetest-dev
01:39 Wayward_One #5734
01:39 ShadowBot https://github.com/minetest/minetest/issues/5734 -- Android crash on startup
02:31 paramat #5733 tested will merge very soon
02:31 ShadowBot https://github.com/minetest/minetest/issues/5733 -- Revert custom player collision box and step height commits by paramat
02:36 paramat nore ^ if you're around
02:37 nore paramat: couldn't we just make a protocol bump and set these if protocol is too old?
02:39 paramat yes, i was discussing with TeT about this, a protocol bump seems good, but a fix is needed urgently, the correct implementation will take time
02:40 nore hmm ok
02:40 nore try not to revert everything though
02:40 nore keep the #defines
02:41 nore use PLAYER_DEFAULT_STEPHEIGHT
02:42 paramat we don't need to keep the defines
02:42 nore it's better code style and it will minimize the changes needed again when doing the new implementation
02:43 paramat PLAYER_DEFAULT_STEPHEIGHT is not used if the PR is reverted
02:44 nore well, you would do s/0.6f/PLAYER_DEFAULT_STEPHEIGHT/
02:45 nore (in localplayer.cpp)
02:45 paramat risky, other literals could be 0.6f
02:45 nore I mean, only at the place when you do the change back
02:46 nore ie. line 348
02:46 paramat ah
02:47 paramat well i either merge this now or someone else does what you suggest, i've spent 2.5 days work on these bugs due to hasty merging on Sat and i've had enough now :]
02:47 paramat *working
02:47 nore we also have that ugly #IF __ANDROID__ thing that I don't know by what it should be replaced (I mean, if you change the stepheight, what should happen on Android?)
02:47 paramat do you want to take over on this PR and merge it tonight?
02:47 nore yeah, just revert, we will figure a better implementation
02:48 nore just merge your thing
02:48 paramat ok
02:48 paramat thanks
02:48 paramat could you +1?
02:48 nore hmmm github seems broken
02:49 paramat no prob will edit later, PR is tested
02:50 paramat we should be more careful at the next meeting
02:52 nore https://www.reddit.com/r/github/comments/69zqxu/is_it_just_me_or_is_js_on_github_broken/ <-- ok so I'm not the only one experiencing github problems :/
02:53 nore well the problem is that the PRs were agreed upon but not really tested :/ we should indeed be more careful next time
02:55 paramat merging
03:01 paramat done. yes that android increase of stepheight should not occur when not touching ground
03:04 nore Yes but even so
03:05 nore If we increase step height, how should the step height on Android?
03:11 octacian__ joined #minetest-dev
03:12 halt_ joined #minetest-dev
03:12 Grandolf joined #minetest-dev
03:14 paramat ah i see
03:14 Hunterz joined #minetest-dev
03:17 paramat maybe .. for 'touching ground', android should have a default of 1.1, if customised use whatever is set
03:34 sofar ugh what a day
03:34 sofar is it tuesday yet?
03:36 VanessaE not for 24 more mins.
03:47 sofar I wish. Still 3 more hours of torture remaining
04:00 benrob0329 Aaand its Tuesday
04:17 sofar !seen pilzadam
04:17 ShadowBot sofar: I haven't seen pilzadam in #minetest-dev.
04:18 sofar I vote for removing pilzadam as minetest-dev
04:18 sofar he hasn't been around for 6+ months
04:18 sofar https://github.com/PilzAdam
04:18 sofar 0 contributions in the last 12 months
04:21 sofar @5735
04:21 sofar #5735
04:21 ShadowBot https://github.com/minetest/minetest/issues/5735 -- Move Pilzadam to previous developers. by sofar
05:02 nerzhul joined #minetest-dev
05:14 Hunterz joined #minetest-dev
05:56 paramat joined #minetest-dev
05:58 red-001 #5736
05:58 ShadowBot https://github.com/minetest/minetest/issues/5736 -- Fix android main menu. by red-001
06:21 Fritigern joined #minetest-dev
06:56 nerzhul merging #5735
06:56 ShadowBot https://github.com/minetest/minetest/issues/5735 -- Move Pilzadam to previous developers. by sofar
06:57 nerzhul difficult to maintain PR outside 153, this evening we really should look at refcent PR some are mergeable
07:05 burli joined #minetest-dev
07:06 burli Hi
07:17 kilbith joined #minetest-dev
08:27 nerzhul joined #minetest-dev
08:28 nerzhul merging #5720 in 15 mins (3 approvals)
08:28 ShadowBot https://github.com/minetest/minetest/issues/5720 -- Add cancel button to password change menu. by red-001
08:37 Zeno` nerzhul, what's the problem with https://github.com/minetest/minetest/pull/5719/files/513089cbdaf9330ac2e0c1f6577ad51adfd551c9#diff-18513665750ef5adf42b5ec29e14162eR4538 ?
08:38 nerzhul Zeno`, they null terminate string, but the remaining area is undefined
08:38 Zeno` and?
08:38 nerzhul it can be a security problem to access random memory
08:39 nerzhul (random previously used memory)
08:39 Zeno` but the code doesn't access random memory
08:39 Zeno` does it?
08:39 nerzhul the code doesn't permit but when allocating char buffer you should ensure it's properly inited to zero (it can be a problem in BSD systems)
08:40 Zeno` I'm pretty sure that if you had = {0} or even memset() after the buffer allocation the compiler is probably going to optimise it away anyhow
08:41 nerzhul Zeno`, absolutely not
08:41 Zeno` it does on Windows and Linux
08:41 burli nerzhul, just one question: Will old mob APIs become obsolete if you change the API?
08:41 Zeno` which is why there are special (non-portable) functions for setting memory to 0
08:41 nerzhul i already saw FreeBSD and OpenBSD problems when area not is properlyu onitialized, it just crash when you read area (pointer valid)
08:42 nerzhul Zeno`, hopefully in C++11 char buf[4096] = {}; is portable to do bzero :p
08:42 nerzhul burli, there is no mod api in core currently
08:42 nerzhul there is just mod doing crap which current lua entity api
08:42 nerzhul it's slow and crappy, but works like it can
08:43 burli nerzhul, I mean your comment here https://github.com/minetest/minetest/issues/5729#issuecomment-300077859
08:43 nerzhul yes i know
08:44 nerzhul i can just say, wait for next release, we will re-do that properly and more things to have a nice way to handle mobs with proper prediction (example: mob:move_to(x,y,z)) (pathfind calculated client side & server side with ticks for the movement)
08:45 burli So my pathfinder is also obsolete
08:45 nerzhul Zeno`, discard my comment as it's a client side issue and there are no BSD users on desktop :)
08:45 nerzhul keep it somewhere, this can be useful coreside if it works properly
08:46 burli Great. Lots of work for nothing
08:50 nerzhul burli, your pathfinder is in Lua ?
08:51 nerzhul and burli not exactly because you are invited to help us to make real mobs API when we start the project in next release, your PF work is appreciated :)
08:52 burli nerzhul, you remember this? https://github.com/minetest/minetest/issues/5479
08:52 burli What does PF mean?
08:52 nerzhul pathfinding
08:53 nerzhul i remember this discussing and i hope we can see this work at a moment
08:53 kilbith no need to talk C++11 to Zeno`, he knows why
08:53 burli I have absolutely no idea about csm
08:53 burli I don't even know how to run a csm mod
08:54 nerzhul burli, what is the relation between CSM and pathfinding, there is not
08:55 burli I assume that the pathfinder on client side works similar than on server side
08:57 Zeno` kilbith, it's because I'm too old
08:58 Zeno` But I still think the compiler in C++11 is allowed to optimise setting memory to 0 away if it wants to
08:58 kilbith and your cognitivity sucks also
08:58 Zeno` kilbith, that's right. Most days I don't even know who I am
08:58 kilbith lol
09:01 nerzhul Zeno`, i don't know for optimization, but in c++11 you can initialize all array or STL containers with = {}; operator
09:02 nerzhul also you can do some nice things, imagine std::vector<std::pair<int,int>> a;
09:02 nerzhul you can do a.push_back({1,1});
09:02 nerzhul :)
09:02 nerzhul it's easier to write STL containers
09:04 nerzhul 5720 merged
09:04 nerzhul ty red-001
09:05 nerzhul red-001, can you explain me what is core.start in #5736 and what is the bug ?
09:05 ShadowBot https://github.com/minetest/minetest/issues/5736 -- Fix android main menu. by red-001
09:05 nerzhul i see the main menu doesn't launch, but why ?
09:21 burli nerzhul, would it be helpful if I try to port the pathfinder to the client side? Or what are you planning on this?
09:23 nerzhul burli, continue to optimize your work, atm we have 0.4.16 to finish and when the project will be opened you will be welcome to contribute with us about it :)
09:24 burli Ok
09:57 YuGiOhJCJ joined #minetest-dev
10:21 lisac joined #minetest-dev
10:34 Karazhan joined #minetest-dev
10:35 Karazhan joined #minetest-dev
10:36 Karazhan joined #minetest-dev
10:39 Karazhan joined #minetest-dev
10:40 Karazhan joined #minetest-dev
10:42 Darcidride joined #minetest-dev
10:43 Karazhan joined #minetest-dev
10:59 Fixer joined #minetest-dev
11:10 DI3HARD139 joined #minetest-dev
11:23 twoelk joined #minetest-dev
13:12 nerzhul joined #minetest-dev
13:13 RobbieF joined #minetest-dev
13:35 kilbith https://lut.im/H98HJ4FUk0/6Uq3dQKB2UQLqtCk.png
13:35 kilbith wth
13:37 kilbith oh, this only happens when pipeworks's enabled
13:37 VanessaE it's not pipeworks, per se.
13:37 kilbith cc: VanessaE, nore
13:37 VanessaE it's the UV-mapped chest textures in mt_game
13:37 VanessaE I have a PR open to put them back right.
13:37 nerzhul kilbith, it's future art :p
13:37 kilbith look, when I disable pipeworks, the chests look back normal
13:38 VanessaE it's.  not.   pipeworks.
13:38 VanessaE see above.
13:39 kilbith I'm not saying it's pipeworks' fault
13:39 kilbith but there's a relation between MTG' chests and this mod
13:39 VanessaE I know.
13:40 VanessaE see my PR in mt_game for a fix.
13:41 kilbith how about adapting downstream stuff to upstream changes then?
13:41 kilbith yeah, I know breaking mods sucks still
13:42 VanessaE because upstream's change was a stupid change (imho)
13:43 kilbith that was certainly not worth the breakage, I agree
13:43 kilbith MTG turned out to be a cesspit long time ago, so meh
13:44 kilbith since PilzAdam left, basically
13:45 kilbith fortunaly I'm paid to work on that shit
13:53 octacian joined #minetest-dev
14:00 cx384 joined #minetest-dev
14:01 * twoelk looksup cesspit
14:07 DS-minetest joined #minetest-dev
14:32 behalebabo joined #minetest-dev
14:39 Player_2 joined #minetest-dev
14:46 Zeno` should MTG be removed from the minetest project?
14:47 kilbith +1
14:47 Zeno` because I am sick to death of seeing notifications about in when I join github when I can't even comment on them
14:47 kilbith let's put it directly in competition with the other subgames
14:47 DS-minetest since you get mtg together with core i dont think so
14:48 Zeno` DS-minetest, yes but core devs cannot have a say about it
14:48 Zeno` so, it should be a different project IMO
14:48 kilbith MTG does not deserve to be in its ivory tower
14:48 DS-minetest hmm, can't you just unwatch mtg?
14:49 Zeno` DS-minetest, not without unwatching engine changes (afaik)
14:49 Zeno` but that's beside the point. Either MTG is part of minetest or it's not
14:50 twoelk hm - if mtg is so bad what should a replacement look like or is the idea ti ship the engine gameless?
14:50 Zeno` and at the moment it's not. So it should be a separate project
14:51 Zeno` twoelk, I have commit privs to MTG but I am not "allowed" to comment or make commits to it
14:51 Zeno` this doesn't make sense
14:51 Zeno` I didn't say it was bad
14:51 twoelk er, parsing error - ??
14:52 twoelk you have privs that are not accepted?
14:52 twoelk can't be the fault of mtg
14:52 Zeno` no. I can do anything I like with the MTG repo. The only reason I don't is out of respect
14:53 Zeno` but... there is are guidelines as to when I (or other coredevs) can intervene
14:53 twoelk guess I missed something :-D
14:53 Zeno` are no guideline*
14:54 Zeno` and that does not seem right
14:54 Zeno` either MTG is part of minetest or it's not
14:54 twoelk ah, so the fault of missing working respectable guidelines
14:55 Zeno` is MTG part of minetest?
14:56 twoelk well being a subgame it might be a subpart in contrast to something parallel
14:57 twoelk but - for me nondev it always was part of mt
14:57 Zeno` Ok, well core devs should be able to contribute to it
14:58 Zeno` and discuss it
14:58 Zeno` etc etc etc
14:58 Zeno` I'm not even sure if celeron55 is allowed to push a commit to MTG
14:59 Zeno` which is kind of crazy
14:59 nerzhul Zeno`, i don't receive MT game notifications because i'm not in the group
14:59 VanessaE mt_game should be treated as part of Minetest. Full Stop.
14:59 twoelk I always thought mtg schould be the first example reference for a modder and as such the code of it should showcase how things are prefered to be done
14:59 nerzhul Zeno`, minetest_game is partially a part of minetest distribution (linux packages, Windows, Android)
15:00 Zeno` nerzhul, so why aren't we involved with it?
15:00 nerzhul for a modder minimal should be enhanced to add examples (i added recently a pseudo unit test in minimal calling player attributes API
15:00 VanessaE nerzhul: "minimal" should be deleted.
15:00 nerzhul personnaly i'm too far from this project, i want to have core side features and many minetest_game discussions are not on my interests
15:01 nerzhul VanessaE, no
15:01 VanessaE there's really no reason for it to exist anymore
15:01 nerzhul minimal it's a developper test
15:01 Zeno` minimal is useful for doing valgrind tests etc
15:01 nerzhul it's nice, minimal content, you can test easily each development
15:01 VanessaE yes, which is exactly why it needs to die
15:01 nerzhul VanessaE, maybe we should drop it for end user distribution but not for us
15:01 twoelk minmal could use some more testing possabilities and tools than
15:01 VanessaE because if you aren't using mt_game for dev testing, you're doing something wrong.
15:01 Zeno` since we don't have any say in MTG then it's not suitable for testing
15:02 nerzhul and i agree with Zeno` valgrind & callgrind like minimal because there are few lua calls permitting to focus on core performance
15:02 nerzhul as many said, minetest is an engine, and when testing an engine i don't want to be polluted by MT game ABM
15:03 VanessaE Zeno`: my point is, if you don't use mt_game for testing, you are not testing with an environment the average user will have in front of him/her.
15:03 VanessaE and that's bad.
15:03 celeron55 minimal should probably be renamed to "example" and the content should be totally redone to better fit the purpose instead of being a snapshot of what minetest_game essentially was at some point
15:03 twoelk still minimal could use more ingame info like an expanded F5
15:03 celeron55 and it shouldn't be distributed to end users
15:04 celeron55 VanessaE: it's sometimes essential to not run an entire game as debugging modes are slower, especially something like valgrind
15:04 celeron55 and you get more control in what happens
15:05 celeron55 some random game content you don't care about when developing won't crash the engine
15:05 VanessaE celeron55: I can respect that.
15:05 VanessaE test with "example" or "minimal" or whatever you wanna call it, then test again with mt_game./
15:06 VanessaE THAT I wouldn't be against.
15:06 celeron55 nobody is arguing against that
15:06 VanessaE good. :)
15:06 VanessaE just sometimes one gets the impression that testing stops at "minimal" more often than not.
15:07 Fixer Zeno`: in notifications just press arrow above to mark them as read and you are done
15:07 nerzhul no, i don't want to test with mt_game, if there is a bug in mt_game it's mt_game problem, minimal should have one example of each function, not hundreds :)
15:08 VanessaE nerzhul: unless the hypothetical bug is in the engine and isn't revealed by testing under minimal.
15:08 VanessaE expecting minimal to have examples of every possible API call is probably excessive
15:08 VanessaE celeron55: it would seem at least one person is arguing against it ;)
15:08 cx384 joined #minetest-dev
15:09 Zeno` VanessaE, there is no point testing with MTG because we're not allowed to fix a bug in it even if we find one
15:09 Zeno` so minimal is what we use
15:09 * twoelk thinks both, minimal and mtg could be improved to serve their purpose better
15:09 VanessaE that needs to change, but by "bug" I mean engine bugs
15:09 VanessaE anything mt_game exposes that minimal doesn't.
15:12 Zeno` Fixer, thanks. I didn't realise how to make all the notifications on github marked as read :p
15:14 Zeno` THe point is... either MTG is part of minetest or it isn't
15:15 Zeno` if it's not then it should be removed from the minetest project on github
15:17 Zeno` if it *is* part of the minetest then any core dev should be able to contribute to it
15:18 twoelk I rather like it within the mt project as you then have one place you can download everything to play. Putting it somewhere else would be too confusing for those new to the project.
15:18 Zeno` not a select few chosen by who knows what method
15:18 rubenwardy joined #minetest-dev
15:19 twoelk i don't understand why you don't let yourself get added to the mtg devs then
15:19 rubenwardy Zeno`: you can select "ignore" on the watch combo box on the Minetest repo
15:19 Zeno` twoelk, I am
15:19 Zeno` twoelk, theoretically
15:19 rubenwardy *Minetest Game
15:20 rubenwardy http://dev.minetest.net/minetest_game_Development
15:20 rubenwardy "Note: To prevent problems with the development like before Minetest Game is maintained by only seven people"
15:20 Zeno` rubenwardy, that's old and out of date
15:20 rubenwardy I don't know what the problems are
15:21 rubenwardy well, http://dev.minetest.net/User:Rubenwardy/Minetest_Game_Development
15:22 rubenwardy also the original link isn't out of date, all still applies
15:22 Zeno` The problem is that onyl SOME core devs are "allowed" to comment or approve/disapprove MTG PRs
15:22 Zeno` and that is not right
15:23 rubenwardy only mtg core devs can vote, but all can comment
15:23 Zeno` who are the mtg core devs and how are they chosen?
15:23 Zeno` that is my point!
15:23 rubenwardy I don't care either way
15:23 Zeno` and based on your comment mtg is a different project. right?
15:24 rubenwardy the same as core devs are chosen - by existing core dev proposals, or self-proposing, and celeron55 ultimately
15:24 twoelk maybe you failed to state your interest at some vital point while things where organized?
15:24 Zeno` rubenwardy, that is not true. I can commit to mtg
15:24 rubenwardy no you can't
15:24 rubenwardy you have github rights
15:24 Zeno` I cna
15:24 Zeno` yeah, because it's part of minetest
15:24 rubenwardy but you would be removed from the minetest organisation if you abused the access righs
15:25 Zeno` really?
15:25 Zeno` where is that stated?
15:25 rubenwardy same as if you merged a PR to minetest/minetest without approval
15:25 rubenwardy ultimately anyway
15:25 Zeno` so it's a different project then
15:25 rubenwardy how are you defining project
15:26 Zeno` "minetest" as a whole
15:26 celeron55 the mtg team was made smaller due to the rather extreme design-by-committee flaws caused to it when all core devs were involved in it
15:26 celeron55 making a game with a huge programming-focused team doesn't quite work
15:26 celeron55 or didn't, in fact
15:27 rubenwardy maybe the compromise here is to let all core devs do patches etc, and approve implementation, but only mtg devs do concept/game direction
15:27 celeron55 i have never chosen the developers of mtg
15:27 celeron55 i just threw it to the people who insisted in developing it after it was officially shut down
15:28 celeron55 that's where the current mtg team originates from
15:29 celeron55 it was re-opened and given to blockmen and some others that forked minetest_next from it
15:31 Zeno` celeron55, so who are the non-core devs able to commit to it
15:31 rubenwardy I agree with making Minetest Game less of the status-quo however. The real power of Minetest isn't the mods imo, but the subgames
15:31 Zeno` and if that's the argument then why are some core devs allowed to and not others?
15:32 rubenwardy Current MTG devs are: sfa_5, no_e, Shado_Ninja, par_mat, s_far, rubenwardy, and Kr_ck
15:32 celeron55 the mtg team chooses the mtg team and it's their responsibility to publish who the team is
15:33 celeron55 the purpose wasn't to put random people in an ivory tower but to give a good place for dedicated people to work on the official subgame that they wanted to work on
15:34 celeron55 without all the hassle of trying to convince the entire core team on individual game ideas
15:35 Zeno` I guess the problem is that with some core devs also being part of the game core devs (all of them?) then the direction of the engine is influenced by their opinions
15:35 celeron55 sounds like a valid concern
15:38 celeron55 maybe it could be enough if the mtg team is informed of the possible bias they have so that they can be mindful of it
15:38 Hunterz joined #minetest-dev
15:40 celeron55 i think it would be good if the mtg team had many non-coredevs
15:41 Zeno` I think it would be good if it had no core (engine) devs, but *shrug*
15:41 celeron55 that's a difficult change to make in a short timespan, but maybe in the long term
15:42 kilbith when you add a feature in MTG, it's important to know whether the engine's ready for it
15:42 rubenwardy mtg dev is practically ran by paramat atm, so that would exclude him
15:42 kilbith that's why it's important to be core-dev in order to maintain a subgame
15:43 Zeno` kilbith, that's not possible atm
15:44 kilbith there's real game designers, etc. etc.
15:44 Zeno` rubenwardy, and that is a problem? paramat would just have to choose his preferred role
15:44 kilbith *there's no
15:45 celeron55 it seems either minetest has literally zero appeal to game designers or they all find something better to do
15:45 kilbith exactly
15:46 Zeno` kilbith, and that's not 100% true. The game maintainers if they need something should request that core engine devs implement whatever they might need. Not the same people saying "we need this" as those who add it to the engine
15:46 kilbith that's right
15:48 kilbith I often read paramat openly refusing any real artist to have an authority on the assets that goes in MTG or not
15:50 kilbith that kind of behavior should stop, people should stick on their domain of expertise and do not self-improvise artist or whatever
15:52 twoelk it often seems that while some devs may to seem unable to move when something is first thrown at them they later after some consideration may decide to change their initially negative stand - and paramat is no exception to this
15:52 kilbith actually he sucks so hard in C++ and Lua that I don't even know why he has an official role in the upstream repos in the first place
15:53 kilbith he's just specialist of a small niche: mapgen
15:53 kilbith and that dofus can vote on anything
15:54 twoelk he does get things happening - sort of
15:56 twoelk and I disagree that the project can only be led by a master coder
15:56 Fixer about permanent fps drop bug - catched it again I think, and literally at the same time (start of the bug) some random person says "lag" in chat (probably meaning fps drop down), not the first time i see this
15:56 twoelk actually that might not be all that usefull to keep <a bunch of lunatics> focused
15:57 Zeno` there is no need for the project to be led by a master coder
15:57 Fixer debug question/begging: please send me a code to display number of particles drawed in F5 graph...
15:57 kilbith there's no need for the project to be led by an incompetent
15:58 Zeno` I don't think anyone has suggested that
15:58 kilbith or at least the incompetent could have the decency to stick on the mapgen stuff and that's all
15:58 twoelk what response are you expecting from using such a phrase?
15:58 kilbith I'm talking to the non-developers here, already
15:58 Zeno` A person who leads a project relies on people other than themselves
15:59 kilbith *not talking
15:59 Zeno` And, I'm going to say it... paramat is not suitable for that role
16:00 Zeno` it's the reason I have not contributed at all in the last 3 months (apart from some very small comments on PRs)
16:00 kilbith yup
16:00 kilbith you're just one of the numerous people disgusted of contributing because of him
16:01 twoelk maybe be polite enough to discuss this further when he is around
16:01 kilbith oh he read us currently, no worry
16:01 Zeno` twoelk, timezones
16:03 Zeno` A recent example. Paramat saying that the ternary conditional operator needs parenthesis around the condition -- i.e. (a == b) ? 1 : 0
16:03 Zeno` that's just silly
16:03 Zeno` (bad example I know)
16:03 kilbith I have plenty of examples in mind about his silliness
16:03 Krock joined #minetest-dev
16:03 nerzhul it's not needed, but i agree taht for readability
16:04 Zeno` nerzhul, it's harder to read
16:04 kilbith he don't even know what compiler he uses
16:04 rubenwardy I find it easier to read
16:04 nerzhul Zeno`, i think it's more taste, i don't complain :)
16:04 rubenwardy however it's a nonissue
16:04 rubenwardy and not something I'd comment on
16:05 Zeno` nerzhul, that was a contrived example. In general parenthesis where they are not needed makes things harder to read
16:06 Zeno` but, meh
16:06 kilbith the essential of his "reviews" is just "spaces around the operator please"
16:06 Hijiri nerzhul: there are plans for engine NPCs?
16:06 nerzhul Hijiri, in sofar and my head yes
16:06 Zeno` you may as well write x = (a) + (b); following that logic lol
16:06 Hijiri nerzhul: alright
16:07 Hijiri so I guess just discussions then and no github issue or anything I could read?
16:07 nerzhul more than just NPC, i have a server side spell API on my fork, which need to be ported to MT and improved with client side abilities
16:07 nerzhul Hijiri, no discussion about this moment, we need to merge many PR for 0.4.16 PR then no more plans atm
16:07 Hijiri alright
16:08 Hijiri client side abilities meaning a specific "abilities" mechanic?
16:08 nerzhul when release will be done i will open projects for: NPC and C++11 switch
16:08 Hijiri or just "client stuff"
16:08 nerzhul Hijiri, imagine: npc:cast_spell('default:fireball', {pos})
16:09 Hijiri what do you mean by client side abilities though?
16:09 Hijiri stuff done through CSM?
16:09 Hijiri I feel like it would be nice to have a way for players to activate abilities without using items
16:10 Hijiri (or using control tables/bitmaps which will get saturated very quickly)
16:12 Hijiri I don't know if that is part of what you meant, but it would be useful if the player is going to be casting spells
16:12 Zeno` I take back my words about paramat. The only issue I have with him is that he seems to control both the engine AND MTG atm
16:13 nerzhul Hijiri, spells and NPC doesn't need CSM
16:13 Zeno` and I don't think that's right
16:13 Zeno` *shrug*
16:13 Hijiri nerzhul: did you mean some kind of engine change then?
16:13 nerzhul Hijiri, maybe we will have a on_cast_spell and on_spell_hit callbacks in CSM, but it's not the first goal, it will be after having a nice NPC API
16:13 Zeno` I know he cares
16:13 Zeno` But sometimes his views are kind of crazy
16:13 nerzhul NPC will require a very huge protocol enhancement to add new features to make NPC entities work withotu lag
16:14 Hijiri alright
16:14 nerzhul instead of telling client (like now) move_direction,acceleration,blah
16:14 nerzhul a goal is to use waypoints: move_to (x,y,z,speed)
16:14 Krock https://github.com/minetest/minetest/issues/3507#issuecomment-300182387  redundant check of a boolean value. It will be true afterwards anyways.
16:14 Hijiri I just think that maybe if there are engine changes specifically for spell stuff, then maybe it would be stuff that can be generalized beyond spells
16:14 Krock ^ nerzhul
16:14 nerzhul and client will calculate path for entities (server too for sending regular new updates)
16:15 Hijiri nerzhul: what if NPCs move in a non-waypointy way sometimes?
16:15 nerzhul Hijiri, yeah, i tell spell, because i'm very World of Warcraft API aware, i maintained a World of warcraft private server for three years 3 years ago
16:16 nerzhul Hijiri, i can't answer all your questions, but we will try to have smooth server side API reducing client entity lag and increasing accuracy
16:16 nerzhul atm it's too early, many PR and bugs should be solved for release
16:16 Hijiri alright, thanks
16:16 Hijiri probably it will be up on Github for discussion later on anyway, I'll be able to comment there
16:17 nerzhul Zeno`, yeah, he's very strange, on the knockback PR he said: NO never useless... etc, and after told, why not in fact
16:17 Hijiri actually not "probably" since you explicitly said you would make a project for it
16:17 nerzhul yeah, just wait for this release, we have some nice PR to merge (before 21st may)
16:17 nerzhul 12 days before feature freeze guys
16:17 Hijiri hope mine gets in
16:17 nerzhul nice PR should be discussed this saturday and merged/enhanced
16:18 Hijiri I stopped bothering to mergebeg every day since it's in the milestone anyway
16:18 nerzhul and recent PR, stop with old PR for 2 weeks please (ShadowNinja)
16:18 nerzhul Krock, i don't think a boolean write is expensive, yes
16:18 nerzhul we are not at this optimization point :p
16:19 nerzhul (but i don't think it's useful to add it to patchset)
16:19 nerzhul i should go, Krock can you merge #5736 in some minutes ?
16:19 ShadowBot https://github.com/minetest/minetest/issues/5736 -- Fix android main menu. by red-001
16:19 Krock nerzhul, yes sure, too trivial. Just seen it and replaced :3
16:19 Krock sure.
16:19 nerzhul ty
16:20 Krock merging in 10 minutes
16:20 nerzhul i hope sfan5 will finish the private metas before feature freeze, it's very important to have this to reduce CSM unneeded lookups in metas for server owners :)
16:21 sfan5 the suggestion is not adressed right
16:21 sfan5 ?
16:21 sfan5 i should probably respond to it
16:22 red-001 nerzhul, #5736 re-adds lines that got accidently removed in https://github.com/minetest/minetest/commit/43d1f375d18a2fbc547a9b4f23d1354d645856ca#diff-22c832bbf307849c913965da6a862b49L192
16:22 ShadowBot https://github.com/minetest/minetest/issues/5736 -- Fix android main menu. by red-001
16:22 nerzhul red-001, i see :)
16:22 nerzhul i missed them
16:23 nerzhul sfan5, i think the remaining point is static_cast to dynamic_cast, else other parts are okay to me
16:23 nerzhul see you later
16:23 sfan5 oh
16:33 Fixer currently playing on mp server, hanging out on my base, memory usage slowly goes up it seems, memory leeks? will look more
16:36 Fixer is there difference on how debug/release builds receive network packets?
16:39 Krock Fixer, only 1 match for NDEBUG in network/, which changes the logging but not the packet handling
16:50 Fixer checked my suspicion, false alarm
17:03 Krock *phew!*
17:07 Fixer this strange fps drop bug sucks so much, i have no idea how to even catch it on debug build, since fps degenerates to very low points
17:15 nerzhul joined #minetest-dev
17:17 nerzhul ty Kray
17:17 nerzhul Krock,
17:17 Krock hehe :P
17:18 nerzhul red-001, can you finish most of your PR before sunday ? feature freeze is at the end of next week
17:18 nerzhul i want #5719 to be merged :)
17:18 ShadowBot https://github.com/minetest/minetest/issues/5719 -- Rework escape/pause menu by red-001
17:21 red-001 working on it
17:33 pandaro joined #minetest-dev
17:34 pandaro left #minetest-dev
17:49 red-001 updated #5719
17:49 ShadowBot https://github.com/minetest/minetest/issues/5719 -- Rework escape/pause menu by red-001
17:50 halt_ joined #minetest-dev
17:51 Grandolf joined #minetest-dev
17:51 twoelk joined #minetest-dev
17:54 rubenwardy joined #minetest-dev
18:01 nerzhul Zeno`, Krock can you review #5719 another time ? or other coredevs ? it's very nice improvement
18:01 ShadowBot https://github.com/minetest/minetest/issues/5719 -- Rework escape/pause menu by red-001
18:02 nerzhul red-001, for std::string damage = g_settings->getBool("enable_damage") ? on : off; can you verify is a const ref can be used for the std::string please ? if yes, please replace
18:06 red-001 note: I used a size of 500 for the char array which slightly more then twice the size of the english controls text
18:06 nerzhul yeah ignor that :)
18:06 red-001 I think that should probaly be enough for most lanuages
18:06 Krock these could be placed into the stream writing
18:07 Krock the compiler then surely wouldn't be unhappy with it
18:07 red-001 krock that might make translating harder
18:07 Krock no:   os << strgettext("- Damage: ") << g_settings->getBool("enable_damage") ? on : off << "\n"
18:07 Krock but that makes it hard to read
18:07 Krock try using const refs
18:12 Fixer idling on multiplayer server in minimise watching a big area: https://i.imgur.com/wUXu2tc.png (mem usage incraesed +120 mb in 50 minutes)
18:12 red-001 done
18:22 kilbith joined #minetest-dev
18:30 nerzhul approved thanks !
18:32 nerzhul nice job on the project on this release red-001 , you have many time as it seems :)
18:33 nerzhul i will merge it in two hours is no objection (taking time for other core devs to re-look at this)
18:38 rubenwardy made a review
18:38 rubenwardy nice improvement though
18:48 nerzhul :) red-001 it's your time heh
18:49 rubenwardy yeah, good work recently
18:56 kaeza joined #minetest-dev
19:12 Fixer i have impression that minetest is memory leaking
19:13 Fixer just afking and watching the area on MP server away from most people: 20:40 - 930mb, 22:10 - over 1.1 Gb
19:18 Fixer just afking and watching the area on MP server away from most people: 20:40 - 930mb, 22:17 - over 1.2 Gb
19:18 Fixer *corrected
19:20 nerzhul Fixer, it depend on mapblocks loaded/modified, valgrind doesn't report memleaks recently
19:20 nerzhul (it's very hard to track client memleaks too)
19:21 Fixer not much activity in my place, i just afk, and sometimes few 1-2 people go by and do a little play
19:55 Fixer 22:32 - over 1.3 gb
19:55 Fixer hmmm
19:56 Fixer having another suspicion, lets bump it up to 2.4 gb right now and measure fps
20:00 Tmanyo joined #minetest-dev
20:09 Fixer https://i.imgur.com/eJbAZVj.png lets see if it unloads some
20:19 Fixer back to afkying in my place and position, observing some mem unload, but slight: https://i.imgur.com/4YisVZv.png
20:29 kilbith joined #minetest-dev
20:37 Fixer more unload, but weak, still lots of mem used, keeps leaking: https://i.imgur.com/JRK9ZrD.png
20:39 Fixer corrected pic: https://i.imgur.com/d9LIGi0.png (vrange 240 was not rendered properly)
20:41 sfan5 nerzhul: nodemeta pr updated
20:48 nerzhul nice looking
20:49 proller joined #minetest-dev
20:50 nerzhul merging #5738 in ~10 mins
20:50 ShadowBot https://github.com/minetest/minetest/issues/5738 -- minetest.deserialize: Throw error when argument not string by SmallJoker
20:52 nerzhul sfan5, just a question, is there any metadata in buildit which can be private?
20:52 nerzhul or minimal ?
20:54 nerzhul added my approval, but minimal & mt_game should be updated after that heh, nice job
21:20 red-001 updated 5719 to escape strings that could break the formspec
21:22 kilbith_ joined #minetest-dev
21:23 RobbieF left #minetest-dev
21:30 ssieb joined #minetest-dev
21:30 proller joined #minetest-dev
21:42 Karazhan joined #minetest-dev
21:47 chatter29 joined #minetest-dev
21:47 chatter29 hey guys
21:47 chatter29 allah is doinf
21:47 chatter29 allah is doing
21:47 chatter29 sun is not doing allah is doing
21:47 chatter29 to accept Islam say that i bear witness that there is no deity worthy of worship except Allah and Muhammad peace be upon him is his slave and messenger
21:47 chatter29 left #minetest-dev
21:48 * red-001 asks for someone to ban chatte29
21:48 nore so they are really typing it each time, lol
21:49 red-001 I wonder why they always use the same nick
21:49 nore no idea
21:50 nore there
21:51 YuGiOhJCJ joined #minetest-dev
21:54 kilbith__ joined #minetest-dev
21:55 ShadowNinja nore: /mode -bq+q chatter29!*@* *!*@gateway/web/freenode* *!*@gateway/web/*
21:55 nore ShadowNinja: hmm, you are not op there? strange
21:55 ShadowNinja (quiet all webchat, not just freenode webchat)
21:56 nore oh ok
21:56 ShadowNinja Yeah, a bunch of people got oped a while ago, but I wasn't around.
21:57 red-001 most of them got de-oped by now
21:57 nore I guess we could op you if celeron55 is fine with it ^
21:59 nore hmmm chatter29 just joined another channel as chatter30
21:59 nore I guess that's why we don't ban them
21:59 red-001 oh thats because I took his nick
21:59 nore haha
22:00 red-001 it's registered so he could ghost me
22:00 nore something like a year ago they were called just chatter
22:00 nore I guess using the nick has been done more than once
22:02 Calinou ShadowNinja: muting the entirety of webchat is rude, IMO
22:02 Calinou oh, this is -dev, I guess it's ok
22:02 Calinou I thought this was the main channel
22:03 nore Calinou: yeah, it's -dev and freenode webchat was already muted
22:05 Taoki joined #minetest-dev
22:17 Karazhan joined #minetest-dev
22:36 paramat joined #minetest-dev
22:37 paramat "maybe the compromise here is to let all core devs do patches etc, and approve implementation, but only mtg devs do concept/game direction" that happens already, anyone can submit PRs
22:37 paramat (still reading logs)
22:41 paramat "rubenwardy > mtg dev is practically ran by paramat atm, so that would exclude him" oops, i know what you meant here, but this will provoke certain people :]
22:41 paramat i'm active but have no more power than any other mtg dev
22:43 paramat oh dear my fan club has been active
22:44 paramat "I often read paramat openly refusing any real artist to have an authority on the assets that goes in MTG or not" just stating my opinion
22:44 kilbith yep, and your fan club really want to meet you to get things straight with you....
22:44 kilbith we are not living far from each other, you know ;)
22:47 paramat "kilbith > there's no need for the project to be led by an incompetent" ! i'm obviously not leading anything
22:47 ShadowBot` joined #minetest-dev
22:48 paramat "Zeno` > And, I'm going to say it... paramat is not suitable for that role" luckily i don't have that role (master coder)
22:48 kilbith think a moment why several people perceive that you're leading  both projects, there's certainly a kernel of truth there
22:51 paramat "it's the reason I have not contributed at all in the last 3 months" that's a delusional personal 'problem with me'
22:51 kilbith it's not personal, it's a common problem with you
22:53 paramat "A recent example. Paramat saying that the ternary conditional operator needs parenthesis around the condition" just a personal opinion for consistency, a suggestion, every dev has opinions and makes suggestions
22:57 paramat "Zeno` > I take back my words about paramat. The only issue I have with him is that he seems to control both the engine AND MTG atm" that's delusional madness
22:58 paramat also implies all other devs are spineless and let me walk all over them
23:00 paramat "kilbith > you're just one of the numerous people disgusted of contributing because of him" the problem here is in your heads, nowhere else
23:07 paramat "kilbith > yep, and your fan club really want to meet you to get things straight with you... we are not living far from each other, you know ;)" almost a personal physical threat
23:08 paramat "kilbith > think a moment why several people perceive that you're leading both projects, there's certainly a kernel of truth there" i'm active and you're delusional
23:10 paramat i've actually had long periods of obsessive personal attacks from kilbith before, it's nothing new
23:11 paramat BTW sun is doing :]
23:17 paramat celeron55 i guess personal threats are against IRC rules?
23:17 paramat ^^^
23:18 paramat half joking is not very comforting
23:26 sofar joined #minetest-dev
23:26 Hijiri Any comment on #5612
23:26 ShadowBot https://github.com/minetest/minetest/issues/5612 -- Allow overriding tool capabilities through itemstack metadata by raymoo
23:27 Hijiri I've got some upvotes and party poppers but no approvals or criticisms of the design
23:27 Hijiri except the code style review by nore
23:27 Hijiri nerzhul I mean
23:27 Hijiri sorry for the highlight
23:28 Calinou there, have one more upvote and party popper
23:28 Calinou you deserved it
23:29 Hijiri :P thanks
23:32 paramat twice before i've had periods of irrational personal attacks from zeno too, however, the rest of the time he is a pleasure to work with
23:34 paramat we do need more subgames, but large subgame teams are a problem as explained before, so it would be good for devs to get into groups to work on multiple subgames
23:34 paramat zeno perhaps start a group and work on a subgame?
23:38 paramat i sort of agree with you that it's weird how some core devs do not have a say in MTG, but having all core devs as MTG devs would probably be worse. multiple subgames created by groups or individuals is best, and takes the focus and pressure off MTG
23:40 paramat this has been asked for, for years, and there is no obstacle, people just have to get working
23:40 paramat and some are already
23:53 Calinou I don't think we need more subgames, quality over quantity is prime here
23:53 Calinou I mean, MTG is close to being a full game… all it needs is mods, and more server admin tools perhaps
23:53 Calinou (but the latter is mostly relevant for multiplayer)
23:53 Calinou mobs*
23:53 Fixer you can merge good mods into game with mod author continuing working on it
23:55 twoelk|2 joined #minetest-dev

| Channels | #minetest-dev index | Today | | Google Search | Plaintext