Time Nick Message 01:07 paramat game#1324 01:07 ShadowBot https://github.com/minetest/minetest_game/issues/1324 -- Default: Add coral nodes and death ABM by paramat 01:07 paramat ^ sofar 01:08 paramat will also make a PR for adding the generic 'bush_leaves' node soon 03:08 paramat will merge game#1320 in a moment 03:08 ShadowBot https://github.com/minetest/minetest_game/issues/1320 -- Bucket: Add optional 'force-renew' bool to registration by paramat 03:13 paramat merging 03:17 paramat merged 03:50 sofar wow someone just redid all the work I did on torches 03:50 sofar even though he knew about it 03:50 hmmmm don't you love those "refactors" 03:51 sofar 6 months of people testing it on servers, too 03:51 sofar I got good feedback 03:51 sofar plus I made the models better, I think 03:51 hmmmm i think the people on minetest game should be more picky with PRs 03:51 sofar I've been critical 03:51 sofar very 03:52 sofar it's wearing me down 03:52 hmmmm well, who merged the latest torch refactor? 03:53 sofar torches in mtg have never been refactored 03:53 sofar they're still vanilla pilzadam 03:53 hmmmm then whatchu talkin bout 03:54 sofar open PR 03:54 hmmmm ahh 03:54 hmmmm ah i see the one 03:54 hmmmm octacian 03:55 hmmmm who is this guy... he made a client side modding PR after i pasted the code i had already and said i've stopped work on it 03:55 hmmmm it looks remarkably similar (but then again, there aren't many *good* other ways of doing it tbf) 04:18 sofar oh well, I just made a competing PR 04:18 sofar may the merit win 04:18 sofar lol 06:10 celeron55 better be concerned whether the best design gets adopted, not whether it's someone's or not 06:11 celeron55 we can just credit everyone who tried :P 06:49 jin_xi i may be repeating myself, but the issue with competing PRs and choosing the best design could be greatly improved if there was a place for devs to actually talk about designs _before_ anyone goes writing code. This is obviously not the case for irc and neither github. 06:50 jin_xi and it might also help new developers cross the initial hurdle easier, seeing as good devs are hard to come by and 'quick' to leave minetest after their big PRs and fixes rot away 08:25 Hijiri I can't figure out where particles get scheduled for drawing 08:26 Hijiri can anybody tell me? 08:28 Hijiri I think I see, irrlicht calls OnRegisterSceneNode before rendering, or something 12:27 red-001 hwo do you run unit tests? 12:27 red-001 how* 12:31 Krock > minetest --run-unittests 12:31 Krock or. 6 failed 12:31 Krock *oh 12:35 red-001 UASSERT(s.getS32("leet") == 1337); 12:38 red-001 test_settings.cpp line 122 12:41 Krock [PASS] testAllSettings - 0ms 12:42 * red-001 has too many PRs open 13:08 dzho oh, is there a limit? 13:09 Megaf nope 13:10 Megaf I mean, there's always a limit 13:10 T4im there's a limit in motivation 13:10 Krock sky is limit too 13:20 T4im hm, that client side mods PR already started out on a pretty outdated ref; loads files, that don't even exist anymore 13:23 T4im (it's rebased by now, although wrongly, but the first time i saw it, he built it on some july ref iirc) 13:24 Krock and the whole thing is not secure at all 14:22 red-001 krock why is it not secure? 14:23 red-001 doesn't it use the mod security added a while back? 14:23 Krock the one which isn't enabled by default yet? 14:30 red-001 from what celron55 said it seemed that mod security is always enabled in the new api 14:31 T4im ShadowNinja opened a PR to do enable it by default; it's not enabled yet 14:31 * T4im might have actually delayed it a bit unintentionally; but is ok with that happening, generally, btw 14:31 red-001 the api as in the client sided scripting 14:32 T4im ah, ok 14:32 red-001 nvm est31 said that not celron55 15:41 red-001 updated #4472 15:41 ShadowBot https://github.com/minetest/minetest/issues/4472 -- Make serverlist searchable. by red-001 15:50 sfan5 wonderful 15:50 sfan5 nrz's unordered map changes made the profile graphs flicker 15:56 sfan5 http://i.imgur.com/8lisQZa.gif 15:58 Krock eeks. I thought the profiler wasn't affected of these changes 16:40 sfan5 nrzkt: mind approving this? http://sprunge.us/OQJO (otherwise this happens http://i.imgur.com/8lisQZa.gif) 16:41 nrzkt sfan5, seems lgtm to bugfix 16:41 sfan5 thanks, will push soon 17:28 sofar sfan5: hah, yeah, bad use of unordered :D 17:31 Krock unordered is bad! 17:36 red-001 why do people not like smart pointers 17:48 sfan5 because c++11 17:50 nrzkt red-001, the problem is smart pointers should only be used when necessary not everywhere 17:50 nrzkt because with smartpointers memory leaks are as easy as forgetting delete: ) 17:51 sfan5 sofar: wouldn't >is_soil == 0 be false with is_soil = false? 17:53 sofar sfan5: get_item_group returns '0' if the soil group is not set 17:53 sfan5 ok thn 17:53 sfan5 *then 17:54 sofar https://github.com/minetest/minetest/blob/3a95054db5c44a3fea6a27d5633f668539e67870/builtin/game/misc.lua#L151 17:54 sofar tl:dr: `return 0` 18:06 Hijiri nrzkt: finally figured out where I was getting build fails from, the "at" method is C++11 apparently 18:21 sfan5 Hijiri: shouldn't travis help you figure that out? 18:21 Hijiri it did 18:21 Hijiri but I didn't undrestand the gcc error message 18:21 Hijiri I had to look at the clang error 18:28 paramat please could devs add their comments about the controversial part of #4620 ? the running of nodeupdate 'on punch' 18:28 ShadowBot https://github.com/minetest/minetest/issues/4620 -- Optimize falling.lua by tenplus1 18:39 paramat ^ nore nrzkt sfan5 Zeno` 18:40 sfan5 IMO nodes don't need to start falling if you punch then 18:40 sfan5 *them 18:42 T4im can't you discuss that in an extra issue? whether or not punching should be removed from them should not prevent fixing of actual issues 18:44 paramat indeed a seperate commit would be good, but tenplus1 is not good with git 18:45 T4im krock recently pointed out, that that new "allow maintainer to edit" checkbox might allow you guys to actually push to those git repos of PR's 18:45 paramat anyway we can alter on merge if necessary 20:03 sfan5 paramat: i really like your floatlands, goob job 20:04 paramat good, a little more tuning to do 20:04 sfan5 dungeons look especially nice in there 20:04 sfan5 do we have an issue on the tracker for putting stuff into dungeons yet? 20:05 sfan5 right now they're just decoration 20:05 paramat some have asked, i think there is an issue 20:08 sfan5 also what about ores in floatlands? 20:09 paramat ore distribution has been changed, above 1024 you get everything with the same densities as below -1024 20:10 paramat so ore-rich floatlands 20:11 sfan5 hm i haven't seen a single ore yes 20:11 sfan5 *yet 20:11 sfan5 i just found a pond of water though, wonderful 20:12 paramat weird, should be lots of ores 20:12 paramat i'll check that 20:17 sfan5 paramat: lots of gravel(?) blobs but no ores: http://i.imgur.com/R6P3hU7.png 20:18 paramat there are ores in the thread screenshot, anyway i'll check when tuning 20:21 paramat latest mtgame? 20:22 sfan5 oops, that must be it 20:23 paramat checf for ore registrations for y = 1024+ 20:34 paramat (check) 20:35 sfan5 nah the p 20:35 sfan5 the outdated mtgame is definitely the issue 20:35 sfan5 it's nice that trees grow even here http://i.imgur.com/aBqMqqR.png 20:48 paramat the standard biomes will be messy though, underwater trees and grass etc. enabling floatlands will enable a tweaked set of biomes that fixes those issues 20:48 sfan5 small overview map of some floatlands: http://i.imgur.com/a7D7MUl.png 20:48 sfan5 looks pretty nice 20:49 paramat there will be larger lakes in places 20:51 * twoelk dreams of nice waterfalls 20:52 sfan5 won't that result in horrible lag? 20:52 paramat water flows through tunnels under the lakes to the floatland base and falls 20:52 paramat nah 20:53 paramat oh well maybe 20:53 paramat if they spread too much over surface 20:54 paramat if you swim in the falling column to keep it loaded you can eventually create a water column down to ground level (takes time though) 20:59 Hijiri what do people think of this solution to transparent particle depth sorting? http://answers.unity3d.com/questions/20984/depth-sorting-of-billboard-particles-how-can-i-do.html 21:00 Hijiri (bounded number of passes of bubblesort) 21:01 paramat we have a problem with particles disappearing behing translucency 21:01 Hijiri or do you think sorting normally every frame would be performant enough 21:02 paramat (behind) 21:02 Hijiri paramat: I'm talking specifically about particles disappearing behind other translucent particles, not nodes or other things 21:02 paramat yeah 21:05 twoelk If above a certain hight wouldn't it be better for falling water to turn to vapour after say 100m or so? 21:06 Hijiri if all particles were inside one big scene node then you could just sort them in there, not sure how it works when each particle is a scene node 21:06 paramat twoelk it usually does that anyway when it hits an unloaded/ungenerated area 21:06 sfan5 wow 21:06 sfan5 you can find diamond at the surface in the floatlands 21:07 paramat yeah identical ores to below y = -1024 21:08 paramat compensates for lack of stone mass 21:10 twoelk cool, I got an old map imported from worldpainter that had too many liquid spills to be playable in minetest. Maybe I should work on that project again. 21:11 paramat also now, falling liquids don't spread outwards on ignore 21:11 paramat the water column just stops 21:11 paramat which can be considered it dissolving into spray 21:12 twoelk I was trying to figure out how to make a special lava that would vapourize water that falls near it - so that I could have a persistant lavalake as base no mater how much water flows onto it 21:13 Hijiri ABM on water with new lava neighbors? 21:14 paramat override the lavacooling ABM 21:14 sfan5 paramat: #4628 21:14 ShadowBot https://github.com/minetest/minetest/issues/4628 -- Use range-based fog instead of z-plane based. by lhofhansl 21:16 paramat looking 21:43 paramat any reviews for #4620 ? 21:43 ShadowBot https://github.com/minetest/minetest/issues/4620 -- Optimize falling.lua by tenplus1 21:48 sfan5 paramat: lgtm 21:58 paramat ok 21:59 paramat will merge it later 22:03 paramat well, maybe it ought to have more input 22:04 Hijiri nrzkt: I moved the erase call into the conditional body 22:04 Hijiri and the build finished 22:09 nrzkt Hijiri, cool thanks 22:53 paramat will merge game#1324 in a moment 22:53 ShadowBot https://github.com/minetest/minetest_game/issues/1324 -- Default: Add coral nodes and death ABM by paramat 22:54 sfan5 wait wait 22:55 sfan5 where's the second approval? 22:55 sfan5 oh from sofar 22:55 sfan5 but IMO we don't need corals 22:56 sofar sfan5: well, um, that can be said about almost any node 22:56 sofar but yes it's cosmetic 22:56 sofar but no, the sea floors are barren, uninterestign areas. 22:56 sfan5 well we can make it useful later 22:56 sfan5 fine with me 22:57 sofar that's the spirit ;) 22:57 paramat ok merging 22:57 sofar paramat: can you close my PR as well then? 22:57 paramat calcium carbonate has many uses 22:57 paramat yes ok 22:59 sofar grind it up for lime fertilizer? 22:59 sofar makes nice kitchen countertops, too 22:59 paramat i have a copy of your coral schem and will be using that, with orange corals added 23:02 paramat merged 23:03 sofar I'll be looking for them :) 23:03 paramat mapgen stuff isn't done though 23:03 paramat yet 23:03 sfan5 hm 23:03 sfan5 paramat: would corals also appear in floatlands? :D 23:04 paramat yeah if wanted 23:04 paramat my intention was to have all resources available up there 23:05 sfan5 some just won't make sense maybe 23:06 paramat eventually the floatlands flag will enable a tweaked set of biomes and decorations that work well with the floatlands 23:06 paramat yeah perhaps 23:06 paramat also of course we could have special biomes up there 23:09 sfan5 that tweaked set will be set by the lua code in mtgame, right? 23:22 paramat yes in mapgen.lua 23:24 sfan5 23:25 paramat i might merge floatlands in a week or so, after some more tuning 23:32 Fixer does coral need slabs/stairs? :trollface: 23:33 sfan5 apropos stairs 23:33 sfan5 paramat: is it possible that dungeons *no longer* have proper stairs? 23:33 sfan5 because they did in 0.4.14 23:34 paramat some dungeons do not have half-node stairs, erm for some reason i temporarily forget 23:35 paramat oh any dongeon with diagonal corridors cannot (desert dungeons) 23:35 paramat (dungeon heh) 23:35 paramat dong eon 23:36 paramat dungeons with 2 node wide corridors cannot because that causes bugs (sandstone dungeons) 23:37 paramat maybe i can fix that later 23:37 sfan5 i observed it in normal dungeons too 23:37 paramat yeah in stone dungeons stairs sometimes become full nodes, don't know why 23:38 paramat it's not new behaviour afaik 23:38 paramat or at least, that code has not been changed 23:49 sofar only stairs are needed for the dead coral node