Minetest logo

IRC log for #minetest-dev, 2015-04-19

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

All times shown according to UTC.

Time Nick Message
00:09 paramat joined #minetest-dev
00:09 hmmmm it really doesn't matter
00:21 est31 fine, now there is only a memory corruption left
00:21 est31 and perhaps dozens of other bugs
00:21 est31 who knows?
00:21 hmmmm heh
00:22 hmmmm requesting testers for linux and windows:  https://github.com/kwolekr/minetest/commit​/5cb19782fee106f8466672f23233cf5083e72e6a
00:22 hmmmm I 100% guarantee I broke shit
00:36 Wayward_Tab joined #minetest-dev
00:44 paramat okay as it seems i have enough approval i will push river water soon
01:04 Player_2 joined #minetest-dev
01:25 paramat now pushing to game
01:29 paramat push complete
01:30 VanessaE damn it paramat
01:30 VanessaE stop improving minetest_game :)
01:30 VanessaE (you're soon gonna force me to resync dreambuilder :P _
01:31 paramat ;P
01:37 Fritigern joined #minetest-dev
01:42 paramat next i will add river water to the biome api and biome definitions so it can be replaced by chosen nodes
01:51 VanessaE btw, why use inventorycube() for the water nodes?
01:54 Lunatrius joined #minetest-dev
01:58 paramat erm, dunno i just copied default:water
01:59 paramat i don't like how flowers reproduce into ever-growing constant densities, i think they should drop 1-2 seeds and be manually re-planted like other crops
02:00 VanessaE I think I suggested something like that before too
02:00 VanessaE but don't they at least limit to perlin?
02:01 paramat um..
02:03 paramat no, they're generated to varying density using perlin
02:03 VanessaE huh.
02:03 paramat but will reproduce to a constant density of 3 flowers per 9x9 node area
02:03 VanessaE ew.
02:04 paramat i plan to add moon-version flowers to my moonrealm mod, they need to be farmable and have seeds
02:05 VanessaE eh, be careful there
02:05 VanessaE there was a "moon flower" debate a while back
02:05 VanessaE (little blue, dimly-glowing flowers as I recall, could be used to make glow dust to be used for light-source recipes)
02:06 VanessaE that debate didn't go well
02:06 paramat yeah i don't mean that =)
02:06 VanessaE oh hell I misread that
02:06 VanessaE heh
02:08 paramat the flower spread abm is very slow so and on a moon you can't just go searching for more
02:08 paramat (remove 'so')
02:08 VanessaE right
02:12 paramat moon cotton too, all this is for dye and coloured wool for carpets and curtains
02:21 paramat hmmmm will you be working on mg_biome or l_mapgen the next couple days? just checking to avoid conflicts
02:39 OldCoder joined #minetest-dev
02:41 paramat #2635 will test soon and push later if acceptable
02:41 ShadowBot https://github.com/minetest/minetest/issues/2635 -- Biome API: Add biome-specific river water by paramat
02:48 paramat next i will add the mapgen alias for river water to mtgame
02:49 Player_2 joined #minetest-dev
02:55 paramat game#495
02:55 ShadowBot https://github.com/minetes​t/minetest_game/issues/495 -- Default/mapgen: Add alias for river water source by paramat
02:59 paramat now pushing to mtgame
03:02 paramat done
03:06 paramat back later
03:06 paramat left #minetest-dev
03:13 est31 if anyone has time please test my SRP PR
03:13 est31 only thing missing now is changing passwors
03:13 est31 passwords*
03:21 est31 About translation PRs, what to do?
03:22 est31 There are currently 2 open PRs for chinese for example
03:22 est31 #1980 and #2634
03:22 ShadowBot https://github.com/minetest/minetest/issues/1980 -- Update Simplified Chinese translation by yyt16384
03:22 ShadowBot https://github.com/minetest/minetest/issues/2634 -- Update chinese translation by fishyWET
03:22 est31 we don't have any way to check them, so rely on the community? Let them rot?
03:23 est31 use google translate and when it doesnt output shit then we merge?
03:26 est31 I mean we just need a standardized way
03:26 est31 best would be some translation cloud or so
03:36 Hijiri joined #minetest-dev
03:53 hmmmm paramat:  yes I think so
03:54 hmmmm hrmm adding a river water node
03:55 hmmmm paramat, is there any other way to do what you need to?
03:55 hmmmm in minecraft, river is a biome as ocean is a biome
03:57 hmmmm ideally when terrain-base biomes get introduced, we'd be able to distinguish this and just set c_water
03:57 Robby joined #minetest-dev
03:58 hmmmm also, minecraft stores biome right in their equivalent of a mapblock
03:59 hmmmm so the mapgen can directly override biome selection
03:59 hmmmm maybe we should start thinking about enhancing how biomes work rather than just building upon a system that's rather weak to begin with
04:07 est31 joined #minetest-dev
04:23 paramat joined #minetest-dev
04:23 * paramat reads
04:24 paramat what you write seems reasonable. i don't deparately need to add river water to the biome api, as we have whats needed to start work on river mapgens, river water could for now be left as default:river_water
04:25 paramat but it would be nice to have until a new system arrives
04:25 paramat for example in tundra rivers become ice glaciers
04:26 paramat it would just make mapgen development prettier as we could use a full set of biomes
04:27 leat joined #minetest-dev
04:30 paramat also, while new mapgens are in development, players can use/test them, for now, with a fully functional biome api. so yeah i would still like to push this
04:31 hmmmm hmmm
04:31 hmmmm see what we have right now is a biome system based on climate only
04:32 hmmmm what i wanted was a biome system based on climate and terrain
04:32 paramat i remember your idea
04:32 hmmmm what you're doing (and you might not even realize it) is adding support for all types of terrain within a single climate biome
04:32 paramat yes
04:33 hmmmm so which direction makes more sense is what I'm wondering
04:34 est31 bye
04:34 hmmmm bye
04:35 hmmmm what do you think, paramat?
04:36 hmmmm should we have the mapgens determine which terrain features are which?
04:36 paramat i have no objection to a new system, and would like to see you experiment with your idea
04:36 hmmmm that would require either saving the output, or effectively running the mapgens each time we want to calculate the biome
04:37 hmmmm honestly the mapgen knows best when it comes to terrain
04:37 hmmmm my algorithmic approaches don't work quite well
04:39 paramat i don't have a strong opinion this, but it's best to try it and see
04:40 hmmmm it's not that easy.  effectively, this requires changing the mapblock serialization format
04:40 hmmmm or at least saving (X, Z) along with each block if it doesn't yet exist
04:41 hmmmm http://minecraft.gamepedia.com/Anvil_file_format  <---- see here, this is the approach that makes sense and is used by minecraft exactly the same way
04:41 paramat until a new system aarives, i feel like using the old system to the full. not having glaciers in tundra while working on a new mapgen will, oddly enough, bug me
04:42 * paramat looks
04:42 VanessaE well a new map format would make your colored lights easier...
04:43 hmmmm a new mapBLOCK format
04:43 hmmmm and colored lights are on the backburner
04:43 VanessaE so I assumed.
04:43 VanessaE minetest has other stuff that must be dealt with first.
04:43 hmmmm christ i have biomes, client-side lua, enhanced emergequeue, all this other stuff
04:44 hmmmm at least there are some people here working on some big things that need work :)
04:44 hmmmm like est31 and SRP
04:44 hmmmm thank you est31.
04:44 hmmmm everybody else sorta pecks at the small stuff or refactoring
04:45 hmmmm also jin xi with the irrlicht native particle spawners
04:46 paramat i'm happy with the current biome system, it's not amazing but okay, don't feel pressure to redesign it all (more work)
04:46 hmmmm well I'll approve of your addition of river water
04:46 paramat okay cool
04:46 hmmmm but this is not going to be the final biome system..  there's not going to be a single biome system
04:47 hmmmm I want there to be multiple versions to encourage experimentation
04:47 paramat multiple systems are my preference
04:47 hmmmm I'm also planning on adding a non-perlin-noise version
04:47 hmmmm blotches like minecraft has
04:47 nore joined #minetest-dev
04:47 hmmmm it really does produce boring shapes, but they're at least more consistently big
04:47 paramat "lovely crinkly edges"
04:48 hmmmm with this system you have a tiny thin strip of snow biome sandwiched in between two desert biomes, or something equally whackery
04:48 hmmmm lovely, init
04:48 paramat personally i don't like the MC jigsaw puzzle
04:48 paramat too regular
04:49 paramat but have nothing against the option
04:49 hmmmm I think I have to draw out on paper what shapes I'd like the biomes to be first
04:49 VanessaE only way you're gonna solve that, hmmmm, is by taking elevation *and* Z location into account
04:49 hmmmm and then come up with some way of making it happen, not the other way around
04:49 VanessaE and that'll piss players off, I'm sure.
04:49 hmmmm ?? what do you mean
04:50 hmmmm you mean Y location?
04:50 VanessaE I mean Z.  as in the closer to center you get, the more lush the landscape.  deserts as you move further out, tundra and ice in the north and south
04:50 VanessaE etc.
04:51 hmmmm that kind of sounds like the addition of the climate noise to the current biome system
04:52 VanessaE I suppose it's similar, bbut like I said it'll piss players off because they'll have huge travel distances to find different biomes
04:52 VanessaE -b
04:52 hmmmm lol I like how MS made GetCurrentDirectory have the same exact parameters and return values as getcwd(), except the switched the two parameters around
04:52 hmmmm jerks
04:53 VanessaE and then of course elevation affects temperature and humidity (or should, idk if it does now)
04:53 paramat large scale climate variation by noise is okay by me, but not linear or determined by centre-edge, or N-S or E-W
04:53 hmmmm agreed
04:53 hmmmm that's why I had it as perlin noise
04:54 paramat i stopped having Y affect temp/humidity in my mapgens because it makes all floatlands tundra, no biome diversity
04:55 hmmmm height should affect temperature, but not by that magnitude
04:55 hmmmm it should be much more subtle
04:55 VanessaE well elevation-dependent, but only up to a point
04:56 hmmmm hmm
04:56 hmmmm does minecraft never have cliffs hanging over rivers?
04:56 VanessaE plus there's the issue of multiple actual realms (i.e. the 1024-node divisions)
04:56 paramat well i always insist on filliing the entire world up to 31000, so even a gentle variation ends up huge at the top
04:57 paramat (filling with rarish floatlands)
04:57 hmmmm you can make an exception for extreme elevations
04:57 hmmmm like making floatland height relative to some baseline
04:57 VanessaE (one doesn't want elevation to turn a realm at +1024 into all snow or whatever)
04:58 paramat we make use of 0 to -31000 but not ~256 to 31000
04:58 paramat so in my lua mapgens i use 3d noise for temp and humidity
05:01 paramat anyway will push my PR soon..
05:07 paramat now pushing 2635
05:12 paramat complete
05:40 hmmmm alrighty
05:40 hmmmm #2637 passed all its checks
05:41 ShadowBot https://github.com/minetest/minetest/issues/2637 -- Porting: Refactor initalizePaths() by kwolekr
05:41 hmmmm anybody?  please review
05:59 paramat i looked but it's beyond me sorry
05:59 hmmmm :/
06:13 paramat http://i.imgur.com/qAtngOY.png more fractal mapgen
06:13 paramat left #minetest-dev
06:16 hmmmm that's the way i thought proller's fractal mapgen was going to look like
06:16 hmmmm this is so much better
06:17 hmmmm and playable
06:41 leat joined #minetest-dev
07:02 Krock joined #minetest-dev
07:10 Hunterz joined #minetest-dev
07:42 err404 joined #minetest-dev
08:01 Yepoleb_ joined #minetest-dev
08:28 kilbith joined #minetest-dev
08:53 ElectronLibre joined #minetest-dev
09:09 jin_xi joined #minetest-dev
09:19 Megaf_ joined #minetest-dev
09:51 Calinou joined #minetest-dev
10:22 Calinou website editors, can you add #minetest-fr to the Other channels section?
10:25 Ritchie joined #minetest-dev
10:41 leat joined #minetest-dev
10:45 kilbith and #minetest-es
10:46 kilbith nvm
10:53 blaze joined #minetest-dev
11:34 leat joined #minetest-dev
11:43 MinetestForFun joined #minetest-dev
11:44 disableclouds joined #minetest-dev
11:53 leat joined #minetest-dev
12:05 celeron55 Calinou: added
12:08 Calinou thanks
12:13 loggingbot_ joined #minetest-dev
12:13 Topic for #minetest-dev is now Minetest core development and maintenance. Chit-chat goes to #minetest. Consider this instead of /msg celeron55. http://irc.minetest.ru/minetest-dev/ http://dev.minetest.net/
12:14 SudoAptGetPlay joined #minetest-dev
12:25 leat joined #minetest-dev
13:02 crazyR_ joined #minetest-dev
13:02 crazyR_ joined #minetest-dev
14:08 ElectronLibre joined #minetest-dev
14:25 rubenwardy joined #minetest-dev
14:30 leat joined #minetest-dev
15:04 cib0 joined #minetest-dev
15:29 MinetestForFun joined #minetest-dev
16:37 SopaXorzTaker joined #minetest-dev
16:45 ElectronLibre joined #minetest-dev
16:47 hmmmm joined #minetest-dev
16:50 johnnyjoy joined #minetest-dev
16:51 johnnyjoy joined #minetest-dev
17:07 cib joined #minetest-dev
17:19 Wayward_Tab joined #minetest-dev
17:43 leat joined #minetest-dev
17:47 est31 joined #minetest-dev
17:49 Guest37259 joined #minetest-dev
17:52 est31 https://github.com/est31/minetest/commit/​b0fd3aabb04a5c464b51e22803f47ada70368512
17:52 est31 ^ merging in 15 minutes
17:53 est31 see comments on this commit too: https://github.com/minetest/minetest/commi​t/5f0b36b8b70faa36b20343342567471c8f50b7f2
17:53 est31 I've used the same comparison that is used to decide whether to start the server
17:54 Krock est31, "// Only time out if we aren't if we to wait for the server we started" re-phrase pls
17:55 Krock Also, I think that's useless because it's only about a ping answer
17:56 est31 oh
17:56 est31 rephrasing
17:57 est31 I think *address != "" needs explanation
17:59 est31 https://github.com/est31/minetest/commit/​f1a41e4f77ce272e8e454710a7a168d95466742c
17:59 Krock This comit is still not required IMO. Were you able to reproduce this problem?
18:00 est31 So you think it can't be reproduced?
18:01 est31 What they said sounded very likely
18:05 est31 yes I am abled to reproduce
18:05 est31 its a very mean one
18:05 est31 first you have to wait
18:05 est31 then the message comes: timed out.
18:05 est31 so you dont wait 10 secs
18:05 est31 but that time you actually would wait too
18:06 est31 due to single threadedness
18:07 est31 and tested, the commit works
18:07 Krock Well then. Continue.
18:08 est31 pushed
18:09 est31 still thanks for the commit
18:09 Krock s/-it/ent/
18:10 est31 ?
18:12 est31 bye now
18:18 ElectronLibre joined #minetest-dev
19:08 hmmmm nobody reviews my commits
19:08 cib0 joined #minetest-dev
19:10 VanessaE hmmmm: because you usually code at a level above everyone else :P
19:10 hmmmm http://i926.photobucket.com/albu​ms/ad104/tuffguy1979/0gNvwh.jpg
19:10 VanessaE heh
19:11 hmmmm i didn't do anything too radical though.. just made things easier to read and simplified some logic
19:11 hmmmm initializePaths() is one of those things which people only tack onto because they're too scared of breaking other platforms
19:12 hmmmm neoasthetic or whatever broke RUN_IN_PLACE for OS X.  as in, he made it read from the cfbundle directory and not relative to the currently running executable
19:12 hmmmm ={ not cool
19:12 hmmmm run_in_place needs to actually mean run_in_place
19:13 VanessaE not good.
19:13 VanessaE of course OS-X is kind of a special case isn't it?  seems like "run-in-place" should pretty much never be 0 in that platform..
19:13 hmmmm i'm not sure
19:13 hmmmm why should run_in_place be different from other unixes?
19:13 VanessaE I always got the impression that on Mac, all programs were basically self-contained in their own folders.
19:14 hmmmm maybe older ones
19:14 VanessaE i.e. none of the usual /usr/* stuff.
19:14 VanessaE oh well maybe
19:14 hmmmm there's a "CFBundle resource location"
19:14 VanessaE my only Mac experience is limited to older ones
19:14 hmmmm executables are stored in /Applications
19:14 hmmmm i think.
19:14 hmmmm I can load up my testing OSX vm
19:15 VanessaE on my old one (runs OS-X 10.4 I think), I seem to recall everything being stored in an Applications folder (wherever that actually is, I never took note), but that each package usually had its own folder therein
19:16 hmmmm osx is such a pain in the ass
19:16 hmmmm everything is like.. different for the sake of being different and it's not necessarily better
19:16 VanessaE it's Apple.
19:16 VanessaE 'nuff said? :)
19:17 hmmmm it's not loading up for whatever reason
19:17 hmmmm uh oh
19:18 VanessaE ?
19:18 hmmmm my VM is stuck on the gray screen with apple logo and the swirly loading animation
19:39 MinetestForFun joined #minetest-dev
19:41 ElectronLibre joined #minetest-dev
19:58 luizrpgluiz joined #minetest-dev
20:13 luizrpgluiz left #minetest-dev
20:15 paramat joined #minetest-dev
20:16 paramat https://github.com/minetest/minetest​_game/pull/493#issuecomment-94309290
20:18 paramat ^ need support for my river water
20:25 leat joined #minetest-dev
20:33 ElectronLibre left #minetest-dev
21:19 paramat left #minetest-dev
22:02 proller joined #minetest-dev
22:04 hmmmm est31:  I just saw that "disable timeout for singleplayer" commit - what's up with that?
22:04 hmmmm why would a lot of mods slow the connection down to that point?  localhost connections should be instantaneous
22:08 Wayward_Tab joined #minetest-dev
22:15 Wayward_Tab joined #minetest-dev
22:23 Wayward_Tab joined #minetest-dev
22:30 Wayward_Tab joined #minetest-dev
22:38 Wayward_Tab joined #minetest-dev
22:45 AndroidKris joined #minetest-dev
22:46 Wayward_Tab joined #minetest-dev
22:54 Wayward_Tab joined #minetest-dev
23:01 Wayward__Tab joined #minetest-dev
23:09 Wayward_Tab joined #minetest-dev
23:17 Wayward_Tab joined #minetest-dev
23:22 hmmmm crap, just found two bugs related to noise
23:25 Wayward__Tab joined #minetest-dev

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