Time Nick Message 00:01 paramat i think i will push these new heat, humidity and mgv7 noise params later. feel free to stop me if this will cause conflicts for you 00:06 hmmmm no go ahead 00:07 hmmmm you must realize that's only a stopgap measure, you can't possibly expect people to adjust noise each time they add a biome 00:11 paramat agreed 00:13 paramat in future perhaps sqrt(number of biomes) will scale spreads as we discussed.. 00:29 hmmmm now, why sqrt() exactly? 00:43 paramat i assume a terrain area of spread by spread where heat and humidity vary across their ranges along the axes, so all biomes are present. keeping biome size constant, spreads are proportional to sqrt(number of biomes) 00:44 paramat hard to describe, but intuitively makes sense to me 00:46 paramat comes from visualizing heat/humidity voronoi diagrams 00:46 hmmmm yeah 00:47 hmmmm I'm going to make a command line option to spit out the voronoi diagrams in image format 00:47 hmmmm given a set of biomes 00:47 paramat nice 00:47 hmmmm but this would depend on the client 00:47 hmmmm because i need irrlicht as a dependency 00:48 paramat with the sqrt(biomes) thing we will need a settable parameter to scale biome size 00:48 paramat a rough biome width 00:50 paramat spread = biomewidth * sqrt(nobiomes) 00:51 paramat both spreads equal 00:52 paramat but because noise biome size will turn out smaller, so param is rough 'max biome size' 00:53 hmmmm this might be easier said than done 00:53 hmmmm increasing the spread decreases the detail so we also need to increase the number of octaves 00:53 hmmmm we can't indefinitely increase the number of octaves, in fact that's quite limited 00:54 hmmmm even 15 octaves would probably exceed the memory limitations on most machines 01:03 paramat ah octaves, with sqrt() spread will only grow slowly, then octaves coould be increased even slower than that 01:04 paramat (could) 01:05 paramat so limited to range 3 to 6 of whatever 01:09 paramat only need 3 more octaves to keep the same detail when going from 4 to 256 biomes, no problem 01:10 paramat will test and push this later #2718 01:10 ShadowBot https://github.com/minetest/minetest/issues/2718 -- Biome API, mgv7: Increase heat/humidity spreads. Improve mgv7 noise parameters by paramat 03:14 paramat pushing #2718 very soon 03:14 ShadowBot https://github.com/minetest/minetest/issues/2718 -- Biome API, mgv7: Increase heat/humidity spreads. Improve mgv7 noise parameters by paramat 03:31 paramat now pushing 03:36 paramat complete 05:02 est31 hey Zeno` ! 05:07 Zeno` hey 05:33 hmmmm hey zeno`, what's been up lately? 06:34 Zeno` not much. Had the flu so I've been in bed a lot :/ 14:36 est31 pushing in 5 mins https://github.com/est31/minetest/commit/6b7fb591cb8d54bc8a43df6adb34a026720b4fe4 14:37 rubenwardy Did you make the ASCII art originally by hand? O_o 14:37 est31 I didnt make it 14:38 est31 only added large parts of it 14:38 est31 but yes I edited them by hand :) 14:38 rubenwardy Ah XD. 14:38 est31 two things are useful though when doing that: 1. kate's block mode, 2. that key over DEL 14:48 Krock write it on the worldoftext site and then select - copypaste :P 14:49 celeron55 i think there are missing arrow heads in the graph 17:21 est31 celeron55, I haven't added arrowheads to arrows of length 1, because there the arrow would only consists of its head. Except I used ↓ which is unicode 17:21 est31 and unicode in c files... 17:23 est31 (that has been previously there too) 17:34 Calinou I've used Unicode in Carbone comments :P 17:35 Calinou why is that a problem? any properly set up editor uses UTF-8 17:36 init unicode and utf8 are different things 17:36 est31 One of the features java bragged about was unicode support even inside source files 17:36 est31 I havent run into issues with unicode inside c files yet 17:36 est31 but still I'm cautious 17:37 est31 you know in two weeks people are showing up "hey this comment only commit from may 24 broke my compiler on openVMS!" 17:37 est31 build* 17:38 est31 as a fun fact, once I've used umlauts inside a project name in java, because, I've thought, everything will work well 17:38 est31 so what happened 17:38 est31 total chaos 17:38 est31 it worked some times 17:39 est31 but there were various bugs 17:39 est31 so I regretted it not using ae or oe instead of ä and ö 17:39 est31 there you have the unicodeness of java :) 17:40 Calinou “Now supports äaö!” Minecraft splash :P 20:30 paramat i will go ahead and make mgv6 snowbiomes default soon 20:31 VanessaE guess that means snowdrift mod will need some changes? ;) 20:44 paramat yes it will need an option 20:48 VanessaE one question though is how do these new snow biomes mesh with SPlizard's snow mod? 20:48 VanessaE (and plants_lib, which uses the same temperature map) 20:49 paramat nore sfan5 perhaps sometime you could add comments to game#520 20:49 ShadowBot https://github.com/minetest/minetest_game/issues/520 -- Default: New sandstone brick texture by paramat 20:50 paramat they don't match at all, sorry, an option will be needed 20:51 VanessaE you should match them by default then 20:51 paramat because the snow mod's locating of snow biomes makes no sense, and mine does =) 20:51 paramat .. biome noise should be taken as temp 20:51 VanessaE that's gonna break maps that use either of the above mods 20:52 VanessaE even if you added a way for a mod to detect if engine-supported snow biomes exist, no mod will ever be able to accurately guess where a snow biome is 20:52 paramat don't think so because old worlds will not have snowbiomes enabled by flags in map-meta.txt 20:53 VanessaE because there will be some parts of the map that are using the older method e.g. snow, plants_lib mods, and some parts using the newer method 20:53 VanessaE that will help 20:54 paramat so essentially only new worlds will get snowbiomes 20:54 VanessaE and if a server owner wants to switch to the newer method? 20:55 paramat they could add the flag into map-meta, and get a few straight-edged biomes between old and new 20:55 VanessaE right, I'd expect that 20:55 VanessaE but the point is it conflicts with how the older mods do it 20:55 VanessaE which means you'll have desert-temperatures in a snow biome 20:55 VanessaE etc. 20:56 paramat yeah mods will need an option added for the new biome system 20:56 VanessaE isn't there some way to identify that a particular mapblock is being generated with the new method? 20:57 paramat erm flags may be checkable.. 20:57 VanessaE I mean something passed along to on_generated 20:58 paramat the issue is here, comments welcome game#519 20:58 ShadowBot https://github.com/minetest/minetest_game/issues/519 -- Should mgv6 snowbiomes be enabled by default? 20:58 paramat erm possibly 20:59 paramat 'get mapgen params' at the start of on-gen? 20:59 paramat not too sure 21:00 VanessaE that might work as long as it's painfully obvious :P that a returned flag basically means "new snow biomes in effect here" or so 21:01 VanessaE and indeed, snowdrift will be a problem 21:01 VanessaE because it can't just check those flags 21:02 VanessaE (assuming the user changes the flags to turn the new biomes on) 21:03 paramat if a server is switched over, all new chunks will have the new biome system, so mods could just be changed to act in the new way from that moment on 21:03 VanessaE but what about old chunks where there's active stuff going on? 21:04 paramat aha 21:04 VanessaE as in snowdrift mod adding snow to existing chunks 21:04 paramat indeed no solution to that 21:04 VanessaE yeah. 21:04 VanessaE the only thing one could do is remove the mod entirely :( 21:05 VanessaE unless, as I said, you can add a flag to the generated chunks 21:05 paramat normally i wouldn't expect any server or world to be switched because of the many straight-edged biomes caused 21:05 VanessaE i.e. something like minetest.get_chunk_flags(mapblock_pos) 21:05 paramat so any switch would cause a mess anyway.. 21:05 VanessaE hah! there have already been screwups like that in the past 21:06 sfan5 paramat: added a comment to game 520 21:06 VanessaE my creative map is the victim of this 21:06 paramat thanks 21:07 VanessaE added my comment as well, just for the record. 21:08 VanessaE paramat: in any case, what I'm really getting at is, adding official snow biomes to the engine like this is a good thing, don't get me wrong 21:08 VanessaE but you want to provide an upgrade path for users who have been using mods to achieve this effect. 21:08 VanessaE (since mods == slow) 21:13 paramat i'll post info in the forum to help modders add an option for the new system 21:14 VanessaE that isn't an "upgrade path" :) 21:14 paramat the headache for modders is unavoidable either way though, as most i expect will turn snowbiomes on even if it isn't default 21:14 paramat hehe 21:14 VanessaE I don't have anything against turning them on by default 21:14 VanessaE since that's only for new maps anyway 21:15 VanessaE nor do I have a problem with adapting e.g. plants_lib to account for it 21:15 VanessaE it's upgrading an old map that poses the quandry. 21:15 paramat yes 21:17 paramat but then it's not advisable to switch an old world over due to the discontinuous biomes, it doesn't justify a complex upgrade path, i wouldn't want to work on that 21:19 paramat modders will just have to do the work themselves if they are crazy and switch =) 21:21 VanessaE on a slightly-related note: there's one thing I've been wanting to do for several months now... reset the maps on a couple of my servers, but keeping the user-built content i.e. laying it all down onto newly-generated terrain, where the terrain matches the seed that the map was supposed to have, to get rid of discontinuous areas where interpretation of mapgen params has changed or gone awry 21:21 VanessaE (my creative map has such things) 21:22 paramat oh man.. 21:22 paramat difficult 21:23 VanessaE that's a hard subject for sure. :) 21:58 celeron55 hmm i wonder if a screenshot taken from minetest+minetest_game requires the attribution+sharealike stuff to be followed 21:59 celeron55 actually i wonder because someone basically asked about that 21:59 VanessaE celeron55: it's generally been held that copyright doesn't apply if the screenshot is low-resolution enough 21:59 celeron55 the screenshot is high-resolution 21:59 VanessaE falls under "fair use" or "fair dealing" or whatever you wanna call it 21:59 celeron55 this same email appears to have been sent to many core developers 22:00 celeron55 they want to use it as a silly background on the cover of a scientific publication 22:01 VanessaE wat 22:01 celeron55 it's ridiculous in all the right ways 22:02 VanessaE off the top of my head, I would say that CC-by-SA applies here 22:05 celeron55 i wonder what that means in practice 22:05 VanessaE an attribute line under the image or in the acknowledgement section maybe 22:05 VanessaE I don't see any reason to need more than that, anyway 22:05 celeron55 what would the acknowledgement line contain 22:06 celeron55 it's not exactly simple as there are many authors 8) 22:06 VanessaE "Screenshot of Minetest engine and minetest_game, CC-by-SA, copyright various, see Minetest source code and LICENSE file." maybe? 22:07 celeron55 "Attribution-ShareAlike 3.0 Unported (CC BY-SA 3.0) Copyright (C) 2010-2015 Minetest contributors"? 22:07 VanessaE that would work too 22:07 celeron55 i guess that's kind of the shortest you can go 22:07 VanessaE but, I'm no lawyer, so... 22:07 * VanessaE hands celeron55 a large grain of salt 22:07 celeron55 well me neither, and i will mention it in my reply 22:08 celeron55 (Do not take this as actual legal advice though; I am not qualified in that and if I was, these few lines would be way more expensive.) 22:08 celeron55 ^ ...in my own style 8) 22:08 VanessaE I like the way slashdot et al. put it with their polls: "if you're using these numbers for anything serious, you're insane." 22:08 VanessaE (or something to that effect) 22:16 celeron55 http://fpaste.org/225186/05773143/ 22:16 celeron55 i think i'll reply with this 22:18 VanessaE "and if I was, these few lines would be way more expensive." heh 22:18 sfan5 using a minetest screenshot as a cover for a science publication 22:18 sfan5 i don't even 22:18 VanessaE sfan5: ikr 22:19 VanessaE celeron55: lgtm 22:20 sfan5 celeron55: s/on the acknowledgement section/in the acknowledgement section/ ? 22:21 celeron55 hmm i wonder if this actually means they'll have to be able to license everything under CC BY-SA that they put over the screenshot including fonts? well i guess or at least hope not, that would be impossible 22:21 sfan5 well 22:21 sfan5 putting text on something is not really a devirative work 22:21 sfan5 you could have the text somewhere without the original work 22:22 VanessaE right, and besides which, it's not being released as its own file anyway 22:22 sfan5 we might be worrying about this too much :P 22:22 celeron55 whatever i'm not wasting more time on this, it's now sent 22:22 celeron55 it's not like they're running into any actual problems anyway 23:29 VanessaE YvesLevier reports the following crash: http://paste.ubuntu.com/11337875