Time Nick Message 03:09 hmmmm found a bug 03:10 hmmmm if a LuaPerlinNoise object is created in Lua, not by EnvRef::get_perlin(), it won't use the map's seed, but only the seed difference 03:11 hmmmm unless that's actually the intended behavior 03:11 hmmmm ah, nevermind, that's intended apparently 03:22 hmmmm hey guys, i am absolutely stumped with this compile error.. http://pastebin.com/7gqzWTYg when I call the constructor on line 95 here, I get this error: no matching function for call to 'LuaPerlinNoiseMap::LuaPerlinNoiseMap(NoiseParams*&, int, v3f&)' 03:22 hmmmm i don't get it, i am not passing any references at all 07:54 hmmmm no ideas? 08:01 hmmmm alright w/e 11:32 Taoki celeron55_: In the latest version, I'm getting a bug which causes water to stop being transparent at times. It didn't happen a few weeks ago however. Any idea what could be causing it and if there's a workaround? 11:43 Taoki Seems it depends on the angle you look at water from 17:07 kahrl_ hmmmm: v3f vs. v3s16? 17:07 kahrl_ I don't remember if there are implicit conversions between them 17:08 hmmmm uuuahh 17:08 hmmmm thanks 18:05 rubenwardy When there is not much space left on the hard drive minetest is on, blocks start to be regenerated 18:05 rubenwardy http://forum.minetest.net/viewtopic.php?id=3246 18:05 rubenwardy Has a fix been made into the c++ code? 18:09 hmmmm i tried fixing it but celeron said my fix was no good because it'd just clog up the blocks to save queue 18:09 hmmmm i haven't taken a closer look 19:15 thexyz anyone? https://github.com/celeron55/minetest/pull/446 19:17 darkrose haven't looked at it, but the first thing that came to mind was an entity with high velocity trying to load several hundred thousand blocks 19:22 darkrose s/hundred// 19:23 rubenwardy http://pastebin.com/daUaKzE3 error 19:26 thexyz darkrose: i think that should be counted as mod's problem 19:28 rubenwardy http://forum.minetest.net/viewtopic.php?pid=66120 19:28 thexyz hm.. 19:29 thexyz last time I had that kind of issues it was related to me messing with threads 19:29 thexyz rubenwardy: any steps to reproduce? 19:33 rubenwardy no 19:33 rubenwardy just playing on server 19:34 rubenwardy actually, I was taking stuff out of a chest 19:50 RealBadAngel hi rubenwardy 19:50 RealBadAngel hows the water+ is doing? 19:52 RealBadAngel thexyz, is the force_load workin? 19:52 rubenwardy good 19:52 rubenwardy added water dropping 19:52 RealBadAngel if it is, its a solution needed by many 19:53 RealBadAngel rubenwardy, i noticed some flaws, when one water node has a slight different level than surround 19:53 RealBadAngel it flips endlessly 19:54 RealBadAngel could you propably use more levels for calculations, and display less? 19:54 RealBadAngel or choose some avg to display 19:55 RealBadAngel like make calculations for 20 levels, display just 10 19:55 RealBadAngel that shall eliminate flicking 19:58 thexyz RealBadAngel: i'll test that one later 20:03 RealBadAngel but imho it shall not be given to an entity 20:03 thexyz why? 20:03 RealBadAngel maybe with groups? 20:04 RealBadAngel entities are being stored and often deleted 20:04 RealBadAngel what if a entity which was supposed to keep area loaded got deleted? 20:04 RealBadAngel whole idea screwed 20:06 RealBadAngel on the other hand node cannot be deleted, and can be named like world anchor or something 20:06 RealBadAngel player can place it and keep factiories or circuits workin 20:07 RealBadAngel and of course remove them by digging 20:25 thexyz groups are not as flexible 20:26 thexyz you can't remove a group from entity 20:27 thexyz this could be useful for carts — only set them `force_load` when it's needed (i.e. cart is moving) 20:28 RealBadAngel not remove group from, remove holder of that group 20:28 RealBadAngel no matter entity or node 20:29 RealBadAngel so if node was dug, unload. unless player is preventin it 20:30 RealBadAngel (cart can fell of the tracks, node can be dug by automat, thats why checks) 20:34 RealBadAngel so if you kill or dig entity from such group, unload check procedure shall be performed. same on create or place. 20:34 RealBadAngel *entity/node 21:19 VanessaE I think before we start allowing more of the world to be forcibly loaded, the emergethread issue will need fixed. 21:20 hmmmm that's literally next on my list 21:20 * VanessaE envisions someone using worldedit to modify a huge section of map, much of which isn't loaded yet. Emergethread will surely barf then. 21:20 VanessaE oh sweet. 21:20 hmmmm almost done with the perlin map things, but for some reason i'm not able to use my LuaPerlinNoiseMap object that i return in lua 21:20 VanessaE you'll figure it out I'm surew 21:21 hmmmm i compared what i'm returning with other functions that return objects and i seem to be doing everything right 21:21 hmmmm i might be missing a metatable or something 21:21 VanessaE RBA and I have been hard at work on moretrees, and I've gotten plantlife up to a fast enough speed that it outruns emergethread very easily now 21:22 hmmmm with the emergethread thing, we're going to have to do a lot of testing on an alternate branch. this isn't going to be a change that we can push to master in 5 minutes 21:22 hmmmm i'm pretty much entirely changing the way a core piece of the whole thing works 21:23 VanessaE I didn't expect it would be 21:27 VanessaE also regarding perlin maps, 21:28 VanessaE you may want to take a look at plantlife and evalute what I'm doing there. it might be a good idea to use comparable settings if possible. 21:28 VanessaE evaluate* 21:37 hmmmm ah solved it, i didn't realize i needed to call Register() 21:37 hmmmm all these gotchas.