Time Nick Message 00:07 Sokomine hi. i pulled latest git and compiled. it does seem a bit bright now (before, mt used to be a bit dark). is there any way to get it slightly darker again? preferably something inbetween :-) 00:19 FavoritoHJS I'm guessing tone mapping got enabled/disabled/changed/whatever between your previous install and your latest commit one. 00:22 Sokomine thanks! do you have any idea what i might change? 00:25 FavoritoHJS tone mapping is (afaik) kinda wip, for example, there isn't anything in the settings other than an enable/disable toggle 00:26 FavoritoHJS actually, what does tone mapping even do? 00:27 FavoritoHJS I recall darks being black when I enabled it, so I guess it's some gamma correction thing, but how does it work exactly 00:29 FavoritoHJS interesting, the advanced settings say it's "Filmic tone mapping", so I guess that's exactly what it does 00:33 FavoritoHJS yup, seems like it affects stuff from 364 onwards in client/shaders/object_shader/opengl_fragment.glsl 00:37 FavoritoHJS interestingly though fog doesn't seem affected, as it's added after the tone map... bug or artistic decision? 00:44 MTDiscord IIRC tone mapping stretches contrast at the mid range, while compressing it at the extremes. It basically creates the impression of stretched contrast but without the "clipping" problem that more linear methods have. 01:35 Sokomine aah. thanks for explaining 10:17 robyn ls 10:41 ROllerozxa cd ~/.minetest 10:42 robyn sorry about that, and hello 10:43 ROllerozxa hi there 10:47 ROllerozxa Sokomine: From what I can tell, tonemapping has changed slightly now with the postprocessing pipeline 12:07 MinetestBot 02[git] 04Wuzzy2 -> 03minetest/minetest: DevTest: Explain purpose of most items in tooltips (#12833) 13c1c6877 https://github.com/minetest/minetest/commit/c1c68775b25314b566a2aae9a304d1b470f5794d (152022-10-09T12:05:22Z) 12:07 MinetestBot 02[git] 04Wuzzy2 -> 03minetest/minetest: Improve documentation for `liquid_alternative_*` (#12810) 13b2a3f53 https://github.com/minetest/minetest/commit/b2a3f53b297d4aec37f8272992f8e9b465d43a43 (152022-10-09T12:05:40Z) 12:07 MinetestBot 02[git] 04fluxionary -> 03minetest/minetest: add an 'equals' method to ItemStack and compatibility w/ lua '==' (#1… 13440d966 https://github.com/minetest/minetest/commit/440d966b939059dfa51604eb68d61eecb12baeb4 (152022-10-09T12:06:09Z) 13:21 Sokomine ROllerozxa: hm. do you have any idea how i may change it to more reasonable lighting? 13:23 rubenwardy gamma 13:37 * Sokomine greets rubenwardy and looks puzzled 14:50 MinetestBot 02[git] 04TurkeyMcMac -> 03minetest/minetest: Optimize lighting calculation (#12797) 139676364 https://github.com/minetest/minetest/commit/9676364c1fe5fe5eae69c55e7d7a45392decfb2d (152022-10-09T14:50:26Z) 14:58 sfan5 oh hey there's a new one of those crypto servers 15:00 ROllerozxa that's... quite the amount of players they've got 15:20 FavoritoHJS oh, you mean one of those play to earn servers that _totally_ isn't botting their playercount? 15:23 FavoritoHJS (for best results, say play to earn with a highly distorted voice a la J. Sterling) 15:26 ROllerozxa it's almost like a server that pays you in ""cash"" for playtime will end up being filled with players AFKing to game the system 17:12 sfan5 Sokomine: the brightness issue has been fixed 17:13 MinetestBot 02[git] 04sfan5 -> 03minetest/minetest: Fix tone mapping being always enabled 137a28f2c https://github.com/minetest/minetest/commit/7a28f2c4fa50957a019a8d44856b7aa0cea306eb (152022-10-09T17:11:04Z) 17:44 MinetestBot 02[git] 04TurkeyMcMac -> 03minetest/minetest: Remove unused MapBlock functionality 138f996e4 https://github.com/minetest/minetest/commit/8f996e4a7cf7b1fa1ca683cbc37327a790e09ddc (152022-10-09T17:43:48Z) 17:44 MinetestBot 02[git] 04TurkeyMcMac -> 03minetest/minetest: Embed data directly in mapblocks 13b3503e7 https://github.com/minetest/minetest/commit/b3503e7853a52a8c16431f6b983e30c9d25951bc (152022-10-09T17:43:48Z) 17:46 Sokomine sfan5: thanks! 17:50 MTDiscord "embed data directly in mapblocks" ... cache hit rate optimization? 17:53 Krock more like removing a level of indirection 17:54 Krock could be that it's easier to cache, not sure what's happening with CPU-side optimizations 17:54 MTDiscord I guess less indirection is nicer conceptually too, though I figured that you'd save a little cache noise that way too. 17:55 MTDiscord it'd be weird if it were like a HUGE improvement, but hopefully it all adds up anyway 17:55 jwmhjwmh I believe I noticed a small improvement in the lighting benchmark's time with data embedded directly in mapblocks. 17:55 MTDiscord Nice. In a mature project, pretty much the only way to make big improvements is to add up small ones. 17:56 Desour most parts of minetest aren't mature in that regard 18:00 Desour for caching, shouldn't you have moved that one mapblock on the stack into a unique_ptr, jwmhjwmh? I kinda doubt that pushing 16ki node data on the hot part of the stack is good for caching 18:01 Krock which stack? 18:02 Krock can we move this to #- dev? 18:02 Desour the call stack 18:02 Desour :O move discussion from #minetest to #minetest-dev ? 18:02 Krock it's development after all 18:03 Krock or whatever... I guess it does not matter 18:03 Desour well, it's just µcro concerns 18:17 MTDiscord microcro concerns? 18:17 Desour µ is mü 18:17 Desour my 18:17 Desour so µcro is mycro 18:18 Desour oh, mu in english https://en.wikipedia.org/wiki/Mu_(letter) 18:34 FavoritoHJS why is this shite crypto server at 52 players? that isn't afk numbers, that screams botting 18:37 FavoritoHJS W00HOO!? ET TU, MY FRIEND!? 18:39 FavoritoHJS I can only hope he's being impersonated... 18:40 MTDiscord I'm joining to take a look at the player activity 18:42 FavoritoHJS Are there any rules on the server list? 18:43 sfan5 no formal ones 18:43 MTDiscord First observation: There do indeed seem to be 52 players on the server. Most are idling (not moving). It's hard to tell whether they are bots or "legit" AFKing players motivated to AFK by fictitious laughable crypto "rewards" which Minetest busywaiting in the background will make up for by driving up their parent's energy bill. 18:43 MTDiscord parents'* 18:44 FavoritoHJS > no formal ones 18:44 FavoritoHJS that can't end well... 18:44 FavoritoHJS goshdarnit I thought that would be sent as one multiline message... DARN YOU KIWIIRC 18:45 MTDiscord FavoritoHJS: propose a rule, sfan5 is on here 18:45 MTDiscord I don't think you could reasonably whack this server with a rule due to what I have outlined above 18:46 FavoritoHJS I'm not a moderator, I don't know how to make a rule that is useful, general yet not exploitable 18:46 MTDiscord Unless the admins admit to botting you can't tell whether it is botting 18:46 MTDiscord esp. whether it's botting by the admins - otherwise I could have servers penalized simply by botting 18:47 ROllerozxa I wouldn't be surprised if it's actually real players, just ones who are afking to get free PAWcoins or whatever they're called 18:47 FavoritoHJS yea, it would make sence it's the whole crypto clown car, that doesn't count as botting 18:48 sfan5 unrelated to the crypto thing it would be useful if someone spent some time to write down a sane ruleset for the server list 18:57 FavoritoHJS As obvious as it is, I suggest something like "Servers on the list shall not be meant to hurt others, lest they be delisted"... but to define "be meant" and "hurt others" in a sufficiently clear way, while ensuring that legitimate mistakes don't mean a delist (or worse...) is probably quite complex. 19:00 FavoritoHJS I do not want rules arbitrary or obscure enough that a 1.19.84-style revolt is even close to a possiblity... then again, having no rules at all will blow up one day, so... 19:11 FavoritoHJS bug report: fog affects your hand, so if you lower it all the way down to 20 your hand gets covered in fog 19:12 Desour that's intended, afaik. 19:12 Desour back when a viewing range of 5 was still possible, I saw this all the time, but stronger 19:14 MTDiscord sorta makes sense after all 19:18 Oblomov yeah, anybody who's been in thick fog knows that's really the case 19:25 MTDiscord Re server list rules, the simplest thing to do is start with the empty set, and add rules as problems are discovered. "Attempting to game the system to promote your server may result in demotion, delisting, or other countermeasures." Keep it pretty simple, because you're going to have to use judgement on it anyway. 19:33 FavoritoHJS the strange thing is that your hand can be in fog, but nodes that are (presumebly) further away aren't. 19:45 FavoritoHJS can't craft an axe "backwards", and it's screwing with my muscle memory, send help 19:46 MTDiscord visit contentdb, there is a mod for that 19:46 MTDiscord https://content.minetest.net/packages/Hugues%20Ross/mtg_craft_reverser/ 19:48 Desour if you craft the axe the wrong way, the sharp side is aimed at you, and you'll hurt yourself when using it 19:49 Desour luatic: > I don't like the style of it; I want my doc comments to be at the appropriate places (e.g. param comments next to params, return comments next to return statements, function description next to function definitions) 19:49 Desour you were referring to ldoc, right? 19:49 MTDiscord yes 19:49 Desour doc comments in ldoc are next to their functions 19:50 Desour what exactly do you mean with the places? 19:50 MTDiscord https://github.com/TheAlgorithms/Lua/blob/main/src/sorting/insertionsort.lua f.E. 19:51 MTDiscord you can put comments in the parameter list of a function 19:51 Desour O_o 19:51 MTDiscord and you can put comments in the explist of a return statement 19:51 Desour that style looks a bit weird to me 19:52 MTDiscord well, I'm probably the only one using it because I'm yet to write a doc generator that recognizes it 19:53 Desour do you have a tool for converting it to html or similar? 19:53 Desour ah 19:53 Desour an advantage of ldoc is that the function doesn't need to actually exist 19:54 Desour also the doxygen-like style is more common 19:54 MTDiscord it may be, but it reminds me of Java 19:54 Desour haha 19:54 Desour that downside 19:55 MTDiscord anyways, gtg 19:56 Desour bb 20:37 ensayia Is it possible to swap the textures of a node in the world? 20:38 MTDiscord no 20:39 ensayia Is it a planned feature or just something not possible in the engine? Any insight? 20:48 rubenwardy ensayia: what are you trying to do? 20:48 ensayia a simple texture swap on action, like if I want a furnace texture to change when I light it, for example 20:48 rubenwardy ok, that's done with multiple nodes 20:48 rubenwardy in MTG, I believe it's default:furnace and default:furnace_on 20:49 ensayia that's what I currently do, it's just annoying managing the metadata because the furnace is a player-facing object and you have to preserve that metadata across swaps 20:49 rubenwardy swap_node won't delete meta 20:50 MTDiscord IIRC swap_node was originally created specifically because of the furnace :-) 20:51 ensayia facing is stored in the param2, which 'defaults' to player facing when the item is place down by a player but swap_node does not preserve that behavior as far as I can see 20:51 MTDiscord swap node probably swaps the whole node, not just the name 20:51 MTDiscord so you take the original node, change the name, and pass that to swap_node 20:52 rubenwardy local node = minetest.get_node(pos) 20:52 rubenwardy node.name = newname 20:52 rubenwardy minetest.swap_node(node) 20:52 rubenwardy * swap_node(pos, node) 20:53 ensayia indeed, but the player placing the node does not write the facing value to param2 when it happens, therefore if the node you are swapping it for has "parameter2" set to "facedir" and it placed by swap_node and not a player it doesn't keep the direction it was originally placed in, because the param2 is not written to on player placement 20:54 ensayia sorry, that last bit was redundant 20:54 rubenwardy you should make sure that the before and after both use the same param2type 20:54 MTDiscord paramtype2 20:54 MTDiscord consistency yay 20:55 ensayia they do 20:55 rubenwardy then the param2 will be kept and the rotation won't be changed 20:55 rubenwardy param2 is how the rotation is stored 20:56 rubenwardy in order for the player to place and rotate a node, it needs to change param2 20:56 ensayia but when 'facedir' rotates the node to face the player that direction isn't saved, it's a function of the property, therefore a node with 'facedir' that was not placed by the player obviously can't face a player and doesn't have the param2 to reference 20:56 rubenwardy it will be saved - facedir is stored in `node.param2` 20:56 rubenwardy there's no other way to rotate a node 20:58 rubenwardy the MTG furnace does the code I wrote: https://github.com/minetest/minetest_game/blob/master/mods/default/furnace.lua#L115 20:59 ensayia if swap_node preserves metadata, why is localizing the node table and swapping out the name needed? 20:59 rubenwardy because metadata is data about the node, not the node itself 21:00 rubenwardy node contains name, param1, and param2 21:00 rubenwardy so getting the node table and only changing name means that you keep param1 and param2 the same 21:00 rubenwardy swap node is documented as " Set node at position, but don't remove metadata" 21:01 ensayia Well, that's the way I'm doing things now I was just seeking to understand. What's the hangup with just swapping a texture? 21:01 rubenwardy the problem with that is where do you store that texture? 21:01 rubenwardy there's no space in the node data 21:01 rubenwardy it would need to be in the meta data, but that impacts performance 21:02 rubenwardy there are long term plans to do something like this, it's called meta set def 21:02 ensayia I see. I use a texture atlas for most everything so it might be worth it to define mutable atlas coords? 21:06 FavoritoHJS a friendly reminder that having 64x64 textures in a world where 16x16 textures are the norm just looks weird.