Time Nick Message 00:02 khonkhortisan poke https://github.com/minetest/minetest/pull/764 00:06 PilzAdam khonkhortisan, reproduced the bug and your fix works 00:06 khonkhortisan ok 00:07 PilzAdam does anyone object to merging this? 00:08 khonkhortisan I don't think anyone will 00:11 PilzAdam khonkhortisan, btw: give your commits a proper commit message and dont name your branches "patch-3" 00:11 khonkhortisan patch-3 is because I went through the github website to make the change 00:11 PilzAdam yes, I know 00:11 PilzAdam still not good 00:12 khonkhortisan you made the other channel talk about me 00:25 PilzAdam if anyone wants to merge https://github.com/minetest/minetest/pull/743 then go ahead, I have nothing against it 00:25 PilzAdam 01:00 hmmmm sweetbomber, i'm not sure i really understand. what do you mean by "prng iterations", exactly? why does this prng have this property where we can know at least one block has a structure if the entire "superblock" has a structure? (this doesn't really sound like any 'random') 01:01 hmmmm also, you seem to be assuming that structures are a pretty rare occurance 01:02 hmmmm when *I* say structure, i mean any discrete, relatively static cuboid of nodes placed at some point on the map - trees, cave spikes, desert temples, village homes, etc. 01:03 Exio are dungeons "structures"? :P 01:04 hmmmm not really sure 01:04 hmmmm they wouldn't be generated using the schematic decoration type, at least 01:06 hmmmm the thing is, i'm not quite sure what to expect with what sweetbomber is working on... but like i said before, it seems that he intends for this to be mostly used for bigger things 01:06 hmmmm so it might be a good idea to just continue on with what i'm doing and leave cutoff handling out 01:10 Exio so, what's up with #754? 01:11 Exio it works, it just had the "wrong commit message" before 01:28 * Exio bugs kahrl 01:28 kahrl oh, hi 01:28 Exio :P 01:28 kahrl I'm in favor of merging it 01:28 kahrl anyone else? 01:51 Tesseract Core devs: To close: https://github.com/minetest/minetest/issues/749 Also this: https://github.com/minetest/minetest/issues/726 01:54 kahrl closed #749 01:55 kahrl #726: "However it is recorded by the closesed player, not the actual placer." is that true? 01:55 kahrl it's worse than I ever even knew 01:56 kahrl 01:57 kahrl server.cpp has a RollbackScopeActor active around the on_use call so it should record the actual placer 01:58 Tesseract kahrl: If not for the bucket it does it for a lot of things, for example blinky plants blinking. 01:59 Tesseract (Limited to a range of 100 IIRC) 02:00 kahrl that's right, if there is no RollbackScopeActor active it tries to guess 02:01 Tesseract Anyway, the whole thing should be redone. Does my set_node_as() Idea sound good? 02:03 Tesseract As nice as being able to rollback worldedit is, it probably slows it down quite a bit. 02:03 kahrl what would be the use case for set_node_as? 02:03 Tesseract (Ans uses a lot of disk space) 02:04 Tesseract Things like buckets that should be recorded. set_node() wouldn't record. 02:04 kahrl buckets already record it because of the RollbackScopeActor 02:05 Tesseract Yes, but other things like buckets. 02:05 kahrl such as? 02:06 kahrl as long as stuff is done in on_use it is recorded with the player that used the item 02:07 Tesseract Yes, I mean out side of un_use, although I can't think of an example ATM. 02:08 Exio maybe something like a remote trigger? (just thinking) 02:09 Exio (a piston doing something because some player triggers the mesecon behind it or w/e) 02:09 Tesseract Yes, like TNT activated remotely. 02:09 kahrl It's tough to track all the possible ways to trigger mesecons 02:09 Tesseract But blinky plant blinks shouldn't be recorded. 02:10 Exio kahrl: i know, i'm just saying a bad example but "valid" :P 02:10 Tesseract ^ 02:11 Tesseract TNT remotes would be easy. 02:12 kahrl For the worldedit issue I hope someone designs the lua MapVoxelManipulator soon 02:12 kahrl Which could either not record into rollback at all or do it more efficiently 02:13 Tesseract Maybe a seperate setting for WE rollback. And a different file or something. 02:14 kahrl wasn't somebody planning a rollback revamp? things like that could be part of that 02:14 Exio the one with the MVM idea :P 02:15 Tesseract It was put in the TODO. But nobody started it. 02:21 Tesseract https://github.com/minetest/minetest/issues/661 02:24 Tesseract This/a derivative never actually got merged: https://github.com/minetest/minetest/pull/608 02:25 Tesseract khonkhortisan: Irrlicht issue? https://github.com/minetest/minetest/issues/607 02:38 sweetbomber the worst thing about using open-source: sometimes you spend half-hour adding a feature you need, and then find out there was an option for it 02:38 sweetbomber in closed source you would just spend half-hour looking for the option 02:39 Tesseract Heh. 02:42 Exio what thing? :P 03:26 khonkhortisan I have no idea where that issue is from. I haven't looked into it. 03:32 Tesseract Well it seems to be an Irrlight issue, so file it in their bug tracker and close it here. 06:37 hmmmm what do you guys think of clay becoming a regular ol' ore that just occurs in sand 06:38 VanessaE why not? 06:38 VanessaE it'll have the same function anyway 06:38 VanessaE as long as it occurs in sheets of course 06:38 hmmmm yeah, it'd be better though. 06:38 hmmmm the clay placement is actually bugged but nobody ever really noticed it 06:39 hmmmm plus, in minecraft, clay occurs at the bottom of the sea as well, not just on the shore next to water 06:39 hmmmm i think it'd be better balanced if there were more clay 06:40 VanessaE absolutely 06:40 VanessaE there's far too little clay now 06:40 VanessaE on any heavily-played server, clay and papyrus are pretty much the first two things to run out 06:40 VanessaE though papyrus grows now, 06:40 hmmmm papayrus runs out because it doesn't regrow 06:40 hmmmm right, so that's not really a problem 06:41 VanessaE so that's no longer an issue 07:09 khonkhortisan Early I learned not to build with brick 17:01 sweetbomber peek this: https://github.com/minetest/minetest/pull/775 17:03 PilzAdam ehm, 90% of this patch is useless 17:03 PilzAdam why do you remove all these spaces? 17:04 sweetbomber only noticed now the 50 removals. 17:05 sweetbomber it seems the file was filled with trailing spaces in the end of lines, and my editor removed thos 17:05 PilzAdam also, leaving old code commented out is unnecessary, we have a git history 17:05 VanessaE sweetbomber: that reminds me, how did my game fair in your profiling? 17:06 VanessaE fare* 17:10 sweetbomber PA: ill fix that 17:10 sweetbomber VanessaE: http://speedy.sh/e52Ga/callgrind.out.vanessa 17:10 sweetbomber VanessaE: use kcachegrind to inspect it 17:10 VanessaE um 17:10 VanessaE that requires some kind of windows app to downloa. 17:10 VanessaE download* 17:10 sweetbomber ? 17:11 sweetbomber omg 17:11 sweetbomber lame site 17:11 sweetbomber wait, ill reupload somewhere else 17:11 sweetbomber i just trusted google 17:11 Exio base64 + pastebin ;P 17:11 PilzAdam google is evil, use ixquick.com 17:11 Exio (base64 if binary) 17:14 sweetbomber https://docs.google.com/file/d/0BzRoxG6zdl1LM19qdU4zVnQyNEk/edit?usp=sharing 17:14 sweetbomber this should do it 17:15 sweetbomber Exio: ill note those 17:15 sweetbomber PA: regarding the trailing spaces, any problem on removing them? 17:16 sweetbomber from the kernel coding style, they are not supposed to be there anyway... 17:16 VanessaE I say leave that part of the patch in there, it's not like the compiler gives two shits about it anyway 17:16 PilzAdam sweetbomber, dont fix code style of code you are not modifying 17:16 VanessaE sweetbomber: got the download. 17:17 sweetbomber ok, ill delete the pr and fix it 17:26 sweetbomber PA: it keeps the 48 additions/deletions 17:27 PilzAdam its fine now 17:27 PilzAdam https://github.com/minetest/minetest/pull/775/files 17:27 sweetbomber ill try to checkout the branch back to see if they disappear 17:27 sweetbomber ok, ill leave it as is, then 17:28 Exio i wonder 17:28 Exio what does it need to get merged? :D 17:28 PilzAdam the second commit shows the diff agains the first one, but the diff against master is fine 17:29 Exio squash then :P 17:29 PilzAdam the one who merges can squash the commits 17:31 sweetbomber as you can see, this commit removes a lot of string searches 17:32 sweetbomber which were using a lot of time 17:35 thexyz sweetbomber: I agree, they should be removed, but not in this pull request 17:37 sweetbomber thexyz: what do you mean? 17:38 thexyz uh, spaces 17:38 thexyz always forget the most important part 17:39 sweetbomber you are talking about the first commit removing a lot of spaces from random lines? the second commit makes the pull request only change 3 lines... 17:40 thexyz that was a reply to this >PA: regarding the trailing spaces, any problem on removing them? 17:40 thexyz anyway, merging it right now 17:43 sweetbomber ok, np. i also think that any general clean-up should be made at all the code in one single pull request aimed at cleanup only 17:44 thexyz that's right 17:44 PilzAdam this will probably cause merge conflicts with other pull request 17:44 thexyz that's right as well 17:49 Exio well, what about the #772 but only shader part? 17:50 PilzAdam Taoki has listed good points to not merge it 18:02 sweetbomber i never worked with 3d engines, but shouldn't it be easy to turn shadows on? 18:02 sweetbomber in my humble opinion, i think that irrlicht was the responsible for rendering shadows on blocks... 18:03 sweetbomber isnt there any option like "shadows = true"? 18:03 sweetbomber i repeat, im a complete noob in 3d rendering engines 18:04 Calinou shadows are made by software 18:05 VanessaE qbbl 18:05 VanessaE bbl too. 18:05 ShadowNinja After running a trailing whitespace stripper: http://pastebin.ubuntu.com/5768594/ 18:07 sweetbomber Calinou: just in minetest, or generaly on 3d software? 18:08 Calinou depends 18:11 ShadowNinja https://github.com/ShadowNinja/minetest/compare/whitespace_strip 18:11 ShadowNinja Compiling... 18:12 PilzAdam what about only the ones where the line is not empty? 18:14 ShadowNinja The whitespace still shouldn't be there. 18:14 ShadowNinja Although I could probably exclude those if you realy wanted. 18:15 ShadowNinja It compiles and runs. 18:15 PilzAdam I mean, look how bloated that diff is 18:16 thexyz ShadowNinja: don't touch lua 18:16 thexyz and other libs 18:17 thexyz no one is going to edit them anyway 18:17 ShadowNinja Yes, I will have to ignore those... 18:19 thexyz and what are we going to do with broken patches? 18:20 Exio well 18:21 PilzAdam if we exclude empty lines then its not such a big problem 18:22 thexyz PilzAdam: what's the point of this then? 18:30 Exio i wonder, is mapblock_mesh.cpp:1202 needed for something? 18:30 Exio https://github.com/EXio4/minetest/commit/f8c8aedfcd01236257e865d26bf6f5f02a4df979 the line removed 18:30 Exio :P 21:29 hmmmm alright, gonna start getting work done 21:30 hmmmm for a given biome, you'd like to list the decorations that go in it, right? 21:31 hmmmm that's just how the loading goes. really, each decoration would have a list of biomes since this is the most efficient and easiest thing to do 21:32 VanessaE what decorations, and what specific conditions within that biome are necessary to produce the given decoration. 21:32 hmmmm thing is, i'm not sure how crazy people are going to go with this; i'd like to keep it a linear search but i know people are going to act stupid and have 256 biomes 21:32 VanessaE trust me, people WILL abuse it. 21:32 hmmmm i don't want to think defensively like this 21:33 hmmmm it's such a small, stupid detail 21:33 hmmmm and i'd like to optimize for the common case 21:35 VanessaE will a 'biome' also define where decorations can go? 21:35 VanessaE I mean, for example, 21:35 VanessaE a "sandy beach" biome versus "a random patch of sand with no water", versus "sand near grass" 21:36 VanessaE will elevation be a factor? temperature? etc etc etc etc 21:36 hmmmm biomes have a specified set of decorations 21:37 hmmmm a decoration is placed if the randomly chosen place is the correct biome, and the node below it is of the correct type 21:37 VanessaE right 21:37 VanessaE but how detailed is the biome def itself? 21:37 hmmmm for example 21:37 VanessaE I mean, are we talking plants_lib levels of detail here? 21:37 hmmmm you have grass that spawn in plains but not on mountains 21:38 hmmmm both have dirt_with_grass as top nodes 21:38 hmmmm i don't know what that level of detail is 21:38 VanessaE hmmmm: https://github.com/VanessaE/plantlife/blob/master/API.txt 21:39 VanessaE line 241 in particular 21:39 VanessaE (the stuff above that's for ABM's only, like 241 describes the mapgen hook) 21:39 hmmmm doesn't have the sradius, and instead of savoid, it's just a single node type where it's allowed 21:40 VanessaE nono skip down. 21:40 hmmmm oh 21:40 hmmmm that's what you define a bioem as 21:40 VanessaE yes. 21:40 hmmmm biome* 21:42 hmmmm your biome definition seems to have some things from the decorations put together 21:43 VanessaE yeah 21:43 VanessaE this all predates your mgv7 work of course 21:43 VanessaE so I expect plenty of differences, since you'll surely find better ways to do this stuff 21:43 VanessaE (nevermind it all being C++)