Time Nick Message 00:02 est31 didnt we raise mapblock format version when the lighting stuff was done? 00:02 est31 the store info in an u8 thing 00:02 est31 or was that not merged? 00:03 Fixer i've opened save on yesterdays mt, mt from 03.12.2016 already had that warning with this map 00:03 Fixer not a problem though 00:04 Fixer I have this https://github.com/minetest/minetest/issues/4885 will post backtrace "soon" 00:07 est31 err what you are doing is almost surely generating a crash 00:07 est31 if a mapchunk is not fully in the map, bad things may happen 00:10 Fixer est31: i've saved that map by local map saving, it was first used as singlenode, then i changed it to v7 00:10 est31 yes 00:10 est31 enabling singlenode is the problem 00:10 est31 local map saving saves non complete chunks 00:10 est31 and apparently mapgen v7 doesnt like that very much 00:11 Fixer est31: is backtrace still useful? 00:11 est31 but I'm not a mapgen pro, ask paramat 00:11 est31 maybe he thinks otherwise, and says this is a bug 00:11 est31 Fixer: ask paramat :) 00:12 Fixer est31: wonder how to work play properly with such saves? 00:12 Fixer est31: but why it saves non complete chunks in the first place? 00:13 est31 Fixer: because the client gets sent mapblocks, not chunks 00:13 est31 and local map saving just saves whatever is on the local env 00:14 Fixer feels hacky 00:14 est31 local map is a big hack unfortunately :( 00:14 paramat i'm not sure about this, sofar may know more 00:15 paramat singlenode works the same way as mgv7 00:16 Fixer but it won't generate past borders 00:16 Fixer hmmmm 00:16 Fixer omg 00:16 Fixer wrong highlight 00:16 Fixer another way is maybe worldediting entire save and pasting into freshly created world 00:17 paramat singlenode generates the same volume as any other mapgen, just places air instead 00:18 Fixer yeah, but imagine that you want to have save, that has functional mapgen with original seed 00:20 Fixer anyone remembers how to run minetest in gdb? 00:20 Fixer nevermind 00:22 lhofhansl #4882 should be in the release 00:22 ShadowBot https://github.com/minetest/minetest/issues/4882 -- Fix computation of view range (in blocks) sent to server by Rogier-5 00:23 paramat should be considered yes 00:23 Fixer and I can't open that world in debug version o_0 00:34 Fixer ok, i guess i crashed it on build with some prs that change mapblock version, will redo it 00:37 Fixer excellent, catched the crash 00:40 Fixer hm 00:43 Fixer crash related to lighting? 00:44 Fixer est31: i need gdb help for sec 00:44 est31 ok 00:44 Fixer est31: I did "thread apply all bt" 00:44 Fixer is this enough? 00:44 est31 did you press enter until it didnt ask you to press more? 00:45 Fixer i pressed but it is soooooooo long 00:45 Fixer bt full gives more info 00:47 Fixer first page of "bt full" http://pastebin.com/raw/hURw8e0R 00:47 Fixer 1500 lines of this light spread and counting 08:12 paramat #4882 tested and works +1 08:12 ShadowBot https://github.com/minetest/minetest/issues/4882 -- Fix computation of view range (in blocks) sent to server by Rogier-5 10:38 Zeno` for those asking about the light LUT... 10:39 Zeno` at the time I added gamma there was compelling pressure for however gamma worked that the current, at that time, light values would remain the same by default 10:40 Zeno` the reason 1.8 was chosen was to at least try and have it at a point where it could be improved upon in the future 10:41 Zeno` but the main reason that it is the way it is was to retain the previous light values 10:42 Zeno` while at the same time doing two things: a) get gamma in there at least (it had been asked for for years); and b) act as a stepping stone 10:43 Zeno` if it wasn't for the strong opposition to changing the default light values then the code would already be like sofar's current PR 10:45 sfan5 paramat: can anything be done to make v7 caves look more like v6 caves? 10:45 sfan5 because IMO they are too narrow to properly explore 10:46 paramat erm, not really. i can check the width again 10:47 paramat i tuned the width by flying underground for a long time and finding the narrowest tuunel, then tuning to make sure the narrowest is walkable 10:47 Zeno` oh and the reason gamma correction is "backwards".... I was asked to do that as well 10:47 Zeno` even though it's just wrong 10:47 paramat the width tends to vary a lot so they still get wide at places 10:47 Zeno` but it should all be in the logs somewhere. In the end I just wanted gamma there. As a start 10:48 paramat but i can check again 10:49 paramat zeno, i see, i linked to this in the thread so people see 10:50 paramat ok so since mgv7 might become default i will check the tunnel width again 10:50 Zeno` thanks :) 10:51 paramat the large caves of mgv6 are in mgv7 and not much changed 10:52 Zeno` paramat, I can't recall exactly why I ended up agreeing to the compromises etc but it may have been close to a feature freeze and having *any* gamma was better than having none at all, so I just accepted people's objections etc and merged what I was able to without conflict. Especially considering how long people had been asking for the feature 10:53 paramat heh i can imagine people insisting on preserving the light values that we are now rejecting 10:54 Zeno` yeah 10:56 Zeno` as for my choice of 1.8 as default gamma I can explain that as well. It was either 1.8 or 2.2 (both kinda standard). 1.0 did not make sense because it would have made future changes harder (everyone would have had to change their configs from 1.0 to either 1.8 or 2.2 to make things look right again). So that choice I can easily explain 10:58 paramat #4847 needs review, seems to fix duplicate objects 10:59 ShadowBot https://github.com/minetest/minetest/issues/4847 -- Fix unnecessary block loading and duplicating objects by juhdanad 11:03 Zeno` (did not make sense as a default gamma I mean... even though code-side it would have, it would have made future changes more painful for users) 15:55 Zeno` what is the commit that suddenly opened console to fullscreen? 15:55 Zeno` I can't find it 16:02 Zeno` https://github.com/minetest/minetest/issues/3956' 16:03 Zeno` #3956 16:03 ShadowBot https://github.com/minetest/minetest/issues/3956 -- F10 console covers the entire viewport/screen 16:03 Zeno` that needs to be re-addressed / talked about again 16:04 Shara I would personally love for height of consol on f10 to be adjustable from options. I currently can't play using an official client due to this. 16:05 Shara Would have posted something there before if I had known an issue had been opened. 16:06 est31 well there is f10 console and the t console 16:06 est31 if you want a small one, use t 16:07 Shara est31: I don't want a small one. I want a nice overview which doesn't drag my eyes right to the bottom of the screen and give me a headache. 16:07 Shara t is also hugely inconvenient for a quick review, because in addition to it being so small, you can't close it from the same key. 16:08 Krock however, the F5 text should be pushed down when T or / is used 16:08 Krock F10 is supposed to be a chat-only view 16:08 est31 so basically, t is too small and f10 too big for you? 16:09 Shara Krock: it's not about wanting to play when using it, it's about the drop down motion of it pulling my eyes too far. 16:09 Zeno` est31, f10 has more text lines shown than my text editor and IRC COMBINED 16:09 Shara est31: Yes. I know some people like it fullscreen and for people with smaller screens it must be an improvement, but it should really be configurable in options. 16:10 Zeno` well, maybe not combined. But in IRC I have about 1/3rd the number of lines shown 16:10 Zeno` and in text editor 1/2 16:11 Shara I have tried to adjust to it many times, but no matter how I try to adapt, it leads to headaches within about half hour of me playing. 16:11 Fixer T-chat menu sucks 16:11 Zeno` I don't need 700 lines of useless chat text. I want f10 console open and be able to do stuff at the same time 16:11 est31 Zeno`: but how are you going to walk then? 16:11 est31 you cant use WASD 16:11 est31 because thats used for entering text now 16:12 Zeno` est31, you can see what's happening in front of you 16:12 est31 so whats bad about t? 16:12 Zeno` instead of being hidden behind a wall of irrelevant text 16:13 Zeno` est31, my main problem with it is that no other app on my computer displays 700 lines of text at once 16:13 Zeno` why should it? 16:13 Krock then don't use F10? 16:13 Krock regular T chat looks fine 16:13 est31 Zeno`: you can open minetest in non full screen 16:13 Shara t chat is useless 16:13 Zeno` Krock, seeing some history is fine... but more than T 16:14 est31 Shara: why? 16:14 Zeno` est31, but it still takes up the entire MT screen 16:14 Shara I want somethign that shows more than that, and that will open and close one one single key 16:14 Zeno` and for what purpose does it use all that space? 16:14 Shara Basically, what f10 used to do before .14 16:14 Krock we could go back in time and use the formspec again :3 16:14 Shara on one* 16:15 Shara Would it be difficult to simply have the height as an option? 16:15 Zeno` it should never have been merged 16:15 est31 not really 16:15 est31 Shara: well you convinced me 16:15 Zeno` unless height was provided as an option and defaulted to 25-30% of the screen height 16:15 est31 lets add an option for the height of f10 16:15 Shara Krock: You know this isn't a new issue for me, because back when it first happened I came to you for help to compile my own client. 16:15 Zeno` it's been an issue for me since it was first merged 16:16 Krock Shara, oh. So you're not playing with this view at all? 16:16 Zeno` if it was configurable, then no big deal 16:16 Shara Krock: I've tried it several times and simply can't adjust. 16:16 Krock :/ 16:17 Zeno` Shara, why should you have to adjust to something that is essentially useless anyway? 16:17 Shara I know Don started to compile his own clients as well as a result of it, though he's not around to ask about his experience now. 16:18 Shara Zeno`: It's fien for me if some peopel prefer it this way, and I don't want to take away from them. Just an option would be really appreciated. 16:18 Shara fine/people* 16:19 Zeno` looking at the comments, most people seemed to want it configurable anyway 16:19 Zeno` and it's something that should be configurable 16:21 Wuzzy I suppose that “feature freeze” also means “beta testers, on your marks!”, right? ;-) 16:22 Wuzzy Could someone please add a News post to https://forum.minetest.net/viewforum.php?f=18 then? 16:22 Wuzzy about the feature freeze, that is 16:23 est31 Shara: can you show a screenshot of your current setuü 16:24 est31 setup 16:24 Shara est31: I have mine set up so f10 behaviour is identical to in .13 16:25 Shara I'm currently just using OldCoder's clients since he includes my changes in his builds. 16:26 Krock Wuzzy, there already are some topics about the latest development 16:26 Shara Will grab a screenshot in a moment anyway 16:26 Krock there are not many people at all who test these feature freeze builds at all 16:27 Krock -"at all" 16:27 Wuzzy there will be even less if its not even announced :P 16:28 Zeno` est31, thanks for the additional label. I guess that's really what it is "a feature request" 16:28 Zeno` personally I hate the current behaviour (and so do others), and others like it 16:28 Fixer Krock: i test 16:28 Krock Fixer, yeah, you do :) 16:28 Fixer Krock: some server owners use latest git 16:28 Zeno` so the middle ground needs to be found I think 16:28 rubenwardy I suggest releasing 0.4.15-rc1 early for testing 16:29 rubenwardy to persuade people to test 16:29 rubenwardy maybe 16:30 Fixer sfan said he will update his builds after freeze, just use this + post a topic about freeze 16:30 Shara est31: .13 client is on left, current one I play from on right, both showing consol size: http://oi65.tinypic.com/2e0sck5.jpg 16:31 Shara As you can see, I basically just aimed to undo the change and nothing else 16:37 sfan5 Wuzzy: an RC build will be released when it's time 16:37 sfan5 the question is when 16:37 Wuzzy I see. :-) 16:38 Wuzzy just give the beta testers enough time to test and the translators enough time to translate. 16:38 Wuzzy take ur time 16:39 rubenwardy 0.4.14's feature freeze was announced: https://forum.minetest.net/viewtopic.php?f=18&t=14102 16:42 Fixer we need to do the same for 0.4.15 asking server owners who use latest to update to really latest post freeze version 16:45 est31 mhh I think I really should update the translation templates 17:28 Wuzzy est31: yes please :) 17:29 Wuzzy translator here: Does anyone here now what is meant with “to instrument” or “instrumentation”? 17:29 Wuzzy it is part of the profiler , apparently 17:55 red-001 updated #4468 17:55 ShadowBot https://github.com/minetest/minetest/issues/4468 -- Load textures from the subfolders in texturepacks. by red-001 18:25 hmmmm i have located paramat's mgflat: http://i.imgur.com/Qfvzx.jpg 18:30 est31 lol 18:33 Krock haha, right 18:35 VanessaE heh 20:07 shivajiva does anyone know of a way to stop a tool being dropped by a player like let's say the admin pick? 20:08 est31 if its an admin pick, you can easily check at each interaction if the player who wields it is the admin 20:08 est31 if the player is not the admin, you can remove the pick from the inv 20:09 shivajiva agreed but I wondered if there is a way to prevent the tool being accidentally dropped by the player 20:22 celeron55 https://paste.fedoraproject.org/504916/48148774/ 20:22 celeron55 "Subject: Are you okay with me putting Minetest on greenlight?" 20:22 celeron55 ehm... maybe not 20:28 Calinou Godot isn't GPL, so they're wrong :) 20:28 Calinou also, I already foresee the DLCs 20:29 Calinou "Minetest DLC: Add Stone" 20:29 shivajiva :/ 20:45 lhofhansl1 Please consider #4881 for 0.4.15. I realize it's not a bug fix, but it does go well with #4811, #4724, and #4746 to reduce server load. 20:45 ShadowBot https://github.com/minetest/minetest/issues/4881 -- Process ABMs in a spherical volume instead of a cuboid by lhofhansl 20:45 ShadowBot https://github.com/minetest/minetest/issues/4811 -- Optimize/adjust blocks/ActiveObjects sent at the server based on client settings. by lhofhansl 20:45 ShadowBot https://github.com/minetest/minetest/issues/4724 -- Improve number of blocks sent and occlusion culling. by lhofhansl 20:45 ShadowBot https://github.com/minetest/minetest/issues/4746 -- Don't limit vertical block send area to 1/2. by lhofhansl 20:54 sfan5 lhofhansl1: your pr is missing the increase of active_block_range 20:54 sfan5 or is that intended? 20:54 sfan5 oh nvm 20:55 lhofhansl1 Paramat requested removing that. 20:55 sfan5 :/ 20:55 sfan5 we need to increase it at some point anyway 20:55 sfan5 you can't think of lower powered devices always 20:56 lhofhansl1 I didn't feel that strongly. I agree 2 is way too small. I find 4 is close to optimal between what one can see and server resources (for my setup anyway). 20:57 lhofhansl1 It would be nice if things like this could auto-tune. For example when ABM times are too long, dial it down. 20:57 lhofhansl1 (hard to get right, though) 20:57 sfan5 that would be pretty easy to add 20:57 sfan5 it already complains about too high abm times on errorstream 20:57 sfan5 j 20:57 lhofhansl1 exactly. :) 20:58 lhofhansl1 I guess one could have an upper and lower bound and have the server tune within that range. 20:59 sfan5 however for this release i think we should take the opportunity to improve it and just ignore lower-powered devices for a second 20:59 lhofhansl1 what should the target be? 200ms when it reports a warning? Or lower, like 100ms? 20:59 sfan5 besides, the abm range can be set lower on android 20:59 sfan5 dunno, the 200ms complain-limit needs some thinking about too 21:00 lhofhansl1 200ms out of every seconds seems like a lot already (but depends on how many cores the server has, too) 21:01 sfan5 indeed 21:17 lhofhansl1 We can discuss on #4892 21:17 ShadowBot https://github.com/minetest/minetest/issues/4892 -- active_block_range (and other setting) should auto-tune based on server performance 21:22 sfan5 oh also 21:22 sfan5 est31, hmmmm: any thoughts about releasing an RC build on the forums for testing? 21:23 hmmmm ¯\_(ツ)_/¯ 21:24 hmmmm actually it might be a good idea. users registered on the mt forums are much more likely to be power users but still somewhat less than somebody who uses development versions 21:25 hmmmm it's not quite the general public, which i'd argue is through package repositories and the main page 21:25 hmmmm i doubt you'll find any bugs you didn't know about before though 21:27 sfan5 well maybe not bugs, but complaints about minor details 21:29 rubenwardy and a pointless discussion about 0.5 21:59 rubenwardy #4893 21:59 ShadowBot https://github.com/minetest/minetest/issues/4893 -- Fix camera jumping on Android when panning past 0/360 mark by rubenwardy 21:59 rubenwardy Wayward_One ^ 22:18 rubenwardy on a role today 22:18 rubenwardy #4894 22:18 ShadowBot https://github.com/minetest/minetest/issues/4894 -- Fix segfault on startup error on Android by rubenwardy 22:57 Fixer_ Calinou got 53 fps on LinuxGaming2 spawn... with shaders disabled on i7 and GTX10XXsomething 8gb ram... i got about 34 fps with core i3 + ATI HD6870 1gb 22:57 Fixer_ 53 fps, Carl 23:08 Fixer_ comparison was crude, Calinou had like 2x more pixels to draw and had fsaa 8x i believe 23:09 Calinou yes, 8x MSAA, though it probably wouldn't be much faster without AA 23:09 Fixer_ i think fsaa does not affect performance much when i tested 23:10 Fixer_ do you know what really SUCKS? 23:10 Fixer_ say you have vrange 240 and 20fps... you lower v_range... 23:10 Fixer_ and you lower it a lot 23:11 Fixer_ below 100, and only than your abismal 19 fps turns to 20... 24 23:11 Fixer_ 40 fps at vrange 50 23:17 Fixer_ https://i.imgur.com/JB9GSaN.png 23:55 paramat game#1450 has been updated with additional necessary changes. sofar 23:55 ShadowBot https://github.com/minetest/minetest_game/issues/1450 -- Default: Add missing 'default_dig_snappy' sound by paramat