Time  Nick         Message
03:03 VanessaE     celeron55_: sicne a generic menusplash no longer works (and since you coded it), there needs to be a way for a texture pack to supply a menusplash image specific to whatever games it wants to cover.
03:03 VanessaE     e.g. gamename_menusplash.png (as well as footer/header).
03:04 VanessaE     either that or the client needs to *not* use a custom game-oriented menu theme.  Either it's a client/server paradigm or it isn't.
03:14 VanessaE     also, the game filter buttons should only be displayed in the single player tab
03:15 VanessaE     maybe also the advanced tab, but definitely NOT in the multiplayer tab
03:15 VanessaE     or the settings/credits tabs
04:07 VanessaE     furthermore, aside from the button image for minetest_game, there should be an overlay identifying it, something like what we had before your change
04:07 VanessaE     or perhaps those really nice ones cisoun did.
04:47 celeron55_   VanessaE: you're wrong; we have to have some special handling for singleplayer games to make things reasonable
04:47 celeron55_   i agree that theming the multiplayer and settings tabs is kind of odd 8)
04:49 celeron55_   kahrl, hmmmm: minecraft's nether has a high ambient light
04:49 VanessaE     quite; I'm working on a custom theme for my game, but I'd still like to be able to override it with HDX
04:49 celeron55_   that is, total darkness is still visible
04:49 hmmmm        right, so what i said
04:50 celeron55_   VanessaE: also i agree with the texture pack thing; somebody can implement it
04:50 VanessaE     you broke, you fix it ;)
04:50 celeron55_   hmmmm: the lighting values in nodes are the same as in caves, but the lighting translation table is just different
04:50 hmmmm        this is a formal call for an addition to the API to set the client's light levels
04:50 celeron55_   that should be easy 8)
04:50 hmmmm        very easy
04:50 hmmmm        and we'd also be getting a fullbright feature while we're at it
04:51 hmmmm        i don't understand why this hasn't been done before
04:51 hmmmm        i would do it, but i feel guilty every time i add some new lua api because that delays sapier's change even more
04:51 hmmmm        besides, this is more of a schoolwork day
04:52 VanessaE     are his changes ready to use yet?
04:52 celeron55_   well i think such changes should be done at a time when others aren't interested in doing things
04:53 hmmmm        that's ridiculous, there will never be a time when nobody would like to add something new
04:54 celeron55_   sure there has been, if only the script api is considered
04:54 celeron55_   for the whole thing, no, but for that part, yes
04:54 hmmmm        for me at least
04:55 hmmmm        there's this decoration nonsense that has to go in
04:55 hmmmm        set client lighting levels
04:55 hmmmm        mapvoxelmanipulator
04:55 hmmmm        uhhh
04:56 hmmmm        oh, the after environment creation callback
04:56 celeron55_   well, i think sapier has to come and ask for a scriptapi freeze when he is ready to update it in a few day's time
04:56 celeron55_   guessing when he will do something is useless
04:57 hmmmm        i thought you sorta took over that
05:00 celeron55_   i was predicting there wouldn't be much interest in adding anything after your hud stuff, but it turned out that was wrong
05:01 hmmmm        not my hud stuff
05:01 hmmmm        i just added polish
05:05 celeron55_   your hud commit then
05:05 celeron55_   it's not relevant who made it and who polished it :P
05:05 celeron55_   (for this discussion)
05:10 kahrl        what exactly does sapier scriptapi commit do again? I keep forgetting
05:11 kahrl        +'s
05:12 celeron55_   mostly move stuff and files around
05:12 kahrl        is it the scriptapi_subfolders branch?
05:13 celeron55_   i don't even know 8)
05:13 kahrl        I wonder if he could implement it as a script rather than a set of commits
05:13 kahrl        so that it could be applied at any time
05:14 kahrl        high chance of breaking something though
05:14 hmmmm        oh totally different but
05:14 hmmmm        what do you guys think about changing the lighting scheme so that there could be light sources with longer ranges
05:15 celeron55_   it also puts the api of minetest.env into minetest (and preserves compatibility)
05:15 celeron55_   hmmmm: how would you do that
05:16 hmmmm        when spreading light sources, don't stop when you hit a greater light value
05:16 hmmmm        so let's say we have some super torch that spreads for... dunno.... 25 blocks
05:17 kahrl        that would actually decrease the light and cause glitches, would it not
05:17 hmmmm        erm
05:17 hmmmm        not if you do it carefully
05:17 hmmmm        it's obviously more complex than i'm making it out to be
05:18 hmmmm        so we have a light source that spreads for 25 nodes; the light values we'll propogate are say, 14, 14, 13, 13, 12, 12, 11, 11, 10, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1
05:18 hmmmm        for example
05:18 hmmmm        it doesn't have to be that way
05:18 hmmmm        which values it duplicates is some implementation specific detail that can be tuned
05:19 hmmmm        the only problem here would be is if you have that light source with a greater spread 'blocked' by a light source that may be brighter even that has a lesser spread
05:20 hmmmm        which is why i was saying that we'd have to keep going until the light value decays completely instead of stopping once an equal or greater light is hit
05:20 hmmmm        obviously this would increase the amount of computation to do lighting
05:21 hmmmm        but 1). it's a strict enhancement; this could be disabled if preferred, and 2). lighting is already a lot faster than it used to be so the impact might not be noticable
05:26 kahrl        how would unspreadLight work?
05:26 hmmmm        i am not really sure what unspreadLight does
05:26 kahrl        wouldn't it have to check a double radius for other light sources
05:27 hmmmm        i completely omitted unspreadLight from my optimized version of lighting
05:27 kahrl        it removes the light emitted by one node
05:27 kahrl        afaik
05:28 hmmmm        ahh. yeah, i guess it's necessary if you're modifying existing lights
05:29 hmmmm        unspreadLight would be more difficult to get working i guess
05:31 hmmmm        what's stopping us from adding dynamic lights right now?
05:34 kahrl        you mean hardware lighting?
05:35 kahrl        or the dynamic lights implementation from the 0.3 days
05:35 hmmmm        hardware dynamic lights, yes
05:35 VanessaE     oooo
05:35 hmmmm        tell me more about the dynamic lights from 0.3 though
05:35 kahrl        I don't know much about it
05:36 kahrl        it added some data to certain mapnodes on the client near the player
05:36 hmmmm        oh..
05:36 hmmmm        dumm
05:36 kahrl        so that nodes near the player would be illuminated if wielding a torch
05:36 hmmmm        if we're doing torches like that, it has to be hardware dynamic lighting
05:36 hmmmm        any other way is really hackish
05:37 kahrl        hardware lighting would be great but you'd still have to run stuff like (un)spreadLight
05:38 hmmmm        not fully hardware lighting
05:40 kahrl        is irrlicht 1.7* still supported by minetest? apparently the shadow volume stuff changed in 1.8
05:42 hmmmm        must be supported!
05:42 hmmmm        absolutely
05:43 hmmmm        until debian stable and freebsd update to irrlicht 1.8, we must support it
05:43 kahrl        guessed so, let me check what actually changed
05:46 kahrl        ok not actually that much
05:46 kahrl        this was fixed in 1.8 and seems relevant: http://sourceforge.net/p/irrlicht/bugs/379/
05:47 kahrl        and "addShadowVolumeSceneNode now replaces an existing shadow"
05:48 hmmmm        hmm
06:47 celeron55_   i think we could get a nice lighting effect if the shader-brightened side of nodes that have sunlight would be dynamically (and smoothly) varied according to sun position
06:47 celeron55_   it's also easy to do
06:48 celeron55_   it'll look odd if you have a straight cave down and sunlight in it, but generally it'd be not odd
06:50 celeron55_   (imo quite useless though, but people want all kinds of useless things)
06:50 RealBadAngel it will also imply another things can be done easily i think, like shadows, lens flare etc
06:51 celeron55_   eh no?
06:51 RealBadAngel why not?
06:51 celeron55_   what i said is just a single sin and cos added to a shader
06:51 celeron55_   that's what i mean by simple
06:52 RealBadAngel ah, i thought you were talkin introducing sun as a light source in scene
06:52 RealBadAngel hardware one
06:53 RealBadAngel seen hmmm talkin about it when i read logs
06:54 celeron55_   light sources aren't simple in our case if you ask me
07:05 RealBadAngel i dont know nothing about lights yet
07:06 RealBadAngel im learnin now how to make day/night transitions to sky
07:16 RealBadAngel http://www.youtube.com/watch?v=TLhO96jeMc4
07:16 RealBadAngel this is pixelstudio, written in c#, MIT licensed
07:17 RealBadAngel how do you like it?
08:48 celeron55_   i think people are too lazy to combine the headers and footers to the game images; i'll add that too when i add texture pack support back
08:49 celeron55_   (at least i am too lazy)
15:07 PilzAdam     anyone against better node placement prediction? https://github.com/minetest/minetest/pull/687
15:10 RealBadAngel ok with me
15:10 sfan5        seems fine to me
15:19 RealBadAngel does anybody know how to pass variables to shaders in minetest?
15:24 Exio         shaders to the pixel or the vertex shader?
15:24 Exio         RealBadAngel: ^
15:24 Exio         er, variables to the *
15:25 RealBadAngel i need to pass textures to shaders
15:25 RealBadAngel vertex shader
15:26 RealBadAngel i got them loaded but have no idea how to pass them to shader
15:26 celeron55_   you mean a few more global shader variables?
15:26 RealBadAngel no, not global
15:26 celeron55_   or something bound to vertices?
15:26 RealBadAngel my own shader variables
15:26 celeron55_   those are your options
15:27 celeron55_   (and the latter is quite limited with irrlicht)
15:27 RealBadAngel http://pastebin.com/wDzmaZnF
15:27 RealBadAngel this is the pixel shader
15:28 RealBadAngel it needs to read from 2 textures and get two variables
15:35 sfan5        RealBadAngel: https://github.com/minetest/minetest/blob/master/src/shader.cpp#L218 or https://github.com/minetest/minetest/blob/master/src/game.cpp#L762
15:37 celeron55_   RealBadAngel: those are global shader variables, see GameGlobalShaderConstantSetter in game.cpp
15:37 celeron55_   altough i don't know how binding textures works in irrlicht
15:41 celeron55_   hmm, i think you may need to use SMaterial::setTexture
15:41 celeron55_   not sure
15:41 celeron55_   and i don't know by what names those would be in the shader either
15:45 celeron55_   maybe that doesn't even work; i recomemend searching for code that does something similar
15:46 celeron55_   this seems to implement such with direct opengl calls: http://code.google.com/p/irrcg/source/browse/trunk/include/IrrCg.h?r=4 http://code.google.com/p/irrcg/source/browse/trunk/source/IrrCg.cpp?r=4
15:46 Exio         irrcg?
15:47 serengeor    celeron55_,  to say which textures for shader to use you need to do something like: gpuservices->setPixelShaderConstant("yourTextureSampler", (float*)&TexAddress, 1);
15:48 serengeor    1 is the texture layer in material
15:53 RealBadAngel invalid cast from type ‘irr::video::SMaterial’ to type ‘float*’
15:53 serengeor    oh wait
15:54 RealBadAngel uh, forgot &
15:54 serengeor    RealBadAngel, http://ideone.com/UKE1K5
15:54 RealBadAngel now it compiles
15:55 serengeor    1 is actually the count
15:55 serengeor    then again, with new irrlicht I think you can upload ints too
15:55 RealBadAngel it blinked with right textures i think (once)
15:56 RealBadAngel but then when in rendering still cannot see textures
15:56 serengeor    should use the callback class I think
15:57 serengeor    RealBadAngel, take a look at this thread: http://irrlicht.sourceforge.net/forum/viewtopic.php?f=1&t=43398
16:01 RealBadAngel http://pastebin.com/zxDxVnMp
16:01 RealBadAngel ive put that in callback
16:02 RealBadAngel last 4 lines
16:03 RealBadAngel and shader blink with proper data received then it doesnt see the textures as usual
16:07 serengeor    RealBadAngel, video::SMaterial texture=m_sky->getMaterial(1);
16:07 serengeor    you should pass only an int to which layer the texture is assigned
16:10 serengeor    int layer = 1; services->setPixelShaderConstant("color", (float*)&layer, 1);
16:10 RealBadAngel so i should keep the textures inside the same mesh
16:11 serengeor    inside same material
16:11 RealBadAngel shader is supposed to build texture out of given two and coords
16:11 RealBadAngel it is set as material layer 0
16:11 RealBadAngel of the mesh
16:13 serengeor    well if you use more than one material for the same mesh you need to draw it multiple times, don't you :?
16:13 RealBadAngel hmmm, its all weird, lets try sending just int there
16:15 serengeor    when material gets set, all textures set to that material get bound. In shader you still need to say which sampler should use which texture, so that's where that " services->setPixelShaderConstant("color", (float*)&layer, 1);" comes in
16:16 serengeor    if multiple materials are used, only textures set to current material are bound
16:17 RealBadAngel passing ints didnt changed anythin
16:18 RealBadAngel http://pastebin.com/iaGhJRuy
16:19 RealBadAngel here are my materials, mesh and shaders
16:19 RealBadAngel shader is hapilly working on some random data
16:20 serengeor    RealBadAngel, why exactly do you use 4 materials?
16:20 RealBadAngel those are color maps (at least 2 of them)
16:21 RealBadAngel i do not display them at all
16:21 RealBadAngel those are source data (colors array) or something like that
16:21 serengeor    RealBadAngel, but you only set textures to those?
16:21 RealBadAngel i need make shader see them
16:22 RealBadAngel shader itself creates displayed texture
16:22 serengeor    you should not use 4 materials if the only thing that changes are textures
16:22 serengeor    instead use one material and set textures to it using: http://irrlicht.sourceforge.net/docu/classirr_1_1video_1_1_s_material.html#a44dc38e7d4a78434d12ffd02e88e3d60
16:23 serengeor    only this way multiple textures can be used
16:23 serengeor    at once in shader
16:23 RealBadAngel you mean to index the materia?
16:23 RealBadAngel *l
16:25 celeron55    umm wat
16:26 celeron55    video::SMaterial texture=m_sky->getMaterial(1);
16:26 celeron55    you make a copy of the material, then give irrlicht pointers to it, then it goes out of scope and you have pointers to invalid memory 8)
16:26 celeron55    (but i'm not sure if that's right in any way in the first place)
16:27 RealBadAngel i havent invented it but using existing minetest code
16:28 BlockMen     celeron55, here the hill you suggested. are you ok with that -> https://dl.dropboxusercontent.com/s/pemo6gcq20cm081/hp_hill.png ?
16:28 RealBadAngel i cannot belive passing pointer or the variable could be so complicated :)
16:29 Exio         you are passing it to the GPU
16:29 RealBadAngel to damn shader
16:29 PilzAdam     BlockMen, that doesnt look god
16:29 Exio         nobody said that was easy
16:29 PilzAdam     *good
16:30 BlockMen     PilzAdam, why?
16:30 PilzAdam     the sky and the perspective are wrong
16:30 serengeor    RealBadAngel, http://ideone.com/kDZLxs
16:30 RealBadAngel i just did the same
16:30 serengeor    this is how your material set-up should look like more or less
16:31 BlockMen     actually the sky and the hill are from one shot, so it should be correct
16:31 BlockMen     but the main grass and the hill grass not (i see now)
16:31 serengeor    RealBadAngel, so does it work now?
16:32 RealBadAngel same
16:32 RealBadAngel how the shader would know to use mat with sky_node mesh?
16:35 RealBadAngel http://pastebin.com/3njixfyq
16:35 RealBadAngel imho mat and mesh are not connected at all
16:39 celeron55    BlockMen: why can't you just leave it as is? :P i don't want to judge these
16:39 celeron55    but the perspective changes do make it look bad
16:39 celeron55    anyhow, may i request you to... just not do anything?
16:40 BlockMen     because its too blurry atm. and to the small changes you said: "add a hill or something"
16:40 celeron55    it wasn't a request, i just wanted to point out that if you insist on doing something, small color changes aren't useful
16:42 serengeor    RealBadAngel, materials are part of node and they get set when node is rendered
16:42 celeron55    *irrlicht node
16:42 celeron55    (/don't make anyone confuse those with minetest's voxel nodes)
16:42 celeron55    -/
16:43 serengeor    yeah, don't really know much about minetest code
16:44 BlockMen     celeron55, in my opinion it should just be a little bit more polished (like that https://dl.dropboxusercontent.com/s/4gznmflv5rdna1d/more_fancy_hp.png). because it looks kinda amateurish (your words :P) and i just want make it a little bit less amateurish.
16:44 BlockMen     an that can only be small changes...
16:45 celeron55    oh by the way, i noticed at some point that the minetest icon isn't really anywhere on the site
16:45 celeron55    i wonder if that is a problem or not
16:46 BlockMen     i would like to have that downloadbox (with the icon) like on my first concept
16:46 BlockMen     *on mian page
16:47 BlockMen     -> https://dl.dropboxusercontent.com/s/z2xndupy27oqj3q/concept_hp_2-3.png
16:50 celeron55    i think all of those are worse than the current
16:51 celeron55    and refuse to continue telling whether things suck or not because i have other things to do
16:52 BlockMen     fine
16:55 kahrl        the only thing I would change about the main page: when you click on the screenshots, display an enlarged version of the screenshot you clicked on (with a link to the screenshots thread)
16:55 kahrl        instead of going directly to the screenshot thread
16:56 Calinou      the "bunch of screenshots" thing is nice but very "inequal"
16:56 Calinou      I'd rather put screenshots in the same way, but with the same video settings for all of them
16:56 Calinou      gameplay + scenery
16:56 Calinou      and a few with mods
16:58 BlockMen     in my opinion the whole screenshot collection (1st pic) looks not good, but w/e
17:48 sfan5        the main menu buttons don't look quite right for me
17:54 Calinou      they look meh, on top of that, why would you want to play minimal :P
17:55 RealBadAngel http://i.imgur.com/xngqHYo.png
17:55 RealBadAngel i start to actually see anything :)
17:55 sfan5        RealBadAngel: is that shader?
17:56 RealBadAngel yup
17:56 sfan5        \o/
17:56 VanessaE     looking good
18:09 celeron55    well you're free to improve the screenshot collection; or maybe replace it with a video or something
18:10 celeron55    the main point of it is to be representative of everything that you can do with the game
18:10 celeron55    or, in more proper words, with minetest
18:20 serengeor    RealBadAngel,  grats
18:20 RealBadAngel thx for help serengeor
18:26 serengeor    np, haven't used irrlicht myself for quite a while now
18:55 celeron55    doing the Something Different (tm) on minetest is somewhat more laborous than i thought; i'm switching to low gear on this
18:55 celeron55    however, it seems to me that voxel area entities would be a real game maker
18:56 Jordach      celeron55, yup
18:56 celeron55    this has been discussed on 4chan before; it's nothing new
18:56 celeron55    but anyhow
18:56 Jordach      and its something we have that mc dont have yet
18:56 Jordach      (this it futuristically speaking)
18:56 celeron55    well, there are reasons why others don't have it
18:56 celeron55    it is that they aren't particularly easy to make
18:57 celeron55    but anyhow, to someone who can do it and has time for doing it: everyone would love you if you did it
18:57 Jordach      just give it a 16^3 chunk to play with
18:59 celeron55    i'd kind of like to do it myself, but i'm still almost continuously in this half-burnout state regarding to minetest and it'd just break me; i'll probably try to be gone for a while now again
19:01 celeron55    but really, if one thinks of voxel area entities in terms of this rather than minetest_game, it makes it kind of totally desirable even in a very crude form: http://c55.me/random/2013-05/screenshot_1726656136.png
19:03 celeron55    hmm...
19:03 * celeron55  feels the spirit of 0.4 pushing him from the back
19:52 celeron55    hmm, yeah, i won't be continuing right now
19:53 celeron55    i seriously have way too many game projects; now i just created like many more 8D
19:53 celeron55    i need to learn finshing faster
19:53 celeron55    +i
19:53 Jordach      terandim is waiting
19:54 celeron55    yes and an another one in addition to that, and then there are many smaller projects
19:55 celeron55    one thing i am wondering now is whether it's too much trouble for the distributions and people that currently like 0.3 that i fork it under a different name
19:55 celeron55    i really want to do certain things that do not suit as a direct continuum for the stable 0.3 series
19:56 celeron55    first of all, i want to remove all code that is restricted to LGPL, because this just allows it in a barely viable way
19:57 celeron55    i rebased the whole thing the one evening and i can mass-rebase authors out of it, and it seems to work surprisingly well
19:57 celeron55    (git is magic; it's magic and there is no doubt of it)
19:58 sfan5        "git: 'magic' is not a git command. See 'git --help'."
19:58 celeron55    it's inherent to all commands!
19:58 sfan5        well, that makes sense
19:59 Jordach      celeron55, My Little Computer: Git is Magic
20:00 celeron55    anyhow, i have way too many things to focus on and i am not getting anything done by trying to focus on everything; however, it's incredibly hard to choose which one to develop
20:01 Jordach      celeron55, russian roulette with the del key
20:01 celeron55    in addition to that, everything i do is pretty much designed to last in time very well, so there's not even any kind of "oh no this is getting old" factor
20:02 celeron55    in addition to that, i would very much like to develop something in finnish for a finnish audience, but how the hell am i going to find the time for that too?!
20:03 celeron55    i need to permanently drop my day job and start working on the thing that is most likely to bring in some money so that i can continue with the others too
20:03 celeron55    ...
20:03 sfan5        seems like you don't have enough time
20:04 Jordach      celeron55, procrastination FTW
20:05 celeron55    this mandatory army substitute slavery thing has consumed way too much time; i'm glad it ends in... oh, it ends in 7 weeks
20:06 celeron55    that brings me to the sad subject that i should go to a dentist before this ends because this makes it free, but who wants that even for free...
20:07 celeron55    uhm... this went terribly offtopic
20:07 Jordach      Chit-chat goes to #minetest.
20:08 Jordach      celeron55, free dental ftw
20:08 * Jordach    understands your pain
20:12 celeron55    (the good thing is, these are the worst problems i have)
20:16 Jordach      celeron55, i have this bitch of a wisdom tooth
20:19 celeron55    (i don't have much problems with my teeth, altough they are likely to recommend pulling out more than the one removed widsom tooth just because... they like to remove teeth i guess)
20:20 Jordach      i've always thought of making a tiny piano from the teeth ive lost
20:27 celeron55    Jordach: might you be interested if i create a channel just for all of my other projects? :P
20:27 Jordach      sure.
20:27 Jordach      i still play JSBombers
20:28 celeron55    does it still work? where is it even...
20:28 Jordach      celeron55, last time i played, it did
20:31 celeron55    it has always needed the node.js server to be running and i don't think i have started it up since a reboot sometime 2011 or 2012
20:32 sfan5        http://jsbombers.mine.bz/ -> "It works! foo bar ..."
20:32 * sfan5      would really like to try out jsbombers
20:36 celeron55    started it :P
20:36 celeron55    hmm, it doesn't seem to work
20:36 sfan5        celeron55: XMLHttpRequest cannot load http://jsbombers.mine.bz:8081/socket.io/1/?t=1367613405850. Origin http://jsbombers.mine.bz is not allowed by Access-Control-Allow-Origin.
20:37 celeron55    let's move this on to #8dromeda
20:54 sfan5        anyone against me commiting what is described here: http://forum.minetest.net/viewtopic.php?pid=88123#p88123
20:54 sfan5        (fixes the favorite server list on win)
21:00 sapier       where's the commit?
21:01 kahrl        sfan5: should I do the same with the code that reads modpack.txt/depends.txt/optdepends.txt?
21:01 sfan5        does that code also use std::ios::binary?
21:01 kahrl        yeah
21:02 sfan5        do that
21:02 kahrl        ok
21:02 kahrl        also how do I squash commits on an existing pull request?
21:02 sapier       git reset HEAD~(number of commits)
21:02 sapier       git add (every file)
21:02 sapier       git commit
21:03 sapier       git push origin --force
21:03 sfan5        git rebase -i HEAD~2; change "pick" line of the commit that fixed something to "fixup"; git push -f
21:03 kahrl        hmm
21:03 sfan5        ^ thats how I do it
21:03 kahrl        okay I wasn't sure if I should force push something that others have already pulled
21:03 sapier       hmm sounds even more easy than the way I do it :)
21:04 kahrl        s/have/might have/
21:05 sapier       https://github.com/minetest/minetest/pull/635 is still to be merged to fix memory leaks
21:05 sfan5        random thing: core devs should have a special title on the forums
21:05 sapier       I know I'm telling this almost every day
21:06 sapier       why?
21:08 sapier       a title doesn't make a better coder or man ;-)
21:10 sfan5        to be regognised easier
21:11 sapier       and why does a "core dev" have to be recognized on forum?
21:11 sfan5        umm...
21:12 sapier       :-) I see .. .seams to me a mertitocracy thing
21:13 sapier       btw is there a schedule for 0.4.7?
21:14 Calinou      1) wait for 0gb.us to sleep, 2) make lava even less renewable, 3) release 0.4.7
21:14 Calinou      <c-fifty-five> and there it is <c-fifty-five> our little release process
21:14 Calinou      :-)
21:15 sapier       :-( not quite convenient to make mods for
21:16 hmmmm        what?? who cares if someone's a core dev or not
21:16 sapier       I've tried to use some mods last days it's been very anoying to find out which one work with what version. As I do mobf myself I know it's not only modders to blame as it's very difficult to do mods for a specific version
21:17 hmmmm        if a core dev is really worthy of respect on the forum, they'd show it in the quality of their responses and what not
21:17 hmmmm        there's no schedule for 0.4.7
21:17 hmmmm        but this time around, the feature freeze is going to be for an ENTIRE WEEK
21:18 hmmmm        i'll merge that itemdef thing later, gotta pass out for a couple hours
21:18 sapier       I know this discussion has been here for some time maybe a schedule would make life easier for modders
21:18 hmmmm        RBA, that screenshot looks pretty awesome
21:19 * sfan5      waits for opinons whether "Optional dependencies" #697 should  be merged
21:20 hmmmm        we were discussing how to pass variables to shaders in like a week ago, you might want to look up that conversation
21:21 hmmmm        i am going to need to pass along the color to mix with the node as well
21:23 hmmmm        why does straybytes keep CTCPing us
21:23 sfan5        technically he CTCP PINGED #minetest-dev
21:25 kahrl        sfan5: I still have to remove std::ios::binary, make it ignore multiple symbols at the end of lines in (opt)depends.txt, and squash the commits
21:25 sapier       did you discuss irrlicht shader resource problem too?
21:27 sapier       irrlicht doesn't support unloading of shaders
21:31 PilzAdam     <sfan5> anyone against me commiting what is described here: http://forum.minetest.net/viewtopic.php?pid=88123#p88123 <- definetly needed; you need to test this first, though
22:01 kahrl        sfan5, others: update https://github.com/minetest/minetest/pull/697
22:01 kahrl        updated*
22:04 * PilzAdam   tests
22:05 PilzAdam     ummm... I will not test any cases of this patch
22:06 PilzAdam     oh, nice, it fixes the dependencies to game mods
22:10 PilzAdam     it seems to work fine
22:13 PilzAdam     kahrl, has another core dev already agreed to this? if yes then I can merge it now
22:13 kahrl        I think sfan5 did
22:14 kahrl        not to the newest version yet though
22:14 PilzAdam     yea, but you have only fixed the code
22:14 PilzAdam     I guess that counts ;-)
22:30 PilzAdam     also tested this: https://github.com/minetest/minetest/pull/686
22:31 PilzAdam     I have set some mapblocks full of leaves with worldedit++ and it works fine
22:31 PilzAdam     (althought the saplings from the decaying leaves crashed my server...)
22:31 PilzAdam     -t
22:32 kahrl        heh
22:35 Exio         lol
22:46 PilzAdam     these 2 seem to be fine too: https://github.com/minetest/minetest/pull/674 https://github.com/minetest/minetest/pull/627
22:54 PilzAdam     ok, tested the first one: I spawned 1326 sand nodes in the air dug one of them, and all fall down correctly
22:55 kahrl        nice
22:55 PilzAdam     umm.. actually that works also without the patch
22:56 * PilzAdam   needs more sand
22:56 Exio         lol
22:56 sapier       there's a simple test for first one
22:56 PilzAdam     I already tried flying away and come back
22:56 PilzAdam     that one works
22:56 sapier       place an entity e.g. a cart somewhere and walk away until it's unloaded
22:57 sapier       then walk back to where the cart should be
22:57 sapier       ok
23:00 PilzAdam     4725 sand nodes doesnt work in current master but it works fine with your patch
23:01 sapier       wow :-) didn't assume this patch helps in increasing numbers
23:01 PilzAdam     (except the FPS drop to 5, but thats unrelated)
23:01 sapier       lol
23:01 kahrl        huh, in clientserver.h the documentation for protocol version 20 has wrong message names
23:01 kahrl        I guess it wasn't updated when they were changed
23:04 PilzAdam     fixed
23:06 Exio         as you are trying pull requests PilzAdam, would you look at https://github.com/minetest/minetest/pull/439 ?
23:07 PilzAdam     we already have merged number 2 of this
23:07 PilzAdam     this needs a rebase
23:09 PilzAdam     number 1 is non trivial to merge since I have rewritten the new_style_water; and IIRC there is an easier way for 3
23:09 PilzAdam     (enable a commented out line again IIRC)
23:10 Exio         ah
23:10 PilzAdam     this also only effects new_style_water
23:10 PilzAdam     oh, and flowing liquids
23:12 PilzAdam     hm, actually only flowing liquids are effected by number 1; so forget what I said about it
23:12 PilzAdam     it still needs a proper rebase
23:13 PilzAdam     it doesnt have a high priority though, since thats are only some minor visual changes
23:14 PilzAdam     -s
23:16 Exio         - what about closing https://github.com/minetest/minetest/pull/216 ?
23:18 PilzAdam     done
23:19 PilzAdam     Exio, anything else :-)
23:19 Exio         hehe
23:20 PilzAdam     I guess Im going to sleep now
23:20 PilzAdam     bye