Minetest logo

IRC log for #minetest-dev, 2016-04-10

| Channels | #minetest-dev index | Today | | Google Search | Plaintext

All times shown according to UTC.

Time Nick Message
00:54 est31 hmmmm, any other problems with the changes?
00:55 hmmmm nothing major, just extremely out of conformance with minetest code style and conventions
00:55 hmmmm but i guess that doesn't really matter for srp
00:55 est31 otherwise I'll merge an even newer version, that uses clang-format
00:55 est31 its closer to minetest code style than before
00:56 est31 e.g. it enforces no space after ( and before )
00:56 est31 and space before *
00:56 hmmmm what is your own personal preference?  j/w
00:56 est31 I've made the guidelines after my personal preference
00:56 est31 https://github.com/est31/csrp-​gmp/blob/master/.clang-format
00:57 est31 so my preferences are aligned by minetest style guides i admit
00:57 est31 note that the last two commits are WIP
00:58 hmmmm is there a "preset" for Linux style?
00:58 est31 usually I dont do push -f, especially for old changes (> 1 day)
00:58 est31 yes
00:58 hmmmm sorta wonder what that does
00:59 est31 http://clang.llvm.org/docs/C​langFormatStyleOptions.html
00:59 est31 copies linux style guide
00:59 hmmmm i mean for CPP
00:59 est31 esp. http://clang.llvm.org/docs/Clang​FormatStyleOptions.html#examples
00:59 hmmmm the minetest code style tries to be linux mostly except expands on c++ specific items
01:00 hmmmm i'm kind of interested in this because it could be integrated with the unit tests and give a warning on miscompliance or something
01:00 hmmmm better to have a robot do dumb things like style checking
01:00 hmmmm so nobody has to say, "oh you missed a space"
01:01 est31 yeah
01:01 hmmmm so what was up with that fit you had a couple days ago
01:01 est31 fit?
01:01 hmmmm from what i was reading in the logs and on various github comments you seemed very angry
01:02 est31 well idk, you turn up, thats cool, then you speak with a contributor about their pr, then you merge it without asking anybody else
01:02 est31 usually you just added a :+1:
01:03 hmmmm because it was small enough to only need 1 approval i thought
01:03 hmmmm certainly wasn't a medium sized PR
01:04 est31 well it had a pretty involved API change in it
01:04 hmmmm when i *just* add a :+1: that usually means i'm lazy and don't feel like going through all the process of downloading the patch and cherry picking it and so on
01:04 hmmmm involved as in, confusing?  sure i'll agree it's confusing
01:04 est31 yes
01:05 hmmmm but paramat has done worse and i was the only one to code review it
01:05 hmmmm nobody seems to have a problem there
01:05 est31 yeah i understand that you then are eager to merge, i have this sometimes too
01:05 hmmmm and it certainly didn't break any reverse compatibility
01:06 est31 usually paramat waits though before its clear that nobody reviews his code
01:06 est31 e.g. mapgen change
01:07 hmmmm no, I'm saying he's modified the API, I'm the only one who reviews it, and there's nobody being upset there
01:07 hmmmm with that situation
01:07 hmmmm i'm sorry it's just that i had no idea that an api change automatically makes it a larger change that requires more reviewers
01:07 hmmmm not quite a change as much as an addition
01:08 hmmmm the word 'change' implies that it would disrupt others from doing their modding
01:09 est31 the main thing that i didnt like was that you didnt respond to my repeated questions
01:09 hmmmm you didn't ask me any questions before i merged it
01:09 est31 completely acted as if i wasnt there
01:09 est31 no no after you merged it
01:10 hmmmm you never asked me anything directly
01:10 hmmmm this is the first time in over 4 days i've been on IRC at the same time you are
01:10 est31 true
01:10 hmmmm soo...
01:10 est31 still I've tried to reach you over various channels
01:10 hmmmm i did act like you weren't there because you weren't there
01:10 est31 either way, I'm okay now with keeping the change in
01:11 hmmmm 'various channels'
01:11 est31 the main critic point I had (bad docs) was fixed
01:11 hmmmm I caught on to what was happening when I reviewed the logs for a completely different reason
01:12 est31 I also asked on github, although I could have remembered that you dont read github notifications
01:12 hmmmm ... no i do not
01:12 hmmmm github notifications are dumb
01:12 hmmmm they alert on literally every single thing anybody does
01:16 jin_xi joined #minetest-dev
01:16 est31 we need a policy for how to deal with conflicts like these: where one dev merged a change and some other thinks it needs more approval
01:16 hmmmm celeron would probably think it's a 'waste of time'
01:16 hmmmm ;;
01:17 paramat anyone agree this can be closed? #2747
01:17 ShadowBot https://github.com/minetest/minetest/issues/2747 -- Random rotation of textures
01:17 est31 I'm only wondering about the disagreement case, not for how to interpret the rules
01:18 hmmmm the amount of approval required for something to be merged should always be the larger of the two in the case of a disagreement
01:18 est31 https://llvm.org/bugs/show_bug.cgi?id=17362
01:18 est31 :(
01:18 hmmmm so if i say something requires 1 approval and you say 2, we go with 2
01:18 hmmmm more careful is always better
01:19 est31 makes sense
01:19 hmmmm if it's already been merged and another developer objects after the fact, then there's not much you can really do aside from raise your objections, have the original developer fix them, fix them yourself, or if it's too egregious of a problem, revert
01:20 hmmmm any of those three actions require approval though, equal to the amount of approval the original change required
01:21 hmmmm (wait, is the last sentence a good policy?)
01:21 hmmmm it should take less approval to undo a change because it can always be re-done
01:21 hmmmm if there is further disagreement
01:22 hmmmm but we don't want a smaller group of developers engaging in a standoff by reverting everybody's commits
01:22 hmmmm in any case i doubt it would matter since common sense ultimately overrules all
01:22 est31 yes, reverting is no fun
01:23 hmmmm these are just rules, not hard rules, but rules nevertheless, for maintaining some semblance of balance and order
01:23 hmmmm and so others have an idea of what to expect under normal circumstances
01:23 hmmmm without rules everybody would get upset due to mismatched expectations
01:25 hmmmm paramat:  yes, that could definitely  closed
01:25 hmmmm be closed*
01:25 paramat ok
01:25 paramat will do so
01:26 hmmmm it's just that closing issues that are valid but hard to fix is really bad
01:26 hmmmm if the main cause is a duplicate of another issue, close it for the reason of being a duplicate and link to the original
01:27 hmmmm even if something is hard to fix doesn't mean we won't fix it, and it's always good to keep it in our issue tracker for visibility
01:27 est31 <hmmmm> it's just that closing issues that are valid but hard to fix is really bad + 1000000 I hope paramat now changes his behaviour
01:27 paramat ok
01:27 hmmmm the PR might get solved later on if we make a big change like switching graphics engines
01:28 hmmmm also that looks reflects bad on us imho because you're basically telling outsiders that if something is too hard you're not going to consider it
01:28 paramat i welcome any checks on my closes, i may have made mistakes :)
01:29 paramat i'm still looking for the duplicate, if i can't find one i'll re-open it
01:31 hmmmm i think you might be able to get a chronologically ordered list of events a user performed on github
01:31 hmmmm i can do that for commits but iunno how to for everything else like commenting or closing issues
01:33 paramat if the duplicate exists it's not well labelled because searching by label didn't help, it's just a case of going through all 600
01:33 paramat thoughts on old issue #1033 ?
01:33 ShadowBot https://github.com/minetest/minetest/issues/1033 -- Mapgen selection needs info text
01:34 hmmmm mixed thoughts... on one hand i think the vague text and no description encourages users to explore each and form their own opinions
01:34 hmmmm on the other hand i can definitely see how a friendlier ui would have descriptions for each
01:34 hmmmm but actual descriptions that are multiple sentences, not (something in parentheses) like the original author of that issue wrote
01:35 paramat i just think extra text should be elsewhere not in the dropdown
01:35 hmmmm of course
01:35 hmmmm it should be in a text pane to the left of the dropdown
01:35 hmmmm err rather a static text label
01:36 paramat well to avoid annoying most users i think don't display any extra info by default
01:37 paramat i also disagree with marking out 'the default' in case it discourages use of the others
01:38 hmmmm i don't think descriptions would annoy the user if the UI is designed correctly
01:47 hmmmm joined #minetest-dev
01:47 paramat re-opened until i find the duplicate (if it exists)
01:48 paramat is it the case that translucency z-sorting would still not work within a mapblock if each mapblock is a made scenenode? i may have misunderstood
02:09 paramat nevermind :) it's not important
02:14 est31 hmmmm, https://github.com/est31/minetest/commit/​2ced546eb1208ac3f47b91afdfbbc1a225520fd9
02:14 est31 if you want to review, you can look at the library commits
02:14 est31 and diff before, and diff after the commits
02:15 est31 and if you want to review the clang-format commit, simply apply clang-format yourself if you dont trust me :)
02:20 hmmmm [09:49 PM] <paramat> is it the case that translucency z-sorting would still not work within a mapblock if each mapblock is a made scenenode? i may have misunderstood
02:20 hmmmm that is correct
02:21 hmmmm the way to get transparency z-sorting to work correctly (under irrlicht, anyway) is to keep a separate scenenode for each translucent material
02:22 paramat ah i see
02:22 hmmmm switching to one-scenenode-per-mapblock won't do anything to help, but it would allow us to use irrlicht occlusion culling
02:22 hmmmm or at least test it
02:23 est31 and celeron55 would cheer because yet another thing to rebase against
02:23 hmmmm this is why you keep your commits small and focused people
02:24 est31 you want me to split up https://github.com/est31/minetest/commit/​2ced546eb1208ac3f47b91afdfbbc1a225520fd9 ?
02:24 est31 i can split it up into its original 6 commits
02:25 est31 err 7
02:25 hmmmm no
02:25 hmmmm there's no advantage to splitting that change up
02:25 hmmmm something like celeron's 55 file change is a nice candidate for splitting
02:26 hmmmm in any case, I hope you know that putting the result of an if statement body on the same line as the if statement is a huge violation of the minetest code style
02:26 hmmmm just FWIW
02:27 est31 ok, wont do it in minetest code
02:27 hmmmm yeah you said you were looking to change it closer to minetest style...
02:27 hmmmm just an FYI
02:28 est31 ok
02:29 est31 I didn't mean that the goal was to mimic minetest style as much as possible, the goal was rather to mimic my preferences
02:29 est31 I edited it until i liked the style
02:29 betterthanyou710 hi
02:29 hmmmm est31, what do you think about this https://github.com/minetest/minetest/pull/3929
02:30 est31 I hope you dont think I'm a horrible person now just because I prefer if on the same line :)
02:30 betterthanyou710 est31 your a core developer right?
02:30 hmmmm i do because you might be hiding something
02:30 betterthanyou710 lol just like batman
02:30 betterthanyou710 dun da da dunnnnnn
02:31 hmmmm if (memcmp(password_hash, stored_hash, 20)) goto success;
02:31 est31 betterthanyou710, yes I am
02:31 hmmmm goto failure;
02:31 est31 hehe
02:31 est31 clang-format would ruthlessly format that however
02:31 hmmmm "WHOOPS" that was a mistake
02:31 betterthanyou710 est31 me and a friend were talking and you haave to be invited to become a core developer right?
02:31 hmmmm RealBadAngel actually pulled something similar to that
02:31 est31 it would make the goto failure on the same line
02:31 betterthanyou710 we made a bet on it xD
02:32 hmmmm if (g_settings->getBool("some_shader_feature_​we_agreed_to_not_be_enabled_by_default"); {
02:32 hmmmm ... do stuff here ...
02:32 hmmmm }
02:32 est31 thats mean
02:32 hmmmm he played ignorance
02:35 est31 betterthanyou710, the rules are a bit more complicated, but yes, in the end you get invited on github, by c55: http://dev.minetest.net/Organisation#Assignment
02:35 betterthanyou710 cool
02:35 betterthanyou710 i also wanted to ask will you guys be doing more texture mods and adding shaders?
02:36 betterthanyou710 cuz one of the only things that minecraft has over minetest is textures
02:36 est31 hrmmmm I discovered a style issue
02:36 est31 https://github.com/est31/minetest/commit/2​ced546eb1208ac3f47b91afdfbbc1a225520fd9#di​ff-2f6369328116e9b654242a7597658a48R1011
02:37 est31 especially the line below
02:37 est31 https://github.com/est31/minetest/commit/2​ced546eb1208ac3f47b91afdfbbc1a225520fd9#di​ff-2f6369328116e9b654242a7597658a48R1012
02:37 est31 I think I will have to tweak the script....
02:37 est31 but before that, I go to bed
02:38 betterthanyou710 ha same
02:39 betterthanyou710 if u have a answer plz msg it to me
02:39 hmmmm your question is off-topic for this channel
02:40 paramat bbl after omnoms game#1022
02:40 ShadowBot https://github.com/minetest​/minetest_game/issues/1022 -- Default/mapgen: Add upper limits to forest biomes and trees by paramat
02:40 paramat left #minetest-dev
02:44 betterthanyou710 sorry hmmmm ill ask on another channel
02:44 kaeza while we're on the topic, VoxelArea is a nice candidate for style cleanup
02:45 hmmmm yeah, it puts a lot of functions way larger than 3 lines into a header file, doesn't it?
02:45 hmmmm speaking of which, util/string.h has a number of offenders there.
02:45 kaeza no, I mean the naming convention
02:45 kaeza all of Minetest is foo.do_something_useful(p1, p2, whatever)
02:46 kaeza VoxelArea is VoxelArea:new{MinEdge=what,MaxEdge=what}
02:46 hmmmm oh you're talking about the lua voxelarea
02:46 kaeza yea
02:46 hmmmm i'd rather not touch it
02:46 kaeza nowhere else is CamelCase used for field names
02:47 kaeza or methods for that matter
02:47 hmmmm you're going to break compatibility and it's just a mirror of the VoxelArea inside the engine
02:47 hmmmm and that's PascalCase, camelCase isLikeThis
02:47 kaeza it shouldn't break compatibility if done right
02:48 hmmmm idk how to do it
02:48 kaeza VoxelArea should really be a function (like ItemStack)
02:48 hmmmm if you can, no problem
02:50 kaeza we should take some time to standardize on the naming conventions for Lua-visible symbols
02:50 hmmmm perhaps
02:50 hmmmm before we just jump in and make everything lower_case_underscore_style i think a vote should be taken to see what people find the most visually pleasing to work with
02:50 kaeza other stuff that really gets on my nerve is that we have register_on_player_whatever, but get_timeofday, or even getvelocity
02:51 kaeza I prefer "PascalCase" for "class" type things and words_with_underscores for fields and methods
02:52 kaeza "class" as in, ItemStack, VoxelArea, perhaps Vector?
02:52 hmmmm sure
02:52 hmmmm make a vote thread about it on the forum even
02:52 hmmmm you want to or should i?
02:53 kaeza my english sucks. could you? :)
02:53 hmmmm that sounds like an excuse
02:54 hmmmm i don't think anybody would be able to tell you were a non-native speaker
02:54 hmmmm from text anyway
02:54 kaeza okay okay, gimme a few mins to come up with something
02:54 hmmmm well ok
03:14 OldCoder kaeza, I will edit, if you like
03:14 OldCoder But I'm sure it will be fine
03:14 OldCoder TBH I had no idea originally of your native language
03:29 sofar :386
03:29 sofar wheps :D
03:38 kaeza hmmmm, PTAL https://gist.github.com/kaeza/0​72684493564eaabfe0d8f42f965e806
03:38 kaeza Cc OldCoder ^
03:39 hmmmm as an aside, I like the ItemStack() notation
03:39 hmmmm it's like a constructor
03:40 hmmmm yeah i agree with your proposal
03:41 hmmmm post looks good
03:48 kaeza let's see how long before it gets Godwin'd #3980
03:48 ShadowBot https://github.com/minetest/minetest/issues/3980 -- Standardizing API naming and calling conventions.
04:08 Zeno` joined #minetest-dev
04:21 Zeno` brb going to laptop
04:22 paramat joined #minetest-dev
04:40 Zeno` joined #minetest-dev
04:43 Zeno` everything lower_case_underscore_style <--- very annoying to type
04:43 Zeno` i.e. I think get_timeofday is better than get_time_of_day
04:44 Zeno` But I don't really want to get involved in this discussion again hehe
05:11 paramat left #minetest-dev
05:50 fling joined #minetest-dev
05:55 Dragonop_ joined #minetest-dev
06:02 asl97 joined #minetest-dev
06:47 Hunterz joined #minetest-dev
07:20 hmmmm it's consistent with everything else though
07:21 hmmmm i will admit it is pretty annoying to type
07:37 Krock joined #minetest-dev
08:32 nrzkt joined #minetest-dev
08:53 Obani joined #minetest-dev
09:16 Zeno` yes, for API function names it certainly needs to be consistent
09:16 Obani would be less confusing indeed
09:22 Calinou joined #minetest-dev
10:16 jin_xi joined #minetest-dev
10:45 proller joined #minetest-dev
10:56 Fixer joined #minetest-dev
10:57 PilzAdam joined #minetest-dev
12:23 est31 joined #minetest-dev
12:23 lisac joined #minetest-dev
12:49 est31 https://github.com/est31/minetest/commit/​ebdffeb2ab5cd36406308eae72a923df1275ad4b
12:49 est31 okay last chance to say something, i will push in ten minutes
12:50 sfan5 all those formatting changes seem useless
12:51 sfan5 and you're still checking fclose and CryptReleaseContext
12:51 sfan5 fine otherwise
12:56 est31 better one check too much than one too few
13:18 iangp joined #minetest-dev
13:25 Obani joined #minetest-dev
13:59 kaadmy joined #minetest-dev
14:12 rubenwardy joined #minetest-dev
14:44 Obani joined #minetest-dev
14:45 nrzkt joined #minetest-dev
15:03 Krock joined #minetest-dev
15:34 rand_user joined #minetest-dev
15:50 hmmmm joined #minetest-dev
16:20 STHGOM joined #minetest-dev
16:54 jin_xi joined #minetest-dev
17:13 proller joined #minetest-dev
17:18 rubenwardy joined #minetest-dev
18:27 Foghrye4 joined #minetest-dev
18:36 nrzkt joined #minetest-dev
18:41 Amaz joined #minetest-dev
18:45 damiel joined #minetest-dev
18:56 Foghrye4 I'm still waiting for my pull request to be accepted. Or, at least, rejected. So, here is a video: https://youtu.be/vWE3fjOgwIc
19:01 sfan5 give it a little title
19:01 sfan5 time*
19:02 hmmmm only 6 days?
19:03 sofar https://github.com/minetest/minetest/pull​s?q=is%3Apr+is%3Aopen+sort%3Acreated-asc
19:03 sofar Jul 9, 2013
19:03 sofar oldest pending PR
19:03 Foghrye4 Why so slow?
19:05 hmmmm because minetest isn't a full time job for anybody here
19:07 Foghrye4 So, new features will be implemented only in 0.4.14?
19:07 sofar after
19:07 Foghrye4 Even after!?
19:08 sofar yes, 0.4.14 is coming out in a short bit, so after that features can be added to master
19:09 hmmmm this is more of a #minetest topic
19:25 bugzapper joined #minetest-dev
19:40 paramat joined #minetest-dev
19:42 paramat please could devs add their opinions to #3774 ? perhaps it would be good to bring this to a resolution
19:42 ShadowBot https://github.com/minetest/minetest/issues/3774 -- Halo: highlight selected face by RealBadAngel
19:47 hmmmm it honestly seems like a bit too much
19:56 nrzkt joined #minetest-dev
20:05 nore hmmmm: yes, but if it is configurable...
20:13 paramat here are my 2 mapgen PRs needed for release #3973 #3974
20:13 ShadowBot https://github.com/minetest/minetest/issues/3973 -- Mgvalleys: Don't let cavegen place biome nodes everywhere by paramat
20:13 ShadowBot https://github.com/minetest/minetest/issues/3974 -- Mgv5: Optimise tunnels, add biome material in entrances by paramat
20:17 iangp joined #minetest-dev
20:22 paramat nore any opinion on game#543 ?
20:22 ShadowBot https://github.com/minetes​t/minetest_game/issues/543 -- Strengthen obsidian glass by tenplus1
20:29 paramat i might merge #2094 later
20:29 ShadowBot https://github.com/minetest/minetest/issues/2094 -- Fix offset being ignored by inventory bar HUD by rubenwardy
20:31 paramat maybe also #3985 since it's trivial
20:31 ShadowBot https://github.com/minetest/minetest/issues/3985 -- Minimap: revert change from RGBA to Indexed by sofar
20:35 PilzAdam paramat, you can merge #3985
20:35 ShadowBot https://github.com/minetest/minetest/issues/3985 -- Minimap: revert change from RGBA to Indexed by sofar
20:35 paramat ok
20:40 paramat closed #3429
20:40 ShadowBot https://github.com/minetest/minetest/issues/3429 -- Configurable damage screen intensity
21:23 DFeniks joined #minetest-dev
21:49 Player_2 joined #minetest-dev
22:08 turtleman joined #minetest-dev
22:09 VanessaE hmmmm: as I recall, one of RBA
22:09 VanessaE ...RBA's goals was more eye candy
22:09 VanessaE done in such a way as to be customizable.
22:09 hmmmm don't you love how the ' is right next to the enter key
22:09 VanessaE so I don't see any harm in adding that feature.
22:09 VanessaE no, I hate it :P
22:10 hmmmm i had a keyboard where there was an indentation between the enter key and the ' which prevented such errors
22:10 hmmmm was my favorite
22:10 VanessaE whatever moron it was who decided on that ought to have been shot
22:10 celeron55 best keyboard design features: ' right next to enter; backspace right next to enter
22:10 VanessaE well I have a 3d printer, I suppose I could create a replacement Enter key with a similar indentation
22:10 hmmmm ya in any case
22:11 hmmmm all this stuff should be configurable but look at how many entries there are in minetest.conf.example
22:11 hmmmm just too many
22:11 hmmmm but it's actually quite typical for games to have a lot of configurable options - the nitty gritty options are more hidden though
22:11 VanessaE celeron55: I had one once that was the MOST irritating design ever - a second backspace key occupying half the footprint normally reserved for the spacebar
22:12 hmmmm like you know in the console how they have sl_some_crazy_option = 5
22:12 celeron55 it's going to always be too many and it can't be solved by forcefully removing them; there needs to be some way for a human to manage the too many settings
22:12 PilzAdam hmmmm, we now have a nice GUI for the settings
22:12 hmmmm well those are usually auto-set with an option that encompasses all similar settings
22:12 PilzAdam where it's possible to hide some advanced settings
22:12 hmmmm like a slider control for visual effects
22:12 VanessaE hmmmm: the current GUI works well for that.  "basic" stuff first, "Advanced" stuff behind a button
22:12 hmmmm no i'm not talking about hiding most of the advanced settings
22:12 Fixer VanessaE, yeah, designers are nuts those days, well designed classical keyboard is too boring it seems, and we need some more mazohism
22:12 hmmmm i'm talking about better settings configuration design in general
22:12 VanessaE you mean more like a "good, better, best" gradation?
22:12 hmmmm yes
22:13 hmmmm so if you had a Visual Effects slider that you could crank all the way up you'd have RBA's presets
22:13 VanessaE I'm not entirely against such a thing, so long as one can still get into, as you put it, the "nitty gritty" if desired.
22:13 hmmmm and if you really wanted to disable this one option, you can, via the text config file
22:13 hmmmm naw'mean?
22:14 celeron55 you don't really need to touch the text file; just open the advanced settings
22:14 celeron55 but same thing
22:14 VanessaE hmmmm: exactly that.
22:14 hmmmm yea
22:14 hmmmm this is what the standard for AAA games is i think
22:14 hmmmm so okay
22:15 hmmmm i guess my complaint isn't with having too many config options, it's more a matter of presenting all these options to the end user
22:15 VanessaE Fixer: I'd have been more happy with one of those weird split-in-half M$ keyboards :P
22:15 hmmmm in an awkward manner
22:15 VanessaE (but I don't think they existed at the time)
22:15 hmmmm in fact you know something, this is one of my ideas for the mapgen noiseparams config settings that never happened
22:15 VanessaE how do you mean?
22:16 hmmmm you remember a long time ago, I wanted just one mapgen, but all the noise parameters were freely configurable
22:16 VanessaE yes
22:16 hmmmm but that would be too many options, so there'd be a dropdown box where you select one of "flat, forests, plains, oceans, islands" etc.
22:16 hmmmm and that enters the presets of noiseparams for that type of terrain
22:17 hmmmm well like that
22:17 hmmmm except for other config options
22:18 VanessaE I could see checkboxes for oceans, deserts, jungles, and other large-scale map features, though.
22:18 VanessaE but yeah, having some presets would be good
22:19 hmmmm the 'select a checkbox for each desired terrain feature' idea already exists in the form of paramat's not-so-simple flat mapgen
22:19 VanessaE well you know what I meant
22:20 VanessaE I so rarely create new worlds, I don't actually know what the current state of the GUI for it is.
22:22 VanessaE though one thing I had wished for at one time was a "preview" of sorts, not unlike you see in some older games like Micropolis/Sim City
22:22 hmmmm we can still do that
22:22 VanessaE Something simple, like that online thing celeron55 had for "making sense of" perlin params
22:23 iangp joined #minetest-dev
22:25 hmmmm yeah
22:25 hmmmm i already made something like that, it's just not a part of minetest
22:26 hmmmm also i have a 3d mode:  http://i.imgur.com/2l13W0U.png
22:26 VanessaE oh that's nice
22:26 hmmmm that's for figuring out appropriate biome values
22:28 hmmmm i think it's more beneficial in the long run to figuring out how to short circuit some logic in minetest to make it generate and display map faster without entering an actual game
22:28 hmmmm s/to//
22:28 hmmmm i coded this up and then compared to minetest's source and it looks really close to mapblock_mesh.cpp
22:29 VanessaE well, you could produce a bare, false-color terrain maybe.  no deco, no trees (well unless it's easy), no lua mapgen hooks, etc
22:30 VanessaE that would certainly short-circuit it
23:01 yang2003 joined #minetest-dev
23:22 DFeniks joined #minetest-dev
23:25 paramat left #minetest-dev
23:42 est31 joined #minetest-dev
23:43 est31 regarding settings, we could improve i think and make some settings per world
23:43 est31 and allow the user to change those settings
23:43 est31 as well as mods to define new settings
23:43 est31 I do like however how we present the current settings to the users
23:44 VanessaE est31: you're taking after paramat :P
23:44 est31 hah lol
23:44 est31 yeah lets close some issues
23:44 est31 what about #3970 , there are no graphical bugs
23:44 ShadowBot https://github.com/minetest/minetest/issues/3970 -- Graphical bugs when drawing inventory hud
23:44 est31 its a lovely characteristic!
23:45 est31 same for
23:45 proller joined #minetest-dev
23:45 est31 #3964
23:45 ShadowBot https://github.com/minetest/minetest/issues/3964 -- lighting glitch
23:45 est31 we must close them all
23:45 est31 592 open issues are too much1
23:45 est31 !
23:45 est31 xD
23:46 hmmmm i don't get what that last issue is about
23:46 est31 probably they mined, and suddenly there was light
23:46 est31 out of nowhere no torch no anything
23:47 hmmmm oh is that what it is
23:47 hmmmm ahh i see it now, that hole in the ceiling looked like it was connecting the surface to the cave he's in
23:48 hmmmm can't do anything there until he tries the latest version
23:50 paramat joined #minetest-dev
23:50 est31 yeah its really one of the kind of bugs you dont want
23:50 est31 probably hard to reproduce, etc.
23:50 est31 but still, perhaps there is hope to solve them
23:55 est31 going again
23:55 est31 bye all

| Channels | #minetest-dev index | Today | | Google Search | Plaintext