| Time |
Nick |
Message |
| 00:46 |
|
v-rob joined #minetest |
| 00:58 |
|
smk joined #minetest |
| 01:04 |
|
Trifton joined #minetest |
| 01:38 |
|
liceDibrarian joined #minetest |
| 02:36 |
|
Lesha_Vel joined #minetest |
| 03:53 |
|
behalebabo joined #minetest |
| 04:00 |
|
MTDiscord joined #minetest |
| 04:22 |
|
v-rob joined #minetest |
| 04:51 |
|
fluxionary joined #minetest |
| 05:06 |
|
fluxionary joined #minetest |
| 05:10 |
|
calcul0n_ joined #minetest |
| 05:44 |
|
s20 joined #minetest |
| 06:25 |
|
definitelya joined #minetest |
| 06:31 |
|
gxt joined #minetest |
| 06:46 |
|
YuGiOhJCJ joined #minetest |
| 07:17 |
|
germ joined #minetest |
| 07:18 |
|
jaca122 joined #minetest |
| 08:20 |
|
mrkubax10 joined #minetest |
| 08:21 |
|
sys4_ joined #minetest |
| 08:22 |
|
Warr1024 joined #minetest |
| 08:46 |
|
Warr1024 joined #minetest |
| 08:59 |
|
appguru joined #minetest |
| 10:00 |
|
Talkless joined #minetest |
| 10:19 |
|
calcul0n joined #minetest |
| 10:20 |
|
s20 joined #minetest |
| 10:42 |
|
appguru joined #minetest |
| 11:00 |
|
erle joined #minetest |
| 11:34 |
|
lissobone joined #minetest |
| 11:34 |
|
mrkubax10 joined #minetest |
| 11:51 |
|
YuGiOhJCJ joined #minetest |
| 11:53 |
|
erle joined #minetest |
| 11:55 |
|
hare_hare_yukai joined #minetest |
| 11:56 |
lissobone |
Greetings. |
| 11:56 |
hare_hare_yukai |
Greetings. |
| 11:58 |
|
appguru joined #minetest |
| 12:24 |
|
qqq joined #minetest |
| 12:27 |
|
book` joined #minetest |
| 12:46 |
|
appguru joined #minetest |
| 13:13 |
|
TheSilentLink joined #minetest |
| 14:11 |
|
lemonzest joined #minetest |
| 14:24 |
|
appguru joined #minetest |
| 14:25 |
|
s20 joined #minetest |
| 14:46 |
|
appguru joined #minetest |
| 15:04 |
|
Lunatrius` joined #minetest |
| 15:17 |
|
doseijin joined #minetest |
| 15:27 |
|
s20 joined #minetest |
| 15:55 |
|
liceDibrarian joined #minetest |
| 16:05 |
|
Talkless joined #minetest |
| 16:05 |
|
TomTom joined #minetest |
| 16:26 |
|
Zambyte joined #minetest |
| 16:53 |
|
erle joined #minetest |
| 17:39 |
|
olliy joined #minetest |
| 17:43 |
|
mazes_83 joined #minetest |
| 17:50 |
mazes_83 |
long time I didn't add new node, I modeled a very basic thing in blender, UV points to 4 different textures, but when I put node in world only first texture is taken in account |
| 17:54 |
erle |
mazes_83 show test case pls |
| 17:54 |
MTDiscord |
<greenxenith> My guess is you didnt export material groups |
| 17:56 |
mazes_83 |
ah greenxenith, I bet it is the correct answer. |
| 18:05 |
mazes_83 |
ty greenxenith, that was the correct guess |
| 18:05 |
MTDiscord |
<greenxenith> Awesome, glad it worked |
| 18:06 |
|
mrkubax10 joined #minetest |
| 18:30 |
|
v-rob joined #minetest |
| 18:48 |
|
jaca122 joined #minetest |
| 19:24 |
|
jaca122 joined #minetest |
| 19:28 |
MTDiscord |
<mnh48> erle : the CJK variant handling is typically done by getting the language someone is using and show the variant appropriately |
| 19:29 |
MTDiscord |
<mnh48> though in terms of using with signs, maybe have a way for the library to know what language they're writing on said sign?? |
| 19:29 |
MTDiscord |
<mnh48> CJK variant is how one font can be used for all three languages and still appear correctly based on language |
| 19:40 |
erle |
mnh48 so how should i change the API? |
| 19:42 |
MTDiscord |
<mnh48> probably finding a way to know what language the user who write the sign is using?? |
| 19:42 |
MTDiscord |
<mnh48> or maybe just put dropbox in the sign menu asking for language lol |
| 19:42 |
erle |
that's not a good answer for a TEXT RENDERING LIBRARY |
| 19:42 |
erle |
look, my interface right now is like this |
| 19:43 |
erle |
text_bitmap_1 = hexdraw.render_text("wð♥𐍈😀!🂐겫") |
| 19:45 |
MTDiscord |
<mnh48> if it's that then add parameter for language code?? |
| 19:45 |
MTDiscord |
<mnh48> maybe like |
| 19:45 |
MTDiscord |
<mnh48> text_bitmap_1 = hexdraw.render_text("早上好中国，现在我有冰淇淋","zh-Hans") |
| 19:45 |
MTDiscord |
<mnh48> then use the language code to decide which variant is used |
| 19:48 |
MTDiscord |
<mnh48> for reference on CJK Variant, this page is displayed with Noto Sans CJK (one font) except for Vietnamese part, look at how the character is different because it's in different language (in HTML, the language can be specified with lang= like <td lang="zh-Hans">) |
| 19:48 |
MTDiscord |
<mnh48> https://cdn.discordapp.com/attachments/749727888659447960/1147981512373960794/2023-09-04_03-42-56_vivaldi.png |
| 19:49 |
muurkha |
erle: this is indeed a major problem with CJK unification in Unicode |
| 19:49 |
muurkha |
the character codes themselves do not provide enough information to allow you to render text properly |
| 19:49 |
muurkha |
imagine if ß were encoded in ISO-8859-1 as "ss" |
| 19:50 |
muurkha |
I guess that's not quite right, because "ss" does occur in German in a few cases |
| 19:50 |
erle |
hexdraw.render_text("早上好中国，现在我有冰淇淋","zh-Hans") is a bad API design |
| 19:51 |
erle |
the second parameter is a dictionary, which makes extensibility much easier |
| 19:51 |
erle |
muurkha, switzerland encodes ß as ss, because they have no letter ß AFAIK |
| 19:52 |
MTDiscord |
<mnh48> I'm just giving example of how language information could be passed on for actual processing, basically do what you want for the actual implementation |
| 19:52 |
erle |
so the jokes write themselves basically |
| 19:53 |
erle |
“genießen in maßen” (enjoy in moderation) vs. “genießen in massen” (enjoy en masse) |
| 19:53 |
erle |
the funniest thing is that a popular book on german words and grammar DER GROẞE DUDEN for a long time had a ß in it but there was no ẞ (capital letter ß) |
| 19:54 |
erle |
so by the rules of the book, it's own entirely uppercase title may have not been legal |
| 19:55 |
MTDiscord |
<mnh48> also this is just CJK variant, which is considerably easier than say, positional glyph like in Arabic lmao .. since it's just one-to-one on what to show for which language |
| 19:56 |
MTDiscord |
<mnh48> for positional glyph, in addition to language, you also need to know what character came before or after what character, and to know if current character need to be join to before, to after, to both, or to none surrounding characters |
| 19:56 |
muurkha |
I feel like positional variants are less of a problem |
| 19:56 |
erle |
yes, so how do you propose i get the CJK variant of a .hex font file |
| 19:56 |
erle |
i mean out of |
| 19:57 |
muurkha |
I mean a medial or final alef is still a medial or final alef whether you're writing in Arabic, Farsi, or Urdu, right? |
| 19:57 |
muurkha |
so even though you don't have enough information in a single code point, you have enough information in the string |
| 19:57 |
erle |
muurkha have you seen the shitty rendering i do with my pure lua unicode renderer? the arabic letters are not joined at all |
| 19:57 |
erle |
i can trivially check out the previous or next letter obv |
| 19:57 |
erle |
like, isn't this like one preprocessing step or so? |
| 19:57 |
MTDiscord |
<mnh48> yes, but final yeh (ي) is different in some languages, as in some language drops the two dots while others retain it |
| 19:58 |
muurkha |
hmm, I didn't realize that |
| 19:58 |
muurkha |
erle: I have terminals that do that too |
| 19:58 |
erle |
so to summarize: a language tag given to the rendering function is a good idea. |
| 19:58 |
muurkha |
unfortunately I don't speak any languages written in Arabic script |
| 19:58 |
erle |
the problem remains how i do anything with it |
| 19:59 |
erle |
mnh48 give me a way to figure out which variant character i need and where it is in unifont, then i put it in |
| 19:59 |
erle |
i do actually query the unicode database for stuff |
| 19:59 |
erle |
so assume i have it |
| 19:59 |
erle |
(see source code) |
| 20:06 |
MTDiscord |
<mnh48> I'm not sure if unifont even carry the variant data like other fonts (they do have a separate Unifont JP font to cover Japanese specifically but that's it) |
| 20:06 |
MTDiscord |
<mnh48> https://unifoundry.com/unifont/index.html |
| 20:06 |
MTDiscord |
<mnh48> see "Unifont Limitations" |
| 20:09 |
|
Wooster joined #minetest |
| 20:23 |
erle |
mnh48 but do i really need a language tag then, if the user could just load the japanese glyphs? |
| 20:24 |
erle |
i have drawn a lot of emoji for unifont, but that was a long time ago |
| 20:26 |
MTDiscord |
<mnh48> not all font release separate versions for different languages, some only release one version for all |
| 20:26 |
MTDiscord |
<mnh48> in which that one version contained the variants in it |
| 20:26 |
erle |
yeah but i only have a loader for the hex font file format |
| 20:26 |
erle |
and given the state of existing sign libraries, i doubt anyone else is going to write a font loader soon |
| 20:26 |
erle |
so i will defer this right now ig |
| 20:26 |
erle |
but thanks for the pointers |
| 20:27 |
erle |
anything else API-wise that you'd like? |
| 20:27 |
erle |
i mean i have tab width |
| 20:27 |
erle |
and kerning yes/no |
| 20:27 |
erle |
though the kerning algorithm is really stupid |
| 20:29 |
MTDiscord |
<mnh48> hmmm.. nothing else for now |
| 20:31 |
erle |
did you look at the example rendering? |
| 20:32 |
Krock |
rendering stuff is freetype's business, not Minetest's |
| 20:32 |
erle |
i think you are mistaken |
| 20:33 |
erle |
Krock have you looked at my thing? |
| 20:33 |
Krock |
please be more specific |
| 20:33 |
erle |
Krock https://git.minetest.land/erlehmann/tga_unicode_encoder |
| 20:33 |
erle |
waid |
| 20:33 |
erle |
wait |
| 20:33 |
erle |
wrong link |
| 20:34 |
erle |
Krock https://git.minetest.land/erlehmann/tga_unicode_renderer |
| 20:34 |
erle |
Krock here is some sample output: https://mister-muffin.de/p/_TxA.png |
| 20:35 |
Krock |
did you implement a font library in Lua? |
| 20:35 |
erle |
yes |
| 20:36 |
erle |
i also accidentally found out that you can reduce the size of textures by 30% over the most optimized optipng by supporting tga.z as a format (and sending normal tga to older clients) |
| 20:36 |
erle |
basically, i found that quake etc. were always right |
| 20:36 |
erle |
for the size of the textures we are talking about |
| 20:37 |
Krock |
I see. The server builds rarely or never come with freetype |
| 20:37 |
erle |
Krock i did it half a year ago, but depression and controlling partner … i released it 6 months after writing |
| 20:37 |
erle |
the goal is unicode on signs |
| 20:37 |
erle |
since i do not have a clue of arabic or hebrew, the thing still sucks for bidirectional text (numbers in hebrew text) or connected letters (arabic) |
| 20:38 |
erle |
but otherwise, i think i am FAR ahead of every signs library |
| 20:38 |
Krock |
I suppose font parsing and rendering is a good exercise for parsing files but it's already possible to generate text on top of Irrlicht-type Images and load those as textures |
| 20:38 |
Krock |
it just hasn't been implemented yet. afaik there's an open PR for that |
| 20:38 |
erle |
okay, but i HAVE implemented something |
| 20:38 |
erle |
my thing is there and it works |
| 20:39 |
erle |
and i have seen the issues and decided it this way because it's actually not too slow |
| 20:39 |
erle |
also this is 100% backwards compatible |
| 20:39 |
erle |
after all, i just generate a texture |
| 20:39 |
erle |
Krock, i'd appreciate input regarding the shape of the API. any suggestions? so far i have the suggestion of a language tag. |
| 20:40 |
erle |
it's only that i can't do much with it right now |
| 20:40 |
erle |
also look into my texture compression benchmarks, shaving 30% of transfer size is no joke |
| 20:41 |
Krock |
that's a good achievement regarding the tools that you're working with |
| 20:41 |
erle |
you mean lua? |
| 20:41 |
erle |
i initially thought lua would be in the way |
| 20:41 |
Krock |
Lua and server-side generated textures that would optimally be sent in text form to clients for rendering on the fly rather than caching every single texutre |
| 20:42 |
erle |
wait what how |
| 20:43 |
erle |
i am pretty sure whatever you envision is only possible if you allow the lua code to execute on the client, which is a bit oof |
| 20:43 |
erle |
i mean, how would you render the funny colored text otherwise |
| 20:43 |
Krock |
there's not much needed to introduce client-sided support: https://github.com/minetest/minetest/pull/12084 |
| 20:43 |
erle |
i am pretty sure this does not do what i do |
| 20:44 |
erle |
like it can do less and also got closed |
| 20:44 |
Krock |
colored text can already be created as string and sent |
| 20:44 |
erle |
i am doing pixel operations on a bitmap that i rendered text in. how are you going to do this in a performant, backwards-compatible way which allows the server to choose the font |
| 20:44 |
erle |
cora has a game that uses the conlang toki pona, she may want her signs to have a toki pona font |
| 20:45 |
erle |
as i said, i looked at existing approaches and rejected them (not said: due to constraints regarding the use of the thing) |
| 20:45 |
Krock |
features in Minetest rarely have been backwards-compatible. old clients are mostly guaranteed to end up with an incomplete experience |
| 20:45 |
erle |
still, i don't get it |
| 20:46 |
erle |
i mean you can do it all using new texmods, but it's not particularly useful |
| 20:46 |
Krock |
you'd rather send the font file as media and not every generated texture |
| 20:47 |
erle |
that's multiple megabytes. a sign texture is probably hundreds of bytes at most. |
| 20:47 |
Krock |
that would also allow texture generation/composition on the fly without manual generation |
| 20:47 |
erle |
you have to make a lot of signs until you get to the point where it makes sense. |
| 20:47 |
erle |
also last time pixel fonts were supported they were ripped out, so i have zero hope for pixel-accurate rendering on the client |
| 20:47 |
Krock |
you'd likely download the font once and use it on multiple servers. over time this will pay out |
| 20:48 |
erle |
people say that for websites that download 2MB js as well lol |
| 20:48 |
Krock |
anyway it's good to see that you found a solution that works for you :) |
| 20:48 |
erle |
look, i could have done it differently actually, using a huge texture and indexing into it using texmods |
| 20:49 |
erle |
but i'm pretty sure huge textures + not-too-modern GPUs = little graphics RAM left |
| 20:49 |
erle |
for example, there is this intel thing where if your GPU memory is nearly full, font rendering goes wonky |
| 20:49 |
erle |
anyways |
| 20:49 |
erle |
Krock API-wise, any ideas? |
| 20:50 |
erle |
Krock you put in utf-8 encoded text and get out a pixel array, any improvements you'd like to see regarding parameters or so? |
| 20:50 |
Krock |
can you generate `[png:` data yet? |
| 20:52 |
Krock |
API-wise I suppose it is pretty intuitive. |
| 20:52 |
erle |
i can not. you can do that, but you will find that a) minetest.encode_png() does not support remotely as many useful features as tga_encoder (for example, you may want your text colormapped or monochrome after all) b) your files get too big with minetest.encode_png() c) geeting to reasonable filesizes (still above tga.z) takes a long encoding time |
| 20:53 |
erle |
basically, there is *no* way this is going to be performant with PNG than with tga or tga.z |
| 20:53 |
erle |
as performant i mean |
| 20:53 |
erle |
(you can try to prove me wrong, but i suggest to try to understand what i did first) |
| 20:54 |
erle |
if you want to put the pixel array in [png, i am not hindering you. i am just saying it's not the best use of your, your players, or your servers time. |
| 20:55 |
erle |
Krock, is there any caching reason for [png maybe? |
| 20:55 |
erle |
like is the cache evicted of the generated texture once it is no longer referenced? |
| 20:56 |
Krock |
AFAIK the resulting texture is kept in memory until leaving to the main menu. that's the case for any loaded or generated texture |
| 20:56 |
erle |
so, do you have any other reason to use it? |
| 20:56 |
Krock |
you'd however gain the flexibility to use a greater variety of characters to e.g. write text on a sign |
| 20:57 |
erle |
as i said, if i wanted to use texmods i'd probably index a huge texture rather than create a lot of smaller ones ig? |
| 20:57 |
Krock |
the signs mod currently uses a sort of character map (flatten out as individual files) which are then put together |
| 20:57 |
erle |
yes, i know |
| 20:57 |
erle |
this approach is pretty limited |
| 20:58 |
erle |
and in fact, you get a much larger overhead than using a single picture that you index into |
| 20:58 |
Krock |
I thought that texmods would use more space than a [png: data blob? |
| 20:58 |
erle |
not sure, but [png has a minimum overhead, of … let me calculate |
| 20:59 |
erle |
probably around 95 useless bytes everytime you use it before you get to the payload |
| 20:59 |
erle |
not sure how many texmods you can put into that |
| 21:00 |
Sokomine |
by any chance tenplus1 hanging around currently? |
| 21:00 |
erle |
depends on your thing |
| 21:01 |
erle |
Krock in any case, how would you do the thing i do where i render text and then apply effects to the texture? |
| 21:02 |
erle |
at the bottom of test.lua i do this |
| 21:03 |
Krock |
in a hacky manner? `[png:` so that it could be used dynamically. in a proper manner? implement it in Minetest. |
| 21:03 |
mazes_83 |
@greenxenith: ty again: https://content.minetest.net/uploads/070998b66e.png |
| 21:04 |
MTDiscord |
<greenxenith> Nice 👌 |
| 21:04 |
erle |
Krock again, HOW would you implement this |
| 21:05 |
erle |
Krock if you use [png that does not solve “i want to render text to a bitmap and then modify the bitmap” |
| 21:05 |
erle |
because texmods are very limited for the purpose |
| 21:05 |
Krock |
if you already "render" it, what do you need to modify afterwards? |
| 21:06 |
erle |
the pixels |
| 21:06 |
Krock |
no shit sherlock |
| 21:06 |
erle |
e.g. to create outlines, shadows etc. |
| 21:06 |
erle |
same way i do in xmaps |
| 21:06 |
erle |
i render a bitmap, then i iterate over the pixels and modify it using filters |
| 21:06 |
erle |
the water texture in xmaps is created that way |
| 21:07 |
Krock |
if this were a new texmod (with font media sending I mentioned earlier), it would need some sort of flags or options to specify the way it is supposed to be rendered. of course, this would no longer allow fine-grained server-side control |
| 21:07 |
erle |
please no new knee-jerk reactions lol |
| 21:08 |
erle |
[png was bad enough |
| 21:08 |
erle |
https://git.minetest.land/erlehmann/tga_unicode_renderer/src/branch/main/test.lua look at line 69 ff |
| 21:08 |
erle |
i doubt you could do this with a texmod |
| 21:09 |
erle |
and about the fonts, how highly do you rate the chance that minetest will support hexfonts? i'd say about zero. |
| 21:09 |
erle |
as last time we had pixel fonts, they got removed |
| 21:09 |
Krock |
text colorization is already a Minetest feature |
| 21:09 |
erle |
look, i am 110% sure it can't do what i do there |
| 21:09 |
erle |
i am not colorizing letters |
| 21:09 |
Sokomine |
ah. working on the display of text ingame? that's good. thanks for doing so! may help a lot |
| 21:09 |
erle |
yes |
| 21:10 |
erle |
Sokomine it is not yet on cdb because i want to have API kinks ironed out |
| 21:10 |
erle |
Sokomine if you look at the examples, anything you might want to have? |
| 21:10 |
erle |
primarily for rendering text on signs |
| 21:11 |
erle |
Sokomine, see https://git.minetest.land/erlehmann/tga_unicode_renderer/src/branch/main/examples.lua line 58 ff |
| 21:11 |
Sokomine |
can't tell yet as i'm not really into it. i only know that it seems to strain the clients a lot |
| 21:12 |
Sokomine |
any improvements in that regard are highly welcome |
| 21:12 |
erle |
what exactly strains the clients, the texmod orgys? |
| 21:12 |
erle |
orgies |
| 21:12 |
Sokomine |
i'm currently a bit busy doing the final polish for yl_speak_up and adding a mod that adds some actual npc (or at least makes them able to be talked to) |
| 21:13 |
Sokomine |
text on signs, posters etc. seems to be part of the problem. there are more factors involved |
| 21:13 |
erle |
well i'd appreciate your input then |
| 21:13 |
erle |
i mean i just spit out pixels and that's it |
| 21:13 |
Sokomine |
in general, servers over time become...hard to bear at spawn due to stress for client and connection |
| 21:14 |
erle |
well i have a pretty shitty pc and so far i have not found any problem with having lots of small textures in cache … but texmods, yeah |
| 21:15 |
Sokomine |
merchant street on the your land server is something to just run away from as fast as possible - which is a pity because it's nicely built. but it's still a horror performance-wise. there are diffrent theories why that is so. everything that improves that is highly welcome |
| 21:15 |
erle |
pretty sure texmods can also be used to crash or hang clients |
| 21:15 |
Sokomine |
and that's not the only server spawn suffering over time. to some degree they all do |
| 21:15 |
erle |
Sokomine so would you be so kind and try to render some text or give me some text you render so i can see if my library can handle it well? |
| 21:16 |
Sokomine |
i'm not rendering any text as such. but if you want to see much text - go to the your land server? has enough posters and signs |
| 21:16 |
erle |
so you do not use signs or posters at all? |
| 21:16 |
erle |
i know that server |
| 21:16 |
erle |
spawn is laggy laggy laggy |
| 21:16 |
erle |
but i never thought it was the signs |
| 21:18 |
erle |
Krock Sokomine here is another rendering that shows why i want it pixel-perfect https://mister-muffin.de/p/nkfF.png |
| 21:18 |
erle |
(the math signs error was there in the source file) |
| 21:18 |
erle |
(or is it an error? i did not modify it) |
| 21:18 |
Sokomine |
i'm currently mostly working with npc and letting them talk. has been quite a while...but that mod is nearing completition (finally!) |
| 21:19 |
erle |
hmm, i should probably get my library into “this week in minetest” right? |
| 21:19 |
erle |
or this month or so |
| 21:19 |
erle |
whatever its called |
| 21:19 |
Sokomine |
looks nice, yes |
| 21:19 |
erle |
what do i need to do for that? |
| 21:20 |
MTDiscord |
<greenxenith> https://github.com/minetest/blog/issues/new/choose |
| 21:20 |
erle |
when is the next issue due? |
| 21:21 |
MTDiscord |
<greenxenith> The post usually gets published midway through the month |
| 21:21 |
erle |
that leaves a bit of time to get the API right lol |
| 21:21 |
erle |
greenxenith you have any opinions on my text rendering thing? |
| 21:21 |
erle |
if so, extrapolate your onions |
| 21:22 |
MTDiscord |
<greenxenith> Sorry, im a little busy with my own project at the moment |
| 21:22 |
erle |
no problem |
| 21:22 |
erle |
Sokomine do you see any other scenarios for text rendering than signs that i should consider? |
| 21:26 |
Sokomine |
signs come in many variants. in the end they're all signs, but from a ton of diffrent mods and in very diffrent style. there are those long ones used for train-stations or shop advertisement, then the posters (very popular), homedecor signs, signs with arrows etc... |
| 21:27 |
Sokomine |
there are also name tags of mobs that...exist. i have no idea in which...relation...or how relevant that is. if you met an npc now and then with a name tag, that probably doesn't matter at all. if you have a scout invasion with hundreds of them - or just someone gone crazy with breeding chicken or ducks... |
| 21:27 |
erle |
name tags are not textures though, or are they? |
| 21:28 |
Sokomine |
i have no idea! they are text. no idea how they're handled engine-wise |
| 21:30 |
erle |
Sokomine btw, i also want to improve xmaps to allow custom maps. any idea for that? |
| 21:30 |
erle |
i mean i prob just need a callback for an algorithm that writes a bitmap |
| 21:38 |
|
v-rob joined #minetest |
| 21:40 |
Sokomine |
don't know offhand what xmaps is |
| 21:41 |
Sokomine |
sorry :-( can't help you there. it's a diffrent area of work. but i'm thankful that you do it! |
| 21:41 |
erle |
Sokomine, xmaps is the ”generate ingame maps” mod that is not mcl_maps |
| 21:41 |
erle |
Sokomine https://content.minetest.net/packages/erlehmann/xmaps/ |
| 21:41 |
Sokomine |
ah. can't say that i have any expertise at all there |
| 21:42 |
erle |
you can pick up the map and it shows your position on it |
| 21:42 |
erle |
and you can put it on the wall |
| 21:42 |
erle |
and it lets you make treasure maps |
| 21:42 |
Sokomine |
ah! that sounds very nice |
| 21:42 |
erle |
see here https://content.minetest.net/uploads/bbb9c6276e.png |
| 21:42 |
erle |
right now it has only one renderer for maps |
| 21:42 |
erle |
which makes this treasure-map-style thing |
| 21:43 |
Sokomine |
might even be something the your land server might like. or almost any survival server for that matter |
| 21:43 |
erle |
yeah but they probably want a different renderer |
| 21:44 |
erle |
i mean the cheapest is just to take the node color |
| 21:44 |
erle |
i think the engine exposed that, right? |
| 21:44 |
erle |
the second-cheapest is taking the node color and elevation into account, like mcl_maps does |
| 21:44 |
erle |
mine is actually pretty convoluted, as you can see from the tree and grass placement on the map |
| 21:46 |
erle |
i also render houses on the maps if there are doors ;> |
| 22:20 |
|
mazes_83 joined #minetest |
| 22:32 |
|
panwolfram joined #minetest |
| 22:54 |
|
amfl2 joined #minetest |
| 23:34 |
|
sparky4 joined #minetest |
| 23:44 |
|
Lesha_Vel joined #minetest |
