Time |
Nick |
Message |
00:34 |
|
MTDiscord joined #minetest |
00:48 |
|
toluene joined #minetest |
01:13 |
|
erle joined #minetest |
01:21 |
|
Thermoriax joined #minetest |
02:02 |
|
Verticen joined #minetest |
02:37 |
|
toluene joined #minetest |
03:06 |
|
Verticen joined #minetest |
03:08 |
|
olliy joined #minetest |
04:00 |
|
MTDiscord joined #minetest |
04:06 |
|
toluene joined #minetest |
04:46 |
|
adam_mc joined #minetest |
04:53 |
|
ball left #minetest |
05:27 |
|
calcul0n_ joined #minetest |
05:37 |
|
Yad joined #minetest |
05:58 |
|
Verticen joined #minetest |
06:08 |
BuckarooBanzai |
does anyone know how to strip color-codes and translations from strings in the engine, for example those here: "(T__builtin)Current day is F4E.E" (i hope the string shows up properly) |
06:09 |
BuckarooBanzai |
i think the control chars were stripped :/ just in case: https://gchq.github.io/CyberChef/#recipe=To_Hexdump(16,false,false,false)&input=GyhUQF9fYnVpbHRpbilDdXJyZW50IGRheSBpcyAbRjQbRS4bRQ |
06:20 |
|
lemonzest joined #minetest |
06:32 |
|
orwell96 joined #minetest |
06:48 |
|
erle joined #minetest |
06:58 |
|
toluene7 joined #minetest |
07:11 |
|
olliy joined #minetest |
08:30 |
|
lemonzest joined #minetest |
08:42 |
|
JerryXiao joined #minetest |
10:07 |
MTDiscord |
<Bastrabun> When I try to compile latest master (51d0ac) of minetest irrlicht on a debian testing machine, it says "CMake Error at source/Irrlicht/CCMakeLists.txt:103 (message): XInput not found. Is that a new dependency? |
10:08 |
MTDiscord |
<ROllerozxa> yeah, xinput2/libxi |
10:10 |
MTDiscord |
<Bastrabun> Ah ... that's why when I installed xinput the error remained. apt can find neither xinput2 nor libxi though |
10:11 |
MTDiscord |
<ROllerozxa> whoops |
10:12 |
MTDiscord |
<Bastrabun> Ah, got it. I found the commit, it checks for libxi-dev. When I install THAT one it let's me continue |
10:13 |
MTDiscord |
<Danil_2461> not nice |
10:13 |
MTDiscord |
<ROllerozxa> lol |
10:13 |
erle |
Bastrabun, create a PR |
10:13 |
MTDiscord |
<Bastrabun> May I suggest to add the library to the dependencies in the doc? |
10:14 |
MTDiscord |
<Bastrabun> A PR? I doubt I'll be able to fix a missing dependency by myself, I know next to nothing about irrlicht ... also, someone arduously added the dependency, they certainly don't need a noob twiddling with it |
10:15 |
MTDiscord |
<ROllerozxa> erle is probably talking about adding it to the list of dependency packages that should be installed |
10:16 |
erle |
yes |
10:16 |
MTDiscord |
<Bastrabun> bonks self ok, that sounds a bit less scary. I'll at least open an issue, if I manage to come to terms with github again, I'll add a PR |
10:18 |
erle |
ROllerozxa any idea how/why this works? https://forum.minetest.net/viewtopic.php?f=47&t=28350&p=412499#p412499 |
10:27 |
MTDiscord |
<ROllerozxa> I would assume deflate compression has an easier time compressing uncompressed TGA images rather than PNG images that are already deflate compressed, but I don't know why that would lead in a noticeable size difference |
10:31 |
erle |
well, even if i tell zip to not compress them, it does not seem to favor png |
10:31 |
erle |
which is weird |
10:31 |
erle |
but it seems that optipng is the wrong thing to do for small textures |
10:38 |
|
specing_ joined #minetest |
10:57 |
|
appguru joined #minetest |
10:58 |
|
degradka[m] joined #minetest |
11:47 |
sfan5 |
BuckarooBanzai: https://github.com/Uberi/Minetest-WorldEdit/blob/abc9efeeb8cccb3e23c055414941fed4a9871b9a/worldedit_commands/init.lua#L125-L151 |
11:47 |
sfan5 |
simple and easy if you only need stripping |
11:47 |
|
definitelya joined #minetest |
11:52 |
|
est31 joined #minetest |
11:54 |
|
fling joined #minetest |
11:54 |
|
natewrench joined #minetest |
11:58 |
sfan5 |
Bastrabun: the dependency list in the docs will be updated soon, at the least before release |
11:59 |
erle |
should be done in the PR that adds a dependency IMO |
12:07 |
sfan5 |
how do I modify the README in minetest/minetest from a commit in minetest/irrlicht? |
12:09 |
sfan5 |
(trick question) |
12:09 |
BuckarooBanzai |
thank you sfan5 ? |
12:09 |
|
fling_ joined #minetest |
12:11 |
erle |
sfan5, submodules or subtrees maybe, both things that minetest development is deathly allergic to |
12:33 |
|
smk joined #minetest |
12:37 |
|
orwell96 joined #minetest |
12:45 |
|
Taoki joined #minetest |
12:48 |
sfan5 |
do submodules allow you to deviate from the fixed commit hash and run a newer tree locally? |
12:56 |
rubenwardy |
yes |
12:56 |
rubenwardy |
when you change the HEAD of a submodule, it shows up in git status and you can then commit it to update the fixed commit hash. But you can also just not commit it |
13:07 |
|
toluene joined #minetest |
13:11 |
|
wolfshappen joined #minetest |
13:13 |
sfan5 |
I guess it wouldn't be that bad then |
13:18 |
|
natewrench joined #minetest |
13:20 |
natewrench |
Is there a mod for minetest that smooths out the cart rides because it jerks your camera around when doing l going diagonally |
13:20 |
|
est31 joined #minetest |
13:21 |
|
natewrench joined #minetest |
13:34 |
|
natewrench joined #minetest |
13:34 |
|
Fixer joined #minetest |
13:50 |
|
est31 joined #minetest |
14:07 |
|
est31 joined #minetest |
14:09 |
|
orwell96 joined #minetest |
15:05 |
|
ronoaldo joined #minetest |
15:08 |
|
ronoaldo joined #minetest |
15:16 |
|
orwell96 joined #minetest |
16:03 |
|
est31 joined #minetest |
16:03 |
|
___nick___ joined #minetest |
16:12 |
|
___nick___ joined #minetest |
16:15 |
|
___nick___ joined #minetest |
16:19 |
|
erle joined #minetest |
16:42 |
|
est31 joined #minetest |
17:15 |
|
sometalgoo joined #minetest |
17:16 |
|
Thermoriax joined #minetest |
17:22 |
|
est31 joined #minetest |
17:24 |
|
fling_ joined #minetest |
17:47 |
|
proller joined #minetest |
18:04 |
|
Yad joined #minetest |
18:19 |
|
Yad joined #minetest |
18:21 |
|
Yad joined #minetest |
18:25 |
|
Trifton joined #minetest |
18:27 |
|
Polo joined #minetest |
18:28 |
|
natewrench joined #minetest |
18:58 |
|
Verticen joined #minetest |
19:05 |
|
natewrench joined #minetest |
19:23 |
erle |
rubenwardy i have learned how to do shaders with openGL 1.4 … can i use this knowledge to improve minetest somehow? https://github.com/minetest/minetest/issues/12569#issuecomment-1191848393 |
19:23 |
|
olliy joined #minetest |
19:23 |
erle |
ah you already answered |
19:26 |
sfan5 |
not sure who brought up shaders |
19:26 |
sfan5 |
the issue is a missing color format in your gpu |
19:27 |
rubenwardy |
I was thrown by the OpenGL 1 |
19:29 |
|
Yad joined #minetest |
19:30 |
sfan5 |
huh I missed that |
19:31 |
erle |
sfan5 the missing color format is a red herring. if i change the color format, i still get shadows and the rendering is still messed up. |
19:31 |
erle |
obviously, i just changed it to some random supported format |
19:31 |
erle |
to figure out if that works |
19:31 |
rubenwardy |
is it possible that the hardware only supports fixed pipeline, and shaders are being software emulated? |
19:31 |
sfan5 |
"if I change random things to make the error go away Minetest is messed up" |
19:32 |
sfan5 |
critical blocker |
19:32 |
erle |
sfan5 no, it's exactly as messed up as before |
19:32 |
erle |
look, it's a shadow, it's gray-ish, a little color format roulette does not make it look meaningfully different |
19:32 |
sfan5 |
without you telling which format you picked I can make no assumptions on whether you broke it or not |
19:33 |
erle |
the shadow renders in any case. just in one case i get an error message. |
19:33 |
sfan5 |
that was not sufficiently clear |
19:34 |
sfan5 |
but it does not matter in any way that your prehistoric GPU cannot render shadows correctly despite apparently offering (fatally broken) shader support to applications |
19:34 |
sfan5 |
this is exactly why I don't bother replying to issues of this kind |
19:34 |
erle |
for some reason other applications manage to render shadows, except for stormbaancoreur, against which i filed a bug |
19:35 |
rubenwardy |
I think the end result is the same in any case: your hardware is too old for us to justify spending time debugging [Won't fix] [Obscure] |
19:35 |
erle |
rubenwardy, how would i know if shaders are software emulated |
19:35 |
erle |
well, i am debugging it obviously |
19:35 |
sfan5 |
its cool that other applications use stencil shadows which work even on ancient hardware |
19:35 |
sfan5 |
Minetest doesn't and never will |
19:36 |
erle |
because it looks shit i assume? |
19:36 |
erle |
rubenwardy, how would i know if the shaders are some kind of software emulation? temperature does not rise noticeably even if i go fullscreen. |
19:37 |
rubenwardy |
who knows |
19:37 |
rubenwardy |
I don't |
19:38 |
sfan5 |
erle: here I copy-pasted a reddit comment: "Modern hardware and rendering techniques are better suited to shadow mapping, so most stencil shadowing information is out of date and based on legacy OpenGL. Stencil shadowing was an invention of the fixed-function era, and generally won't perform very well when used with shader-driven rendering - with |
19:38 |
sfan5 |
the possible exception of Doom 3's esoteric shadowing technique, but newer versions of id's engines abandoned that design anyways." |
19:38 |
|
natewrench joined #minetest |
19:38 |
erle |
i see |
19:58 |
erle |
rubenwardy why do the zipgrep results expire? |
20:01 |
rubenwardy |
They're stored in redis as celery worker results |
20:02 |
|
orwell96 joined #minetest |
20:04 |
erle |
am i correct that for shadows i would have to render a scene from the sun's POV but culling front faces and then doing a depth test and then draw from tha cameras POV and project the depth test texture on the scene somehow? |
20:23 |
|
natewrench joined #minetest |
20:25 |
|
natewrench joined #minetest |
20:28 |
|
natewrench joined #minetest |
20:40 |
|
natewrench joined #minetest |
20:54 |
|
natewrench joined #minetest |
20:59 |
|
specing_ joined #minetest |
21:08 |
|
est31 joined #minetest |
21:19 |
|
natewrench joined #minetest |
21:44 |
|
erle joined #minetest |
21:57 |
|
hatman joined #minetest |
21:59 |
hatman |
Hi guys , I am wondering if there is a way to scan infront of a node say by 3 nodes ? I know of minetest.get_objects_inside_radius but that is a scan all around the node rather than just infront of it for example. Does anyone have any ideas as to a way I might be ble to achieve this please :) |
22:01 |
MTDiscord |
<MisterE> Yes. minetest.get_node(pos) add 3 to whichever direction (x or z) you want to scan in |
22:02 |
hatman |
Thanks I hadnt thought of that I will look into it :) |
22:03 |
hatman |
I also just realised I could look at the code for the technic quarry as it does similar to what Iam looking for:0 |
22:03 |
MTDiscord |
<Benrob0329> That sounds like a job for raycasting |
22:03 |
hatman |
im looking to detect entitys in a 3x3 area infront of a machine |
22:05 |
hatman |
i guess manually going through each of the 9 positions and checking the +y :) I( was just trying to think how to achieve the for loop to iterate through them :) |
22:12 |
MTDiscord |
<MisterE> Oh, my method only works for nodes, yeah, what benrob said |
22:16 |
|
comrad joined #minetest |
22:22 |
hatman |
Interesting thanks Benrob0329 :) I havent used it yet but just found the commit adding it wich tells me all I need to know for now to atleast try and get something working :) Thanks again :) |
22:23 |
hatman |
Thanks for your reply to MisterE |
22:34 |
|
panwolfram joined #minetest |
22:45 |
MTDiscord |
<Benrob0329> hatman: https://gitlab.com/benrob0329/ikea/-/blob/master/mods/ikea_tools/init.lua#L44 |
22:45 |
MTDiscord |
<Benrob0329> along with: https://github.com/minetest/minetest/blob/master/doc/lua_api.txt#L1554 |
22:46 |
MTDiscord |
<Benrob0329> along with: https://github.com/minetest/minetest/blob/master/doc/lua_api.txt#L5605 |
22:46 |
MTDiscord |
<Benrob0329> Raycast is an alias to minetest.raycast |
22:47 |
erle |
rubenwardy I think i have figured it out: GLSL was introduced as an extension to openGL 1.4 and later the functions were slightly renamed for OpenGL 2.0, but the functionality is essentially the same. |
22:52 |
erle |
rubenwardy also, once a shader exceeds a specific instruction count, i get extremely bad software rendering and minetest does exceed this instruction count1 |
22:52 |
erle |
so its basically the same issue as the mali 400 thing |
22:54 |
erle |
and maybe the solution should not be to make it impossible to enable shaders, but to make it impossible to enable shaders that exceed the shader model that the GPU supports |
22:54 |
|
orwell96 joined #minetest |
22:57 |
erle |
basically, the important thing is not “do you have openGL (ES) 2.0?” or “do you have GLSL?”, as you can have both and it can still not work |
23:14 |
|
fling joined #minetest |
23:18 |
erle |
in conclusion … all of you should be forced to use 10 year old underpowered netbooks, maybe people would write more performant software then ^^ |
23:28 |
|
MTDiscord joined #minetest |
23:35 |
|
natewrench joined #minetest |
23:36 |
natewrench |
Hah I should frame this and put this in my room https://imgur.com/a/fhSZCPT |
23:41 |
|
natewrench joined #minetest |