| Time |
Nick |
Message |
| 00:04 |
|
SFENCE joined #minetest |
| 00:39 |
|
SFENCE joined #minetest |
| 00:53 |
|
AwesomeAdam54321 joined #minetest |
| 00:58 |
|
SFENCE joined #minetest |
| 01:04 |
|
SFENCE joined #minetest |
| 01:30 |
|
silverwolf73828 joined #minetest |
| 01:39 |
|
SFENCE joined #minetest |
| 01:44 |
|
Verticen joined #minetest |
| 01:51 |
|
vampirefrog joined #minetest |
| 01:58 |
|
SFENCE joined #minetest |
| 02:05 |
|
SFENCE joined #minetest |
| 02:16 |
|
AwesomeAdam54321 joined #minetest |
| 03:05 |
|
SFENCE joined #minetest |
| 04:00 |
|
MTDiscord joined #minetest |
| 04:31 |
BuckarooBanzai |
sfan5: i like the rising adoption of mods with integration tests :) |
| 04:55 |
|
sparky4 joined #minetest |
| 04:59 |
|
YuGiOhJCJ joined #minetest |
| 05:48 |
|
TomTom joined #minetest |
| 06:11 |
|
gregon joined #minetest |
| 07:11 |
|
sparky4 joined #minetest |
| 07:59 |
|
jaca122 joined #minetest |
| 08:17 |
|
mrkubax10 joined #minetest |
| 10:02 |
|
gregon joined #minetest |
| 10:12 |
|
ireallyhateirc joined #minetest |
| 10:23 |
|
ireallyhateirc joined #minetest |
| 10:26 |
|
silverwolf73828 joined #minetest |
| 10:34 |
|
ireallyhateirc joined #minetest |
| 11:11 |
|
Juri joined #minetest |
| 11:11 |
|
Niklp joined #minetest |
| 11:25 |
|
est31 joined #minetest |
| 11:27 |
|
cx384 joined #minetest |
| 11:32 |
|
Desour joined #minetest |
| 11:37 |
|
cheek_pain joined #minetest |
| 11:38 |
|
gregon joined #minetest |
| 12:27 |
|
illwieckz joined #minetest |
| 13:44 |
|
gregon joined #minetest |
| 13:59 |
[MatrxMT] |
<Blockhead256> filler bots posting on forums again https://forum.minetest.net/viewtopic.php?p=438759#p438759 |
| 14:07 |
|
Desour joined #minetest |
| 14:36 |
|
AwesomeAdam54321 joined #minetest |
| 14:41 |
|
SFENCE joined #minetest |
| 14:45 |
|
SFENCE joined #minetest |
| 14:53 |
|
SFENCE joined #minetest |
| 14:55 |
|
SFENCE joined #minetest |
| 15:03 |
|
SFENCE joined #minetest |
| 15:11 |
|
SFENCE joined #minetest |
| 15:22 |
|
SFENCE joined #minetest |
| 15:52 |
|
boingman joined #minetest |
| 15:53 |
MinetestBot |
[git] grorp -> minetest/minetest: Fix incorrect documentation of new-style particlespawner size property 5d18b6f https://github.com/minetest/minetest/commit/5d18b6fcd0960fa495164bd8ac91b39e2d3b4b2e (2024-08-27T15:51:29Z) |
| 15:54 |
|
jaca122 joined #minetest |
| 16:17 |
|
SFENCE joined #minetest |
| 16:19 |
|
gregon joined #minetest |
| 16:25 |
|
SFENCE joined #minetest |
| 16:30 |
|
SFENCE joined #minetest |
| 16:41 |
|
SFENCE joined #minetest |
| 16:49 |
|
SFENCE joined #minetest |
| 16:53 |
|
bodiccea joined #minetest |
| 16:58 |
|
bodiccea joined #minetest |
| 16:59 |
|
SFENCE joined #minetest |
| 17:03 |
|
SFENCE joined #minetest |
| 17:07 |
|
Talkless joined #minetest |
| 17:09 |
|
SFENCE joined #minetest |
| 17:11 |
|
vampirefrog joined #minetest |
| 17:20 |
|
SFENCE joined #minetest |
| 17:23 |
|
gregon joined #minetest |
| 17:51 |
|
SFENCE joined #minetest |
| 17:58 |
|
SFENCE joined #minetest |
| 18:01 |
|
kamdard joined #minetest |
| 18:13 |
|
SFENCE joined #minetest |
| 18:30 |
|
SFENCE joined #minetest |
| 18:42 |
|
ireallyhateirc joined #minetest |
| 18:44 |
|
SFENCE joined #minetest |
| 18:49 |
|
boingman joined #minetest |
| 18:50 |
|
blaub joined #minetest |
| 18:50 |
blaub |
swift110-mobile: hey what's your channel again? |
| 18:55 |
|
jaca122 joined #minetest |
| 19:15 |
|
SFENCE joined #minetest |
| 19:19 |
|
SFENCE joined #minetest |
| 19:21 |
|
kamdard joined #minetest |
| 19:25 |
MTDiscord |
<greenxenith> Discord bridge may reconnect a few times |
| 19:26 |
|
___nick___ joined #minetest |
| 19:32 |
blaub |
does minetest have enemies now? |
| 19:34 |
MTDiscord |
<greenxenith> I mean sure, plenty of people dont like the engine |
| 19:35 |
MTDiscord |
<greenxenith> Unless you are referring to mobs in a game, which would be weird because Minetest is an engine, not a game |
| 19:49 |
|
MTDiscord joined #minetest |
| 19:50 |
blaub |
very narcisstic answer |
| 19:50 |
|
blaub left #minetest |
| 19:51 |
MTDiscord |
<greenxenith> blaub: I am sorry you feel that way. What would you like to be different? |
| 19:51 |
ireallyhateirc |
not sure how that's narcisstic tbh |
| 19:51 |
MTDiscord |
<greenxenith> Ah, they left. |
| 19:51 |
jonadab |
In general, people who really want a PvE combat-oriented game, tend not to like the engine, because none of the combat-oriented games / mods that exist are very good. |
| 19:52 |
jonadab |
Not sure whether that is a weakness in the engine, or just a sign that the community writing content for it doesn't care that much about PvE combat. |
| 19:53 |
ireallyhateirc |
yeah, obviously the engine needs to be adjusted to support many different types of voxel games instead of a vague minecraft-like experience |
| 19:53 |
|
silverwolf73828 joined #minetest |
| 19:53 |
ireallyhateirc |
but we're getting there |
| 19:53 |
MTDiscord |
<greenxenith> It already does support other types of voxel games, but no one wants to make them |
| 19:54 |
ireallyhateirc |
I'm making one, please be patient!!! |
| 19:54 |
jonadab |
Personally, I like the fact that the games we have, tend to have a really solid peaceful-mode experience, complete with a meaningful power progression and the ability to farm large quantities of resources for building. |
| 19:55 |
jonadab |
Certain other voxel games, which shall remain nameless here but are implemented in Java, are absolute pants for peaceful mode play. |
| 19:55 |
jonadab |
Because too many of their infrastructure/farm things are based on mob drops. |
| 19:56 |
jonadab |
Minetest has never fallen into that particular trap. |
| 19:56 |
ireallyhateirc |
greenxenith, I'd argue that Minetest still has a way to go, for example dehardcoding stuff like camera, player physics and such |
| 19:56 |
MTDiscord |
<greenxenith> I never said there wasnt miles of room to improve |
| 19:57 |
MTDiscord |
<greenxenith> All I said was other types of voxel games are completely viable already |
| 19:58 |
ireallyhateirc |
well I guess, I always end up banging my head against a concrete wall when I try implementing something non-minecraftish though |
| 19:58 |
ireallyhateirc |
less with every release though |
| 19:59 |
ireallyhateirc |
My game is going to be a psychological horror taking place in a procedurally generated maze city |
| 19:59 |
MTDiscord |
<greenxenith> The longer you work with the engine the easier it gets to work around it |
| 19:59 |
|
___nick___ joined #minetest |
| 20:00 |
ireallyhateirc |
true, I'm doing minetest for 2 years only so far |
| 20:00 |
MTDiscord |
<greenxenith> Some of us are jaded when gauging ease of use |
| 20:12 |
ireallyhateirc |
I think blaub was thinking about MTG which left a pretty ugly legacy for Minetest's reputation, but that's not a good reason to call someone names and leave lol |
| 20:15 |
|
MTDiscord joined #minetest |
| 20:15 |
MTDiscord |
<greenxenith> I assumed as much, which is why I try to remind people that Minetest is not the game. Game mechanics dont go in an engine. |
| 20:17 |
|
kamdard joined #minetest |
| 21:05 |
|
sparky4 joined #minetest |
| 21:26 |
[MatrxMT] |
<Bram> Can somebody help me understand how the gennotify thing works during the mapgen process? |
| 21:27 |
[MatrxMT] |
<Bram> I don't understand how or why it is used. If you know that you want to generate a temple somewhere, why not build it right away and move it forward? Is there a benefit to this? Or does it do something else? |
| 21:27 |
ireallyhateirc |
afaik gennotify works like this: the engine generates something and passes that data to lua |
| 21:28 |
ireallyhateirc |
say you register some decorations like a type of a plant or something |
| 21:28 |
ireallyhateirc |
then gennotify can tell you where exactly that plant was generated |
| 21:30 |
[MatrxMT] |
<Bram> So it's more like a marker for yourself and for potential other mods that mess with the mapgen? |
| 21:31 |
ireallyhateirc |
yes |
| 21:32 |
ireallyhateirc |
I think you can also get biomemap and heightmap with it |
| 21:32 |
ireallyhateirc |
nope |
| 21:33 |
ireallyhateirc |
I confused two different things, biomemaps and heightmaps are another thing |
| 21:33 |
[MatrxMT] |
<Bram> No, that seems like something else. |
| 21:33 |
ireallyhateirc |
You can also do your own custom gennotify with the mapgen env for a custom lua mapgen |
| 21:34 |
[MatrxMT] |
<Bram> I was thinking of using it, but I haven't figured out yet how. |
| 21:34 |
ireallyhateirc |
it's quite easy, the first thing you need to define a mapgen script - a file that runs in the mapgen env |
| 21:35 |
[MatrxMT] |
<Bram> I'm currently trying to migrate my cave generation mod, which works well but is kinda slow (1s per 80x80x80 chunk) to work in an async environment that Minetest 5.9.0 now supports: git.noordstar.me/Bram/noordstar_caves |
| 21:35 |
ireallyhateirc |
you do it with minetest.register_mapgen_script("mymod/init.lua") (or some other file) |
| 21:36 |
ireallyhateirc |
then in mymod/init.lua you can do stuff and most importantly use this: minetest.register_on_generated(function(vmanip, minp, maxp, blockseed)) |
| 21:36 |
[MatrxMT] |
<Bram> I'm also trying to figure out how to share my personal global variable. I've defined a noordstar_caves.registered_biomes, noordstar_caves.registered_decorations and noordstar_caves.registered_shapes but they (naturally) don't get shared with the environment for free - unlike the standard versions of those global variables. |
| 21:36 |
ireallyhateirc |
then you can work with the voxel manipulator object and hence change map data |
| 21:37 |
ireallyhateirc |
> kinda slow (1s per 80x80x80 chunk) |
| 21:37 |
ireallyhateirc |
that's extremely slow, I could probably do that in 30ms |
| 21:37 |
[MatrxMT] |
<Bram> Yes, I'm aware on how the rest of the mapgen objects work. :) It's mostly that the gennotify documentation is kinda vague on what it does and what it's used for? |
| 21:38 |
ireallyhateirc |
I generate 100 mapchunks in 1s |
| 21:38 |
ireallyhateirc |
In Exile for example we use it to tell us where animal eggs were spawned |
| 21:38 |
[MatrxMT] |
<Bram> What are you doing on those mapchunks? |
| 21:38 |
ireallyhateirc |
Drawing roads for my city game |
| 21:39 |
ireallyhateirc |
though could be a bad example because it's 2D |
| 21:39 |
ireallyhateirc |
but I also do tasks like snow placing in the vm, worst code runs 30-100ms |
| 21:40 |
[MatrxMT] |
<Bram> Yeah, I don't think they'd compare. I need to carve cave shapes based on the registered_shapes variable, then decorate them using the registered_biomes, and then sprinkle stuff on top using registered_decorations. |
| 21:40 |
ireallyhateirc |
I'd have to look at what exactly your mod does though and see if I could rewrite it to be faster |
| 21:41 |
ireallyhateirc |
have this: https://gist.github.com/sfan5/1a124eb496a39734f53fb4daffc47bee |
| 21:42 |
|
sparky4 joined #minetest |
| 21:42 |
ireallyhateirc |
example code that uses the mapgen env |
| 21:48 |
[MatrxMT] |
<Bram> Wowie, thanks! ❤️ |
| 21:49 |
[MatrxMT] |
<Bram> I would recommend looking at API.md to get a glimpse of what it is doing: git.noordstar.me/Bram/noordstar_caves/src/branch/main/API.md |
| 21:49 |
[MatrxMT] |
<Bram> My best performance is about 1.51 calculations per node, which averages it on my laptop to about 600ms per chunk. |
| 22:02 |
ireallyhateirc |
you have your own flat array data type? |
| 22:04 |
MTDiscord |
<luatic> Bram: apply https://content.minetest.net/packages/jwmhjwmh/jitprofiler/ to find out where time is spent |
| 22:05 |
MTDiscord |
<luatic> at a glance however i already see that you're using abstractions too liberally. almost none of them are "zero cost". you need to think roughly in terms of the work your lua code results in. |
| 22:05 |
MTDiscord |
<luatic> for example you are iterating positions and turning each into an index individually. it is much faster to iterate indices with the respective strides. |
| 22:06 |
ireallyhateirc |
yeah, I can see that too. Function calls and such are expensive, I limit that to minimum in all my VM stuff |
| 22:07 |
[MatrxMT] |
<Bram> Yes. The documentation didn't seem to explain why the chunks were larger than 80x80x80. The minp and maxp did not reflect the vm size. |
| 22:07 |
sfan5 |
the vmanip contains one extra block at each side to accomodate overgrowth. |
| 22:07 |
sfan5 |
you should only fill [minp, maxp] with nodes |
| 22:08 |
[MatrxMT] |
<Bram> Even more. The vmanip contains a strip of 8 blocks on all sides. |
| 22:08 |
[MatrxMT] |
<Bram> 16, my mistake. The flat array is of length 1.404.928. |
| 22:09 |
[MatrxMT] |
<Bram> Which is 112 x 112 x 112 = (16 + 80 + 16) ^ 3 |
| 22:10 |
[MatrxMT] |
<Bram> The documentation reveals nothing about this, leaving me to debug my code for hours because my test generations wouldn't line up and always be skewered by 32 nodes per axis. |
| 22:11 |
ireallyhateirc |
that's yet another thing - mapchunk offset |
| 22:11 |
ireallyhateirc |
meaning the grid starts at xyz: -32 |
| 22:12 |
ireallyhateirc |
I guess I could take a look at the VM stuff documentation and add extra explanation to save other people hours of debugging |
| 22:16 |
[MatrxMT] |
<Bram> Keep in mind that the strip of 16 blocks always appears in at least one other chunk's generation phase again - so it's not like those strip nodes need to be addressed right away. |
| 22:18 |
[MatrxMT] |
<Bram> I will definitely try this. I might need to do some fixing on Linux for that - Minetest always gets a Segmentation fault on my Linux machine upon boot, _(using the Nix package manager)_ and I'm currently developing everything in Windows. But I'd love to use tooling like that to optimize the engine. |
| 22:22 |
ireallyhateirc |
hmmmm I'm running Guix but I have Nix installed too so I could try idk |
| 22:22 |
[MatrxMT] |
<Bram> I haven't looked into it for a while - I can come back to it when I have more context to share. Who knows, maybe it's even my fault that it doesn't run. |
| 22:24 |
ireallyhateirc |
well Nix is reproducible so it's probably the package that's broken or you're simply running it in a weird and improper way |
| 22:24 |
ireallyhateirc |
you're running NixOS or just Nix as a package manager? It's good to update package definitions btw, but I don't know Nix well enough to tell you the command |
| 22:25 |
[MatrxMT] |
<Bram> Nix as a package manager on EndeavourOS. |
| 22:26 |
[MatrxMT] |
<Bram> I am having some issues with certain apps not running due to some weird rendering issues that might be related to OpenGL. And I'm not smart enough to understand/solve those. That might be to blame for the Segmentation fault, too. |
| 22:27 |
ireallyhateirc |
use Minetest's flatpak first and check if it works. If it doesn't then something's likely wrong with drivers/firmware and not with Nix |
| 22:32 |
|
panwolfram joined #minetest |
| 22:55 |
|
sparky4 joined #minetest |
| 23:05 |
|
Eragon joined #minetest |
| 23:09 |
|
SFENCE joined #minetest |
| 23:17 |
|
Verticen joined #minetest |
| 23:44 |
|
ireallyhateirc joined #minetest |
| 23:45 |
|
lhofhansl joined #minetest |
