Time |
Nick |
Message |
01:27 |
|
DI3HARD139 joined #minetest-hub |
03:15 |
|
AndroBuilder joined #minetest-hub |
03:37 |
|
AndroBuilder joined #minetest-hub |
03:39 |
|
AndroBuilder_ joined #minetest-hub |
03:48 |
|
AndroBuilder joined #minetest-hub |
03:54 |
|
AndroBuilder joined #minetest-hub |
04:13 |
|
AndroBuilder joined #minetest-hub |
04:26 |
|
RichardTheTurd joined #minetest-hub |
05:30 |
|
RichardTheTurd joined #minetest-hub |
07:53 |
|
sniper338 joined #minetest-hub |
08:47 |
|
DI3HARD139 joined #minetest-hub |
08:58 |
|
Jordach joined #minetest-hub |
09:10 |
|
tenplus1 joined #minetest-hub |
09:10 |
tenplus1 |
hi folks |
10:03 |
shivajiva |
morning tenplus o/ |
10:04 |
tenplus1 |
hi shiva, hows things |
10:04 |
shivajiva |
good thanks, things are looking up :) |
10:04 |
tenplus1 |
glad to hear :) |
10:05 |
shivajiva |
have you enjoyed being off work? |
10:05 |
tenplus1 |
I updated a few mods earlier and added settingtypes.txt to them so you can edit .conf settings within the SETTINGS tab |
10:05 |
tenplus1 |
yeah, am finding time to do many things now |
10:05 |
shivajiva |
that sounds sweet, didn't know you could do that |
10:06 |
tenplus1 |
krock suggested it yesterday for Pova, thought I'd do the same for my other mods |
10:07 |
shivajiva |
how long has that been possible? |
10:07 |
tenplus1 |
0.4.16 stable I think |
10:09 |
shivajiva |
heh good to know it can be done :) |
10:10 |
tenplus1 |
there's prolly a lot of new features I'm unaware of in 0.4.16 still... |
10:10 |
tenplus1 |
let alone 0.5x which I haven't tested yet |
10:11 |
shivajiva |
guess we need to scour the documentation for the release and keep a closer eye on 0.5 changes =) |
10:12 |
tenplus1 |
0.4.17 has some good stuff added as well, but few ppl report issues |
10:15 |
shivajiva |
http://dev.minetest.net/Changelog |
10:17 |
tenplus1 |
? 0.4.17 isnt on there |
10:17 |
shivajiva |
I thought 0.4.17 was a mistake |
10:18 |
tenplus1 |
0.4.17 will be the next release with some of the 0.5x additions and fixes added |
10:18 |
tenplus1 |
0.5x is total dev for new features... |
10:21 |
rubenwardy |
some = 1 or 2 |
10:21 |
tenplus1 |
hi ruben |
10:24 |
tenplus1 |
https://github.com/minetest/minetest/issues/6542 |
10:28 |
tenplus1 |
I cant even find any 0.4.17 builds to test |
10:30 |
sfan5 |
there's some on my builds page |
10:30 |
tenplus1 |
hey sfan, checking now |
10:30 |
tenplus1 |
for linux ? |
10:31 |
sfan5 |
winddows |
10:31 |
sfan5 |
+d |
10:31 |
tenplus1 |
buuuu... lol... |
10:31 |
sfan5 |
.. |
10:31 |
sfan5 |
you can always compile it yourself on linux |
10:32 |
tenplus1 |
is it 0.4-backport thats officially 0.4.17 ? |
10:32 |
shivajiva |
tenplus doesn't compile builds does he? |
10:32 |
tenplus1 |
not on this old thing :P |
10:32 |
tenplus1 |
my net-top is optimized for desktop use only, not compiling |
10:33 |
shivajiva |
ten if you need a specific rev I could try but what works for me may not run on your system |
10:34 |
tenplus1 |
no worries, am just checking 0.4- backport to see what changes are in there... |
10:46 |
tenplus1 |
most of the lua changes I'm using anyhoo with mod overrides :D |
11:17 |
|
Fixer joined #minetest-hub |
11:17 |
tenplus1 |
o/ fixer |
11:26 |
Fixer |
\o |
11:28 |
|
Krock joined #minetest-hub |
11:28 |
Krock |
hi tenplus1 |
11:28 |
tenplus1 |
hey krock :) |
11:29 |
tenplus1 |
the new item_entity.lua code makes it harder to implement water flowing for items |
11:29 |
Krock |
!tell shivajiva You should consider updating the 3d_armor mod on your skyblocks server |
11:29 |
MinetestBot |
Krock: yeah, yeah |
11:30 |
Krock |
tenplus1, well yes but at least it's now better structured and supports slippery nodes (0.5.0-dev) |
11:31 |
tenplus1 |
yeah, saw taht... |
11:31 |
tenplus1 |
will squeezze it in somehow tho :D |
11:32 |
Krock |
after all you can still overwrite the entire on_step function entirely |
11:32 |
Krock |
but that will make it harder to update later |
11:32 |
tenplus1 |
I have the 0.5x version of it and will add additional features like before (water flow, move items to space if block placed on top) |
11:36 |
tenplus1 |
fixed :) w00t! |
11:44 |
tenplus1 |
slides on ice, deletes in void, burns in lava, water pushes items and if block placed on top of item it moves to closest air block :) |
11:44 |
tenplus1 |
nom time :D |
11:46 |
|
Raven262 joined #minetest-hub |
11:48 |
tenplus1 |
hey raven |
11:52 |
|
Fixer joined #minetest-hub |
11:54 |
tenplus1 |
wb |
11:54 |
Raven262 |
Hi ten |
11:54 |
tenplus1 |
:P |
11:55 |
Krock |
!tell shivajiva The creative inventory leak exists there too. I suggest patching asap :) |
11:55 |
MinetestBot |
Krock: I'll pass that on when shivajiva is around |
11:55 |
tenplus1 |
leak ? |
12:05 |
Krock |
free items from the infinite creative inventory |
12:05 |
Krock |
for everybody. was fixed months ago |
12:06 |
tenplus1 |
so certain servers arent running the udpated version ? |
12:07 |
* rubenwardy |
sighs |
12:07 |
rubenwardy |
https://forum.minetest.net/viewtopic.php?p=317365#p317365 |
12:07 |
|
longerstaff13 joined #minetest-hub |
12:07 |
tenplus1 |
enrico causing hassle on other servers now |
12:07 |
tenplus1 |
hi longer |
12:07 |
longerstaff13 |
Hi |
12:08 |
Krock |
tenplus1, well yeah. it's not contained in the last stable so there are surely many unpatched servers there |
12:09 |
Krock |
s/s there/s online/ |
12:09 |
Krock |
tenplus1, they're on #minetest btw :P |
12:09 |
tenplus1 |
who is, enrico ? |
12:10 |
Krock |
rubenwardy, http://3.bp.blogspot.com/-vgsccj-2ANU/VTFnvs4dLdI/AAAAAAAAPcE/IizjXwM4UPo/s1600/facebook-thumbs-star-wars-luke-skywalker.jpg |
12:10 |
Krock |
tenplus1, yes |
12:10 |
tenplus1 |
glad I dont go in there :P |
12:10 |
tenplus1 |
had enough of him on Xanadu |
12:10 |
rubenwardy |
he literally said "Hi" |
12:10 |
Krock |
Quality player |
12:12 |
tenplus1 |
heh |
12:29 |
tenplus1 |
github.com/tenplus1/builtin_item updated |
12:29 |
|
Mr_Pardison joined #minetest-hub |
12:30 |
tenplus1 |
hi mister |
12:30 |
Mr_Pardison |
ahoy |
12:30 |
ThomasMonroe |
:P |
12:30 |
tenplus1 |
hi thomas |
12:30 |
ThomasMonroe |
hi tenplus1 |
12:52 |
|
aerozoic joined #minetest-hub |
12:52 |
tenplus1 |
hi aerozoic |
12:52 |
aerozoic |
wazup tenplus1 |
12:52 |
tenplus1 |
o/ |
13:01 |
shivajiva |
thanks krock :D |
13:01 |
MinetestBot |
shivajiva: Apr-27 11:29 UTC <Krock> You should consider updating the 3d_armor mod on your skyblocks server |
13:01 |
MinetestBot |
shivajiva: Apr-27 11:55 UTC <Krock> The creative inventory leak exists there too. I suggest patching asap :) |
13:39 |
Krock |
np shivajiva. The latter one can be patched using the backport-0.4 MTG branch or applying the patch manually |
13:41 |
shivajiva |
think I'll go for applying the patch manually at this stage, don't want to update everything just yet |
13:48 |
shivajiva |
presumably this is #1840 |
13:48 |
ShadowBot |
https://github.com/minetest/minetest/issues/1840 -- Singlenode: Failed to resolve 'mapgen_water_source' |
13:57 |
Krock |
yes, game#1840 |
13:57 |
ShadowBot |
https://github.com/minetest/minetest_game/issues/1840 -- Creative: Prevent unauthorized item access by SmallJoker |
13:59 |
shivajiva |
heh wondered how to do that :P |
14:32 |
|
sniper570 joined #minetest-hub |
14:33 |
|
sniper570 joined #minetest-hub |
14:37 |
|
Mr_Pardison joined #minetest-hub |
14:37 |
Mr_Pardison |
hai |
14:37 |
tenplus1 |
hi mister :P |
14:38 |
ircSparky |
is minetest.find_path heavy on ram or CPU? |
14:39 |
rubenwardy |
more RAM |
14:39 |
rubenwardy |
as you need to read the map, and also store an open list |
14:39 |
tenplus1 |
has error message been fixed in chat for pathfinding when it cant find one ? |
14:39 |
rubenwardy |
but it's blocking |
14:44 |
ircSparky |
do you know offhand if open_ai is much heavier than mobs redo? |
14:45 |
tenplus1 |
open_ai works well :) but is very simple... mobs redo api has many more features to use but can be tailored to run fast |
14:46 |
ircSparky |
does mobs redo have advanced pathfinding? |
14:46 |
ircSparky |
open_ai disnt have any hostile mobs, and thats what i need, but I need mobs that are harder to loose |
14:46 |
tenplus1 |
yes, has 2 level pathfinding... level 1 finds player through mazes... level 2 breaks walls/builds up to player etc |
14:47 |
ircSparky |
does i tneed to be enabled? or just the version I have dosnt haveit |
14:48 |
tenplus1 |
it's there, in mob registration you set pathfinding level |
14:49 |
ircSparky |
ok |
14:49 |
ircSparky |
ill try it out |
14:50 |
tenplus1 |
if you get stuck check out the example .lua's for monsters and animals, or look at api.txt |
14:50 |
ircSparky |
do most monsters haveleve 1? |
14:50 |
tenplus1 |
some of the smarter one's... the majority use no pathfinding and dumb directional to follow player |
14:52 |
ircSparky |
ok which ones are the smart ones? :P |
14:53 |
tenplus1 |
sand/stone monster and oerkki |
14:54 |
ircSparky |
I mean, ill mostly have zombies, but zombies are usually know how to properly turn a corner :P |
14:54 |
ircSparky |
kk |
14:54 |
tenplus1 |
we have a mob called "Evil Bonny" on Xanadu that has level 2 pathfinding and kills everyone she's near |
14:58 |
Mr-Pardison |
tenplus1: what kind of weapons are there on Xanadu? |
14:58 |
rubenwardy |
contributions welcome to fix the engine pathfinder |
14:59 |
rubenwardy |
also, would be awesome to make a parallel vesion |
14:59 |
tenplus1 |
defaults, crystal, lava pick, super axe |
15:00 |
Mr-Pardison |
no spears? |
15:01 |
tenplus1 |
bows n arrows |
15:01 |
tenplus1 |
snowballs :D eggs |
15:03 |
tenplus1 |
players can use monsters as weapons also... sneaky players |
15:06 |
Mr-Pardison |
sounds like something I would do :) |
15:07 |
tenplus1 |
ehehe |
15:08 |
tenplus1 |
I made 5x sky islands with a water source and a different tree for each... so far 4 players are trying to survive on them :D |
15:08 |
Mr-Pardison |
is that 5th island claimed yet? |
15:09 |
tenplus1 |
nah, u want ? |
15:09 |
Mr-Pardison |
sure. |
15:10 |
Mr-Pardison |
try a survival server other than DL (need to diversify my experiences). |
15:11 |
tenplus1 |
am there now if you wanna join I'll to ya |
15:11 |
Mr-Pardison |
k |
15:12 |
Mr-Pardison |
what's the port for it? (my client refuses to load any other server names than what I've joined) |
15:13 |
Mr-Pardison |
nvr mind. |
15:13 |
tenplus1 |
lol |
15:27 |
rdococ |
Yayâ„¢ |
15:30 |
tenplus1 |
hi rdococ |
15:34 |
ircSparky |
so, the mobs_monsters are able to follow a player through a maze? |
15:35 |
ircSparky |
or you need smart_mobs? |
15:35 |
tenplus1 |
mobs redo api has pathfinding that allows a mob to find a player through mazes... |
15:35 |
ircSparky |
hm |
15:35 |
tenplus1 |
but of a set size :) |
15:35 |
tenplus1 |
dont want them super OP |
15:36 |
ircSparky |
yea |
15:37 |
ircSparky |
https://www.youtube.com/edit?o=U&video_id=FUZ0GOWFj8I |
15:37 |
ircSparky |
they dont seem to be working the best for me |
15:37 |
ircSparky |
idk why |
15:37 |
tenplus1 |
lol |
15:38 |
rdococ |
Huh, I didn't know Lua 5.4 was ut |
15:38 |
ircSparky |
um |
15:38 |
rdococ |
s/ut/out/ |
15:38 |
ircSparky |
wups |
15:38 |
tenplus1 |
only attacking mobs use pathfinding... so make a glass maze 3x high so it cant jump over... punch sand monster outside and quikly go inside, but leave an entrance for them to find |
15:38 |
rdococ |
I'm gonna read its changelog |
15:38 |
ircSparky |
https://www.youtube.com/watch?v=FUZ0GOWFj8I |
15:39 |
ircSparky |
ah ok |
15:39 |
ircSparky |
lemme try that |
15:39 |
rdococ |
tenplus1: I think passive mobs could do with pathfinding too |
15:39 |
|
twoelk joined #minetest-hub |
15:39 |
rdococ |
well, some of them, anyway |
15:40 |
tenplus1 |
hi twoelk |
15:40 |
twoelk |
o/ |
15:40 |
rdococ |
Also, when you attack a chicken, other chickens should be spawned that attack you like in TLoZ :P |
15:40 |
tenplus1 |
once pathfinding and line of sight functions improve then all mobs could ise them |
15:40 |
tenplus1 |
lol rdococ, chickens and rabbits runaway from player, unless you are owner |
15:40 |
rdococ |
tenplus1: so chickens are chicken? |
15:41 |
rdococ |
There could be a rare deer mob that drops cool stuff, and an even rarer unicorn mob that drops a unicorn horn |
15:41 |
rdococ |
Both would run away from players if they don't hold sneak |
15:42 |
* twoelk |
wonders wether mobs could deteckt nodes placed by players and avoid these as if they smelled dangerous as animals may do in real life |
15:43 |
ircSparky |
so when do sand monsters and such go into pathfinding mode? |
15:44 |
tenplus1 |
when they attack a player and a timer goes off saying it'd been idle too long |
15:45 |
rdococ |
Certain mobs could be drawn towards artificial light sources |
15:46 |
twoelk |
bsssssssszzzzzzz |
15:46 |
rdococ |
twoelk: Imagine a horde of zombies that sees someone's torchlight in the distance . . . |
15:46 |
ircSparky |
ok |
15:46 |
ircSparky |
sounds would be nice too |
15:46 |
tenplus1 |
yeah... was thining of redoing walking cycle to move towards certain blocks/light etc |
15:46 |
ircSparky |
if they could hear sounds |
15:46 |
ircSparky |
yea the maze test worked |
15:47 |
twoelk |
ah, my circling sharky jellyfish idea arises again |
15:47 |
tenplus1 |
:P |
15:48 |
rdococ |
Zombies would be attracted to artificial light sources as they have a higher chance of finding players there, those weird desert skeleton things could maybe be able to 'smell' blocks that are artificial and don't normally spawn in their biome (e.g. wood) |
15:48 |
rdococ |
Hehe, that would be funny because skeletons and probably weird desert skeleton things don't have noses |
15:49 |
tenplus1 |
am trying to think of a way to spawn mobs without using ABM's... |
15:49 |
twoelk |
https://forum.minetest.net/viewtopic.php?p=241321#p241321 |
15:50 |
tenplus1 |
ooh jellyfish :P |
15:50 |
rdococ |
Maybe zombies or jellyfish could change their behavior based on whether they're alone or in a horde. E.G. a lone zombie probably would run away from an armored player, but a horde would move towards them |
15:50 |
tenplus1 |
we have fish and sharks handy for now, might be worth adding those nd turtles |
15:54 |
ircSparky |
would it be heavy for each mob to check line of sight every second or half second? |
15:55 |
rdococ |
Odd, I can't tell if Lua 5.4 is out or not |
15:55 |
twoelk |
some sort of situation related herd behaviour that can react to changes would be really cool, although pretty creepy as it might get close to resembling an ai |
15:55 |
rubenwardy |
no |
15:55 |
rubenwardy |
line of sight is pretty optimised |
15:55 |
rdococ |
Each mob should store a detailed neural network |
15:56 |
ircSparky |
totally |
15:56 |
ircSparky |
ok |
15:57 |
tenplus1 |
if only :P |
15:57 |
rdococ |
GMO crops mod: you'd be able to splice DNA and create new crops |
15:57 |
ircSparky |
tenplus1, should i try to make the level 1 pathfinding use line of sight instead of a timer? or maybe make a new level? or do you want to code it yourself? |
15:57 |
twoelk |
can one querry how many mobs are active at any given time on a running map? |
15:59 |
tenplus1 |
api uses line of sight already |
15:59 |
ircSparky |
oh, i thought you said it used a timer :P |
16:00 |
rdococ |
yay, line of sightâ„¢ |
16:00 |
tenplus1 |
it uses all sorts, line of sight for attack and pathfinding |
16:03 |
ircSparky |
hm |
16:03 |
ircSparky |
do you know what the main difference between it and open_ai pathfinding system is? |
16:05 |
tenplus1 |
open_ai pathfinds towards certain blocks only, not players |
16:06 |
ircSparky |
so, the block the player is standing on? |
16:06 |
ircSparky |
or |
16:06 |
ircSparky |
in |
16:07 |
tenplus1 |
open_ai has sheep for instance look for grass blocks or wheat nodes to move around... it heads towards areas a sheep is at home :P |
16:07 |
ircSparky |
ah |
16:07 |
tenplus1 |
in mobs redo it's up to the mob where it wants to go, only using pathfinding to find player |
16:07 |
ircSparky |
well most of the npcs also follow players when they hold dry grass |
16:07 |
tenplus1 |
but I was gonna add a way to head towards certain blocks in api at some point |
16:09 |
ircSparky |
you must not have seem my comparison video? |
16:09 |
tenplus1 |
you made a video ? |
16:09 |
ircSparky |
i mean, I guess its not very polite of me to just paste youtube links |
16:09 |
ircSparky |
https://www.youtube.com/watch?v=FUZ0GOWFj8I |
16:09 |
ircSparky |
im not rickrolling you though :P |
16:09 |
tenplus1 |
heh |
16:10 |
ircSparky |
the first time works for the stone monster |
16:10 |
ircSparky |
not the rest though |
16:10 |
ircSparky |
and i have one trying to get to me in that exact spot right now :P |
16:11 |
|
Jordach joined #minetest-hub |
16:12 |
tenplus1 |
hi Jordach |
16:12 |
tenplus1 |
ircSparky: pathfinding needs work :P yeh |
16:12 |
|
Mr_Pardison joined #minetest-hub |
16:12 |
tenplus1 |
wb mister |
16:13 |
ircSparky |
yea |
16:13 |
Mr_Pardison |
gracias. |
16:13 |
ircSparky |
but of course, open_ai has the best pathfinding ive ever seen |
16:14 |
ircSparky |
so it might be unfair to compare with mobs_redo |
16:14 |
Mr_Pardison |
but have you seen closed_ai? :P |
16:14 |
ircSparky |
:P |
16:15 |
tenplus1 |
lol |
16:15 |
ircSparky |
sometimes they get stuck in the middle of a open area though |
16:15 |
ircSparky |
and it would be nice if they could get through doors |
16:15 |
tenplus1 |
yeah doors kinda suck for that |
16:16 |
ircSparky |
does find_path treat them as solid nodes? |
16:16 |
tenplus1 |
yeah |
16:16 |
tenplus1 |
pathfinding really needs a list of nodes that it can ignore (doors, slabs etc) |
16:16 |
Mr_Pardison |
IMO if they went through doors, then that would defeat the purpose of them at night |
16:16 |
|
Darcidride_ joined #minetest-hub |
16:16 |
ircSparky |
yea |
16:17 |
ircSparky |
and "open" doors can be used to close a different direction |
16:17 |
tenplus1 |
hi Darcidride_ |
16:18 |
|
Fixer joined #minetest-hub |
16:18 |
tenplus1 |
wb fixer |
16:21 |
Fixer |
ty |
16:21 |
tenplus1 |
open_ai has a newer framework which is really good... mobs redo still has compatibility with the older simple mobs framework |
16:21 |
tenplus1 |
if features could be ported across :)) |
16:22 |
ircSparky |
ah, nice |
16:22 |
ircSparky |
I think its the most used mob framework |
16:23 |
tenplus1 |
for ease of use, yes... features, yes... but I still want it to be better |
16:23 |
|
benrob0329 joined #minetest-hub |
16:23 |
tenplus1 |
hi benrob |
16:23 |
benrob0329 |
Hi tenplus1 |
16:24 |
tenplus1 |
o// |
16:24 |
benrob0329 |
I definitely didn't stay up way too late last night making this thing https://gitlab.com/benrob0329/simpdown |
16:25 |
tenplus1 |
heh |
16:25 |
benrob0329 |
But hey, now I don't have to write my website in html |
16:25 |
tenplus1 |
even better :P |
16:26 |
tenplus1 |
back soon, making noms :P |
16:29 |
|
Krock joined #minetest-hub |
16:31 |
tenplus1 |
wb |
16:31 |
Krock |
ty |
16:35 |
rdococ |
Lua 5.4 is in developyayâ„¢ |
16:36 |
tenplus1 |
oooh |
16:36 |
tenplus1 |
mt still doesnt use 5.3 yet |
16:37 |
rdococ |
I'm pretty sure Minetest doesn't use 5.2 yet, nevermind 5.3 |
16:38 |
tenplus1 |
ehehe |
16:38 |
rubenwardy |
Lua 5.2+ sucks |
16:38 |
tenplus1 |
how come ? |
16:38 |
rubenwardy |
No LuaJIT support |
16:38 |
tenplus1 |
buuuu |
16:39 |
rdococ |
does Lua 4.0 have LuaJIT? |
16:40 |
sofar |
luajit is only 5.1 I think |
16:40 |
|
Krock joined #minetest-hub |
16:55 |
benrob0329 |
isn't that more the fault of LuaJIT, and not Lua? |
16:59 |
tenplus1 |
they wont admit to it :D |
16:59 |
rdococ |
X sucks because Y refuses to support itâ„¢ |
17:00 |
rdococ |
Totally makes senseâ„¢ |
17:01 |
rubenwardy |
LuaJIT hasn't got any maintainers anyway |
17:01 |
tenplus1 |
o.O open_ai spawning is intruiging... it has a 70 second timer which then loops through every player and then loops through every mob to do a find_nodes_in_area(20 radius), choose random node and spawn mob... |
17:01 |
rubenwardy |
and for me, LuaJIT is Lua |
17:01 |
tenplus1 |
you'd think that 5.3 would do away with separate luajit and just have an option to enable |
17:02 |
rdococ |
rubenwardy: So for you, Lua has no maintainers? :P |
17:02 |
rubenwardy |
basically |
17:02 |
rdococ |
hm, does Minetest require luajit? |
17:03 |
benrob0329 |
No |
17:03 |
|
longerstaff13 joined #minetest-hub |
17:03 |
tenplus1 |
o/ longer |
17:03 |
benrob0329 |
In fact, numerous problems tend to occur when its compiled with LuaJIT on heavy servers |
17:03 |
tenplus1 |
the 2mb limit ? |
17:04 |
rubenwardy |
that's because GC64 isn't stable yet |
17:04 |
Mr_Pardison |
tenplus1: does the latest release of mobs_redo prevent mobs from glitching out through nodes? |
17:04 |
longerstaff13 |
o/ |
17:04 |
Mr_Pardison |
\o |
17:05 |
tenplus1 |
it tries to... the mt engine itself has a glitch where entities use stepheight to glitch through solid walls |
17:05 |
rdococ |
I wonder if Minetest supports Lua 5.2+, seeing as it doesn't require LuaJIT |
17:06 |
benrob0329 |
So using something that tends to crash the game, has very few maintainers, refuses to support new versions of the standard, and who's added speed boost does nothing for most mods is worth not supporting Lua 5.2+ ? |
17:06 |
benrob0329 |
That sounds like some bs right there, ngl |
17:06 |
rubenwardy |
doesn't crash that much |
17:06 |
rubenwardy |
if your mods aren't bloated |
17:07 |
benrob0329 |
Unless you use one if the heavy mods its supposed to help make faster |
17:07 |
rubenwardy |
I'd rather have the 64x speed boost |
17:07 |
rdococ |
> does nothing for most mods |
17:07 |
rdococ |
> 64x speed boost |
17:07 |
rdococ |
does not compute |
17:07 |
benrob0329 |
For what, defining nodes? |
17:07 |
rdococ |
globalsteps, ABMs |
17:07 |
rubenwardy |
heavy on memory :P |
17:08 |
tenplus1 |
how do I tell minetest to use lua isntead of luajit ? |
17:08 |
rdococ |
(although, to be fair you should be avoiding globalsteps whenever possible anyway) |
17:08 |
benrob0329 |
ABMs are used rarely anymore, and globalsteps need to be lightweight anyways |
17:08 |
benrob0329 |
tenplus1: compile time option |
17:09 |
tenplus1 |
whats different between using globalstep and mintest.after to run same function ? |
17:09 |
tenplus1 |
benrob0329: damn, was gonna test a game without jit to see |
17:09 |
benrob0329 |
Globalstep is repeated every step |
17:09 |
benrob0329 |
After is only repeated once |
17:09 |
rdococ |
^ |
17:09 |
benrob0329 |
Er, done once |
17:09 |
rdococ |
essentially |
17:10 |
tenplus1 |
so even if I setup a globalstep with a 1 second timer wouldnt that help ? |
17:10 |
rdococ |
globalsteps don't have a timer parameter so if you want a timer you'll have to keep track of it yourself |
17:10 |
benrob0329 |
It will only run the code once a second, but will repeat the check every step |
17:10 |
rubenwardy |
I've never had an OOM with LuaJIT |
17:10 |
tenplus1 |
timer = 0 ; globalstep function ; timer = timer + dtime ; if timer < 1 then return end ; timer = 0 ; do something |
17:11 |
rdococ |
minetest.after is better for things that don't run every step or so, because it minimizes the number of function calls per step |
17:11 |
benrob0329 |
after() just delays a function |
17:11 |
rdococ |
^ |
17:11 |
benrob0329 |
rubenwardy: tagged: WORKSFORME |
17:11 |
rdococ |
but you can define a function that uses minetest.after to call itself |
17:11 |
rubenwardy |
basically |
17:11 |
rdococ |
resulting in a loop |
17:11 |
tenplus1 |
yeah, did that for stamina poisoning |
17:12 |
benrob0329 |
Ive done it for smooth Alpha transitions :P |
17:12 |
tenplus1 |
:P |
17:12 |
rdococ |
I like exhaustion poisoning (takes 1 point of exhaustion away every second or so) because it makes you less tired |
17:13 |
rdococ |
rubenwardy: hm, what do you think about a CTF map that takes place in space with planets? players would have to use their planet's resources to create bridges from planet to planet |
17:13 |
rubenwardy |
interesting |
17:13 |
rubenwardy |
maybe islands would be better |
17:13 |
rdococ |
but islands aren't in space |
17:13 |
rubenwardy |
or float lands |
17:13 |
tenplus1 |
sounds a good idea, can you do planetoids with biomes ? |
17:13 |
rubenwardy |
MT can't do space |
17:13 |
sofar |
everything is in space, technically |
17:13 |
benrob0329 |
Idea for sprint: you always sprint, but your speed decreases with stamina (and stamina with food) |
17:13 |
rdococ |
Floatlands might be a good compromise |
17:14 |
rdococ |
sofar: technicallyâ„¢ |
17:14 |
rubenwardy |
you'd expect a vacuum and low G |
17:14 |
tenplus1 |
lol |
17:14 |
benrob0329 |
Technically everything is moving too |
17:14 |
tenplus1 |
pova's base overrides can set gravity to a lower value :) |
17:14 |
rdococ |
benrob0329: technically everything can be said to be moving from one reference frame and not from its own reference frame |
17:14 |
tenplus1 |
could wear say antigrav boots to compensate |
17:14 |
rdococ |
s/one/at least one/ |
17:14 |
rdococ |
CTF space map: everyone suffocates repeatedly upon spawning |
17:15 |
tenplus1 |
haha |
17:15 |
benrob0329 |
Lol |
17:15 |
Mr_Pardison |
XD |
17:15 |
tenplus1 |
brb |
17:15 |
rdococ |
rubenwardy: although, I wonder how low gravity would affect the game |
17:16 |
rdococ |
PvP would essentially have a third dimension to it as you have to work harder to aim up and down at the enemy |
17:16 |
benrob0329 |
Eh, just jump over peopl |
17:17 |
rdococ |
They could jump up to you |
17:17 |
benrob0329 |
We don't have collision thougb |
17:17 |
rubenwardy |
would be interesting |
17:17 |
rubenwardy |
Minetest won't like it though |
17:17 |
rdococ |
benrob0329: but then they can reach you to hit you |
17:17 |
rubenwardy |
maybe an underwater map would be cool too |
17:17 |
rubenwardy |
with gas bubbles |
17:18 |
rubenwardy |
and bases so far apart you need to build glass tunnels |
17:18 |
Mr_Pardison |
methane bubbles? :P |
17:18 |
benrob0329 |
*boom* |
17:18 |
rdococ |
gas bubbles could be implemented as walk-thru translucent nodes |
17:18 |
sofar |
cosmic rays constantly bombarding players with radiation making them mutate slowly altering their physical properties randomly |
17:19 |
benrob0329 |
Thats morbid, but also really cool sounding |
17:19 |
sofar |
it'll kill players over time, but they also might get stronger or faster |
17:20 |
benrob0329 |
Or die instantly from radiation poisoning |
17:20 |
rdococ |
Or you get cancer and die slowly and painfully |
17:20 |
sofar |
instantly? |
17:21 |
sofar |
whatever happened to slow, painful deaths? |
17:21 |
rdococ |
I mentioned them |
17:21 |
rdococ |
;) |
17:21 |
rubenwardy |
cancer -> disables hp regen |
17:21 |
benrob0329 |
cancer -> you get teleported to a Roblox server |
17:22 |
Krock |
cancer -> you're suddenly cured |
17:22 |
Mr_Pardison |
cancer -> discord. |
17:22 |
Mr_Pardison |
no wait. |
17:23 |
Mr_Pardison |
cancer -> fortnite -> hype |
17:23 |
rdococ |
lol |
17:24 |
rdococ |
When people say "XXX is cancer" without elaborating as to why, it bugs mwe |
17:24 |
rdococ |
s/mwe/me/ |
17:24 |
rdococ |
for example, I don't think fortnite or roblox ever caused people to die slow painful deaths or caused cells to multiply out of control |
17:24 |
rdococ |
okay, roblox might have but idk about fortnite |
17:25 |
Mr_Pardison |
fortnite is all hype. |
17:25 |
Mr_Pardison |
ergo, cancer. |
17:26 |
rdococ |
I'm pretty sure fortnite is a game of some sort, meaning that it is not literally composed of hype. I think you're talking metaphorically, but I don't completely understand so can you elaborate a little further? |
17:26 |
rubenwardy |
fortnite is pretty fun though |
17:26 |
rubenwardy |
basically a rip off of CTF |
17:26 |
tenplus1 |
seen it being played on yt, not bad |
17:27 |
Mr_Pardison |
I don't get ppls' obsession over it when it's just your basic build and fight game with fancy graphics and weapons. |
17:28 |
rdococ |
I'm looking on Wikipedia |
17:28 |
rdococ |
apparently Fortnite is a paid sandbox survival game while Battle Royale is a gamemode that can also be gotten for free as a standalone game? |
17:29 |
rdococ |
it wasn't acquired by Microsoft, so it's already better than YKW |
17:31 |
tenplus1 |
Q. should mobs despawn when stepping on "ignore" node ? |
17:32 |
|
rubenwardy joined #minetest-hub |
17:32 |
rdococ |
tenplus1: nah. the only way that could happen is if a mob was on top of an unloaded mapblock boundary and I don't think it should be removed just because the ground it's on isn't loaded yet |
17:39 |
rubenwardy |
huh |
17:40 |
rubenwardy |
I got 17.10 not 18.04 |
17:40 |
rubenwardy |
odd |
17:40 |
rdococ |
I have 18.04 |
17:40 |
tenplus1 |
ditto |
17:40 |
rdococ |
well, xubuntu 18.04 |
17:40 |
rubenwardy |
I ran do-release-upgrade, but it didn't give me 18.04. Hasn't it been released? |
17:40 |
rdococ |
speaking of which, I think 18.04 was released yesterday |
17:40 |
rubenwardy |
-d makes it find it |
17:40 |
rubenwardy |
odd |
17:40 |
tenplus1 |
rubenwardy: sudo update-manager -d |
17:41 |
rdococ |
maybe look to see what the -d flag does |
17:41 |
rubenwardy |
this is on a server |
17:41 |
rubenwardy |
d means development version |
17:41 |
tenplus1 |
and only works when current setup is up to date |
17:44 |
rubenwardy |
"The upgrade has aborted. The upgrade needs a total of 117 M free |
17:44 |
rubenwardy |
space on disk '/boot'. Please free at least an additional 35.5 M of |
17:44 |
rubenwardy |
disk space on '/boot'. You can remove old kernels using 'sudo apt |
17:44 |
rubenwardy |
autoremove' and you could also set COMPRESS=xz in |
17:44 |
rubenwardy |
/etc/initramfs-tools/initramfs.conf to reduce the size of your |
17:44 |
rubenwardy |
initramfs. " |
17:44 |
rubenwardy |
ew |
17:44 |
rubenwardy |
oops |
17:44 |
rubenwardy |
stupid new lines |
17:44 |
tenplus1 |
do you add your own /boot partition ? |
17:45 |
rubenwardy |
no, from host |
17:45 |
rubenwardy |
which is stupid for a server, as you'd expect to find a lot on /var/www |
17:47 |
rubenwardy |
/ also only has 15GB |
17:47 |
rubenwardy |
/home has 985GB |
17:47 |
rubenwardy |
was wondering why those lines weren't sending ^ |
17:47 |
tenplus1 |
eek... I set / to 50gb and the rest to /home |
17:47 |
rubenwardy |
that would be logical |
17:49 |
|
Raven262 joined #minetest-hub |
17:50 |
tenplus1 |
wb raven |
17:50 |
Raven262 |
ty ten |
17:52 |
* rdococ |
updates his operating systemâ„¢ |
17:53 |
tenplus1 |
Ubuntu 18.04 final release was yesterday :) |
17:53 |
Krock |
quick! distro upgrade! |
17:54 |
* Mr_Pardison |
levels up his distro |
17:54 |
tenplus1 |
:) |
17:54 |
rdococ |
I'm a time traveller from the future |
17:54 |
rdococ |
Linux 5.0.0 is all the rage |
17:56 |
rdococ |
in 2030, GNU acquires Mojang from Microsoft at the end of an incredibly unlikely chain of events |
17:57 |
tenplus1 |
lol |
17:57 |
ircSparky |
is Elon Musk the supreme leader of Mars? |
17:57 |
Mr_Pardison |
and how would that snowball start? |
17:57 |
Mr_Pardison |
ircSparky: one day he might be. |
17:57 |
Mr_Pardison |
but I'd take over and give it to the ppl for free :D |
17:57 |
tenplus1 |
and Trump finally builds his wall.... on the moon |
17:58 |
ircSparky |
does he deport mexicans to the moon? |
17:58 |
rdococ |
a wall wouldn't work anyway |
17:58 |
tenplus1 |
lol yeah... they'd just get a cheap flight to USA for holiday and not go back :P |
17:58 |
rdococ |
there are these magic white birds in the sky that can carry tons of people, they're called "airplanes" |
17:59 |
ircSparky |
yea we have those |
17:59 |
rdococ |
time to restartâ„¢ |
18:04 |
tenplus1 |
https://libreboot.org/ |
18:07 |
rdococ |
I couldn't find installation instructions specific to my computer/its bootloader/I'm just stupid |
18:07 |
sofar |
MS doesn't sign the UEFI bootloader |
18:07 |
sofar |
MS signs the windows loader |
18:08 |
sofar |
and the MS pubkey is used by UEFI to validate that the windows loader was signed by it |
18:08 |
rdococ |
ah |
18:08 |
sofar |
but UEFI itself isn't signed |
18:08 |
* rdococ |
bashes his head against a wall |
18:08 |
sofar |
take out the keys from UEFI, and you're trusting nobody |
18:08 |
sfan5 |
intel employee defends microsoft |
18:08 |
sfan5 |
classy |
18:08 |
sfan5 |
/s |
18:08 |
sofar |
troll |
18:08 |
sofar |
you know well how it works |
18:09 |
rdococ |
I have secure boot disabled on this laptop |
18:09 |
sfan5 |
don't you see the /s I put |
18:09 |
sofar |
btw it's your hardware vendor that puts the UEFI firmware on the motherboard |
18:09 |
sofar |
I do, but I made sure others also know ;) |
18:10 |
rdococ |
I'm not that familiar with bootloaders |
18:10 |
sofar |
UEFI without secure boot is just fine, as long as you trust your motherboard firmware vendor |
18:10 |
sofar |
you can also put in your own keys |
18:10 |
sofar |
and prevent someone from possibly tampering with your OS |
18:10 |
sofar |
at least, you can detect it that way |
18:12 |
sofar |
btw most of UEFI is open source and you can see how it works (https://www.tianocore.org/) |
18:12 |
rdococ |
Unicorns |
18:12 |
sofar |
almost all EFI implementations are based on tianocore |
18:13 |
rdococ |
rubenwardy: are there size restrictions for CTF maps? |
18:21 |
Calinou |
↠proud user of disabled Secure Boot :p |
18:21 |
tenplus1 |
lol, hi Cal |
18:21 |
Calinou |
thankfully, my hardware still lets me do it |
18:22 |
Calinou |
(there's reports of some motherboard models not allowing disabling Secure Boot, even on x86) |
18:22 |
Calinou |
my old laptop required putting an UEFI password first in order to disable Secure Boot |
18:22 |
Calinou |
the password I set was "123" :-) |
18:23 |
rubenwardy |
rdococ: 225 max |
18:24 |
rubenwardy |
225*y*225 |
18:24 |
tenplus1 |
am prolly 1x bios update away from uefi forever |
18:24 |
rubenwardy |
That's r < 115 |
18:25 |
rdococ |
what about y? |
18:26 |
rdococ |
rubenwardy: what about y? |
18:26 |
rubenwardy |
Err, don't go mad |
18:26 |
|
Mr_Pardison joined #minetest-hub |
18:27 |
Mr_Pardison |
ty. |
18:27 |
rubenwardy |
It's usually smaller though |
18:28 |
rdococ |
I meant what 'y' is in '225*y*225' |
18:28 |
rubenwardy |
yeah |
18:28 |
rdococ |
I'm mad as in insane but not as in angry :P |
18:29 |
rubenwardy |
there's no coded limit unlike X and Z |
18:29 |
tenplus1 |
y = 37.952 |
18:29 |
rubenwardy |
but bigger values will increase load time |
18:29 |
rubenwardy |
so maybe max 200 |
18:29 |
rdococ |
ha |
18:29 |
rdococ |
s/ha/ah/ |
18:29 |
rubenwardy |
but that should be rare, and you should justify it |
18:29 |
rubenwardy |
and I will also need to make sure it's not too slow |
18:30 |
rubenwardy |
two hills has a y of 140 |
18:30 |
rubenwardy |
bridge is the same |
18:34 |
rdococ |
I'm trying to figure out how to set a world to use 'dummy' backend |
18:34 |
rdococ |
I don't have a 'minetestserver' file I can execute and using './minetest --migrate [blah blah]' doesn't seem to work |
18:35 |
rubenwardy |
literally just change backend in world.mt to dumy |
18:35 |
rubenwardy |
*dummy |
18:36 |
rdococ |
ah, world.mt didn't exist until I opened the world. I think. |
18:37 |
rdococ |
or I'm just an idiot |
18:37 |
rubenwardy |
set up windows builds on my CI for my game |
18:38 |
rubenwardy |
source/server/worker/threadworker.hpp:34:8: error: ‘mutex’ in namespace ‘std’ does not name a type |
18:38 |
rubenwardy |
silly Ubuntu being outdated :'( |
18:38 |
rubenwardy |
hopefully the update will fix it |
18:38 |
rubenwardy |
normal gcc works fine |
18:38 |
rubenwardy |
so it's just mingw that's outdated |
18:43 |
|
Mr-Pardison joined #minetest-hub |
18:50 |
|
rubenwardy joined #minetest-hub |
18:51 |
ircSparky |
im looking through the mob_redo api |
18:51 |
tenplus1 |
:P |
18:52 |
ircSparky |
if self.hornytimer > 240 then |
18:52 |
ircSparky |
self.child = false |
18:52 |
ircSparky |
wtf |
18:52 |
tenplus1 |
mobs can be bred, have children :) |
18:52 |
ircSparky |
lo |
18:52 |
ircSparky |
l |
18:52 |
rdococ |
Yayâ„¢ |
18:52 |
rubenwardy |
woo 18.04 on my server! |
18:52 |
rubenwardy |
lol! |
18:52 |
tenplus1 |
nice one rubem :P |
18:52 |
rdococ |
so if a young mob becomes horny enough he becomes an adult? |
18:53 |
sofar |
that code is wrong on many levels |
18:53 |
rdococ |
sofar: exactly what I was thinking |
18:53 |
sofar |
who wrote it? |
18:53 |
tenplus1 |
lol, self.hornytimer controls 3 different things, hearts above mod meaning it can breed, timer until ti can breed again, and child growing up status |
18:53 |
rdococ |
I feel like that might add several more levels of wrong |
18:54 |
tenplus1 |
lolol, it's just a variable name :D |
18:54 |
|
Jordach joined #minetest-hub |
18:54 |
tenplus1 |
wb Jordach |
18:54 |
rdococ |
so the mob becoming an adult is equivalent to it being able to breed for the first time? |
18:54 |
sofar |
seriously though, variable names *matter* |
18:54 |
rdococ |
and the mob becomes an adult by getting really horny |
18:54 |
tenplus1 |
sofar: agree'd, and this means what it means :P |
18:54 |
sofar |
do you honestly think people will be able to read your code objectively if you label your variables "swastika" and "redskin" ? |
18:55 |
tenplus1 |
why would I use those ? |
18:55 |
sofar |
whatever happened to "fertile" |
18:55 |
rdococ |
I don't think "hornytimer" is on the same level as wrong as "swastika" and "redskin" |
18:55 |
tenplus1 |
heh... my code is commented so you know exactly what's happening |
18:55 |
sofar |
look |
18:55 |
rdococ |
also, can mobs become infertile/ |
18:55 |
sofar |
I'm not part of the PC police here |
18:55 |
rdococ |
s/\//?/ |
18:55 |
tenplus1 |
and self.hornytimer is exactly that... how long the mob is horny for so it can breed... |
18:56 |
sofar |
the name is thick |
18:56 |
sofar |
very, very thick |
18:56 |
tenplus1 |
lol |
18:56 |
tenplus1 |
what woudl ya make it ? |
18:56 |
rdococ |
sofar: hehe, PC police... it kind of sounds like it could be recursive. PoliCe police |
18:56 |
rubenwardy |
*SJWs |
18:56 |
sofar |
mating, fertile, etc. |
18:56 |
rdococ |
I'm not clear on the definition of SJW though |
18:56 |
sofar |
reproductiontimer |
18:57 |
sofar |
etc. |
18:57 |
sofar |
plenty of normal ways of saying it |
18:57 |
sofar |
heck even `aroused` sounds better |
18:57 |
rdococ |
surely a grand title like "justice warrior" wouldn't be given to someone whose complaints about the way others use terms are extreme? |
18:57 |
tenplus1 |
self.repo_timer ... bam! guy breaks through your door to steal the car back :DDDDD hah |
18:57 |
ircSparky |
testosteronetimer |
18:57 |
|
IhrFussel joined #minetest-hub |
18:57 |
tenplus1 |
hi fussel |
18:58 |
IhrFussel |
tenplus1, not sure if it's still like that in the newest mobs redo version but slow animals with jump enabled cannot climb up nodes...only when I make them faster |
18:58 |
rdococ |
ircSparky: wouldn't that suggest that all the mobs are male? then how would they create child mobs without advanced technology? |
18:58 |
tenplus1 |
lol rdoc |
18:58 |
rdococ |
:O |
18:58 |
tenplus1 |
we cant assume their genders :D |
18:58 |
rdococ |
lol |
18:59 |
rdococ |
the mobs have a secret underground laboratory where they construct eggs |
18:59 |
rdococ |
well, eggs and embryos |
18:59 |
ircSparky |
males just generally have more, if i am understanding correctly |
18:59 |
IhrFussel |
They simply jump at the same spot, they cannot reach the next node (in my version at least) |
18:59 |
rdococ |
ircSparky: lolwat :P |
18:59 |
ircSparky |
idk |
19:00 |
rubenwardy |
women have testosterone |
19:00 |
rubenwardy |
just less of it |
19:00 |
rdococ |
IhrFussel: it's fun to jump repeatedly |
19:00 |
rdococ |
rubenwardy: true, I suppose |
19:00 |
IhrFussel |
In particular tamed mobs that try to follow the player |
19:00 |
IhrFussel |
It's annoying to go with them on exploration tour cause they are stuck if they fall down 1 node |
19:00 |
rdococ |
IhrFussel: just make them super fast ;) |
19:01 |
tenplus1 |
fussel, set stepheight to 1.1 in the mob registry or add to jump_height so they can clear a block better |
19:01 |
tenplus1 |
I've been getting complaints about mobs being able to escape captivity so much I limited them too much it seems |
19:04 |
tenplus1 |
if I make stepheight 1.1 so they can walk up hills then they get stuck inside walls... mt engine bug with entities |
19:05 |
ircSparky |
self.path.way = minetest.find_path(s, p1, 16, self.stepheight, dropheight, "A*_noprefetch") |
19:06 |
ircSparky |
we kno dey wey |
19:06 |
ircSparky |
:P |
19:06 |
tenplus1 |
lol |
19:06 |
rdococ |
I'm bored so I'm experimenting with lua 5.3.4 |
19:06 |
ircSparky |
what are you experimenting? |
19:06 |
rdococ |
not much yet |
19:06 |
rdococ |
I'm looking on the lua-users wii |
19:06 |
rdococ |
s/wii/wiki/ |
19:06 |
tenplus1 |
install love2d and play with gui's :P |
19:07 |
rdococ |
there are cool things like mutable functions |
19:09 |
IhrFussel |
tenplus1, jump_height 6 means 0.6 nodes? |
19:10 |
tenplus1 |
set_velocity({x = x, y = self.jump_height, z = z}) |
19:11 |
IhrFussel |
Is velocity in nodes? |
19:11 |
rubenwardy |
yes |
19:11 |
tenplus1 |
then yeah, 0.6 nps |
19:12 |
tenplus1 |
thing is, the mob jumps high enough to make the step up.. I gotta make it move forward at the apex of the jump |
19:13 |
IhrFussel |
My "hack" is if self.owner ~= "" and self.order == "follow" then v.y = 11 end |
19:14 |
IhrFussel |
That way it should prevent animals from jumping through walls if not following their owner |
19:15 |
tenplus1 |
lol, that's a high jump |
19:15 |
IhrFussel |
1.1 nodes...is that wrong? |
19:18 |
tenplus1 |
mob will jump very high into air |
19:18 |
IhrFussel |
Is there a better self. value to check if the animal is currently following a player? |
19:19 |
tenplus1 |
k, I've added a minetest.after when mob jumps to nudge it forward at ape |
19:19 |
tenplus1 |
*apex |
19:19 |
tenplus1 |
testing |
19:20 |
IhrFussel |
How do you know whether you need to increase X or Z velocity? |
19:21 |
tenplus1 |
it's all calculated in do_jump |
19:21 |
IhrFussel |
Ah...will look at it. The tamed mobs save the player they follow in self.following correct? |
19:22 |
tenplus1 |
jumping up works fine now :D |
19:23 |
tenplus1 |
git updated, try new version |
19:24 |
tenplus1 |
it puts a slight delay after jump which then moves mob forwards to get up that 1 block :D |
19:24 |
tenplus1 |
without the use of stepheight |
19:27 |
Calinou |
rubenwardy: Fedora is an excellent OS for MinGW builds, I use it for my Godot CI builds |
19:27 |
Calinou |
GCC 7 out of the box :) |
19:27 |
rubenwardy |
:'( |
19:27 |
tenplus1 |
18.04 has gcc7 |
19:27 |
Calinou |
for portable Linux builds, you can use Ubuntu 14.04 with a GCC 7 PPA, and statically link libstdc++/libgcc |
19:27 |
rubenwardy |
My computer has gcc7 |
19:27 |
rubenwardy |
(Manjaro) |
19:27 |
Calinou |
you can run Docker containers in Travis CI if you really need to, but you should probably use GitLab CI instead |
19:27 |
rubenwardy |
tenplus1: really? |
19:27 |
Calinou |
(which uses Docker) |
19:28 |
rubenwardy |
maybe it's because I'm using Docker? |
19:28 |
tenplus1 |
yeh, currently sitting at gcc 7.3.0 |
19:28 |
rubenwardy |
no no |
19:28 |
rubenwardy |
I'm talking about mingw-w64-gcc |
19:28 |
rubenwardy |
not gcc |
19:28 |
Calinou |
unfortunately, I haven't found a way to make portable Linux applications from a recent distribution in all cases |
19:28 |
rubenwardy |
it's the windows cross compiler |
19:28 |
Calinou |
(i.e. with graphical applications that link to OpenGL libraries) |
19:29 |
Calinou |
it can be done for CLI apps, but not for GUI applications |
19:29 |
rubenwardy |
companies like Steam and Unity seem to do it well |
19:29 |
Calinou |
(https://github.com/intoli/exodus) |
19:29 |
Calinou |
rubenwardy: they just build on old distributions :P |
19:29 |
rubenwardy |
they probably compile on Ubuntu |
19:29 |
Calinou |
which is what you should do for now |
19:29 |
rubenwardy |
oh? |
19:29 |
Calinou |
my pipe dream is building on a recent distribution so I can use recent tech all I want |
19:29 |
rubenwardy |
what's the issue? |
19:29 |
Calinou |
but Ubuntu 14.04 + GCC 7 PPA is good enough for this, just remember to link libgcc and libstdc++ statically |
19:29 |
Calinou |
(the license allows it) |
19:30 |
Calinou |
it should be enough for C++14 at least, not sure about C++17… I haven't tested |
19:30 |
rubenwardy |
well, I use C++14 |
19:30 |
Calinou |
(GCC 7 has full C++17 support as far as I know, but I don't know if stuff will break when the binary is run on old distributions) |
19:30 |
Calinou |
I'm writing my LearnOpenGL code in C++17 mode… not that I actually use features from it anyway :P |
19:30 |
rubenwardy |
do you know how I can get a newer mingw-gcc version? |
19:30 |
Calinou |
(I definitely use C++11 features and possibly at least one C++14 feature) |
19:31 |
Calinou |
there should be PPAs for it, but I advise you to build from Fedora instead |
19:31 |
Calinou |
nothing forces you to use the same container for all target platforms :) |
19:31 |
Calinou |
https://github.com/Calinou/godot-builds-ci |
19:31 |
Calinou |
I'm using Fedora 27 for all platforms except Linux (Ubuntu 14.04), macOS and iOS (which use Travis CI instead of GitLab CI) |
19:32 |
Calinou |
it's also good for making WebAssembly and Android builds |
19:32 |
rubenwardy |
I use self-hosted CI on my server |
19:33 |
rubenwardy |
with the builds done in docker containers |
19:33 |
Calinou |
ah, I see |
19:33 |
Calinou |
it should be easy enough to run Fedora containers, no? |
19:33 |
rubenwardy |
not sure |
19:34 |
rubenwardy |
I mean, Fedora doesn't have apt, right? |
19:34 |
rubenwardy |
I'd have to rewrite most of my scripts |
19:34 |
rubenwardy |
I'd rather find a PPA |
19:34 |
rubenwardy |
or build mingw-w64 myself |
19:36 |
Calinou |
it uses DNF instead of apt, yeah |
19:36 |
Calinou |
I don't know if there's a good MinGW PPA |
19:36 |
rubenwardy |
doesn't appear to be |
19:41 |
rdococ |
there seems to be no such thing as first-class scopes |
19:46 |
Mr-Pardison |
perhaps there are first-rate ones |
19:47 |
rubenwardy |
Calinou, tenplus1: turns out the docker container is running 16.04, not 18.04 |
19:47 |
rubenwardy |
I did delete and recreate them |
19:48 |
tenplus1 |
:P# |
19:48 |
Calinou |
keep in mind your binary won't run on old distributions |
19:49 |
Calinou |
PSA https://www.gimp.org/news/2018/04/27/gimp-2-10-0-released/ |
19:49 |
tenplus1 |
ooh, havent tested new gimp out yet :))) thanks for reminding me |
19:50 |
benrob0329 |
Calinou: I was using Fedora, then I tried Debian testing |
19:51 |
rubenwardy |
Calinou: jesus, I'm building for Windows not Linux! |
19:52 |
|
Jordach joined #minetest-hub |
19:53 |
tenplus1 |
wb Jordach |
19:53 |
rubenwardy |
argh |
19:53 |
Jordach |
microfiber cloth + 99% isopropyl alchohol |
19:53 |
tenplus1 |
:P |
19:53 |
Jordach |
that shit cleans screens |
19:54 |
rubenwardy |
do you know how I can override the ubuntu version of a docker parent image? |
19:54 |
Jordach |
linux magic |
19:54 |
rubenwardy |
I'm inheriting from this: https://github.com/JetBrains/teamcity-docker-agent |
19:54 |
rubenwardy |
which uses u16.04 |
19:54 |
Jordach |
ayy LTS |
19:54 |
Calinou |
rubenwardy: sorry, I thought you were building for Linux |
19:54 |
Calinou |
I read it wrong |
19:54 |
Calinou |
I mean, both platforms |
19:56 |
Krock |
no idea.. check for docker updates? |
19:56 |
rubenwardy |
I guess I'm going to have to fork both of these repos |
19:57 |
sfan5 |
ayy docker |
19:58 |
Krock |
ayy lmao sfan5 got triggered ^^ |
19:58 |
sfan5 |
yes |
19:59 |
|
Mr_Pardison joined #minetest-hub |
19:59 |
rubenwardy |
actually, with tagging I can replace a parent of a packet out of my control |
20:00 |
tenplus1 |
wb mister |
20:00 |
Mr_Pardison |
t.hanks. |
20:00 |
Mr_Pardison |
something is going on with the internet here..... |
20:09 |
rubenwardy |
so, the answer to my issue is to make my own image with 18.04 and java8, then tag it as the base image for the CI container |
20:10 |
tenplus1 |
nite folks o/// |
20:10 |
|
tenplus1 left #minetest-hub |
20:14 |
rdococ |
A scope would be a special case of associative array which has a 'parent'. If you try to access a scope's variable that is undefined, then it will look in its parent instead for that variable name |
20:14 |
rdococ |
e.g. "function () { var a = 3; return (function () { return a })() }" in the inner function, 'a' does not exist in its scope so it checks the parent scope, which does have 'a' |
20:15 |
rdococ |
This would also occur with assignment (assigning to 'a' will assign to the first variable 'a' it finds, checking the current scope first then its parent, and then its parent), unless you use a keyword like 'local' |
20:16 |
rubenwardy |
makes sense |
20:16 |
rubenwardy |
that's how my compiler/interpretter worked |
20:17 |
rdococ |
Additionally: Function calls would treat 'return' statements as implicit parameters holding the caller's continuation (a function-like object that, when invoked, returns from that function), essentially turning calls into call/cc and also meaning that the call stack would be part of the scoping system |
20:45 |
|
nerzhul joined #minetest-hub |
20:54 |
|
Mr_Pardison joined #minetest-hub |
21:17 |
* rubenwardy |
is now running Linux 4.16 on his laptop |
21:18 |
rubenwardy |
it's really cool how easy it is to switch on Manjaro https://i.rubenwardy.com/J2diI.png |
21:20 |
rdococ |
I am using xubuntu 18.04 with Linux 4.15 |
21:20 |
rubenwardy |
yeah, that's what ubuntu comes with |
21:20 |
rdococ |
which is good, as 4.15 is a fairly recent release |
21:21 |
rubenwardy |
it has the spectre fixes in it, which makes it slower |
21:21 |
rubenwardy |
4.16 has improvements on that |
21:21 |
rubenwardy |
it'll be backported though |
21:21 |
rubenwardy |
may already have been |
21:21 |
rdococ |
I'm bored and experimenting with lua |
21:22 |
rubenwardy |
4.15=28/Jan/18 4.16=1/April/18 |
21:22 |
rubenwardy |
oh, fart |
21:22 |
rubenwardy |
$ sudo apt update |
21:22 |
rubenwardy |
sudo: apt: command not found |
21:22 |
rdococ |
Sounds smelly |
21:23 |
Krock |
rdococ, then fite some players in CTF |
21:23 |
rdococ |
rubenwardy: That doesn't sound good |
21:23 |
rdococ |
Krock: What if I loze? :c |
21:23 |
Krock |
rdococ, then grab a new sord and gun to try again |
21:23 |
rubenwardy |
I'm not using a debian based OS anymore, so no apt |
21:23 |
rubenwardy |
the command I was looking for was pacman -Syyu |
21:23 |
rdococ |
sounds apt |
21:24 |
Krock |
< insert -Syyu meme here > |
21:26 |
* rubenwardy |
adds pacman -Syyu to his crontab |
21:27 |
|
Gael-de-Sailly joined #minetest-hub |
21:29 |
Fixer |
pacman commands are so shitty |
21:30 |
Calinou |
4.15.17-300.fc27.x86_64 |
21:30 |
Calinou |
and yeah, my laptop feels a bit slower lately :/ |
21:30 |
Calinou |
desktop is fine still |
21:32 |
rubenwardy |
> reuben |
21:32 |
rubenwardy |
TRIGGERED |
21:32 |
rubenwardy |
https://forum.minetest.net/viewtopic.php?p=317400#p317400 |
21:32 |
rdococ |
that's a weird way to do a 2x2 crafting table |
21:33 |
rdococ |
yet I think that might actually work |
21:34 |
Krock |
reuben, that method is not CSM-safe |
21:35 |
rubenwardy |
no |
21:41 |
xerox123 |
https://en.wikipedia.org/wiki/Reuben_sandwich |
21:41 |
rdococ |
delicious? |
21:42 |
xerox123 |
I guess so, possibly meat overload |
21:43 |
rdococ |
warning: meat overload |
21:45 |
rdococ |
yay |
21:48 |
rdococ |
I'm working on a linked list implementation |
21:48 |
rdococ |
should list[3] return list.tail.tail.tail or list.tail.tail.tail.head? |
21:51 |
rdococ |
It's a Krock |
21:59 |
* benrob0329 |
starts to setup his website on a more permanent setup |
22:00 |
rdococ |
It's a benrob0329 |
22:00 |
rubenwardy |
it's a yay |
22:00 |
rdococ |
I am slowly infesting everyone around me with the yay |
22:00 |
rdococ |
Soon, everyone will say yay |
22:02 |
rdococ |
I'm working on a linked list script in Lua |
22:02 |
rdococ |
it adds linked lists that use zero-indexingâ„¢ |
22:04 |
sofar |
rdococ: you should't allow [] usage on a linked list |
22:04 |
sofar |
x = List() |
22:04 |
sofar |
x.insert(elem) |
22:04 |
sofar |
x:insert(elem) |
22:05 |
sofar |
etc.. |
22:05 |
rubenwardy |
* List:new() |
22:05 |
rdococ |
sofar: whyZat |
22:05 |
|
Jordach joined #minetest-hub |
22:05 |
sofar |
x:next():get_value() |
22:05 |
sofar |
something like that |
22:06 |
rdococ |
so for the fourth item of a list I need x:next():next():next():get_value()â„¢ |
22:06 |
sofar |
rdococ: [] isn't a linked list operator |
22:06 |
sofar |
rdococ: sure, yes |
22:06 |
rubenwardy |
or get() |
22:06 |
sofar |
or x:get_by_index(4) |
22:06 |
rdococ |
sofar: that's long for x[4] though |
22:07 |
rdococ |
at the moment you can access the head and the tail directly |
22:07 |
sofar |
then your implementation is probably broken |
22:07 |
rdococ |
e.g. "x.head = 42" will work |
22:07 |
sofar |
x:head() and x:tail() should always be used |
22:07 |
sofar |
x:head():set(value) |
22:07 |
rubenwardy |
you know Lua has operator overloading, right? |
22:07 |
sofar |
is he overloading? |
22:07 |
rubenwardy |
it's still kinda wrong to have [], but not too wrong |
22:07 |
rdococ |
yes |
22:07 |
rdococ |
I am overloading |
22:07 |
sofar |
I still wouldn't start with overloading [] tbh |
22:08 |
rubenwardy |
it's the kinda wrong that you shouldn't worry too much about in Lua |
22:08 |
rdococ |
I'm confused |
22:08 |
rdococ |
I need more elaboration on what kind of wrong it i |
22:08 |
rdococ |
s/it i/it is/ |
22:09 |
rdococ |
I also have x(n) - the difference is that x(n).head == x[n] |
22:09 |
* xerox123 |
has exams in a couple weeks >.< |
22:19 |
sofar |
I've never felt the need for a linked list in lua |
22:19 |
sofar |
in C, sure |
22:20 |
rubenwardy |
it's an academic exercise |
22:20 |
rubenwardy |
obv |
22:21 |
sofar |
well I've used LL in several work projects |
22:21 |
sofar |
so I'd like to see what he cooks up in Lua |
22:21 |
rubenwardy |
I've not used it much |
22:22 |
rubenwardy |
mostly use vectors or deques in C++, as I didn't tend to change the contents |
22:22 |
sofar |
systemd-bootchart uses a doubly linked list, for instance |
22:24 |
rdococ |
The way my linked list script works is that the head and tail are exposed as properties of the table returned from List.new |
22:24 |
rdococ |
allowing you to do whatever with them |
22:24 |
sofar |
geesh |
22:24 |
sofar |
can you stop swearing? |
22:24 |
sofar |
I mean, that's just mean |
22:25 |
rdococ |
wat |
22:25 |
rubenwardy |
you don't store the tail |
22:25 |
rubenwardy |
if it's a doubly linked list, you don't even have a tail |
22:25 |
sofar |
rdococ: I'm just saying that doing that kind of stuff when you design a class, are just reasons why you would instantly fail your data structures 101 class |
22:25 |
rdococ |
? |
22:26 |
sofar |
exposing your head/tail should really be done ONLY if you can show a computational advantage in a real world use case |
22:26 |
sofar |
not when you design a basic class |
22:26 |
rdococ |
I should just give up programming |
22:27 |
sofar |
lol, no |
22:27 |
sofar |
just take the learning experience |
22:27 |
rdococ |
I don't see the point of having functions to set the head and tail as opposed to just exposing them |
22:27 |
sofar |
you never should set the head |
22:27 |
sofar |
the head is internally kept |
22:28 |
sofar |
you give out a reference for it (a new list, essentially) |
22:28 |
sofar |
but you never expose it |
22:28 |
rdococ |
I still don't completely understand |
22:33 |
rubenwardy |
OOP relies a lot on keeping things private if they should be private |
22:33 |
rubenwardy |
like, as a class maker you decide what the things that use you should be able to access |
22:33 |
rubenwardy |
and you should choose it in a way that dissuades hacky behaviour |
22:34 |
rdococ |
Lua is the definition of hacky behavior |
22:34 |
rubenwardy |
well, true |
22:45 |
rdococ |
sofar: although, pretending that I'm swearing only really served to confuse me |
23:52 |
Fixer |
HEY< LITTLE DRUM! |