| Time |
Nick |
Message |
| 00:03 |
|
khonkhortisan joined #minetest |
| 00:14 |
|
roboman2444 joined #minetest |
| 00:14 |
|
roboman2444 joined #minetest |
| 00:45 |
|
harrison joined #minetest |
| 00:49 |
|
harrison joined #minetest |
| 00:53 |
|
WindHero joined #minetest |
| 00:57 |
|
Guest67987 joined #minetest |
| 01:24 |
|
frecel joined #minetest |
| 01:42 |
|
GTRsdk joined #minetest |
| 01:44 |
|
zat joined #minetest |
| 01:47 |
RealBadAngel |
https://www.youtube.com/watch?v=vF9FZx2f8Uw |
| 01:47 |
RealBadAngel |
what do you think about it? |
| 02:28 |
|
fu-fu joined #minetest |
| 02:31 |
GTRsdk |
RealBadAngel, looking at lava? |
| 02:32 |
RealBadAngel |
yes |
| 02:32 |
GTRsdk |
looks trippy |
| 02:33 |
NakedFury |
when looking at 22 seconds and up it looks great |
| 02:33 |
RealBadAngel |
http://i.imgur.com/CwfeNc1.png |
| 02:34 |
RealBadAngel |
some bumpmapping added |
| 02:34 |
NakedFury |
damn good lava |
| 02:35 |
NakedFury |
really like how it looks |
| 02:36 |
NakedFury |
we need this special effect for minetest lava |
| 02:36 |
NakedFury |
https://www.youtube.com/watch?v=xExdEXOaA9A#t=31 |
| 02:42 |
fu-fu |
Would I be right in guessing the only way to preserve variables between server restarts is in files? |
| 02:46 |
fu-fu |
Also, very pretty. |
| 02:46 |
|
exio4 joined #minetest |
| 02:46 |
exio4 |
hi |
| 02:46 |
fu-fu |
ho |
| 02:47 |
Exio4 |
dammit |
| 02:47 |
fu-fu |
timmad |
| 02:48 |
Exio4 |
hai hai anyway |
| 02:48 |
Exio4 |
-exio4- VERSION xchat 2.8.8 Linux 2.6.37-slitaz [i686/4,01GHz/SMP] |
| 02:48 |
Exio4 |
kinda weird |
| 02:48 |
Exio4 |
32bits and 4ghz |
| 02:48 |
Exio4 |
4012! |
| 02:48 |
fu-fu |
Huh. |
| 02:48 |
fu-fu |
Why? |
| 02:49 |
Exio4 |
because noone sane uses 32bit + 4ghz + 12gb of ram |
| 02:49 |
Exio4 |
Mem: 3257 135 3121 0 10 |
| 02:49 |
Exio4 |
i can use... 3257 mb of ram |
| 02:50 |
fu-fu |
Recompile kernel with kexec, and see if you can spontaniously switch it to 64bit? |
| 02:57 |
RealBadAngel |
https://www.youtube.com/watch?v=58lS8ptwNkU |
| 02:57 |
RealBadAngel |
^^still lava |
| 02:57 |
RealBadAngel |
with bumpmapping ofc |
| 02:58 |
VanessaE |
you need to invert that bump-map |
| 02:58 |
VanessaE |
the cracks look like ridges |
| 02:58 |
VanessaE |
looks good though |
| 02:59 |
RealBadAngel |
this is generated on the fly one |
| 03:02 |
|
donat joined #minetest |
| 03:20 |
|
kizeren2 joined #minetest |
| 03:23 |
|
zyxwvuts joined #minetest |
| 03:24 |
|
zyxwvuts left #minetest |
| 03:39 |
|
asl joined #minetest |
| 05:08 |
|
dhbiker joined #minetest |
| 05:20 |
|
PenguinDad joined #minetest |
| 05:42 |
donat |
how can I remove all entities from the world? |
| 05:43 |
Vazon |
./clearobjects would do it |
| 05:43 |
Vazon |
at least i believe so |
| 06:04 |
|
werwerwer joined #minetest |
| 06:07 |
|
AnotherBrick joined #minetest |
| 06:12 |
|
AnotherBrick joined #minetest |
| 06:16 |
|
grrk-bzzt joined #minetest |
| 06:32 |
donat |
thanks, it worked |
| 06:38 |
|
AnotherBrick joined #minetest |
| 06:43 |
|
grrk-bzzt joined #minetest |
| 06:43 |
|
grrk-bzzt joined #minetest |
| 06:45 |
|
grrk-bzzt joined #minetest |
| 06:46 |
|
asie joined #minetest |
| 06:47 |
|
sebastia joined #minetest |
| 06:53 |
|
AnotherBrick joined #minetest |
| 06:58 |
|
LemonLake joined #minetest |
| 07:01 |
|
aheinecke joined #minetest |
| 07:08 |
|
werwerwer joined #minetest |
| 07:23 |
|
AnotherBrick joined #minetest |
| 07:33 |
|
asie joined #minetest |
| 07:35 |
ThatGraemeGuy |
any objections to pipeworks-related questions here or is there a better place? |
| 07:36 |
|
Akagi201 joined #minetest |
| 07:41 |
|
Akagi201 joined #minetest |
| 08:00 |
sfan5 |
meow |
| 08:00 |
sfan5 |
hi everyone |
| 08:00 |
PenguinDad |
hi sfan5 |
| 08:01 |
|
Yepoleb_ joined #minetest |
| 08:03 |
|
Kalista joined #minetest |
| 08:18 |
|
khor_ joined #minetest |
| 08:29 |
|
shadowzone joined #minetest |
| 08:31 |
shadowzone |
Ok is there anyone here who is a op on the other #minetest? |
| 08:36 |
sfan5 |
other #minetest? |
| 08:36 |
sfan5 |
the one on inchra? |
| 08:36 |
shadowzone |
Yes. |
| 08:37 |
shadowzone |
there was a netsplit about 5 to 7 minutes ago. |
| 08:43 |
|
Calinou joined #minetest |
| 08:47 |
|
Krock joined #minetest |
| 08:47 |
Krock |
hi |
| 08:47 |
|
asie joined #minetest |
| 08:48 |
shadowzone |
Hi. |
| 08:48 |
|
kalist4 joined #minetest |
| 08:50 |
Calinou |
hi |
| 08:59 |
|
Kalista joined #minetest |
| 09:00 |
shadowzone |
VanessaE |
| 09:18 |
|
alket joined #minetest |
| 09:23 |
|
Tux[Qyou] joined #minetest |
| 09:34 |
Calinou |
happy 3rd birthday 0.2_20110731 |
| 09:41 |
Krock |
:D |
| 09:42 |
Calinou |
tomorrow, my forum account is 3 years old |
| 09:43 |
Calinou |
in 3 days, I killed an oerkki for the first time 3 years ago |
| 09:44 |
Krock |
in 1 day, I'll be playing with fire |
| 09:48 |
|
Pest joined #minetest |
| 09:53 |
|
AnotherBrick joined #minetest |
| 09:53 |
|
Vargos joined #minetest |
| 10:08 |
|
AnotherBrick joined #minetest |
| 10:27 |
|
alket_ joined #minetest |
| 10:28 |
|
Tux[Qyou] joined #minetest |
| 10:33 |
|
Weakence joined #minetest |
| 10:44 |
|
restcoser joined #minetest |
| 11:05 |
|
Jousway joined #minetest |
| 11:15 |
|
proller joined #minetest |
| 11:15 |
|
Jordach joined #minetest |
| 11:16 |
Jordach |
ello |
| 11:18 |
|
ImQ009 joined #minetest |
| 11:21 |
|
SylvieLorxu joined #minetest |
| 11:23 |
RealBadAngel |
hi Jordach |
| 11:23 |
|
AnotherBrick joined #minetest |
| 11:31 |
|
shadowzone joined #minetest |
| 11:36 |
PenguinDad |
Jordach: Dingbats Man was here :P |
| 11:42 |
RealBadAngel |
https://www.youtube.com/watch?v=v52f-1YsSAM |
| 11:42 |
RealBadAngel |
^^ new look |
| 11:47 |
RealBadAngel |
http://i.imgur.com/uMzaqOn.png and a screenshot |
| 11:51 |
|
CheapSeth joined #minetest |
| 11:56 |
|
asie joined #minetest |
| 11:58 |
|
CheapSeth left #minetest |
| 12:15 |
Jordach |
RealBadAngel, you still don't understand that shaders aren't meant just for Haven right |
| 12:16 |
VanessaE |
Jordach: he showed me that shader using default 16px lava texture also - it looked pretty good |
| 12:17 |
VanessaE |
https://www.youtube.com/watch?v=vF9FZx2f8Uw |
| 12:17 |
VanessaE |
there. |
| 12:17 |
VanessaE |
the two nodes in the corner are flowing lava, still using the default texture. |
| 12:17 |
VanessaE |
only visible for a few seconds though |
| 12:18 |
VanessaE |
and also with HDX: http://i.imgur.com/1YHRQMJ.png |
| 12:29 |
Jordach |
i don't see it |
| 12:30 |
VanessaE |
bbl |
| 12:32 |
|
kizeren joined #minetest |
| 12:32 |
Jordach |
i had an oh pls moment |
| 12:34 |
Jordach |
!tell Calinou https://cdn.mediacru.sh/wcs5Tq-Yr0tD.png |
| 12:34 |
MinetestBot |
Jordach: yeah, sure, whatever |
| 12:35 |
|
fu-fu joined #minetest |
| 12:35 |
Jordach |
and August tomorrow |
| 12:40 |
|
Megaf joined #minetest |
| 12:43 |
|
asie joined #minetest |
| 12:55 |
|
harrison joined #minetest |
| 12:57 |
|
jojoa1997 joined #minetest |
| 13:10 |
|
Alex_Mercer joined #minetest |
| 13:13 |
|
dafull97 joined #minetest |
| 13:14 |
|
harrison_ joined #minetest |
| 13:14 |
|
UukGoblin joined #minetest |
| 13:16 |
|
zash joined #minetest |
| 13:17 |
|
VargaD joined #minetest |
| 13:17 |
|
VanessaE joined #minetest |
| 13:17 |
|
^v joined #minetest |
| 13:17 |
|
bas080 joined #minetest |
| 13:17 |
|
crafterbox joined #minetest |
| 13:23 |
|
kahrl joined #minetest |
| 13:23 |
|
AnotherBrick joined #minetest |
| 13:31 |
|
Taoki[laptop] joined #minetest |
| 13:31 |
|
luizrpgluiz joined #minetest |
| 13:32 |
luizrpgluiz |
hi |
| 13:33 |
|
Taoki[laptop] joined #minetest |
| 13:36 |
luizrpgluiz |
does anyone here mods? |
| 13:37 |
|
Jousway joined #minetest |
| 13:39 |
|
luizrpgluiz left #minetest |
| 13:47 |
|
smoke_fumus joined #minetest |
| 13:51 |
|
kizeren joined #minetest |
| 13:55 |
|
Amaz joined #minetest |
| 13:56 |
|
jojoa1997 joined #minetest |
| 14:01 |
RealBadAngel |
Jordach, for christ sake, shaders are doing on high res sources in the first place |
| 14:02 |
RealBadAngel |
2nd, if you find this lava shader wrong, youre just fuckin blind |
| 14:02 |
|
hmmmm joined #minetest |
| 14:02 |
RealBadAngel |
imho thats the most nice thing i ever made |
| 14:02 |
RealBadAngel |
hmmm, here? |
| 14:02 |
Jordach |
RealBadAngel, not everyone likes fucking >16px |
| 14:03 |
PenguinDad |
Jordach: that's why there are settings to disable these shaders |
| 14:04 |
RealBadAngel |
well, that lava is 2048px |
| 14:04 |
RealBadAngel |
:P |
| 14:05 |
RealBadAngel |
whats funny that lava shader is way faster than default 16px animated :P |
| 14:06 |
RealBadAngel |
but anyway, you can feed the shader with any base texture |
| 14:06 |
RealBadAngel |
as VanessaE mentioned even 16px looks quite interesting |
| 14:10 |
|
kizeren2 joined #minetest |
| 14:11 |
|
frecel joined #minetest |
| 14:21 |
|
twoelk joined #minetest |
| 14:27 |
|
Krock joined #minetest |
| 14:28 |
Krock |
hi |
| 14:30 |
PenguinDad |
iH :kcorK |
| 14:30 |
|
Calinou joined #minetest |
| 14:30 |
MinetestBot |
Calinou: 07-31 17:34 UTC <Jordach> https://cdn.mediacru.sh/wcs5Tq-Yr0tD.png |
| 14:31 |
Calinou |
lol |
| 14:31 |
Calinou |
fun fact, when I first played TF2, I thought “BLU†was a typo |
| 14:32 |
* Jordach |
considers making a random node mod |
| 14:34 |
Calinou |
if you want to make a mod, make on that makes the /give command smart |
| 14:34 |
Calinou |
searches in registered names, to not have to type mod name |
| 14:34 |
Calinou |
can type partial name, auto-completion |
| 14:34 |
Calinou |
sfan5 said this was possible |
| 14:34 |
Jordach |
Calinou, random nodes as in potential conbinations of natual looks |
| 14:34 |
|
reactor joined #minetest |
| 14:36 |
|
asie joined #minetest |
| 14:38 |
reactor |
Evening! |
| 14:38 |
Krock |
afternoon! |
| 14:38 |
reactor |
proller: heard about the new law? |
| 14:45 |
Krock |
Jordach, srsly? still at 3999? |
| 14:48 |
PenguinDad |
Krock: Jordach was configured with --enable-the-force :P |
| 14:48 |
Krock |
:/ |
| 14:48 |
Krock |
seems like sticked the count at 3999 |
| 14:49 |
Krock |
wut? A 16 bit DOS application tells me about too less virtual ram? |
| 14:50 |
Jordach |
>16bit >needs more ram |
| 14:50 |
Krock |
>Jordach |
| 14:50 |
Krock |
>need more posts |
| 14:50 |
Jordach |
1 more * |
| 14:51 |
Krock |
why 1 more? |
| 14:51 |
Jordach |
3999 :P |
| 14:51 |
Krock |
97 posts |
| 14:51 |
Krock |
until 4096 |
| 14:52 |
|
NakedFury joined #minetest |
| 14:52 |
Jordach |
you do realise i'm timing 4k with something right] |
| 14:54 |
RealBadAngel |
Jordach, so compare it to real 16px (or something like that): minecraft lava shader: https://www.youtube.com/watch?v=0QT7Qqza9Sw |
| 14:54 |
RealBadAngel |
:PPPPPP |
| 14:55 |
Jordach |
#notashader |
| 14:56 |
Jordach |
https://www.youtube.com/watch?v=-ihfvu9Z4o8 not a fucking "real" light source |
| 14:58 |
RealBadAngel |
that lava is ugly as millenium old dogs shit |
| 14:59 |
Jordach |
RealBadAngel, and |
| 14:59 |
RealBadAngel |
you may be good in blender but you dont have a bit of taste :P |
| 14:59 |
Jordach |
try and make the modulation and distortion based on the resoulution of the image |
| 14:59 |
reactor |
Speaking of blender. |
| 14:59 |
Jordach |
that's used for the shader material |
| 14:59 |
reactor |
Everything is grey in Cycles -- is this a known problem? |
| 15:00 |
Jordach |
errrr |
| 15:00 |
* Jordach |
doesn't use Cycles |
| 15:00 |
Jordach |
best ask in #blender |
| 15:00 |
Jordach |
or http://reddit.com/r/blender |
| 15:00 |
RealBadAngel |
Jordach, lava is combination of 3 textures and some math |
| 15:00 |
RealBadAngel |
its not so easily scaleable |
| 15:00 |
Jordach |
RealBadAngel, stop hardcoding things D: |
| 15:11 |
twoelk |
hi Krock, you use windows don't you? |
| 15:16 |
|
Calinou joined #minetest |
| 15:19 |
twoelk |
anybody want to try a mapper related batchfile I have been tinkering with? |
| 15:20 |
Calinou |
your Windows seems legit, twoelk :P |
| 15:20 |
* twoelk |
would like to know if this works on other pc's https://www.dropbox.com/s/tji2lhjknpcnmq0/wizapmapper.zip |
| 15:20 |
Calinou |
Windows Vista Ultimate Service Pack 2 (Build 6002) |
| 15:22 |
|
mati1 joined #minetest |
| 15:22 |
mati1 |
hi |
| 15:22 |
mati1 |
dzindybry |
| 15:22 |
twoelk |
one of several OS's I use ;-P |
| 15:24 |
|
Hirato_ joined #minetest |
| 15:25 |
|
ecutruin joined #minetest |
| 15:26 |
Jordach |
hai Ecu :3 |
| 15:26 |
* sfan5 |
meows at Jordach |
| 15:27 |
Jordach |
sfan5, [16:25:00] <Microchip> Jordach: Can't, I'm busy. (I just git pulled, since I've been playing BFD instead of Minetest_game) [16:25:05] <Microchip> Jordach: Way better imo. |
| 15:27 |
sfan5 |
I don't decide which games get into the release |
| 15:27 |
|
jojoa1997 joined #minetest |
| 15:27 |
sfan5 |
that info is worthless for me |
| 15:27 |
jojoa1997 |
hi |
| 15:27 |
* sfan5 |
licks jojoa1997 |
| 15:28 |
* jojoa1997 |
pets sfan5 |
| 15:30 |
Calinou |
well, you're always busy with BFD |
| 15:31 |
Calinou |
jojoa1997, if you want to give Stunt Rally a try, you should probably start it downloading when you have time |
| 15:31 |
|
harrison joined #minetest |
| 15:32 |
Jordach |
i'm an idiot as usual |
| 15:32 |
Jordach |
i was doing it wrong |
| 15:33 |
|
harrison joined #minetest |
| 15:35 |
jojoa1997 |
Calinou nah i ahve other things |
| 15:36 |
crazyR |
has anyone set the mesecons PISTON_MAXIMUM_PUSH settings higher than the default 15? |
| 15:47 |
Krock |
Always look up on someone if you're not helping them down. |
| 15:48 |
PenguinDad |
Krock: Never look up to someone unless you're helping them down. :P |
| 15:48 |
Krock |
Always look up on everybody if you're not pushing them down. |
| 15:48 |
Krock |
PenguinDad ^ |
| 15:50 |
PenguinDad |
Never look up everybody unless you're spying them :D |
| 15:50 |
Krock |
omg. such opposite |
| 15:55 |
Calinou |
https://cdn.mediacru.sh/qBA6Twlv2OzT.png → Geany colour scheme update |
| 15:55 |
Calinou |
more yellow keywords |
| 15:56 |
* PenguinDad |
hugs Calinou |
| 15:56 |
sfan5 |
!tell hoodedice Droid Sans as a programming font? https://cdn.mediacru.sh/2OKDvLIqQhog.png ew why not monospace fonts? |
| 15:56 |
MinetestBot |
sfan5: I'll pass that on when hoodedice is around |
| 15:57 |
Calinou |
use a fixed-width font |
| 15:57 |
Calinou |
sfan5, DejaVu Sans Mono or Droid Sans Mono or Fira Mono |
| 15:57 |
Calinou |
your colour scheme makes key words hard to read |
| 15:57 |
sfan5 |
tell that hoodedice |
| 15:57 |
Calinou |
almost impossible :s |
| 15:57 |
sfan5 |
I use ency code r normally |
| 15:57 |
sfan5 |
envy* |
| 15:57 |
Calinou |
I tried to make everything as readable |
| 15:57 |
Calinou |
proprietary font :p |
| 15:57 |
sfan5 |
https://cdn.mediacru.sh/bFRUGiqUrOT1.png |
| 15:57 |
Calinou |
also ugly anyway |
| 15:58 |
Calinou |
too thin, feels like 1995 |
| 15:58 |
PenguinDad |
I usually use Droid Sans Mono |
| 15:59 |
sfan5 |
droid sans mono looks |
| 15:59 |
sfan5 |
meh |
| 15:59 |
* Jordach |
sits there with Consolas |
| 15:59 |
Jordach |
or was it courier |
| 16:00 |
sfan5 |
Jordach: consolas is a bit too thick and courier is just ugly |
| 16:01 |
Calinou |
DejaVu Sans Mono looks smooth, could be slightly more spaced between lines |
| 16:01 |
Calinou |
but at least you can show lots of lines on a screen |
| 16:01 |
Calinou |
a cave connecting two water pools (seen a few months ago in Carbone): https://cdn.mediacru.sh/J50XJ3eI-HTw.jpg |
| 16:03 |
Jordach |
bullshit alarm https://cdn.mediacru.sh/zN_Ukyr-cns9.png |
| 16:03 |
Jordach |
i can't be fucked to write a titanic single liner |
| 16:04 |
sfan5 |
meow |
| 16:05 |
Jordach |
i just want a font without AA and hinting |
| 16:06 |
|
asie joined #minetest |
| 16:06 |
sfan5 |
that doesn't really depend on the font |
| 16:06 |
* Krock |
just tested minecraft demo and it not very happy. he got the same fps as in MT (even if MT should have less graphical "operations") |
| 16:06 |
sfan5 |
I can set that in my xfce settings |
| 16:06 |
Jordach |
Krock, 280fps in the snapshots |
| 16:07 |
* Krock |
opens everytime the inventory when trying to walk fast :3 |
| 16:07 |
Calinou |
what's wrong with AA and hinting?! |
| 16:07 |
Calinou |
these things aren't your enemies |
| 16:07 |
sfan5 |
^ |
| 16:08 |
* Krock |
's try to enter /giveme default:hoe_stone failed (well, he could try it with mese, but that's not expectabletowork) |
| 16:08 |
Jordach |
Krock, https://twitter.com/SeargeDP/status/491594238329499648 |
| 16:08 |
Calinou |
Krock, meh, E-as-inventory-key is way smarter, that should've been default in Minetest |
| 16:08 |
Calinou |
I use R as use key |
| 16:08 |
Calinou |
very good |
| 16:08 |
Calinou |
G as drop key |
| 16:08 |
* twoelk |
always has inventory on E and fast on F |
| 16:08 |
Calinou |
this is layout-agnostic, also works in AZERTY the same way it works on QWERTY |
| 16:08 |
Calinou |
I have full range on F |
| 16:08 |
PenguinDad |
Krock: that's why I set the inventory key to I in CMc |
| 16:08 |
PenguinDad |
*MC |
| 16:08 |
* Krock |
has MT defaults and it happy with it |
| 16:09 |
Krock |
Jordach, your grpahics card is too good for your eyes |
| 16:09 |
Krock |
the screen also can't display such framerate |
| 16:09 |
twoelk |
nah mt defaults is two hands |
| 16:10 |
PenguinDad |
MC default is two hands too |
| 16:10 |
Krock |
now I get why MC alway looks other than MT on screenshots.. the nodes have a bigger shadow! |
| 16:13 |
twoelk |
the most dangerous key in mt is Q, way to close to W |
| 16:13 |
Calinou |
Q is a crap key |
| 16:13 |
Calinou |
it changes on AZERTY, is poorly placed |
| 16:13 |
twoelk |
on qwertz that is |
| 16:14 |
twoelk |
erm "too close to"? |
| 16:15 |
Jordach |
does minetest.dig_node(pos) drop the item on the floor when used? |
| 16:15 |
Calinou |
https://github.com/minetest/minetest_game/issues/293 → BlockMen powa! |
| 16:15 |
Jordach |
(i'm in need of the method used for dropping when inv full) |
| 16:16 |
Krock |
Jordach, for my tests, it only dug them. don't kno about item_drop |
| 16:16 |
Jordach |
bleh, i mean literally use the code else where that makes the item drop |
| 16:17 |
Jordach |
when your inventory is completely full |
| 16:17 |
twoelk |
overflowing inventory? |
| 16:18 |
twoelk |
drop the most valuable first :-D |
| 16:18 |
Jordach |
twoelk, you know mining stone until it starts falling onto the floor |
| 16:19 |
twoelk |
yeah thats when I start storing stuff in the crafting grid |
| 16:19 |
Jordach |
twoelk, well i just want the items falling on the floor bit |
| 16:19 |
twoelk |
bags in default ftw |
| 16:19 |
|
SoniEx2 joined #minetest |
| 16:21 |
twoelk |
I love the vacuming effect when you approach dropped items with room in the inventory |
| 16:23 |
|
AnotherBrick joined #minetest |
| 16:24 |
* twoelk |
is musing about a vacumcleaner mod that deletes the items it sucks up |
| 16:24 |
|
PilzAdam joined #minetest |
| 16:25 |
Jordach |
i suppose core.get_item_drops can be used as a substitute |
| 16:27 |
Jordach |
PilzAdam, i'm trying to drop items from an abm, eg, remove the node, the drop drops from where the abm activated |
| 16:31 |
PenguinDad |
Papyrus is the rarest thing in my subgame :D |
| 16:33 |
|
Exio joined #minetest |
| 16:34 |
|
Vazon joined #minetest |
| 16:34 |
twoelk |
why make papyrus rare? |
| 16:35 |
PenguinDad |
twoelk: it's actually a bug =) |
| 16:35 |
twoelk |
oops |
| 16:36 |
twoelk |
I think papyrus is rather usefull and quite often too rare on servers |
| 16:36 |
Krock |
papyrus blocks are needd. |
| 16:36 |
|
jojoa1997 joined #minetest |
| 16:37 |
twoelk |
hm 9 papyrus=papyrusblock=bambus? |
| 16:37 |
Krock |
papyrus != bambus |
| 16:37 |
Jordach |
i have the feeling this might not work |
| 16:37 |
Jordach |
http://paste.debian.net/112965/ |
| 16:38 |
Krock |
bambus != propr english |
| 16:39 |
twoelk |
minetest uses proper english? |
| 16:39 |
Krock |
Jordach, why "hardcoding" thoseliquids? just go through all registered nodes |
| 16:39 |
Jordach |
items aren't dropping |
| 16:39 |
Krock |
minetest much english proper |
| 16:40 |
Krock |
ah I see, use group:liquid |
| 16:40 |
Jordach |
Krock, can't be bothered |
| 16:40 |
Jordach |
a simple list is faster |
| 16:40 |
Krock |
k |
| 16:40 |
twoelk |
proper can mean in german "all shiney and well made" |
| 16:41 |
PenguinDad |
yay! just increasing two numbers fixed it |
| 16:41 |
Krock |
can a can be CleAN? |
| 16:41 |
Jordach |
seriously - i'm trying to get the drops of a node in that list and make sure it puts them in the world |
| 16:44 |
|
init joined #minetest |
| 16:53 |
Jordach |
WOOO |
| 17:02 |
Jordach |
items now drop from being replaced by flowing liquids |
| 17:04 |
|
asie joined #minetest |
| 17:04 |
|
LemonLake joined #minetest |
| 17:04 |
Calinou |
cool |
| 17:04 |
Calinou |
ABM? |
| 17:04 |
Jordach |
yes |
| 17:04 |
Jordach |
kinda heavy |
| 17:04 |
LemonLake |
yes yes |
| 17:04 |
LemonLake |
heavy is |
| 17:04 |
Jordach |
mowing the fucking lawn: https://cdn.mediacru.sh/9Npnm7_gqEwe.png |
| 17:05 |
Jordach |
i may have brutally murdered a builtin script |
| 17:05 |
Jordach |
(and moving it to misc) |
| 17:05 |
Jordach |
(if anyone asks, my usual address is running it) |
| 17:07 |
|
SoniEx2 joined #minetest |
| 17:08 |
|
q66 joined #minetest |
| 17:12 |
Calinou |
pull request to minetest_game, I guess |
| 17:15 |
|
Exio joined #minetest |
| 17:15 |
sfan5 |
!up jordach.minetest.net |
| 17:15 |
MinetestBot |
jordach.minetest.net:30000 is up (402ms) |
| 17:20 |
|
grrk-bzzt joined #minetest |
| 17:26 |
Jordach |
mew? |
| 17:26 |
* sfan5 |
meows at Jordach |
| 17:27 |
Jordach |
56fps in BFD :p |
| 17:27 |
Jordach |
(with the VM engaged) |
| 17:29 |
|
jp__ joined #minetest |
| 17:29 |
sfan5 |
Jordach: spotify:track:2vBQaIronZnOIpZBwSHQYb |
| 17:32 |
Krock |
!up minetest.ntdll.net |
| 17:32 |
MinetestBot |
minetest.ntdll.net:30000 is up (84ms) |
| 17:33 |
Jordach |
turns out theres lots of builtin code that can be used elsewhere |
| 17:33 |
sfan5 |
Random thing: servers.minetest.net now supports IPv6 (for the web server, IPv6 for servers was already supported) |
| 17:36 |
|
PenguinDad joined #minetest |
| 17:38 |
jp__ |
[ENTER] key still crash MT when pressed... |
| 17:38 |
sfan5 |
we know |
| 17:38 |
sfan5 |
but nobody wants to fix it |
| 17:39 |
PenguinDad |
Roboto Condensed still crashes gimp in debian |
| 17:40 |
Sokomine |
hm. if i turn shaders on (for waving plants and water), my fan audibly has more to do. guess i'll better stick to shaders off again for testing...and turn them on only for actual playing |
| 17:42 |
Calinou |
Sokomine, shaders on, without waving stuff |
| 17:42 |
Calinou |
is what I use, to fix water rendering and such |
| 17:42 |
Calinou |
but yeah, shader-less rendering is almost the same now |
| 17:43 |
Sokomine |
i like the waving plants. it makes the landscape more enjoyable. and the waving water is also decorative. but for testing, that's not really required |
| 17:44 |
* Sokomine |
is generating lots of villages in order to see if they work out as desired |
| 17:44 |
* sfan5 |
is meowing |
| 17:44 |
* PenguinDad |
pets sfan5 |
| 17:45 |
* sfan5 |
purrs |
| 17:45 |
* Sokomine |
pets sfan5 as well and puts a glass of milk on the table |
| 17:46 |
* sfan5 |
noms the milk |
| 17:46 |
Calinou |
meh, I hate waving stuff, bad for gameplay, distracting |
| 17:47 |
jp__ |
+1 |
| 17:47 |
Calinou |
Minetest running without FPS limit doesn't make the card any noisier, due to obnoxious CPU limitation |
| 17:47 |
|
rubenwardy joined #minetest |
| 17:47 |
Calinou |
(but I limit FPS to make it use less power) |
| 17:47 |
Sokomine |
the waving trees are very odd and confusing, yes. as for plants, i like it. but i rarely play survival as such |
| 17:47 |
rubenwardy |
Hi all! |
| 17:47 |
Jordach |
rubenwardy, ello |
| 17:47 |
Jordach |
i did quite a bit of mowing the lawn ;3 |
| 17:48 |
PenguinDad |
hey rubenwardy |
| 17:48 |
Sokomine |
it's only a very small increase in fan speed. it's audible, but only when it's quiet. after all this is still just a celeron g1610 doing the job...not a fancy graphics card :-) |
| 17:48 |
Sokomine |
hi rubenwardy |
| 17:48 |
|
mati1 left #minetest |
| 17:48 |
Sokomine |
jordach: real one or in bfd? |
| 17:48 |
Jordach |
BFD# |
| 17:48 |
Sokomine |
ah :-) |
| 17:48 |
Jordach |
rubenwardy, Sokomine https://cdn.mediacru.sh/9Npnm7_gqEwe.png |
| 17:48 |
Jordach |
water and other liquids remove plants and other stuff ;3 |
| 17:49 |
Sokomine |
i'm currently puzzling over how to best place fields around the villages generated by nores mapgen. turning the buildings further out into fields helps only to a limited degree. it still looks too odd |
| 17:50 |
Calinou |
my CPU fan speed never increases, or I don't hear that increase |
| 17:50 |
Calinou |
the latter probably |
| 17:50 |
Sokomine |
hmm. i don' like that removing part of water. better add a real harvester :-) a node that travels on raillike nodes and has two "arms" stretching sideways 3 nodes with what looks like 3 tubes from pipeworks would be nice |
| 17:50 |
Calinou |
it's not PWM, but motherboard regulates it |
| 17:50 |
Calinou |
runs at ~600 rpm at idle (minimum setting) |
| 17:51 |
Jordach |
Sokomine, plants don't block water literally, and neither do flowers |
| 17:51 |
Sokomine |
i'm using the boxed fan as that was said to be good enough (and quiet enough) for that cpu. and it is. it's really just a slow increase of noise |
| 17:52 |
Sokomine |
hmm. but they don't get harvested by water either. the water just flows through |
| 17:52 |
Calinou |
woah, boxed fan is crap |
| 17:52 |
Calinou |
for small dual cores it's acceptable, nothing more |
| 17:52 |
Sokomine |
remember, it's a cheap cpu that doesn't generate much heat anyway |
| 17:52 |
Calinou |
but if you want real quietness, you need something better |
| 17:52 |
Calinou |
but at least boxed fan is easy to install |
| 17:52 |
Sokomine |
exactly. it's the smallest dual core anyway |
| 17:53 |
Calinou |
unlike the heatsink I use :/ |
| 17:53 |
Calinou |
(and it's not the hardest-to-install one ever seen by mankind) |
| 17:53 |
Sokomine |
yes. that was an argument for it as well: with that cheap cpu, i didn't have to worry much about cooling. easy install + quiet + low power consumption...works quite well most of the time :-) |
| 17:54 |
Sokomine |
sometimes i wish for a dedicated graphics card though. it'd perhaps allow me to look further or to get at least more fps on far |
| 17:54 |
VanessaE |
Jordach, RealBadAngel: I showed Abe that minecraft lava shader, he thought it looked good. then showed him the minetest one with the 16px textures in the corner and he was like "WHOA!". |
| 17:54 |
Jordach |
16px doesn't prove it |
| 17:54 |
VanessaE |
Jordach: this video doesn't? https://www.youtube.com/watch?v=vF9FZx2f8Uw |
| 17:55 |
VanessaE |
Jordach: what would prove it then? aside from needing to see the code of course |
| 17:55 |
Jordach |
VanessaE, it doesn't even look like the original texture |
| 17:55 |
VanessaE |
it is the original texture. |
| 17:55 |
Jordach |
and RBA is hardcoding patterns |
| 17:55 |
VanessaE |
the patterns come from a normalmap I believe |
| 17:55 |
Jordach |
a simple noise texture solves everything |
| 17:55 |
VanessaE |
similar to what's done with the water surface shader |
| 17:55 |
VanessaE |
you can replace them via a texture pack |
| 17:56 |
VanessaE |
the overlaying texture comes from the textures you supply in your game/texture pack |
| 17:56 |
Sokomine |
vanessae: tell it inocudom. he loves improvements on the graphics side :-) |
| 17:56 |
VanessaE |
I had him try it with default textures and with the textures supplied in HDX. |
| 17:56 |
Sokomine |
sounds like a practical soulution |
| 17:57 |
Jordach |
water doesn't have a fucking normal map IRL |
| 17:57 |
Jordach |
i don't want graphics from 1998 |
| 17:57 |
LemonLake |
Jordach: normal maps don't fucking exist IRL |
| 17:57 |
VanessaE |
Jordach: no, but it has rippled on the surface. the only sane, efficient way to model something like water is to use normalmapping. |
| 17:57 |
VanessaE |
that's how all modern games do it |
| 17:57 |
LemonLake |
but do you know how expensive it would be to calculate REAL water? |
| 17:57 |
Jordach |
efficient |
| 17:57 |
Sokomine |
hmm, those graphics back then wheren't bad. graphics' not everything |
| 17:57 |
LemonLake |
reaaallly fucking expensive |
| 17:58 |
Jordach |
LemonLake, the perlin noise can be used |
| 17:58 |
VanessaE |
Jordach: realtime perlin noise would be ridiculously expensive |
| 17:58 |
Jordach |
VanessaE, horse shit |
| 17:58 |
VanessaE |
it can't be done, not even with a Titan or tesla. |
| 17:58 |
Jordach |
if blender cycles can fucking do it, you're doing it wrong |
| 17:58 |
Jordach |
and other video games as well |
| 17:59 |
VanessaE |
just because they can doesn't mean they *should* |
| 17:59 |
Calinou |
Tesla is not necessarily faster than GeForces, VanessaE |
| 17:59 |
Calinou |
they have some more professional features, that's all |
| 17:59 |
Calinou |
(but less than Quadros… and they usually don't have video outputs) |
| 17:59 |
VanessaE |
Calinou: I know, just going by the benchmarks |
| 18:00 |
Calinou |
Jordach, water using normal maps looks quite good |
| 18:01 |
VanessaE |
Jordach: you don't expend 100 CPU cycles doing something if you can do it in 10 cycles and spend 100K in image data to get it done. Not if you're doing it in a speed-critical render loop. |
| 18:01 |
VanessaE |
in blender, it just doesn't matter., |
| 18:01 |
VanessaE |
in a video game, it fucking matters more than ANYTHING. |
| 18:01 |
LemonLake |
^ |
| 18:01 |
Jordach |
other video games already produce non repeating water normals |
| 18:01 |
VanessaE |
rule number one in speed critical code: memory is cheap. CPU cycles aren't. |
| 18:01 |
Jordach |
bullshit |
| 18:01 |
Jordach |
1998 thinking |
| 18:02 |
VanessaE |
WHAT |
| 18:02 |
VanessaE |
that's even more important now than ever |
| 18:02 |
Jordach |
memory is your fucking chokepoint |
| 18:02 |
VanessaE |
we can't speed CPUs up anymore, and GPUs are starting to top out now too |
| 18:02 |
VanessaE |
(and when I say "CPU cycles" I also, by extension, mean GPU cycles) |
| 18:03 |
VanessaE |
if memory is your choke point, your memory is too damned slow or your workload is too big |
| 18:03 |
VanessaE |
cache misses are the problem if your memory is too slow |
| 18:03 |
VanessaE |
what's the common size of an L1 cache now? |
| 18:03 |
VanessaE |
64k? 256? I can't remember |
| 18:04 |
Calinou |
you could tile a texture on several node spaces |
| 18:04 |
rubenwardy |
CPU cycles are cheap. Data transfer is not. |
| 18:04 |
Calinou |
we really need that in Minetest, good to make textures that tile well |
| 18:04 |
rubenwardy |
Most of the time the CPU is waiting for RAM memory |
| 18:04 |
|
Gregor3000 joined #minetest |
| 18:04 |
Calinou |
(eg. 64 × 64 texture on 4 × 4 × 4 stone area) |
| 18:04 |
VanessaE |
Calinou: procedural texturing would help things too |
| 18:04 |
Calinou |
not necessarily procedural |
| 18:04 |
VanessaE |
but I see what you're saying |
| 18:04 |
Calinou |
but this would help a lot already, to have textures that tile well |
| 18:04 |
PenguinDad |
Calinou: http://sprunge.us/gAMB |
| 18:05 |
Calinou |
what's this, why? |
| 18:05 |
* Jordach |
sees a sokomine |
| 18:06 |
Calinou |
VanessaE, actually, we do get about 10 % performance increase each year on CPU, and about 20 % per year for GPU |
| 18:06 |
Calinou |
(we'll have 20 nm in 2015… everyone promised it for 2014, but it'll only be for 2015, same goes for 14 nm CPUs) |
| 18:07 |
Calinou |
DDR 4 next year too, probably |
| 18:07 |
Calinou |
so IGPs will be faster (since they rely heavily on system RAM) |
| 18:07 |
|
rmilan joined #minetest |
| 18:07 |
PenguinDad |
patch for carbone makes the lava cooling code more readable |
| 18:07 |
VanessaE |
well 10-20% is okay but really, that's hardly Moore's Law compliant. |
| 18:08 |
VanessaE |
point is, stop wasting CPU/GPU cycles to do something that can be done faster in other ways |
| 18:09 |
PenguinDad |
^ Calinou |
| 18:09 |
Calinou |
OK |
| 18:09 |
VanessaE |
if you can look up something faster in a table than you can calculate it, then use the table. |
| 18:10 |
VanessaE |
and that's not 1998 thinking, that's 1982 thinking, when we still used microcomputers and knew how to squeeze every last cycle out of the computer because we HAD to. |
| 18:10 |
VanessaE |
it's a lost art and honestly, it's sad :( |
| 18:10 |
rubenwardy |
Most of the time the CPU is waiting for RAM memory |
| 18:10 |
VanessaE |
rubenwardy: much of the time, but I wouldn't go so far as to say "most". |
| 18:11 |
VanessaE |
if it were "most", programs wouldn't be deadlocked at 100% in userspace doing whatever they do (it would be "system" time) |
| 18:11 |
rubenwardy |
As long as the table fits in the RAM cache, it should be fine |
| 18:11 |
VanessaE |
exactly |
| 18:12 |
|
Exio joined #minetest |
| 18:12 |
* twoelk |
misses well writen html not relaying on grafics |
| 18:13 |
Sokomine |
i'm curious as to how your mapgen now works, jordach :-) |
| 18:14 |
Calinou |
pushed, seems to work |
| 18:14 |
Calinou |
<VanessaE> it's a lost art and honestly, it's sad :( |
| 18:15 |
Calinou |
human time isn't cheap, computer time is |
| 18:15 |
VanessaE |
yeah but even human time done for fun? |
| 18:15 |
Calinou |
twoelk, flat design relies on graphics less than skeuomorphic design. |
| 18:15 |
VanessaE |
why does *every fucking thing* have to be measured in <insert currency here>? :( |
| 18:15 |
Calinou |
VanessaE, programming isn't fun to many people |
| 18:15 |
Calinou |
in maptools:copper_coin? :p |
| 18:16 |
VanessaE |
hah |
| 18:16 |
Calinou |
twoelk, so, we do use less images and such than before, because flat design is the new cool thing |
| 18:16 |
Calinou |
(and it's good) |
| 18:16 |
Exio |
Calinou, the problem is when people saying "meh fuck it" |
| 18:16 |
Calinou |
pages load faster, easier to design, less relying on image editing software |
| 18:16 |
VanessaE |
you could have at least said gold coins :P |
| 18:16 |
Exio |
"i could do it in 3 hours, but i'll do it in 2 hours and half" |
| 18:16 |
Exio |
"AND HAVE A 700% PERFORMANCE TRADEOFF" |
| 18:16 |
twoelk |
(oh so that is how you spell relies) |
| 18:16 |
Exio |
the problem is that computer time IS cheap, not infinite |
| 18:17 |
Calinou |
https://cdn.mediacru.sh/sm8h7j6G1fTr.png |
| 18:17 |
Calinou |
how the colour scheme looks now, with Minetest code |
| 18:17 |
twoelk |
long times ago I argued with adobe guys on computer fairs that you should not solve everything with graphics |
| 18:18 |
twoelk |
they listened but shoved me away when potential custumers showed up |
| 18:19 |
VanessaE |
Calinou: if I spend too long reading light-on-dark, I get raster burn :-/ |
| 18:19 |
VanessaE |
(ironically IRC and all my terminals are light-on-dark) |
| 18:19 |
VanessaE |
(but do it in an editor or on the web and ugh..) |
| 18:19 |
Calinou |
oh, I love light-on-dark |
| 18:20 |
Calinou |
I made the text bright grey, not white |
| 18:20 |
Calinou |
background is dark grey, not black |
| 18:20 |
Calinou |
to reduce contrast |
| 18:20 |
Calinou |
looks very easy on the eyes to me, all text is actually readable (unlike many syntax colourings where there will be cherry-coloured text on dark grey) |
| 18:20 |
Calinou |
comments are visible, they're even highlighted, so you don't forget them :D |
| 18:26 |
PenguinDad |
an invisible comment would be useless :D |
| 18:27 |
twoelk |
toggling comments can be usefull to see the flow of the code better |
| 18:29 |
Exio |
woo |
| 18:29 |
Exio |
argentina's economy is fucked up |
| 18:30 |
Exio |
can anyone tell me what your news channels are saying @ TV? |
| 18:33 |
|
paramat joined #minetest |
| 18:35 |
Calinou |
PenguinDad, many colour schemes make comments hard to read, or make them too discreet |
| 18:35 |
Calinou |
see https://medium.com/@MrJamesFisher/your-syntax-highlighter-is-wrong-6f83add748c9 for comments |
| 18:35 |
Calinou |
making them highlighted is probably a better idea |
| 18:37 |
Exio |
Calinou, anyway, if you need to use a lot of comments, something is wrong |
| 18:37 |
Exio |
unless you're coding in ASM |
| 18:37 |
|
rubenwardy joined #minetest |
| 18:37 |
Calinou |
it's good to have comments for stuff that may seem obvious, in 6 months it probably won't be obvious |
| 18:37 |
VanessaE |
Exio: no news to watch here (no TV); what's going on in Argentina? |
| 18:38 |
twoelk |
last news I heard about Argentina was "economy is in trouble" :-( |
| 18:39 |
Exio |
Calinou, use better function names, proper variables, try to solve just a single thing in a single function, make them have less side effects (noone if possible..) |
| 18:39 |
VanessaE |
twoelk: not good. comparable to the crisis the US had since 2007, or more like US Great Depression in 1929? |
| 18:40 |
Exio |
what you 1929 our 2001? |
| 18:40 |
Exio |
your* |
| 18:40 |
twoelk |
the word that made me listen was "insolvence" |
| 18:40 |
Exio |
it isn't going to be like 2001, if everything goes well, we'll have a super-devaluated currency! |
| 18:41 |
VanessaE |
Exio: 1929 and a few years beyond led to mass homelessness and unemployment. it was a HUGE crash for the US economy (and the rest of the world suffered then too) |
| 18:41 |
Exio |
VanessaE, what was the unemployment? |
| 18:41 |
twoelk |
didn't hear (car-radio) to what that applied though |
| 18:41 |
Exio |
+50%? |
| 18:41 |
Calinou |
Exio, you should still comment your code |
| 18:41 |
Calinou |
and add good spacing too |
| 18:41 |
Calinou |
code is like writing, it should feel good to read |
| 18:42 |
Exio |
Calinou, "x = randomNumber(seed) // get a random number into x, using <seed>" |
| 18:42 |
Exio |
ths? |
| 18:42 |
Exio |
this*? |
| 18:42 |
twoelk |
1929 was part of the fertilizer for the brown 30's in Germany |
| 18:43 |
Calinou |
I don't do that, Exio |
| 18:43 |
Calinou |
sounds = default.node_sound_wood_defaults(), -- Intended. |
| 18:43 |
Calinou |
eg. this for Mese block |
| 18:43 |
Calinou |
so that people don't report this as a bug, and to not make myself surprised when I look at it later |
| 18:43 |
Exio |
tell me what kind of comments do you need to specify in a normal high-language nowdays |
| 18:43 |
Exio |
Calinou, you should define a local function |
| 18:43 |
Exio |
"sounds" |
| 18:43 |
Jordach |
most of my animated chest code is UN commented |
| 18:44 |
Exio |
or a local table, or stuff |
| 18:44 |
Exio |
and make the code be obvious |
| 18:44 |
Exio |
also, try to avoid the "how the computer should solve this problem" and instead abstract the "low-level stuff" as much as possible |
| 18:45 |
Exio |
anyway, that isn't possible in dynamic languages :( |
| 18:45 |
Exio |
Calinou, (Generalized) Algebraic data types + Phantom types |
| 18:45 |
Jordach |
biggest lie of all time https://cdn.mediacru.sh/sNfHdtVkoiOf.png |
| 18:46 |
LemonLake |
Jordach: if it takes that long it means the subtitles are corrupted |
| 18:47 |
Jordach |
LemonLake, i mean the building of it |
| 18:47 |
Jordach |
>minute >takes 3 minutes |
| 18:48 |
Exio |
at least it finishes |
| 18:48 |
Exio |
if i do that here it'd take 15 minutes and then the system would crash |
| 18:48 |
Exio |
probably... |
| 18:49 |
Exio |
we need a hs to lua compiler! |
| 18:49 |
Exio |
or maybe a super-set of lua with a typesystem and algebraic datatypes |
| 18:49 |
Exio |
sounds kinda nice |
| 18:50 |
Exio |
a super-set of lua :D |
| 18:50 |
Exio |
sadly lua isn't lisp... |
| 18:50 |
Exio |
or it'd be easy ;P |
| 18:50 |
PenguinDad |
we need a [insert your favourite] to lua compiler! |
| 18:50 |
PenguinDad |
+ language |
| 18:51 |
Exio |
imo a superset of lua would be kinda neat |
| 18:51 |
Calinou |
MoonScript? |
| 18:51 |
Calinou |
you could write Minetest mods in it |
| 18:51 |
Calinou |
there's also Vortex, by q66 |
| 18:51 |
Calinou |
q66, (is that thing useful?) |
| 18:52 |
Exio |
Calinou, what about the types? |
| 18:52 |
* PenguinDad |
throws http://squirrel-lang.org/ in the channel |
| 18:52 |
Exio |
a typesystem + adts would be very useful when writing mods |
| 18:53 |
q66 |
<Calinou> q66, (is that thing useful?) |
| 18:53 |
q66 |
no |
| 18:53 |
q66 |
i put it on hiatus |
| 18:53 |
q66 |
with that i'll redesign it... one day |
| 18:53 |
q66 |
also moonscript sucks :P |
| 18:53 |
q66 |
just use lua |
| 18:54 |
Exio |
vortex sounds like what i would like to use |
| 18:54 |
q66 |
also integrating a typesystem with tables and metatables is painful and probably near impossible |
| 18:54 |
Exio |
i wouldn't want a sound typesystem |
| 18:55 |
q66 |
doesn't matter if sound or unsound |
| 18:55 |
q66 |
tables are dynamic |
| 18:55 |
q66 |
so they cannot easily be typed |
| 18:55 |
Exio |
yeah, i still think some kind of type safety in a few cases would be possible |
| 18:55 |
q66 |
i was experimenting with this |
| 18:55 |
q66 |
it's difficult |
| 18:56 |
q66 |
in my game engine i use a superset of lua https://github.com/quaker66/luacy |
| 18:56 |
q66 |
not typed though |
| 18:56 |
q66 |
just a few things i found handy |
| 18:57 |
paramat |
hmmmm is there any way to control or predict the order of 'on generated' functions of multiple terrain processing mods? Also, intersecting mod update with screenshot https://forum.minetest.net/viewtopic.php?f=11&t=9521&p=149438#p149438 |
| 18:58 |
rubenwardy |
paramat: it is FIFO |
| 18:58 |
rubenwardy |
So depends on loading of mods |
| 18:59 |
Krock |
paramat, +1, I support this idea |
| 18:59 |
rubenwardy |
you could probably do minetest.registered_on_generated.insert(1, function) -- inserts at first. |
| 19:00 |
paramat |
thanks |
| 19:00 |
rubenwardy |
* minetest.registered_on_generated.insert(function, 1) |
| 19:04 |
paramat |
can this 'loading of mods' be controlled by their dependancies? |
| 19:05 |
Krock |
yes ^ |
| 19:06 |
Krock |
but idk about the on_generated function |
| 19:08 |
rubenwardy |
If you depend on a mod, the do their on_generated first |
| 19:08 |
rubenwardy |
*they |
| 19:08 |
VanessaE |
rubenwardy: it's not FIFO either. |
| 19:08 |
VanessaE |
not if vmanip is involved. |
| 19:08 |
rubenwardy |
Huh? |
| 19:08 |
|
exoplanet joined #minetest |
| 19:08 |
VanessaE |
see the recent compatibility problem between snow mod and plants_lib |
| 19:09 |
VanessaE |
I think there's an issue filed against minetest engine for it |
| 19:09 |
rubenwardy |
It is run in FIFO, but there is no guarantee that the nodes are saved in time for your callback. |
| 19:09 |
VanessaE |
that's the problem. |
| 19:09 |
VanessaE |
either Lua is single-threaded or it isn't. |
| 19:09 |
rubenwardy |
So it is as if it is simultaneous. |
| 19:10 |
paramat |
no one seems to know for sure, that's why i'm asking the expert |
| 19:10 |
VanessaE |
simultaneous is a problem |
| 19:10 |
VanessaE |
a BIG problem |
| 19:11 |
PenguinDad |
paramat: hmmm isn't the only expert ;) |
| 19:11 |
rubenwardy |
You question was about the order of functions, which is pretty easy. FIFO. The order of application of changes is harder. |
| 19:11 |
paramat |
ah i see |
| 19:12 |
|
init joined #minetest |
| 19:13 |
Calinou |
FIFO? |
| 19:13 |
|
asie joined #minetest |
| 19:14 |
PenguinDad |
Calinou: First-In First-Out |
| 19:15 |
Calinou |
ah |
| 19:15 |
|
asie joined #minetest |
| 19:17 |
|
asie joined #minetest |
| 19:19 |
rubenwardy |
A quick google would tell you :P |
| 19:20 |
Calinou |
not Google |
| 19:20 |
Krock |
DDG |
| 19:21 |
rubenwardy |
DDG doesn't return anything relevant |
| 19:21 |
rubenwardy |
(exaggeration) |
| 19:21 |
PenguinDad |
rubenwardy: it does for me :/ |
| 19:21 |
Exio4 |
layer 8 problems |
| 19:23 |
* sfan5 |
licks Exio4 |
| 19:24 |
PenguinDad |
Layer 8 is often poorly implemented :D |
| 19:25 |
paramat |
it could be a race condition, so i will code mods to behave properly whatever the order of vmanip execution |
| 19:30 |
LemonLake |
why not have some sort of global table or something? |
| 19:33 |
VanessaE |
paramat: all I know is when snow mod started using vm's, plants_lib stopped working with it because snow's vm actions erase plants_lib's actions entirely. it's since been patched to at least allow plants_lib's non-snow-biome actions to work, but it's still partly busted. |
| 19:33 |
VanessaE |
it's an engine fault, because the two mods are meant to work together. |
| 19:35 |
VanessaE |
(and plants_lib does NOT use vm's) |
| 19:35 |
|
Exio joined #minetest |
| 19:40 |
Jordach |
fuck |
| 19:40 |
Jordach |
we're halfway through the year already |
| 19:40 |
VanessaE |
yep |
| 19:40 |
Krock |
yeah. noticed that too |
| 19:40 |
Krock |
some hours ago, it was sprin |
| 19:40 |
Krock |
g |
| 19:42 |
* sfan5 |
meows at Jordach |
| 19:43 |
Jordach |
GTV5 for PC looks fucking insane |
| 19:44 |
Jordach |
GTA* |
| 19:44 |
LemonLake |
oh wow it does |
| 19:45 |
hmmmm |
paramat: callbacks are performed in the order the callbacks were registered |
| 19:45 |
VanessaE |
hmmmm: nope.avi |
| 19:45 |
Calinou |
Jordach, like its hardware requirements, probably |
| 19:46 |
* Jordach |
pokes VanessaE about making a data cache for BFD that's moderately upto date |
| 19:46 |
VanessaE |
(maybe they're *supposed* to, but it doesn't work in practice :) ) |
| 19:46 |
hmmmm |
okay, why |
| 19:46 |
Exio |
then something is fucked up? |
| 19:46 |
hmmmm |
VanessaE: if you're saying that I'm wrong, that implies there is a bug somewhere |
| 19:47 |
VanessaE |
hmmmm: not saying you're wrong, per se. Rather that the engine is. |
| 19:47 |
VanessaE |
lemme search a sec. |
| 19:47 |
Exio |
he just said what the engine should do |
| 19:47 |
VanessaE |
there it is |
| 19:47 |
hmmmm |
the engine does not call each individual callbacks, it only calls builtin.lua:core.run_callbacks() |
| 19:47 |
VanessaE |
https://github.com/minetest/minetest/issues/1354 |
| 19:48 |
VanessaE |
hmmmm: this issue explains it better. |
| 19:48 |
VanessaE |
I don't like calling you wrong :) |
| 19:50 |
VanessaE |
Jordach: I'm gonna redo the caches soon; the main one is based mostly around dreambuilder's media but some of that is "outdated" relative to what you might want |
| 19:50 |
hmmmm |
wait a minute |
| 19:50 |
paramat |
thanks hmmmmm |
| 19:50 |
hmmmm |
let's think about this rationally for a minute |
| 19:50 |
hmmmm |
mapgen is finished, Map::finishBlockMake is called, VoxelManipulator is not destroyed at this point, but merely the contents have been blit back |
| 19:50 |
hmmmm |
THEN the callbacks are called in order of their registration |
| 19:51 |
VanessaE |
hmmmm: yes - but those callbacks' mapgen operations are *ignored* |
| 19:51 |
VanessaE |
or something. |
| 19:51 |
hmmmm |
an on_generated callback may obtain the mapgen object voxelmanip from that thread and the LuaVoxelManip wrapper is created around it |
| 19:52 |
hmmmm |
so let's say callback #1 uses this |
| 19:53 |
hmmmm |
at this point it should not need to call read_from_map because the contents are still loaded from the mapgen |
| 19:53 |
hmmmm |
but they still need to call write_to_map after they're done |
| 19:54 |
hmmmm |
okay |
| 19:54 |
hmmmm |
so the problem can exist if the voxelmanip-using callback happens after the set_node call one |
| 19:55 |
VanessaE |
more or less yeah |
| 19:55 |
hmmmm |
callback #1 does some set_node, so the Map is updated but the mapgen's VoxelManipulator contents are not |
| 19:55 |
VanessaE |
see: |
| 19:55 |
hmmmm |
then callback #2 using the voxelmanip thinks |
| 19:55 |
hmmmm |
"oh I can save a step and not read from the map!" |
| 19:55 |
VanessaE |
in the past, the snow mod used set_node()/get_node() and plants_lib opt-depended on it, so it just worked. |
| 19:55 |
hmmmm |
now let's be real here |
| 19:55 |
hmmmm |
this can NOT be fixed |
| 19:55 |
VanessaE |
now snow mod uses vm's to do its mapgen work |
| 19:55 |
Jordach |
VanessaE, what's the cURL line? |
| 19:56 |
LemonLake |
Jordach: ____ |
| 19:56 |
hmmmm |
it would require each mapgen's internal voxelmanip object to be updated from within Map |
| 19:56 |
* Jordach |
claps |
| 19:56 |
LemonLake |
More accurately: ____) |
| 19:56 |
VanessaE |
Jordach: remote_media = http://192.99.11.10/creative-survival-media/ |
| 19:56 |
hmmmm |
but the voxelmanipulator is not some external, persistant object |
| 19:56 |
Jordach |
VanessaE, got it |
| 19:56 |
hmmmm |
it lives for only one makeChunk call, and then gets deallocated |
| 19:57 |
sfan5 |
VanessaE: why no index.mth? |
| 19:57 |
VanessaE |
hmmmm: but we're talking about *one* mod that uses vm's within on_generated, competing with *one* mod that uses only traditional set_node/get_node calls within on_generated |
| 19:57 |
hmmmm |
if you were to really update each voxelmanipulator then you'd need the envlock back |
| 19:57 |
VanessaE |
sfan5: too lazy to fix. |
| 19:57 |
hmmmm |
or you'd need to add a new lock for mapgen voxelmanipulators |
| 19:57 |
Jordach |
remote_media = minetest.digitalaudioconcepts.com/ |
| 19:57 |
Jordach |
^ VanessaE |
| 19:57 |
hmmmm |
vanessae: okay, so we are, what's the point? |
| 19:57 |
Jordach |
just waiting for the folder name :P |
| 19:58 |
sfan5 |
VanessaE: https://gist.github.com/sfan5/6351560 |
| 19:58 |
VanessaE |
hmmmm: the point is that the changes are being written out-of-order relative to the mods' reads |
| 19:58 |
sfan5 |
Jordach: you probably need a http:// |
| 19:58 |
VanessaE |
hmmmm: you've got mod A reads map, mod B reads map, mod B writes map, mod A writes map. Ergo mod B's changes got overwritten. |
| 19:58 |
VanessaE |
hmmmm: it should be mod A reads, mod A writes, B reads, B writes. |
| 19:58 |
Jordach |
sfan5, at least my clipboard is synced between Host and Guest |
| 19:59 |
Krock |
we need priorities. |
| 19:59 |
hmmmm |
perhaps there should be a new voxelmanip object method that would just return a bool that is set if it's "dirty" |
| 19:59 |
hmmmm |
so then some code would be like |
| 19:59 |
hmmmm |
vm = minetest.get_mapgen_object("voxelmanip") |
| 19:59 |
hmmmm |
if vm.is_dirty() then |
| 19:59 |
hmmmm |
vm.read_from_map(emin, emax) |
| 19:59 |
hmmmm |
?? |
| 19:59 |
Krock |
else ? |
| 20:00 |
hmmmm |
if it's not in a dirty state, then the VM hasn't been invalidated yet and you don't need to do anything |
| 20:00 |
VanessaE |
hmmmm: you're proposing putting that in mod B's on_generated calls? |
| 20:00 |
hmmmm |
in all on_generated calls from this point forward |
| 20:00 |
VanessaE |
that's messy |
| 20:01 |
hmmmm |
got a better idea then? |
| 20:01 |
VanessaE |
we're talking about mods that *don't use* vm's at all here, being overwritten by mods that do |
| 20:01 |
PenguinDad |
Krock: else print("Yay, the voxelmanip isn't dirty!") |
| 20:01 |
PenguinDad |
end |
| 20:01 |
Krock |
:D |
| 20:02 |
VanessaE |
e.g. a mod that generates terrain using vm's is keeping a mod that uses traditional calls to add stuff to the terrain from being able to actually add its stuff because the terrain is overwriting those changes |
| 20:02 |
VanessaE |
despite the second mod depending on the first |
| 20:02 |
VanessaE |
(ergo the second mod always loads and runs last) |
| 20:02 |
hmmmm |
got a better idea then? |
| 20:02 |
VanessaE |
not at present |
| 20:03 |
VanessaE |
except that if there's a vm object "open" for a given mod, I can only say to lock it |
| 20:03 |
hmmmm |
perhaps there can be a convenience wrapper that keeps getting the data at one call, and validates the VM contents |
| 20:03 |
hmmmm |
like |
| 20:03 |
hmmmm |
vm.get_current_data() |
| 20:03 |
VanessaE |
if you're somehow running this stuff concurrently, or there's any way at all for another mod to read/write the map and there's a vm "open" for a given mapblock, you're gonna have to envlock that mapblock so that shit gets done in the order that it should, until the vm is "released" |
| 20:04 |
hmmmm |
lol |
| 20:04 |
VanessaE |
(please excuse me if I don't know the terms properly) |
| 20:04 |
hmmmm |
vanessa, none of this is concurrent. |
| 20:04 |
hmmmm |
alright |
| 20:04 |
VanessaE |
if it's not concurrent, then how are these two mods possibly able to move from a read A, write A, read B, write B pattern, ---> to read A, read B, write B, write A when neither mod has any control over that? |
| 20:05 |
hmmmm |
vanessae: no mod shall ever do that latter pattern |
| 20:05 |
VanessaE |
but that's exactly what's happening. |
| 20:05 |
hmmmm |
it'll have unpredictable results |
| 20:05 |
hmmmm |
no no |
| 20:05 |
VanessaE |
yes yes |
| 20:05 |
hmmmm |
mod A is not reading |
| 20:05 |
hmmmm |
it's just using data that had been invalidated thanks to the mod using set_data() |
| 20:05 |
hmmmm |
the only thing doing reading to the VM is the mapgen |
| 20:06 |
VanessaE |
"read" means in this case either getting a vm object or doing a get_node() call. |
| 20:06 |
hmmmm |
read to the mapgen |
| 20:06 |
hmmmm |
er voxelmanip |
| 20:06 |
hmmmm |
whatever |
| 20:06 |
hmmmm |
trust me, this is plenty clean |
| 20:07 |
hmmmm |
I'm going to add a new method to LuaVoxelManip called get_current_data() that'll check to see if the VM contents have been invalidated, and if so, make a new call to read_from_map() before finally returning the result of get_data() |
| 20:07 |
VanessaE |
point is, mod A does all it's work in its on_generated call, then mod B gets to run its on_generated calls. But mod B's on_generated calls, which I have confirmed are in fact being run, are being completely overwritten by mod A's |
| 20:07 |
hmmmm |
so you'll be able to do like |
| 20:07 |
hmmmm |
data = vm.get_current_data() |
| 20:07 |
|
grrk-bzzt joined #minetest |
| 20:07 |
|
grrk-bzzt joined #minetest |
| 20:07 |
hmmmm |
instead of data = vm.get_data() |
| 20:07 |
hmmmm |
it might be a little bit slower but |
| 20:07 |
paramat |
to clarify i don't particularly need control over the order, just curious. vanessa's problem is more important |
| 20:07 |
hmmmm |
it'll be consistent and there will be no data loss |
| 20:08 |
hmmmm |
VanessaE: what is the logical issue with my solution? |
| 20:08 |
hmmmm |
paramat: no, but you brought up an important issue |
| 20:08 |
hmmmm |
lol I'm supposed to be doing work right now :) |
| 20:08 |
VanessaE |
hmmmm: if your solution requires any changes to any mod, it's already wrong |
| 20:08 |
VanessaE |
that's the thing |
| 20:09 |
hmmmm |
there is literally NO other way to fix this |
| 20:09 |
VanessaE |
what? |
| 20:09 |
VanessaE |
when the callback exits, the vm should be deleted/released/marked dirty, whatever |
| 20:09 |
hmmmm |
yeah but |
| 20:09 |
hmmmm |
the thing that'll set the VM to dirty would be the set_node() call |
| 20:09 |
hmmmm |
hmmm |
| 20:10 |
hmmmm |
I can't set something to dirty in between callbacks, you'd have to modify builtin to do that |
| 20:10 |
hmmmm |
in run_callbacks() where the loop is |
| 20:12 |
* hmmmm |
ponders if what he's about to do is threadsafe |
| 20:15 |
hmmmm |
this invalidated flag would need to be in the actual VoxelManipulator object |
| 20:15 |
|
zat joined #minetest |
| 20:15 |
hmmmm |
since that's the only object persistent across all callbacks whereas LuaVoxelManip wrappers are not |
| 20:16 |
hmmmm |
there is one LuaVoxelManip object for each callback that gets associated with the very same VM |
| 20:17 |
VanessaE |
the thing is, you've got potentially dozens of other callbacks happening. that's why I'm dead set against adding to the API to fix this |
| 20:17 |
hmmmm |
then fix it yourself |
| 20:17 |
VanessaE |
wat? |
| 20:17 |
hmmmm |
i'm not a miracle worker |
| 20:17 |
VanessaE |
wtf? |
| 20:18 |
VanessaE |
look, this is simple - the callback happens, Lua gets its vm object, makes its changes, writes the results and exits, right? |
| 20:19 |
VanessaE |
I mean from an abstract point of view, that's what happens |
| 20:20 |
VanessaE |
when that callback exits, there should be nothing standing in the way of another callback coming along, twiddling some nodes on the now-written results (because the mod load order said "ok, this mod's callback comes next in the list") |
| 20:21 |
VanessaE |
someone needs to at least remove "not a bug" from the issue report. this IS an engine bug. I've been over my code with a fine-toothed comb. SPlizard has been over his code. I've looked at it too. |
| 20:21 |
VanessaE |
there's NO way this should be happening. |
| 20:21 |
hmmmm |
are you sure you're understanding the actual issue? |
| 20:22 |
hmmmm |
the mapgen voxelmanip contents stay the same as they were right after the core mapgen finished its job |
| 20:22 |
hmmmm |
mod A is writing to the map with set_node() |
| 20:22 |
VanessaE |
yes, I get that |
| 20:22 |
VanessaE |
nononono |
| 20:22 |
hmmmm |
this actually commits the changes immediately |
| 20:22 |
VanessaE |
stop there |
| 20:22 |
hmmmm |
what. |
| 20:22 |
VanessaE |
you're misunderstanding me. |
| 20:23 |
VanessaE |
Mod A gets a VM object and makes its changes. |
| 20:23 |
VanessaE |
it writes the map. |
| 20:23 |
hmmmm |
ok |
| 20:23 |
VanessaE |
Mod B comes along with set_node() and makes its changes. |
| 20:23 |
hmmmm |
sure |
| 20:23 |
hmmmm |
now what precisely is the problem |
| 20:23 |
VanessaE |
but somehow, mod B's changes, which took place *after* mod A's VM was written, are being *ignored* |
| 20:24 |
hmmmm |
now hold on a minute |
| 20:24 |
VanessaE |
and Mod B is NOT using a VM. |
| 20:24 |
hmmmm |
how do you *know* that mod B was called after |
| 20:24 |
VanessaE |
because Mod B depends on mod A? |
| 20:24 |
VanessaE |
(opt-depend via depends.txt that is) |
| 20:25 |
hmmmm |
did you actually print out to the console in on_generated and verify that, yes, mod A on_generated is called first, before mod B's? |
| 20:25 |
hmmmm |
because the scenario in which you depicted is quite impossible |
| 20:25 |
VanessaE |
yes, I did. |
| 20:25 |
VanessaE |
we went over it and over it. |
| 20:25 |
VanessaE |
because mod B (in this case it's plants_lib but the name is irrelevant) contains callbacks that depend on mod A's nodes. |
| 20:26 |
hmmmm |
as long as mod A is the one doing the voxelmanip stuff, this should not happen |
| 20:26 |
VanessaE |
right. |
| 20:26 |
VanessaE |
but it does. |
| 20:26 |
hmmmm |
I bet you're using voxelmanip wrong |
| 20:26 |
hmmmm |
there's literally no other logical way for this to happen |
| 20:26 |
hmmmm |
we're talking about logic here |
| 20:27 |
VanessaE |
well SPlizard is in charge of the snow mod ("mod A"), not me, but I've looked at the code and I couldn't see anything wrong with it. |
| 20:27 |
hmmmm |
link me? |
| 20:27 |
VanessaE |
no one even answered his last question from a month ago :( |
| 20:28 |
VanessaE |
https://forum.minetest.net/viewtopic.php?f=11&t=2290&p=142223#p142223 -- and the issue, https://github.com/minetest/minetest/issues/1354 |
| 20:28 |
VanessaE |
ah, that wasn't the link I wanted actually, but hey, that's a good place to start reading. |
| 20:29 |
VanessaE |
(I meant to paste the link to the first page and ultimately the github sources, which are here: https://github.com/Splizard/minetest-mod-snow/ ) |
| 20:30 |
|
smoke_fumus joined #minetest |
| 20:30 |
VanessaE |
I believe the code you'll be most interested in is here: |
| 20:30 |
VanessaE |
https://github.com/Splizard/minetest-mod-snow/blob/848f30543d08570368fd59df9f97b8d6d5ee02f3/src/mapgen_v6.lua#L19 |
| 20:30 |
VanessaE |
this is before his hack to try to get around this issue. |
| 20:31 |
hmmmm |
hmm |
| 20:31 |
hmmmm |
well it seems okay so far |
| 20:31 |
hmmmm |
what the heck is wrong with his perlin noise calls though |
| 20:31 |
VanessaE |
but it's pretty simple. he gets his vm object, modifies it, calcs lighting, and writes it. |
| 20:31 |
VanessaE |
idk, I'm sure he's since improved his perlin code. |
| 20:32 |
hmmmm |
why does he create a brand new perlin noise object for every single tile, why does he not use the perlin maps to begin with |
| 20:32 |
* hmmmm |
scratches head |
| 20:32 |
VanessaE |
the current version, post-hack is here: https://github.com/Splizard/minetest-mod-snow/blob/master/src/mapgen_v6.lua#L22 |
| 20:33 |
hmmmm |
the original mapgen code looks alright |
| 20:33 |
VanessaE |
still the same thing, with write-to-map if'd for the hack. |
| 20:33 |
VanessaE |
hm. he generates assho----er iceholes ;-) |
| 20:35 |
VanessaE |
now, you can compare that to plants_lib's calls if you want to, but they're basically the same deal - on_generated --> wrapper function --> bunch of get_node calls filtered through perlin and biome checks, followed by set_node, function calls, or spawn_tree() calls as needed, and then exit. no vm's there. |
| 20:35 |
VanessaE |
https://github.com/VanessaE/plantlife_modpack/blob/master/plants_lib/init.lua#L111 |
| 20:36 |
Calinou |
(unrelated to current discussion) I found this, may be worth reading for modders: http://blog.codinghorror.com/the-best-code-is-no-code-at-all/ |
| 20:37 |
|
Halo joined #minetest |
| 20:42 |
Exio |
Calinou, make the code more modular using high-order-functions and without caring too much about the underlying hardware, abstract the problem with nice syntax, etc :P (?) |
| 20:42 |
Exio |
function composition is quite nice |
| 20:43 |
hmmmm |
vanessae, you don't need to ever call voxelmanip.update_map() if you're calling from a mapgen voxelmanip object |
| 20:43 |
Krock |
sfan5, would you be so bored and may cross compile dillo-3.0.4 for windows? :3 |
| 20:45 |
hmmmm |
vanessae, could it be that there is more than one mod that uses voxelmanip? |
| 20:46 |
VanessaE |
hmmmm: if there is, I am not aware of it, however in this case no I'm certain that only snow mod uses vm's in the tests SPlizard and I were doing. |
| 20:46 |
VanessaE |
as I recall from the initial investigation, we or someone else narrowed it down to *just* snow + plantlife modpack |
| 20:47 |
VanessaE |
and still it bugged out |
| 20:47 |
paramat |
update_map is for the non mapgen object voxelmanip only |
| 20:47 |
VanessaE |
maybe you could mention the vm.update_map() comment on the issue? |
| 20:48 |
paramat |
sure |
| 20:48 |
VanessaE |
I meant hmmmm :) |
| 20:48 |
VanessaE |
but whoever :)( |
| 20:51 |
hmmmm |
really have no idea |
| 20:52 |
hmmmm |
i don't see how it could possibly happen |
| 20:54 |
VanessaE |
hmmmm: feel like trying it out? :) just comment out snow mod mapgen_v6.lua line 307 (and the corresponding end at 324), and run it with plantlife and undergrowth modpacks (easier to see if it's working that way) |
| 20:56 |
VanessaE |
unless he's added more hacks, what you should get is fairly plain ordinary minetest_game terrain, with snow biomes here and there, as if plantlife and undergrowth aren't even present. |
| 20:56 |
VanessaE |
or rather, what you're likely to get. |
| 20:57 |
VanessaE |
(what you should get is a wildly organic, natural, rough terrain with lots of plants and such) |
| 20:59 |
VanessaE |
and no, plants_lib's depends.txt doesn't mention snow mod (because I didn't keep it after the attempt failed) |
| 21:04 |
|
ImQ009 joined #minetest |
| 21:09 |
Exio |
VanessaE, you use xubuntu, no? |
| 21:09 |
Exio |
i need a dark theme, know any good one? |
| 21:09 |
|
werwerwer joined #minetest |
| 21:11 |
VanessaE |
I use xubuntu but with a light theme |
| 21:12 |
VanessaE |
("xfce-4.2" theme to be exact) |
| 21:12 |
VanessaE |
maybe xfce-dusk might work for you |
| 21:13 |
Exio |
my eyes already hurt |
| 21:13 |
Exio |
and i'm using a not-so-light theme.. |
| 21:13 |
|
proller joined #minetest |
| 21:14 |
* Jordach |
bagged 5 free games from steam |
| 21:14 |
PenguinDad |
o_O I just had a download speed of 200KB/s |
| 21:16 |
Calinou |
I'm surprised when I reach 500 KB/s |
| 21:16 |
|
paramat left #minetest |
| 21:17 |
Exio |
i'm surprised when i reach 30kb/s |
| 21:17 |
PenguinDad |
Calinou: that's my_usual_speed * 8 |
| 21:17 |
|
jin_xi joined #minetest |
| 21:18 |
Exio |
!c 500/8 |
| 21:18 |
MinetestBot |
62.5 |
| 21:19 |
PenguinDad |
Exio: I was talking about the 200KB/s |
| 21:19 |
Exio |
,wa 200/8 |
| 21:19 |
Krock |
!c 22/7 |
| 21:19 |
MinetestBot |
3.142857142857143 |
| 21:19 |
Exio |
!c 200/8 |
| 21:19 |
MinetestBot |
25.0 |
| 21:19 |
Krock |
omg. it has decimals! |
| 21:19 |
Krock |
*she |
| 21:22 |
jin_xi |
sup |
| 21:22 |
Jordach |
soap |
| 21:23 |
Krock |
catsoup |
| 21:25 |
Krock |
<3 http://www.whatsmyuseragent.com/ |
| 21:25 |
jin_xi |
you guise into dwarf fortress? new release is out |
| 21:25 |
jin_xi |
it har really nice trees |
| 21:25 |
jin_xi |
s |
| 21:27 |
VanessaE |
but..but...moretrees :( |
| 21:28 |
Krock |
lua (vm) generated trees are also nice |
| 21:28 |
jin_xi |
well, ascii graphics leaf more to the imagination but they are huge, bloom in spring and all that seasonal crap and they have roots! |
| 21:28 |
Jordach |
Krock, schematic trees are even faster |
| 21:29 |
Jordach |
just ask BFD |
| 21:29 |
|
AnotherBrick joined #minetest |
| 21:29 |
Krock |
BFD, are they faster? |
| 21:29 |
VanessaE |
<BFD> no fucking way :P |
| 21:30 |
jin_xi |
it always amazes me how dynamic df's ascii graphics are. ever tried embarking on a beach? surfs up |
| 21:31 |
|
Miner_48er joined #minetest |
| 21:31 |
Jordach |
VanessaE, schems are raw nodes |
| 21:31 |
VanessaE |
so? |
| 21:32 |
jin_xi |
nodules crudes |
| 21:32 |
Krock |
raw. |
| 21:32 |
Krock |
raw meat node? |
| 21:34 |
Jordach |
VanessaE, updated my media cache? |
| 21:35 |
VanessaE |
YOUR cache? :P |
| 21:35 |
Jordach |
well, you know |
| 21:35 |
Jordach |
Sokomine actually waited on UDP |
| 21:35 |
VanessaE |
I'll take care of it later tonight |
| 21:36 |
Jordach |
thank you :3 |
| 21:36 |
Jordach |
under bfdmedia |
| 21:36 |
VanessaE |
ehm... |
| 21:36 |
VanessaE |
you know, you could do it yourself too :P |
| 21:36 |
* Jordach |
doesn't have anywhere to put it for cURL |
| 21:40 |
PenguinDad |
VanessaE: does this look correct to you? https://cdn.mediacru.sh/sayfkMjUVvzt.png |
| 21:41 |
VanessaE |
compared to what? |
| 21:42 |
VanessaE |
oh. |
| 21:42 |
VanessaE |
did you comment-out the two lines I mentioned? |
| 21:42 |
|
ImQ009 joined #minetest |
| 21:43 |
PenguinDad |
VanessaE: just one line is commented out |
| 21:43 |
VanessaE |
what one line? |
| 21:44 |
VanessaE |
you have to comment-out lines 307 and 324 of snow mod mapgen_v6.lua (and delete/re-generate the map) for the glitch to show itself |
| 21:44 |
|
fu-fu joined #minetest |
| 21:45 |
VanessaE |
those lines are a hack SPlizard added to work around this conflict. all they do more or less is turn off the snow mod's vm if it's not actually in a snow biome area |
| 21:46 |
VanessaE |
which ultimately is a little bit faster, but in theory it shouldn't need to be turned off at all. nevertheless with those lines commented out you shouldn't be able to reproduce that screenshot. |
| 21:46 |
|
Exio joined #minetest |
| 21:46 |
VanessaE |
and with or without those lines, in a snow biome you should be able to see moretrees small firs. but you won't because of the conflict. |
| 21:47 |
|
crazyR joined #minetest |
| 21:49 |
|
Exio4 joined #minetest |
| 21:50 |
|
werwerwer joined #minetest |
| 21:51 |
|
jojoa1997 joined #minetest |
| 22:10 |
|
grrk-bzzt joined #minetest |
| 22:18 |
PenguinDad |
I couldn't investigate anything except that his code quite messy :/ |
| 22:18 |
PenguinDad |
+ is |
| 22:19 |
VanessaE |
PenguinDad: well just comment out those two lines, delete the map, let it re-gen, and take a screenshot of the same area |
| 22:19 |
VanessaE |
then show it to me |
| 22:19 |
VanessaE |
I'll be able to tell you if it's still glitching |
| 22:20 |
VanessaE |
(it's entirely possible for the glitch to have already been fixed in the engine but I doubt it in this case :P ) |
| 22:22 |
PenguinDad |
there was only one way for me to be able to see undergrowth and plantlife nodes naturally generated |
| 22:22 |
VanessaE |
which is? |
| 22:23 |
Exio4 |
\o/ |
| 22:23 |
PenguinDad |
I had to comment out line 316 |
| 22:24 |
VanessaE |
same difference. |
| 22:24 |
VanessaE |
that'll just be the equiv of not having snow installed, really |
| 22:24 |
VanessaE |
so you just experienced the glitch I was talking about. |
| 22:25 |
VanessaE |
congrats :) |
| 22:26 |
PenguinDad |
interesting but I'm going to sleep now |
| 22:30 |
VanessaE |
hmmmm: still think it's being done wrong in the Lua side? ;) |
| 22:31 |
|
Hirato_ joined #minetest |
| 22:32 |
hmmmm |
I really don't know |
| 22:32 |
hmmmm |
i stopped looking at it a while ago |
| 22:36 |
|
Eater4 joined #minetest |
| 22:38 |
VanessaE |
hmmmm: could you at least unmark the "not a bug" from the issue |
| 22:38 |
VanessaE |
(and correspondingly, unmark it from "feature request") |
| 22:40 |
Jordach |
tl;dr deathly spread IS OP |
| 22:45 |
Eater4 |
VanessaE: can you try to do /op on isis.inchra.net #minetest |
| 22:46 |
VanessaE |
Eater4: how? I don't have op there. |
| 22:46 |
VanessaE |
stupid services have been down forever |
| 22:47 |
Eater4 |
VanessaE: try, you might have Oo privs there, witch means you can op anywhere |
| 22:47 |
VanessaE |
nope.avi |
| 22:48 |
Exio4 |
>.< |
| 22:48 |
Eater4 |
only if shadowninja oped my bot! |
| 22:52 |
Eater4 |
I have control over #botwar though |
| 23:03 |
|
fling joined #minetest |
| 23:04 |
|
jojoa1997 joined #minetest |
| 23:14 |
Jordach |
00:14:15: ERROR[main]: Map::setNode(): Not allowing to place CONTENT_IGNORE while trying to replace "mapgen:deathly_grass_2" at (-212,2,144) (block (-14,0,9)) |
| 23:14 |
Jordach |
^ wat |
| 23:19 |
SylvieLorxu |
s/deathly/deadly/ |
| 23:25 |
VanessaE |
Jordach: your node is undefined at that point. |
| 23:26 |
VanessaE |
it places IGNORE if you try to set a node with a bad/typoed name |
| 23:26 |
VanessaE |
(and then fails with that error) |
| 23:28 |
twoelk |
for those runnung the windows version of the Minetest-mapper, you may want to test this https://www.dropbox.com/s/tji2lhjknpcnmq0/wizapmapper.zip (just added some validity checking) |
| 23:30 |
Exio4 |
:( |
| 23:30 |
Exio4 |
i can't install moar stuff |
| 23:46 |
|
twoelk joined #minetest |
| 23:49 |
|
Exio joined #minetest |
| 23:55 |
|
luizrpgluiz joined #minetest |
| 23:55 |
luizrpgluiz |
hi |
