| Time |
Nick |
Message |
| 00:03 |
hmmmm |
RealBadAngel, well that's very odd, because there's no logical way that can happen |
| 00:03 |
hmmmm |
vanessa, if you and pilzadam can't find a workaround to the variable number of arguments, i'll remove that |
| 00:04 |
PilzAdam |
I alrady pushed the solution for that |
| 00:04 |
hmmmm |
does that fix third party mods too? |
| 00:04 |
VanessaE |
hmmmm: it's already fixed and seems to work fine for e.g. mesecons |
| 00:04 |
hmmmm |
okay, great |
| 00:06 |
RealBadAngel |
hmmmm, it looks like placing tree nodes is skipped, but not apples |
| 00:06 |
hmmmm |
again, not possible |
| 00:06 |
hmmmm |
make_tree is never even called |
| 00:06 |
RealBadAngel |
so who placed all those apples there? :) |
| 00:06 |
hmmmm |
something else must've happened |
| 00:07 |
RealBadAngel |
your update with trees happened |
| 00:07 |
hmmmm |
i'll look at it myself |
| 00:07 |
RealBadAngel |
disable trees and create new world |
| 00:09 |
PilzAdam |
RealBadAngel, I cant reproduce this |
| 00:10 |
hmmmm |
i'm looking for floating apples, and i'm not finding |
| 00:10 |
hmmmm |
pilzadam.... shadows on water seem blocky now |
| 00:10 |
hmmmm |
i think it was your update with new_style_water perhaps |
| 00:10 |
hmmmm |
shadows are still smooth on not-water |
| 00:12 |
hmmmm |
well, good thing i can't find the bug that RBA is talking about, because that would be completely illogical |
| 00:15 |
|
PilzAdam joined #minetest-dev |
| 00:27 |
PilzAdam |
hmmmm, http://irc.minetest.ru/minetest-dev/2013-01-23#i_2832524 |
| 00:31 |
RealBadAngel |
hmmm, will try to hunt the bug later |
| 00:32 |
RealBadAngel |
now im busy polishing 6d facedir for release |
| 01:14 |
|
jin_xi joined #minetest-dev |
| 01:50 |
hmmmm |
hey pilzadam, it'd be neat if we had a cooling sound effect for when water touches lava |
| 02:00 |
hmmmm |
also, when chunksize != 5, on_generate seems to give problems; it calls PseudoRandom.next() with max < min, but it doesn't tell me *where* the error happened |
| 02:00 |
hmmmm |
anybody have an idea on how to track something like this down? |
| 02:06 |
hmmmm |
seems to happen with basically any chunksize greater than 5.. specifically the numbers are 62, 64, -62, -64, or some variation thereof, if there's any significance to that. it happens in one of the generate_ores. |
| 02:34 |
hmmmm |
on an unrelated note, i tried setting buildable_to to false for content_ignore, and that didn't fix the node placement problem at all.. it might be that the initial analysis of the problem was wrong |
| 02:34 |
|
ShadowNinja joined #minetest-dev |
| 02:35 |
RealBadAngel |
hmmm, now we are dealing with prerotated node's textures |
| 02:35 |
RealBadAngel |
like pistons, or stuff like this |
| 02:35 |
RealBadAngel |
its pain in the ass and not correct |
| 02:36 |
RealBadAngel |
question is we shall go in direction that node shall be defined with nonrotated textures only |
| 02:36 |
RealBadAngel |
and rotations will make facedir only |
| 02:36 |
RealBadAngel |
or take care of sins of the past and got code partially fucked up |
| 02:37 |
RealBadAngel |
facedir alone cannot check if a texture is prerotated |
| 02:38 |
VanessaE |
better honor the manual rotations |
| 02:38 |
hmmmm |
we should do things the right way perhaps |
| 02:38 |
VanessaE |
you never know what kind of weird corner case might need it |
| 02:38 |
RealBadAngel |
no corner cases this way |
| 02:38 |
RealBadAngel |
0 rotations when definin a node |
| 02:38 |
RealBadAngel |
simple as that |
| 02:39 |
RealBadAngel |
is 0 simple? isnt it? |
| 02:39 |
RealBadAngel |
:) |
| 02:39 |
|
khonkhortisan joined #minetest-dev |
| 02:39 |
VanessaE |
corner case: |
| 02:39 |
VanessaE |
well, no strike that |
| 02:39 |
RealBadAngel |
thx to prerotations we cannot make it right |
| 02:39 |
RealBadAngel |
we got right textures flipped |
| 02:41 |
RealBadAngel |
note: problem is raising only when prerotated node is given new extra facedir values |
| 02:41 |
RealBadAngel |
with old range it is ok |
| 02:42 |
RealBadAngel |
but trying to rotate prerotated and trying to keep it lookin ok, fucks up properly defined ones |
| 02:43 |
RealBadAngel |
and introducing flips on textures |
| 02:43 |
RealBadAngel |
VanessaE, go on |
| 02:43 |
VanessaE |
nevermind. |
| 02:43 |
RealBadAngel |
:) so? |
| 02:43 |
VanessaE |
I withdraw the question, your honor :D |
| 02:43 |
RealBadAngel |
so, lets do it right |
| 02:44 |
RealBadAngel |
hmmmm, agree? |
| 02:44 |
hmmmm |
yes, remove manual rotations and do things the right way |
| 02:44 |
VanessaE |
but only when they're applied to a node with facedir |
| 02:44 |
RealBadAngel |
aye sir |
| 02:44 |
hmmmm |
thank you :) |
| 02:45 |
hmmmm |
we're finally cleaning crap up |
| 02:45 |
khonkhortisan |
yep |
| 02:45 |
RealBadAngel |
VanessaE, we r talkin facedir only now |
| 02:45 |
VanessaE |
right. |
| 03:33 |
VanessaE |
now, side note: can we please have the game drop focus from the world when the F10 console is open? |
| 03:33 |
VanessaE |
(so that moving the mouse moves a mouse pointer instead of panning/tilting) |
| 03:41 |
khonkhortisan |
that's really easy to do |
| 03:41 |
khonkhortisan |
I think? |
| 03:41 |
VanessaE |
beats me. |
| 05:20 |
|
proller joined #minetest-dev |
| 05:41 |
hmmmm |
you know, i find it absolutely hilarious that luajit was dismissed back then |
| 05:41 |
hmmmm |
it's being used by quite a lot of people now and they all say there is a very noticable improvement in everything |
| 05:41 |
|
troller joined #minetest-dev |
| 05:42 |
hmmmm |
in my mind, anybody who says that a speed improvement isn't worth it is a complete idiot and must be disregarded for the safety of the project |
| 06:01 |
|
proller__ joined #minetest-dev |
| 06:05 |
|
proller joined #minetest-dev |
| 06:21 |
|
loggingbot_ joined #minetest-dev |
| 06:21 |
|
Topic for #minetest-dev is now Minetest core development and maintenance. Chit-chat goes to #minetest. Consider this instead of /msg celeron55. http://irc.minetest.ru/minetest-dev/ http://dev.minetest.net/ |
| 06:25 |
|
proller__ joined #minetest-dev |
| 08:59 |
celeron55 |
23:37:17 < troller> idea about sources: maybe use automatic formater like astyle or indent or... ? |
| 08:59 |
celeron55 |
i shot this one down years ago, and still will |
| 08:59 |
celeron55 |
i think the programmer should have the freedom to format his code whatever way he wants |
| 08:59 |
celeron55 |
it's a human right |
| 09:00 |
celeron55 |
using automatic formatter is like removing the steering wheel from a car; it just isn't fun anymore |
| 09:13 |
|
proller joined #minetest-dev |
| 09:52 |
|
PilzAdam joined #minetest-dev |
| 10:20 |
|
realbadangel_ joined #minetest-dev |
| 10:20 |
|
ShadowNi1ja|Away joined #minetest-dev |
| 10:22 |
|
Exio joined #minetest-dev |
| 11:00 |
|
darkrose joined #minetest-dev |
| 11:00 |
|
darkrose joined #minetest-dev |
| 11:01 |
|
jin_xi joined #minetest-dev |
| 12:38 |
|
proller joined #minetest-dev |
| 12:44 |
|
hmmmm joined #minetest-dev |
| 13:23 |
|
troller joined #minetest-dev |
| 13:31 |
|
troller joined #minetest-dev |
| 13:53 |
|
serengeor joined #minetest-dev |
| 14:11 |
|
Taoki joined #minetest-dev |
| 14:25 |
realbadangel_ |
hi all |
| 14:26 |
realbadangel_ |
hmmmm, agree about luajit. especially when speedup cost absolutely nothin |
| 14:30 |
realbadangel_ |
we are all looking for ways to speed up our code. hard |
| 14:31 |
realbadangel_ |
and it seems like at least 100% is lyin on the street. just to pick it up |
| 14:53 |
hmmmm |
hopefully with the ores-in-core and decorationdef, we can remove the call to on_generated completely if nothing is registered, trimming a decent 150ms from chunk generation time |
| 14:54 |
hmmmm |
150ms doesn't seem like a lot until you realize that there's more than one chunk being generated |
| 14:54 |
realbadangel_ |
hell more than one |
| 14:56 |
realbadangel_ |
btw i thought a bit about apple bug |
| 14:56 |
realbadangel_ |
i think i know from where it came from |
| 14:57 |
|
proller joined #minetest-dev |
| 14:57 |
realbadangel_ |
vanessae disabled tree generation by changing aliases to air |
| 14:58 |
realbadangel_ |
now propably we do have new mapgen alias: apple , right? |
| 14:58 |
hmmmm |
ah, that explains it |
| 14:58 |
hmmmm |
yes, apple |
| 14:59 |
realbadangel_ |
tree is being generated but to air |
| 14:59 |
PilzAdam |
mapgen_apple |
| 14:59 |
realbadangel_ |
btw, what a waste of cycles... |
| 14:59 |
hmmmm |
it is a waste of cycles until we figure out the decorationdef crap |
| 15:00 |
hmmmm |
that's going to be the ultimate solution to all of this, i imagine |
| 15:00 |
realbadangel_ |
dont imagine. bet |
| 15:00 |
hmmmm |
well i don't know because i don't have any of that coded out, only the ore generation |
| 15:01 |
realbadangel_ |
havin an idea really helps to get it coded |
| 15:03 |
realbadangel_ |
start from definin empty biomes |
| 15:03 |
realbadangel_ |
and then lets fill them with contens just |
| 15:04 |
hmmmm |
well the original idea behind decorationdef was to have a biome contain one or more of these, and the trees (or other things) would be PseduoRandomly scattered throughout the biome |
| 15:04 |
hmmmm |
eliminating the need for tree perlin noise |
| 15:04 |
hmmmm |
and so on |
| 15:04 |
hmmmm |
but how do i frame this into a mapgen v6 context, with no biome support |
| 15:05 |
realbadangel_ |
does v7 support it? |
| 15:05 |
hmmmm |
well yes, that's the whole point of v7 and that's what makes it so different |
| 15:05 |
hmmmm |
the biomes |
| 15:06 |
realbadangel_ |
remember today mornin? my question? |
| 15:06 |
hmmmm |
what was your question? |
| 15:06 |
realbadangel_ |
about doin something right |
| 15:06 |
hmmmm |
remove the manual rotations? |
| 15:06 |
realbadangel_ |
yes |
| 15:06 |
hmmmm |
i was in that conversation |
| 15:06 |
hmmmm |
i said yes |
| 15:06 |
realbadangel_ |
do it right |
| 15:07 |
realbadangel_ |
so |
| 15:07 |
hmmmm |
oh |
| 15:07 |
realbadangel_ |
do it |
| 15:07 |
hmmmm |
that'd change a little too much and it'd be a whole different mapgen |
| 15:07 |
|
troller joined #minetest-dev |
| 15:07 |
hmmmm |
it still can be done right, i just need to think harder of how i'm going to do it |
| 15:07 |
* realbadangel_ |
marks in the chan thin red line |
| 15:07 |
realbadangel_ |
dare to cross? :) |
| 15:08 |
hmmmm |
meh |
| 15:08 |
hmmmm |
i don't |
| 15:08 |
realbadangel_ |
chicken :P |
| 15:09 |
realbadangel_ |
but seriously |
| 15:09 |
realbadangel_ |
such decisions have to be made |
| 15:09 |
realbadangel_ |
otherwise we will have all the time, desert, desert, some desert, beach, oh tree, and water |
| 15:10 |
hmmmm |
again - just wait for mapgen v7 |
| 15:10 |
hmmmm |
anyway i have an idea |
| 15:10 |
hmmmm |
a part of decorationdef could be placement style |
| 15:10 |
hmmmm |
perlin noise placed, pseudorandomly placed, so on |
| 15:11 |
realbadangel_ |
can i have an idea? |
| 15:11 |
hmmmm |
for example like we have right now with deserts and cactuses, that would be defined by a model, its perlin noise occurance, its div size |
| 15:11 |
hmmmm |
huh. |
| 15:11 |
hmmmm |
oh nevermind, i don't even need a placement style |
| 15:12 |
realbadangel_ |
plant noise. for whole world |
| 15:12 |
hmmmm |
everything follows from perlin noise + pseduorandom div placement |
| 15:12 |
hmmmm |
that would be included with this idea |
| 15:12 |
realbadangel_ |
then when noise tells you to plant something |
| 15:12 |
hmmmm |
in order to have "plant noise for the whole world" you just define all these decorations and use the same exact perlin noise parameters |
| 15:12 |
realbadangel_ |
check where you are |
| 15:12 |
realbadangel_ |
to put correct plant |
| 15:13 |
hmmmm |
alright yes this is good |
| 15:13 |
hmmmm |
i can do this |
| 15:13 |
realbadangel_ |
so the main noise is the trigger |
| 15:13 |
realbadangel_ |
to plant |
| 15:14 |
hmmmm |
no no |
| 15:14 |
hmmmm |
you can have separate noises |
| 15:14 |
realbadangel_ |
main plant noise |
| 15:14 |
hmmmm |
why would you want that? |
| 15:15 |
realbadangel_ |
to be it an ecosphere |
| 15:15 |
realbadangel_ |
undepedent |
| 15:15 |
hmmmm |
but you do understand that 'plants' can be absolutely anything, including mushrooms in caves and so on |
| 15:15 |
hmmmm |
stones |
| 15:15 |
realbadangel_ |
exactly |
| 15:15 |
hmmmm |
glowstone things in the nether |
| 15:16 |
hmmmm |
so it would be weird if you had all these things exactly correlated |
| 15:16 |
realbadangel_ |
when ecosphere tells you to grow |
| 15:16 |
realbadangel_ |
check where you are |
| 15:16 |
hmmmm |
but it's not necessarily a biological process |
| 15:16 |
realbadangel_ |
and what can be grown here |
| 15:17 |
realbadangel_ |
when youre close to the borders of biomes |
| 15:17 |
realbadangel_ |
make it fuzzy |
| 15:17 |
hmmmm |
but there's only two biomes, desert and normal |
| 15:17 |
hmmmm |
lol, i'm talking about V6. |
| 15:17 |
realbadangel_ |
im talkin bout the future :) |
| 15:19 |
realbadangel_ |
btw could you take a look at chasm and riverbeds mods? |
| 15:19 |
realbadangel_ |
c55 asked me to check chasm with luajit and i found that mods very nice |
| 15:20 |
hmmmm |
i would, but not today, i have a big paper due tomorrow that i didn't even start on |
| 15:20 |
realbadangel_ |
sure |
| 15:20 |
realbadangel_ |
but fyi those things shall be hardcoded |
| 15:20 |
realbadangel_ |
imho at least |
| 15:21 |
realbadangel_ |
such things makin landscape |
| 15:21 |
realbadangel_ |
and bringing variety |
| 15:21 |
hmmmm |
what, the chasms? |
| 15:21 |
realbadangel_ |
and theyre pretty cool |
| 15:21 |
realbadangel_ |
chasms and riverbeds |
| 15:22 |
hmmmm |
chasms would be neat, they can be added to mapgen v6 if done well enough, just that you'd need to specify the 'chasm' flag to enable |
| 15:22 |
hmmmm |
just as long as it doesn't change anything fundamental with v6 |
| 15:22 |
realbadangel_ |
take a look at mod |
| 15:22 |
realbadangel_ |
its pretty simple |
| 15:22 |
hmmmm |
the reason why proller's modifications had to go into a separate mapgen was because it modified several fundamental structures and what not |
| 15:24 |
realbadangel_ |
stoppin sun and moving earth was also a fundamental change |
| 15:24 |
realbadangel_ |
and hard to make |
| 15:47 |
|
proller joined #minetest-dev |
| 15:53 |
PilzAdam |
wtf https://travis-ci.org/minetest/minetest/builds/5628096 |
| 15:54 |
hmmmm |
probably just a hiccup |
| 16:01 |
PilzAdam |
do we want this? https://github.com/minetest/minetest/pull/409 |
| 16:03 |
khonkhortisan |
yes, I should be able to grab a ladder in front of my eyes with my hands, and not have to jump onto it with my feet |
| 16:03 |
PilzAdam |
ehm |
| 16:03 |
khonkhortisan |
unless I misunderstand what that does |
| 16:04 |
PilzAdam |
you do :-) |
| 16:05 |
PilzAdam |
it fixes "elevators" like this where you only have to hold sneak and jump: http://www.zimg.eu/i/250259109 |
| 16:06 |
|
Jordach joined #minetest-dev |
| 16:28 |
|
Calinou joined #minetest-dev |
| 16:41 |
ShadowNinja |
PilzAdam: I think fixing it is fine now that we can go fast on ladders |
| 16:46 |
VanessaE |
I disagree, sneak elevators is unique to minetest, and doesn't really differ from what someone could do (in theory) in reality |
| 16:46 |
VanessaE |
(aside from them not needing any supports in the game) |
| 16:47 |
VanessaE |
besides, fast on ladders is busted anyway |
| 16:47 |
VanessaE |
you should not be able to shoot 20m up off the top. |
| 16:48 |
khonkhortisan |
that should be reserved for swimming |
| 16:48 |
Jordach |
i would like to highlight realistically, a person who takes in parkour could actually use a sneak elevator |
| 16:49 |
jin_xi |
while carrying 30m^3 of stone and wielding an axe |
| 16:49 |
Jordach |
still possible |
| 16:50 |
VanessaE |
wtf is parkour? |
| 16:50 |
Jordach |
VanessaE, shortest point for A > B |
| 16:50 |
VanessaE |
oh |
| 16:51 |
Jordach |
which means: http://www.youtube.com/watch?v=p-OUNPqRzII |
| 16:52 |
khonkhortisan |
that looks dangerous, even in a game |
| 16:52 |
Jordach |
marktraceur does it |
| 16:54 |
hmmmm |
or more realistically, http://www.youtube.com/watch?v=KwqW_lg7X68 |
| 16:56 |
khonkhortisan |
that double wall flip was almost a headkick |
| 16:58 |
VanessaE |
*watches hmmmm's video* |
| 17:00 |
VanessaE |
(whoever ripped this video needs lessons in smooth framerates) |
| 17:00 |
hmmmm |
it's a really old video, is it not? |
| 17:00 |
VanessaE |
dunno |
| 17:00 |
hmmmm |
uploaded 2006 |
| 17:01 |
VanessaE |
obviously ripped from a video tape though |
| 17:16 |
thexyz |
great, windows.h defines some weird std::min/std::max replacements |
| 17:16 |
thexyz |
http://stackoverflow.com/questions/2789481/problem-calling-stdmax |
| 17:17 |
hmmmm |
let's use MYMIN and MYMAX |
| 17:17 |
thexyz |
we can (std::min)(a, b) as well |
| 17:18 |
thexyz |
or #define NOMINMAX somewhere |
| 17:18 |
thexyz |
which is better, i guess |
| 17:18 |
thexyz |
before including windows.h, that is |
| 17:19 |
khonkhortisan |
If it's different across oses, the winner is the way that's most common. |
| 17:19 |
thexyz |
https://github.com/FessWolf?tab=activity |
| 17:20 |
khonkhortisan |
not the best place to put a comment, right in the middle of the function call |
| 17:21 |
PilzAdam |
thexyz, Id say #define NOMINMAX to prevent doing the same error again |
| 17:22 |
thexyz |
I agree, now, where does it include windows.h? |
| 17:23 |
celeron55 |
put it directly in the cmake configuration |
| 17:23 |
celeron55 |
trying to make it always go before windows.h via source files is a pain |
| 17:23 |
khonkhortisan |
debug.h filesys.cpp gettext.h porting.h (twice!) socket.cpp |
| 17:24 |
thexyz |
mapgen_indev.obj : error LNK2001: unresolved external symbol "public: virtual void __thiscall MapgenV6::makeChunk(class BlockMakeData *)" (?makeChunk MapgenV6@@UAEXPAVBlockMakeData@@@Z) |
| 17:24 |
thexyz |
odd |
| 17:24 |
khonkhortisan |
winresource.rc |
| 17:25 |
celeron55 |
thexyz: http://stackoverflow.com/questions/9017573/define-preprocessor-macro-through-cmake |
| 17:28 |
celeron55 |
maybe http://paste.dy.fi/Y8k |
| 17:28 |
celeron55 |
umm actually, not there but upper so that it's in the WIN32 block 8) |
| 17:30 |
hmmmm |
khonkhortisan, what are those, places where windows.h is included? |
| 17:31 |
hmmmm |
it'd be nice if someone defined _WINNT_VERSION and remove those #ifndef blah #define blah 0x0600s |
| 17:31 |
hmmmm |
i needed that so MingW would compile with OpenThread() being used |
| 17:31 |
khonkhortisan |
yes |
| 17:34 |
khonkhortisan |
Does it matter if luajit compiles to a .so? Can I still copy it to liblua.a? It seemed to work. |
| 17:36 |
hmmmm |
not really sure |
| 17:37 |
|
iqualfragile joined #minetest-dev |
| 17:40 |
thexyz |
that's pointless |
| 17:40 |
khonkhortisan |
what, luajit? |
| 17:40 |
thexyz |
no |
| 17:40 |
thexyz |
msvc linking problems |
| 17:46 |
hmmmm |
thexyz, if you figure out why that virtual function won't link, tell me, because i don't see why it wouldn't work either |
| 17:49 |
thexyz |
sure |
| 17:51 |
PilzAdam |
hmmmm, do you mean stonebricks (as added to minetest_game) by "Smoothstone" in the TODO in the wiki? |
| 17:52 |
hmmmm |
i guess... are those the half-height slabs that you stack onto eachother to make a full, smooth slab? |
| 17:52 |
hmmmm |
what i'm talking about was in minecraft classic |
| 17:52 |
PilzAdam |
kinda |
| 17:52 |
hmmmm |
i guess that's it then |
| 17:53 |
hmmmm |
anyway did you ever find a cooling sound for when lava and water meet? |
| 17:54 |
|
proller joined #minetest-dev |
| 17:55 |
Exio |
hmmmm: hm, i don't think you meant those, http://forum.minetest.net/viewtopic.php?id=509 (the stone squares) |
| 17:56 |
hmmmm |
oh i don't |
| 17:56 |
hmmmm |
if i find a picture i'll show you |
| 17:57 |
hmmmm |
back when i actually played minetest, i remember i ripped the texture from minecraft and defined my own node that looked like i wanted |
| 17:57 |
Exio |
the bricks are using the texture of cobblestone |
| 17:57 |
Exio |
stone bricks, not bricks * :P |
| 18:01 |
|
rubenwardy joined #minetest-dev |
| 18:01 |
khonkhortisan |
Can I change "unkn own block" to "unkn own node"? |
| 18:04 |
khonkhortisan |
I think I will. |
| 18:37 |
|
troller joined #minetest-dev |
| 18:48 |
|
BlindBanana joined #minetest-dev |
| 18:54 |
|
BlindBanana joined #minetest-dev |
| 21:35 |
troller |
its possible to make multithread mapgen? |
| 21:35 |
VanessaE |
that's being worked on |
| 21:35 |
PilzAdam |
mapgen is already multthreaded, isnt it? |
| 21:35 |
VanessaE |
it already is multi-threaded now |
| 21:36 |
VanessaE |
but the multi-threading capability is not considered "stable" yet. |
| 21:36 |
Exio |
what is the definition of 'stable'? |
| 21:37 |
VanessaE |
Exio: works without risking lockups and uses all available CPU when (and only when) needed? |
| 21:37 |
troller |
now its uses only one cpu at 100% |
| 21:37 |
Exio |
i don't get it |
| 21:37 |
Exio |
troller: emerge_threads=<threadnumber> |
| 21:37 |
troller |
i want to 4-8 paralel blocks gen |
| 21:37 |
troller |
thanks |
| 21:37 |
VanessaE |
Exio: at present, some parts of the mapgen have sort of build-in waiting periods for no apparent reason. |
| 21:38 |
Exio |
for no apparent reason? you mean lua? :) |
| 21:38 |
VanessaE |
in other places, some parts of the mapgen get stuck waiting for other parts to finish, but really shouldn't. |
| 21:38 |
VanessaE |
hmmmm can explain it better, but that's good enough i thinl |
| 21:38 |
VanessaE |
thnk |
| 21:38 |
VanessaE |
.... |
| 21:38 |
Exio |
hehe |
| 21:38 |
Exio |
poor VanessaE |
| 21:39 |
Exio |
the "single thread performance" here is not perfect, and a phenom has more IPC |
| 21:39 |
Exio |
*but* if i use the other cores, i see a speedup |
| 21:39 |
Exio |
it can be a lot faster, i don't say it can't be |
| 21:47 |
VanessaE |
I currently run my server with two mapgen threads. Don't wanna break it. |
| 21:47 |
Exio |
my singleplayer works with 8 threads |
| 21:48 |
PilzAdam |
what is the default? 2*cores - 2? |
| 21:48 |
Exio |
1 |
| 21:48 |
Exio |
if "enabled autodetect", cores-2 (or 1 if negative/0) |
| 22:21 |
|
jojoa1997|Tablet joined #minetest-dev |
| 22:21 |
jojoa1997|Tablet |
when will mobs be added back. i kno they were in lua but why get rid of them when they were faster |
| 22:25 |
Exio |
the old mobs were in the core, engine, C++ |
| 22:26 |
Exio |
there are some pull requests for adding path-finding algorythms, what should be added soon(tm) |
| 22:26 |
Exio |
i'd say, don't panic and keep touched with minetest! |
| 22:26 |
Exio |
k |
| 23:20 |
|
iqualfragile joined #minetest-dev |
| 23:35 |
Exio |
something about the issues in the repository |
| 23:35 |
Exio |
does anyone with 'admin' perms in it is going to "clean" and "mark" them? |
| 23:35 |
Exio |
there are a lot of what are, or resolved, or are feature requests, or.. and so |
| 23:36 |
Exio |
having a way for having only bugs, or only feature requests and so would be nice |
| 23:37 |
PilzAdam |
yea, I can go through them tomorow or so |
| 23:37 |
Exio |
thanks |
| 23:46 |
Exio |
hahaha |
| 23:46 |
Exio |
314 files changed, 43793 insertions(+), 14772 deletions(-) |
| 23:46 |
Exio |
my "fork" |
