| Time |
Nick |
Message |
| 00:37 |
|
doserj joined #minetest-dev |
| 01:07 |
Exio |
talking about the mapgen, any one know why it isn't using/changing/whatever my mapgen alias? |
| 01:08 |
Exio |
i changed tree/leaves/apple to air, and water source to lava |
| 01:08 |
Exio |
and i still see trees and water :P |
| 01:08 |
Exio |
in a new generated world btw |
| 01:41 |
hmmmm |
realbadangel, what |
| 01:59 |
hmmmm |
Exio, that's because you probably changed the mapgen alias lines in mapgen.lua; those aliases get overwritten by the ones on the bottom of init.lua |
| 02:23 |
Exio |
ah |
| 03:59 |
|
dannydark joined #minetest-dev |
| 04:00 |
dannydark |
hmmmm you about? |
| 04:01 |
dannydark |
I've been playing around with your new mapgen work with the v6 mapgen enabled, its nice and fast only had one issue that keeps happening and though I would report it |
| 04:01 |
dannydark |
after a few minutes of generating new parts of the map I get: http://paste.ubuntu.com/1461072/ |
| 04:02 |
dannydark |
I have merged your code with 0.4.4 so It could be something to do with that |
| 04:03 |
hmmmm |
indeed, looks unrelated to my code |
| 04:03 |
hmmmm |
'loadBlock' |
| 04:04 |
dannydark |
ok, I'll have another look at the server.cpp and manually merge your changes into a fresh copy of the file |
| 04:04 |
hmmmm |
i made a lot of changes everywhere, and this is based on an old version, so i do not expect it to be a straightforward task of merging the two |
| 04:05 |
hmmmm |
i think in total i modified ~2k LOC |
| 04:05 |
dannydark |
yeah when I did a merge I think only a couple of conflicts appeared, just not sure if server.cpp was one of those files ^_^ |
| 04:06 |
dannydark |
infact I should have a merge log somewhere on my linux system |
| 04:06 |
dannydark |
other than this (which might not be related to your code) I have to say the map generation is very fast (using mapgen v6) |
| 04:06 |
dannydark |
I haven't had chance to test v7 yet ^_^ |
| 04:07 |
hmmmm |
don't bother, it's shit |
| 04:07 |
dannydark |
haha ok |
| 04:07 |
dannydark |
btw is v6 using any of your biome data? |
| 04:07 |
hmmmm |
no |
| 04:07 |
hmmmm |
v6 just doesn't support biomes |
| 04:07 |
dannydark |
because the beaches seem well bigger and better than they where previously :S |
| 04:07 |
hmmmm |
unless you really want to call the desert thing a biome |
| 04:08 |
hmmmm |
yeah, those aren't beaches.. you're running minimal, right? |
| 04:08 |
dannydark |
no lol... |
| 04:08 |
hmmmm |
minimal doesn't have desert sand so it falls back to regular sand |
| 04:08 |
hmmmm |
oh |
| 04:08 |
hmmmm |
i didn't change any functional parts of the mapgen, just made it faster |
| 04:08 |
dannydark |
I added your biome configs to the mapgen.lua in the minetest_game |
| 04:09 |
dannydark |
maybe its just this seed then haha, well I will take note of it for use later |
| 04:09 |
dannydark |
just seemed like the beaches where better, oh also forests....like proper ones similar to how jungles used to look but with normal trees |
| 04:10 |
hmmmm |
sometime in between 0.4 and 0.4.3, all the neat features like dungeons and jungles were removed |
| 04:10 |
dannydark |
its the first time I've seen these sort of things in 0.4.4 :S so just assumed it was your biome additions or something |
| 04:11 |
hmmmm |
like, i'll be honest, even though dungeons spawned mid air occasionally, they were still pretty interesting |
| 04:11 |
dannydark |
yeah I thought so too |
| 04:11 |
hmmmm |
I am going to try to bring them back but in a less buggy manner |
| 04:11 |
dannydark |
I'd rather have glitchy dungeons than non at all haha...infact why where jungles removed? :S never understood that |
| 04:12 |
hmmmm |
i'm not really sure |
| 04:12 |
dannydark |
but deserts added :S |
| 04:13 |
dannydark |
oh wait...is the mapgen v6 that you have in your fork from 0.4.3? |
| 04:13 |
hmmmm |
yes |
| 04:13 |
dannydark |
ahhh that explains it then |
| 04:13 |
hmmmm |
was anything changed since 0.4.3? |
| 04:13 |
hmmmm |
i told them not to change anything |
| 04:13 |
dannydark |
well maybe not, but like I say the beaches are much nicer now and there is forests being generated |
| 04:14 |
dannydark |
infact I will grab that seed and test in my other copy of minetest that is based on upstream |
| 04:14 |
hmmmm |
i'll have to check it out |
| 04:14 |
hmmmm |
er, yea you do that |
| 04:14 |
hmmmm |
my plans for bringing the v6 mapgen up-to-date include adding back all the old features that were removed and adding switches for them |
| 04:15 |
dannydark |
yeah I saw the smoothing setting you added, Thank you! ^_^ |
| 04:15 |
hmmmm |
so you'd be able to have no caves, dungeons, jungles, or something |
| 04:15 |
hmmmm |
or any combination |
| 04:15 |
hmmmm |
trees in general |
| 04:16 |
dannydark |
yeah sounds good, I saw you mention something about the v5 mapgen a few days back too |
| 04:16 |
hmmmm |
RBA is working on a 'treegen' system that procedureally generates them from an L-system, which is pretty neat.. I want to have each biome to have one or more "DecorationDef" structures associated with them |
| 04:17 |
hmmmm |
each DecorationDef would let you choose what type of decoration it is; you can have an RBA tree, for example, a jungle tree, a regular tree, or a stone, or custom decorations that you generate in Lua |
| 04:17 |
hmmmm |
and of course you'd choose the frequency distribution of each |
| 04:17 |
dannydark |
yeah that would be nice, I've seen of the stuff RBA has been working on looks impressive. I like how it completely changes the feel of the game |
| 04:18 |
hmmmm |
it's great stuff |
| 04:18 |
hmmmm |
we just partially fixed a problem where caves were carving out pieces of his trees |
| 04:18 |
hmmmm |
a side effect of this fix is that generation of above-ground land is faster |
| 04:19 |
hmmmm |
(that is, it doesn't even bother calculating where to place caves in mid-air) |
| 04:19 |
dannydark |
oh nice, yeah I've just started going through the logs and have seen some of the mese images (cave-air replaced with mese) |
| 04:21 |
hmmmm |
that was so we can actually see the problem |
| 04:22 |
dannydark |
so is it worth me applying the patch you linked? |
| 04:22 |
hmmmm |
yes |
| 04:22 |
hmmmm |
it wasn't a patch though, it was just a bit of code and surrounding context |
| 04:22 |
dannydark |
goodgood I'll get that done shortly then |
| 04:23 |
hmmmm |
i couldn't do a diff because my source file i did it in was almost completely different from the ones they're working with |
| 04:24 |
dannydark |
yeah thats fine I'll merge it with my local copy of your fork |
| 04:30 |
dannydark |
ok I'm convinced there both different will grab some screenshots and upload one sec |
| 04:32 |
dannydark |
Both maps have the same seed and I'm near enough in the same position, upstream 0.4.4: http://ompldr.org/vZ3Q1bA and your fork: http://ompldr.org/vZ3Q1bQ |
| 04:32 |
dannydark |
hmmmm^ |
| 04:32 |
hmmmm |
oh no |
| 04:32 |
hmmmm |
that means i messed something up |
| 04:33 |
dannydark |
the second image seems to have jungles/forests |
| 04:33 |
dannydark |
dense forests* |
| 04:33 |
hmmmm |
those aren't forests, that's my mapgen being stupid |
| 04:33 |
dannydark |
ah :( but I liked them lol |
| 04:34 |
hmmmm |
those are almost completely different |
| 04:34 |
dannydark |
reminded me of when we had jungles with hundreds of fireflies flying around |
| 04:37 |
dannydark |
they look like this: http://ompldr.org/vZ3Q1bg haha minus the weird white texture issue with the leaves :S hmm |
| 04:38 |
hmmmm |
can you do me a favor? |
| 04:38 |
dannydark |
yeah |
| 04:38 |
hmmmm |
check on a 0.4.3 version with that same seed and same position |
| 04:39 |
dannydark |
ok give me a sec I'll grab it |
| 04:39 |
hmmmm |
thanks |
| 04:40 |
hmmmm |
welp i found a bug |
| 04:40 |
hmmmm |
i never calculate the noise for trees |
| 04:40 |
hmmmm |
but that doesn't explain the massive difference in the terrain |
| 04:40 |
dannydark |
ah lol |
| 04:42 |
hmmmm |
thanks btw |
| 04:46 |
dannydark |
no worries ^_^ heres a screenshot from 0.4.3: http://ompldr.org/vZ3Q1cg |
| 04:47 |
hmmmm |
heh |
| 04:47 |
hmmmm |
do you see the difference between 0.4.3 and 0.4.3? |
| 04:48 |
hmmmm |
i wonder what caused that |
| 04:48 |
hmmmm |
my stuff is completely screwed up though |
| 04:49 |
hmmmm |
the base terrain looks close enough, the difference is that my terrain is a lot less hilly |
| 04:49 |
dannydark |
yeah I noticed that, just checking again now with a new world and no mods to make sure no mods are causing any differences |
| 04:49 |
hmmmm |
so there's something busted with the high/low selector, the steepness factor, or the higher terrain noise |
| 04:49 |
hmmmm |
gotta look into it |
| 04:49 |
hmmmm |
thanks a bunch btw |
| 04:51 |
* VanessaE |
peeks in, shouts something random at hmmmm and ducks back out ;) |
| 04:52 |
dannydark |
hmm just tried again 0.4.4 without the mods and had the same carved out hill as in the first screenshot :S and no worries ^_^ I didn't see anyone else respond in the logs saying they had tested |
| 04:52 |
dannydark |
so though I would have a try myself ^_^ hehe |
| 05:23 |
dannydark |
hmmmm: well I compared the differences between your server.cpp version against the upstream and the only things missing from yours are some cURL and protocal stuff |
| 05:23 |
dannydark |
it seems I'm getting the segfault after mapgen->makeChunk(&data); has run |
| 05:23 |
dannydark |
although it only happens after flying around the map for a bit |
| 05:23 |
hmmmm |
can you paste your version of server.cpp? |
| 05:24 |
dannydark |
yeah one sec |
| 05:25 |
dannydark |
http://paste.ubuntu.com/1461162/ |
| 05:25 |
dannydark |
thats the upstream version with your changes |
| 05:29 |
hmmmm |
hrmm |
| 05:30 |
hmmmm |
oh my, did you remember to add the modifications of map.cpp? |
| 05:31 |
dannydark |
yeah |
| 05:31 |
dannydark |
or well....at least I'm sure I got them all I will double check now |
| 05:32 |
dannydark |
also unrelated MSVS2012 spits out "\src\biome.cpp(99): warning C4715: 'BiomeDefManager::createBiome' : not all control paths return a value" a few times when compiling |
| 05:32 |
hmmmm |
oh yeah |
| 05:33 |
hmmmm |
dumb warning |
| 05:33 |
dannydark |
yeah haha |
| 05:33 |
hmmmm |
i'm doing a switch on an enum and handling all cases |
| 05:34 |
hmmmm |
logically it is not possible, save for corruption, for a control path to not return a value |
| 05:34 |
hmmmm |
i'll fix it though |
| 05:35 |
hmmmm |
yeah, sorry, i am not seeing an assertion failing on line 400 |
| 05:35 |
hmmmm |
the line the diagnostic message points to is bogus |
| 05:36 |
hmmmm |
could you run it under a debugger perhaps? |
| 05:37 |
dannydark |
yeah my version of server.cpp that the debug output was from had line breaks missing :S so I'm just trying to get it to throw it again now with the server.cpp with line breaks so I can get the line |
| 05:44 |
dannydark |
http://paste.ubuntu.com/1461181/ ok thats the debug.txt output with the same server.cpp as I pasted above, I will see if I can setup a debug build now |
| 05:46 |
dannydark |
ah smeg just got to nip out and defrost missus car for her then will be back in about 10 mins or so. |
| 06:12 |
|
SpeedProg joined #minetest-dev |
| 06:25 |
dannydark |
hmmmm: back sorry, ok debugging minetest via MSVS at the moment and its thrown "Access violation reading location 0x0000007F46C47C40." on line 380 of mapgen_v6.cpp |
| 07:02 |
dannydark |
hmmmm: just another update with regards to the issue i'm having, I've just downloaded your fork and compiled it as is without any changes and the error also gets thrown on that, so its not related to my other version being 0.4.4 |
| 07:03 |
dannydark |
It can sometimes happen within seconds of entering the map, but I seem to always be able to trigger it by flying around the map |
| 07:05 |
dannydark |
something else I just noticed with this build (not sure if its related) I just restarted the game after it crashed and the map didn't load at first for a few seconds then when it loaded only a single chunk/block loaded |
| 07:06 |
dannydark |
and nothing else for at least 5/7 minutes then it just started loading the rest of the map |
| 07:06 |
dannydark |
actually this could be something else I think I read something in the logs about a similar issue a few days ago |
| 07:29 |
hmmmm |
hrmm |
| 07:31 |
hmmmm |
okay that's not good |
| 07:31 |
hmmmm |
can you tell me what p2d, node_min, and ystride are at the time of the crash? |
| 07:39 |
|
yunfan joined #minetest-dev |
| 07:50 |
dannydark |
Yeah sorry, p2d {X=-192 Y=-32 } irr::core::vector2d<short> | node_min {X=-192 Y=-32 Z=-1312 } irr::core::vector3d<short> | ystride 80 int |
| 08:00 |
hmmmm |
oh my god |
| 08:00 |
hmmmm |
celeron you. |
| 08:00 |
hmmmm |
grrrrrrrrrr |
| 08:01 |
hmmmm |
another mapgen bug we never found |
| 08:02 |
hmmmm |
i probably didn't notice it because I never went that far in the Z direction |
| 08:02 |
hmmmm |
dannydark, on line 379 change p2d.Y - node_min.Z to p2d.Y - node_min.Y |
| 08:02 |
dannydark |
ok will do |
| 08:03 |
hmmmm |
you know what, I don't really think people are going to notice or care about their maps being 100% coherent with previous versions |
| 08:03 |
hmmmm |
for the sake of sanity, i am going to fix all the places where he has X,Y instead of X,Z |
| 08:05 |
dannydark |
yeah thing is eventually it will be replaced with v7 anyway so as long as v6 works I think that will be enough, plus with it being much faster anyway I think people will welcome it |
| 08:05 |
dannydark |
I've just compiled and testing now |
| 08:11 |
dannydark |
hmmmm: not sure that fixed it, still crashed out at the same line |
| 08:12 |
hmmmm |
of course |
| 08:13 |
hmmmm |
because in add mud he uses x,z instead of being consistently wrong with x,y |
| 08:13 |
hmmmm |
hrmmm |
| 08:13 |
hmmmm |
okay, change line 379 back to what it was, and then on line 654, change v2s16(node_min.X, node_min.Y) to v2s16(node_min.X, node_min.Z) |
| 08:14 |
dannydark |
ok will do |
| 08:14 |
hmmmm |
that will change it for certain |
| 08:14 |
hmmmm |
s/change/fix/ |
| 08:15 |
hmmmm |
sorry if i'm not paying attention as well as i should |
| 08:16 |
dannydark |
haha its fine, I'm playing BF3 in between compiling etc ^_^ |
| 08:16 |
dannydark |
ok compiled will test now |
| 08:20 |
dannydark |
:( nope crashed out again, same line |
| 08:20 |
hmmmm |
WTF |
| 08:20 |
hmmmm |
did you remember to change 379 back to what it was? |
| 08:20 |
thexyz |
how about naming the fixed one mapgen v6.5 or something |
| 08:20 |
dannydark |
yeah |
| 08:20 |
dannydark |
should I pastebin the file? |
| 08:22 |
hmmmm |
get_biome is used in 3 places, 1 in ground level generation, that's x,z, then one for the cave count, that's x,y but fixed to x,z, then there's the add mud thing, that's x,z, then there's adding dirt and gravel to the underground, which is x,y aha. |
| 08:22 |
hmmmm |
i was wrong, it's used in 4 places |
| 08:22 |
hmmmm |
it's the last one that has to be the problem |
| 08:22 |
hmmmm |
okay, line 987, change v2s16(node_min.X, node_min.Y) to v2s16(node_min.X, node_min.Z) |
| 08:23 |
hmmmm |
third time's a charm :) |
| 08:23 |
dannydark |
haha compiling now |
| 08:28 |
dannydark |
yep reckon thats sorted it ^_^ |
| 08:28 |
hmmmm |
great |
| 08:29 |
hmmmm |
thanks again |
| 08:29 |
hmmmm |
i've gotta go |
| 08:29 |
dannydark |
no worries |
| 08:29 |
dannydark |
byes |
| 09:02 |
RealBadAngel |
hi |
| 09:30 |
|
dannydark left #minetest-dev |
| 10:13 |
|
jin_xi joined #minetest-dev |
| 11:17 |
|
doserj joined #minetest-dev |
| 11:27 |
RealBadAngel |
shit, i trimmed those caves at least :) |
| 11:27 |
RealBadAngel |
they now neither wont generate above ground level, nor destroy ground level |
| 11:28 |
RealBadAngel |
but can happen that entrances to caves will be generated |
| 11:37 |
|
doserj joined #minetest-dev |
| 12:00 |
|
doserj joined #minetest-dev |
| 12:08 |
|
serengeor joined #minetest-dev |
| 12:40 |
|
doserj joined #minetest-dev |
| 13:00 |
|
doserj joined #minetest-dev |
| 13:28 |
|
celeron56 joined #minetest-dev |
| 13:48 |
|
rubenwardy joined #minetest-dev |
| 14:06 |
RealBadAngel |
well, now caves generates much nicer |
| 14:06 |
|
doserj joined #minetest-dev |
| 14:08 |
RealBadAngel |
and its way faster imho |
| 14:12 |
|
SpeedProg joined #minetest-dev |
| 14:25 |
|
rubenwardy joined #minetest-dev |
| 14:35 |
RealBadAngel |
server with tweaked cave generation and L-system trees: 77.91.18.205:30000 |
| 14:35 |
RealBadAngel |
im pretty pleased with the results :) |
| 15:07 |
darkrose |
bit odd in places, and sometimes the trunk doesn't touch the ground all the way around, pretty nice though |
| 15:25 |
celeron55 |
hmm |
| 15:25 |
celeron55 |
http://minetest.net/wiki/doku.php?id=roadmap:roadmap |
| 15:25 |
celeron55 |
this page is still surprisingly relevant |
| 15:29 |
celeron55 |
maybe some more longer-term stuff should be planned and put there |
| 15:29 |
celeron55 |
...dunno |
| 15:29 |
RealBadAngel |
celeron55, ive tweaked a bit cave generation, its much smoother and without annoyin artifacs |
| 15:30 |
celeron55 |
how is it smoother and what artifacts |
| 15:30 |
celeron55 |
those are the most vague terms you could have possibly used |
| 15:32 |
RealBadAngel |
ok, ive cut down number of generated caves by half |
| 15:33 |
RealBadAngel |
not allowing generation of them above the ground |
| 15:33 |
RealBadAngel |
allow adding of dirt/gravel only in caves below water level |
| 15:33 |
RealBadAngel |
and some other minor changes |
| 15:34 |
RealBadAngel |
effect you can check here: 77.91.18.205:30000 |
| 15:35 |
RealBadAngel |
i really like it |
| 15:39 |
RealBadAngel |
btw new trees are also here |
| 15:43 |
RealBadAngel |
btw those words were for how it feels now |
| 17:11 |
|
hmmmm joined #minetest-dev |
| 17:25 |
|
rubenwardy joined #minetest-dev |
| 17:37 |
|
nyuszika7h joined #minetest-dev |
| 18:08 |
|
rubenwardy1 joined #minetest-dev |
| 18:11 |
|
SpeedProg1 joined #minetest-dev |
| 18:24 |
|
VanessaE joined #minetest-dev |
| 19:27 |
VanessaE |
https://github.com/celeron55/minetest_game/pull/78 |
| 19:32 |
VanessaE |
basically to fix irritating stuff like this: |
| 19:32 |
VanessaE |
http://digitalaudioconcepts.com/vanessa/hobbies/minetest/screenshots/screenshot_3462887985.png |
| 19:34 |
darkrose |
follwo |
| 19:34 |
VanessaE |
also, that little one-line fix to make the server stop choking on the emerge thread works mostly - I still had to restart my server today because the world just refused to load for me again, in two different regions. |
| 19:34 |
VanessaE |
yeah I know, I didn't catch that typo until after I pushed. |
| 19:35 |
VanessaE |
(but the server no longer appears to lock up anymore) |
| 19:37 |
VanessaE |
I suppose brick should also follow facedir but I figured wood was enough for now. |
| 19:38 |
darkrose |
I get that problem with cobble too... but at that rate everything will need facdir, so meh |
| 19:38 |
VanessaE |
I thought about that |
| 19:39 |
VanessaE |
I don't see any immediate problem with doing that though |
| 19:39 |
VanessaE |
just on the base materials that really need ti anyway |
| 19:43 |
VanessaE |
*looks at default materials again* |
| 19:43 |
thexyz |
RealBadAngel: where's that server located? my ping to it is about 1.5sec |
| 19:43 |
VanessaE |
yeah, just wood, bricks, cobble, and *maybe* stone need facedir since they all have a "grain" to them |
| 19:45 |
thexyz |
RealBadAngel: i can host it (but not sure if that's needed) |
| 19:46 |
VanessaE |
if he runs it locally, it's on Poland. |
| 19:46 |
VanessaE |
in* |
| 19:55 |
thexyz |
https://github.com/celeron55/minetest/pull/365 nice one, though i don't understand why does he #ifdef USE_IPV6 in so many places |
| 19:56 |
VanessaE |
beats me, IPv6 (in general) always seemed like overkill to me. |
| 19:56 |
darkrose |
640k is all anyone will ever need! |
| 19:57 |
VanessaE |
heh |
| 19:57 |
thexyz |
yes, yes, we'll never run out of ipv4 |
| 19:57 |
thexyz |
wait, we already did |
| 19:57 |
VanessaE |
well, truth be told? add two more octets to IPV4 and that would have been enough until the death of the sun, but noooooo they go for something that resembles a MAC address on serious steroids |
| 19:58 |
thexyz |
i assume that guys who created ipv4 thought the same |
| 19:59 |
darkrose |
relevant: https://xkcd.com/865/ |
| 19:59 |
VanessaE |
possibly, but ipv4 is already 32 bit = a bit under 4 billion addresses if you filter out LAN IPs and loopback and such. 48 bits would have been 65k as many addresses. |
| 19:59 |
VanessaE |
do we really need every microwave, TV, toaster, and blender to have an IP address? :-) |
| 20:00 |
VanessaE |
darkrose: haha |
| 20:00 |
darkrose |
my loungeroom clock has a public IP address |
| 20:01 |
VanessaE |
oh excuse me, I forgot to mention clocks and radios too. :-) |
| 20:01 |
thexyz |
VanessaE: why not? |
| 20:01 |
VanessaE |
thexyz: lack of necessity? |
| 20:03 |
VanessaE |
a little NAT magic and each home only needs 1 IP address. there are 7 billion people in the world but, what, half as many actual homes? a quarter as many? |
| 20:03 |
thexyz |
that's what "why not" was referred to |
| 20:03 |
VanessaE |
(and I mean NAT run inside the home e.g. from a PC or router or so) |
| 20:03 |
thexyz |
we don't need NAT anymore with ipv6 |
| 20:03 |
VanessaE |
yes I know that |
| 20:04 |
VanessaE |
but IPV6 is a solution looking for an unsolved problem |
| 20:04 |
thexyz |
that's its advantage too |
| 20:04 |
thexyz |
(this conversation is too unrelated for this channel anyway) |
| 20:04 |
VanessaE |
yeah |
| 20:17 |
celeron55 |
getting completely rid of NAT is awesome |
| 20:19 |
VanessaE |
admittedly it would be nice not to need it, but for now it serves a useful purpose. |
| 20:19 |
celeron55 |
1 ip per home is ridiculously little; you're supposed to be able to host anything from anywhere to the internet |
| 20:19 |
celeron55 |
NAT is not useful because of anything else but the unintended limtiations of ipv4 |
| 20:20 |
celeron55 |
being able to host anythin to the internet from anywhere without any configuration in between is the whole idea of IP |
| 20:20 |
celeron55 |
+g |
| 20:20 |
VanessaE |
well sure, but like I said, does your toaster, microwave, fridge, etc. really need to be connected to the 'net? :-) |
| 20:20 |
celeron55 |
ipv6 makes that happen once again after a long time of silly hacks |
| 20:20 |
celeron55 |
VanessaE: i have multiple computers |
| 20:21 |
VanessaE |
as do I |
| 20:21 |
VanessaE |
5 devices that are internet-capable here. |
| 20:21 |
VanessaE |
(6 if you could my spare box) |
| 20:21 |
VanessaE |
count* |
| 20:22 |
VanessaE |
I don't have a big problem with needing to move beyond IPv4, it's the way its implemented that seems..stupid to me |
| 20:22 |
VanessaE |
I mean, IPv6 is what, eight 16-bit figures? |
| 20:22 |
celeron55 |
i'm not going to argue more |
| 20:23 |
celeron55 |
i'll just say i am glad you are not the one designing internet standards 8) |
| 20:23 |
* VanessaE |
shrugs. this is off-topic for here anyway |
| 20:24 |
celeron55 |
http://article.gmane.org/gmane.linux.kernel/1414106 |
| 20:24 |
celeron55 |
8) |
| 20:24 |
VanessaE |
as for on-topic, Pilzadam not being here, what's your reaction to https://github.com/celeron55/minetest_game/pull/78 ? |
| 20:25 |
VanessaE |
hah! |
| 20:28 |
celeron55 |
dunno, i guess there needs to be some rule of what is allowed to have facedir in upstream if that is made |
| 20:28 |
celeron55 |
or alternatively everything should have |
| 20:28 |
VanessaE |
my guess is that if it has a visible, obvious "grain", it should have facedir, but that's me,. |
| 20:29 |
VanessaE |
I mean, no point in doing that with sand, water, etc. |
| 20:29 |
thexyz |
aww, such cute mail from Linus, again |
| 20:31 |
celeron55 |
does cobblestone have "obvious grain"? |
| 20:31 |
celeron55 |
it really depends on the textures |
| 20:31 |
Exio |
hahaha |
| 20:31 |
Exio |
nice link celeron55 |
| 20:31 |
celeron55 |
so this is kind of a fuzzy thing |
| 20:31 |
VanessaE |
celeron55: well kinda yeah - in that it looks like stone bricks |
| 20:31 |
VanessaE |
so they sort of "go" in a given direction |
| 20:32 |
celeron55 |
but they don't really go so much that somebody would care, compared to wood |
| 20:32 |
VanessaE |
no, probably not |
| 20:32 |
VanessaE |
that one is questionable for sure |
| 20:32 |
VanessaE |
I've been wondering why you went with the brick-like look instead of some sort of round-ish stones? |
| 20:33 |
celeron55 |
because of random |
| 20:33 |
VanessaE |
haha |
| 20:33 |
thexyz |
oh, btw, VanessaE, it seems bug your reported (with MT crashing after lost focus) just happened for me |
| 20:33 |
celeron55 |
you're free to propose a more cobblestone-ish texture |
| 20:34 |
thexyz |
https://gist.github.com/de40695061d3a4ca0f7e |
| 20:34 |
VanessaE |
oh I might at some point - it's not that important, merely academic curiosity. |
| 20:35 |
VanessaE |
thexyz: ah, at least it ain't just my setup doing it. How far did you have to go into my suggested setup to reproduce it? |
| 20:35 |
thexyz |
i just tried to connect to RBA server |
| 20:35 |
thexyz |
to see his trees |
| 20:36 |
VanessaE |
ah |
| 20:36 |
VanessaE |
they're nice looking, if a bit plain. |
| 20:39 |
celeron55 |
has anyone thought anything of the trees gameplay-wise? |
| 20:40 |
VanessaE |
in what way? |
| 20:40 |
celeron55 |
like, in survival, do you spend the first half an hour whacking down one huge tree? |
| 20:40 |
celeron55 |
the small trees are handy, so to say |
| 20:41 |
VanessaE |
his code also has amller trees |
| 20:41 |
VanessaE |
smaller* |
| 20:41 |
VanessaE |
about the same size as the defaults |
| 20:42 |
VanessaE |
but yeah, you make a good point |
| 20:42 |
VanessaE |
how hard would it be for the game to actually fell a tree when the trunk is cut? |
| 20:43 |
VanessaE |
I don't mean the timber mod, per se, but something that makes a tree fall down horizontally. |
| 20:45 |
celeron55 |
depends on the fanciness |
| 20:45 |
VanessaE |
right |
| 20:47 |
celeron55 |
if you just remove the tree, rotate it 90 degrees and place it down to where it falls, with something to somehow bend/remove the leaves and branches that press against the ground (so that the trunk ends up almost fully on the ground), that would work gameplay-wise |
| 20:48 |
celeron55 |
the nontrivialities are: 1) where to implement it, 2) how to detect when a tree is supposed to fall |
| 20:48 |
VanessaE |
yes, I like that idea. |
| 20:49 |
VanessaE |
where? in C++ for certain. How to detect? you'd have to trace from dirt or stone upward until you find nothing but air all around your trace path |
| 20:49 |
VanessaE |
if a trunk is more than half way cut, it wouldn't be out of the question to make it fall |
| 20:49 |
RealBadAngel |
hi, sorry, was afk |
| 20:50 |
celeron55 |
that is not really implementable purely in C++ |
| 20:50 |
VanessaE |
can I change the subject for a minute so I can throw an idea out regarding mese? |
| 20:51 |
RealBadAngel |
thexyz, you can try to join now |
| 20:51 |
VanessaE |
Read this, gimme your thoughts by /msg so as not to clutter the channel: http://pastebin.com/ZPdnCvKg |
| 20:51 |
celeron55 |
but some C++ could be implemented that can rotate and do that stuff to a tree |
| 20:52 |
celeron55 |
the thing that looks if a tree trunk was cut should be in lua, and it would then call that thing |
| 20:52 |
VanessaE |
mmmmm |
| 20:52 |
VanessaE |
well I guess but only if it can be done incredibly fast |
| 20:53 |
VanessaE |
the question then becomes, how does one avoid mistaking a vertical stack of tree trunks (e.g. in a house) for a tree? |
| 20:53 |
VanessaE |
I guess look for air surrounding all four sides? |
| 20:53 |
celeron55 |
they're connected to something from elsewhere than ground |
| 20:54 |
VanessaE |
(air or leaves..) |
| 20:55 |
celeron55 |
i guess the detection of whether a tree trunk that was cut down belongs to a tree and if that tree was completely cut should be done in C++ too, then |
| 20:55 |
celeron55 |
it probably should be as generic as is feasible |
| 20:55 |
VanessaE |
as long as a modder could hook into it, I see no real problem with making it as much C++ as possible, aside from who has to code it :D |
| 20:57 |
RealBadAngel |
trunk is generic, put abm of trunk nodes inside |
| 20:57 |
celeron55 |
well, lua would then call something like minetest.is_tree(p); minetest.is_cut_tree({x=p.x, y=p.y+1, z=p.z}); minetest.fall_tree({x=p.x, y=p.y+1, z=p.z}) |
| 20:57 |
RealBadAngel |
if they lost connection trigger fall |
| 20:58 |
VanessaE |
fell_tree() |
| 20:58 |
VanessaE |
(to cut a tree down is to fell it. obscure English at its best.) |
| 20:58 |
VanessaE |
Ilike it. |
| 20:59 |
celeron55 |
the C++ side of the engine can check stuff based on the tree and leaves groups |
| 20:59 |
VanessaE |
yes |
| 20:59 |
RealBadAngel |
lsystem trunk is a cross, only one in middle should have abm on it |
| 20:59 |
RealBadAngel |
and can check others around |
| 21:00 |
RealBadAngel |
and count them for example |
| 21:00 |
VanessaE |
RealBadAngel: the only way to do that is to make the center a different node from the surroundings |
| 21:00 |
RealBadAngel |
yes |
| 21:00 |
VanessaE |
(trivial to do, of course) |
| 21:00 |
RealBadAngel |
same lookin |
| 21:00 |
VanessaE |
that *would* make it easier to isolate what is or is not a trunk |
| 21:00 |
VanessaE |
especially if the "different" node is not available by crafting or creative inventory. |
| 21:01 |
RealBadAngel |
typical lsystem tree has trunk 5 height, so it makes 20 trunk nodes, 5 with abms |
| 21:01 |
celeron55 |
ABMs don't make any sense for that |
| 21:01 |
RealBadAngel |
lets make a rule 3/5 are cut, or 10/20 total for tree to fell |
| 21:02 |
celeron55 |
having events trigger from breaking a tree block is much more efficient |
| 21:02 |
RealBadAngel |
yeah? |
| 21:02 |
RealBadAngel |
how would you differ whats a tree and what is a buildin made of tree? |
| 21:02 |
celeron55 |
... |
| 21:04 |
RealBadAngel |
could be funny anyway seein tree goin out of a shack ;) |
| 21:15 |
VanessaE |
celeron55: regarding the link I pointed you to with the convo between Jordach and I, please examine this pull request if you are so inclined (yes, I know it's PilzAdam's purview, but he ain't here at the moment) |
| 21:15 |
VanessaE |
https://github.com/celeron55/minetest_game/pull/79 |
| 21:15 |
RealBadAngel |
about tree rotation its not a big problem |
| 21:16 |
VanessaE |
rotating 90 degrees is trivial enough |
| 21:16 |
VanessaE |
rotate from the head of the tree, then make all nodes fall like entities. |
| 21:17 |
RealBadAngel |
have you any idea how much enities gonna fall from a really big tree? |
| 21:17 |
VanessaE |
a lot. |
| 21:17 |
VanessaE |
hundreds, I would assume. |
| 21:17 |
RealBadAngel |
PilzAdam was cutting one pine tree for several minutes |
| 21:17 |
VanessaE |
yeah but he was trying to harvest the entire thing |
| 21:18 |
RealBadAngel |
and got over 3 stacks of trunks |
| 21:18 |
VanessaE |
not just fell it |
| 21:18 |
RealBadAngel |
i guess that leaves would be like 2 times more |
| 21:19 |
VanessaE |
yeha |
| 21:19 |
VanessaE |
yeah* |
| 21:20 |
RealBadAngel |
guess what will happen then |
| 21:22 |
VanessaE |
kaboom? :-) |
| 21:23 |
RealBadAngel |
lotsa messages |
| 21:23 |
RealBadAngel |
and entities deleted |
| 21:24 |
VanessaE |
only if those entities exceed 49 per node space I thought? |
| 21:24 |
RealBadAngel |
we are talkin bout circa 1k entities |
| 21:26 |
VanessaE |
naw |
| 21:26 |
VanessaE |
not *that* many |
| 21:26 |
|
doserj joined #minetest-dev |
| 21:26 |
VanessaE |
a few hundred at most I'd have thought |
| 21:27 |
VanessaE |
especially if you "empty out" the trees as we discussed before |
| 21:27 |
VanessaE |
(branches ending in leaves, rather than *buried* in leaves) |
| 21:27 |
RealBadAngel |
we can chop one big oak if you want |
| 21:27 |
celeron55 |
for comparison, a mapblock is 16x16x16; that is 4096 nodes |
| 21:28 |
VanessaE |
hrm. |
| 21:28 |
VanessaE |
I guess it does add up kinda fast doesn't it. |
| 21:28 |
VanessaE |
how many entities can the engine realistically move all at once? |
| 21:28 |
VanessaE |
assuming a slow computer of course |
| 21:29 |
celeron55 |
dunno |
| 21:35 |
RealBadAngel |
i collected already 8 stacks of leaves, and im far from end |
| 21:39 |
RealBadAngel |
forget entities |
| 21:39 |
RealBadAngel |
got already 1,3k of leaves and its like 25% of the whole tree, maybe less |
| 21:39 |
VanessaE |
wow |
| 21:39 |
VanessaE |
way too many leaves and too much wood I think |
| 21:40 |
VanessaE |
this is gonna make the leafdecay mod barf. |
| 21:40 |
VanessaE |
:() |
| 21:40 |
VanessaE |
:) |
| 21:40 |
RealBadAngel |
hehe |
| 21:42 |
RealBadAngel |
i guess leafdecay is not good idea for such kind of trees |
| 22:42 |
RealBadAngel |
ok, if anybody still thinkin bout cuttin the trees and make them fell as entities: cutting one big oak gave me 5236 leaves and 253 tree nodes |
| 22:43 |
VanessaE |
what about quickly rendering the tree into a model and making *that* fall? :-) |
| 22:43 |
VanessaE |
ok maybe not, just joking anyway |
| 22:43 |
RealBadAngel |
ehehe |
| 22:43 |
VanessaE |
so much for falling as entities |
| 22:44 |
RealBadAngel |
took bout 20 minutes to cut it |
| 22:44 |
RealBadAngel |
and several mese picks |
| 22:45 |
RealBadAngel |
of course cuttin smaller oak would be way faster |
| 22:45 |
RealBadAngel |
ive chosen just the biggest one |
| 22:45 |
serengeor |
I wanna see that on video ;o |
| 22:46 |
RealBadAngel |
join the server and chop one yourself |
| 22:46 |
RealBadAngel |
77.91.18.205:30000 |
| 22:46 |
serengeor |
uhm, I guess I could do that, though tbh I haven't really played minetest :/ |
| 22:47 |
serengeor |
which client version do I need? |
| 22:47 |
RealBadAngel |
0.4.4 |
| 22:49 |
VanessaE |
0.4.4-release since RBA's server build is somewhat old :-) |
| 22:50 |
RealBadAngel |
a week and old |
| 22:50 |
RealBadAngel |
:) |
| 22:53 |
serengeor |
k, waiting for it to compile |
| 23:38 |
VanessaE |
is there a way to tell MT to ignore an alias for the sake of crafting something? |
| 23:39 |
|
jaseg joined #minetest-dev |
| 23:46 |
|
iqualfragile joined #minetest-dev |
| 23:48 |
|
Jeija joined #minetest-dev |
| 23:48 |
|
Jeija left #minetest-dev |
