| Time |
Nick |
Message |
| 00:00 |
nolsen |
#L377 is more specific |
| 00:00 |
RealBadAngel |
movement was choking, after restart it got a bit better |
| 00:00 |
nolsen |
and #L392 |
| 00:00 |
nolsen |
and so on |
| 00:02 |
RealBadAngel |
btw, what does "Node highlighting" option do in reanimated simple settings menu? |
| 00:02 |
Fixer |
highlights node |
| 00:03 |
RealBadAngel |
are you completely sure? |
| 00:03 |
Fixer |
hold on |
| 00:03 |
RealBadAngel |
iirc theres no such setting in game ;) |
| 00:04 |
Fixer |
enable_node_highlighting = true |
| 00:04 |
Fixer |
hmm |
| 00:04 |
RealBadAngel |
node_highlighting = box/halo .... |
| 00:04 |
Fixer |
RealBadAngel, it turns halo thing |
| 00:04 |
Fixer |
RealBadAngel, at least in stable 0.4.13 |
| 00:05 |
RealBadAngel |
so this patch was meant for 0.4.13 stable or what? |
| 00:05 |
Fixer |
RealBadAngel, it was push now = fix later iirc |
| 00:05 |
|
misprint joined #minetest-dev |
| 00:06 |
Fixer |
RealBadAngel, simple settings will need fixes |
| 00:06 |
RealBadAngel |
built again, only minimal, everything seems fine now |
| 00:07 |
|
OldCoder joined #minetest-dev |
| 00:08 |
Fixer |
https://github.com/minetest/minetest/pull/3740/files |
| 00:08 |
Fixer |
https://github.com/minetest/minetest/pull/3740/files#diff-b61e3e14dbe70423022f4ee065fa3f94R215 |
| 00:23 |
RealBadAngel |
awww, i forgot that paramat removed normal maps from game |
| 00:24 |
RealBadAngel |
no wonders, relief mapping on auto gen is damn slow |
| 00:26 |
sofar |
I've found that hand-painting normal maps in blender is likely the best way to do it |
| 00:26 |
|
OldCoder joined #minetest-dev |
| 00:26 |
sofar |
I'm wondering if there's an easy way to do that for lots of textures |
| 00:27 |
RealBadAngel |
im using gimp to paint shapes, filter, blur etc |
| 00:27 |
RealBadAngel |
and to process maps AwesomeBump |
| 00:28 |
RealBadAngel |
more on the autogen if somebody would know: it was meant to genereate a maps only in case regular map is not supplied |
| 00:29 |
RealBadAngel |
having it on to make each and every texture out there is plain dumb |
| 00:30 |
RealBadAngel |
not to mention that results of autogenerated one is always way behind good hand made one |
| 00:31 |
RealBadAngel |
sofar, there are ways to batch it, but results are not any good imho |
| 00:31 |
sofar |
I don't think that's a good solution |
| 00:32 |
sofar |
did you ever make a default_normalmaps texture pack with *just* the normalmaps that were in minetest_game? |
| 00:32 |
RealBadAngel |
auto gens, batching are just workarounds to get support fast and easy |
| 00:32 |
RealBadAngel |
but it wont work nor look any good |
| 00:32 |
RealBadAngel |
ofc i did |
| 00:33 |
RealBadAngel |
3 yrs ago? |
| 00:33 |
RealBadAngel |
but point is normals are not texture packs stuff |
| 00:34 |
RealBadAngel |
texturing in games requires having set of textures for one surface |
| 00:34 |
RealBadAngel |
up to several ones |
| 00:34 |
RealBadAngel |
normal, height, specular etc |
| 00:35 |
RealBadAngel |
you cant call one of them to be a single standalone TEXTURE |
| 00:35 |
RealBadAngel |
so treating normals the same way as diffuse ones is wrong |
| 00:37 |
RealBadAngel |
its like you wanted to separate alpha channel and put all the alpha channel files in texture pack folder ;) |
| 00:48 |
sofar |
no, but the normal maps are a function of the texture |
| 00:54 |
RealBadAngel |
ideal would be xcf with layers |
| 00:55 |
RealBadAngel |
layer 0 - diffuse, 1 normal, 2 specular and so on |
| 00:55 |
RealBadAngel |
single file for all the components |
| 00:57 |
RealBadAngel |
i wonder how hard would be to add xcf support |
| 01:00 |
sofar |
ImageMagick has an XCF reader module |
| 01:01 |
sofar |
you could use their library, even |
| 01:02 |
RealBadAngel |
that would be really handy |
| 01:03 |
RealBadAngel |
and would solve mess with components once and for all |
| 01:05 |
RealBadAngel |
will have to take care of it one day, good TODO point |
| 01:14 |
|
linkedinyou joined #minetest-dev |
| 01:30 |
|
DevBox joined #minetest-dev |
| 02:05 |
|
dulrich joined #minetest-dev |
| 02:25 |
RealBadAngel |
seems like i do have VBO working quite ok |
| 02:26 |
RealBadAngel |
since half an hour ive visited a few servers, started few singleplayer worlds, switching through them |
| 02:26 |
RealBadAngel |
without closing client ofc |
| 02:28 |
RealBadAngel |
ram usage normal, no crashes |
| 02:28 |
RealBadAngel |
fps almost all the time at 60 |
| 02:29 |
RealBadAngel |
at 240 and shitload of trees too :) |
| 02:30 |
RealBadAngel |
http://i.imgur.com/dE3nsJU.png |
| 02:34 |
kaadmy |
nice |
| 02:37 |
|
Dragonop joined #minetest-dev |
| 03:21 |
|
STHGOM joined #minetest-dev |
| 03:29 |
RealBadAngel |
#3763 |
| 03:29 |
ShadowBot |
https://github.com/minetest/minetest/issues/3763 -- Mapblock mesh: Allow to use VBO by RealBadAngel |
| 03:29 |
RealBadAngel |
who wanna test? :) |
| 03:34 |
sofar |
what am I looking for? |
| 03:34 |
sofar |
oh hmmm |
| 03:38 |
RealBadAngel |
adding screenshots atm |
| 03:38 |
RealBadAngel |
still havent closed the client |
| 03:38 |
RealBadAngel |
its propably 2 hrs already |
| 03:40 |
sofar |
what options do I need to enable to see a difference? |
| 03:40 |
sofar |
normal maps? |
| 03:40 |
sofar |
bumpmapping? |
| 03:41 |
RealBadAngel |
theres option to turn it on in ingame/basic |
| 03:41 |
RealBadAngel |
enable_vbo |
| 03:42 |
RealBadAngel |
client/graphics/ingame/basic |
| 03:43 |
sofar |
so I should see a difference without bumpmapping and parrallax disabled? |
| 03:44 |
RealBadAngel |
yeah |
| 03:44 |
RealBadAngel |
find a place you had low FPS and check the difference |
| 03:44 |
RealBadAngel |
or just increase view range |
| 03:45 |
RealBadAngel |
and compare FPS with vbo and without |
| 03:47 |
RealBadAngel |
VBO is about storing meshes in VRAM, what effects are you using is not that important |
| 03:47 |
sofar |
ohhh |
| 03:47 |
RealBadAngel |
well, bumpmapping and parallax have impact, tangent space vertices takes more memory |
| 03:49 |
RealBadAngel |
sofar, and? |
| 03:49 |
RealBadAngel |
any difference? |
| 03:57 |
sofar |
seems fps is stuck at 50-60 with 500+ viewing range for me with vbo enabled |
| 03:57 |
sofar |
I have to push it to get it down to 30 :) |
| 03:57 |
sofar |
but, I suggest running this against ridic. servers like just test |
| 03:59 |
RealBadAngel |
i was there for several time last 2 hours |
| 03:59 |
RealBadAngel |
see screenies in PR |
| 04:00 |
sofar |
significant improvement |
| 04:00 |
sofar |
like, 28fps -> 40+ |
| 04:01 |
RealBadAngel |
i suggest running top in another window and watch memory usage |
| 04:01 |
RealBadAngel |
for me its all the time ~2gig |
| 04:02 |
RealBadAngel |
with peaks up to 3 (lotsa blocks loaded) |
| 04:04 |
sofar |
2108virt 1261res |
| 04:05 |
VanessaE |
it's a small improvement for me. for my usual Dreambuilder testing world "benchmark" scene (RealBadAngel knows which one), drawtime was 34-36 before with a view range in the high 90's (that build predated the view range auto-tune removal). Now it's about 27 at 100m (git HEAD as of now, with RBA's PR) |
| 04:06 |
VanessaE |
(so 35 fps now. used to be less than 30 in this scene before) |
| 04:06 |
RealBadAngel |
still a gain |
| 04:07 |
VanessaE |
not anywhere near the gains I saw with Pilzadam's original VBO code though. |
| 04:07 |
VanessaE |
(and the menu indicates VBO is enabled) |
| 04:08 |
RealBadAngel |
its hardly different to PA's code |
| 04:08 |
RealBadAngel |
either you set it or not |
| 04:08 |
* VanessaE |
shrugs |
| 04:08 |
VanessaE |
just reporting my findings. |
| 04:08 |
RealBadAngel |
there are other factors that have impact on that |
| 04:09 |
RealBadAngel |
like tangent vertices/regular ones |
| 04:09 |
VanessaE |
not sure which of those modes I'm set for. |
| 04:09 |
RealBadAngel |
check shaders enabled/bump/parallax disabled |
| 04:09 |
VanessaE |
shaders are enabled, but no effects. |
| 04:09 |
RealBadAngel |
only bumpmapping and parallax can trigger tangent space on |
| 04:09 |
VanessaE |
default textures. |
| 04:10 |
RealBadAngel |
btw, youre still on open source drivers? |
| 04:10 |
VanessaE |
nop |
| 04:11 |
sofar |
do you need me to test on my Intel HD GFX system? |
| 04:12 |
RealBadAngel |
sofar, sure |
| 04:12 |
RealBadAngel |
test what you can |
| 04:12 |
RealBadAngel |
just be sure youre using OpenGL |
| 04:12 |
VanessaE |
using catalyst drivers, Debian 8. looks like they're v15.20.3 |
| 04:12 |
sofar |
of course that's using OpenGL, lol |
| 04:13 |
RealBadAngel |
i will have to put there condition, so folks wont try it with DX or software for example ;) |
| 04:15 |
RealBadAngel |
VanessaE, only difference to c55/PA patch is how i handle removing hardware buffers |
| 04:15 |
VanessaE |
I assumed something like that was in play. good deal. |
| 04:15 |
RealBadAngel |
(and when i demand reuploading to VRAM) |
| 04:16 |
VanessaE |
btw, turning off shaders GREATLY increases my FPS. |
| 04:16 |
VanessaE |
as in 52-53 (draw time of about 17-18) at 100m on the same scene. |
| 04:16 |
VanessaE |
52-53 FPS that is. |
| 04:16 |
RealBadAngel |
i know the reason propably |
| 04:17 |
RealBadAngel |
theres a code somwhere to fix that |
| 04:17 |
VanessaE |
sounds like you need to enable/apply it then |
| 04:18 |
RealBadAngel |
thats gregorycu patch about setting shader uniform variables |
| 04:18 |
RealBadAngel |
it has to be reworked and merged ASAP |
| 04:18 |
VanessaE |
#3534 ? |
| 04:18 |
ShadowBot |
https://github.com/minetest/minetest/issues/3534 -- Replace vector with vector by gregorycu |
| 04:19 |
RealBadAngel |
filter closed PR by him |
| 04:19 |
RealBadAngel |
something "superfluous" and "shaders" |
| 04:19 |
VanessaE |
only one result among closed, #2156 |
| 04:19 |
ShadowBot |
https://github.com/minetest/minetest/issues/2156 -- Optimise MapBlockMesh functions by gregorycu |
| 04:20 |
RealBadAngel |
https://github.com/minetest/minetest/pulls?utf8=%E2%9C%93&q=is%3Aclosed+is%3Apr+author%3Agregorycu |
| 04:21 |
VanessaE |
@2320 |
| 04:21 |
VanessaE |
#2320 |
| 04:21 |
ShadowBot |
https://github.com/minetest/minetest/issues/2320 -- Fix superflous shader setting updates by gregorycu |
| 04:21 |
RealBadAngel |
2309 and 2320. later on rubenwardy tried again with this code |
| 04:21 |
VanessaE |
wonder why github didn't find that right away |
| 04:21 |
RealBadAngel |
i manually edit filters adding author and "closed" |
| 04:23 |
RealBadAngel |
this code does the right thing, but is overengineed |
| 04:23 |
sofar |
building on my Intel HD4000 |
| 04:23 |
RealBadAngel |
factories, templates god knows what else |
| 04:24 |
RealBadAngel |
i will have to simplify it |
| 04:24 |
RealBadAngel |
good that code doesnt milk cows just by the way |
| 04:24 |
VanessaE |
fwiw, if I load up a bunch of map in my viewing direction, and go back to the starting point, the 'benchmark' scene still pulls 30 fps at 190m. that's pretty good, nearly double what I'm used to there. |
| 04:25 |
VanessaE |
er 180m |
| 04:25 |
VanessaE |
(specifically, that's about an 80 percent speed boost) |
| 04:28 |
RealBadAngel |
thats nice |
| 04:28 |
VanessaE |
something I noticed btw: my setup has abysmal 2d performance outside of MT as well, so anything you can do to eliminate 2d drawing events will help. |
| 04:29 |
VanessaE |
(it's noticeable in screensavers, for example. 3d performance outside of MT, including there, is fine) |
| 04:29 |
RealBadAngel |
theres not much left |
| 04:30 |
RealBadAngel |
VanessaE, how about memory usage? |
| 04:30 |
VanessaE |
I didn't look |
| 04:30 |
kaadmy |
i get abysmal 3d and 2d rendering |
| 04:30 |
sofar |
seems to work fine on Intel HD graphics |
| 04:30 |
sofar |
no issues, was entirely playable |
| 04:31 |
RealBadAngel |
VanessaE, if something went wrong you propably would notice it then quickly |
| 04:31 |
RealBadAngel |
for me client is running for >2hrs |
| 04:31 |
VanessaE |
RealBadAngel: right after starting the world, looking in the direction of the usual test scene, 1.5 GB with default textures. (this is a Dreambuilder world). |
| 04:31 |
RealBadAngel |
many server joins, shitload of worlds created etc |
| 04:32 |
RealBadAngel |
top says 2,4 gig atm, im on just test |
| 04:34 |
VanessaE |
about 2.3 GB after loading up a bunch of map. |
| 04:39 |
sofar |
My Intel HD gfx box is sitting at 1.0gb virt 450mb res |
| 04:39 |
sofar |
at spawn on just test with viewing range 100 |
| 04:40 |
RealBadAngel |
how about speed? |
| 04:42 |
sofar |
15-20fps, it slows down as more of the map loads |
| 04:42 |
sofar |
it starts at 40 |
| 04:42 |
RealBadAngel |
intel isnt a speed deamon ;) |
| 04:42 |
VanessaE |
that just means your view range is greater than your mapblock load range |
| 04:42 |
VanessaE |
(which is also why I try to always mention the drawtime as well as FPS) |
| 04:42 |
|
kaeza joined #minetest-dev |
| 04:42 |
RealBadAngel |
set distance to 100-150 and compare then |
| 04:42 |
sofar |
this is a 2nd gen Intel HD gfx |
| 04:43 |
sofar |
I need to get my hands on a 5th gen |
| 04:43 |
sofar |
or 6th :) |
| 04:43 |
RealBadAngel |
hehehe |
| 04:44 |
sofar |
actually it's a 3rd gen core i7 |
| 04:44 |
sofar |
I can't keep em apart anymore :) |
| 04:44 |
sofar |
I have a 4th gen in my NUC and one at work |
| 04:45 |
sofar |
the 6th gen NUCs are coming in the office soon |
| 04:45 |
VanessaE |
sofar: now try VE-Creative spawn. |
| 04:45 |
RealBadAngel |
VanessaE, VBO wont help there much |
| 04:46 |
VanessaE |
or better, "Basic" East City spawn |
| 04:46 |
RealBadAngel |
instant reloading of meshes :( |
| 04:46 |
VanessaE |
RealBadAngel: eh? |
| 04:46 |
RealBadAngel |
#3166 |
| 04:46 |
ShadowBot |
https://github.com/minetest/minetest/issues/3166 -- Send to clients only changed node metadata instad of whole mapblock by RealBadAngel |
| 04:46 |
VanessaE |
oh that |
| 04:46 |
VanessaE |
why isn't that merged yet? |
| 04:47 |
RealBadAngel |
its not finished |
| 04:47 |
VanessaE |
ok. |
| 04:48 |
RealBadAngel |
have to finish that asap |
| 04:48 |
sofar |
:) |
| 04:48 |
RealBadAngel |
fuck the water, thats way more important |
| 04:49 |
RealBadAngel |
water being visually fucked up have no impact on performance |
| 04:49 |
RealBadAngel |
this does. |
| 04:50 |
RealBadAngel |
before i go back to coding renderer i have to fix all those bottlenecks |
| 04:50 |
RealBadAngel |
i need ALL the power i can get ;) |
| 04:51 |
RealBadAngel |
2 fps here, 3 fps there and voila, time for water reflections found. for example ;) |
| 04:53 |
VanessaE |
I could gain another ~10 fps if I turn off desktop compositing, but I kinda need that (alt-mousewheel zooming, very useful for my bad eyes) |
| 04:54 |
VanessaE |
you should look into that, if there's a way to gain those 10 fps without disabling compositing. |
| 04:54 |
VanessaE |
(i.e. if there's something MT is doing that invokes the desktop compositing code) |
| 04:54 |
RealBadAngel |
i will check that |
| 04:56 |
VanessaE |
there was at one time something where the engine was rapidly varying the content of the window decorations, to show FPS, which ironically cost FPS by doing so. Long since removed, but I wonder if MT is doing other, similar stuff that just isn't easily visible. |
| 04:56 |
VanessaE |
but that was literally years ago. |
| 04:56 |
|
Dragonop_ joined #minetest-dev |
| 04:57 |
sofar |
RealBadAngel: 30 minutes now and my intel box is still only at 1gb virt minetest ram |
| 04:57 |
RealBadAngel |
so it looks like VBO doesnt have memory leaks problems anymore |
| 04:58 |
VanessaE |
I'll give it my +1, since it works and appears to not have problems (as if that's of any value :P ) |
| 04:58 |
VanessaE |
but someone more qualified than me will have to review and give it a real +1 |
| 05:09 |
RealBadAngel |
btw i do know how to make toggle to full screen |
| 05:10 |
VanessaE |
..and make it behave properly with dual-/triple-head? :) |
| 05:10 |
RealBadAngel |
thats quite easy |
| 05:10 |
VanessaE |
good |
| 05:10 |
RealBadAngel |
you just need to restart irr device |
| 05:10 |
sofar |
no issues at VE-basic spawn |
| 05:11 |
VanessaE |
sofar: in East City ? |
| 05:11 |
sofar |
1700mb virt mem use now tho! :) |
| 05:11 |
sofar |
I don't know |
| 05:11 |
sofar |
spawn? |
| 05:11 |
VanessaE |
(the oldest of old spawns) |
| 05:11 |
sofar |
teleport me there, lol |
| 05:11 |
sofar |
I don't know the way |
| 05:11 |
RealBadAngel |
problem is that device, driver etc are kept by the functions in the code as member variables |
| 05:11 |
VanessaE |
go to the current spawn, use the Travelnet booth near there to get to East City Central |
| 05:11 |
RealBadAngel |
that has to change |
| 05:12 |
RealBadAngel |
device, driver, scene manager etc have to be global |
| 05:12 |
VanessaE |
sofar, on my way.. |
| 05:12 |
sofar |
found it |
| 05:12 |
RealBadAngel |
no hiding stuff in the pockets |
| 05:12 |
VanessaE |
I see you already did. :) |
| 05:12 |
RealBadAngel |
then we just call the global func to shut irr down, restart it and voila |
| 05:14 |
VanessaE |
you sure that would work? |
| 05:14 |
RealBadAngel |
sure |
| 05:20 |
sofar |
hmm OpenGL 3.0 |
| 05:20 |
sofar |
can't I get more on Intel HD nowadays? |
| 05:25 |
sofar |
still under 2gb ram virt. I think that's enough cooking. |
| 05:32 |
sofar |
hmm |
| 05:32 |
sofar |
it's stuck shutting down |
| 05:32 |
VanessaE |
it does that to me too, sometimes. |
| 05:33 |
sofar |
oh, there it goes |
| 05:33 |
sofar |
#0 0x00007f84e05bd70d in pthread_join () from /usr/lib/libpthread.so.0 |
| 05:33 |
ShadowBot |
sofar: Error: Delimiter not found in "HTTP Error 404: Not Found" |
| 05:33 |
sofar |
#1 0x000000000057c2a8 in Thread::wait() () |
| 05:33 |
sofar |
#2 0x000000000074f050 in httpfetch_cleanup() () |
| 05:33 |
ShadowBot |
https://github.com/minetest/minetest/issues/1 -- GlowStone code by anonymousAwesome |
| 05:33 |
ShadowBot |
https://github.com/minetest/minetest/issues/2 -- Burned wood |
| 05:34 |
sofar |
weird, httpfetch timeout at shutdown? |
| 05:34 |
sofar |
anyway, my wife just came home from a 4-day trip |
| 05:34 |
* sofar |
afk |
| 05:43 |
sofar |
github down? losing 100% packets on "xo.net" route parts :( |
| 05:44 |
sofar |
entire west-to-east coast link down for me :( |
| 05:44 |
VanessaE |
works for me. |
| 05:44 |
VanessaE |
wonder what happened |
| 05:46 |
sofar |
frontier seattle - chicago link down |
| 05:47 |
VanessaE |
yeah, it's showing up in github's status graphs too as a sudden drop in response time |
| 05:47 |
VanessaE |
(== less traffic I guess) |
| 05:48 |
sofar |
SEA-ATL link is OK |
| 05:49 |
sofar |
heh, it was up 2 packets, now down again |
| 05:49 |
sofar |
back now |
| 05:50 |
sofar |
anyway, afk |
| 06:38 |
|
Hunterz joined #minetest-dev |
| 07:32 |
|
Alduin_ joined #minetest-dev |
| 08:23 |
|
nrzkt joined #minetest-dev |
| 08:31 |
|
ud1 joined #minetest-dev |
| 08:56 |
|
Megaf joined #minetest-dev |
| 09:33 |
|
asl97 joined #minetest-dev |
| 10:03 |
|
Indohmo joined #minetest-dev |
| 10:04 |
|
Indohmo joined #minetest-dev |
| 10:08 |
RealBadAngel |
n |
| 10:12 |
nrzkt |
m |
| 10:13 |
Indohmo |
o |
| 10:25 |
|
blaze joined #minetest-dev |
| 10:57 |
|
RealBadAngel joined #minetest-dev |
| 11:16 |
|
nrzkt joined #minetest-dev |
| 11:16 |
|
Samson1 joined #minetest-dev |
| 11:33 |
|
OldCoder joined #minetest-dev |
| 11:45 |
|
VanessaE joined #minetest-dev |
| 11:45 |
|
exoplanet joined #minetest-dev |
| 12:16 |
|
Fixer joined #minetest-dev |
| 12:45 |
Indohmo |
RealBadAngel, would it be possible to make main_menu music continue while the loading ? |
| 12:46 |
RealBadAngel |
idk |
| 12:47 |
Fixer |
._. |
| 12:47 |
Fixer |
VanessaE, does your benchmark has technic and stuff? |
| 12:47 |
Fixer |
VanessaE, also, maybe you are limited by CPU |
| 12:51 |
RealBadAngel |
she has 6 or 8 cores cpu, so rather not |
| 12:51 |
Fixer |
hmhmh |
| 12:51 |
RealBadAngel |
and the gpu is r9 |
| 12:52 |
Fixer |
iirc r9 is somewhat slow in linux? |
| 12:52 |
RealBadAngel |
looks like so |
| 12:52 |
Fixer |
one of the best performance was hd 6870 (that i have btw) |
| 12:52 |
Fixer |
bad thing is lack of vulkan support but that way off the charts now anyway |
| 12:53 |
Fixer |
RealBadAngel, good place to test VBO is LinuxGaming 2 |
| 12:53 |
Fixer |
RealBadAngel, in 0.4.12 it was eating 6GB of RAM, 0.4.13 - 2GB |
| 12:54 |
RealBadAngel |
for servers env i need to finish that meta thingy, im planning to do that today |
| 12:55 |
RealBadAngel |
wont go to sleep but i will finish that damn thing |
| 12:55 |
Fixer |
RealBadAngel, RAM usage of client* |
| 12:55 |
Fixer |
my build is ready |
| 12:56 |
Fixer |
will test vbo as much as i can |
| 12:56 |
RealBadAngel |
Fixer, i had client with VBO running for a few hours |
| 12:56 |
RealBadAngel |
had no problems with memory usage |
| 12:57 |
Fixer |
RealBadAngel, RAM usage is key thing to test (but i have 8Gig anyway), thank ctulhu i've update my PC before 2x devaluation |
| 12:58 |
Fixer |
RealBadAngel, don't see VBO thing in minetest.conf.example |
| 12:59 |
Fixer |
i see it ingame config, enabled by default |
| 12:59 |
Fixer |
ok |
| 13:05 |
RealBadAngel |
and? |
| 13:05 |
Fixer |
RealBadAngel, mapgen stutters are horrible in singleplayer, while fps can be 30-60 at 240 (30 if huge forest) |
| 13:06 |
Fixer |
RealBadAngel, RAM usage in huge forest is 1GB (as before) |
| 13:06 |
Fixer |
RealBadAngel, will look much more |
| 13:06 |
RealBadAngel |
ok |
| 13:08 |
RealBadAngel |
going out, be beack in an hour |
| 13:09 |
Fixer |
RealBadAngel, lots of stuttering in mp when loading blocks in j-t not far from spawn |
| 13:23 |
Fixer |
RealBadAngel, right now from J-T on my usual config got +20% fps boost, on other configs - up to +35% |
| 13:27 |
|
casimir joined #minetest-dev |
| 13:31 |
|
ElectronLibre joined #minetest-dev |
| 13:43 |
|
iqualfragile joined #minetest-dev |
| 13:47 |
Fixer |
RealBadAngel, more testing: 17-35% fps boost at this moment |
| 13:55 |
|
lubuntu_ joined #minetest-dev |
| 13:56 |
|
Wayward1 joined #minetest-dev |
| 14:10 |
|
Player_2 joined #minetest-dev |
| 14:13 |
|
Megaf joined #minetest-dev |
| 14:21 |
Fixer |
RealBadAngel, why gain is so large in singleplayer? +80% and more |
| 14:22 |
|
hmmmm joined #minetest-dev |
| 14:27 |
|
kaadmy joined #minetest-dev |
| 14:32 |
|
paramat joined #minetest-dev |
| 14:37 |
Megaf |
gain? |
| 14:41 |
Megaf |
what do you mean Fixer ? |
| 14:48 |
Fixer |
Megaf, fps increased |
| 14:57 |
paramat |
!tell hmmmm sheet ore PR in case you're interested https://github.com/minetest/minetest/pull/3762 might merge later |
| 14:57 |
ShadowBot |
paramat: O.K. |
| 15:01 |
paramat |
can anyone +1 #3741 ? code is already reviewed just needs a 2nd +1 |
| 15:01 |
ShadowBot |
https://github.com/minetest/minetest/issues/3741 -- Mgvalleys: Add Dry Riverbeds by duane-r |
| 15:05 |
|
asl97 joined #minetest-dev |
| 15:07 |
paramat |
!tell hmmmm PR for PcgRandom https://github.com/minetest/minetest/pull/3746 |
| 15:07 |
ShadowBot |
paramat: O.K. |
| 15:08 |
|
kaeza joined #minetest-dev |
| 15:08 |
|
Megaf joined #minetest-dev |
| 15:12 |
|
est31 joined #minetest-dev |
| 15:12 |
Indohmo |
paramat, it seems like you are the only one merging all those PRs :( |
| 15:12 |
Indohmo |
oh ! |
| 15:13 |
Indohmo |
mea culpa, est31 does too :p |
| 15:18 |
paramat |
nore sfan5 any comments on removing 'dig up' from cacti for this reason? https://github.com/minetest/minetest_game/pull/847#issuecomment-187224272 |
| 15:18 |
|
asl97 joined #minetest-dev |
| 15:18 |
paramat |
will merge game#858 later |
| 15:18 |
ShadowBot |
https://github.com/minetest/minetest_game/issues/858 -- Door protection fix, and fallback compatibility code for mods. by sofar |
| 15:19 |
Fixer |
RealBadAngel, posted in 3763 |
| 15:20 |
kaadmy |
paramat: can't cactuses be attached nodes? |
| 15:20 |
kaadmy |
will that break existing worlds? |
| 15:20 |
paramat |
will consider it |
| 15:21 |
kaadmy |
oh attached nodes can't be placed on each other |
| 15:21 |
est31 |
attached nodes? |
| 15:21 |
kaadmy |
yeah |
| 15:21 |
kaadmy |
like torches |
| 15:21 |
est31 |
wats that |
| 15:21 |
est31 |
wallmounted |
| 15:21 |
est31 |
ah |
| 15:21 |
kaadmy |
they pop into drops when dugbup |
| 15:21 |
Indohmo |
is #3739 OK ? |
| 15:21 |
ShadowBot |
https://github.com/minetest/minetest/issues/3739 -- Add support for Mumble positional audio by ShadowNinja |
| 15:21 |
kaadmy |
or unattached |
| 15:22 |
est31 |
last time I checked it was WIP |
| 15:30 |
Indohmo |
#2375 ??? |
| 15:30 |
ShadowBot |
https://github.com/minetest/minetest/issues/2375 -- Fix superflous shader setting updates by rubenwardy |
| 15:34 |
paramat |
i think RBA may be intending to rework that PR |
| 15:36 |
Fixer |
why on earth VBO gives so huge fps boost on singleplayer, but very slight in mp :/ |
| 15:37 |
Fixer |
well 20-35% is not slight, but singleplayer got much more |
| 15:42 |
|
Fixer joined #minetest-dev |
| 15:51 |
Indohmo |
What's new with it ? https://github.com/minetest/minetest/pull/3208 |
| 15:55 |
Indohmo |
this* |
| 16:20 |
RealBadAngel |
back |
| 16:20 |
RealBadAngel |
fixer because of metadata issue |
| 16:21 |
RealBadAngel |
servers kills clients with it |
| 16:22 |
RealBadAngel |
paramat, yes, i have to rework it, its very important pr |
| 16:24 |
kaeza_ |
RealBadAngel: long texture strings also cause lots of FPS problems. I haven't looked too much at the code, but can't the texture be cached somewhere? |
| 16:24 |
|
Dragonop joined #minetest-dev |
| 16:25 |
RealBadAngel |
kaeza_, textures ARE cached |
| 16:25 |
kaeza_ |
long as in signs lib long ;) |
| 16:25 |
RealBadAngel |
rotfl |
| 16:25 |
RealBadAngel |
dont tell me bout signs lib, please |
| 16:25 |
RealBadAngel |
this mod is a diseaster |
| 16:25 |
RealBadAngel |
total abuse ^2 |
| 16:26 |
kaeza_ |
yes it is, but what's the alternative? |
| 16:26 |
RealBadAngel |
sooner we will have proper signlike drawtype the sooner we can get rid of this monstrous code |
| 16:27 |
RealBadAngel |
alternative is the specialized drawtype |
| 16:27 |
kaeza_ |
^[text:foo would be nice |
| 16:27 |
RealBadAngel |
theres no other way |
| 16:27 |
RealBadAngel |
maybe i will be able to get that working before .14 release |
| 16:28 |
kaeza_ |
uh, maybe sspecial_tiles hack? |
| 16:28 |
RealBadAngel |
but that depends on that if i can fix 3166 |
| 16:28 |
RealBadAngel |
no hacks, please |
| 16:28 |
RealBadAngel |
we have suffered from hacks too much already |
| 16:28 |
RealBadAngel |
see inventory images |
| 16:29 |
RealBadAngel |
either this will be made proper way or none |
| 16:29 |
kaeza_ |
well, I mean a special tiles mode like we have animations |
| 16:30 |
RealBadAngel |
place where u put texture is not important |
| 16:30 |
RealBadAngel |
important is what happens to those textures |
| 16:31 |
RealBadAngel |
theyre got stacked on client side, never deleted |
| 16:31 |
RealBadAngel |
well, they got deleted. on client shutdown |
| 16:32 |
RealBadAngel |
proper signlike means drawtype that will get text, render it to texture on client side, keep the texture as long as its needed |
| 16:33 |
RealBadAngel |
on update trash the texture and render new |
| 16:33 |
RealBadAngel |
generating textures server side and sending them to clients all the time is SICK |
| 16:34 |
RealBadAngel |
that mod shall be blocked long time ago |
| 16:34 |
VanessaE |
I keep telling you, RealBadAngel, you can't just make it "signlike" |
| 16:34 |
VanessaE |
otherwise you have no yard signs, no hanging signs, etc. |
| 16:35 |
VanessaE |
it HAS to be something that can be applied to any node/mesh shape. |
| 16:35 |
RealBadAngel |
idk if it "has", one thing i know that signs lib shall not exists at all |
| 16:36 |
VanessaE |
yes, it has to be. |
| 16:36 |
VanessaE |
any replacement has to be able to be a complete superset. |
| 16:37 |
VanessaE |
doesn't matter what else it can do if it can't replace the colors, formatting, sign shapes, etc that are possible with signs_lib |
| 16:37 |
RealBadAngel |
engine doesnt have to imitate some sick mod |
| 16:37 |
|
Jeija joined #minetest-dev |
| 16:37 |
VanessaE |
in fact, yes it does. |
| 16:37 |
RealBadAngel |
engine has to provide proper ways of doing things |
| 16:37 |
VanessaE |
you don't replace a de-facto standard with an inferior product and expect the replacement to be well-received. |
| 16:38 |
RealBadAngel |
proper drawtype is not a replacement |
| 16:38 |
RealBadAngel |
and i wouldnt dare calling total abuse a standard |
| 16:38 |
|
Krock joined #minetest-dev |
| 16:38 |
VanessaE |
it's a standard. |
| 16:39 |
VanessaE |
most servers have signs_lib I'm sure by now |
| 16:39 |
RealBadAngel |
not from point of view of the engine |
| 16:39 |
VanessaE |
the engine doesn't define what is, or isn't standard. the users do.. |
| 16:39 |
VanessaE |
(or the server owners, to be more exact) |
| 16:40 |
|
hmmmm joined #minetest-dev |
| 16:40 |
RealBadAngel |
for start i wil write a generic thing |
| 16:41 |
RealBadAngel |
we will see of what its gonna be capable |
| 16:41 |
RealBadAngel |
but definitely i wont have in mind one mod coding it |
| 16:42 |
RealBadAngel |
and definitely it wont be compatible |
| 16:42 |
VanessaE |
then it will fail and no one will use it. |
| 16:43 |
Indohmo |
https://github.com/minetest/minetest/pull/3208 ? |
| 16:43 |
RealBadAngel |
sorry but i cant code another stuff that i know that will kill performance |
| 16:43 |
Indohmo |
Would be better than teleporting the player IMO :p |
| 16:43 |
VanessaE |
RealBadAngel: no one said anything about performance-killing. |
| 16:43 |
nrzkt |
the more important is to have a good design for signs. signs_lib works but the most != the best in this case |
| 16:44 |
VanessaE |
the whole damn point is to let the modder put the text anywhere they want, fitting a specific size/shape of bounding box, and color it in any way they want. |
| 16:44 |
RealBadAngel |
thats propably how its gonna work, but forget about ANY compability with signlib |
| 16:45 |
VanessaE |
signs_lib can be modified to be compatible with the engine's way of doing things. |
| 16:45 |
VanessaE |
the point is, don't lose any features. |
| 16:45 |
RealBadAngel |
idk if i can code from start any feature that some mod has |
| 16:45 |
RealBadAngel |
that wont be even my goal |
| 16:45 |
VanessaE |
what? |
| 16:46 |
VanessaE |
what's so complicated about letting a mod specify the size, shape, and position of the text bounding box, and put the text there? With color and very basic formatting? |
| 16:46 |
RealBadAngel |
i dont intend even trying to read signs lib code before i made engine code |
| 16:46 |
RealBadAngel |
i know whats have to be done and i will just code it |
| 16:47 |
est31 |
gonna push this in 10 minutes https://github.com/est31/minetest/commit/a26970cdd4aec995416566cd75dd271c6b485f16 |
| 16:47 |
RealBadAngel |
if modders were so smart to abuse the engine so much to get those signs, they will propably made even more fancy signs with new code |
| 16:47 |
est31 |
nrzkt, it should make android jenkins build again |
| 16:48 |
RealBadAngel |
and easier |
| 16:48 |
RealBadAngel |
est31, nrzkt can you take a look on VBO pr? |
| 16:48 |
VanessaE |
RealBadAngel: again, and again, AND AGAIN! let the modder specify the size, shape, and position of the text bounding box. Let the engine put the specified text in that space, with color support. THAT IS ALL! |
| 16:49 |
VanessaE |
jesus fuck man, why are you overcomplicating the problem?! |
| 16:49 |
RealBadAngel |
you want the custom meshes |
| 16:49 |
RealBadAngel |
i want text to texture |
| 16:49 |
RealBadAngel |
thats a difference |
| 16:50 |
VanessaE |
of course I want custom meshes. you want to out text in a fixed position with absolutely no variation. |
| 16:50 |
VanessaE |
by doing that you guarantee signs_lib will never be obsoleted. |
| 16:50 |
VanessaE |
put* |
| 16:50 |
RealBadAngel |
just wait for the PR |
| 16:50 |
nrzkt |
est31, cool |
| 16:50 |
RealBadAngel |
until then discussion is pure academic |
| 16:51 |
Jeija |
For some reason https://git.io/v2f6j seems to break request_insecure_environment for me. Calling print("environment:" .. dump(minetest.request_insecure_environment())) directly from init.lua of a secure.trusted_mod gives me nil. Can someone confirm? |
| 16:52 |
VanessaE |
greetz, Jeija. |
| 16:52 |
Jeija |
Hi VanessaE |
| 16:53 |
Indohmo |
#2852 #2827 |
| 16:53 |
ShadowBot |
https://github.com/minetest/minetest/issues/2852 -- Make players respect makes_footstep_sound in the object properties by TeTpaAka |
| 16:53 |
ShadowBot |
https://github.com/minetest/minetest/issues/2827 -- Allow overriding the hand by TeTpaAka |
| 16:53 |
Jeija |
I just compared minetest with the commit and without, and it doesn't work with the commit. |
| 16:53 |
Jeija |
Tested both LuaJIT and bundled Lua. |
| 16:55 |
est31 |
okay then file a bug |
| 16:55 |
VanessaE |
RealBadAngel: like I said, if the modder can't write a BRAND NEW MOD that creates an ordinary mesh and places a text box somewhere in that node, then it's not going to work and it's not going to obsolete signs_lib. plus, it has to obsolete his ^^^^ digiline LCD screens too. |
| 16:56 |
VanessaE |
the whole point is that it should be possible to reduce signs_lib to just a collection of custom sign models, letting the engine handle everything else. |
| 16:56 |
|
kaeza joined #minetest-dev |
| 16:57 |
RealBadAngel |
est31, nrzkt ive put WIP label on the VBO pr but the code is plain dumb. and it looks like i got rid of the flaws old c55/PA VBO patch |
| 16:57 |
kaeza |
okay. back on PC |
| 16:58 |
RealBadAngel |
there are no memory leaks, i even had client running for several hours connecting all the time to different servers, creating single player worlds all over |
| 16:58 |
RealBadAngel |
without closing the client at all, and memory usage was stable |
| 17:01 |
nrzkt |
RealBadAngel, i need to merge all you rprevious commits in my branch before looking at this |
| 17:02 |
|
Hunterz joined #minetest-dev |
| 17:02 |
sofar |
VanessaE: I'd love to create a new SIGNLIKE nodetype that the client can render text on... it requires irrlicht font rendering so it should all be possible |
| 17:02 |
sofar |
VanessaE: picking the font will be the hardest part to get through a PR ;) |
| 17:02 |
RealBadAngel |
sofar, i do have renderer ready |
| 17:02 |
RealBadAngel |
thats not a problem |
| 17:02 |
est31 |
it should work on android |
| 17:02 |
est31 |
thats my main concern |
| 17:03 |
VanessaE |
sofar: well you could always take signs_lib's font and generate a TTF out of it ;) |
| 17:03 |
sofar |
you'd have to make a nodebox + text surface nodetype |
| 17:03 |
est31 |
and it should not do "meta set nodedef" not yet |
| 17:03 |
RealBadAngel |
est31, it will work on any platform |
| 17:03 |
sofar |
I think a nodebox drawtype extension "NDT_TEXT" that has a planar surface for text would do fine |
| 17:04 |
sofar |
so you can place the text surface on any part of the nodebox |
| 17:04 |
RealBadAngel |
sofar, i want it to be more generic |
| 17:04 |
VanessaE |
sofar: that's precisely what I've been saying |
| 17:04 |
RealBadAngel |
i want a tile property, text instead of texture |
| 17:05 |
RealBadAngel |
so you can put the tile on whatever you want |
| 17:05 |
RealBadAngel |
even on imported mesh |
| 17:05 |
kaeza |
If you're going the TTF rendering way, please provide a facility to add colored text. That is one of the features loved in signs_lib (along with variable width fonts) |
| 17:06 |
RealBadAngel |
so you can put that on any model you want |
| 17:06 |
sofar |
RealBadAngel: how do you go from tile to 'this is the surface where you should render the text' ? |
| 17:06 |
RealBadAngel |
just define tiles as usual, but instead of giving texture, pass there text |
| 17:06 |
sofar |
kaeza: should be trivial to add |
| 17:07 |
sofar |
RealBadAngel: but, how to tell it where to draw the text? |
| 17:07 |
VanessaE |
sofar: RealBadAngel wants a drawtype that a big solid cube with one side dedicated to text display :P |
| 17:07 |
VanessaE |
that's* |
| 17:07 |
sofar |
and how do you 'change' the text without changing the tiles? is that even possible? |
| 17:07 |
RealBadAngel |
how do you define which face is using which tile? |
| 17:07 |
RealBadAngel |
same way |
| 17:07 |
sofar |
I don't want cubes |
| 17:08 |
sofar |
I want arbitrary faces with text |
| 17:08 |
RealBadAngel |
meshes are also using tiles |
| 17:08 |
sofar |
yes but how do I place the text on one of the mesh faces? |
| 17:09 |
VanessaE |
RealBadAngel: show us a proposed node def table. |
| 17:09 |
RealBadAngel |
define uv in blender for example |
| 17:09 |
VanessaE |
wat |
| 17:09 |
sofar |
I guess I don't see how tiles would work. I am missing some coords that I think are needed |
| 17:09 |
RealBadAngel |
on a face where do you want sign text |
| 17:09 |
RealBadAngel |
and instead of supplying texture for that face, give there text |
| 17:09 |
RealBadAngel |
engine will render it |
| 17:09 |
sofar |
I also think that this method should be generic and work for both mesh, nodebox and possibly others |
| 17:10 |
VanessaE |
and..um.. how does the mod or user change the text? you can't change a tile after it's defined. |
| 17:10 |
sofar |
^^ |
| 17:10 |
RealBadAngel |
thats a whole trick ;) |
| 17:10 |
VanessaE |
why isn't the text stored in a metadata item instead? |
| 17:10 |
sofar |
that's what I think is the proper solution |
| 17:10 |
VanessaE |
what sense is there in storing it in a tile slot, when that tile slot is gonna be needed for node textures anyway? |
| 17:11 |
sofar |
Q: is it arbitrary node data? A: yes -> node metadata is the proper storage |
| 17:11 |
RealBadAngel |
yes, its gonna be stored in meta |
| 17:11 |
RealBadAngel |
but now it simply cant be stored |
| 17:11 |
VanessaE |
since when? |
| 17:11 |
RealBadAngel |
until i fix it |
| 17:11 |
RealBadAngel |
since always |
| 17:11 |
VanessaE |
what do you think a default sign (no signs_lib) uses to store the sign text now? |
| 17:12 |
sofar |
RealBadAngel: the basics are (1) where does the text go (surface, coords), and (2) what is the text content |
| 17:12 |
sofar |
(1) I can see multiple solutions |
| 17:12 |
RealBadAngel |
moving shit in a chest deletes whole metas of a mapblock |
| 17:12 |
VanessaE |
so what? |
| 17:12 |
VanessaE |
we're not talking about that. |
| 17:12 |
RealBadAngel |
editing one sign too |
| 17:12 |
RealBadAngel |
thats a big problem |
| 17:12 |
VanessaE |
that's a bit problem for another discussion |
| 17:12 |
RealBadAngel |
all your meta is lost |
| 17:12 |
sofar |
implementation problem |
| 17:12 |
VanessaE |
big* |
| 17:12 |
sofar |
not a concept problem |
| 17:13 |
VanessaE |
focus on defining what the damn node can store, where the text goes, and what it looks like |
| 17:13 |
sofar |
conceptually node metadata is fine |
| 17:13 |
RealBadAngel |
its a valid reason why it cannot be done yet |
| 17:13 |
sofar |
RealBadAngel: efficiently |
| 17:13 |
sofar |
anyway, tea time |
| 17:13 |
RealBadAngel |
imagine whole mapblock of signs regenerating their textures |
| 17:13 |
VanessaE |
look, if signs_lib can do it, the engine should be able to do it 100 times better. |
| 17:14 |
RealBadAngel |
another diseaster |
| 17:14 |
RealBadAngel |
dont remind me signs lib all the time please |
| 17:14 |
RealBadAngel |
sings are entities in the first place with it |
| 17:14 |
VanessaE |
then stop making this sound like it's gonna be one-colored text on one specific shape and size of node |
| 17:14 |
RealBadAngel |
we are not talking about entities anymore |
| 17:15 |
kaeza |
beetlesigns_lib, beetlesigns_lib, beetlesigns_lib |
| 17:15 |
RealBadAngel |
i just said its gonna be generic, to use on any model |
| 17:15 |
VanessaE |
RealBadAngel: then tell us HOW damn it |
| 17:15 |
RealBadAngel |
havent mentioned colors even :P |
| 17:15 |
VanessaE |
I asked for a node def example |
| 17:16 |
RealBadAngel |
how can i give you sample nodedef when theres no code yet? lol |
| 17:16 |
VanessaE |
um, because you need a plan? |
| 17:16 |
sofar |
RealBadAngel: huh, wait |
| 17:16 |
kaeza |
RealBadAngel, it's called a "prototype" |
| 17:16 |
sofar |
RealBadAngel: the *client* should render the fonts |
| 17:16 |
sofar |
RealBadAngel: why would that be a problem? |
| 17:16 |
sofar |
<RealBadAngel> imagine whole mapblock of signs regenerating their textures |
| 17:17 |
kaeza |
that is a different issue altogether |
| 17:17 |
sofar |
designing the API before the implementation is usually a good exercise |
| 17:17 |
RealBadAngel |
sofar, that is exactly what was happening to my renderer |
| 17:18 |
RealBadAngel |
whole damn meta is deleted all the time by unrelated stuff |
| 17:18 |
VanessaE |
RealBadAngel: stop talking about what the client's doing with metadata right now. we don't care if opening a triggers some regen somewhere else on the map. |
| 17:18 |
sofar |
I'm not saying it's going to be correct, but I think it's better than starting coding without a plan |
| 17:18 |
VanessaE |
we care about how the fucking thing is to be coded independently of that |
| 17:18 |
RealBadAngel |
for sake of damn backwards compability with clients from 0.3 era |
| 17:18 |
sofar |
RealBadAngel: well, implement partial per-node meta updates then ;) |
| 17:18 |
kaeza |
also, what I meant previously by "caching" a texture is to store an (incref'd) ITexture* somewhere instead of looking up the texture every damn frame |
| 17:18 |
VanessaE |
if opening a chest* |
| 17:18 |
RealBadAngel |
sofar, 3166 is the code |
| 17:19 |
sofar |
kaeza: text should only be rendered on mesh update, not every frame |
| 17:19 |
RealBadAngel |
that effectively fixes the problem |
| 17:19 |
RealBadAngel |
i need just to polish the code and make it more effective |
| 17:19 |
sofar |
I know, 3166 is a good improvement |
| 17:20 |
RealBadAngel |
it works already |
| 17:20 |
RealBadAngel |
epixel is using it since half a yr or something like that |
| 17:20 |
sofar |
there's quite a few PR's like that |
| 17:20 |
sofar |
the list predict one is great too |
| 17:20 |
sofar |
can't believe it's just sitting there :( |
| 17:20 |
RealBadAngel |
i had no time before to finish it |
| 17:21 |
kaeza |
sofar, oh wait, I'm thinking of other stuff |
| 17:21 |
RealBadAngel |
and its not that easy |
| 17:21 |
RealBadAngel |
but hey, ive fixed many old stuff lately |
| 17:22 |
RealBadAngel |
vbo was here for yrs already |
| 17:22 |
RealBadAngel |
and now you can play with it |
| 17:26 |
RealBadAngel |
sofar, but we can work on the signs together once im done with meta |
| 17:27 |
RealBadAngel |
maybe we could get together somethign even more generic |
| 17:27 |
|
32NAAC280 joined #minetest-dev |
| 17:27 |
RealBadAngel |
two heads are better than one |
| 17:28 |
sofar |
fixes schmixes |
| 17:28 |
sofar |
they don't count until they merge :) |
| 17:31 |
RealBadAngel |
i think i will start on 3166 today |
| 17:32 |
RealBadAngel |
all other stuff is merged, vbo needs shitload of testing to be sure |
| 17:32 |
RealBadAngel |
i can spend day or two only on meta stuff |
| 17:42 |
|
nrzkt joined #minetest-dev |
| 17:47 |
|
nrzkt joined #minetest-dev |
| 17:57 |
|
est31 joined #minetest-dev |
| 18:06 |
|
Calinou joined #minetest-dev |
| 18:11 |
|
Gory joined #minetest-dev |
| 18:17 |
|
nrzkt joined #minetest-dev |
| 18:38 |
|
numzero joined #minetest-dev |
| 18:45 |
|
DevBox joined #minetest-dev |
| 18:49 |
|
numzero_ joined #minetest-dev |
| 18:55 |
|
ud1 joined #minetest-dev |
| 18:57 |
|
hmmmm joined #minetest-dev |
| 19:30 |
|
Dragonop joined #minetest-dev |
| 20:05 |
|
paramat left #minetest-dev |
| 20:07 |
|
Robby joined #minetest-dev |
| 20:18 |
|
fireglow joined #minetest-dev |
| 20:21 |
|
Warr1024 joined #minetest-dev |
| 20:22 |
|
Ritchie joined #minetest-dev |
| 20:25 |
|
Sokomine joined #minetest-dev |
| 20:27 |
|
Megaf joined #minetest-dev |
| 20:37 |
|
est31 joined #minetest-dev |
| 21:00 |
|
paramat joined #minetest-dev |
| 21:00 |
paramat |
game#867 (WIP) |
| 21:00 |
ShadowBot |
https://github.com/minetest/minetest_game/issues/867 -- Default/mapgen: Make forest clearings larger and more common WIP by paramat |
| 21:09 |
|
DFeniks joined #minetest-dev |
| 21:11 |
|
dfelinto joined #minetest-dev |
| 21:11 |
|
ssieb joined #minetest-dev |
| 21:43 |
|
Wuzzy joined #minetest-dev |
| 22:12 |
|
Indohmo joined #minetest-dev |
| 22:14 |
|
dfelinto_ joined #minetest-dev |
| 22:19 |
paramat |
game#866 |
| 22:19 |
ShadowBot |
https://github.com/minetest/minetest_game/issues/866 -- Remove dig_up() from cactus. by MT-Modder |
| 22:19 |
|
paramat left #minetest-dev |
| 22:42 |
|
est31 joined #minetest-dev |
| 23:29 |
|
STHGOM joined #minetest-dev |
| 23:29 |
|
STHGOM_ joined #minetest-dev |
| 23:36 |
|
Taoki joined #minetest-dev |
| 23:53 |
|
Taoki joined #minetest-dev |
