| Time |
Nick |
Message |
| 00:00 |
|
proller__ joined #minetest-dev |
| 00:00 |
|
bigfoot547 joined #minetest-dev |
| 00:03 |
|
bigfoot547_ joined #minetest-dev |
| 00:31 |
|
Arthricyn1 joined #minetest-dev |
| 00:38 |
|
Arthricyn joined #minetest-dev |
| 00:41 |
|
Arthricyn1 joined #minetest-dev |
| 00:48 |
|
Arthricyn joined #minetest-dev |
| 00:52 |
|
Arthricyn1 joined #minetest-dev |
| 00:55 |
|
Arthricyn joined #minetest-dev |
| 00:57 |
|
Arthricyn1 joined #minetest-dev |
| 01:00 |
|
Arthricyn joined #minetest-dev |
| 01:10 |
|
Arthricyn1 joined #minetest-dev |
| 01:15 |
|
Arthricyn joined #minetest-dev |
| 01:16 |
|
ThomasMonroe joined #minetest-dev |
| 01:16 |
|
ThomasMonroe joined #minetest-dev |
| 01:25 |
|
Arthricyn1 joined #minetest-dev |
| 01:28 |
|
Arthricyn joined #minetest-dev |
| 01:29 |
|
kaeza joined #minetest-dev |
| 01:38 |
|
Arthricyn1 joined #minetest-dev |
| 01:39 |
|
Tmanyo joined #minetest-dev |
| 02:57 |
|
octacian__ joined #minetest-dev |
| 03:04 |
|
MoNTE48 joined #minetest-dev |
| 03:30 |
|
Lunatrius` joined #minetest-dev |
| 04:10 |
|
octacian joined #minetest-dev |
| 04:52 |
|
MoNTE48 joined #minetest-dev |
| 05:01 |
|
DI3HARD139 joined #minetest-dev |
| 05:01 |
|
DI3HARD139 joined #minetest-dev |
| 05:32 |
|
YuGiOhJCJ joined #minetest-dev |
| 05:43 |
|
nerzhul joined #minetest-dev |
| 06:26 |
|
Icedream joined #minetest-dev |
| 06:27 |
|
vifino joined #minetest-dev |
| 06:40 |
|
MoNTE48 joined #minetest-dev |
| 06:41 |
|
Hunterz joined #minetest-dev |
| 06:46 |
|
Bobr2 joined #minetest-dev |
| 07:12 |
|
kilbith joined #minetest-dev |
| 07:32 |
|
behalebabo joined #minetest-dev |
| 07:34 |
|
Raven262 joined #minetest-dev |
| 07:54 |
|
paramat joined #minetest-dev |
| 08:06 |
|
behalebabo joined #minetest-dev |
| 08:23 |
|
behalebabo joined #minetest-dev |
| 08:29 |
|
MoNTE48 joined #minetest-dev |
| 08:32 |
|
behalebabo joined #minetest-dev |
| 08:44 |
|
octacian_ joined #minetest-dev |
| 08:57 |
|
lisac joined #minetest-dev |
| 09:11 |
|
kilbith joined #minetest-dev |
| 09:28 |
|
Darcidride joined #minetest-dev |
| 09:29 |
|
kilbith joined #minetest-dev |
| 09:34 |
|
kilbith joined #minetest-dev |
| 10:17 |
|
MoNTE48 joined #minetest-dev |
| 10:20 |
|
DS-minetest joined #minetest-dev |
| 10:23 |
|
juhdanad joined #minetest-dev |
| 10:25 |
|
rubenwardy joined #minetest-dev |
| 10:27 |
|
kilbith joined #minetest-dev |
| 10:46 |
|
proller__ joined #minetest-dev |
| 10:51 |
|
fireglow joined #minetest-dev |
| 10:51 |
|
jin_xi joined #minetest-dev |
| 10:58 |
|
Fixer joined #minetest-dev |
| 10:59 |
|
lisac joined #minetest-dev |
| 11:05 |
|
juhdanad joined #minetest-dev |
| 11:08 |
|
Darcidride joined #minetest-dev |
| 11:27 |
juhdanad |
Is it possible to debug a mod in an IDE? (use breakpoints, read variables etc...) |
| 11:35 |
|
ThomasMonroe joined #minetest-dev |
| 11:35 |
|
ThomasMonroe joined #minetest-dev |
| 11:52 |
|
DS-minetest joined #minetest-dev |
| 12:01 |
|
lisac joined #minetest-dev |
| 12:08 |
|
proller joined #minetest-dev |
| 12:23 |
|
proller joined #minetest-dev |
| 12:27 |
|
RobbieF joined #minetest-dev |
| 12:39 |
|
proller joined #minetest-dev |
| 12:53 |
|
nerzhul joined #minetest-dev |
| 13:00 |
|
proller joined #minetest-dev |
| 13:04 |
|
kilbith joined #minetest-dev |
| 13:31 |
|
DS-minetest joined #minetest-dev |
| 13:40 |
|
DS-minetest joined #minetest-dev |
| 13:40 |
|
est31 joined #minetest-dev |
| 13:42 |
|
est31 left #minetest-dev |
| 14:01 |
|
proller joined #minetest-dev |
| 14:14 |
KaadmY |
Should type casts be "joined" when their parent variable? |
| 14:14 |
KaadmY |
Ie. `float x = (float)otherNumber` |
| 14:14 |
KaadmY |
Or `float x = (float) otherNumber` |
| 14:15 |
KaadmY |
The code I'm looking at uses the former |
| 14:15 |
KaadmY |
And I don't think the style guidelines mention that |
| 14:15 |
|
paramat joined #minetest-dev |
| 14:16 |
KaadmY |
Hullo |
| 14:16 |
paramat |
our code uses joined |
| 14:16 |
KaadmY |
Okay |
| 14:26 |
paramat |
an idea for more intuitive crafting recipes for inner/outer stairs https://github.com/minetest/minetest_game/pull/1820#issuecomment-313696928 |
| 14:31 |
KaadmY |
Having a builtin microblock system would be so awesome |
| 14:31 |
KaadmY |
So you could make block "schematics" with a 2x2x2 grid |
| 14:32 |
|
kilbith joined #minetest-dev |
| 14:39 |
|
juhdanad joined #minetest-dev |
| 14:48 |
|
ThomasMonroe joined #minetest-dev |
| 14:48 |
|
ThomasMonroe joined #minetest-dev |
| 14:55 |
|
juhdanad joined #minetest-dev |
| 14:57 |
|
Karazhan joined #minetest-dev |
| 15:05 |
VanessaE |
paramat: you might consider "splitting" the faces on the foot of the stairs as is done with the top. |
| 15:05 |
VanessaE |
so that the wood grain bevels together |
| 15:13 |
paramat |
i'm not sure if that graphic is of MT, i'll check what the PR does |
| 15:15 |
|
octacian__ joined #minetest-dev |
| 15:16 |
|
Darcidride joined #minetest-dev |
| 15:17 |
|
octacian joined #minetest-dev |
| 15:19 |
VanessaE |
oh ok |
| 15:20 |
VanessaE |
still, it's easily done |
| 15:21 |
VanessaE |
if you wanted to be fancy-schmancy, you'd add a second material to the model that looks like it' overlaid on the diagonal seam :) (that way wooden stairs could have an extra strip of wood over the seam, bricks could have a line of mortar, etc) |
| 15:21 |
VanessaE |
it's* |
| 15:22 |
VanessaE |
(well, tumeninodes would have to do that rather) |
| 15:24 |
paramat |
GreenXenith made the .obj models for the PR, guess it would be texture-wrap stuff |
| 15:24 |
VanessaE |
nah it's easier than that |
| 15:25 |
VanessaE |
I'll put a comment in the PR |
| 15:26 |
paramat |
ok |
| 15:27 |
|
kilbith joined #minetest-dev |
| 15:27 |
juhdanad |
Any thoughts on #5717? |
| 15:27 |
ShadowBot |
https://github.com/minetest/minetest/issues/5717 -- Reserve `param1` for lighting |
| 15:29 |
|
kilbith joined #minetest-dev |
| 15:29 |
kilbith |
MT appears to be blurred on a HiDPI screen :( |
| 15:30 |
juhdanad |
Pixellated, or evenly blurred? |
| 15:30 |
kilbith |
both I think |
| 15:31 |
juhdanad |
Maybe you enabled bilinear filtering on the options pane? |
| 15:31 |
kilbith |
nope, fresh install |
| 15:32 |
juhdanad |
Are the edges of the nodes sharp? |
| 15:32 |
kilbith |
and the cursor ain't locked, camera keeps moving |
| 15:32 |
kilbith |
no, I'm telling you I have *not* enabled any filtering |
| 15:33 |
kilbith |
seriously, I'm not a random Joe |
| 15:33 |
juhdanad |
I mean: only the textures are blurred, or the geometry is also blurred? |
| 15:33 |
kilbith |
a screenshot will speak better |
| 15:35 |
|
slendyman joined #minetest-dev |
| 15:36 |
kilbith |
crap, even in full-maxed window, the resolution remains low |
| 15:37 |
kilbith |
this means I have to click n pixels aside a button to click it |
| 15:37 |
kilbith |
see what I mean? |
| 15:38 |
juhdanad |
Yes: the contents of the window are the same size as before, and there is an empty border between the old and new window frame. |
| 15:39 |
kilbith |
yes |
| 15:40 |
kilbith |
https://lut.im/O2fmo23KeG/nuzZNSeGa2sIBJlh.png |
| 15:40 |
juhdanad |
But then it is not blurred, but has wrong size, right? |
| 15:40 |
kilbith |
this is more crappy than this screenshot |
| 15:40 |
kilbith |
I set a window size of 1920x1024 |
| 15:41 |
kilbith |
on a 13 inches screen |
| 15:41 |
KaadmY |
Probably non power of two? |
| 15:41 |
KaadmY |
And do you have undersampling? |
| 15:42 |
kilbith |
no |
| 15:42 |
kilbith |
NPOT only applies to opengl 1.x |
| 15:42 |
juhdanad |
kilbith: The image you posted is sharp. |
| 15:42 |
kilbith |
no really it looks like crap from there |
| 15:43 |
KaadmY |
Hm? |
| 15:43 |
juhdanad |
Did you make the screenshot with f12? |
| 15:43 |
kilbith |
yes |
| 15:43 |
juhdanad |
Then it is really the negine's problem. |
| 15:43 |
juhdanad |
*engine |
| 15:44 |
kilbith |
the screenshot makes it render nice |
| 15:44 |
kilbith |
https://lut.im/H5IyAbxCsJ/NVKXszkAZ3usyx5g.png |
| 15:44 |
juhdanad |
The screenshot should be the same size as your window. |
| 15:44 |
kilbith |
here on the screenshot you don't see the aliasing on the text |
| 15:44 |
kilbith |
no, it assures you it looks different in game |
| 15:45 |
kilbith |
err, s/it/I |
| 15:45 |
juhdanad |
For some reason Minetest renders to a 1280x666 pixel large canvas, then stretches it to your screen. |
| 15:46 |
kilbith |
do you have a HiDPI screen anyway? |
| 15:46 |
juhdanad |
No, mine is smaller. |
| 15:47 |
kilbith |
ubuntu is (partly) fucked on a such screen as well |
| 15:49 |
juhdanad |
What if you resize your window to be smaller? |
| 15:50 |
kilbith |
the issues remain |
| 15:51 |
juhdanad |
And does the screenshot become smaller? |
| 15:52 |
|
Darcidride joined #minetest-dev |
| 15:54 |
kilbith |
I've managed to take a screenshot with Win10 |
| 15:54 |
kilbith |
https://lut.im/4um98iUwns/tQ0DddzKVSsLb0q6.png |
| 15:54 |
kilbith |
see the text is fucked |
| 15:54 |
kilbith |
images are aliased |
| 15:56 |
kilbith |
conclusion: MT doesn't fit to the modern anymore |
| 15:56 |
kilbith |
*modern world |
| 15:57 |
juhdanad |
Still, could you make a screenshot with F12 when your window is small? |
| 15:57 |
kilbith |
no, F12 cheats with the actual rendering I'm seeing |
| 15:57 |
kilbith |
while the screenshot tool from Win10 doesn't |
| 15:58 |
kilbith |
and on a small window the issues persist |
| 15:58 |
juhdanad |
I just would like to know if the size of the screenshot depends on the window size. |
| 15:58 |
juhdanad |
And if yes, how. |
| 15:59 |
kilbith |
alright |
| 16:02 |
kilbith |
well apparently not |
| 16:02 |
kilbith |
window size of 800x600: https://lut.im/dfO0apUoZA/bWEBdFgnuSh85SzJ.png |
| 16:03 |
juhdanad |
But your first screenshot (1280x666) was smaller than your window size, right? |
| 16:04 |
kilbith |
still fuscked up: https://lut.im/QZvMPRVdko/7Z9vdaoqIvnD11hV.png |
| 16:04 |
kilbith |
juhdanad, yes |
| 16:04 |
kilbith |
no idea why |
| 16:04 |
juhdanad |
I'm looking at the code... |
| 16:05 |
kilbith |
be back later, will attempt a Fedora install |
| 16:07 |
VanessaE |
if text is blurry, wouldn't that suggest a problem with the desktop environment or font rendering outside of minetest? |
| 16:07 |
|
Foz joined #minetest-dev |
| 16:07 |
juhdanad |
I don't think, as the screenshot size is not matched to the window size. |
| 16:08 |
VanessaE |
fair enough (this is why I use printscreen, then I can be 100% sure that what I share is what I actually see) |
| 16:09 |
|
red-001 joined #minetest-dev |
| 16:10 |
juhdanad |
This mismatch indicares a bug in the rendering engine. |
| 16:10 |
juhdanad |
*indicates |
| 16:12 |
|
kilbith joined #minetest-dev |
| 16:12 |
juhdanad |
Half-done inventory overlays: http://imgur.com/a/QOZBv |
| 16:12 |
VanessaE |
nice |
| 16:12 |
kilbith |
VanessaE: everything is fucked, not only the text |
| 16:13 |
VanessaE |
kilbith: can you take a whole screenshot? minetest window at like 90% of the screen size or something, with desktop visible behind it? |
| 16:13 |
VanessaE |
i.e. alt-printscreen or whatever the key is |
| 16:13 |
kilbith |
even gtk2 windows are fucked |
| 16:14 |
kilbith |
VanessaE: why would you want to see the desktop |
| 16:14 |
juhdanad |
kilbith: maybe you could try an other renderer? |
| 16:15 |
VanessaE |
I actually don't care about the desktop, but doing so gives us a base for comparison |
| 16:15 |
kilbith |
well opengl is the only viable option |
| 16:15 |
VanessaE |
because we can also see the window decorations and so on |
| 16:16 |
VanessaE |
I mean if gtk rendering is similarly messed up, I'm having a hard time blaming minetest :) |
| 16:16 |
kilbith |
gtk **2** doesn't support HiDPI |
| 16:17 |
kilbith |
free softwares badly support newer hardware |
| 16:17 |
VanessaE |
when I last checked, it was up to your OS and its graphics rendering to handle that |
| 16:17 |
kilbith |
no, it also depends on the GUI framework |
| 16:18 |
VanessaE |
minetest you can easily set the GUI scale to 2x or something, if your screen is so high-def, but I use 1600x1200 screens here with zero rendering problems, so unless you're going far higher, well... |
| 16:18 |
kilbith |
only gtk3 and qt5 support hidpi |
| 16:19 |
juhdanad |
In your minetest.conf, is screen_w and screen_h correct? |
| 16:19 |
kilbith |
it's proportionally a 4K screen on a 15.6 screen |
| 16:20 |
kilbith |
juhdanad: yes, set to 1920x1024 currently |
| 16:21 |
kilbith |
fullscreen is also fucked |
| 16:21 |
|
lisac joined #minetest-dev |
| 16:23 |
|
proller joined #minetest-dev |
| 16:24 |
|
kilbith joined #minetest-dev |
| 16:25 |
KaadmY |
Is completely dropping the CPU lighting pipeline an option at all? |
| 16:25 |
KaadmY |
In favor of GPU lighting and shading? |
| 16:26 |
KaadmY |
Most people are CPU bottlenecked now anyway |
| 16:26 |
KaadmY |
And shaders shouldn't be a performance issue |
| 16:26 |
KaadmY |
Even on mobile devices, shaders are probably faster than mass CPU calculations |
| 16:27 |
juhdanad |
KaadmY: the problem is that in Minetest you can easily place 10000 light sources. |
| 16:27 |
juhdanad |
Traditional lighting does not help there. |
| 16:27 |
red-045 |
depends on what you mean by completely |
| 16:27 |
red-045 |
even if we use traditinal lighting |
| 16:28 |
KaadmY |
red-045: shaders always-on, all lighting on the GPU |
| 16:28 |
KaadmY |
Minus torches and light sources maybe |
| 16:28 |
red-045 |
we still need to calculate lighting vaules |
| 16:28 |
VanessaE |
juhdanad, KaadmY there's also the problem of being able to read the lighting |
| 16:28 |
KaadmY |
Since passing a crapton of lights to the GPU is slow |
| 16:28 |
KaadmY |
red-045: yes, that still |
| 16:28 |
VanessaE |
i.e. mods need to read param1 sometimes, or get_light() or whatever the call is |
| 16:28 |
KaadmY |
But meshgen only happens when the mesh changes, ie. dig or place stuff |
| 16:28 |
KaadmY |
You'd still need to propogate lighting CPU-side for mods |
| 16:29 |
red-045 |
and if you add nice directional lighting and all which I assume is why you want this |
| 16:29 |
KaadmY |
Yes |
| 16:29 |
KaadmY |
t |
| 16:29 |
red-045 |
then the ligthing the user sees will not be consistent with what mods see |
| 16:29 |
KaadmY |
Just curious, because most people are CPU bottlenecked but we still do almost everything CPU-side even though we can almost do everything GPU-side already |
| 16:30 |
KaadmY |
red-045: light falloff can be very custom |
| 16:30 |
VanessaE |
the problem is servers don't have GPUs. |
| 16:30 |
juhdanad |
But if you remove torches and light sources, only the sun can cast directional light. |
| 16:30 |
KaadmY |
VanessaE: I think I've said like twice that you'd still need to calculate light propagation on the CPU |
| 16:30 |
KaadmY |
:P |
| 16:30 |
juhdanad |
(this is what MC shader packs implement, if I remember well) |
| 16:30 |
VanessaE |
KaadmY: so this makes three times :P |
| 16:30 |
KaadmY |
Yep :D |
| 16:31 |
juhdanad |
kilbith: I have no idea why Minetest is blurred, sorry. |
| 16:31 |
red-001 |
http://irc.minetest.ru/minetest-dev/2016-01-23 |
| 16:32 |
kilbith |
well, I'm rather sorry for you to play with this shit |
| 16:32 |
juhdanad |
For me it runs well. |
| 16:32 |
paramat |
there has been a huge amount of talk about hardware lighting in this channel in the past, especially RBA and hmmmm, it's very non-trivial. RBA's idea was a mix of CPU / light level and GPU |
| 16:33 |
kilbith |
if I wasn't a professional, MT would have been only a souvenir |
| 16:34 |
KaadmY |
Well I've been trying to implement this since yesterday |
| 16:34 |
KaadmY |
And every time I'm blocked by Minetest's codebase |
| 16:34 |
KaadmY |
This is the 3rd method I've been trying to get around problems with state |
| 16:35 |
KaadmY |
BTW couldn't you use the vertex color to pass the light level to the shaders? |
| 16:35 |
KaadmY |
Since they're not used for actual coloring anymore |
| 16:35 |
juhdanad |
Vertex colors do pass the light level to the shaders, in a compressed format. |
| 16:36 |
KaadmY |
As alpha? |
| 16:36 |
KaadmY |
The main problem I'm having right now is disabling CPU shading if shaders are enabled |
| 16:37 |
KaadmY |
But there's no global state I can use to save shader state without g_settings, which is too inefficient and a bad idea anyway |
| 16:38 |
red-045 |
which file are you working in? |
| 16:38 |
KaadmY |
mesh.cpp, content_mapblock.cpp, mapblock_mesh.cpp, game.cpp, nodes_shader/opengl_vertex.glsl |
| 16:39 |
KaadmY |
And content_mapblock.h, mapblock_mesh.h |
| 16:39 |
red-045 |
why does it need to be global? |
| 16:39 |
KaadmY |
Because the lighting functions are global for some reason and no shader context is passed in |
| 16:39 |
KaadmY |
In mesh.cpp:43 |
| 16:40 |
KaadmY |
Currently even if shaders are on, shading is CPU-side and baked into the vertex colors |
| 16:40 |
juhdanad |
Kaadmy: if you enable shaders, light is calculated on the GPU. |
| 16:41 |
KaadmY |
juhdanad: nope |
| 16:41 |
KaadmY |
Try commenting out that function's contents |
| 16:41 |
KaadmY |
And enable shaders |
| 16:42 |
red-001 |
GameGlobalShaderConstantSetter? |
| 16:43 |
juhdanad |
You misunderstood that function! That is for making certain faces of nodes slightly darker, it has nothing to do with shaders! |
| 16:43 |
KaadmY |
juhdanad: but that should be GPU side, but it's always CPU side anyway |
| 16:43 |
KaadmY |
All the GPU currently does in terms of lighting is day/night lighting |
| 16:44 |
juhdanad |
Yes, this is the only thing that is dynamic in a map block. The other things are static. |
| 16:46 |
juhdanad |
The vertex format is: alpha=day light/all light; r, g, b=(day light+night light)/2 |
| 16:53 |
KaadmY |
Is it too messy to have a global function pointer that I swap out for another function if shaders are enabled? |
| 16:54 |
KaadmY |
That would probably be by far the fastest solution |
| 16:54 |
KaadmY |
But also one of the messiest |
| 16:56 |
juhdanad |
I suggest you to do it as you can, then when you make a PR, devs will request some changes anyways. |
| 17:05 |
|
Krock joined #minetest-dev |
| 17:22 |
|
kilbith joined #minetest-dev |
| 17:33 |
KaadmY |
Yay using a global function pointer swap works beautifully |
| 17:33 |
KaadmY |
It doesn't work if you change it ingame though, dunno if that's good or bad |
| 17:34 |
juhdanad |
I'm happy to read that. |
| 17:34 |
KaadmY |
Yeah shaders don't update normally either if you change them while ingame |
| 17:44 |
paramat |
#5748 is updated and fixed, can anyone review? |
| 17:44 |
ShadowBot |
https://github.com/minetest/minetest/issues/5748 -- Underwater plantlike drawtype by numberZero |
| 17:45 |
juhdanad |
paramat: I have rebased #4421, thanks for your support! However, I don't see why you added the two approvals label. |
| 17:45 |
ShadowBot |
https://github.com/minetest/minetest/issues/4421 -- Expose getPointedThing() as Raycast by juhdanad |
| 17:48 |
|
lisac joined #minetest-dev |
| 17:54 |
paramat |
erm .. |
| 17:55 |
|
lisac joined #minetest-dev |
| 17:55 |
paramat |
zeno and smalljoker +1 |
| 17:56 |
juhdanad |
Yes, really, I did not see Zeno's approval. |
| 17:56 |
paramat |
nerzhul said it doesn't have to wait for the CSM stuff so it can be merged |
| 17:56 |
juhdanad |
That is waiting for a long time to be merged... |
| 17:57 |
paramat |
i'll merge it tonight |
| 17:57 |
juhdanad |
It is hard to create a client-sided object reference system, and would make the PR more complicated. |
| 17:57 |
paramat |
yeah |
| 17:59 |
paramat |
sfan5 #5533 is updated, could you check it? it seems you might (or might not) have issues with it |
| 17:59 |
ShadowBot |
https://github.com/minetest/minetest/issues/5533 -- Sneak: Strip down version by SmallJoker |
| 18:07 |
|
Natechip joined #minetest-dev |
| 18:08 |
paramat |
i'll merge #6103 #4421 within 2 hours |
| 18:08 |
ShadowBot |
https://github.com/minetest/minetest/issues/6103 -- Mapgen Carpathian: Add lava_depth parameter by vlapsley |
| 18:08 |
ShadowBot |
https://github.com/minetest/minetest/issues/4421 -- Expose getPointedThing() as Raycast by juhdanad |
| 18:13 |
KaadmY |
Who made the carpathian mapgen? |
| 18:13 |
KaadmY |
I haven't heard of it until it got merged a few days ago |
| 18:15 |
|
bit_lySLH2uSZHed joined #minetest-dev |
| 18:16 |
|
bit_lySLH2uSZHed left #minetest-dev |
| 18:21 |
* red-001 |
was hoping deepbook was going to be book` |
| 18:22 |
|
ThomasMonroe joined #minetest-dev |
| 18:22 |
|
ThomasMonroe joined #minetest-dev |
| 18:24 |
|
nerzhul joined #minetest-dev |
| 18:24 |
|
ThomasMonroe joined #minetest-dev |
| 18:24 |
|
ThomasMonroe joined #minetest-dev |
| 18:27 |
|
numzero joined #minetest-dev |
| 18:29 |
|
Tmanyo joined #minetest-dev |
| 18:33 |
numzero |
does anyone work on RenderingEngine now? |
| 18:34 |
numzero |
nerzhul? I remember you refactored that recently. |
| 18:36 |
paramat |
the new mapgen is by vlapsley |
| 18:37 |
KaadmY |
Never heard of them |
| 18:37 |
|
Player_2 joined #minetest-dev |
| 18:47 |
|
Natechip joined #minetest-dev |
| 18:47 |
Noxarivis[m] |
Why does the worldgen make dungeons? |
| 18:52 |
paramat |
c55 coded them into early MT |
| 18:53 |
juhdanad |
The real question is: why are dungeons empty? |
| 18:54 |
nerzhul |
numzero, yes ofc |
| 18:54 |
nerzhul |
juhdanad, dungeons should be improved to add more fun, add mines can be nice too |
| 18:55 |
juhdanad |
Exploding mines in dungeons? Cool! |
| 18:55 |
KaadmY |
Dungeons should be dangerous but rewarding |
| 18:55 |
KaadmY |
Hm good idea juhdanad :P |
| 18:55 |
juhdanad |
(I was joking) |
| 18:55 |
KaadmY |
Having random trip mines in dungeons :P |
| 18:56 |
juhdanad |
I think nerzhul meant long shafts with ores by 'mines' |
| 18:56 |
KaadmY |
Basically long shafts? |
| 18:57 |
juhdanad |
Yes. |
| 19:02 |
paramat |
as far as i know early MT had mobs by default that would appear in dungeons, maybe dungeon loot too (not sure) |
| 19:05 |
paramat |
but now mobs and loot are optional mods |
| 19:09 |
KaadmY |
Oerkki and Dungeon Masters |
| 19:10 |
juhdanad |
VanessaE, are you there? |
| 19:13 |
paramat |
Krock please could you review/approve game#1820 ? code is good and is tested. ideally i'd like to merge it later tonight with game#1818 |
| 19:13 |
ShadowBot |
https://github.com/minetest/minetest_game/issues/1820 -- Add corner stair nodes 4 by TumeniNodes |
| 19:13 |
ShadowBot |
https://github.com/minetest/minetest_game/issues/1818 -- Saplings: Reduce grow time to ABM equivalent by paramat |
| 19:13 |
paramat |
once merged i plan to add a dedicated placement function for the inner/outer stairs |
| 19:17 |
numzero |
nerzhul, so you continue cleaning RenderingEngine, right? I won’t touch it now then. |
| 19:17 |
|
DI3HARD139 joined #minetest-dev |
| 19:28 |
|
proller joined #minetest-dev |
| 19:35 |
|
proller joined #minetest-dev |
| 19:35 |
|
kilbith joined #minetest-dev |
| 19:36 |
sfan5 |
paramat: loot should not be an optional mod tho |
| 19:36 |
sfan5 |
minetest_game is too boring |
| 19:36 |
red-001 |
& |
| 19:36 |
red-001 |
^ |
| 19:37 |
KaadmY |
^ |
| 19:37 |
KaadmY |
^ |
| 19:37 |
KaadmY |
^ |
| 19:38 |
paramat |
dungeon loot would be good, just a case of someone writing a PR (i'm busy enough) |
| 19:41 |
paramat |
i think there's a mod in the forum, maybe by blockmen |
| 19:44 |
kilbith |
so MT w/ a HiDPI screen on linux looks barely fine with a lot of tweaks |
| 19:45 |
kilbith |
the HUD images are not scaled though |
| 19:45 |
|
Hunterz joined #minetest-dev |
| 19:48 |
|
proller joined #minetest-dev |
| 19:52 |
nerzhul |
numzero, what do you want to do ? |
| 19:52 |
nerzhul |
numzero, i will continue some PR to move things to the RenderingEngine singleton class yes, i don't think this evening |
| 19:52 |
numzero |
some refactoring |
| 19:52 |
numzero |
and postprocessing, but refactoring seems to be more important now |
| 19:52 |
nerzhul |
juhdanad, i mean gold mines :D |
| 19:52 |
nerzhul |
not exploding mines :p |
| 19:53 |
numzero |
:D |
| 19:53 |
numzero |
nerzhul, which PR? |
| 19:55 |
nerzhul |
no pr atm |
| 19:55 |
numzero |
nvm |
| 19:56 |
nerzhul |
numzero, for postprocessing what do you want to have in RenderingEngine ? |
| 19:56 |
nerzhul |
Device and Driver are mainly in RenderingEngine singleton atm |
| 19:57 |
nerzhul |
only Game have the real pointer too for Device, for Driver some other class but this can be moved in future pr. I will focus on guiEnv and master scene in future pr |
| 19:59 |
juhdanad |
nerzhul: there are still some files which do not use RenderingEngine, for example wieldmesh.cpp |
| 20:00 |
juhdanad |
(I just didn't know if you were aware of that) |
| 20:00 |
nerzhul |
juhdanad, yes wieldmesh has a specific thing |
| 20:00 |
juhdanad |
A specific thing? |
| 20:01 |
nerzhul |
i can't remember what but it was normal if i didn't do it atm because i depend on another thing if i remember |
| 20:01 |
nerzhul |
(on week ago i'm lazy :p) |
| 20:01 |
numzero |
er, want to have? I will need Driver, shader access (don’t remember which class gives that), and will need to keep some textures in the RenederingEngine (similar to current undersampling or stereo rendering, but want it to be cleaner) |
| 20:02 |
nerzhul |
ShaderSrc |
| 20:02 |
nerzhul |
which can be a singleton too attached to Rendering Engine |
| 20:02 |
numzero |
but the main concern is that adding postprocessing changes code path significantly |
| 20:02 |
numzero |
so conflicts with refactoring |
| 20:02 |
nerzhul |
texturesrc can be a singleton too (cf celeron55) |
| 20:03 |
nerzhul |
where is your part ? |
| 20:06 |
numzero |
depends |
| 20:07 |
numzero |
I want it to be supported for all modes (to move some code to postprocessing), but before refactoring that means writing everything here and there, in all these draw_* methods |
| 20:07 |
nerzhul |
if you want to refactor RenderingEngine::draw_* you are free :) |
| 20:08 |
numzero |
as I have other things to do, I’d prefer refactoring to be done first |
| 20:08 |
numzero |
er... |
| 20:08 |
numzero |
oh, okay |
| 20:09 |
numzero |
will do that (not now, I’m lazy too =) |
| 20:10 |
numzero |
one question: won’t you mind if 3d modes will require shaders? |
| 20:15 |
|
Taoki joined #minetest-dev |
| 20:23 |
|
proller joined #minetest-dev |
| 20:24 |
|
Darcidride joined #minetest-dev |
| 20:24 |
|
proller joined #minetest-dev |
| 20:26 |
|
proller joined #minetest-dev |
| 20:28 |
KaadmY |
In a shader, how do you figure out a vertex's position relative to the camera? |
| 20:29 |
juhdanad |
There is a global vector which tells the camera position. |
| 20:29 |
|
Fixer joined #minetest-dev |
| 20:29 |
KaadmY |
Which? |
| 20:31 |
|
proller joined #minetest-dev |
| 20:32 |
juhdanad |
I think m_eye_position_vertex |
| 20:33 |
KaadmY |
In the shader? |
| 20:33 |
juhdanad |
No, sorry, it is eyePosition. |
| 20:34 |
Hijiri |
#6108 what does "real overlay support" mean? Does this mean that using overlay texture modifiers doesn't make a new texture for the combined texture? |
| 20:34 |
ShadowBot |
https://github.com/minetest/minetest/issues/6108 -- Make crack an overlay |
| 20:34 |
KaadmY |
Ah |
| 20:34 |
Hijiri |
Or is there a special setting for node overlays now |
| 20:34 |
juhdanad |
In minetest/client/shaders/nodes_shader/vertex.glsl you can find an example. |
| 20:34 |
KaadmY |
opengl_vertex.glsl* |
| 20:34 |
juhdanad |
Hijiri: what kind of setting? |
| 20:34 |
Hijiri |
I don't know |
| 20:35 |
Hijiri |
Is the issue referencing the stuff described in "Soft texture overlay" in the docs? |
| 20:36 |
|
proller joined #minetest-dev |
| 20:36 |
Hijiri |
I think it is, nevermind |
| 20:37 |
KaadmY |
Gah, no matter what I'm doing the specular highlight is always in screen space |
| 20:38 |
juhdanad |
Screen space? |
| 20:39 |
Hijiri |
will mods be able to register nodes with different specular coefficients? |
| 20:39 |
KaadmY |
Well specular has to work first |
| 20:39 |
KaadmY |
For now I'll just hardcode specular to a sane value |
| 20:39 |
KaadmY |
Just to add a bit of depth to geometry |
| 20:42 |
|
proller joined #minetest-dev |
| 20:46 |
|
proller joined #minetest-dev |
| 20:51 |
|
proller joined #minetest-dev |
| 20:59 |
|
jcalve joined #minetest-dev |
| 21:02 |
VanessaE |
juhdanad: what's up? |
| 21:03 |
|
proller joined #minetest-dev |
| 21:03 |
|
paramat joined #minetest-dev |
| 21:04 |
juhdanad |
Is game#1784 still valid? |
| 21:04 |
ShadowBot |
https://github.com/minetest/minetest_game/issues/1784 -- Crash problem in beds mod, bad `param2` values |
| 21:05 |
|
Gael-de-Sailly joined #minetest-dev |
| 21:06 |
VanessaE |
looking at mt_game commits, nothing jumps out at me as something that would have fixed it |
| 21:06 |
VanessaE |
I assume it's still valid yeah |
| 21:08 |
|
RobbieF left #minetest-dev |
| 21:08 |
|
RobbieF joined #minetest-dev |
| 21:10 |
|
proller joined #minetest-dev |
| 21:10 |
|
kilbith joined #minetest-dev |
| 21:13 |
|
kilbith joined #minetest-dev |
| 21:13 |
paramat |
yes game1784 is not fixed yet |
| 21:18 |
|
proller joined #minetest-dev |
| 21:21 |
|
proller joined #minetest-dev |
| 21:24 |
paramat |
will merge #4421 #6103 in a moment |
| 21:24 |
ShadowBot |
https://github.com/minetest/minetest/issues/4421 -- Expose getPointedThing() as Raycast by juhdanad |
| 21:24 |
ShadowBot |
https://github.com/minetest/minetest/issues/6103 -- Mapgen Carpathian: Add lava_depth parameter by vlapsley |
| 21:27 |
|
proller joined #minetest-dev |
| 21:28 |
kilbith |
http://irc.minetest.ru/minetest-dev/2017-06-29#i_4993629 |
| 21:28 |
kilbith |
^ this question has not been answered yet |
| 21:30 |
|
MoNTE48 joined #minetest-dev |
| 21:34 |
paramat |
merging |
| 21:39 |
|
proller joined #minetest-dev |
| 21:39 |
sfan5 |
paramat: i'll try to look at stripped down sneak soon, however i'll be on holiday |
| 21:40 |
sfan5 |
also a question |
| 21:40 |
paramat |
merged |
| 21:40 |
paramat |
ok |
| 21:40 |
sfan5 |
do you think it would be beneficial to be able to restrict csm on a per-player basis? |
| 21:41 |
nerzhul |
sfan5, i think it's overkill to have a such feature. Many server side features are globally limited |
| 21:41 |
sfan5 |
well dunno |
| 21:41 |
sfan5 |
a server admin might want to use oredetect |
| 21:41 |
|
proller joined #minetest-dev |
| 21:43 |
paramat |
doesn't seem important to have |
| 21:44 |
sfan5 |
i don't see any downside |
| 21:46 |
|
bigfoot547 joined #minetest-dev |
| 21:48 |
paramat |
maybe no downside, but i think the work and code needs justification, and i don't see much yet |
| 21:50 |
paramat |
a server admin could use a serverside oredetect mod |
| 21:50 |
sfan5 |
true |
| 21:53 |
|
Megaf joined #minetest-dev |
| 21:54 |
|
Megaf_ joined #minetest-dev |
| 21:54 |
nerzhul |
flavour require more directions to be finished & merged |
| 21:54 |
red-001 |
can always be implemented latter if there is a demand for it |
| 21:55 |
|
ThomasMonroe joined #minetest-dev |
| 21:55 |
|
ThomasMonroe joined #minetest-dev |
| 22:33 |
|
Arthricyn1 joined #minetest-dev |
| 22:36 |
|
Arthricyn joined #minetest-dev |
| 22:40 |
|
Arthricyn joined #minetest-dev |
| 22:41 |
|
Arthricyn joined #minetest-dev |
| 22:59 |
|
Arthricyn joined #minetest-dev |
| 23:05 |
|
Arthricyn1 joined #minetest-dev |
| 23:09 |
|
proller joined #minetest-dev |
| 23:24 |
|
Player-2 joined #minetest-dev |
| 23:30 |
|
Arthricyn joined #minetest-dev |
| 23:33 |
|
Arthricyn1 joined #minetest-dev |
| 23:47 |
|
Arthricyn joined #minetest-dev |
| 23:50 |
|
Arthricyn1 joined #minetest-dev |
| 23:52 |
|
Taoki joined #minetest-dev |
| 23:53 |
|
Foz joined #minetest-dev |
