Time |
Nick |
Message |
00:42 |
|
proller joined #minetest-dev |
00:42 |
|
kilbith joined #minetest-dev |
01:20 |
|
jwmhjwmh joined #minetest-dev |
01:34 |
|
kilbith joined #minetest-dev |
03:41 |
|
YuGiOhJCJ joined #minetest-dev |
04:00 |
|
MTDiscord joined #minetest-dev |
04:43 |
|
calcul0n_ joined #minetest-dev |
05:51 |
|
Menchers joined #minetest-dev |
08:05 |
|
appguru joined #minetest-dev |
08:22 |
|
Warr10247 joined #minetest-dev |
09:04 |
|
Warr1024 joined #minetest-dev |
09:18 |
|
olliy joined #minetest-dev |
10:17 |
|
Fixer joined #minetest-dev |
11:06 |
|
Fixer_ joined #minetest-dev |
11:13 |
|
kilbith joined #minetest-dev |
11:18 |
|
kilbith joined #minetest-dev |
11:52 |
|
kilbith joined #minetest-dev |
11:53 |
|
Taoki joined #minetest-dev |
12:45 |
kilbith |
https://imgur.com/a/Zy9sFkG |
12:45 |
kilbith |
wth is this shit |
12:45 |
kilbith |
when I set undersampling to 2 |
12:46 |
MTDiscord |
<ROllerozxa> oops |
12:46 |
celeron55 |
well, it *is* undersampling. if you don't mind the other stuff |
12:49 |
kilbith |
https://i.imgur.com/6ji2jch.png |
12:49 |
kilbith |
fuck man |
12:50 |
kilbith |
MT is designed with the blueprint of the Linux desktop: fucking broken |
12:51 |
celeron55 |
there's probably been tens of PRs related to the rendering pipeline and people have probably forgotten to test undersampling in pretty much all of them |
12:51 |
MTDiscord |
<ROllerozxa> and just like the linux desktop, it works on my machine™️ |
12:51 |
sfan5 |
if shiny PRs aren't merged you complaim, if shiny PRs are merged and have occasional bugs you complain |
12:51 |
sfan5 |
so which is it? |
12:51 |
celeron55 |
anyway, just figure it out and fix it. that's the way it goes |
12:52 |
celeron55 |
the first question is, can you find an old version where it isn't broken |
12:53 |
kilbith |
okay, there's something wrong with mesh culler |
12:53 |
kilbith |
+the |
12:54 |
sfan5 |
it looks like the z buffer is inverted |
12:55 |
kilbith |
question is, is it only happening on certain GPUs? |
13:00 |
sfan5 |
just updated and it's also broken |
13:00 |
sfan5 |
the bloom PR maybe? |
13:02 |
celeron55 |
it works for me on intel, and i can't test on nvidia because prime broke once again in this version of fedora |
13:04 |
|
Desour joined #minetest-dev |
13:04 |
kilbith |
ok there's something with nvidia |
13:05 |
Desour |
^ can't reproduce (with amd gpu) |
13:05 |
kilbith |
definitely nvidia |
13:05 |
celeron55 |
(of course, i can't test using 4K resolution though) |
13:06 |
celeron55 |
(but that doesn't seem like a likely factor) |
13:06 |
kilbith |
otherwise x2048 would have spotted that bug (he uses AMD) |
13:06 |
sfan5 |
it's broken for me on amd |
13:06 |
sfan5 |
as of 22cbc058080127445b69d2137e346ad52f8b3387 |
13:06 |
sfan5 |
with either OpenGL or OpenGL ES |
13:07 |
sfan5 |
(pretty irrelevant info since I bet nobody else has a working GLES build set up) |
13:07 |
|
kilbith joined #minetest-dev |
14:23 |
MTDiscord |
<savilli> Works for me on nvidia and intel gpu |
16:22 |
|
kilbith joined #minetest-dev |
17:17 |
|
vampirefrog joined #minetest-dev |
17:40 |
|
appguru joined #minetest-dev |
17:58 |
|
appguru joined #minetest-dev |
18:18 |
|
olliy joined #minetest-dev |
18:35 |
MTDiscord |
<x2048> I use Intel and AMD on Linux with OSS drivers, use undersampling daily and check 3d modes from time to time. |
19:50 |
MTDiscord |
<x2048> sfan5: is only undersampling broken for you? is it still broken when shaders are disabled? |
20:29 |
|
kilbith joined #minetest-dev |
20:42 |
|
Fixer joined #minetest-dev |
21:55 |
schwarzwald[m] |
<sfan5> "(pretty irrelevant info since..." <- I did at one point I think, on AMD here as well. |
22:26 |
|
proller joined #minetest-dev |
22:35 |
|
panwolfram joined #minetest-dev |
22:49 |
|
proller joined #minetest-dev |
23:02 |
|
YuGiOhJCJ joined #minetest-dev |