| Time |
Nick |
Message |
| 00:43 |
|
schwarzwald[m] joined #minetest-dev |
| 04:00 |
|
MTDiscord joined #minetest-dev |
| 05:09 |
|
fluxionary joined #minetest-dev |
| 05:20 |
|
behalebabo joined #minetest-dev |
| 05:50 |
|
calcul0n_ joined #minetest-dev |
| 07:32 |
|
Flitzpiepe joined #minetest-dev |
| 08:01 |
|
calcul0n_ joined #minetest-dev |
| 09:34 |
|
calcul0n joined #minetest-dev |
| 10:21 |
|
appguru joined #minetest-dev |
| 10:21 |
|
calcul0n joined #minetest-dev |
| 10:32 |
|
fluxionary joined #minetest-dev |
| 10:51 |
|
HuguesRoss joined #minetest-dev |
| 10:52 |
Baytuch |
Good morning |
| 11:13 |
|
MTDiscord joined #minetest-dev |
| 11:36 |
|
asdflkj_sh joined #minetest-dev |
| 11:56 |
erle |
has any of you used semgrep? i used it to find dupes in mcl games, but otherwise not so far |
| 12:36 |
|
fossdev2 joined #minetest-dev |
| 13:02 |
|
proller joined #minetest-dev |
| 13:17 |
|
Fixer joined #minetest-dev |
| 15:09 |
sfan5 |
merging #12442 in 10m |
| 15:09 |
ShadowBot |
https://github.com/minetest/minetest/issues/12442 -- Add unittests for entity lifecycle and callbacks by sfan5 |
| 15:25 |
erle |
sfan5 what was the on_death issue? does it fix any reported bug? |
| 16:51 |
sfan5 |
see the first commit message |
| 17:40 |
|
Noisytoot joined #minetest-dev |
| 18:48 |
|
calcul0n joined #minetest-dev |
| 19:29 |
MTDiscord |
<x2048> Merging #12021 in a few minutes |
| 19:29 |
ShadowBot |
https://github.com/minetest/minetest/issues/12021 -- Add description of privileges by x2048 |
| 19:33 |
|
calcul0n joined #minetest-dev |
| 19:35 |
|
calcul0n joined #minetest-dev |
| 20:11 |
|
x2048 joined #minetest-dev |
| 20:13 |
x2048 |
erle: do you know why many MCL2 models have randomized normals? This is related to #12430 |
| 20:13 |
ShadowBot |
https://github.com/minetest/minetest/issues/12430 -- Minetest recalculates normals if they don't match the winding order |
| 20:22 |
erle |
x2048 no. truth be told, the provenance of the models is highly questionable, as in: there is no automated build process as far as i know. i think people just load them into blender or whatever and export them (no idea how). last person i can remember having changed something is fleckenstein, i suggest to ask him about the tooling. |
| 20:23 |
erle |
x2048 did you look at the blender files and try to export them yourself? |
| 20:23 |
x2048 |
no, I did not dig that deep :) |
| 20:24 |
x2048 |
I'll ask around on discord, just wanted to check with you first. thank you |
| 20:25 |
erle |
x2048 these are the models in question i assume https://github.com/22i/minecraft-voxel-blender-models |
| 20:26 |
erle |
oh no, the instructions are horrible |
| 20:27 |
erle |
> test animation by pressing ALT-A and export using special minetest B3D exporter |
| 20:27 |
erle |
x2048 this could be the smoking gun … or a red herring. hehe https://github.com/minetest/B3Dexport |
| 20:49 |
Pexin |
smoked herring mmm |
| 21:16 |
x2048 |
It seems this use of vertex.normal is wrong: https://github.com/minetest/B3DExport/blob/master/B3DExport.py#L1101. Blender shows vertex normals as an average of the neighboring faces |
| 21:18 |
erle |
x2048 well, if it is wrong *there* then it is wrong everywhere, not just in mcl |
| 21:19 |
x2048 |
Yes, I've caught this in many MTG, some RePixture and some Juanchi models. |
| 21:19 |
MTDiscord |
<luatic> You'd actually have to duplicate the vertices if the faces use different normals. |
| 21:19 |
MTDiscord |
<luatic> Or just omit the normals and let the engine recalculate them. |
| 21:19 |
x2048 |
They are duplicated anyway, because that loop is per-face. |
| 21:20 |
MTDiscord |
<luatic> anyways x2048, that tool is very outdated |
| 21:20 |
MTDiscord |
<luatic> so I wonder whether https://github.com/GreenXenith/io_scene_b3d has the same bugs |
| 21:21 |
x2048 |
Same line: https://github.com/GreenXenith/io_scene_b3d/blob/master/export_b3d.py#L1128 |
| 21:21 |
MTDiscord |
<luatic> oof then |
| 21:22 |
x2048 |
Well, it behaves the same, and I suspect this is the problem. I can try to change it and see if it fixes the problem. |
| 21:54 |
|
kaeza joined #minetest-dev |
| 22:03 |
|
HuguesRoss9 joined #minetest-dev |
| 22:04 |
|
sfan5_ joined #minetest-dev |
| 22:06 |
|
sofar_ joined #minetest-dev |
| 22:09 |
|
kaeza joined #minetest-dev |
| 22:25 |
|
proller joined #minetest-dev |
| 22:32 |
|
panwolfram joined #minetest-dev |
| 22:45 |
erle |
x2048 if the engine has to recalculate the normals anyway, what do you gain? loading speed if the export is fixed? |
| 23:34 |
|
proller joined #minetest-dev |
