Time Nick Message 17:48 Krock Would someone from the doc team please be so nice to merge https://github.com/luanti-org/modtools/pull/15 ? Thanks! 17:48 Krock wait I am stupid. This is not a doc repo 18:05 sfan5 planning to merge #16634 later if no more comments 18:05 ShadowBot https://github.com/luanti-org/luanti/issues/16634 -- Automatically choose multiple emerge threads for singlenode by sfan5 18:09 Krock sfan5: do you know whether this is intentional behaviour? https://krock-works.uk.to/u/rec-2025-11-13_19.08.13.mp4 18:09 Krock (the shadow intensity varies... somehow at random) 18:10 sfan5 that doesn't look intentional 18:10 sfan5 also why do your rails cast shadows 18:10 Krock opengl3 moment ig 18:10 Krock there's no rail shadows for "opengl", but the shadow intensity still varies 18:11 sfan5 it will vary if you have artifical light nearby 18:12 Krock yes indeed! In newly generated map areas, there's no such variation 18:13 sfan5 then that's a feature 18:13 sfan5 for the rails this is how it should look https://0x0.st/Kpj7.jpg 18:13 sfan5 taken with opengl3 and the last commit from #16670 18:13 ShadowBot https://github.com/luanti-org/luanti/issues/16670 -- [no squash] Clean up some GLSL code + more CI by sfan5 18:13 Krock that's the PR I'm testing right now 18:14 Krock renderdoc time ig 18:17 sfan5 be warned that the shadow map does partial updates (not fully redrawn every frame) 18:18 Krock I was about to say that I have no idea what to look for because they appeared together with the sand texture 18:19 sfan5 if you find the call where they are drawn to the shadow map then they shouldn't be solid there 18:25 Krock or I just set m_force_update_shadow_map = true to have it update in all frames 18:26 sfan5 yes 18:28 Krock rendered to ShadowMapSampler, step 213 18:30 Krock * 243, although I still have no idea what to look for... here's the renderdoc file, sfan5 - if it is of any help for you: https://n.uguu.se/GuSFBRvB.gz 18:32 sfan5 the source texture is not bound 18:32 sfan5 it should be bound to the "ColorMapSampler" 18:33 sfan5 sampling from an unbound texture apparently returns opaque black 18:34 sfan5 that however is because ColorMapSampler is the wrong type 18:35 sfan5 your shaders don't match this commit revision 18:36 Krock AAAAAAh not again. Thank you. Yes, that's really it. 18:37 Krock I had switched those back to a backup to compare an older Luanti installation. Apparently they were compatible enough to not cause errors. 18:37 Krock whereas in builtin you'd generally run into obvious errors when there's a mismatch 21:50 MTDiscord just going to nicely ask if https://github.com/luanti-org/luanti/pull/14103 could get a core dev review as its rebased now, a pretty useful thing everyone has to hack around for now, and is a few years old (not 4 yet, i knot) 22:47 cheapie Is there a plan/timeline for making opengl3 the default video driver in place of opengl? Now that it's actually worth using it seems logical to do that eventually, but at the same time I understand if there's a desire to wait until the texture array feature has had time to bake for a release or two 22:54 MTDiscord there is one blocker issue 22:54 MTDiscord https://github.com/luanti-org/luanti/issues/15524 22:55 cheapie ah 22:57 cheapie I remember seeing that one a while back, assumed it was just crappy scaling somewhere along the line and didn't think to try with opengl 22:58 MTDiscord prs welcome 😉 22:59 cheapie You say that like you expect me to understand graphics programming :P 23:08 sfan5 if someone PR'd even an AI-made solution (provided it works) that would be better than none