Time Nick Message 11:25 Krock Why does an insecure CPCSM env exist? 11:26 sfan5 the original _G doesn't disappear as long as there's a reference to it 11:26 Krock I meant - are there any use-cases for it? 11:26 sfan5 no 11:27 Krock also fair - it is needed for the secure wrappers.... 11:28 sfan5 CSM only needs access to the original debug.getinfo IIRC, but that can easily be solved differently 11:28 Krock I'll just propose a change for server mods and then adapt CSM once that concept is approved 11:30 sfan5 before that we'll need to get the fix for https://github.com/orgs/luanti-org/discussions/157 in 11:30 Krock that's what I'm currently doing, based on your feedback 11:30 sfan5 because otherwise you get changes that cannot be individually be backported 11:30 sfan5 (we want to backport the fix to a patch release obviously, but the other commits can stay for later) 11:30 sfan5 ah, ok 17:36 Guest34 Should the version scheme for openGL in the render info text always be major.minor? My NVidia driver seems to report an extra "patch" zero at the end. 17:38 whosit I have 4.6.0 on windows and 4.6 on linux... 17:39 Guest34 With a NVidia driver? 17:39 whosit intel/amd 17:41 Guest34 Hmm. Afaik openGL uses major.minor as their version scheme. 17:42 Guest34 I already got a patch to cut the extra zero, if that's wanted. 17:43 whosit I'm not an engine dev, but why cut it if driver reports it... 17:46 Guest34 Per https://wikis.khronos.org/opengl/History_of_OpenGL their last patch version was in 1998! Why? I was confused why the render info text was longer when starting Luanti with a dGPU and figured it might make sense to unify it on the client side. 17:46 * user333_ 's old laptop supports up to 1.4... 17:47 Guest34 Woah, that's old! 17:48 user333_ indeed 17:48 Guest34 I guess for "accuracy" reasons, it shouldn't be cut - as you said. I just wondered :P 17:49 user333_ integrated graphics of a 15+yo atom aren't great :p 17:49 Guest34 I thought my Intel Core 2 Duo ThinkPads are old … 17:49 user333_ at least they're x86_64 17:50 user333_ unlike mine 17:50 sfan5 what the renderer text says doesn't really matter 17:51 Guest34 So keep it as is? 18:21 [MatrxMT] is it normal that, whilst downloading dynamic media, mapblocks stop loading until it's done downloading? I've noticed entities movement and chat work as usual, it's just the map that freezes 18:39 lhofhansl Will merge #16945 in a few 18:39 ShadowBot https://github.com/luanti-org/luanti/issues/16945 -- Bring back low priority block events in a safe way. by lhofhansl 18:40 lhofhansl Also would appreciate opinions on #16569. Happy to close if there's no interest. 18:40 ShadowBot https://github.com/luanti-org/luanti/issues/16569 -- Add movement smoothing to CAOs by lhofhansl 19:10 whosit does smooting works by making things move 2x slower, or I'm misunderstanding the code? 19:11 whosit (I won't claim to understand it at all ;P) 19:12 whosit (it kinda explains snapping/teleportation of falling item...) 19:13 lhofhansl No, it is spreading out movement over more frames, but it's additive, so no information is lost (when there is a collision, it needs to be precise, and switches back to precise frames). 19:14 whosit spreading movement over frames sounds like slowing down?.. 19:20 whosit I probably need to learn more about how interpolation works... 19:23 lhofhansl Entities are not slowed down. Acceleration is spread over a few frames (which removes the jerkiness). When a collision is expected acceleration is precise again. 19:25 lhofhansl Merged #16945 19:25 ShadowBot https://github.com/luanti-org/luanti/issues/16945 -- Bring back low priority block events in a safe way. by lhofhansl