Time Nick Message 04:06 MTDiscord anyone know if the builtin locale files & template are up to date? 04:06 MTDiscord if the strings are freeze'd then might want to run the translation script for builtin so translators can start translating builtin for next release 04:06 MTDiscord (looking at repo the template is like 8 months old) 04:06 MTDiscord if the strings are not freeze'd yet, please remember to run the translation script against builtin when nearing the day of release to give time for translators making PR for it 05:00 srifqi last time I checked it (a few days ago), there are no new strings in builtin 05:02 srifqi ah, the Discord bridge just disconnected right before I sent it 05:02 srifqi last time I checked it (a few days ago), there are no new strings in builtin 05:04 srifqi running `python util\mod_translation_updater.py builtin` does not add nor remove any string 05:04 srifqi the order is different, though 05:05 srifqi yup, just run the script and there are no new strings 05:15 MTDiscord ah, so it's just some kind of shuffling around of the existing strings 05:28 srifqi I wonder whether we should just use the ordering from the script (when a new string is added) 05:30 srifqi since the script orders it based on string location and file name 05:38 MTDiscord I was wondering if there was any support/excitement for https://github.com/minetest/minetest/issues/14040 because I know exactly what I need to do and how I need to do it. I am checking this information before I start working on the PR 05:42 Mantar not sure there's a big demand; I think most people are waiting for gltf to get here, or they have b3d tooling set up already 05:43 Mantar but easier tooling would answer one of the major complaints about b3d 05:46 Mantar The other major complaint I've heard is its inefficiency. Apparently file size grows too quickly with increasing poly count, much more than other formats. That might be something to do with the binary exporters, and if so, JSON could fix that too 05:50 MTDiscord I think this has to do with the fact that b3d is embedding a tree list with C strings and byte counts within it 05:51 MTDiscord If you load up a b3d file in vscode, you can actually see the cstrings everywhere 05:51 Mantar ah 05:52 Mantar interesting 05:52 MTDiscord I think I'm just going to be dumb and make the PR anyways 05:54 Mantar yeah, it might help somebody out there get modelling 07:42 MTDiscord If at least one person's day has been improved by the modeling format, I will have done my job correctly 08:27 Mantar that's a good way to think about it! ^_^ b 11:46 MTDiscord Mantar: I think that might be due to the exporter writing weights for every vertex for every bone. 11:46 MTDiscord (the blender exporter, that is) 11:47 MTDiscord you can often achieve significant space savings just by re-encoding with modlib, since it strips zero weights 14:21 Krock Meeting in ~40 minutes 15:00 srifqi hello 15:00 Krock Meeting ping for those who reacted. sfan5 Desour srifqi 15:01 Krock hello srifqi :) 15:02 srifqi :D 15:02 Krock First topic: https://github.com/minetest/minetest/milestone/22 15:02 Krock !topic 15:02 ShadowBot Krock: FEATURE FREEZE IN EFFECT | Minetest core development and maintenance. Chit-chat goes to #minetest. https://dev.minetest.net/ https://irc.minetest.net/ https://github.com/minetest 15:02 Krock !title 15:02 ShadowBot 5.8.0 Milestone · GitHub 15:03 Krock #13961 is slightly annoying but not a blocker 15:03 ShadowBot https://github.com/minetest/minetest/issues/13961 -- Strange behavior when dragging&dropping from MTG creative inventory (since #13146) 15:03 srifqi issue 13822 already has fixes: PR 13872 and even PR 14029 15:03 Krock so far only #13872 and the credits are open tasks 15:03 ShadowBot https://github.com/minetest/minetest/issues/13872 -- Formspec: Pass the second-touch event as is by srifqi 15:04 srifqi i don't know enough to have opinion about issue 14007 15:04 Krock I tried to find a difference between that issue and the PR that is supposed to fix it 15:05 Krock apparently I did not do it properly because there was no difference 15:06 srifqi ... and i don't have device with older OpenGL version 15:07 Krock it would be possible to use OpenGL's environment variables to force a version by software 15:07 Krock although that does not emulate whatever bugs there are with old hardware 15:08 Krock I see. Desour also couldn't reproduce on the (presumably) AMDGPU driver, just like me 15:08 Desour if you use the mesa software renderer, (or zink, I think), you can get the issue that anisotropic filter enables bilinear filter 15:08 Desour I could not yet reproduce the other fixed issue(s) 15:11 srifqi which issue? 15:11 Krock it would also be interesting to know whether the "LGTM" in #14016 are actually tested or just a "code LGTM, ship it." 15:11 ShadowBot https://github.com/minetest/minetest/issues/14016 -- [no squash] Return texture filter settings to previous state by sfan5 15:12 Krock srifqi: #14007 there are a few filtering issues, apparently. 15:12 ShadowBot https://github.com/minetest/minetest/issues/14007 -- Texture filtering regressions (biilinear and trilinear filtering broken, anisotropic filtering results inconsistent depending on GPU) 15:12 srifqi oh, i thought it was another issue 15:14 Desour one other issue is that for Warr1024 apparently the bilinear minification doesn't work anymore 15:15 Krock what does "minification" mean in this context? 15:15 Desour the 3rd issue is, I think, just that some users liked the smooth look 15:16 Desour Krock: when the texel (pixel of the texture) is smaller than a screen pixel, minification happens 15:16 Desour aka downscaling 15:16 Krock alright. I just saw the relevant comment 15:19 Krock there should be a difference in the wielded item rendering depending on the filter settings 15:21 Krock lol I'm not sure whether this guy is serious or not: https://github.com/minetest/minetest/issues/13822#issuecomment-1813864090 15:22 Krock haha anyway. the remaining issues should be pretty close to be fixed 15:22 srifqi i mean, that is still a valid workaround :D 15:23 Krock sure they're technically not wrong 15:24 srifqi by the way, why is PR #13872 in 5.9.0 milestone while the issue that will be fixed by it is in 5.8.0? 15:24 ShadowBot https://github.com/minetest/minetest/issues/13872 -- Formspec: Pass the second-touch event as is by srifqi 15:24 Krock srifqi: do you have an idea why #13872 is marked as 5.9.0 milestone but its issue got the 5.8.0 milestone? something must've gone wrong there 15:24 ShadowBot https://github.com/minetest/minetest/issues/13872 -- Formspec: Pass the second-touch event as is by srifqi 15:24 Desour haha 15:24 srifqi that is also my question 15:24 Krock good timing. I think it should be set to 5.8.0 15:24 Krock I don't know why rubenwardy changed it 15:25 Krock objections? 15:25 rubenwardy I don't remember changing that, maybe it was accidental 15:25 Krock alright :) so 5.8.0 it is. 15:26 Krock this is somewhat important for 5.8.0 to not ruin the Android players' experience 15:26 Krock also rubenwardy while you're here .. would you like to review/test the PR in the next few days? 15:27 Krock assuming the texture filters PR also makes it way through in that time span we could finally head towards a release 15:30 Krock is there any other PR that should be discussed? 15:30 Krock bugfixes, maintenance, ...? 15:33 srifqi none from me 15:37 Krock alright then. thank you for participating. I'll head over to propose a fix for #14044 15:37 ShadowBot https://github.com/minetest/minetest/issues/14044 -- ParticleSpawner (and maybe sound) IDs Reused 15:38 srifqi thank you for organising 15:39 Krock np 15:40 srifqi oh, i just remembered. will the next meeting about releasing 5.8.0? 15:40 srifqi ... or that depends on the milestone (which already past 7 days)? 15:41 Desour iirc someone wanted that we release before november, because of the game jam 15:41 srifqi ... and it's already November 15:41 Desour december* 15:41 srifqi ah, i see 15:42 srifqi so ... as soon as the milestone is achieved? 15:42 Desour Krock: FYI, for server sound handle reuse, there was already #12780 (not merged) 15:42 ShadowBot https://github.com/minetest/minetest/issues/12780 -- Put server sound handles into userdata by Desour 15:42 Krock srifqi: hopefully the release is before the nex tmeeting 15:42 Krock depends on the milestone progress. 15:43 Krock Desour: sound IDs were recycled too? 15:43 Desour (someone was ruben: ) 15:44 * srifqi is looking forward to the release 15:44 Desour Krock: they still are, but only after the counter would overflow. so it wasn't considered worth fixing, iirc 15:45 Krock well okay that's not too much of a concern 15:45 Krock unless those IDs are not skipped and forcefully reused 15:45 Krock in that case it should be fixed 15:48 Krock Server::nextSoundId() 15:48 Krock they are forcefully reused, which is pretty bad 16:03 rubenwardy Krock: if it's ready then don't wait for me 16:04 Krock rubenwardy: okay thank you for clarifying. It was mainly a question because you requested a review. 17:32 MTDiscord I, too, hope for a release in November for the sake of the jam, if at all possible :] 17:53 sfan5 do we have a changelog yet? 17:54 MTDiscord start of one: https://dev.minetest.net/Changelog#5.7.0_.E2.86.92_5.8.0 18:13 MTDiscord Krock: I think the proper solution would be to autoincrement IDs, keeping the ID -> spawner mapping in an unordered map. Then the logic in nextSoundId becomes trivially. Overflowing is not a problem if you use a 64-bit type. On the Lua side of things, ints up to 2^53 will be accurate. Both of these numbers are high enough that I can guarantee you that it is not physically possible to hit them. 18:14 Krock @luatic I already opened a PR. feel free to comment.#14045 18:14 ShadowBot https://github.com/minetest/minetest/issues/14045 -- Server: avoid re-use of recent ParticleSpawner and Sound IDs by SmallJoker 18:19 MTDiscord commented 18:21 Krock oh that was already the reply to my PR? sorry. I misunderstood the intend. 18:21 Krock it would be possible to increase the bit with, but only for newer clients due to network protocol limitations 18:21 Krock because particle spawners are not only removed on server/lua side but also client-side 18:47 MTDiscord ah I see, of course 🤦 18:48 MTDiscord deleted my comment 18:58 MTDiscord It's probably not a problem (just a little suboptimal) if deletion requests are sent to the client when things already expired client side. The client needs to be able to gracefully handle receiving requests to delete things "late" due to network latency anyway. 20:17 sfan5 I didn't know multicrafts mainmenu looked that good 20:22 MTDiscord what specifically looks good/the best about it to you? 20:24 sfan5 proper and consistent design, font, reasonable use of icons 20:24 MTDiscord ah, thanks