Time Nick Message 11:50 [MTMatrix] Reminder for the staff to vote for the new name of Minetest 11:51 ROllerozxa the node mcnodeface engine 15:30 MTDiscord I think the tag "Rebase needed" should be removed in my Dual wielding PR if it was rebased already few days ago? 16:07 sfan5 opinions needed re #14483 16:07 ShadowBot https://github.com/minetest/minetest/issues/14483 -- [no squash] Irrlicht import by Desour 16:22 Krock I'd prefer submodules but well that's not wanted apparently 16:24 sfan5 the long-term plan isn't to keep irrlicht separate 21:55 MTDiscord idea 21:55 MTDiscord what if irrlicht was separate and people can mod other render engines in 21:55 MTDiscord ill move that to lounge chat 21:57 celeron55_ what if... irrlicht was separate for 14 years and nobody did that 22:19 MTDiscord It is a little concerning to me that even up until now I haven't heard the reason for absorbing Irrlicht articulated even once. 22:20 MTDiscord It's probably been discussed somewhere, it just seems like everybody takes it for granted, and doesn't talk about it. 22:21 MTDiscord I've not been in favour of it because I value the ability to contribute to an upstream project very highly. The nature of open source is that users can contribute something back to improve the project. The Irrlicht community is still active, so severing ourselves from them seems like sort of a drastic and potentially controversial decision to me. 22:22 MTDiscord Pretty sure the idea is that by the time it is merged into the engine, it wont really be irrlicht any more 22:22 MTDiscord Maybe it's because core devs use different communication channels and don't share those discussions with the community. 22:24 MTDiscord So we think it'll be faster to redo the architecture of the existing Irrlicht engine compared to writing our own? 22:24 MTDiscord frankly, between reading here and github issues/prs, have been well informed. i suggest you do the same 22:25 celeron55_ irrlicht isn't maintained anymore and is grossly outdated in every possible way. by absorbing it into the engine we can get rid of it. the unneeded parts will be shed off and the necessary parts will get modernized 22:25 MTDiscord That does make sense. 22:25 celeron55_ this process has been going for a long time already. it was started years ago 22:25 MTDiscord In my past experience, Minetest's GitHub has contained a lot of controversy, drama, rude language, and bikeshedding. I don't consider it a trustworthy source of information anymore. 22:26 celeron55_ getting rid of the separate repo is a major milestone in the process, of course 22:26 MTDiscord The core devs sometimes post useful information there, but I don't have time to filter through all the spam. I'm a college student. 22:28 celeron55_ that's where most development discussion happens, though. it's the obvious place for it, and it's designed for developers so developers like it. it may be a bit much for those wanting to just follow along though 22:28 MTDiscord so basically you dont want to read the one of the two places (irc and github) that development is done. yet you complain here about not being informed 22:29 MTDiscord Perhaps it was wrong of me to complain here. I am sorry. 22:30 MTDiscord for context, it is stated at the head of the irrlichtmt roadmap issue > IrrlichtMt is meant to be copied into Minetest some day, ideally at a point where we can throw away much of the code because we would use SDL2 + our own OpenGL(ES) renderer by then. https://github.com/minetest/irrlicht/issues/107 22:30 MTDiscord I have read that thread multiple times. It doesn't contain any information about the rational or decision making, and what alternatives were rejected and why. 22:31 MTDiscord Also, ideally is italicized, indicating that it's not a realistic expectation. 22:31 MTDiscord The issue is a good summary of the work that needs to be done to achieve a goal that was already decided upon somewhere else. 22:32 MTDiscord there was an issue in minetest/minetest that had iirc what you are looking for. talking about upstream irrlicht taking forever to release, etc. suppose i have to find it for you 22:38 Mantar irrlicht is still maintained though? https://sourceforge.net/p/irrlicht/code/HEAD/tree/ lists the last update as as 2024-03-16 22:38 MTDiscord ah, here we go: https://github.com/minetest/minetest/issues/10738 22:39 MTDiscord mantar: i quote the above issue, which is 3 ish years old now > 1.9 has been in-dev since like 2013, it has a lot of big changes 22:40 Mantar that's very different from "irrlicht isn't maintained anymore" 22:41 Mantar more correct to say "the next release is stuck in development hell" 22:41 MTDiscord Ok 22:43 MTDiscord iirc 10738 was one of the things that sparked off minetest forking irrlicht (note the orginal comment has been since edited to reflect the fork, read orginal version). you would have to dig in here for additional conversations as well as some other issues 22:46 MTDiscord Writing a modern OGL 4 backend is something I'm considering since it might enable proper order-independent transparency, which in turn enables some instancing optimizations. It'd probably be a massive chunk of work though. 22:50 MTDiscord additionally, minetest probably isnt the first or the last to fork/replace irrlicht. most notable fork continuing on as a project is probably https://github.com/Devsh-Graphics-Programming/Nabla