Time  Nick       Message
11:02 ROllerozxa Could I get page deletion permissions on the Minetest Wiki? I've been going through and done a large cleanup and updating effort of the wiki (removing obsolete content, linking to ContentDB instead of maintaining outdated and duplicate content, updating things to be more CURRENT_YEAR...), and it'd be useful if I could actually delete all the pages that have been proposed for deletion, and also go through the uploaded images as there are a lot
11:02 ROllerozxa of images that are now unused
11:02 ROllerozxa For reference, here's the category of delete proposed pages: https://wiki.minetest.net/Category:Proposed_for_deletion and list of unused files: https://wiki.minetest.net/index.php?title=Special:UnusedFiles&limit=500&offset=0
20:36 kilbith    x2048' octablock-ssaa: https://i.imgur.com/xDwEnMg.png
20:36 kilbith    which improves the long distance terrain look
20:37 sfan5      that screenshot looks 100% like terasology
20:40 kilbith    https://i.imgur.com/34lD0mR.png
20:41 kilbith    it is *minetest*  https://i.imgur.com/XT4U7ln.png
20:41 Krock      it's definitely Minetest because there's aliasing effects in the water
20:42 kilbith    and you better be reviewing and no more delaying x2048' PR
20:42 sfan5      I can add extra delay if you want
20:43 kilbith    if you want we can just go ahead
20:43 kilbith    what it is shitty mentality
20:44 MTDiscord  <ROllerozxa> hot damn, proper antialiasing?
20:50 celeron55  talking about looks, have you seen this water trick apercy posted on the forums? https://forum.minetest.net/viewtopic.php?p=420342#p420342
20:51 kilbith    x2048 could simply decide to fork MT and tear the community in two
20:51 kilbith    he has the power to do it so
20:51 kilbith    so let's be nice and reactive with him
21:05 celeron55  13133 is a bit slower for me on my intel integrated gpu, in my latest hilly test scenario that i happened to load into, where more occlusion culling was possible before
21:07 celeron55  it's a rather obvious effect on hardware limited more by vertices than drawcalls. not sure how much anyone cares
21:08 kilbith    can you leave some solid data comparison on the PR
21:09 celeron55  don't have time now, have to sleep. but i don't expect it to be fixed with any reasonable amount of effort, and i also don't expect the PR to be discarded
21:09 celeron55  so i won't waste my time
21:09 celeron55  i just hope people are aware of this
21:16 celeron55  the difference was very slight though and the test was very haphazard, so don't put any value on that for now. i'll try to dig up some actual problems for a minute and then go
21:18 celeron55  the diffeerence is very small for me even on my dedicated gpu so it's not easy to have much opinions about this
21:19 kilbith    I wonder how less drawcalls can induce less performances
21:29 celeron55  this test scene with 400 range, on my nvidia gpu, renders at drawtime=60ms with upstream/master and at drawtime=57ms with 13133. so, tiny difference. that's with medium shadows. disabling shadows doubles the performance for both but that's not what i happened to have set up
21:30 celeron55  i was more concerned about weird things happening at very small view ranges but don't see anything of concern
21:31 celeron55  so i'm going to be neutral about this
21:35 celeron55  i'm more interested about your opinions about the screenshot link i posted