Time Nick Message 00:00 user333_ that's on 5.15.0-dev-0f943e581 02:42 [MatrxMT] I only asked about what Minecraft has because I want to raid them for potential ideas. This should be obvious from the amount of shade I throw at Mineclone-family games for being raw clones 02:43 [MatrxMT] If anything, having first-class map downloading and launching features can be another relatively easy win over Minecraft, like the main killer feature of not having to do any pre-download to join a server 03:24 repetitivestrain Blockhead256: and in my not so humble estimation, the mcl*s don't deserve the flak you throw at them 03:24 repetitivestrain :-) 03:26 [MatrxMT] I'll never say they're technologically deficient, only unoriginal :) 03:27 repetitivestrain yes, but conversely minecraft _is_ original, and therefore we're still realizing an original conception, just with different technology and with a moral advantage 08:26 [MatrxMT] pgimeno: Thanks for looking more into this. I will see what I can debug further, worst case Debian drops the Luanti i386/i686 build, since the architecture has reduced support within Debian now. 08:54 celeron55 "reduced support" is kind of a silly term as what happens is nothing works when you don't fix the weakest link 08:57 [MatrxMT] celeron55: I don't understand your message 09:02 [MatrxMT] Ah, now I follow. Yep, I guess the plan of course is to ditch 32-bit x86 eventually so they just need to fix enough bugs to keep the lights on until someone pulls the plug. 11:15 pgimeno as I suspected, the tests pass with CMAKE_BUILD_TYPE=Debug 12:14 pgimeno apparently, in the failing test setRotationRadians is receiving some ridiculous arguments (in the order of -1e+34). This might well be a compiler bug. 12:24 sfan5 compiler bugs are less likely than you think 12:25 [MatrxMT] pgimeno: Shouldn't setRotationRadians be able to handle very large values like that? I assume those are possible random values 12:49 pgimeno twrightsman in that test they are limited to 0..2*PI 12:49 SwissalpS does anybody have an idea how to find mods that set 'data' field in itemstack meta? 12:49 pgimeno and yes in theory 12:50 [MatrxMT] Ah sorry pgimeno, I see that bound limit now 13:09 Elouin Hi, is it fine to use a single database for world/player/auth data in postgres or does it need to be a distinct one for each? 13:13 [MatrxMT] depends on if there are table name overlaps at least. But I think it's not hard to set up multiple 'bases on the same instance 13:13 [MatrxMT] https://docs.luanti.org/for-server-hosts/database-backends/ 13:13 pgimeno twrightsman: My bad, it seems some kind of problem in gdb or my use of it. The argument to setRotationRadians seems to be OK. However, there's this conundrum. With this patch: http://www.formauri.es/personal/pgimeno/pastes/catch_setRotationRadians.diff a single line is printed on every run, and it has reasonable arguments. 13:13 [MatrxMT] it's not always a good idea to use postgres for everything though 13:13 [MatrxMT] https://github.com/luanti-org/luanti/issues/16181 13:14 pgimeno (Sorry about the lack of indentation, it's how I distinguish debug code so I can easily locate and remove it after debugging) 13:15 pgimeno with "reasonable" defined as "within the intended range of 0..2*PI" 13:21 [MatrxMT] But the test still fails? 13:21 sfan5 Elouin: best to use separate ones 13:21 pgimeno yes 13:21 pgimeno the NaN propagates all the way to the matrix, and the test fails when comparing the matrices 13:23 [MatrxMT] Ah duh I misread your code, it is printed because you made it print when NaN 13:24 pgimeno yes, again sorry about the lack of indentation 13:25 [MatrxMT] All good, your advance warning helped me 13:53 Elouin sfan5: okay. thanks. 19:46 Menchers lol nice nick ireallyhateirc 19:47 ireallyhateirc thanks 19:47 * Menchers wonders if idespisediscord is taken yet 19:47 jonadab Likely. 19:47 dudz ircisgreat 19:48 Menchers also hi jonadab 19:48 Menchers I only knew you from the nethack channels 19:48 Menchers lol 19:48 jonadab IRC has the advantage of being an open protocol, so you aren't stuck with some horrific proprietary app like Discord. 19:48 Menchers yup 19:48 Menchers client I'm using this week: quassel 19:48 jonadab irssi is at least decently configurable, mostly. 19:49 Menchers still prefer hexchat but it's unmaintained now 19:49 dudz weechat here 19:49 Menchers tried to like irssi 19:49 Menchers might give it another go 19:49 Menchers ability to run in a detached session on a shell server is a definite plus 19:50 Menchers wouldn't need ZNC 19:53 Menchers at least a couple luanti servers are IRC clients :D 19:53 Menchers your-land.de has an IRC client mod 19:55 dudz havnt worked out how to buy a hiuse there yet 19:56 dudz *house 20:27 jonadab Yeah, irssi in a screen session is just easier to setup than znc or whatever. 22:31 erle given that the dithering filter can be like 20% of the framerate i find it at least a bit weird that it is turned on by default 22:35 erle attempting to fix banding artifacts in postprocessing but not wanting gamma-correct rendering (which just *may* reduce the amount of banding, WHO KNOWS lol) is a bit weird though 22:39 erle as i understand it … unlike i am mistaken (enlighten me if you know better!), gamma-incorrect rendering will basically always reduce the dynamic range (which leads to more banding)