Time Nick Message 02:32 paramat pow! game#1025 02:32 ShadowBot https://github.com/minetest/minetest_game/issues/1025 -- Default: New grass and side textures by paramat 02:33 VanessaE no. 02:33 VanessaE too saturated and way too smooth 02:33 VanessaE they look like..plastic 02:34 hmmmm sorry, i agree 02:34 hmmmm nevermind the textures, what about sounds? 02:35 hmmmm those need way more work imho 02:35 hmmmm i have not updated minetest_game in a while, but i'm betting the sounds are all still eggshells crackling or differently pitched taps 02:36 VanessaE I always have my sound muted when I play at all, but I think they still are. 02:37 paramat meh 02:37 VanessaE sorry, paramat. they suck :) 02:39 hmmmm what do you guys think about 3979? 02:39 hmmmm #3979 02:39 ShadowBot https://github.com/minetest/minetest/issues/3979 -- Android settings optimization by MoNTE48 02:41 Zeno` I was thinking about that last night and tried them out. They do seem better on my Android at least 02:41 paramat some of the settings are possibly a good idea 02:42 hmmmm 90% of those setting changes shouldn't (theoretically) make much of an impact at all 02:42 Zeno` and yeah, changing settings on Android is not very easy (most users wouldn't know where to start) 02:42 hmmmm wondering if this MoNtE guy actually tested each setting 02:42 hmmmm also all of this is so incredibly subjective 02:42 Zeno` well, that's a different question :) 02:42 hmmmm you're not going to get any decent quality information unless you collect performance data on users' devices 02:43 paramat view range 25 seems too small 02:43 hmmmm exactly 02:43 Zeno` much too small 02:43 hmmmm and that's probably where 95% of the performance boost is coming from 02:43 hmmmm that single setting change 02:43 Zeno` 98% :) 02:43 hmmmm and that's a bit too much of a trade-off 02:44 hmmmm setting fps_max to 30 doesn't do anything 02:44 paramat 3d clouds off seems ok 02:44 hmmmm 3d clouds off, sure, but that only improves fps by like... 2 02:44 paramat yeah 02:44 hmmmm nevertheless it's a less significant detail that the users won't likely miss 02:45 hmmmm setting active_block_range to 1 is a great way to nerf gameplay 02:45 hmmmm we can strike that change out, and it's not likely to do anything either 02:45 hmmmm in fact, many of these options only make a difference if they're in singleplayer 02:46 hmmmm would be worthwhile to collect information on how often android players play singleplayer vs. multiplayer 02:49 paramat client unload unused data timeout seems too low 02:50 hmmmm max_objects_per_block won't practically improve performance 02:50 hmmmm 95% of this change is snake oil. i'm sorry, i won't agree to merging it. 02:50 paramat agreed 02:50 hmmmm so that's my -1 02:51 Zeno` paramat, how is the feature freeze progressing? 02:52 paramat well, https://forum.minetest.net/viewtopic.php?f=18&t=14102&p=213709#p213694 02:53 paramat it's waiting on the milestones 02:53 Zeno` The Ancient Romans have a lot to answer for their invention of milestones 02:55 Zeno` #2094 can probably be merged? That's one out of the way 02:55 ShadowBot https://github.com/minetest/minetest/issues/2094 -- Fix offset being ignored by inventory bar HUD by rubenwardy 02:56 hmmmm did you review it 02:56 Zeno` reviewed yes, tested no (because there are already 2 approvals) 02:56 paramat i reviewed it and was planning to merge it tonight 02:56 hmmmm okay hold on 02:57 hmmmm if past history has taught me anything, it's that if somebody didn't test it because they assumed somebody else tested it, it's likely that nobody actually tested it 02:57 paramat heh, rubenwardy tested it 02:58 paramat i didn't 03:00 Zeno` I test everything before I do an actual merge 03:02 paramat please can anyone review #3973 and #3974 ? some final mapgen stuff for release 03:02 ShadowBot https://github.com/minetest/minetest/issues/3973 -- Mgvalleys: Don't let cavegen place biome nodes everywhere by paramat 03:02 ShadowBot https://github.com/minetest/minetest/issues/3974 -- Mgv5: Optimise tunnels, add biome material in entrances by paramat 03:04 Zeno` I am going to an appointment in an hour but tonight I will review and test all of the remaining milestones not tested by the time I get back 03:05 Zeno` is #3948 critical for 0.4.14? 03:05 ShadowBot https://github.com/minetest/minetest/issues/3948 -- Escape more strings: formspecs, item descriptions, infotexts... by Ekdohibs 03:05 hmmmm i think i reviewed that yesterday but i forget what happened 03:05 hmmmm oh right, est's comments weren't addressed yet 03:06 Zeno` it seems there are a lot of questions that remain unanswered for a PR that's not all that old 03:07 Zeno` I hate character sets 03:08 paramat well i don't understand why it's so essential for release 03:08 hmmmm woo 03:08 hmmmm that's a messed up HUD if I ever saw one 03:09 hmmmm can confirm the bug at least, now let's confirm the fix 03:09 hmmmm and then i need to confirm what the fix is doing 03:09 Zeno` paramat, that's why I was asking if it was critical for the release or not. To me it would seem prudent to delay it 03:09 hmmmm it doesn't seem kosher exactly, because leftuppercorner is supposed to be the screen offset 03:10 hmmmm I guess 3948 could wait... 03:10 hmmmm I want sofar's node falling improvement in though 03:11 paramat that grass is close in saturation and brightness to 0.4.10 grass, however i expect others will find it too bright, so i'll redo it :/ 03:11 hmmmm could we merge that right now? i think if there's a bug with it we'd see that right away given everybody playing the current head of the development branch 03:11 VanessaE you could always... use HDX grass ;) 03:11 hmmmm although what would be best is to write a unit test for that algorithm 03:13 Zeno` the falling node improvement is very nice 03:13 Zeno` it's not in the milestones though 03:13 hmmmm so? 03:14 hmmmm the milestone thing is just something that nerzhul decided on doing 03:14 Zeno` so those are all I'm looking at right this moment 03:14 Zeno` oh ok... 03:14 Zeno` I was looking because paramat said the freeze/release was delayed because the milestones had not been met 03:14 hmmmm i would imagine that has more to do with the bugs with blocker status 03:15 Zeno` #3894 should be removed from the milestones (it's still a WIP for a start) 03:15 ShadowBot https://github.com/minetest/minetest/issues/3894 -- Add serverside noclip enforcement by est31 03:15 paramat it's not in milestone because nerz thinks it should be delayed 03:16 Zeno` hmmmm, yeah I'll look into the blocker ones after I get back from my appointment (I guess you're looking at them as well) 03:17 paramat (.. i refer to 3938) 03:17 paramat i'd like 3938 in for release, we have weeks to test it 03:17 Zeno` Ok, I'll ask my question in a different way. What can we do to make the release happen more quickly? 03:18 paramat work on the milestones :) 03:18 Zeno` ok, well get the milestones up-to-date :p add 3938 to the list 03:19 Zeno` :) 03:19 hmmmm wow i really don't remember hud looking this ugly 03:19 paramat 3894 is somewhat needed for the related meshdoors bug, est was working on it 03:19 hmmmm a nuclear bomb went off in this file 03:20 Zeno` probably that nuclear blast mod 03:20 hmmmm i need to do some reading 03:20 paramat ok added 3938 to milestones 03:21 hmmmm ahh, I get why offset exists now 03:27 Zeno` hang on 03:28 Zeno` is there a feature freeze or not? 03:28 paramat not 03:28 Zeno` I dunno why I was told there was when I came back then 03:28 paramat but focussing on stuff essential for release 03:28 VanessaE MT is "refrigerated" right now :) 03:28 paramat feature 'chilled' 03:28 VanessaE on the door shelf next to the juice, I believe paramat said before 03:29 Zeno` <Zeno`> paramat, how is the feature freeze progressing? 03:29 Zeno` it could have at least been clarified when I said that 30 minutes ago lol 03:29 hmmmm what the hell... is it just me or is screenshotting broken now? 03:29 Zeno` I was told that there was a feature freeze by at least TWO core devs on my first day back that there was a feature freeze 03:29 Zeno` and several times since then 03:29 VanessaE hmmmm: what commit are you at? 03:29 hmmmm "Saved screenshot to './screenshot_blah.png'" shows up in the chat console, but not in the chat area 03:30 hmmmm and I can't find the screenshot.. 03:30 hmmmm current 03:30 paramat sorry i thought my forum post explained freeze is delayed, also see channel banner 03:30 Zeno` no wonder things are slow when half the devs think there is a freeze and half don't :P 03:30 VanessaE hmmmm: works fine at commit 01ae43c 03:30 Zeno` as if we look at things like channel topic and the forum hehe :) 03:31 VanessaE at least, it dropped the screenie in my home dir. 03:31 Zeno` ok, well this changes things, I am going into full development and debugging mode as soon as I get home tonight 03:31 Zeno` instead of just dawdling around :) 03:32 paramat i have noticed that colours are changed in screenshots 03:32 hmmmm oh weird 03:33 hmmmm the CWD is a totally different directory 03:34 hmmmm seems like a KDE problem 03:34 hmmmm this is the sample code with rubywarden's patch applied: http://i.imgur.com/NAHe7rD.png ??? 03:36 Zeno` umm. nice? 03:37 hmmmm well then again, I guess it does fix the offset 03:46 Zeno` yeah 03:46 Zeno` the artifacts are a different bug 03:46 Zeno` hmm or are they 03:47 hmmmm they are different, 2094 isn't supposed to fix 3970 03:47 Zeno` yeah ok I got confused just because of what I had open in my browser... had to scroll up in the conversation 03:48 hmmmm they were reported by the same person, one messed up looking hud screenshot was in the other issue, one was a pr the other was an issue, etc. it's easy to get confused here 03:50 paramat shall we re-add #3848 to milestones? hmmmmm you seemed to want to work on it? it was something RBA was working on for this release so it would be good if possible 03:51 ShadowBot https://github.com/minetest/minetest/issues/3848 -- Send only changed node metadata to clients instead of whole mapblock by est31 03:51 hmmmm no that's not going to be possible due to the protocol change 03:52 hmmmm it would've been nice, because it has a substantial impact 03:54 VanessaE I thought that one used one of the optional protocol extensions that are used for other stuff? 03:56 hmmmm I'm going to merge 3938 03:56 Zeno` I agree 03:56 hmmmm if there is a bug with it, trust me, we'll know in less than 6 hours 03:56 Zeno` I just went through it 03:57 Zeno` seems like a fairly straightforward unwinding and the code *looks* fine 03:57 hmmmm we'll get at least 3 different habitual crack users^H^H^H^H^H^H^H^H^H^Hminetest users complaining about how minetest just got broken after they updated 03:57 Zeno` it's not even unwinding... it's still recursive just maintaining its own stack 03:57 hmmmm would feel better about it if it had a unit test though :) 03:57 Zeno` I am not writing a unit test for that hehe 03:57 hmmmm me neither 03:58 Zeno` especially if it has to be written in Lua 03:58 hmmmm but lua in general is pretty easy to unit test because you can hook functions by simply doing "=" 03:58 kaadmy hmmmm: terminal-based irc client? ;) 03:58 Zeno` I guess so, but let's just merge it :) 03:58 hmmmm kaadmy: yea 03:58 kaadmy we'll get at least 3 different habitual crack users^H^H^H^H^H^H^H^H^H^Hminetest users complaining about how minetest just got broken after they updated 03:59 Zeno` well it's better broken earlier so it can be fixed earlier! 03:59 kaadmy i forget what C-H does 04:00 hmmmm backspace on a misconfigured vt220 04:00 Zeno` hjkl are for cursor movement 04:00 Zeno` in vi/vim anyway 04:03 hmmmm found a bug 04:04 Zeno` you're a habitual crack^D^D^D^D^D minetest dev 04:04 hmmmm or maybe not 04:05 Zeno` gotta go; be back tonight 04:05 hmmmm when you're adding a node adjacent to floating falling nodes, is the falling supposed to stop at a block border? 04:08 hmmmm so the problem seems to be that a falling node caused by another node update does not update the adjacent falling nodes 04:10 paramat sofar ^ 04:11 hmmmm hrmm, it's debatable if this is a bug or not actually 04:12 hmmmm did the original nodeupdate work on falling nodes adjacent to the placed node? 04:13 paramat i think so 04:13 hmmmm ahh yes, looking at the code, it did 04:14 sofar hmmm? 04:14 hmmmm it updated the 3x3x3 cube enclosing the updated node, this only updates the 6 neighbors 04:14 hmmmm no diagnoal nodes 04:14 sofar old code went 3x3x3, new code goes 6 ways 04:14 sofar that gets all adjacent 04:15 sofar diagonal is not adjacent 04:15 hmmmm paramat, what do you think? 04:15 hmmmm should it update the 27 neighbors or 7? 04:15 paramat well updating the 6 will get all 27 after 3 iterations 04:16 hmmmm what?? 04:16 sofar right 04:16 hmmmm oh i see what you're saying. that will only work if the adjacent node isn't supported either 04:17 hmmmm what i'm saying is that this is functioning slightly different from what i naturally expected from older versions of minetest 04:17 sofar the walking matrix can be expanded 04:17 hmmmm not saying this is wrong behavior, it's just different from how it used to be and it might make other users complain about a bug 04:17 sofar I'd like to keep the 4 corners out, but adding simple diagonals seems fine 04:17 paramat the diagonal neighbours will still be updated, but indirectly through combinations of non-diagonal updates 04:18 hmmmm alright we'll keep it as-is 04:18 paramat but i don't understand the PR fully 04:18 sofar I don't blame you, lol 04:18 sofar it's a recursive tail call optimization 04:19 paramat i'm just the mapgen bitch 04:19 hmmmm if it really is a problem for users we can just add the other 20 neighbors to the nodeupdate_neighbors list and change the numeric literal '7' to '27' 04:19 sofar paramat: essentially, instead of using the function call stack, we use an array in lua 04:19 sofar hmmmm: yes 04:19 sofar it would be really trivial to add those in 04:19 hmmmm ok merging then 04:19 paramat btw #3985 can be merged 04:19 ShadowBot https://github.com/minetest/minetest/issues/3985 -- Minimap: revert change from RGBA to Indexed by sofar 04:19 paramat trivial 04:20 hmmmm agreed 04:20 hmmmm i'll merge that 04:25 hmmmm alright any more approvals, give them to me now, otherwise i'm going to work on #3970 04:25 ShadowBot https://github.com/minetest/minetest/issues/3970 -- Graphical bugs when drawing inventory hud 04:25 hmmmm any more PRs in need of approval* 04:26 sofar #3891 ? 04:26 ShadowBot https://github.com/minetest/minetest/issues/3891 -- Implement client node dig prediction. by sofar 04:26 sofar well no 04:27 sofar need to wait until the noclip anticheat gets merged first 04:27 sofar at least, that was the deal 04:27 paramat #3973 #3974 04:27 ShadowBot https://github.com/minetest/minetest/issues/3973 -- Mgvalleys: Don't let cavegen place biome nodes everywhere by paramat 04:27 ShadowBot https://github.com/minetest/minetest/issues/3974 -- Mgv5: Optimise tunnels, add biome material in entrances by paramat 04:28 paramat these are possibly my last 2 mapgen engine things needed for release 04:31 paramat i can merge those myself if approved 05:10 hmmmm hey paramat, next time, if you're moving a function to a different location, don't make any changes to it 05:10 hmmmm because now you can't see the diff of what parts actually changed and what was moved 05:10 paramat good point 05:11 paramat i noticed the diff is a bit confusing 05:11 hmmmm actually, nvm, your new version of generateCaves is substantially different from the old one, so it doesn't make much of a difference 05:12 paramat still, function move is best a seperate commit i guess 05:12 hmmmm so you iterate over (x, z) in columns in this new version 05:14 hmmmm one minor nuance i should point out with your approach here: 05:15 hmmmm a large cave from CaveV5 could get carved from that area and not have its entrance get biomed 05:15 paramat yes 05:16 paramat oh but large psrand caves are only deep underground in mgv5 05:16 paramat but indeed you are correct for mgv7 05:18 paramat i'll consider it 05:18 paramat later 05:32 paramat thanks for comments 05:38 paramat i'll work on those issues 05:49 nore About #3948, it's not critical but if it isn't merged, it means the features it allows will have to wait even longer 05:49 ShadowBot https://github.com/minetest/minetest/issues/3948 -- Escape more strings: formspecs, item descriptions, infotexts... by Ekdohibs 05:50 hmmmm anybody have any opinions on $2085? 05:50 hmmmm #2085 05:50 ShadowBot https://github.com/minetest/minetest/issues/2085 -- Disable font scaling on non-Android platforms by cheapie 05:50 nore What's more, it addresses a few issues with the previous pr I made that was merged 05:52 nore hmmmm: sorry, gtg now :/ 05:52 hmmmm bye 05:52 hmmmm I don't know why 2085 is waiting exactly 05:52 * nore is sick and in his bed 05:55 paramat sorry i have to go eat 06:13 hmmmm 3970 has been around since at least january 2014 06:21 sofar eat at this hour? where is paramat? midnight taco bell run? 06:22 sofar lol 06:22 hmmmm i think it's 8 am for british pepole 06:22 hmmmm care for a cuppa? 06:24 sofar ahhh I didn't realize he's in the UK 07:00 hmmmm geh 07:00 hmmmm there's no way to rotate an image in irrlicht without rendering it as a 3d object 07:00 hmmmm i'm gonna have to chop up gui_hotbar.png, but it might break reverse compat with older clients 07:07 VanessaE hmmmm: actually, I think now's a good time for older-client-breaking changes 07:08 hmmmm well you can still have compatibility, it's just that you'd have to start distributing two versions of the gui hotbar anymore 07:08 VanessaE I really think it's a good idea to get that stuff into 0.4.14, since we already have one or two older-client-breaking changes in place as it is 07:08 hmmmm either way this isn't a blocker 07:08 VanessaE (connected nodeboxes, for example) 07:09 hmmmm it's a bug that's been in there for a long time 07:10 hmmmm also, are there any hotbar backgrounds that are asymmetrical? 07:10 VanessaE none that I can think of 07:11 hmmmm i think the way mt currently handles inventories and how deeply integrated everything is with the inventorylist is very monolithic 07:12 hmmmm i want to be able to have a mod that's an RPG, and you can switch between several weapons or spells in your inventory, but each is broken up into its own inventory 07:12 hmmmm and i want to be able to have one inventory on the left side of the screen, and the other on the right side, going downward 07:12 nrzkt VanessaE, if you look at protocol changes, clients before 0.4.10 should be forbidden because they could not use breath for example 07:13 VanessaE I would tend to agree with you on that, sure 07:13 hmmmm this is so easy to do but it requires a rewrite 07:13 hmmmm ugh 07:13 VanessaE of course 0.4.10 was a long time ago 07:13 nrzkt hmmmm, agreed the inventory is very very monolithic 07:13 srifqi what do you mean by monolithic? 07:14 hmmmm having multiple itembars as it exists right now is basically useless 07:14 hmmmm i don't even know why you're able to add one via lua 07:14 hmmmm ABSOLUTELY 07:14 hmmmm USELESS 07:14 Obani hmmmm, I think#3985 can be closed ^^ 07:14 hmmmm this is why 3970 is not a blocker and should be set to low priority 07:15 Zeno` hmmmm, what makes you say that 3970 has been around for more than 2 years? 07:15 nrzkt i think we should stop all those breaking changes and start a feature freeze, because huge rework is needed on master as i read on irc. 07:15 Zeno` and why don't I experience it hehe 07:15 nrzkt it's time to have a new MT step 07:16 Zeno` nrzkt, until this afternoon I thought we *were* in feature freeze so I agree 07:16 hmmmm Zeno`, what mods have you used that add an extra itembar to your hud? 07:16 hmmmm seriously? 07:16 Zeno` there is never going to be a release if we wait 07:16 Zeno` hmmmm, oh ok... so I need an extra itembar? 07:16 sofar I could see a use for armor 07:16 Zeno` hmmmm, I haven't tried that at all 07:16 hmmmm Zeno`: Exactly 07:16 hmmmm again, this ONLY affects lua-added itembars 07:16 Zeno` hmmmm, well that's the only answer I was after 07:17 hmmmm there is literally no reasonable usecase for a lua-added itembar 07:17 hmmmm because again, it's so monolithic 07:17 Zeno` no wonder I'd never seen it 07:17 Zeno` hmmmm, I wasn't saying you were wrong saying it had been around for 2 years 07:17 Zeno` I just wanted clarification; sorry if I wasn't clearer 07:18 Obani you mean the default item bar from hud ? 07:18 Zeno` Maybe that info should be added to the bug report 07:19 Zeno` now back to why this stupid linux installation wants UEFI when I've never needed it prior to this version of Fedora :/ 07:19 hmmmm actually you know what 07:19 hmmmm I do remember a point at which somebody added more than 8 items to the itembar 07:19 hmmmm and it went onto a second row 07:19 hmmmm and it worked correctly 07:20 VanessaE dreambuilder 07:20 VanessaE carbone also. 07:20 hmmmm right 07:20 Obani hmmmm, item bar size is configurable in minetest.conf anyway 07:20 VanessaE and when it does that, it breaks the HUD background 07:20 Obani And it's very useful 07:20 hmmmm well those are broken after somebody added this gee-whiz way of customizing the itembar background 07:20 VanessaE (the image just stretches to fit. imho it ought to tile..or something) 07:20 hmmmm IDK who did it but certainly wouldn'tve passed my review 07:21 Obani I don't understand the interest of adding another item bar as you can already choose its size in mt.conf :x 07:21 hmmmm https://github.com/minetest/minetest/commit/439079cad8f15028fffebb05aed5ae6c0da9bfab 07:22 hmmmm this is the commit that caused 3970 07:22 hmmmm the bug has been present since 0.4.8 07:22 Zeno` size == number of slots? 07:25 hmmmm yeah 07:25 hmmmm okay one blocker left then, 3901 07:28 ssieb Zeno`: Is your hard drive formatted as GPT? 07:29 Zeno` ssieb, I suppose it must be 07:30 Zeno` ssieb, but why does it currently work without /boot/uefi (I just have /boot) 07:30 Zeno` but yes it's GPT 07:30 ssieb no idea. What version do you have currently? 07:30 Zeno` fedora 21 07:31 ssieb And you're booted in BIOS (CSM?) mode, not UEFI? 07:31 Zeno` correct 07:31 Zeno` I don't mind making an extra directory (/boot/uefi) if that's all it is... 'cause there is /boot already 07:31 ssieb How did you get it in that configuration? :-) 07:32 hmmmm can anybody here reproduce #3901? 07:32 ShadowBot https://github.com/minetest/minetest/issues/3901 -- Air bubbles not showing up under water 07:32 ssieb It's actually /boot/efi, but that has to map to the EFI partition. 07:32 Zeno` ssieb, I probably added it. I do remember using chroot and doing most of the f21 installtion by hand hehe 07:32 ssieb oh! 07:32 ssieb well, then, you're just asking for it! :-P 07:33 ssieb hmmmm: I haven't had that happen with recent builds 07:33 hmmmm if more people can verify the bug doesn't work, perhaps wait until we have kilbith try it again, i think we might be good for release 07:33 hmmmm and this time, whoever makes the windows builds is gonna save the .pdbs or else 07:33 Zeno` ssieb, I don't use grub either (my m/board lets me choose a boot device so there is no need) 07:34 Zeno` so maybe I'm on my own again. *sigh* 07:34 ssieb I'd say you've gone pretty far off the standard track 07:34 Zeno` oh well, I'll do it after dinner 07:34 Zeno` ssieb, that's the way I roll :3 lol 07:35 hmmmm i installed freebsd manually one time 07:35 hmmmm it was not pleasant 07:37 Zeno` I'd better make some disk images first 07:37 Zeno` hmmmm, I thought freebsd was your main OS 07:37 hmmmm yea it is 07:38 Zeno` oh ok :) 07:38 hmmmm i was reading the instructions from my laptop at the time 07:38 hmmmm these days, you can install zfs as the root fs right from the installer. no manual installation necessary! imagine that 07:39 hmmmm technological progress people 07:39 hmmmm we're in the future now 07:39 hmmmm little 1s and 0s flying everywhere 07:40 Zeno` that's how I installed f21... laptop next to me, chrooting, doing silly stuff 07:41 Zeno` I suppose I could just fall into the trap and enable UEFI 07:41 Zeno` makes me very uneasy but I guess the flood gates are siezed open now anyway 07:42 asl97 pretty sure fedora 21 reach EOL 07:42 Zeno` but then I'll probably have to reinstall Win7 as well 07:42 Zeno` asl97, it has 07:42 Zeno` asl97, I've been manually keeping it up to date but that approach is not working now 07:43 Zeno` ok, I have dual bios so I can probably have one as traditional BIOS and one UEFI 07:43 Zeno` so I don't have to reinstall win7 as well 07:43 Zeno` wish me luck 12:02 Fixer vrange 50 for android will be slideshow on my tablet 12:02 Fixer it is already working badly with 15 12:02 Fixer lenovo yoga 12:02 Calinou how old and low-end is it? 12:03 Fixer kitak 4.4, probably 2 years old 12:03 Fixer kitkat* 12:03 Fixer middle-end 12:04 PilzAdam Fixer, why do you even play it on your device, then? 12:04 Fixer i don't 12:04 Fixer i just tested minetest on it few times 12:04 Fixer i play on my PC ofc 12:05 Fixer i'm not a mazohist 12:05 Fixer >_> 12:06 PilzAdam well, why are you telling us that a vrange of 50 is too high, when your device wouldn't run Minetest anyway? 12:06 Fixer it is running it, but fps is low 12:07 Fixer and it is middle-end one iirc, a lot of people will have worse experience 12:07 Fixer 50 v_range on android is bad idea 12:11 PilzAdam if your device doesn't even run with a vrange of 15 properly, then Minetest is not playable on it, at all 12:11 PilzAdam so we shouldn't even bother with your experiences 12:12 Fixer let me say it has somewhere near 24 fps with vrange 15 12:12 Fixer let me try it now in fact 12:14 PilzAdam I personally wouldn't consider a vrange of 15 playable, even it was running at 60 FPS 12:15 Calinou Minetest wasn't designed for mobile devices after all 12:16 Calinou to be honest, it probably should never have been released for Android 12:16 Calinou the experience is just so bad 12:16 Calinou tiny GUI elements, poor inventory management 12:16 Calinou and low performance 12:21 srifqi Because the GUI is still heavily-based for desktop. Even for tablet 10.1", it still a little bit hard to play. 12:21 Calinou GUI elements typically remain tiny in game even with that 12:22 Fixer tested in singleplayer, fdroid build, 0.4.13, ~24-26fps with mostly v_range 20, yoga tablet 10 (b8000h) / MTK MT8389 QuadCore 1.2GHz, 1GB RAM, 1280*800 13:48 est31 IMO we should just get this HTML thing for HUD/formspecs 13:48 est31 then we can make the modders make responsive GUIs 13:48 est31 people who play full screen complain that the buttons are too large 13:48 est31 people on mobile devices complain that they are too small 13:49 est31 just lets add HTML rendering with lua access to the DOM 13:52 est31 and #3948 is not that important... after all one would need lecacy handling for clients either way or another 13:52 ShadowBot https://github.com/minetest/minetest/issues/3948 -- Escape more strings: formspecs, item descriptions, infotexts... by Ekdohibs 13:52 est31 so the only thing we save is a protocol bump 13:54 nrzkt est31: note irrlicht has a xml parser included to help us 13:55 nrzkt i already started to think about a XML interface, it's a huge PR to do :) 14:29 est31 I rather add no more additional dependencies on irrlicht 14:29 nrzkt then you will add another library for XML parsing ? lol 14:30 nrzkt as you want, libxml could be useful but is this portable ? 14:39 PilzAdam don't forget the javascript engine 14:39 PilzAdam Mineweb 2.0 14:40 est31 PilzAdam, they solved problems already, why should we rebuild something they built decades ago? 14:41 PilzAdam est31, hm? my comment was for pure entertainment reasons 14:41 est31 ah okay then 14:41 est31 :) 14:42 PilzAdam I mean, I would still try to write a browser in Minetest, since we already have curl support... 14:43 nrzkt PilzAdam, i prepare malicious code then xD 14:44 sofar well 14:44 sofar it would give us client-side scripting if you added javascript support 14:44 jin_xi why not guile... 14:46 Calinou_ why not QuakeC 14:46 Calinou_ nobody needs structs 14:47 Calinou_ to est31: he should be aware that libRocket (the most popular "HTML in games" library) is notorious for being CPU-intensive, it has been used in games like Unvanquished and Warsow 14:49 asl97 is minetest still broken on the fdroid and playstore? 15:37 Calinou_ broken? 15:43 Fixer i have stable one from fdroid, and it runs fine 15:43 asl97 something about libiconv.so 20:08 VanessaE [04-11 16:07] err... the latest minetest daily has a glitch with hotbar, it appears in top left corner of screen 20:23 hmmmm don't see it. 20:35 VanessaE just passing it along, I haven't checked it myself yet 20:41 Fixer i will check it 20:41 Fixer have newest build at hand 20:42 Fixer bug reproduced :/ 20:43 hmmmm try again with no mods 20:43 Fixer you mean with minimal? 20:43 hmmmm eh 20:43 Fixer reproduced again 20:43 hmmmm you're testing on a totally plan minetest_game? 20:43 Fixer _yes_ 20:44 Fixer and minimal dev test 20:44 Fixer problem :/ 20:44 hmmmm what resolution are you playing minetest at 20:44 PilzAdam hmmmm, I can reproduce it, too 20:44 Fixer 1280*1024 minus start bar 20:44 hmmmm what platform are you using 20:45 Fixer windows 7sp1 20:45 hmmmm pilzadam, what about you? 20:45 PilzAdam hmmmm, bisecting now, but it's probably eae33951ad8b13cb335fca8eaec73185959eb44e 20:45 hmmmm what is your resolution, and what platform are you using? 20:45 Fixer actual rendering is 1280*978 or so 20:45 PilzAdam hmmmm, you are the only one that can't reproduce it? 20:45 hmmmm i wouldn'tve merged it if i was able to reproduce the said bug 20:46 Amaz The hotbar is in the right place for me 20:46 PilzAdam what is your setup` 20:46 PilzAdam *? 20:47 PilzAdam eae33951ad8b13cb335fca8eaec73185959eb44e is indeed the offending commit 20:47 PilzAdam all games I tried, even ones that have a custom hotbar, are being displayed incorrectly 20:47 Fixer my screenshot https://i.imgur.com/jKrqicX.png 20:49 sofar PilzAdam: why does it not surprise me... fix the offset -> nobody was properly using it, so they're all 0? 20:49 PilzAdam rubenwardy, *ping* 20:49 rubenwardy I've seen 20:49 PilzAdam what was your test setup? 20:49 rubenwardy if the offset is 0,0, it shouldn't make a difference 20:49 rubenwardy this is the only line that could change things: https://github.com/minetest/minetest/commit/eae33951ad8b13cb335fca8eaec73185959eb44e#diff-af29bcc6c7dda4b2f32c1b2e0c0dba42R217 20:49 PilzAdam it seems that every single game is broken; whether they override the itembar or not 20:50 PilzAdam what have you tested this with? 20:50 paramat can anyone review game#1024 ? sfan5 ShadowNinja 20:50 ShadowBot https://github.com/minetest/minetest_game/issues/1024 -- Books: Split text per page relatively to number of lines by kilbith 20:51 rubenwardy I tested with minimal originally, then minetest_game when rebasing 20:51 PilzAdam hud_scaling is 1.0 for me 20:51 paramat i might merge game#543 later if nore doesn't object 20:51 rubenwardy I can't reproduce when using minetest_game 20:51 ShadowBot https://github.com/minetest/minetest_game/issues/543 -- Strengthen obsidian glass by tenplus1 20:51 nore paramat: fine for me 20:52 paramat ok, wishing you health :) 20:53 sofar nore: I wish you success in massacring billions of unwanted germs! 20:53 nore thanks :) 20:54 PilzAdam rubenwardy, maybe there's a problem with getDisplayDensity()? 20:54 * nore goes back to his bed to get better 20:55 rubenwardy All default settings, I just cleared minetest.conf 20:55 hmmmm oh 20:55 hmmmm maybe it's because I'm using an outdated minetest_game 20:55 hmmmm let me try minimal.. 20:56 hmmmm or not 20:56 PilzAdam getDisplayDensity() is 0.989583 for me 20:57 PilzAdam if I just comment it out then everything works fine 20:58 rubenwardy So it is that line 20:58 hmmmm ohh... so it's https://github.com/minetest/minetest/commit/eae33951ad8b13cb335fca8eaec73185959eb44e#diff-af29bcc6c7dda4b2f32c1b2e0c0dba42R217 20:59 PilzAdam ugh, why can't I << a v2s32 into infostream? 20:59 hmmmm because it doesn't have an << operator 20:59 PilzAdam but why? has nobody every thought this could be useful? 21:01 PilzAdam seems that the *= operator in that line is not working as expected 21:01 PilzAdam splitting it into 2 lines, and doing X and Y separately works fine 21:02 PilzAdam probably because the right-hand-side of the *= is cast into an s32? 21:02 rubenwardy damn overloading 21:02 PilzAdam so 1.0f * 0.989583f gets cast into 0 21:05 PilzAdam https://github.com/PilzAdam/minetest/commit/73d4538d9463b349eedc767c47d1750a164e2b7b 21:05 PilzAdam ^ hmmmm okay to push? 21:08 hmmmm that fixes it? 21:09 hmmmm yeah I guess that's good 21:09 hmmmm also i guess i wasn't affected because my display density was > 1 21:09 sofar I can't deny that that indeed looks correct 21:09 hmmmm wait, how does does that fix work even? 21:09 rubenwardy I have a high def screen too 21:10 PilzAdam hmmmm, v2s32 only overrides *= with an s32 argument; thus the floating point number get's cast to 0 21:10 sofar and it shifts everything top-left... 21:11 sofar it's missing a *= operator for int 21:11 sofar (which also would have been a good fix) 21:11 Fixer i can try it, needs 20 min though 21:11 PilzAdam sofar, v2s32 is an irrlicht type, no? 21:11 PilzAdam I don't want to patch Irrlicht 21:11 sofar yes, so you're not likely to fix it there ;) 21:12 PilzAdam the fix is merged; if you find something better then revert it 21:13 Fixer well, i will compile it and check now 21:30 paramat is anyone working on #3894 ? it is apparently important for release 21:30 ShadowBot https://github.com/minetest/minetest/issues/3894 -- Add serverside noclip enforcement by est31 21:32 paramat needed for game#954 21:32 ShadowBot https://github.com/minetest/minetest_game/issues/954 -- locked doors useless on servers 21:37 paramat sofar if possible could you combine the ABMS in game #999 ? 21:37 ShadowBot https://github.com/minetest/minetest/issues/999 -- Send die message to all by proller 21:38 sofar paramat: said I would, didn't I? 21:38 sofar must have slipped my mind, yes, I will update that. 21:38 paramat cool thanks 21:53 ShadowNinja BTW, I'm updating the Android build system to use Gradle instead of Ant (Gradle's been the prefered build system for a while). 21:58 Calinou cool 23:19 Fixer crap, i forgot to check the fix :S 23:20 Fixer DOES NOT FIX 23:21 Fixer situation intensifies 23:22 Fixer proof: https://i.imgur.com/jy2VbUU.png 23:31 * Fixer bumps hmmmm and runs 23:37 Fixer lol 23:38 rubywarden I'm not here 23:39 rubywarden That new line of hud scaling is just problematic. I can't remember why it's there. May it should just be removed :/