Minetest logo

IRC log for #minetest-dev, 2016-10-15

| Channels | #minetest-dev index | Today | | Google Search | Plaintext

All times shown according to UTC.

Time Nick Message
00:29 Robby joined #minetest-dev
00:54 misprint joined #minetest-dev
02:21 paramat joined #minetest-dev
02:31 kaeza joined #minetest-dev
02:38 paramat ehh .. the 'jungletrees simple' texture (for use with 'simple' leaf rendering) is a completely different colour to the standard texture. will fix
03:54 Tmanyo joined #minetest-dev
04:22 paramat left #minetest-dev
04:26 SloanOnLinux joined #minetest-dev
04:26 SloanOnLinux joined #minetest-dev
04:39 Hunterz joined #minetest-dev
04:52 Foghrye4__ joined #minetest-dev
05:26 MrIbby joined #minetest-dev
05:50 nrzkt joined #minetest-dev
06:13 MrIbby joined #minetest-dev
07:11 nerzhul_ joined #minetest-dev
07:13 nrzkt joined #minetest-dev
07:16 jin_xi joined #minetest-dev
08:05 kruug joined #minetest-dev
08:27 Calinou_ joined #minetest-dev
08:29 red-001 joined #minetest-dev
08:29 fling_ joined #minetest-dev
08:38 ssieb joined #minetest-dev
08:40 misprint joined #minetest-dev
08:51 SloanOnLinux joined #minetest-dev
08:51 sofar joined #minetest-dev
08:51 Megaf joined #minetest-dev
08:53 Megaf joined #minetest-dev
08:57 Krock joined #minetest-dev
08:58 Foghrye4__ Hi all! I've got assertion fail on C++ side on sending bone position on client by calling Lua function "set_bone_position". Is this bug known?
08:58 Foghrye4__ *to client
09:02 Krock what's the exact error message?
09:07 Foghrye4__ One second, i'm trying to repeat that bug.
09:08 Foghrye4__ What is biggest float in Lua?
09:08 Krock it's say it's inf
09:09 Krock *I'd say
09:09 Foghrye4__ serialize.h:278: void writeF1000(irr::u8*, irr::f32): Assertion `i >= ((float)(s32)((-0x7FFFFFFF - 1) / 1000.0f)) && i <= ((float)(s32)((0x7FFFFFFF) / 1000.0f))' failed.
09:10 Krock 1E308 is ~near to infinite
09:10 Foghrye4__ 1e100 was enought. It seems get_look_vertical() and/or get_look_horizontal() sometimes return numbers out of range.
09:13 red-001 could be caused by a bug in the client
09:15 red-001 are packets checked for integrity?
09:16 red-001 afaik get_loo
09:17 red-001 afaik get_look asks the client for the information
09:18 Krock it doesn't ask. it uses the last received look dir change from the client
09:20 Foghrye4__ Anyway, issue is submitted: #4635
09:20 ShadowBot https://github.com/minetest/minetest/issues/4635 -- Assertion fail on bone position update
09:21 Krock Foghrye4__, don't forget to provide your Minetest version (+ commit, if possible)
09:21 SloanOnLinux joined #minetest-dev
09:21 SloanOnLinux joined #minetest-dev
09:22 Krock it's sometimes a nice addition :)
09:24 Foghrye4__ Krock, done.
09:31 Krock2 joined #minetest-dev
09:35 red-001 does github prefer licence or license?
09:36 Krock joined #minetest-dev
09:36 Foghrye4__ red-001, dunno about github, but google translate prefer "license".
09:37 Krock "The license"  or "to licence something"
09:37 sofar_ joined #minetest-dev
09:37 Krock o/ sofar_
09:39 red-001 any options on game#1331 ?
09:39 ShadowBot https://github.com/minetest​/minetest_game/issues/1331 -- Use unregister_item() to remove tnt:tnt if tnt is disabled. by red-001
09:39 ptv joined #minetest-dev
09:41 Krock that change makes no difference, it's just moving code
09:44 red-001 It makes a diffrence if another mod uses tnt_register
09:45 Krock in that case there would be TNT even if it was disabled
09:45 Krock that's not the target
09:49 red-001 you could always disable the mod
09:49 red-001 I don't think there are any tnt mods that add something apart from tnt
09:50 red-001 A server owner might want to disable default tnt and use a diffrent mod for it
09:50 Krock still, if tnt is disabled it is disabled. no matter how many other tnt mods you have
09:50 troller joined #minetest-dev
09:51 Krock the other tnt mods usually use their own API
09:51 Krock because tnt is quite new in MTG
10:06 Foghrye4__ joined #minetest-dev
10:08 Megaf joined #minetest-dev
10:12 Guest53631 joined #minetest-dev
10:16 Foghrye4 joined #minetest-dev
10:28 jin_xi joined #minetest-dev
10:42 sfan5 pushing https://github.com/minetest/minetest/pull/4633 in 10 minutes
10:42 sfan5 (it already has an approval and is pretty trivial too)
10:52 sfan5 pushed
10:56 Zeno` joined #minetest-dev
10:59 ElectronLibre joined #minetest-dev
11:05 proller joined #minetest-dev
11:13 nrzkt Zeno`, sfan5, if you get time #4612 needs some reviews, it's not finished yet
11:13 ShadowBot https://github.com/minetest/minetest/issues/4612 -- [WIP] Player class cleanup / PlayerSAO - LocalPlayer ownership by nerzhul
11:49 red-002 joined #minetest-dev
11:50 red-002 joined #minetest-dev
12:02 davisonio joined #minetest-dev
12:13 est31 joined #minetest-dev
12:24 red-001 joined #minetest-dev
12:24 Fixer joined #minetest-dev
12:26 turtleman joined #minetest-dev
12:44 T4im joined #minetest-dev
13:00 red-001 #4551 needs another review
13:00 ShadowBot https://github.com/minetest/minetest/issues/4551 -- Trim whitespaces from the input of the `/privs` command. by red-001
13:06 est31 red-001, done
13:07 ElectronLibre joined #minetest-dev
13:17 red-001 thanks
13:22 Taoki joined #minetest-dev
14:17 MrIbby joined #minetest-dev
14:28 ElectronLibre joined #minetest-dev
14:32 Taoki joined #minetest-dev
14:40 hmmmm joined #minetest-dev
14:51 red-001 joined #minetest-dev
15:12 HonoredGlory joined #minetest-dev
15:16 KaadmY joined #minetest-dev
16:19 Fixer do we still have OGL 2.1 shaders support for minetest?
16:21 Fixer Decided to run minetest on laptop from 2008 with OGL 2.1 and now it does not render blocks with shaders enabled, I'm not sure if it was working earlier either
16:21 Fixer irrlicht gives me two errors:
16:21 Fixer Irrlicht: GL_INVALID_OPERATION
16:21 Fixer Irrlicht: Could not bind Texture
16:21 Fixer Intel Integrated
16:24 Fixer Minetest 0.4.14-09cefc3 (32 bit, Win 7 SP1)
16:25 Fixer_ joined #minetest-dev
16:26 Krock I remember that it worked with my old Radeon card (OGL 2.1)
16:28 sfan5 joined #minetest-dev
16:28 Fixer_ that might be another pro-DisableShaders case
16:29 Fixer_ hmm, i will try few older version from sfan5 just for fun
16:29 Krock use MSVC, it has a better performance!
16:29 Krock * MSVC binaries
16:32 Wayward_One joined #minetest-dev
16:35 Fixer_ Krock: that was msvc one
16:47 Krock ah okay
17:13 red-001 how would I fix github seeing a rebase as adding new commits?
17:16 twoelk joined #minetest-dev
17:26 Foghrye4 How do you rebase on a first place, red-001
17:26 Foghrye4 &
17:26 Foghrye4 ?
17:27 xunto joined #minetest-dev
17:28 red-001 there are diffrent ways to rebase?
17:29 Krock yes, but the other method is called reset&stash
17:29 xunto joined #minetest-dev
17:30 red-001 pull+rebase?
17:30 red-001 I think thats what I did
17:30 red-001 I use a gui
17:30 T4im pull == fetch + merge
17:31 T4im you want a fetch + rebase
17:32 Krock Git merge is on Complexity Level 3/6: https://xkcd.com/1667/
17:32 T4im heh
17:32 red-001 I think I use fetch+rebase
17:32 twoelk is pull actually branch specific?
17:32 T4im yea
17:33 T4im either by parameter or config (which in some situation is auto-set into your .git/config)
17:34 red-001 the alt-text is the best part
17:34 red-001 how do I generate a diff from the last commit?
17:36 T4im do you have the sha-id from the last commit before the failed rebase?
17:37 red-001 nope
17:37 Krock I could do it... just with no plans how the commands look like
17:37 Krock :3
17:38 Foghrye4_ joined #minetest-dev
17:39 T4im well you should be able to get it with the help of "git reflog"
17:41 T4im once you do you can use "git reset --hard <id>" to get the local repository to that state
17:41 T4im and local branch
17:48 ElectronLibre joined #minetest-dev
17:48 T4im ah, you rebased upstream master onto your branch, not the other way around?
17:48 T4im that might be easier fixed with just rebasing and removing all those unnecessary commits then
17:49 red-001 ok I got the last good commit
17:49 red-001 what do I do now?
17:50 red-001 reset --hard?
17:50 T4im is it 43ea3605 by any chance?
17:50 T4im eh, or 115d44884d4
17:52 red-001 yeah it is
17:52 T4im then "git reset --hard 115d448" should reset you back to before the failed rebase, yes
17:56 red-001 it worked
17:56 red-001 thanks
18:06 red-001 could someone reopen #4551
18:06 ShadowBot https://github.com/minetest/minetest/issues/4551 -- Trim whitespaces from the input of the `/privs` command. by red-001
18:06 red-001 I seem to have closed it accidently will trying to fix a diffrent issue with git
18:10 red-001 anyone?
18:14 T4im the bigger problem seems to be, that that PR now has no commits
18:15 T4im ah, that's probably because of the close
18:15 T4im nbm
18:15 T4im nvm*
18:16 Krock lol, you seems to have compared minetest:master with your master branch
18:16 Krock *seem
18:16 T4im at the time of the close they seem to have been the same state, yea
18:16 T4im after the close it won't be updated with new changes anymore
18:31 Fixer_ back to my issue, on default config minetest on OGL 2.1 laptop works, issue was with mipmapping (it is disabled by default though)
18:38 Fixer_ please reopen this: https://github.com/minetest/minetest/issues/910
18:40 twoelk isn't it better to open a new one and reference the old one along with the forum page he mentioned?
18:40 Fixer_ twoelk: i've already added that info to 910
18:41 Fixer_ no need to multiply tons of dupes
18:47 red-001 just opened a new pr #4636
18:47 ShadowBot https://github.com/minetest/minetest/issues/4636 -- Trim whitespaces from the input of the `/privs` command. by red-001
18:50 est31 twoelk, for git beginners its an option, but its not really good, because you lose all the discussion
18:58 paramat joined #minetest-dev
18:59 twoelk Fixer mabe you could also add the effect that forcing a font change had to the description
18:59 sfan5 well duh
19:00 sfan5 Fixer_: not sure whether this is the problem, but lots of translations are empty https://github.com/minetest/minetes​t/blob/0.4.14/po/he/minetest.po#L92
19:00 Fixer_ sfan5: could be the reason
19:01 twoelk well he did see something when using the windows default font
19:02 twoelk enough to notice it was ltr
19:04 SloanOnLinux joined #minetest-dev
19:04 SloanOnLinux joined #minetest-dev
19:05 sfan5 well the book is probably missing fonts
19:05 sfan5 language = ja works fine on 0.4.14 on windows
19:05 sfan5 ^ Fixer_, paramat
19:05 Fixer_ i will try it
19:05 sfan5 i just did
19:06 Fixer_ ja works for me
19:06 Fixer_ he is not
19:06 Fixer_ hebrew is not working*
19:07 paramat it looks like gregorycu will not be completing #4530 please could someone complete this? est31 hmmmm highlighting you since you have the most input in the review
19:07 ShadowBot https://github.com/minetest/minetest/issues/4530 -- Fix superflous shader setting updates by gregorycu
19:07 Fixer_ languages with empty text probably -> be, ko, nb, he
19:08 Fixer_ hmmm
19:08 Fixer_ be has text
19:08 sfan5 looks like ko is missing font too
19:08 Fixer_ ko works
19:09 Fixer_ english text
19:09 sfan5 try pressing new
19:09 Fixer_ aha
19:10 Fixer_ why not fallback to english in untranslated places?
19:10 sfan5 nah
19:10 sfan5 it's doing that
19:10 twoelk are tose strings really empty?
19:11 sfan5 Fixer_: "Cancel" is translated: https://github.com/minetest/minetes​t/blob/0.4.14/po/ko/minetest.po#L80
19:11 sfan5 but probably due to fonts the field is empty
19:13 sfan5 aha
19:13 sfan5 if you replace liberationsans.ttf with a hebrew font it works
19:14 twoelk including direction?
19:15 sfan5 dunno
19:17 Fixer_ sfan5: english words with hebrew font? so it is font problem?
19:18 sfan5 Fixer_: see screenshot in #minetest
19:25 sfan5 looks like Noto Sans has support for japanese, russian, korean and hebrew
19:25 sfan5 maybe we can replace droid sans with it
19:28 Foghrye4_ joined #minetest-dev
19:28 est31 http://www.zdnet.com/article/google​s-open-source-noto-free-font-covers​-800-languages-including-dead-ones/
19:28 est31 noto is #1
19:28 ShadowBot https://github.com/minetest/minetest/issues/1 -- GlowStone code by anonymousAwesome
19:29 Fixer_ what I found after crude testing
19:29 Fixer_ he (no text), ko (no text on some buttons), ky (few missings symbols?)
19:31 Foghrye4_ est31, Nerzhul asked on your opinion here: #4611
19:31 ShadowBot https://github.com/minetest/minetest/issues/4611 -- Fix crash on attaching player to entity by Foghrye4
19:49 Darcidride joined #minetest-dev
19:49 Darcidride joined #minetest-dev
19:51 sfan5 hm there's no single font file for noto
19:51 Fixer_ today I did some small performance testing on OpenGL 2.1 laptop, it gives nice perspective about low performance with default config that has shaders enabled
19:51 Fixer_ ^^ results - https://i.imgur.com/g44trWl.png
19:52 sfan5 maybe we can disable shaders based on gpu
19:52 Fixer_ only basic shader was enabled
19:52 sfan5 because i can't observe this with nvidia
19:53 Fixer_ sfan5: i observe big fps drop off on ATI and Intel
19:53 minetestusr joined #minetest-dev
19:53 Fixer_ it halfes fps on both machines :/
19:53 paramat remember vertex tangent issue, huge performance loss for no reason
19:53 Fixer_ paramat: it does not account for all fps loss imo
19:53 Fixer_ paramat: 20% fps loss yes, but not 50-60%
19:54 Fixer_ also, performance went down a little in 0.4.14-dev when basic shader is enabled
19:55 Fixer_ -15% down
19:55 Fixer_ probably because of that tangent spaces thing
19:55 mimetestrtl joined #minetest-dev
19:55 mimetestrtl testing
19:55 Krock success
19:55 mimetestrtl http://pastebin.com/WyCPS3TU <-- this is a code example using ICU if you want to try to add proper RTL support (including in stuff like chat, etc... ) in mimetest
19:56 Fixer_ no shaders performance is muuuch better
19:56 Fixer_ i still think shaders should be disabled by default, even after 4530 merge
19:56 paramat i agree
19:56 twoelk sfan5, why did you close #910 but in #4637 state it is a different issue? is the invisible font issue fixed?
19:57 ShadowBot https://github.com/minetest/minetest/issues/910 -- Russian/Japanese/Chinese localization is terribly broken on Windows
19:57 ShadowBot https://github.com/minetest/minetest/issues/4637 -- Minetest doesn't support RTL languages (Hebrew)
19:57 Fixer_ seems like 4530 does not help much slow machines that have other bottlenecks
19:57 Fixer_ 10% at most
19:58 sfan5 twoelk: japanese works so the description of 910 isn't accurate anymore
19:58 sfan5 the problem with ko or ky are missing fonts
19:58 sfan5 the problem with he (or arabic if we have that) is text direction
19:58 sfan5 910 describes a different bug
20:00 ElectronLibre joined #minetest-dev
20:00 mimetestrtl he has a problem of missing text as well
20:00 twoelk so all hebrew text is displayed in some direction?
20:00 sfan5 hebrew is right to left
20:00 mimetestrtl there are 2 issues
20:00 mimetestrtl a. hebrew is missing by default and b. when changing the font to a hebrew-supported one, it's mirrored (LTR) instead of RTL
20:03 mimetestrtl if somebody wants to try that ubidi code in gettext.cpp i can see if it works ok
20:04 est31 #4638
20:04 ShadowBot https://github.com/minetest/minetest/issues/4638 -- Don't enable translations for languages that are broken
20:07 minetestusr left #minetest-dev
20:08 twoelk shouldn't that rather be "don't ship with broken languages"?
20:09 paramat MT is most likely to be used on a low power device, and low FPS makes a bad first impression, the lack of basic shader effects will not be missed because newbies won't be aware of them, so it's obvious basic shaders should be disabled by default.
20:10 Fixer_ paramat: maybe more people should test that, like actual developers %)
20:11 Fixer_ paramat: i only gave my results on 2 machines with ati and old intel
20:11 est31 twoelk, changed
20:13 Out`Of`Control joined #minetest-dev
20:14 Out`Of`Control left #minetest-dev
20:32 mimetestrtl btw, that language support is not just locale stuff yeah ? it's in the chat system, etc...
20:50 red-001 could someone review #4472  ?
20:50 ShadowBot https://github.com/minetest/minetest/issues/4472 -- Make serverlist searchable. by red-001
20:54 mimetestrtl is there a info on minetest uptake ? also in which C++ is it written ?
20:56 Calinou_ mimetestrtl: it is not C++11/14 yet
20:57 sfan5 generally we use C++03
20:57 mimetestrtl ouch
20:57 sfan5 though we have some code that gets enabled when you compile with C++11 support
20:57 sfan5 e.g. using std::unordered_map over std::map if possible
20:57 paramat T4im and all, can #4614 be merged? there seems to be a discussion over an issue
20:57 ShadowBot https://github.com/minetest/minetest/issues/4614 -- Enable mod security by default by ShadowNinja
20:59 paramat i'll merge 4352 4554 4636 later
21:05 paramat closed #4526
21:05 ShadowBot https://github.com/minetest/minetest/issues/4526 -- Don't print join/leave messages in singleplayer (again) by LNJ2
21:06 paramat ShadowNinja can #4614 be merged?
21:06 ShadowBot https://github.com/minetest/minetest/issues/4614 -- Enable mod security by default by ShadowNinja
21:08 paramat well i'll merge it in 2 hours with the other 3 if no objections
21:52 T4im paramat: from my side, yea
21:54 sofar joined #minetest-dev
22:04 STHGOM joined #minetest-dev
22:05 DI3HARD139 joined #minetest-dev
22:14 paramat ok
22:18 betterthanyou710 joined #minetest-dev
22:40 ElectronLibre joined #minetest-dev
22:44 Fritigern joined #minetest-dev
22:46 Tmanyo joined #minetest-dev
23:21 ElectronLibre joined #minetest-dev
23:23 AntumDeluge joined #minetest-dev
23:30 Grandolf joined #minetest-dev
23:45 sofar gonna try updating the carts stuff
23:47 paramat joined #minetest-dev

| Channels | #minetest-dev index | Today | | Google Search | Plaintext