Minetest logo

IRC log for #minetest-dev, 2017-09-18

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

All times shown according to UTC.

Time Nick Message
00:00 paramat this code was updated recently in a 'clever' but maybe too clever way, maybe that causes unexpected behaviour
00:00 paramat i didn't like the clever code but it was merged
00:01 paramat celeron55 "the ideal solution would be to not switch things between physical and non-physical from lua; it's stupid"
00:01 paramat seems good to me, there can be lots of entities
00:02 paramat if possible that should be kept i feel
00:17 paramat ~tell DTA7 can you confirm that also checking vel.x/z at line 151 of item_entiy.lua causes unwanted behaviour? because that change seems necessary
00:17 ShadowBot paramat: O.K.
00:19 Natechip joined #minetest-dev
00:19 Natechip joined #minetest-dev
00:28 DTA7 joined #minetest-dev
00:38 DTA7 paramat: Later, when !is_physical, the item is stopped, which would then not happen anymore
00:38 DTA7 This is what I meant by "The items keep moving like on a slippery node, but without deceleration in x- and z-direction"
00:39 paramat ok thanks
00:40 paramat this is what i mean by overly clever code :]
00:40 paramat but we can fix the logic at the same time as adding the vel.x/z checks
00:42 asl97 joined #minetest-dev
00:42 paramat the logic has been built around the error in L151 it seems
00:43 DTA7 Then stop it simply if not def or def.walkable?
00:45 DTA7 This would not fix the TNT bug however
00:51 rubenwardy the vector library in the API is so illogical
00:51 rubenwardy how was this merged
00:51 * rubenwardy add fixing it to his to-do list
00:52 DTA7_ joined #minetest-dev
01:02 paramat i can't follow the new clever code, DTA7 can you open an issue about this?
01:03 DTA7 joined #minetest-dev
01:04 paramat i see you added a comment but a new issue would be better
01:05 paramat DTA7 ^
01:06 DTA7 About the TNT bug? Or about the physical property?
01:08 paramat both, perhaps in 1 issue if they're closely related
01:09 DTA7 Will do
01:12 paramat and about L151 vel.x/z checks too, which seem needed
01:29 paramat rubenwardy can we close #6383 ?
01:29 ShadowBot https://github.com/minetest/minetest/issues/6383 -- Remember last selected slot on the hotbar
01:29 rubenwardy there's a surprising amount of support
01:29 rubenwardy I'll comment, maybe leave it while
01:30 paramat bad ideas always get support :]
01:33 thePalindrome Why do you think windows is still used ;)
01:34 rubenwardy paramat, maybe leave it an extra day
01:34 paramat sure
02:00 DTA7 Done
02:01 DTA7 I can make a pull request tomorrow
02:27 loggingbot_ joined #minetest-dev
02:27 Topic for #minetest-dev is now Minetest core development and maintenance. Minetest 0.4.16 released! Chit-chat goes to #minetest. http://irc.minetest.net/minetest-dev/ http://dev.minetest.net/
02:28 sfan5 joined #minetest-dev
05:07 Hijiri this issue has been around for a long time https://github.com/minetest/minetest/issues/2813
05:07 Hijiri is it an irrlicht problem?
05:18 paramat #2813
05:18 ShadowBot https://github.com/minetest/minetest/issues/2813 -- Attachments experience movement delays, triggered randomly
05:18 Hunterz joined #minetest-dev
05:20 Hijiri I think I found a fix
05:20 Hijiri maybe
05:25 paramat i have no idea about the cause of that issue
05:26 Hijiri no, this didn't fix it
05:40 paramat related to #1334 ?
05:40 ShadowBot https://github.com/minetest/minetest/issues/1334 -- Make attached entities move more smoothly
05:43 paramat i'm not sure if that code in 1334 is correct or not
05:45 Hijiri paramat: I don't think so, it looks like it does something with server objects
05:47 paramat ok
06:42 lisac joined #minetest-dev
07:01 paramat sorted out leveled nodebox and added missing documentation #6434
07:01 ShadowBot https://github.com/minetest/minetest/issues/6434 -- Leveled nodebox: Change levels from 1/63rds to 1/64ths by paramat
07:29 Hijiri #5612 is rebased and ready for more reviews by the way
07:29 ShadowBot https://github.com/minetest/minetest/issues/5612 -- Allow overriding tool capabilities through itemstack metadata by raymoo
08:18 Fritigern joined #minetest-dev
08:20 paramat joined #minetest-dev
08:24 proller joined #minetest-dev
08:48 nerzhul joined #minetest-dev
09:02 proller joined #minetest-dev
09:59 paramat will merge #6434 in a few mins
09:59 ShadowBot https://github.com/minetest/minetest/issues/6434 -- Leveled nodebox: Change levels from 1/63rds to 1/64ths by paramat
10:07 paramat merging
10:12 paramat done
11:20 troller joined #minetest-dev
11:28 Fixer joined #minetest-dev
11:36 ThomasMonroe joined #minetest-dev
11:42 CalebDavis joined #minetest-dev
11:45 Megaf joined #minetest-dev
11:58 YuGiOhJCJ joined #minetest-dev
13:01 johnnyjoy joined #minetest-dev
13:10 srifqi joined #minetest-dev
13:31 Megaf joined #minetest-dev
13:34 troller joined #minetest-dev
13:41 CalebDavis joined #minetest-dev
13:51 Fixer hmmm, linux 4.14 has zstd support
13:54 Xio joined #minetest-dev
13:55 antims joined #minetest-dev
14:07 troller joined #minetest-dev
14:12 Wuzzy joined #minetest-dev
14:15 Fixer zstd also changed a license just month ago to better one IMO
14:15 Fixer details here https://github.com/minetest/minetest​/issues/4495#issuecomment-330231496
14:23 ThomasMonroe joined #minetest-dev
14:40 Megaf joined #minetest-dev
14:44 proller joined #minetest-dev
15:29 Hunterz joined #minetest-dev
15:47 proller joined #minetest-dev
16:00 proller joined #minetest-dev
16:03 CalebDavis joined #minetest-dev
16:14 Grandolf joined #minetest-dev
16:20 Player_2 joined #minetest-dev
16:51 Megaf joined #minetest-dev
17:17 Fixer joined #minetest-dev
17:26 nerzhul joined #minetest-dev
17:33 Krock joined #minetest-dev
17:46 proller joined #minetest-dev
18:24 lisac joined #minetest-dev
18:31 Grandolf joined #minetest-dev
18:54 ssieb joined #minetest-dev
18:55 ThomasMonroe joined #minetest-dev
19:13 Xio joined #minetest-dev
19:40 Raven262 joined #minetest-dev
20:03 CalebDavis joined #minetest-dev
20:28 ^v joined #minetest-dev
20:44 domall joined #minetest-dev
20:46 domall left #minetest-dev
21:40 sofar got a protocol implementation question
21:40 sofar when MT sends nodedefs to the client
21:41 sofar does it make a large buffer and put all the nodedefs in there, and then send it letting the socket fragment it as needed?
21:41 sofar or does it make lots of small packets?
21:41 proller joined #minetest-dev
21:43 sfan5 all of them
21:43 sfan5 https://github.com/minetest/minetes​t/blob/master/src/server.cpp#L1488
21:43 sofar how about media?
21:45 sofar hmmm seems it makes 5k packets
21:45 sfan5 it does
21:45 sofar I'm wondering what the impact will be if I make tilesheets for my game
21:45 sofar instead of 200+ 16x16 textures, put them in a 256x256 tilesheet
21:46 sfan5 saves some overhead
21:46 sofar I'd need extra bytes for the texture def, but I'd reduce the number of iterations over media files by a significant amount
21:47 sofar my server is at ~650 media files atm, so it's not trivial
21:49 sofar does the client use openGL smartness to map those tiles?
21:50 sfan5 texture atlas support was dropped many versions ag
21:50 sfan5 o
21:50 sofar euh, so it makes a new 16x16 gl texture for every tile? ouch
21:50 sfan5 if you do want to use a tilesheet the client will need to chop it up into single textures via ^[sheet
21:50 sofar yeah, that was my plan
21:51 sofar will still help loading times, I bet, though
22:06 sofar nore: do you remember the max length of a texture string?
22:07 sfan5 why do you even need to know ಠ_ಠ
22:07 nore sofar: well I guess it can't go over 65536
22:07 sfan5 the absolute maximum would be 2^16
22:08 nore If you start to use a single texture, it will increase quite a lot the length of the texture string for signs
22:08 nore Might even become a limitation
22:09 Miner_48er joined #minetest-dev
22:13 sofar looks like we capped our signs at 1000chars
22:14 sofar just going over formspec user input validation
22:23 sofar we've got some interesting possible exploits with formspecs
22:26 sofar come to think of it, I'm worried about the potential for abuse for most subgames :)
22:35 Beton joined #minetest-dev
22:35 Beton joined #minetest-dev
22:45 DI3HARD139 joined #minetest-dev

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