Time Nick Message 00:16 blert2112 Was a spatialindex problem, disabled it and it went past. Now I'm dealing with cURL issues. Arg! 00:17 VanessaE heh 00:17 VanessaE what are you building on anyway? 00:21 blert2112 Win7 minGW64 00:22 paramat #3171 00:23 ShadowBot https://github.com/minetest/minetest/issues/3171 -- Firelike drawtype: Simplify and improve by paramat 00:24 VanessaE paramat: game#672 00:24 ShadowBot https://github.com/minetest/minetest_game/issues/672 -- water mist by pinkysnowman 00:28 VanessaE paramat: you should give my idea a try regarding plantlike (copy the polys, offset their pitch by some amount without changing their yaw) 00:28 VanessaE (so that each pair forms a "V" shape) 01:06 paramat VanessaE interesting, perhaps for a new drawtype that is used less often. plantlike needs to remain simple though because it is so heavily used in mapgen and affects fps 01:06 VanessaE paramat: that's the idea, yeah 01:07 paramat firelike grass is another option 01:07 VanessaE regular plantlike for most things, plantlike_fancy or some such for others 01:07 VanessaE I'm not a big fan of the square pattern that firelike creates 01:08 VanessaE (besides, the individual faces are too unevenly spaced for grass et al.) 01:08 paramat yeah 01:09 est31 ever heard of settings? 01:09 est31 people with fast boxes can tune them up people with slow boxes leave them the way they are 01:09 est31 everybody is happy 01:10 VanessaE est31: won't work for plantlike drawtype. 01:12 est31 why 01:12 VanessaE because it's used for some stuff that would only look right if the drawtype remains exactly as it is right now (such as table legs, leaves in realtest, spider webs, and other things). :) 01:12 est31 so why is then there discussion at all 01:13 VanessaE well, this proposed "fancy" drawtype would make the plants look thicker and mostly eliminate the problem you mentioned of not being able to see them well from above. 01:13 est31 leaves in realtest are laughable 01:13 est31 they really are 01:13 VanessaE (versus RBA's idea to just extrude) 01:13 RealBadAngel about plantlike to make it more lovely setting param2 on place would do 01:13 est31 super low FPS 01:13 est31 and if you look to the sky, you see the sun 01:14 est31 even in dense forests 01:14 RealBadAngel why do you want to make things harder? 01:14 RealBadAngel ive made that to bring more variety to plants 01:14 est31 what do you mean exactly with copy the polys VanessaE 01:15 est31 there are no polygons for plantlike, currently, no? 01:15 RealBadAngel est31, thats the price, but fps gain is huge 01:15 paramat we could have an option for an alternative drawtype for grass, no problem 01:15 RealBadAngel damn 01:16 RealBadAngel its and option aviable for half a yr already 01:16 VanessaE est31: instead of two faces in an X shape (as seen from above), there would be four faces arranged into two v-shaped pairs, with the pairs still offset in an X shape 01:16 est31 what 01:16 RealBadAngel use it for christ sake 01:16 est31 RealBadAngel, what is an option 01:17 VanessaE the two faces in each V would be tilted away from one another by somewhere around 30 or 40 degrees. 01:18 est31 would look good I guess for small plants, but large plants will still have issues 01:18 VanessaE yeah 01:18 est31 also, we would still have the backface problem 01:18 paramat well as you know i find plantlike fine-rotation is very unsuitable for default mapgen, fine for more specialist uses 01:19 est31 hrmm it seems to be abled to render the backfaces 01:19 est31 so nvm here 01:25 VanessaE est31: I think you're right about realtest's leaves. 01:25 VanessaE I should change those back. 01:32 VanessaE done. 01:40 RealBadAngel est31, when param2 is set to something other than zero plant is rotated 01:41 RealBadAngel 0-179 01:42 RealBadAngel atm game is not using that feature 02:16 paramat hi hmmmm, fractal mapgen is almost ready for merge but i have a bemusing bug described here https://github.com/minetest/minetest/pull/3088#issuecomment-134002097 perhaps sometime you could look at the code? no rush i know you're busy 02:31 paramat back later 02:49 RealBadAngel paramat. with ur change to waiving platlike is more like a grid 02:50 RealBadAngel please do consider again possibly testing rotations to the plantlike 02:54 est31 man trees grow faster than fire spreads 04:45 est31 Tesseract, could you teach the bot to respond to website issues? 04:45 est31 Tesseract, e.g. you say website#11 04:46 est31 Tesseract, and it links to https://github.com/minetest/minetest.github.io/issues/11 05:04 T4im minetest/minetest.github.io#11 05:04 T4im hm, no github style references either :/ 13:35 nrzkt https://www.phoronix.com/scan.php?page=news_item&px=EA-Frostbite-No-Linux 13:35 nrzkt oops :p 13:35 nrzkt wrong chan :) 13:48 Krock you destroyed at least 6 hours and 15 minutes of silence! 14:22 nrzkt :p 15:54 rubenwardy hmmmm: no blocks are loading when I join my server, it's possible this is an effect of minetest.delete_blocks, is there any debugging I can do or should I just restart this server? 17:09 rubenwardy hmmmm: no blocks are loading when I join my server, it's possible this is an effect of minetest.delete_blocks, is there any debugging I can do or should I just restart this server? 17:09 rubenwardy s/it's/is it/ 18:47 hmmmm rubenwardy, I explained why that happens 18:47 hmmmm if you want it to be faster then you need to set sqlite_synchronous = 0 18:48 rubenwardy I'm the only player online 18:48 rubenwardy no mapblocks are loading 18:51 rubenwardy it just looks like a vacuum 18:51 rubenwardy disabling noclip means I can't move 18:52 hmmmm rubenwardy, would you like me to explain again? 18:52 rubenwardy It's not loading slowly, it's not loading at all 18:52 hmmmm yes, I never said that it loads slowly 18:52 hmmmm it isn't loading at all because delete_area is still running 18:53 rubenwardy delete area usually only takes 1 second 18:53 rubenwardy and the server freezes during that time - 18:53 rubenwardy why would it still be running? 18:53 hmmmm delete area takes 1 second for how large of an area? 18:54 rubenwardy 60x60x60 18:54 rubenwardy f5 is logging ~3 packets per line 18:54 hmmmm right now how about how long it takes to delete the size your game uses 18:55 rubenwardy about 1.5 seconds, which is normal 18:55 rubenwardy map is still a vacuum afterwards 18:55 hmmmm well I tested your capture the flag mod and it takes about 40-60 seconds to delete the map 18:55 rubenwardy I use an SSD 18:55 hmmmm after it's done, the mapblocks reload as intended 18:56 hmmmm i am not able to replicate your issue 18:56 rubenwardy and the server uses a smaller arena than the upstream mod 18:56 hmmmm and then I set sqlite to fully asynchronous, which makes delete_area very fast, around 5 seconds 18:56 hmmmm making it work exactly like it's supposed to 18:57 rubenwardy this is the first time it's happened in 3 weeks 18:58 rubenwardy current uptime of 19h 18:59 hmmmm alright 18:59 rubenwardy I'm telling you about it as you asked me to tell you if it happened again - I can just restart the server and carry on as usual 18:59 hmmmm let's assume there's a problem that is not delete_area's inherent slowness 19:00 hmmmm the way I would debug a server freeze is by running minetestserver in a debugger, and then breaking in the debugger when you notice the pause 19:00 hmmmm then looking at a backtrace to see which function the serverthread is in 19:00 rubenwardy the server is running in GDB 19:00 hmmmm ctrl+C breaks it 19:01 est31 note when running with gdb 19:01 est31 make it forward siginfo signals 19:01 rubenwardy this is the script I use: http://digitalaudioconcepts.com/vanessa/hobbies/minetest/Server-scripts/minetestserver-basic.sh 19:01 est31 instead of handling it itself 19:01 est31 because otherwise it kills minetest 19:01 est31 leading in envargsend errors and worse 19:01 rubenwardy and http://digitalaudioconcepts.com/vanessa/hobbies/minetest/Server-scripts/gdbopts 19:02 hmmmm that's the default behavior est 19:02 T4im hmmmm: would that asynchronous setting also speed up clearobjects? 19:02 hmmmm for me, anyway 19:02 est31 hmmmm, for me it isnt 19:02 hmmmm T4im, only a little bit 19:02 hmmmm the thing about clearobjects is that it needs to read, modify, then write back the blocks 19:03 hmmmm the read and modify would still be slow operations because the database abstraction layer uses a synchronous api 19:03 T4im ah, ok, thanks :) 19:03 hmmmm write back would happen instantaneously 19:03 rubenwardy I'm running the server using a bash file which logs its output to a file, rather than to the terminal 19:03 rubenwardy done &>> $MOREDEBUG 19:03 hmmmm what delete_area does is just deletes everything indiscriminantly so it appears to happen instantaneously 19:03 rubenwardy That's a problem as I can't access the gdb terminal 19:04 hmmmm rubenwardy, sounds like the only thing you can do here is just restart the server without redirection 19:04 est31 rubenwardy, store the PID of the gdb in a file 19:04 est31 then kill that 19:04 est31 or send it sigint 19:06 rubenwardy done, now what? 19:06 est31 well it should be killed now 19:07 rubenwardy great 19:07 rubenwardy you can use: gdb -p 7214 19:07 rubenwardy but I've killed it, so I can't now 19:08 rubenwardy :'( 19:16 OldCoder rubenwardy, it will be all right 19:17 OldCoder Mourn the dead process and move on 19:17 OldCoder :P 19:18 Amaz Or, as Beowulf would recommend, avenge it :P 21:20 est31 okay reviewed translations, should work now 21:27 est31 pushed