00:12 < bridge> #embed is awesome 01:12 < bridge> #include 01:12 < bridge> #include 01:12 < bridge> #include 01:12 < bridge> #include "engine/external/json-parser/json.h" 01:12 < bridge> #include "engine/shared/json.h" 01:12 < bridge> #include 01:12 < bridge> 01:12 < bridge> error because gameclient and curl are in conflit or something like, can i use http library in the project to do a Sync request? 01:26 < bridge> you can't really do a sync request from that code 01:26 < bridge> it'd mean the client freezes until then 02:25 < bridge> ``` 02:25 < bridge> 2023-05-14 00:24:35 I assert: /ddnet-source/src/engine/server/register.cpp(517): curl version < 7.77.0 does not support registering via both IPv4 and IPv6, set `sv_register ipv6` or `sv_register ipv4` 02:25 < bridge> Illegal instruction 02:25 < bridge> ``` 02:25 < bridge> what's that 02:25 < bridge> funny curl doesn't allow that since 08:24 <+ChillerDragon> hm how do computers work i get a weird glitchy vertical line while playing teeworlds or any game in windowed or fullscreen for that matter. No cpu core uses more than 10% the ddnet client process is as 14% my gpu is at 70% peek and uses 4% mem for the ddnet client 08:25 <+ChillerDragon> is this possibly a issue with my screen and not with my hardware being slow? 08:25 <+ChillerDragon> i have 400 fps and it feels smooth but i get this glitchy line which i always thought means the gpu can not keep up 08:27 <+ChillerDragon> eh horizontal line i mean 08:28 <+ChillerDragon> its been running smooth with this setup for years im not sure what to replace i tried downgrading to linux-lts and nvidia-lts that did not help im thinking about buying a new screen but maybe im stupid xd 08:30 <+ChillerDragon> i would be more confident its just the screen if vim didnt start to take 3s to write a file 08:32 < bridge> You mean tearing? 08:35 <+ChillerDragon> lmme duckduckgo that 08:36 <+ChillerDragon> yes looks good 08:36 <+ChillerDragon> but i turned on vsync in tw and it didnt help 08:38 < bridge> Then your driver refuses to allow vsync for some reason 08:38 <+ChillerDragon> fak this driver guy 08:38 < bridge> Without vsync the fix is. Buy higher refresh rate monitor xd 08:38 <+ChillerDragon> i have the pay2win nvidia propriatary malware og driver 08:39 <+ChillerDragon> ok i will buy a monitor 08:39 <+ChillerDragon> weird that this monitor worked fine for almost a decade 08:39 < bridge> Maybe u changed gnome compositor settings? 08:39 <+ChillerDragon> maybe the update did 08:39 <+ChillerDragon> what is compositor 08:39 < bridge> The monitor is not broken 08:39 < bridge> U probably always played with vsync 08:39 <+ChillerDragon> he broke of old age? 08:40 <+ChillerDragon> ok thanks jupsti babe i will buy new monitor that cost more than my graphiccard 08:40 < bridge> Compositor is basically the thing that makes ur windows look like windows xd 08:40 <+ChillerDragon> good financial advice once again 08:40 <+ChillerDragon> i have linux 08:40 <+ChillerDragon> no windows 08:40 <+ChillerDragon> xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxD 08:40 < bridge> Trollodragores 08:41 <+ChillerDragon> ye my windows are broken for sure too 08:41 <+ChillerDragon> https://zillyhuhn.com/cs/.1684046489.png 08:41 <+ChillerDragon> see my vscode windowing all over the place xd 08:42 <+ChillerDragon> and there is no way to make it stop neither reboot resize or reopen 08:42 <+ChillerDragon> but its electron so hey gnome might not be responsible 08:42 < bridge> U use Arch right? 08:42 <+ChillerDragon> yes BTW 08:43 < bridge> Well no hate against it. But ofc u are always a beta tester with it 08:43 <+ChillerDragon> wowo 08:43 <+ChillerDragon> arch shaming 08:43 <+ChillerDragon> yea ik thats why this window also doesnt really bother me 08:43 <+ChillerDragon> but the glitchy does and i assume those important things always relate to the driver and kernel 08:44 <+ChillerDragon> and arch offers the easy switch to the lts driver and kernel which makes you travel back in time 08:44 <+ChillerDragon> but this did not help this time .-. 09:17 < bridge> when u use vk backend & enable vsync, does it mention in F1 console or terminal that it could not enable FIFO mode? 09:17 < bridge> chillerdragon: ^ 09:27 <+ChillerDragon> no 09:27 < bridge> thats weird xd 09:27 <+ChillerDragon> wait its because its working 09:28 < bridge> 🤔 09:28 <+ChillerDragon> ok but thats even worse 09:29 <+ChillerDragon> i play better when i had 8 beers than w vsyncx 09:29 <+ChillerDragon> but it did seem to have fixed the tearing 09:29 < bridge> ok 09:29 <+ChillerDragon> so buying a new screen it is jupsti? 09:30 < bridge> i give u best hack 09:30 < bridge> wait 09:30 <+ChillerDragon> maybe vsync is only broken for opengl or after reboot it worked now idk 09:31 < bridge> in vulkan_backend.cpp search the function GetPresentationMode 09:31 < bridge> before 09:31 < bridge> VKIOMode = g_Config.m_GfxVsync ? VK_PRESENT_MODE_FIFO_KHR : VK_PRESENT_MODE_IMMEDIATE_KHR; 09:31 < bridge> add 09:31 < bridge> VKIOMode = VK_PRESENT_MODE_MAILBOX_KHR; 09:31 < bridge> return; 09:31 <+ChillerDragon> watf 09:31 <+ChillerDragon> hak 09:32 < bridge> but pssst 09:33 <+ChillerDragon> so replace the line i guess 09:33 < bridge> doesnt matter 09:33 <+ChillerDragon> yeye 09:33 <+ChillerDragon> hmm not the best hack 09:33 <+ChillerDragon> still sluggish 09:34 < bridge> it should prevent tearing 09:34 < bridge> while allowing 400fps 09:34 <+ChillerDragon> ah so vsync off 09:34 < bridge> its basically what wayland does 09:34 < bridge> oh yes 09:34 < bridge> well it should be off normally lol 09:34 <+ChillerDragon> hmm 09:34 <+ChillerDragon> vsync off still caps me at 60 09:34 < bridge> nani 09:35 <+ChillerDragon> https://zillyhuhn.com/cs/.1684049711.png 09:35 < bridge> https://cdn.discordapp.com/attachments/293493549758939136/1107209595232997477/image.png 09:35 < bridge> u forgot the return 09:35 <+ChillerDragon> a 09:36 <+ChillerDragon> true? 09:36 <+ChillerDragon> ah ye u sent scren 09:37 <+ChillerDragon> holy shit 09:37 <+ChillerDragon> tthast works 09:37 <+ChillerDragon> why is it not a cfg? 09:38 < bridge> bcs it only works with vulkan 09:38 < bridge> and i am too lazy to add bloat 09:38 <+ChillerDragon> i see 09:38 < bridge> the vsync option should be a dropdown 09:38 <+ChillerDragon> gaming with custom patch it is 09:38 <+ChillerDragon> thanks juspti the gfx guru 09:39 <+ChillerDragon> ok still a bit wonkey 09:39 < bridge> yes 09:39 <+ChillerDragon> i think ill just buy the new screen 09:39 < bridge> it still adds latency 09:40 < bridge> have u tried to use unlimited fps 09:40 < bridge> it doesnt prevent it 09:40 < bridge> but makes it a tiny bit better 09:40 <+ChillerDragon> wot 09:40 < bridge> important also 09:40 < bridge> cl_refresh_rate 0 09:41 <+ChillerDragon> 480 xd 09:41 <+ChillerDragon> wowo 600 fps 09:41 <+ChillerDragon> but tearing back 09:41 <+ChillerDragon> i think i added the 480 limit once to fix tearing 09:42 < bridge> ah yeah u play on 4k screen right? 09:42 <+ChillerDragon> poor mans vsync 09:42 <+ChillerDragon> yes 09:42 < bridge> then rip 09:42 < bridge> 4k and 60fps lmao 09:42 <+ChillerDragon> ikr 09:42 <+ChillerDragon> i gotta buy 4k 144hz 09:42 < bridge> that must be horror gaming combo 09:42 < bridge> more like 4k 240hz 09:42 <+ChillerDragon> its been smooth so far actually 09:42 <+ChillerDragon> i can not affort 4k 240hz 28" 09:42 <+ChillerDragon> afford* 09:42 < bridge> yes u can 09:42 <+ChillerDragon> ok finanical advisor if u say so 09:43 < bridge> xd 09:43 < bridge> 1125 € 09:43 < bridge> that's free af 09:43 <+ChillerDragon> uff 09:43 <+ChillerDragon> i was more thinking 700 09:43 < bridge> but 32'' 09:43 <+ChillerDragon> 32 hmm 09:43 <+ChillerDragon> im used to 28 09:44 <+ChillerDragon> send link 09:44 <+ChillerDragon> i was thinking this https://www.amazon.de/Gigabyte-Monitore-Monitor-M28U-EK-3840X2160/dp/B09778LZGQ 09:44 < bridge> https://www.amazon.de/Samsung-S32BG850NU-UHD-Aufl%C3%B6sung-Reaktionszeit-Bildwiederholrate/dp/B09YV3LYTM/ 09:45 < bridge> urs 144hz, not future proof 09:45 < bridge> invest into my future 09:45 <+ChillerDragon> e wot 09:45 < bridge> if u buy expensive now, i can buy cheap in 3 years 09:45 <+ChillerDragon> 144hz is fine for my eyes 09:45 <+ChillerDragon> i used 60hz forever 09:45 <+ChillerDragon> and it good gaming xd 09:45 < bridge> its much better 09:45 < bridge> but will still see tearing, just less annoying 09:45 <+ChillerDragon> also the screen u sent is broken 09:45 <+ChillerDragon> it has a dent 09:46 < bridge> with 240hz its almost gone 09:46 < bridge> better would be 480hz 09:46 < bridge> invest into 480hz 09:46 <+ChillerDragon> me buying expensive high end stuff makes it cheaper for u in the future? 09:46 <+ChillerDragon> if you say so mr financial pro 09:46 < bridge> yeah the market will shift to it eventually then 09:47 < bridge> if everyone would buy 8k tv, they would get mainstream 09:47 < bridge> but nobody wants it 09:47 <+ChillerDragon> yea but curved is cursed 09:47 < bridge> xd 09:47 < bridge> never used one 09:47 < bridge> bcs i fear the curse 09:47 <+ChillerDragon> i would try emacs before i try curved screen 09:47 < bridge> but for me 27'' it is 09:47 <+ChillerDragon> i gotta get 28" so i can zooz better in fng 09:48 < bridge> u can simply go towards the screen with ur eyes 09:48 < bridge> lifehack 09:48 <+ChillerDragon> hm 09:48 <+ChillerDragon> vr 09:49 <+ChillerDragon> anyone ever tried tw vr gaming? xd 09:50 < bridge> https://www.amazon.de/ASUS-PG27AQN-27-Zoll-esports-Monitor/dp/B0BL82FYV4/ 09:50 < bridge> 09:50 < bridge> my next monitor 09:50 < bridge> 360hz IPS panel 09:50 < bridge> nice 09:50 <+ChillerDragon> uff the price bro 09:50 < bridge> no, i dont have vr setup 09:50 < bridge> u gotta invest, and feel stupid after 2 years 09:50 < bridge> else nothing changes 09:50 <+ChillerDragon> xd 09:51 <+ChillerDragon> progressive screen pioneer 09:51 <+ChillerDragon> its not even 4k .-. 09:51 < bridge> i dont need 4k anyway 09:51 <+ChillerDragon> which helps with fps 09:51 <+ChillerDragon> but still 09:52 < bridge> i dont sit 1 centimeter away from my screen 09:52 <+ChillerDragon> u just said 09:52 < bridge> i prefer more hertz 😄 09:52 <+ChillerDragon> i am big zoozer 09:52 < bridge> yes u need 4k 09:52 < bridge> u need 16k tbh 09:52 <+ChillerDragon> i loose all my fps anyways with -30 zoom on Back in Time 3 09:52 < bridge> then all the 4k tee skins finally have a purpose 09:52 <+ChillerDragon> ye tru 09:53 <+ChillerDragon> but i zoom out 09:53 <+ChillerDragon> not in 09:53 <+ChillerDragon> so i basically could downscale 09:53 < bridge> i should write a compute shader to speedup zooming out rendering 09:53 < bridge> but so much work <.< 09:54 <+ChillerDragon> you mean more fps while zoomed out due to saving on rendering pixels that dont fit on the screen anyways? 09:54 <+ChillerDragon> that be soo amazing 09:54 <+ChillerDragon> i buy u a beer if u do 09:55 < bridge> yeah basically like that, but it would involve writing a custom rasterizer, or ray tracer (not like 3d thing), optimized for 2d 09:55 <+ChillerDragon> very pog 09:55 < bridge> add 20k lines of code, no problem xd 09:55 <+ChillerDragon> i noticed the ddrace pro scene slowly picked up zooming 09:56 <+ChillerDragon> playing with 3 times zoomed out on average and replacing pause with zoom on scroll is such a game changer 09:56 <+ChillerDragon> idk how people ever hammerfly through huuughe parts 09:56 <+ChillerDragon> w out zoom 09:57 < bridge> game sense 09:57 <+ChillerDragon> i can fall 10k tiles get a hammerhit done at the bottom 10/10 09:59 <+ChillerDragon> also when tees fall next to you from far above with zoom you can easily predict a hook on most falling speeds where regular zoominners wouldnt even manage to react at all let alone hit the hook 10:00 <+ChillerDragon> but yea you need the hardware for it 10:01 <+ChillerDragon> you can always try zooming in you will realize how much harder the game gets and thats comparable to how much easier it gets when zoomed out 10:11 <+ChillerDragon> are ger2 servers not in http master? 10:11 <+ChillerDragon> ``curl -s https://master1.ddnet.org/ddnet/15/servers.json | jq . | grep ger2`` 10:11 <+ChillerDragon> https://zillyhuhn.com/cs/.1684051889.png 10:12 <+ChillerDragon> some people reported they were not able to find the full multeasymap server 10:14 <+ChillerDragon> https://zillyhuhn.com/cs/.1684052038.png 10:14 <+ChillerDragon> ah no im stoopid xd 10:31 < bridge> https://www.reddit.com/r/ProgrammerHumor/comments/13gt6co/standagainstfloats/ 10:31 < bridge> https://cdn.discordapp.com/attachments/293493549758939136/1107223560528867368/rlcj5w7gjpza1.png 10:32 < bridge> time for analog computers xd 10:32 < bridge> then AI would be so much better 😄 10:32 < bridge> just woke up 10:33 < bridge> time to code ddnet again 10:33 < bridge> what did i miss 10:33 < bridge> what r u gonna add into ddnet 10:33 < bridge> or is it the same PR 10:33 < bridge> chiller and me discussed our plan to conquer the world 10:33 < bridge> idk im gonna look 10:33 < bridge> and im gonna merge https://github.com/ddnet/ddnet/pull/6560 10:33 < bridge> cuz learath didnt review when he said 10:33 < bridge> im gonna test it first tho xd 10:42 < bridge> oh nice 10:42 < bridge> this fixes the color laser for draggers 10:43 < bridge> !!! Enjoy the most profitable financial market (crypto market ) as you get 100% profit...and you can also make up to $100k or more in 3days send me a private message and ask me HOW on TGhttps://t.me/VincentKadar 10:44 < bridge> !!! Enjoy the most profitable financial market (crypto market ) as you get 100% profit...and you can also make up to $100k or more in 3days send me a private message and ask me HOW on TGhttps://t.me/VincentKadar 11:11 < bridge> Excellent meme, completely correct 11:12 < bridge> I did take a look, I just didn't merge it, thought you wanted to 11:12 < bridge> ah, but u didnt state anywhere u did review it xd 11:12 < bridge> xD 11:12 < bridge> learath busy man bro 11:12 < bridge> yeah 11:12 < bridge> and im taking over ddnet 11:12 < bridge> he has to fight the eternal darkness 11:13 < bridge> and no one cant stop me 11:13 < bridge> Yeah, extremely busy doing things I absolutely detest. So happy to be part of the living 11:13 < bridge> join the undead 11:17 < bridge> bloat merges 11:25 < bridge> @Ryozuki would u say hiding a bug is better than panic? 11:25 < bridge> hiding = do not crash 11:25 < bridge> u mean the pr i merged? 11:25 < bridge> i would say having no ub is better than anythign else 11:25 < bridge> then we can think about that 11:26 < bridge> then we could use dbg_asserts 11:26 < bridge> make pr 11:26 < bridge> instead of hiding the bug 11:26 < bridge> as some c grandpa once said, talk with code not words 11:26 < bridge> or smth like that idk 11:26 < bridge> well seems like all devs here disagree with me anyway 11:26 < bridge> :gigachad: 11:26 < bridge> i dont agree or disagree 11:27 < bridge> im just on a holy crusade against ub 11:27 < bridge> @Jupeyy_Keks review my pr and approve it https://github.com/ddnet/ddnet/pull/6585 11:27 < bridge> then i make a pr with ur assert 11:27 < bridge> if u tell me what uwant 11:28 < bridge> it changes server code 11:28 < bridge> that is like selfkilling 11:28 < bridge> i tested it ^tm 11:30 < bridge> later tonight after learath reviewed it 11:30 < bridge> [ PASSED ] 200 tests. 11:30 < bridge> [ FAILED ] 1 test, listed below: 11:30 < bridge> [ FAILED ] Sql/SingleScore.LoadPlayerData/SQLite, where GetParam() = 0x556ce7ecc060 11:30 < bridge> 11:30 < bridge> 1 FAILED TEST 11:30 < bridge> f 11:30 < bridge> xd 11:32 < bridge> dude this test framework sucks 11:32 < bridge> it doesnt tell me the expected value 11:32 < bridge> say it 11:32 < bridge> rust is better 11:33 < bridge> 😎 11:33 < bridge> Expected equality of these values: 11:33 < bridge> m_pPlayerResult->m_Data.m_Info.m_Time 11:33 < bridge> Which is: -1 11:33 < bridge> 0.0 11:33 < bridge> Which is: 0 11:33 < bridge> [ FAILED ] Sql/SingleScore.LoadPlayerData/SQLite, where GetParam() = 0x5608e0708060 (0 ms) 11:33 < bridge> found 11:33 < bridge> makes sense 11:33 < bridge> cuz it changed i guess 11:36 < bridge> fixed 11:36 < bridge> so never 11:37 < bridge> iks de 11:43 < bridge> What are we disagreeing on? 11:45 < bridge> Is this a followup or does it replace #5960 completely? 11:45 < bridge> https://github.com/ddnet/ddnet/pull/5960 11:45 < bridge> #6497 11:45 < bridge> followup, i just fixed conflicts and the test 11:45 < bridge> https://github.com/ddnet/ddnet/pull/6497 11:46 < bridge> I thought we resolved that conflict. You can refuse to launch without a data dir imo but you shouldn't invoke UB while doing so 11:47 < bridge> well they add this weird patch to allow headless client work without data dir 11:47 <+ChillerDragon> headless pog 11:47 < bridge> xd 11:47 < bridge> It also fixes the issues if your normal client is missing some files 11:47 < bridge> it should hang the client if you are missing a language index 11:48 < bridge> @Learath2 do u want me to take over https://github.com/ddnet/ddnet/pull/5558 11:48 < bridge> anyway, to me this is a if(... == nullptr) patch 11:48 < bridge> 11:48 < bridge> and they are usually a sign of bad design to me 11:48 < bridge> 11:48 < bridge> you can ofc disagree 11:48 < bridge> it shouldn't hang the client if you are missing a language index 11:48 < bridge> take over as in fix conflicts xd 11:49 < bridge> i mean ddnet isnt the pinacle of design anyways 11:49 < bridge> it uses c++ to begin with 11:49 < bridge> :justatest: 11:49 < bridge> xD 11:49 < bridge> i agree, but its more effort to revert such stuff than to design it from the start 11:50 < bridge> write cpp2rust converter and run :justatest: 11:50 < bridge> just make a pr and handle it however u want, but no ub 11:50 < bridge> some day in 3000 years i release ddnet-best-vk-edition 11:50 < bridge> and then i can die 11:51 < bridge> Mh. I forgot about that one. I'll finish it up 11:51 < bridge> good good 11:51 < bridge> get to work slave 11:51 < bridge> Might aswell do something I enjoy for once 11:51 < bridge> nice 11:52 < bridge> https://github.com/ddnet/ddnet/pull/5737 11:52 < bridge> this is such a big pr 11:52 < bridge> and no conflicts 11:52 < bridge> but im 2 lazy to review it 11:52 < bridge> Yeah I 1 million % disagree. Who hates null guards even wtf 11:52 < bridge> i dont know editor much anyway 11:53 < bridge> You should contact Chairn and ask for him to help you get a job in the academia 11:54 < bridge> yeah who doesnt likes to spam a if(... == nullptr) to every single code line in existence just bcs someone from 1000 years agos decided it can be null 11:54 < bridge> yeah 11:54 < bridge> even in rust i try not to use too many Option 11:54 < bridge> i am not saying it never makes sense ofc 11:54 < bridge> to a reasonable degree 11:54 < bridge> but in c++ its too dangerous 11:54 < bridge> a option is safe 11:55 < bridge> yeah but passing it around everywhere just sucks xD 11:55 < bridge> Anyway, if you don't like the null guard there, move it up the callstack. When you don't find a data dir print an error and quit 11:55 < bridge> Or assert 11:55 < bridge> and unwrapping a option is a panic, a null deref a segfault 11:55 < bridge> yes, totally fine with that 11:55 < bridge> there it makes sense 11:56 < bridge> yeah we know how awesomely awesome awesome-rust is, not awesomely enough it still doesn't make awesome fixes more awesome 11:56 < bridge> ok 11:56 < bridge> awesome 11:57 < bridge> awesome 11:57 < bridge> a awesome fix is awesome by def tho 11:59 < bridge> @marmare_314 Do you have time to update https://github.com/ddnet/ddnet/pull/6546 ? 12:01 < bridge> Can one of you rust enthusiasts work on https://github.com/ddnet/ddnet/pull/6279 ? 12:02 < bridge> at work i just use https://github.com/taiki-e/cargo-llvm-cov 12:11 < bridge> ChillerDragon: zoom on mousewheel is goated. 12:14 <+ChillerDragon> nono have you tried? 12:14 <+ChillerDragon> i scroll on mousewheel and zoom on shift+mousewheel its really pog 12:14 <+ChillerDragon> ah i mean i change weapons on regular mousehweel 12:16 <+ChillerDragon> so you can see me doing a emote when i zoom out its very sophisticated 12:20 < bridge> OH 12:20 < bridge> @Robyt3 i think i found out why 12:20 < bridge> its a bug in rust i think 12:20 < bridge> https://github.com/rust-lang/rust/issues/95825 12:20 < bridge> the ci fails on the doc tests 12:21 < bridge> ok nvm 12:21 < bridge> im blind and cant read 12:21 < bridge> it doesnt fail on the doc test 12:21 < bridge> xd 12:21 < bridge> /usr/bin/ld: /home/runner/work/ddnet/ddnet/headless/debug/deps/libddnet_test-5eaca437e040b524.rlib(linereader.cpp.o):(.data.rel+0x20): undefined reference to `__gcov_merge_add' 12:22 < bridge> this looks like a cpp error not rust tho 12:22 <+ChillerDragon> rust fan manages to blame rust bug on c++ 12:25 < bridge> It was caused by adding rust to the build though 12:25 < bridge> i think its missing -lgcov 12:25 < bridge> in the rustflags 12:25 < bridge> ill try 12:25 < bridge> > because the cmake-init-env overwrites outside env variables 12:32 < bridge> @Robyt3 lol asan triggered on ci 12:32 < bridge> https://github.com/ddnet/ddnet/actions/runs/4971879116/jobs/8896710547?pr=6586 12:32 < bridge> is this cuz the branch is old? 12:33 < bridge> lol is this not rebased on master? 12:33 < bridge> this also randomly happened in https://github.com/ddnet/ddnet/pull/6565 12:34 < bridge> i rebased now, without conflicts 12:34 < bridge> ok, so it wasn't my PR making the bug happen more frequently 12:35 < bridge> @Robyt3 if u can take a look at https://github.com/ddnet/ddnet/pull/6585 12:35 < bridge> @fokkonaut this interests u doesnt it 12:36 < bridge> yes 12:37 < bridge> I still need it xd 12:38 < bridge> review it again 12:45 < bridge> @Ryozuki do u want to fix smth? 12:45 < bridge> fix what? 12:45 < bridge> do you need a task? 12:45 < bridge> im trying to fix the ci coverage rn 12:46 < bridge> but if u know something ez tell 12:46 < bridge> its very complex 12:46 < bridge> then no 12:46 < bridge> sad 12:46 < bridge> what is it tho 12:47 < bridge> make mouse movement more accurate 12:47 < bridge> 12:47 < bridge> floats are a problem, e.g. if u have high FPS and add small values to a float it increases the chance of inaccuracy a lot 12:48 < bridge> also find out how on different operating systems the relative mouse movement reporting changes based on resolution and DPI 12:48 < bridge> SDL documentation is shit like always 12:48 < bridge> so we should document ourself 12:48 < bridge> I think a newer SDL version makes some improvements for more accurate mouse movement 12:48 < bridge> well even ms documentation isnt satisfying to me tbh 12:48 < bridge> https://learn.microsoft.com/en-us/windows/win32/api/winuser/ns-winuser-rawmouse 12:49 < bridge> its not clear how i should image the relative mouse movement exactly based on the mouse DPI settings 12:51 < bridge> i mean a mouse with polling rate of 8000 hertz should probably report very slight movements 12:51 < bridge> 12:51 < bridge> how do you even display that in a integer? 12:51 < bridge> 12:51 < bridge> i doubt they thought about 2023 when they designed the API xd 12:52 < bridge> In SDL3: "Mouse events use floating point values for mouse coordinates and relative motion values. You can get sub-pixel motion depending on the platform and display scaling." 12:52 < bridge> ah nice 12:52 < bridge> but why float 12:52 < bridge> why not double 12:52 < bridge> cant trust floats xd 12:52 < bridge> or fixed point floats, even better 12:53 < bridge> floats should be enough for pixel values 12:53 < bridge> how accurate can you even move a mouse? 12:53 < bridge> not even perfection is good enough for ddnet 12:54 < bridge> i dunno, but the sensors have like 23k dpi, and 8000 hertz 12:54 < bridge> i guess its pretty accurate, and i dont trust floats 12:54 < bridge> https://tenor.com/view/what-do-they-mean-random-numbers-gif-10654449 12:55 < bridge> i mean wtf even is sub-pixel in this regard? 12:55 < bridge> do they use the window screen size or what? 12:55 < bridge> i'd have thought relative mouse movement is unrelated to that 12:56 < bridge> meh 12:56 < bridge> it failed again 12:56 < bridge> out of ideas 12:57 < bridge> sad, but thanks for trying 12:57 < bridge> @Jupeyy_Keks Any idea how we get this assertion now? https://github.com/ddnet/ddnet/issues/5143#issuecomment-1546735484 12:58 < bridge> `Found non empty text container with index 681 with 6 quads 'Team 0'` 12:58 < bridge> Doesn't look like a `DeleteTextContainer` is missing 12:59 < bridge> ah nice, do you know when we even display Team 0 12:59 < bridge> scoreboard? 12:59 < bridge> nah 13:00 < bridge> where tf do we display team 0 xD 13:00 < bridge> kill message 13:00 < bridge> i never read this in my entire life in the client (except chat) XD 13:00 < bridge> is LDFLAGS and LFLAGS different? 13:00 < bridge> it's new from the team kill messages 13:00 < bridge> LFLAGS ? never heard 13:00 < bridge> but it uses the existing kill message code, so it should delete the containers 13:00 < bridge> > no, LFLAGS are additional flags for lex (as in lex & yacc). LDFLAGS is 13:00 < bridge> > used when calling the linker. They are both used for implicit rules. 13:01 < bridge> LFLAGS 13:01 < bridge> 13:01 < bridge> Extra flags to give to Lex. 13:01 < bridge> lol 13:01 < bridge> ok then we somehow clear kill msgs 13:01 < bridge> 13:01 < bridge> i guess 13:01 < bridge> without noticing 13:01 < bridge> in worst case memory overflow, sadly kaffeine who could reprod didnt find more 13:02 < bridge> u rewrite ddnet in latex? nice 13:02 < bridge> no 13:02 < bridge> u know what a lexer is right 13:02 < bridge> oh yeah misread xd 13:03 < bridge> lmao, good brain 13:03 < bridge> need more gehirnjogging 13:04 < bridge> what 13:04 < bridge> https://cdn.discordapp.com/attachments/293493549758939136/1107262243772575804/message.txt 13:04 < bridge> lmao 13:05 < bridge> https://futurefive.co.nz/uploads/story/2020/01/21/Braintrain.webp 13:05 < bridge> this 13:05 < bridge> xd 13:05 < bridge> in english its called brain age 13:05 < bridge> it was pretty popular in germany xd 13:05 < bridge> `Team %d` is also used in the scoreboard, but only with `TextEx` 13:05 < bridge> i suspect textex anyway 13:06 < bridge> but do we ever print out team 0? 13:06 < bridge> ah, we don't 13:06 < bridge> so it has to be the kill message text 13:07 < bridge> I enjoyed the tas at agdq, is gud https://www.youtube.com/watch?v=mSFHKAvTGNk&t=1000s 13:08 < bridge> my only thesis on this is: 13:08 < bridge> we run out of cmd space in graphics_threaded while being tabbed out, when background rendering is disabled 13:08 < bridge> 13:08 < bridge> but 1. that shouldn't happen very quickly 13:08 < bridge> 2. i didnt find anything 13:08 < bridge> i could try to reduce the buffer size to check if i can reprod 13:08 < bridge> yeah true, saw that before, really awesome 😄 13:16 < bridge> mh seems like the running out of cmd is relatively safe as expected 13:17 < bridge> where exactly in the kill messages do we print the t0? 13:17 < bridge> 13:17 < bridge> do i need to update my server for that? 13:17 < bridge> i only see my tee and name 13:18 < bridge> it's easiest to just add some server code to keep sending team kill messages 13:18 < bridge> you need to be in a team with at least another tee I think 13:19 < bridge> and how do u do that in team 0 😂 13:19 < bridge> hmmmmmm 13:28 < bridge> asan 13:30 < bridge> wtf, why is it happening so often now 13:45 < bridge> maybe u changed the code layout so its more often triggered^^ 13:52 < bridge> @Ryozuki Can you merge it anyway, so the usage of `os_version_str` is correct again? I'll work on a fix for the ASAN issue next. 13:53 < bridge> ye 13:55 < bridge> @Robyt3 i have an idea 13:55 < bridge> is the crashlog from you? 13:55 < bridge> it's from teini94 13:56 < bridge> can u ask him if he toggles nameplates visisbility or similar 13:57 < bridge> @teini94 ^ can you check this, about the recent crash popup and the `Found non empty text container with index 681 with 6 quads 'Team 0'` error 13:59 < bridge> And if not. Do u still have the broken Auto demo 14:39 < bridge> @Ryozuki thanks for wasting your time on cpp just to make ddnet great, what a sacrifise 15:04 < bridge> big sacrifice 15:48 < bridge> When rework the collisions? hurts to look at 15:48 < bridge> https://cdn.discordapp.com/attachments/293493549758939136/1107303491652812870/image.png 15:49 < bridge> When rework the collisions? hurts to look at (jk) 15:49 < bridge> https://cdn.discordapp.com/attachments/293493549758939136/1107303491652812870/image.png 15:49 < bridge> gonna do my own collision check on my system, hopefully it doesn't require me to write 30+ codes for it 16:17 <+ChillerDragon> omg i just did a typo in my email xd when signing up for a web account i now registered that typoed email lmao anyone else got typo versions of the email registered just in case? xd 16:19 < bridge> no xd 16:21 <+ChillerDragon> ryo you need that your gamertag and irl name is a single typo no way ppl get that right the first time 16:22 <+ChillerDragon> royzuki@yahoo.com 16:23 <+ChillerDragon> gotta singup heinrich5990@gmail.com and tell deen i forgot ssh password 17:36 < bridge> :kek: 17:44 <+ChillerDragon> @heinrich5991 monthly request to get snap into dissector but this time i actually it only ending up in the json any chance you could rename the kind from data to something custom like snap_data? https://github.com/heinrich5991/libtw2/blob/89ec0167e7ad1c6fa60d96788d2005a04aee43d4/gamenet/generate/spec/teeworlds-0.7.5.json#L957 17:44 <+ChillerDragon> need* 17:45 <+ChillerDragon> years of irc i and i still proof read my messages after sending them gg 17:46 < bridge> im reviving edlang again 17:46 < bridge>