00:10 <+bridge> the "fly" gametype: https://github.com/unique-clan/fly or https://github.com/SushiTee/teeworlds/tree/Fly 02:08 <+bridge> when whas /tp x y cganget to /tpxy x y? 02:08 <+bridge> i thought we were clear on the status of the /tp overloads 02:09 <+bridge> when whas /tp x y changed to /tpxy x y? 09:41 <+bridge> morning ddneters 09:41 <+bridge> or ddnutters 09:44 <+bridge> 🔩 09:50 <+bridge> #7765 and #7813 09:50 <+bridge> https://github.com/ddnet/ddnet/issues/7765 09:50 <+bridge> https://github.com/ddnet/ddnet/pull/7813 11:38 <+bridge> @learath2 11:38 <+bridge> https://cdn.discordapp.com/attachments/293493549758939136/1257269125131145266/GRY5kKHbcAAKHzA.png?ex=6683ca9e&is=6682791e&hm=0e456aeff8e94fbf82e26a4a0dfffb0ff01a5e20c771b0143ec549f049aa8c3c& 11:56 <+bridge> If I ever achieve something with C, this is how I want to be depicted for future generations 12:09 <+bridge> is there incoming any updates for phone version of game? 12:11 <+bridge> beautiful 12:43 <+bridge> https://github.com/miniupnp/miniupnp/pull/762 12:43 <+bridge> Soon upnp contributor 12:44 <+bridge> that one extra thread increasing buildtime 100% :poggers2: 12:44 <+bridge> chillerdragon: bet that's not on your list yet!! XD 12:44 <+bridge> yep 13:04 <+bridge> <0xdeen> I got a new server, so it's already more than twice as fast! 13:05 <+bridge> :poggers: !!! 13:11 <+bridge> <0xdeen> @fokkonaut Can we get your 128 player code in DDNet btw? 13:12 <+bridge> Of course. I'm a little worried how to port it smoothly though x) It's not the cleanest solution, but one that works well for me regarding most aspects (like team_states, dummy hammer, etc) 13:12 <+bridge> & 0.7 ofc 13:14 <+bridge> <0xdeen> Considering how many servers we have full at 63 players now, I feel like we should give it a try. For the client I'd just add a scroll wheel in scoreboard for now, no other way to show all well. 13:14 <+bridge> I have no client implementation 13:14 <+bridge> <0xdeen> I know 13:14 <+bridge> ah 13:14 <+bridge> Yea, I can give it a try 13:14 <+bridge> <0xdeen> Thank you! 13:14 <+bridge> I mean, porting it 13:14 <+bridge> np 13:15 <+bridge> For now, possibly only as a draft, I'll not clean up the code and instead port it as I use it. 13:15 <+bridge> Can be cleaned up in the future by me or someone else if everything proves to work 13:18 <+bridge> Then we'll have an overview diff at least to see all affected places 13:20 <+bridge> oh god 13:20 <+bridge> 100 players in multeasymap 13:24 <+bridge> 128* 13:25 <+bridge> the 28 players will be racing the actual map or afking 13:33 <+bridge> There is a 0.7 impl for that. Cherrypicking and adding smooth-scrolling via mousewheel would Work nicely im sure (or Just ask bamcane, who initially implemented that) 13:33 <+bridge> https://github.com/teeworlds-community/teeworlds/pull/94 13:45 <+bridge> can you call yourself software engineer without a degree? 13:45 <+bridge> i always had this doubt, but well names are just names 13:53 <+bridge> Mh, you can call yourself one, but there might be situations where you technically might not be allowed to certify something. No idea if there are such requirements anywhere on earth though 13:54 <+bridge> @learath2 are u a software engineer 13:54 <+bridge> am i one 13:54 <+bridge> what is life 13:54 <+bridge> https://tenor.com/view/cliphy-confused-about-life-feelings-gif-13385960 13:54 <+bridge> xd 13:54 <+bridge> Lol 13:54 <+bridge> random gif 13:55 <+bridge> I'm a computer engineer. Not just software 13:55 <+bridge> oh computer 13:55 <+bridge> where is the baseline in knowledge 13:55 <+bridge> to be called one 13:55 <+bridge> or its only the certificate of graduation 13:55 <+bridge> a self learner cant be called one? 13:56 <+bridge> You are probably only missing the random engineering stuff in terms of knowledge 13:57 <+bridge> I actually do not know. Perhaps there are some legal contexts in which you might not want to call yourself one 13:57 <+bridge> https://edgl.dev/ 13:57 <+bridge> i changed the desc, i remembered this interaction https://discord.com/channels/252358080522747904/293493549758939136/1255848838468730922 xd 13:57 <+bridge> omg it says home 13:57 <+bridge> stupid html 13:58 <+bridge> Like maybe certifying software for safety critical stuff. Idk if spain lets computer people certify stuff, very country dependent 14:00 <+bridge> well i didnt mean it for the certification stuff xd 14:01 <+bridge> how is ur golang work going btw+ 14:01 <+bridge> how is ur golang work going btw 14:03 <+bridge> <0xdeen> @remakepower Are you interested in porting this to DDNet?! 14:13 <+bridge> Eh, I already learned all the golang I was interested in, which made it pretty boring now 14:13 <+bridge> iirc in canada you need a engineering license for you to be an engineer. 14:14 <+bridge> I'm so not made for work. I see it as a learning opportunity. Then I get bored when I'm done ingesting information 14:14 <+bridge> startup company when 14:22 <+bridge> What a number, Heinrich Used His vacationMoney for good i See :gigachad: 14:58 <+bridge> I guess we should move towards `std::function` maybe? 15:03 <+bridge> U First :poggers2: 15:06 <+bridge> at some point, we could probably move to some standardized localization library so we don't have to write tools ourselves 15:06 <+bridge> check out https://temporal.io for interesting tech. 15:07 <+bridge> ask the person who wrote the changelog to include it. probably too late for you now 15:11 <+bridge> new blog post https://edgl.dev/blog/rust-fn-size-trick/ 15:12 <+bridge> i mentioned this before but i figured i would put it in a post 15:16 <+bridge> i get screen tearing in the ddnet client even with vsync on; ideas? 15:17 <+bridge> linux or windows? 15:17 <+bridge> vulkan? 15:17 <+bridge> linux vulkan 15:17 <+bridge> wayland? 15:17 <+bridge> x, nvidia 15:17 <+bridge> odd xd 15:17 <+bridge> what does gfx_asyncrender_old say 15:17 <+bridge> it should be 1 15:17 <+bridge> i got a hunch wayland will fix it 15:18 <+bridge> it is 1 15:18 <+bridge> I also spell it, but some people say sequel ^^ 15:18 <+bridge> gfx_render_thread_count 15:18 <+bridge> oh is this new 15:18 <+bridge> it says 3 for me 15:18 <+bridge> for me too 15:18 <+bridge> @jupeyy_keks are u alive? any idea 15:19 <+bridge> @timakro do u use a compositor 15:19 <+bridge> does it have like some vsync setting 15:19 <+bridge> i don't know i just use gnome 15:19 <+bridge> > What fixed for me any type of VSync issues in Gnome shell was this extension: 15:19 <+bridge> > 15:19 <+bridge> > https://extensions.gnome.org/extension/1873/disable-unredirect-fullscreen-windows/ 15:19 <+bridge> > 15:19 <+bridge> > https://github.com/kazysmaster/gnome-shell-extension-disable-unredirect 15:19 <+bridge> > 15:19 <+bridge> > The description may be misleading, but since installing it I no longer experience Vsync issues in Firefox and during gaming. Try it, maybe it will help you. It is easy to disable it so I do not think it will cause issues for you. 15:19 <+bridge> from reddit 15:19 <+bridge> It'll do the same thing, with potentially more overhead since it's even less specific 15:20 <+bridge> wayland may fix it too 15:20 <+bridge> also with vsync off my gpu fans make audible noise 15:20 <+bridge> u can limit fps in ddnet 15:20 <+bridge> i limit to 1000 15:20 <+bridge> there is a limit for cpu and gpu 15:20 <+bridge> gfx refresh rate 15:20 <+bridge> and cl_refresh rate 15:21 <+bridge> https://www.phoronix.com/news/RegreSSHion-CVE-2024-6387 15:21 <+bridge> new ssh vulnerability! 15:21 <+bridge> published today 15:21 <+bridge> > A signal handler race condition within the OpenSSH server could lead to unauthenticated remote code execution. Various versions of OpenSSH going back years are affected on Linux. 15:22 <+bridge> > 15:22 <+bridge> > OpenSSH versions earlier than 4.4p1 are vulnerable to this signal handler race condition unless they are patched for CVE-2006-5051 and CVE-2008-4109. 15:22 <+bridge> > 15:22 <+bridge> > Versions from 4.4p1 up to, but not including, 8.5p1 are not vulnerable due to a transformative patch for CVE-2006-5051, which made a previously unsafe function secure. 15:22 <+bridge> > 15:22 <+bridge> > The vulnerability resurfaces in versions from 8.5p1 up to, but not including, 9.8p1 due to the accidental removal of a critical component in a function. 15:22 <+bridge> ChillerDragon: thanks that you reported an issue as well, that helped not forgetting it 15:23 <+bridge> i think i just turn vsync off and set fps limit to 900 and tearing will be so small that it's not visible 15:24 <+bridge> I guess the proper solution would be to inspect the executables to see what dlls they rely on. mind creating an issue? 15:25 <+bridge> As soon as I find the bar in this airport 15:26 <+bridge> maybe 0? because the first `a` is of type `bool`, but the assignment returns `()`, so the second `a` is of type `()` 15:27 <+bridge> oh its rewind time 15:28 <+bridge> this seems more of an issue with our ping calculation and less of an issue of needing to inform all game servers of our player's IP address. we shouldn't save ping values that are just spikes 15:29 <+bridge> ChillerDragon: cool! where do I find it? 15:29 <+bridge> ah, you posted it a few lines later 15:36 <+bridge> @learath2 where is the discord forum at? :kek: 15:37 <+bridge> When I get back. I had to travel 15:37 <+bridge> @learath2 can u get some reimburstment for delays cuz meteorology? 15:37 <+bridge> on flights 15:39 <+bridge> Not for meteorology. If it wouldn't have been safe to operate the flight they won't compensate you beyond a replacement flight 15:40 <+bridge> If they make scheduling mistakes, they will 15:42 <+bridge> Ow okay 16:15 <+bridge> isn it fat af? 16:15 <+bridge> isnt* 16:16 <+bridge> also how does it solve problem of erasing the type 16:16 <+bridge> It is because it has to assume the worst. And the stl kind does not have a move constructor iirc, so it's double bad 16:17 <+bridge> it should go to gym 16:17 <+bridge> its kinda funny that std::function can be casted to function pointer if held without any context 16:17 <+bridge> but can't just become function itself 16:18 <+bridge> it can know the type of puserdata if we use a enum and a tag right? 16:18 <+bridge> this reminds me of rust enum_dispatch 16:18 <+bridge> oh lord 16:18 <+bridge> why should it know the type xd 16:18 <+bridge> why not 16:18 <+bridge> also why not use IGameServer 16:22 <+bridge> oh, coool! 16:23 <+bridge> soon ill be able to link real C functions and use them in my lang 😏 16:23 <+bridge> but only simple types xd 16:24 <+bridge> nicee 16:24 <+bridge> @milkeeycat does ur lang compile to asm directly without llvm? 16:24 <+bridge> it generates nasm 16:24 <+bridge> i can also call C methods in edlang :owo: 16:24 <+bridge> i just dp use c 16:25 <+bridge> oh 16:25 <+bridge> i just do use c 16:25 <+bridge> why not generate assembly? 16:26 <+bridge> ```rust 16:26 <+bridge> 16:26 <+bridge> pub fn main() -> i64 { 16:26 <+bridge> let a: i64 = 0; 16:26 <+bridge> 16:26 <+bridge> let b: &mut i64 = &mut a; 16:26 <+bridge> *b = 2; 16:26 <+bridge> 16:26 <+bridge> return a; 16:26 <+bridge> } 16:26 <+bridge> 16:26 <+bridge> ``` 16:26 <+bridge> its edlang not rust :justatest: 16:26 <+bridge> ```rust 16:26 <+bridge> 16:26 <+bridge> pub fn main() -> i64 { 16:26 <+bridge> let mut a: i64 = 0; 16:26 <+bridge> 16:26 <+bridge> let b: &mut i64 = &mut a; 16:26 <+bridge> *b = 2; 16:26 <+bridge> 16:26 <+bridge> return a; 16:26 <+bridge> } 16:26 <+bridge> 16:26 <+bridge> ``` 16:27 <+bridge> Are you sure? 16:28 <+bridge> master server entry also works for 0.7 16:29 <+bridge> xd 16:31 <+bridge> ChillerDragon: thanks! 16:32 <+bridge> ``` 16:32 <+bridge> // C 16:32 <+bridge> typedef struct { 16:32 <+bridge> uint8_t one; 16:32 <+bridge> uint8_t two; 16:32 <+bridge> uint8_t three; 16:32 <+bridge> } Foo; 16:32 <+bridge> 16:32 <+bridge> void do_smth(Foo *self, uint16_t param1) { 16:32 <+bridge> // do some important stuff 16:32 <+bridge> } 16:33 <+bridge> 16:33 <+bridge> // My lang 16:33 <+bridge> struct Foo { 16:33 <+bridge> uint8_t one; 16:33 <+bridge> uint8_t two; 16:33 <+bridge> uint8_t three; 16:33 <+bridge> 16:33 <+bridge> // use function from C as a method 16:33 <+bridge> extern "C" void do_smth(uint16_t); 16:33 <+bridge> }; 16:33 <+bridge> 16:33 <+bridge> Foo foo = Foo { 16:33 <+bridge> ... 16:33 <+bridge> ... 16:33 <+bridge> ... 16:33 <+bridge> }; 16:33 <+bridge> foo.do_smth(69420); 16:33 <+bridge> ``` 16:33 <+bridge> i also had an idea to do smth like that but i dunno it that makes sense xd 16:33 <+bridge> ChillerDragon: just noticed that variables.h and config_variables.h have been merged. kind of funny that the TODO to do this was there for 13 years :D 16:37 <+bridge> @heinrich5991 how can u donate 509€ 16:37 <+bridge> For a russian server 16:50 <+bridge> pong 16:51 <+bridge> https://cdn.discordapp.com/attachments/293493549758939136/1257347747028402276/image.png?ex=668413d7&is=6682c257&hm=4cb167a0e9c3caf258ca239caf25c5a114d80bf95f5cf794e0d1715e49e5c065& 16:51 <+bridge> ehm do you get my webchat messages? 16:53 <+bridge> Nop 16:59 <+bridge> wot list xd 17:01 <+bridge> oh wow all my msgs to timakro lost 17:01 <+bridge> i am spamming this channel since 15 hours 17:02 <+bridge> @timakro: it has this projectile hitting bug since 13 years? xd 17:02 <+bridge> Then you can just edit the code to remove projectiles hitting right? 17:02 <+bridge> or do you want to avoid compiling such a old source and work only with the binary? 17:02 <+bridge> reminds me of some story where microsoft lost some source code and binary patched some bug in one of their products or something like that xd 17:03 <+bridge> @solayer: everything is possible some things are more work than others. Why should your client do? I built a py lib covering most of the 0.6 and 0.7 protocol parsing but its not a full client yet 17:03 <+bridge> @timakro: builds fine on my machine :D woah is that battle fly? i thought it was fastcap 17:03 <+bridge> Ayo any chance someone wants to proof read my snapshot delta documentation and check for correctness and possible improvements? @patiga the random issue owner bot has assigned you to this task :p 17:03 <+bridge> https://chillerdragon.github.io/teeworlds-protocol/07/snap_items.html#snapshot_deltas 17:06 <+bridge> @timakro: yea having git lens which shows the commit date really made me wanna open an issue on ddnet finally fixing variables.h xd 17:08 <+bridge> @0xdeen: woah i did not expect you to get interested in 128. That would be so cool! What happend to the collison O(O^N\*420) problems? ddos attack surface? potato vps? I mean F-DDrace is running smooth since years but not sure if it proofs that it will be flawless for ddnet. 17:09 <+ChillerDragon> ping 17:10 <+bridge> pong 17:11 <+bridge> `err_cannotsendtochan` fml 17:13 <+bridge> ok somehow my js irc client became read only xd 17:14 <+bridge> ill be a lurker from now on then 17:37 <+bridge> chillerdragon: athanks 17:44 <+ChillerDragon> @solayer what should your client do? are you still planning to do it? 18:11 <+bridge> ChillerDragon: 18:11 <+bridge> > Those three values are packed integers so their size might be 1 to 4 bytes. 18:11 <+bridge> packed ints can be up to 5 bytes long (not '1 to 4') 18:11 <+bridge> all values in the snap delta are packed integers, so the sentence is a bit misleading 18:11 <+bridge> 18:11 <+bridge> maybe write that you are talking about the network connection. in demos, the deltas apply differently and the keyframes have a differently encoded snapshot 18:11 <+bridge> explicitly say that `size` is the amount of i32/ints 18:11 <+bridge> maybe point out that type_id, id id thrown together in the deleted items, but separate integers in the item diffs 18:12 <+bridge> you could create a visual overview over the entire snap delta, like you did for the single snap item 18:12 <+bridge> some of those might be nits, please take them as suggestions :) 18:20 <+bridge> if you want, you could try to reorder the paragraphs: 18:20 <+bridge> First put into context what a delta/snapshot is, when the server sends it (end of paragraph 'All snapshot items are deltas') 18:20 <+bridge> Then describe the layout of the full delta (start of ('All snapshot items are deltas, 'Snapshot data format') 18:20 <+bridge> Then go over to the snapshot items 18:20 <+bridge> 18:20 <+bridge> (though that might be a bit too much change, I think its also good rn) 18:31 <+bridge> nice! my touch screen will work with ddnet? ^^ (also @_noby) 18:33 <+bridge> could probably be fixed by @learath2, I think he proxies the webhook from github 18:34 <+bridge> I filter some. I'll have to take a look at why it happens 18:35 <+bridge> ChillerDragon: yes! tests are cool, and designing a nice test system is hard. e.g. I don't know how to test UI applications. but it'd be really nice if the ddnet client was also properly tested (e.g. firefox definitely has UI tests) 18:35 <+bridge> egyt: is Hz different in different languages? 18:47 <+bridge> OMGGGGG things get real 18:47 <+bridge> why it's so huge :justatest: 18:47 <+bridge> there more 07 stuff than everything else 18:50 <+bridge> translation os phonetically possible, as some languages like russian don't have the Latin H. but usually with such unit's it's international. 18:50 <+bridge> not sure what it would improve to put the gameservers into containers… they're already a single executable. probably mostly makes debugging more difficult 18:50 <+bridge> is 18:51 <+bridge> linux repositories contain fully open-source binaries of ddnet 18:51 <+bridge> Hz in russian is Гц 18:51 <+bridge> cuz Герц 18:52 <+bridge> Would make our deployment simpler and more reproducible 18:52 <+bridge> H translates to G 18:52 <+bridge> cuz no H 18:52 <+bridge> it should have been be Херц but idk it just got more popular ig 18:52 <+bridge> hmmm. shouldn't it try to install from source then? 18:53 <+bridge> Hamburger classic Gamburger 18:53 <+bridge> Why not Х do you happen to know? 18:53 <+bridge> xd 18:53 <+bridge> I use zfs everywhere. works quite nicely. sometimes I can't upgrade the kernel to the latest version 18:53 <+bridge> different pronouncation did the thing 18:53 <+bridge> I'm now running btrfs, so far so good 18:54 <+bridge> X is more of a Ch 18:54 <+bridge> nop 18:54 <+bridge> or a spanish J of Juan 18:54 <+bridge> Ch sounds way off 18:54 <+bridge> timing mostly, I believe 18:55 <+bridge> the german Ch, not the english Ch. 18:55 <+bridge> wait is Hertz = Görtz 18:55 <+bridge> It's like the "ch" in "Loch" or German "Bach" yeah 18:55 <+bridge> Charakter f.e. 18:56 <+bridge> ChillerDragon: aged like milk :p 18:56 <+bridge> what's correct way to pronounce Heinrich? 18:56 <+bridge> Hi 18:56 <+bridge> Hi-n 18:57 <+bridge> that sounds like a weird thing… probably every language has zealots. C and C++ definitely do 18:57 <+bridge> Heinrich = Генрих 18:58 <+bridge> `%s/foo/bar/c`. the `c` makes it selective 18:58 <+bridge> what's `caw`? I know `ci"` which deletes everything in double quotes 18:58 <+bridge> Assuming SDL touch input works for your platform, I didn't have any luck with Windows 18:59 <+bridge> гаынрищ 18:59 <+bridge> what 18:59 <+bridge> hi-n-risch 18:59 <+bridge> yes 18:59 <+bridge> a is a"round" 18:59 <+bridge> mor or less 18:59 <+bridge> So it also gets the quotes 19:00 <+bridge> juan!! 19:00 <+bridge> I still believe the х is much closer to h than g. G is not even a fricative anymore 19:00 <+bridge> Sounds absurd 19:00 <+bridge> its easier to pronounce g than h as first consonant 19:01 <+bridge> X in catalan sounds like sh 19:01 <+bridge> i think 19:01 <+bridge> Xavi 19:01 <+bridge> oh its more like ch 19:01 <+bridge> Good thing we don't have "x" in Esperanto 19:01 <+bridge> well depends on the position 19:01 <+bridge> the english ch 19:01 <+bridge> ? 19:01 <+bridge> baixa vs Xavi 19:01 <+bridge> xd 19:02 <+bridge> Xavi is said like cha·vee 19:02 <+bridge> k 19:02 <+bridge> and baixa 19:02 <+bridge> is like bye-sha 19:02 <+bridge> it will 19:02 <+bridge> Baixa feels more chinese than catalan 😄 19:02 <+bridge> lol no 19:03 <+bridge> baixa aquí 19:03 <+bridge> what are you doing? sounds interesting ^^ 19:03 <+bridge> its like saying come here 19:03 <+bridge> baixa aqui 19:04 <+bridge> I just noticed my french is godawful now. It was never great but now I sound wrong too 😄 19:04 <+bridge> libffi didnt work cuz i dont know how tou emulate the i128 bit type with 16 byte align 19:04 <+bridge> I need to refrench 19:04 <+bridge> do u know spanish? 19:04 <+bridge> Can't you specify an alignment for a specific struct? 19:04 <+bridge> i always find funny spanish is among the top spoken languages 19:04 <+bridge> I know no spanish at all 19:04 <+bridge> key. will be hard otherwise 19:05 <+bridge> Vamos 19:05 <+bridge> a la playa 19:05 <+bridge> https://cdn.discordapp.com/attachments/293493549758939136/1257381525821853747/image.png?ex=6684334c&is=6682e1cc&hm=2e1f359ad8beac3d770ef6dbc1ec0a44134ecc2882969f359b97453b99dbf055& 19:05 <+bridge> open-source apps are allowed on ios 19:05 <+bridge> despacito 19:06 <+bridge> The spanish kinda conquered the world, killed the natives, planted their own people there and left 19:06 <+bridge> @heinrich5991 https://github.com/fokkonaut/F-DDrace/commit/5fae8676a091ea3237f53d6c6fbcbbaa57453344 19:06 <+bridge> I don't verify the packet currently 19:06 <+bridge> but it cant break anything, it's just a hash being sent 19:06 <+bridge> haha 19:06 <+bridge> its said the spanish once had enough territory that the sun was always up 19:06 <+bridge> or maybe i made that up 19:07 <+bridge> @learath2 lot of south americans joke about spain needing to return the stolen gold xd 19:07 <+bridge> but they dont know most south americans are immigrants from europe 19:07 <+bridge> like argentinians, most have italian parents or grandpas 19:08 <+bridge> No you did not. It was used to refer to the spanish then the british empires at their times 19:08 <+bridge> what do u think about this invention 19:08 <+bridge> https://cdn.discordapp.com/attachments/293493549758939136/1257382431019765781/51hWU14BiNL._AC_SL1468_.jpg?ex=66843424&is=6682e2a4&hm=8e9a1a741fdc5fad63d6c06ddd276c2b6beec6018aeefa58b27a9f3e6f5eae48& 19:09 <+bridge> modern fan but no noise? xd 19:09 <+bridge> It's 😎 19:09 <+bridge> i want to buy one 19:09 <+bridge> Depending on price I might consider it 19:09 <+bridge> found this fro 280€ 19:09 <+bridge> they a bit expensive 19:09 <+bridge> No way on earth am I paying 280 for a fan 19:09 <+bridge> chillerdragon: yea. I (accidentally) took a break from ddnet, but it was good, I think. I'll review this PR if I happen to go through PRs again 19:10 <+bridge> it also filters 19:10 <+bridge> It better 19:10 <+bridge> anyway the key point is its way more silent 19:10 <+bridge> If it also cleans, I might consider it 19:11 <+bridge> this one is rowenta 19:11 <+bridge> but the expensive one is from Dyson 19:11 <+bridge> @learath2 wanna see the funniest shit 19:11 <+bridge> get the 800€ dyson one 19:11 <+bridge> this is a dyson invention 19:12 <+bridge> https://cdn.discordapp.com/attachments/293493549758939136/1257383217086992444/Dyson-zone-1_trans_NvBQzQNjv4BqdUtpIx0in8Ny-w3cMmD2wZrqcnqs16-SvTUWJfMdbUU.webp?ex=668434e0&is=6682e360&hm=2f2024afd23b808a2f824ce451f88afeaceca4f26db99e47c52e92059b3183dc& 19:12 <+bridge> imagine going out with this 19:12 <+bridge> its a air filter 19:12 <+bridge> cyberpunk vibes 19:12 <+bridge> https://edgl.dev/blog/rust-fn-size-trick/ cool blog post. could you maybe link the last example to github? 19:12 <+bridge> I am so enjoying my new laptop. 16 cores and 64g of ram is so cool 19:12 <+bridge> > everything is possible some things are more work than others. 19:12 <+bridge> 19:12 <+bridge> - ChillerDragon 2024 19:12 <+bridge> > everything is possible some things are more work than others. 19:12 <+bridge> 19:12 <+bridge> \- ChillerDragon 2024 19:12 <+bridge> gotta kneel before this godly creature. 19:12 <+bridge> it'll do the same thing, but we could use closures and get type safety 19:12 <+bridge> nice 19:14 <+bridge> it was hard to pay the servers, I donated because I wanted ddnet rus to stay up (like all other servers, too) 19:14 <+bridge> Oh also, neverssl.com is quite useful if you ever find yourself struggling to connect to a hotspot outside 19:15 <+bridge> https://cdn.discordapp.com/attachments/293493549758939136/1257383995801604158/image.png?ex=66843599&is=6682e419&hm=0689dd51eac588a06b373a646b2a037db9532a71a1827c8d1bb61a4935bfdf00& 19:15 <+bridge> my post got some traction kek 19:15 <+bridge> it happens because github is broken. I also get the email twice 19:15 <+bridge> Ah, well my stateless filter can't handle that 19:16 <+bridge> Maybe I can add some state in ram 19:16 <+bridge> nah, always h → г in transcription AFAIK 19:17 <+bridge> yes, russian "х" is pronounced more like german "ch" than "h" AFAIK 19:17 <+bridge> You are a legend mate 19:17 <+bridge> гаинрих? 19:18 <+bridge> you know russian? 19:18 <+bridge> sound-wise this is close imo 19:18 <+bridge> Dynamic type safety is pretty meh. I don't think this has ever been an issue, it's very common to pass a void* for context and get it back 19:19 <+bridge> no need for state. just block the github-merge-queue updates. these are the duplicates 19:20 <+bridge> If you have a minute to add my new ssh key I can take a look right now 19:20 <+bridge> хаынрищ 19:20 <+bridge> even closer 19:21 <+bridge> you pronounce heinrich differently from how I do it ^^ are you german? 19:21 <+bridge> <0xdeen> the initial ping can also be arbitrarily wrong 19:21 <+bridge> I am German indeed 19:21 <+bridge> sure. what SSH key should I give root access to ddnet servers? 19:22 <+bridge> I see how that might be sus 😄 19:22 <+bridge> https://github.com/learath2.keys the very bottom one 19:22 <+bridge> <0xdeen> I'd be interested in running it the most powerful servers we have 19:23 <+bridge> oh TIL https://github.com/edg-l.keys 19:23 <+bridge> wdym initial ping? currently, the issue is that we use a sample of 1 to set the ping. if the internet has a bad moment at the last ping we tried, then we'll save that. what we could instead do would be use an average and maybe even stop taking that into account after some time 19:24 <+bridge> <0xdeen> The estimation of distance we show when you never connected to a server before 19:26 <+bridge> Yeah that one can be wrong since the geolocation db isn't perfect 19:26 <+bridge> looks like deen prefers rsa 19:26 <+bridge> heinrich too 19:26 <+bridge> I should move to ec keys 19:26 <+bridge> they're much shorter 19:26 <+bridge> robyt uses rsa too 19:26 <+bridge> yeah 19:27 <+bridge> old defaults, I guess 19:27 <+bridge> I guess while you are connected to the server, we can sample the ping every once in a while, and keep a couple dozen samples, discarding outliers at a speed proportional to their distance to the mean or sth 19:27 <+bridge> <0xdeen> More like EC was not available when I set up my keys 19:27 <+bridge> https://github.com/ChillerDragon.keys 19:27 <+bridge> as expected 19:27 <+bridge> chiller dragon uses rsa, and has a fuck ton keys XD 19:27 <+bridge> ah. I guess we should update the geoip db. we're using a years-old version 19:28 <+bridge> <0xdeen> IPs keep changing owners faster and faster, I think this is impossible to get perfect 19:28 <+bridge> <0xdeen> But we still should, where is it running? 😄 19:28 <+bridge> all literature i have met use "Генрих" 19:28 <+bridge> pronounces the same way as written 19:28 <+bridge> can you also document a way to add overrides? I spotted turkey in USA once but didn't know how to fix 19:29 <+bridge> ah, didn't know its official transcription. but that is not how it's pronounced in german 19:30 <+bridge> ye, jxsl posted right pronounciation there somewhere 19:30 <+bridge> хаынрищ 19:30 <+bridge> ye, jxsl posted right pronunciation there somewhere 19:30 <+bridge> there's `/home/httpmaster/locations.csv` which it currently uses. it's a very bad format. I think I should update it to use a better format, then we can also get easier updates 19:31 <+bridge> coming from north western germany 19:31 <+bridge> my north-west german interpretation most definitely doesn't have a щ at the end. but a х ^^ 19:31 <+bridge> ah, nvm. north east 19:32 <+bridge> I'll code it, I feel like it right now 19:32 <+bridge> it's rust \o/ 19:36 <+bridge> <0xdeen> @.insanity_ Thanks a lot! Since POL is currently not running I used it for the other servers, hope that's fine 19:39 <+bridge> @0xdeen Do I need to fix the tests? 19:39 <+bridge> Build runs through 19:39 <+bridge> Hm, are you sure? I see someone also added a direct discord webhook to the ddnet org, maybe that is causing the doubles? 19:40 <+bridge> that's what I get in emails, anyway. always "deen closed the issue", then "github-merge-queue closed the issue" 19:40 <+bridge> dunno why this is broken for months 19:41 <+bridge> <.insanity_> ik dw 19:41 <+bridge> <0xdeen> yes, please 19:42 <+bridge> Can't see why it fails 19:42 <+bridge> ig clang tidy 19:42 <+bridge> Ah, I see. Pull requests are not going through filter-events. We get those directly, only branch creates and pushes I handled by hand. Idk what the issue was that we fixed with it 19:42 <+bridge> also could you change all `ClientID` back to `ClientId`? 19:43 <+bridge> there's an attached log with a backtrace 19:43 <+bridge> bruh, this is just for testing purpose 19:43 <+bridge> <0xdeen> @fokkonaut seems pretty clear: 19:43 <+bridge> <0xdeen> ``` 19:43 <+bridge> <0xdeen> thread 'main' panicked at src/engine/console.rs:63:1: 19:43 <+bridge> <0xdeen> assertion failed: executed == 67 19:43 <+bridge> <0xdeen> ``` 19:43 <+bridge> rust? 19:43 <+bridge> idk 19:43 <+bridge> never used that 19:43 <+bridge> it's like any other language 19:43 <+bridge> where do i find that test? 19:43 <+bridge> <0xdeen> I'll try to fix it 19:44 <+bridge> didnt work with ddnet base in a while 19:44 <+bridge> the output contains the line 19:44 <+bridge> and the file 19:44 <+bridge> hmm. but it seems to be wrong (the line) 19:44 <+bridge> XD 19:44 <+bridge> but I find a 67 in the file 19:45 <+bridge> Thanks 19:45 <+bridge> the issue seems to be that `command all` isn't executed 64 times anymore, but 128 times 19:46 <+bridge> so you probably have to add 64 to that number 19:46 <+bridge> (btw, you can run the tests locally using `cmake --build build --target run_tests`) 19:46 <+bridge> ah 19:46 <+bridge> no 128 would be correct 19:47 <+bridge> the expected output should change 19:47 <+bridge> yes 19:47 <+bridge> that's what I mean 19:47 <+bridge> cool, ty, do you know if it works from VS? 19:47 <+bridge> VS code or VS? 19:47 <+bridge> i feel like a boomer still using that shit 19:47 <+bridge> VS :( 19:47 <+bridge> lmao 19:48 <+bridge> make me move to vs code in the future pls 19:48 <+bridge> hmmm. not sure tbh. is there a `run_tests` target somewhere? 19:48 <+bridge> @ryozuki !remindme 100days 19:48 <+bridge> never seen that, but i'll check it out later 19:48 <+bridge> (generally, I like msvc better than visual studio code) 19:48 <+bridge> Hm, it did not happen here, maybe it is fixed? :thonk: 19:48 <+bridge> u mean vs? 19:49 <+bridge> yes 19:49 <+bridge> vs code is getting worse with every new release 19:49 <+bridge> Hmm 19:49 <+bridge> I'm pretty unsure about what IDE to main 19:49 <+bridge> just keep on waiting a little bit longer and you might join the new hyped editor 19:49 <+bridge> i recommend a good text editor + terminal 19:50 <+bridge> not really feasible on windows, I think 19:50 <+bridge> u wont believe me, but im using windows terminal (not cmd) 19:50 <+bridge> sublime text and msys2 are godsent 19:50 <+bridge> I'm using vs code but the last release introduced quite a few crash bugs that I have at least triggered myself 19:50 <+bridge> If it has to be an IDE I keep hearing good things about CLion from people that are not stupid 19:51 <+bridge> yea that one seems to be good 19:51 <+bridge> sublime text was great 19:51 <+bridge> i use vscode and neovim 19:51 <+bridge> no issues 19:51 <+bridge> the best IDE is linux 19:51 <+bridge> rg, fd, cargo, vim 19:51 <+bridge> vscode has issues with blocking your cargo build IIRC 19:51 <+bridge> If you are okay with the terminal being part of your workflow just any text editor with LSP support is great 19:51 <+bridge> got caught by clion indexing ddnet codebase for 15 minutes, uninstalled after 19:51 <+bridge> is what I saw from people using vscode 19:51 <+bridge> lmao 19:51 <+bridge> lol 19:51 <+bridge> i guess clion doesnt use clangd? 19:51 <+bridge> clangd is p fast 19:51 <+bridge> it does 19:52 <+bridge> Yeah, I just gave up on these kinds of things. I'm very comfortable with my terminal tools and just add stuff to my neovim as I need it 19:52 <+bridge> +1 19:52 <+bridge> im trying on emacs but emacs is too powerful to be ide 19:52 <+bridge> alway wanted to try neovim 19:53 <+bridge> but the pain is not painful enough 19:53 <+bridge> man, that convo escalated really quick 19:53 <+bridge> https://developercommunity.visualstudio.com/t/cmake-project-missing-option-to-run-custom-targets/442367 19:53 <+bridge> can you check whether there's a "target view" in solution explorer? 19:54 <+bridge> it died as fast as it escalated 19:54 <+bridge> xD 19:54 <+bridge> in here? 19:54 <+bridge> https://cdn.discordapp.com/attachments/293493549758939136/1257393970741379213/image.png?ex=66843ee3&is=6682ed63&hm=2cf4beb8d6a444bb4a27cc4a225c168332970da1fc0ee1116055388689313f67& 19:54 <+bridge> If you are not already kinda comfy in vim, I really wouldn't suggest it. It's like a guaranteed loss in productivity for the first month or so where you'll fumble around with motions 19:55 <+bridge> Can't seem to find it 19:55 <+bridge> yes. run the `run_tests` or `run_rust_tests` 19:55 <+bridge> @blaiszephyr teach @fokkonaut how you found it 19:55 <+bridge> :kek: 19:55 <+bridge> :D 19:56 <+bridge> @0xdeen Will you host it on GER10? 19:56 <+bridge> I do my occasional server hopping and fixing our monitoring stuff using vim on some aix server in the worst case. But not using it for productive development of stuff. just editing some config files or perl scripts 19:56 <+bridge> If the double github pull request notification happens again, ping me and I'll filter it. I think it was fixed on the github side though 19:56 <+bridge> @fokkonaut 19:56 <+bridge> https://cdn.discordapp.com/attachments/293493549758939136/1257394519230644304/Screenshot_2024-07-01_175640.png?ex=66843f66&is=6682ede6&hm=319f89456959f43ae933993f72dc8e2675bf0b295145646103aafaa3edbfc318& 19:57 <+bridge> https://cdn.discordapp.com/attachments/293493549758939136/1257394578131124355/image.png?ex=66843f74&is=6682edf4&hm=8703c7634778415d280656d74b20564a0959e5c19820244e42211e6c0c3bc31a& 19:57 <+bridge> fixed on the github side? I doubt it. this issue has happened to my email ever since we started using the github merge queue 19:57 <+bridge> what VS version is that 19:57 <+bridge> 2019 19:57 <+bridge> :kek: 19:57 <+bridge> im using 2022 19:57 <+bridge> ._. 19:57 <+bridge> I just recently updated it from a more older 2019 version 19:57 <+bridge> :pepe_holy: 19:57 <+bridge> https://discordapp.com/channels/252358080522747904/293493549758939136/1257392443071004783 19:58 <+bridge> One day I decided to stop updating it, cuz every update broke more features and removed stuff 19:58 <+bridge> but it should still show up in there as `CMake-Zielansicht` 19:58 <+bridge> click on that and it should open up all targets 19:58 <+bridge> <0xdeen> I wanted to try UKR first, then RUS, then GER etc 19:58 <+bridge> I love how this modern web application is so shit that even it's link contraction doesn't work properly 19:58 <+bridge> Alright 19:58 <+bridge> doubleclick on that and it should open up all targets 19:58 <+bridge> <0xdeen> Mainly so more people go to UKR server 😄 19:58 <+bridge> doubleclick on that and it should open up all targets for your Project 19:58 <+bridge> :D 20:00 <+bridge> fokko its about the 128 player support right? 20:00 <+bridge> yea 20:00 <+bridge> lemme run tests for ya - before you loose your sanity 20:00 <+bridge> deen fixed it 20:00 <+bridge> god damn it, i wanted to be useful 20:00 <+bridge> :feelsbadman: 20:01 <+bridge> np, u can port my mod to ddnet base 20:01 <+bridge> 10€ 20:01 <+bridge> offer of ur life 20:01 <+bridge> 20€ and 2 worn pieces of lacoste socks 20:01 <+bridge> not whole socks, just pieces 20:01 <+bridge> <0xdeen> I'll provide another fix 20:01 <+bridge> i only have puma socks, i think 20:01 <+bridge> same - the 20pair packs are insane 20:02 <+bridge> 8€ or smth 20:02 <+bridge> xD 20:02 <+bridge> right 20:02 <+bridge> oh that's why your tests are failing, you are supposed to be wearing programming socks 20:02 <+bridge> You have this enum: 20:02 <+bridge> https://registry.khronos.org/vulkan/specs/1.3-extensions/man/html/VkPresentModeKHR.html 20:02 <+bridge> 20:02 <+bridge> If one of the enum types doesnt work that is a driver thing. 20:02 <+bridge> Vsync is VK\_PRESENT\_MODE\_FIFO\_KHR, ddnet tries that. 20:02 <+bridge> Nothing i can do xd 20:02 <+bridge> 20:02 <+bridge> Also summer holidays for me, so I am more alive than ever 😏 20:02 <+bridge> jupsti is alive! 20:02 <+bridge> reminds me of my co-worker, he has a cup which says "I drink and I know things" :D 20:03 <+bridge> summer holiday already? mine starts on 29th july >:( 20:03 <+bridge> afk 20:04 <+bridge> ok nvm 20:04 <+bridge> too lazy 20:04 <+bridge> i have a cool "might contain bugs" coffee cup 20:04 <+bridge> I dont :( 20:04 <+bridge> I want a teeworlds cup with all y'all tees 20:04 <+bridge> leak address, full name and social security number, i send 20:05 <+bridge> @dariusv someone wants a cup!!!!! 20:05 <+bridge> i live on mars 20:05 <+bridge> ok i tell elon to send one with next shuttle 20:05 <+bridge> lmao 20:05 <+bridge> it wont survive the thrust tho 20:05 <+bridge> too fragile 20:06 <+bridge> We still talked in-game yesterday, you can't miss me already xddd 20:06 <+bridge> afk 20:07 <+bridge> fokko's first name is Elongates 20:07 <+bridge> and his last name is Tusk 20:07 <+bridge> WHAT, no it's obviously elongatees 20:07 <+bridge> sorry. typo 20:07 <+bridge> be forgiven 20:08 <+bridge> much obliged 20:08 <+bridge> you'r (@ chillerdragon: ) welcome 20:08 <+bridge> @fokkonaut i believe now 20:08 <+bridge> https://cdn.discordapp.com/attachments/293493549758939136/1257397540064591983/2024-07-01_18-08-07.mp4?ex=66844236&is=6682f0b6&hm=44ac2cb83efd3f7194aaaad340ed47d0368fff645ed92efdfa8e601946628247& 20:09 <+bridge> pro 20:09 <+bridge> https://github.com/DDNetPP/DDNetPP/blob/3424492d720c695a6dc91194bb666657b313fc7a/src/game/server/ddracechat_ddpp.cpp#L4906 20:09 <+bridge> :dab1: :dab2: :dab3: 20:09 <+bridge> :troll: 20:09 <+bridge> please edit the spaces out 20:10 <+bridge> `spaces` ? - i can make a super small video showcasing that stuff if that's what you mean :D 20:25 <+bridge> Emacs ftw 20:25 <+bridge> @0xdeen can you also start a susterism server or back in time 3 whatever 20:25 <+bridge> Would be nice to see Insane + Novice 20:26 <+bridge> <0xdeen> There seem to be some serious problems still, uninitialized memory, larger allocations for db inserts etc 20:26 <+bridge> <0xdeen> It'd be sad for Insane people to lose their runs because of crashes etc 20:26 <+bridge> true that, maybe Brutal? Stronghold or smth 20:26 <+bridge> thats not too hard/long 20:27 <+bridge> simple down 3 sounds like a fun way to enjoy 128 people 20:27 <+bridge> Ye 20:27 <+bridge> lets go haha, people know its Beta 20:27 <+bridge> hoepfully 20:27 <+bridge> its not really deadly, and i can imagine it'll be really fun 20:28 <+bridge> Yea 20:29 <+bridge> <0xdeen> Done 20:29 <+bridge> nice! 20:30 <+bridge> `2024-07-01 20:29:59 I chat: : (invalid): Do you know someone who uses a bot? Please report them to the moderators.` 20:30 <+bridge> lmao 20:30 <+bridge> <0xdeen> 12% CPU usage of one core with 70 players, so far so good 😉 20:30 <+bridge> <0xdeen> Yeah, we have to fix that too @fokkonaut 20:30 <+bridge> Hehe, completely forgot about that 20:30 <+bridge> glhf! 20:31 <+bridge> <0xdeen> I'll go afk for a while, taking bets on whether the server will crash 😄 20:31 <+bridge> tees sometimes weirdly flash and appear as weapons but other than that it seems to run really smoothly 20:31 <+bridge> im not able to shoot/hammer :D help! 20:31 <+bridge> <:v> https://cdn.discordapp.com/attachments/293493549758939136/1257403210239442954/image.png?ex=6684477e&is=6682f5fe&hm=19840ede0c7e7899ddbbf97a40466386472085546b958c2bcc6d3e73bbcb10ba& 20:31 <+bridge> 20€ dono to a server of your choice if it crashes 20:31 <+bridge> <:v> is that normal ? 20:31 <+bridge> I'll put 5$ on it'll crash 20:32 <+bridge> :) 20:32 <+bridge> thats ddnet code 20:32 <+bridge> ah 20:32 <+bridge> well 20:32 <+bridge> :P 20:32 <+bridge> XD 20:34 <+bridge> @0xdeen I say it won't crash 20:34 <+bridge> but idk ddnet's code nowadays enough in order to tell haha 20:35 <+bridge> But I can see some tees flying around here and there 20:36 <+bridge> never seen that on my srv even with 128 20:36 <+bridge> ye its really weird 20:36 <+bridge> compared to your server 20:36 <+bridge> Hm, I wonder why. Code is more or less same 20:36 <+bridge> fokko is your source even based off of latest ddnet? 20:36 <+bridge> I see hammer and gun entities flying around instead of tees sometimes :monkaS: 20:36 <+bridge> just little differences i think 20:36 <+bridge> no, 0.7 vanilla 20:37 <+bridge> :kek: 20:37 <+bridge> i see 20:37 <+bridge> thats supposed to happen when the idmap is full, so if there is a tee missing in the map that can happen 20:37 <+bridge> I'm unsure whether we want to allow these servers to make ranks 😄 20:37 <+bridge> There might be some exotic bugs there 20:37 <+bridge> probably not 20:37 <+bridge> @0xdeen Can you set sv_mapupdaterate 15? 20:37 <+bridge> it's on 5 in ddnet i think, and my server uses 15 20:38 <+bridge> i can see some maps where over 70 tee's are enough to skip insane amounts :D 20:38 <+bridge> <0xdeen> @fokkonaut Why do tees closeby to me stay weapons? 20:38 <+bridge> @fokkonaut this shouldnt really happen with such a low number of tees next to you no? 20:38 <+bridge> https://cdn.discordapp.com/attachments/293493549758939136/1257405026586857563/screenshot_2024-07-01_20-37-55.png?ex=6684492f&is=6682f7af&hm=023208bbc409af641f9c98b2396480305a6ee0ff19a83bce0d51a9031678bcc7& 20:38 <+bridge> hi 20:38 <+bridge> <0xdeen> sure 20:38 <+bridge> It's possible that there's a bug, try moving your cursor away, and resume 20:38 <+bridge> How can I become an admin? 20:39 <+bridge> deen 20:39 <+bridge> <:v> https://discord.com/channels/252358080522747904/293493549758939136/1257403210428321832 same for me 20:39 <+bridge> Straight to admin dam 20:39 <+bridge> no moderator 20:39 <+bridge> or tester 20:39 <+bridge> <:v> he trys to speedrun ddnet staff 20:39 <+bridge> <0xdeen> done 20:40 <+bridge> thanks, definitely less jumpy 20:40 <+bridge> <0xdeen> Try the 80 player server, they always stay 20:40 <+bridge> i am on ukr 20:40 <+bridge> this is fun - LGTM - ! 20:40 <+bridge> https://cdn.discordapp.com/attachments/293493549758939136/1257405535536283840/2024-07-01_18-34-35-00.04.51.434-00.04.57.827.mp4?ex=668449a9&is=6682f829&hm=8e4dc5b7c223d199b877d96c4d21c7bc2e37644557438dc0c134d9e53928fdea& 20:40 <+bridge> You should ask in #questions next time, but anyway, you need to make an application when we open applications again. Which should be soon 20:41 <+bridge> Yea, there's a bug that can happen, didn't know it happens so frequently. Definitely happens less when tees are more spread apart 20:41 <+bridge> looks so funny XD 20:41 <+bridge> https://cdn.discordapp.com/attachments/293493549758939136/1257405788734095370/screenshot_2024-07-01_20-41-34.png?ex=668449e5&is=6682f865&hm=bbdcf6554bf944dfc48db00f157dc364ef0925b92c09f79ff04848aecc06a29f& 20:41 <+bridge> jupsti isnt here is he? he had that on his impl aswell and got a fix for that iirc 20:41 <+bridge> seeing those walking hammers and guns trying to beat a map 20:42 <+bridge> I think you mentioned that you don't keep track of distance 20:42 <+bridge> Exactly 20:43 <+bridge> it just skips rendering completly, also hook's are rendered as lasers? XD 20:43 <+bridge> When do you decide to stop snapping characters? When the snapshot is full? 20:43 <+bridge> yup 20:43 <+bridge> serverside 20:43 <+bridge> Is it expected that IDs for moderating are wrong 20:44 <+bridge> No, usually only when a new player crosses your networkjclipped viewdistance, he will get added and replaces a tee that is not in your distance 20:44 <+bridge> i seriously thought if you server was running this so well then it will go way easier on ddnet lol 20:44 <+bridge> yes 20:44 <+bridge> so when people are too tight in one spot (according to zoom distance "close"), then that can happen 20:44 <+bridge> zooming in can fix all issues too, i think 20:44 <+bridge> tzooming and a little and moving cursor away 20:44 <+bridge> yeah this is a thing on bw aswell, you cant trust the "show clientID" indicator in scoreboard 20:45 <+bridge> I made a command which makes it possible to "trust" the id 20:45 <+bridge> <0xdeen> Implement 128 players client-side instead, all this playermapping is way too messy 20:45 <+bridge> but since i could only moderate 64 tees then, i reverted it and just use status "name" 20:45 <+bridge> status fok 20:45 <+bridge> and what is the max amount of players you render? 20:45 <+bridge> ofc 64, rest is gun+laser 20:45 <+bridge> can the client even render more than 64? 20:46 <+bridge> no 20:46 <+bridge> yes, if you change it 20:46 <+bridge> not without a change 20:46 <+bridge> xd 20:46 <+bridge> This might not be possible without extensive changes. We might be at the edge of the snaps size anyway 20:46 <+bridge> yeah i mean obviously :D 20:46 <+bridge> I guess we should collect statistics on that ^^ 20:46 <+bridge> I can report no issues on my server, using thousands of objects for plots, free drawing areas, etc 20:46 <+bridge> I don't think we have anything already set up to split snaps up either 20:47 <+bridge> Dropped weapons, dropped guns. You just have to see which things you snap first 20:47 <+bridge> poke chiller - he loves to play around with that 20:47 <+bridge> so the rest can get filled up 20:47 <+bridge> and ddnet is not close to reaching the limit 20:47 <+bridge> but character objects are large, if there is a way to force the server to snap a very different character quickly, and we don't gracefully break up snaps you could dos the server with a couple tees 20:47 <+bridge> deen status on the server load atm? we peak at 83/127 atm 20:48 <+bridge> deen status on the server load atm? we peak at 83/127 20:48 <+bridge> Oh, true I forgot. I was only sending half the clients (64) 20:48 <+bridge> mhh 20:48 <+bridge> <0xdeen> 15% of one CPU core, totally fine 20:48 <+bridge> 90 now - damn 20:48 <+bridge> @fokkonaut what about those skin updates? 20:48 <+bridge> that were stressing the server a lot 20:49 <+bridge> idk, it might be fine, but we definitely need to run the numbers, if 128 full characters don't fit, I think someone should introduce a snap splitting mechanism or enlarge snaps somehow 20:49 <+bridge> with rainbow and stuff 20:49 <+bridge> sad, nobody joining simpel down 20:49 <+bridge> 128 player scoreboard would also "work" without scrolling (imagine twice as many players in this screenshot) :justatest: 20:49 <+bridge> https://cdn.discordapp.com/attachments/293493549758939136/1257407844454760458/screenshot_2024-07-01_20-49-00.png?ex=66844bcf&is=6682fa4f&hm=224af3c892f1a06104530c07e534409aeffdb5d10551acab43142fbb74f0c066& 20:50 <+bridge> i dig it :kek: 20:50 <+bridge> xd 20:50 <+bridge> :PepeCross: 20:51 <+bridge> Also ignore the bundle of stuck debug dummies :kek: 20:51 <+bridge> reminds me of jupsti's screenshot 20:55 <+bridge> im literally having the time of my life with this - can we keep it? 20:55 <+bridge> https://cdn.discordapp.com/attachments/293493549758939136/1257409297743085600/asdasdasdasdasdasdasd.mp4?ex=66844d2a&is=6682fbaa&hm=7f311fff031a168a65f3842fafcd8e8e3daf6b12b84714fc0f9de567dbe43e55& 20:57 <+bridge> a hammer moving around on it's own using lasers as a hook is trippy 21:01 <+bridge> oh nice 21:01 <+bridge> some people started joining impel down 21:02 <+bridge> need more 21:02 <+bridge> im running 4 tees on impel and multeasy 21:05 <+bridge> @0xdeen cant identify the bug rn why it's not exactly like on my srv, but ye, we need client impl anyways 21:06 <+bridge> https://maskray.me/blog/2024-06-30-integrated-assembler-improvements-in-llvm-19 21:12 <+bridge> @0xdeen i pushed a fix 21:13 <+bridge> that's what happens when u introduce new variables and rename things to fit the other source base better 21:13 <+bridge> xd 21:15 <+bridge> @heinrich5991 could you update the servers with the new commit? 21:15 <+bridge> I'll leave that to @0xdeen 21:15 <+bridge> shit 21:15 <+bridge> hes afk 21:16 <+bridge> I'll look at it 21:16 <+bridge> Thanks! 21:16 <+bridge> you can DM me details why this is urgent 21:16 <+bridge> (if it is) 21:16 <+bridge> its a oneliner 21:16 <+bridge> cuz it solves uninitialized garbage memory 21:16 <+bridge> lmao 21:20 <+bridge> <0xdeen> building 21:20 <+bridge> nice! 21:20 <+bridge> phew 21:20 <+bridge> should definitely be better now after that 21:21 <+bridge> now you can probably leave the mapupdaterate at 5, but would be cool at 15 again so i can compare it better 21:26 <+bridge> will you restart the servers? 21:29 <+bridge> wait will fokko's pr be staying there until full 64 -> 128 codebase migration (client support, 128 teams instead of 64 and so on) 21:30 <+bridge> It's mostly an experiment. It can stick around for a bit to evaluate what would need to be done to have full proper 128 player support 21:30 <+bridge> any issues? 21:31 <+bridge> Single tick reload grenade weapon tune can cause much bigger snapshots than 128 players afaik 21:32 <+bridge> It's required to be restarted quickly, as this uninitialized memory makes players jump/appear in wrong positions 21:33 <+bridge> it's like the core breaking point :D 21:33 <+bridge> We just need to know what happens when we go over max snap size and if the client recovers gracefully 21:34 <+bridge> good point 21:34 <+bridge> Also look into if we can just split the snap so there is no recovery needed 😄 21:34 <+bridge> I think not though. We can't just wait for the more packet 21:35 <+bridge> Perhaps a way to ensure the things that didn't make it into the last snap get snapped fully the next 21:35 <+bridge> Is a single snap meant to fit into a single network packet? Where does the size limit come from 21:36 <+bridge> Don't have the code in front of me. I only know there is a limit 21:36 <+bridge> you could increase the limit theoretically 21:36 <+bridge> there's already splitting of snaps into multiple network messages 21:37 <+bridge> Ah nice 21:37 <+bridge> yea, it's just a size change in the client 21:37 <+bridge> but someone should run a calculation first 21:37 <+bridge> What's the concern for making the limit really big? DoS? 21:37 <+bridge> i checked it out in the server, you can increase the size easily, but the client has to have the same change. 21:38 <+bridge> I guess self-DoS of the server 21:38 <+bridge> when it spawns so many entities that it self-DoSes because it sends too many things 21:38 <+bridge> so mapping will be required to support old clients? 21:38 <+bridge> or ddnet will drop old ones :poggers: 21:38 <+bridge> cant support old clients 21:38 <+bridge> at those font sizes you can also use a third column instead 21:38 <+bridge> yes, mapping would be required for old clients 21:38 <+bridge> no 21:39 <+bridge> have u seen my 1000 player scoreboard? 21:39 <+bridge> thats not possible @heinrich5991 21:39 <+bridge> i could still read it on my screen 21:39 <+bridge> wdym 21:39 <+bridge> What's not possible 21:39 <+bridge> this pr actually adds 128 support for old clients 21:39 <+bridge> because theyre 64p clients 21:39 <+bridge> mapping of player IDs is necessary for 64 players. the snapshot size cannot be increased for old clients 21:39 <+bridge> the snapshot is applied though. 21:39 <+bridge> mapping of player IDs is necessary for 64 players on old clients. the snapshot size cannot be increased for old clients 21:39 <+bridge> yes, thats what i am saying 21:40 <+bridge> its not possible to increase it for old clients, like in mapping it somehow 21:40 <+bridge> hey jopsti 21:40 <+bridge> me too. but it seemed like you thought we were disagreeing. thus I said the same thing, but more verbose 21:40 <+bridge> yea, it looked liked it, because zhn seemed to be supporting you 21:40 <+bridge> hey gerdoe, since when did u rename back 21:40 <+bridge> since u said this 21:40 <+bridge> matrix thingie im still zhn 21:40 <+bridge> oh k 21:41 <+bridge> heino u understand ? xd 21:43 <+bridge> i was talking about this though 21:43 <+bridge> https://cdn.discordapp.com/attachments/293493549758939136/1257421277942251614/image.png?ex=66845852&is=668306d2&hm=0396177651086a7a7c51061dab93438bdddb4ae77a743aa7469c298eb33043e3& 21:43 <+bridge> like we all three said same thing idk 21:43 <+bridge> thats only the case if would support old clients 21:44 <+bridge> i said literally what you said xd 21:44 <+bridge> i said literally the same thing you said xd 21:44 <+bridge> thought ur talking about the snaps 21:44 <+bridge> nah :D 21:44 <+bridge> where is deen tho 21:44 <+bridge> :( 21:55 <+bridge> <.insanity_> i didnt even see you there 21:55 <+bridge> rude :( 21:56 <+bridge> <.insanity_> :feelsbadman: 22:16 <+bridge> @learath2 @heinrich5991 could someone of you update the beta servers.... 22:16 <+bridge> Poor foko 22:20 <+bridge> Elongatees :'/ 22:21 <+bridge> given that currently 75 players play on it, I'd prefer not to… 22:21 <+bridge> ITS BROKEN AF 22:21 <+bridge> tf 22:21 <+bridge> the uninitialized memory does garbage 22:22 <+bridge> its the core of everything which is uninitialized here xd 22:22 <+bridge> they'll understand 22:22 <+bridge> and be happy after it 22:24 <+bridge> announce it with a broadcast 22:27 <+bridge> I locked the servers, they'll restart once they empty out 22:30 <+bridge> hah 22:30 <+bridge> i mean it says beta so they were warned xd 22:30 <+bridge> cool, simpel down is restarted i believe 22:31 <+bridge> exactly 22:31 <+bridge> and on multeasy its the best to test rn 22:31 <+bridge> im just curious if any exploit comes out of this 22:31 <+bridge> does it run on the official database? 22:31 <+bridge> i should get ready to hack 22:32 <+bridge> hackerman 22:32 <+bridge> @heinrich5991 does it run the new commit ? 22:32 <+bridge> check /info 22:32 <+bridge> am about to 22:32 <+bridge> cool 22:33 <+bridge> restart the russian multeasy already :D it has 4 players 22:34 <+bridge> https://cdn.discordapp.com/attachments/293493549758939136/1257434238995791944/image.png?ex=66846464&is=668312e4&hm=4ba212528014171a52f9dd5bd6153d0124129a9bd324671f317a33a0a0336dd3& 22:34 <+bridge> stalker 22:35 <+bridge> seems like 128p servers are famous, i should run 512p server 22:35 <+bridge> I think it's mostly the map ^^ 22:35 <+bridge> this could also easily be exceeded to 256 22:36 <+bridge> the server run at 12-15% with 90p 22:36 <+bridge> change 5 vars, done 22:36 <+bridge> depends if the network code scales 22:36 <+bridge> if all players are active 22:36 <+bridge> yea, true 22:36 <+bridge> heinrich, can you check the mapupdaterate and what it's set to currently? 22:37 <+bridge> meh. I don't actually want to live-support this experiment… 22:37 <+bridge> k 22:37 <+bridge> its either 5 or 15 22:38 <+bridge> Some math tells me that we will be barely able to do 512p, let alone FullHD 22:39 <+bridge> FullHD server omg, that is a good name 22:39 <+bridge> will be famous af 22:39 <+bridge> u can make a frame 22:39 <+bridge> with tees 22:39 <+bridge> that would be cool 22:40 <+bridge> Beta UKR players dropping under 30almost 22:40 <+bridge> Well, we don't have 2073600 players, so no FuldHD 22:40 <+bridge> haha true 22:40 <+bridge> Well, we don't have 2073600 players, so no FullHD, unfortunately 22:40 <+bridge> 1080p is fullhd 22:42 <+bridge> 25p left 22:42 <+bridge> "Bad Apple!!" in DDNet? Sounds crazy 22:43 <+bridge> isnt it real to make already 22:43 <+bridge> but 1000 lines patch for 128p support, uff 22:43 <+bridge> 22:43 <+bridge> if ddnet code wouldnt suck it would be exactly changing one config variable instead xd 22:43 <+bridge> ye 22:43 <+bridge> but this is only for supporting old client 22:43 <+bridge> thats why its so much 22:43 <+bridge> never knew that it was client issue 22:44 <+bridge> everything uses MAX\_CLIENTS 22:44 <+bridge> everything are hardcoded arrays in ddnet 22:44 <+bridge> is MAX_PLAYERS defined xd 22:44 <+bridge> it scales in almost no aspect 22:44 <+bridge> also, collision code is quadratic 22:44 <+bridge> yeah 22:44 <+bridge> u can use distance squared 22:44 <+bridge> dd-pg ftw 22:45 <+bridge> wont help 22:45 <+bridge> it introduces physics change, no? 22:45 <+bridge> i send what helps, and i am sure it can be applied to ddnet if carefully done 22:45 <+bridge> Distance squared isn't going to fix the quadratic part 22:45 <+bridge> true 22:45 <+bridge> nope 22:45 <+bridge> well not what makes collision code non-n² 22:45 <+bridge> Quadratic part is that every tee has to test if it collides with any other tee 😄 22:46 <+bridge> ye 22:46 <+bridge> the only edge case is if tees stack 22:46 <+bridge> N tees doing n checks. N^2 :Celebrate: 22:46 <+bridge> but that is an edge case with basically every structure 22:46 <+bridge> worst case\* 22:46 <+bridge> @heinrich5991 can u see if the ukr server is restartable ? its 18 left on lock 22:51 <+bridge> not yet 22:51 <+bridge> not yet ready to restart 22:53 <+bridge> ddnet is not released in the play store? 22:53 <+bridge> The original android build was pretty janky to begin with 22:54 <+bridge> DDNet will likely never be on the Play Store 22:54 <+bridge> D: 22:55 <+bridge> yea the original android build had blatant graphics issues 22:55 <+bridge> anyone could upload it to play store 22:55 <+bridge> Fdroid release when? 22:55 <+bridge> fdroid! 🙂 22:55 <+bridge> why not playstore? 22:55 <+bridge> like when you'd switch out & back in lots of buffers would be full of random shit and they wouldn't be corrected by the game 22:56 <+bridge> graphics buffers i mean 22:56 <+bridge> video mammaries 22:56 <+bridge> the new client is more epyc anyway, bcs it has vk support 22:56 <+bridge> 😏 22:56 <+bridge> i just finished spending 2 hours cleaning up a smoothie that i spilled all over my desk 22:56 <+bridge> and didn't even get any of 22:57 <+bridge> tja 22:57 <+bridge> ama 22:57 <+bridge> Requires too much effort to satisfy the whims of Google's review bots, so I won't be investing any effort into that. Any other store front that allow auto-updating seems better than Google. 22:57 <+bridge> i have/had apps on playstore 22:57 <+bridge> it's not that hard 22:58 <+bridge> just all the questions about user data etc are bit annoying 22:58 <+bridge> TIL 22:58 <+bridge> but it's a onetime thing 22:58 <+bridge> no 22:58 <+bridge> google is pestering you like every year 22:58 <+bridge> they certainly aren't the first thing outsiders think of when they hear about C++ :P 22:58 <+bridge> I see. but someone else could try 22:58 <+bridge> to update your java sdk apps 22:58 <+bridge> well sometimes 22:59 <+bridge> or whatever sdk they provide 22:59 <+bridge> I think you're in a bubble. I know people who don't have that opinion about rust ^^ 22:59 <+bridge> but it's only about the target SDK 22:59 <+bridge> we target the newest android 22:59 <+bridge> so they would not annoy you for a few years 22:59 <+bridge> awesome 😁 implementation issue fixed by romething completely unrelated 22:59 <+bridge> if there is a ci/cd pipeline, I'd guess that's just uploading tue release the 22:59 <+bridge> awesome 😁 implementation issue fixed by something completely unrelated 22:59 <+bridge> but fdroid! 22:59 <+bridge> the, then 23:01 <+bridge> cmon, last 13 players r sleeping 23:01 <+bridge> i might say the same for you. i know more non-rust devs who vocally feel that way about it than not 23:01 <+bridge> lmao fokko never seen you so motivated for smth 23:01 <+bridge> fokko, actively tell them to leave the server for an update :0 23:01 <+bridge> sell them those juicy bugfixes 23:02 <+bridge> lie that there are new features incoming 23:02 <+bridge> I just want the oneliner bug fix to be fixed cuz it fixes all the problems people can have (almost) 23:02 <+bridge> xd 23:02 <+bridge> its literally breaking everything 23:02 <+bridge> UB 23:02 <+bridge> and its a beta server 23:02 <+bridge> dunno why cant just restart 23:02 <+bridge> with 13 player 23:02 <+bridge> cant, its passworded 23:02 <+bridge> xd 23:02 <+bridge> the chat is passworded? 23:03 <+bridge> THE SERVER 23:03 <+bridge> I did not see that when I was there D: 23:03 <+bridge> oh 23:03 <+bridge> it's added so they leave 23:03 <+bridge> Insee 23:03 <+bridge> xd 23:03 <+bridge> why the heck did you leave 23:03 <+bridge> well in this discord for sure. but generally i have to say the programming language doesn't matter. i know lot of maintainers of other projects that are rather ignorant to bug reports & suggestions 23:03 <+bridge> bruh 23:03 <+bridge> but it also doesn't really matter and this sort of thing really just depends on your exposure. can't deny that this perception exests, and is very prominent 23:04 <+bridge> actually this is the discord that i feel hates on rust for it the least 23:04 <+bridge> https://cdn.discordapp.com/attachments/293493549758939136/1257441792778047540/image.png?ex=66846b6d&is=668319ed&hm=26c266812b1631c5df2fa9b0d57b3f8d4c94a74cae73aa6641f2a6781846ffb9& 23:04 <+bridge> most of what i'm thinking about is from a community i was in years ago 23:04 <+bridge> i found that message again 23:04 <+bridge> lul 23:05 <+bridge> you still on that beta server @cyberfighter ? 23:05 <+bridge> ye 23:05 <+bridge> im on some dead one 23:05 <+bridge> sick 23:05 <+bridge> https://cdn.discordapp.com/attachments/293493549758939136/1257442001885073528/image.png?ex=66846b9f&is=66831a1f&hm=92b7adf29fa7033ee58a2418c5249b19a0841e1a8534a51a4ff65d63a02c12c0& 23:05 <+bridge> ah i thought that this server is most rust elitest xD 23:05 <+bridge> tell those people to oeave for a restart 23:05 <+bridge> leave 23:05 <+bridge> so not about hate 23:05 <+bridge> the get new features 23:05 <+bridge> it had 0 already 23:05 <+bridge> amd bigfixes 23:06 <+bridge> :'/ 23:06 <+bridge> my comment about bug reports etc is also more about ppl feel too elitest 23:06 <+bridge> instead of just being pragmatic 23:07 <+bridge> i mean teeworlds is also best.. but damn the code sucks 😬 23:07 <+bridge> that's elitist :0 23:07 <+bridge> it works, hehe 23:07 <+bridge> wdym 23:08 <+bridge> :justatest: 23:08 <+bridge> ever since you drove voxel away there hasnt been much rest elitism backlash 23:09 <+bridge> Rust has lots of evangelists. That's what creates the feeling. You ask a question about C/C++ in any programming space and some dude will pull up like a jehovas witness asking you if you've heard of Rust our lord and savior 23:09 <+bridge> C also has lots 23:09 <+bridge> altho they are a bit older and maybe lurk more in irc 23:09 <+bridge> Jehovah's* 23:10 <+bridge> i think he was not mentaly stable, unrelated to this chat and probs stressed irl 23:10 <+bridge> ever since you drove voxel away there hasnt been much less rust elitism backlash 23:10 <+bridge> yes but you did drive him away 23:10 <+bridge> You won't find our kind leaking outside of our own spaces 23:10 <+bridge> 😂😂😂😂 23:10 <+bridge> never watch a video called rust vs cpp 23:10 <+bridge> amem 23:10 <+bridge> amen 23:10 <+bridge> rust al gaib! 23:11 <+bridge> you found something that genuinely got on his nerves, as benign as it would seem, and continued to push until he had to leave 23:11 <+bridge> not rly 23:11 <+bridge> maybe that community wasn't that cool then, if it hates on random things. things to hate on is a very easy thing to build a group identity around 23:11 <+bridge> yes 23:11 <+bridge> dune reference 👍 23:11 <+bridge> yeah. i left when it became cool to hate on me :^) 23:12 <+bridge> @mpft: when will you learn rust 23:12 <+bridge> i was actually quite friend with voxel, i even paid him a comission for 50€ (way more than what he asks) at some point he begun to spiral down bad, he had a shitty job, etc 23:12 <+bridge> the ddstats logo was the comm 23:13 <+bridge> :owo: 23:13 <+bridge> if u truly felt like he was a friend you would have respected his boundaries 23:13 <+bridge> but making him mad was just so funny right 23:13 <+bridge> well friends come and go, and no i didnt make him mad on purpose rly 23:14 <+bridge> xd 23:14 <+bridge> And even that didn't stop you from destroying them with the rust spam. Truly 😈 23:14 <+bridge> :gigachad: 23:14 <+bridge> tbf voxel was mad at a lot of things xDDD 23:14 <+bridge> i still think its totally not related 23:14 <+bridge> he also always been microsoft best 23:14 <+bridge> but i do miss him 23:14 <+bridge> he was more mad at ddnet dev not being the way he wanted 23:14 <+bridge> Oh yeah, voxel loved microsoft too 23:15 <+bridge> not "progressing" 23:15 <+bridge> reminds me of lynn 23:15 <+bridge> he/she/it got real mad too 23:15 <+bridge> i knew this would come xDDDDDDDD 23:15 <+bridge> lets push it down pls 23:16 <+bridge> you mean put it aside 23:16 <+bridge> is push it down a expression? 23:16 <+bridge> til 23:16 <+bridge> idk actually 23:16 <+bridge> nah but the intent is clear 23:16 <+bridge> I never met Lynn but I heard a lot about them 23:17 <+bridge> he did some commit recently 23:17 <+bridge> or she 23:17 <+bridge> some simple edit 23:17 <+bridge> but complained others did not make what they wanted 23:17 <+bridge> but still alive, epyc 23:17 <+bridge> i remember making some commentary gender-neutral or something 23:17 <+bridge> voxel atleast made some nice changes 23:17 <+bridge> yeah but i am also always mad 23:18 <+bridge> bcs of backward compability 23:18 <+bridge> im also always mad 23:18 <+bridge> same 23:18 <+bridge> at memory unsafety :gigachad: 23:18 <+bridge> and ppl that have no vulkan capable gpu 23:18 <+bridge> pf 23:18 <+bridge> :troll: 23:18 <+bridge> and no avx512 23:18 <+bridge> you literally pointing at me 23:18 <+bridge> pff 23:19 <+bridge> who are you? 23:19 <+bridge> your worst nightmare 23:19 <+bridge> liking microsoft sounds sane to me. 23:19 <+bridge> opengl 1.x gpu user 23:19 <+bridge> my whole day dream 23:19 <+bridge> why u not on discord? 23:19 <+bridge> llvm pipe users 23:19 <+bridge> after my summer holidays 23:19 <+bridge> :justatest: 23:19 <+bridge> i'm back on discord 23:19 <+bridge> https://cdn.discordapp.com/attachments/293493549758939136/1257445589537263646/image.png?ex=66846ef6&is=66831d76&hm=0cfe9fbf3dfb12743a19ec5c1349a6b7149d2a815289038d14ce29cd47407d9a& 23:19 <+bridge> serverbrowser bug. is it known? 23:20 <+bridge> which version do you use? 23:20 <+bridge> after changing the quicksearch thing, it was fixed 23:20 <+bridge> 18.1, I guess I should update 23:20 <+bridge> I thought this was fixed long ago 23:20 <+bridge> if someone will try to fix it please can you also fix width of server name on 5:4 23:20 <+bridge> even longer than 18.1 iirc 23:20 <+bridge> longer ago/before 23:21 <+bridge> do you literally have a 5:4 screen? 23:21 <+bridge> trust me I'm a native English speaker 23:21 <+bridge> :justatest: 23:21 <+bridge> are u a csgo user 23:21 <+bridge> who plays csgo anymore it's all about cs2 23:21 <+bridge> nah i do have 21:9 one but im playing on 1:1, 5:4 mostly 23:21 <+bridge> we really need ingame aspect ratio support. 23:21 <+bridge> 23:21 <+bridge> changing native resolution sounds bit annoying 23:21 <+bridge> better view for vanilla 23:22 <+bridge> "intent was clear" 23:22 <+bridge> what would it do jopsti 23:22 <+bridge> wdym 23:22 <+bridge> it would give u 5:4 resolution but only for ingame rendering 23:22 <+bridge> so UI stays at "normal" resolution 23:22 <+bridge> nothing xD 23:23 <+bridge> would it stretch? can't imagine it would without stupid overhead 23:23 <+bridge> but the monitor does it in real time 23:23 <+bridge> it stretches, no overhead 23:23 <+bridge> good 23:23 <+bridge> it's probably faster than the monitor 23:23 <+bridge> @gerdoe: https://github.com/ddnet/ddnet/pull/7940 23:24 <+bridge> you should test it out :D 23:24 <+bridge> gfx\_ingame\_aspect\_ratio 1250 is what you would need 23:24 <+bridge> wtf 23:24 <+bridge> u already made it?? 23:24 <+bridge> yeah long ago when we had yet another 5:4 ui bug 23:24 <+bridge> yay updated 23:24 <+bridge> that is a rly good idea actually 23:25 <+bridge> jupstar gigachad 23:25 <+bridge> https://github.com/ddnet/ddnet/pull/7940/commits/ea2076a615284132aa32de061b0fe42ff6a74ed9 23:25 <+bridge> drop native 5:4 and implement it ingame 23:25 <+bridge> i mean it's 2 lines of code 23:25 <+bridge> lmao 23:25 <+bridge> jupstar when amd fdr and dlss 23:25 <+bridge> for 100k fps 23:25 <+bridge> lmfao 23:25 <+bridge> fsr 23:25 <+bridge> xd 23:25 <+bridge> zd 23:25 <+bridge> 480p ddnet 23:25 <+bridge> xd 23:25 <+bridge> i assume it's slower than direct rendering for entities 23:25 <+bridge> likely 23:25 <+bridge> bcs the "AI" is not a cheap operation 23:26 <+bridge> like any upscaling/AA it'd need at least a little extra time, no? 23:26 <+bridge> yeah 23:26 <+bridge> @heinrich5991 why no ranks 23:27 <+bridge> I don't trust it yet, and teehistorian hasn't been fixed 23:27 <+bridge> xD 23:27 <+bridge> k 23:27 <+bridge> it'll not crash 23:27 <+bridge> dont worry, didnt even crash with uninitialized memory 23:27 <+bridge> what time is it in Germany rn? do u have multiple timezones? 23:28 <+bridge> no 23:28 <+bridge> 23:28 23:28 <+bridge> ah 23:28 <+bridge> late 23:28 <+bridge> u 23:28 <+bridge> i am watching POR-SVN xdd 23:28 <+bridge> PORN 23:28 <+bridge> SV 23:28 <+bridge> :justatest: 23:28 <+bridge> xd 23:29 <+bridge> BTW just kinda popped in my head, but what (in the code) is stopping ddnet from having >64 player support? 23:29 <+bridge> did spain win 23:29 <+bridge> yeah 23:30 <+bridge> they play against germany next 23:30 <+bridge> 😬 23:30 <+bridge> 64 is like hardcoded everywhere 23:30 <+bridge> fixed arrays everywhere 23:30 <+bridge> fixed size 23:30 <+bridge> ye no dynamic mem 23:31 <+bridge> just do a small vec 23:31 <+bridge> first 64 on stack 23:31 <+bridge> just do a normal vec xd 23:31 <+bridge> then moved to heap 23:31 <+bridge> overengineering fun 23:31 <+bridge> it is sometimes 23:32 <+bridge> who doesnt get mad excitement when abusing traits in rust 23:32 <+bridge> i'd recommend against the tinyset crate 23:32 <+bridge> it's buggy xd 23:32 <+bridge> xd 23:32 <+bridge> wtf? that can't be it 23:32 <+bridge> 1 search and replace 23:33 <+bridge> its also in the protocol iirc 23:33 <+bridge> ah 23:33 <+bridge> if we have no backward comp 23:33 <+bridge> everything would be easy 23:33 <+bridge> suddenly coding would be fun 23:33 <+bridge> and not pure pain 23:33 <+bridge> xddd 23:33 <+bridge> the sun would rise 23:33 <+bridge> flowers bloom 23:34 <+bridge> penalty shooting 23:34 <+bridge> rip 23:34 <+bridge> if we accept no backward comp i promise to be active dev again 23:35 <+bridge> I think then we lose heinrich 23:35 <+bridge> :justatest: 23:35 <+bridge> You two are incompatible 😄 23:35 <+bridge> we should just do 2 separate ddnet projects 23:35 <+bridge> we could truly make ddnet 2 23:35 <+bridge> as voxel wanted 23:36 <+bridge> i am ok with compability between database and maps 23:36 <+bridge> at least for now 23:36 <+bridge> well and physics 23:36 <+bridge> well that is my plan anyway lmao 23:36 <+bridge> We shouldn't. You think it's fun now, but when the novelty of breaking compat wears off we'll have one extra dead fork with hundreds of hours of devtime wasted 23:37 <+bridge> why fork it 23:37 <+bridge> make it main 23:37 <+bridge> :gigachad: 23:37 <+bridge> why do you assume the new version wouldnt be liked more? 23:37 <+bridge> If we had like 2 dozen activeish devs it may work out. But with just a handful of us it'd definitely be a disaster 23:37 <+bridge> we have me 23:38 <+bridge> Either way one of the forks will waste hundreds of hours 23:38 <+bridge> ddnet-cpp wouldnt need to be updated a lot anyway 23:38 <+bridge> just as it is now 23:40 <+bridge> ddnet 2 plan: 23:40 <+bridge> 23:40 <+bridge> - remove backcompat code and monkeypatches 23:40 <+bridge> - remove ogl 1 23:40 <+bridge> - std 23 23:40 <+bridge> - future proof map format 23:40 <+bridge> - maybe more mod friendly 23:40 <+bridge> - rust for protocol codegen? 23:40 <+bridge> - port to postgres, remove mysql 23:40 <+bridge> - add a predictable physics test framework 23:41 <+bridge> - accounts 23:41 <+bridge> :poggers2: 23:41 <+bridge> there is rust std 23? 23:41 <+bridge> 😬 23:41 <+bridge> xd 23:41 <+bridge> edition 2024 in october 23:42 <+bridge> game Launcher for better updates 23:42 <+bridge> Personally I'd want to keep ogl1 forever. Let people use their toasters in peace 23:43 <+bridge> nah 23:43 <+bridge> OGL1 is soo annoying 23:43 <+bridge> the editor is so much faster without OGL1 23:44 <+bridge> can we make ogl1 behind a macro def 23:44 <+bridge> xd 23:44 <+bridge> It is annoying but it atleast runs on old hw 23:44 <+bridge> and then xd 23:44 <+bridge> we should do a hardware survery 23:44 <+bridge> 2014 was 10!!! years ago 23:44 <+bridge> we oldnguyst 23:45 <+bridge> we old guys 23:45 <+bridge> browsers use ANGLE, which kinda is GLES on top of multiple backends afaik. 23:45 <+bridge> These older devices also have buggy gl3 implementations which you'll blacklist and then tell people to buy new computers 23:45 <+bridge> so i think it's not really needed 23:45 <+bridge> I know you better than you 23:45 <+bridge> they can just use old ddnet 23:45 <+bridge> since they use old hardware 23:45 <+bridge> makes sense 23:45 <+bridge> xD 23:45 <+bridge> They cant, you also want to remove all server backcompat 23:46 <+bridge> but i dont remove ddnet-cpp at all 23:46 <+bridge> as said multiple versions 23:46 <+bridge> 2 ddnets 23:46 <+bridge> * ddnet for true gamers 23:46 <+bridge> * ddnet for middle agers 23:46 <+bridge> medieval ddnet 23:47 <+bridge> People who can afford good hardware* 23:47 <+bridge> <._.spook._.> is there a thing that makes your screen darken for a certain period of time 23:47 <+bridge> When rtx50xx requirement? 23:48 <+bridge> well i am a maximist for overall happyness. 23:48 <+bridge> 23:48 <+bridge> if 99% of the players profit from it, then yes, poor 1% for staying in the past 23:48 <+bridge> everyone can afford modern enough hardware 23:48 <+bridge> 99% of players dont use the ogl1 backend. So they wont be affected badly by us maintaining it. Perfect 23:49 <+bridge> you are correct, 99% dont use it, still we have to support it.. for exactly no gain at all 23:50 <+bridge> Anyway, do it however you like for your personal project. I definitely won't +1 a pr that removes ogl1 anytime soon 23:51 <+bridge> you dont listen to me anyway 23:51 <+bridge> i wont touch ddnet-cpp 23:51 <+bridge> i want to split ddnet into 2 clients 23:51 <+bridge> and 2 servers 23:51 <+bridge> and if u want to give support for OGL1, do that 23:52 <+bridge> your time, your choice 23:52 <+bridge> just like u want to make 1% of players happy, you can make me happy by supporting my plan 23:52 <+bridge> i mean in the end that is what you want isnt it? 23:52 <+bridge> everyone should be happy 23:55 <+bridge> Support in what way? I'm definitely not hosting a version of the server that players can't connect to without the very latest jupjupnet client. So can't help you on serverside adoption. 23:55 <+bridge> 23:55 <+bridge> Best I can do is wish you good luck 23:55 <+bridge> mh weird, you want me sad 23:55 <+bridge> It would atleast be interesting to see how many fps you can pull out of the newest gpus 23:55 <+bridge> where is your happyness maximism now 23:56 <+bridge> you really think it's about fps alone? 23:56 <+bridge> it's about all the code bloat 23:56 <+bridge> ddnet imo is very hard to maintain 23:56 <+bridge> all the network extensions 23:56 <+bridge> all the backends 23:56 <+bridge> all the useless features ppl want and are disabled by default 23:56 <+bridge> Well that too, but you can't just slash out some features and say you reduced bloat. You did but we lost features too. Backcompat is a feature 23:56 <+bridge> see my project also sucks in many ways 23:57 <+bridge> but at least these 3 things i tried to kinda fix 23:57 <+bridge> I think it belongs in it's own codebase tbh. But it is a feature nonetheless 23:57 <+bridge> backcomp is just egoism of some ppl 23:57 <+bridge> ddnet 18 and ddnet 9 are 99% the same 23:57 <+bridge> what do you miss in ddnet 18 that is in ddnet9? 23:58 <+bridge> ddnet 18 works on all computers too, soon even android 23:58 <+bridge> Fwiw I always played on the very latest client. It's not me that needs the backcompat 23:59 <+bridge> The only thing I have an issue with about backcompat is that the burden of maintainance is on all of us, while very few actually use it and even less among the devs