00:00 < bridge> <heinrich5991> yes
00:05 < bridge> <Jupstar ✪> One qol advantage of having an index is that you can show the database content inside the client ui easily
00:05 < bridge> <Jupstar ✪> https://cdn.discordapp.com/attachments/293493549758939136/1301321207580463215/IMG_20241031_000457.jpg?ex=67240d51&is=6722bbd1&hm=5710988d81ee054ce96364e1f4dba5f1726c3272e58bd895584a5b8f896425c4&
01:42 < bridge> <fr4gues> "\xff\xff\xff\xffgetstatus" not working on ddnet?
01:46 < bridge> <heinrich5991> what is this? ^^
01:46 < bridge> <heinrich5991> you can get status of all teeworlds and ddnet servers from https://master1.ddnet.org/ddnet/15/servers.json
02:09 < bridge> <fr4gues> the objective is to connect to a game server using the UDP protocol, identify yourself as a player and collect information from the server, I tested it with other games and got a response, but with ddnet there is no return
02:18 < bridge> <fr4gues> Is there a specific command for ddnet?
08:19 < ws-client> <ChillerDragon> @fr4gues if you want to identify yourself as a player to the server you have to speak the protocol and do a full token handshake. Its more than sending one message. But as heinrich already said the publicly listed servers data is served as json.
08:19 < ws-client> <ChillerDragon> @fr4gues what data exactly do you need?
12:49 < bridge> <milkeeycat> Github contributions are orange now :boo:
12:49 < bridge> <milkeeycat> https://cdn.discordapp.com/attachments/293493549758939136/1301513475411021987/image.png?ex=6724c061&is=67236ee1&hm=5853820b22278b7f1ead5e59d8e6320dac96d87554054d71fc0eaa44d37468fa&
13:20 < bridge> <Jupstar ✪> I am more scared about how empty my contribution history is :boo:
13:27 < bridge> <tsfreddie> imagine someone who can't tell the difference because it is empty
13:45 < bridge> <Jupstar ✪> @robyt3 hello android pro.
13:45 < bridge> <Jupstar ✪> 
13:45 < bridge> <Jupstar ✪> Have you ever made/have experience with an app with high multi threaded workload?
14:06 < bridge> <learath2> All my contributions are in filthy for profit private repositories :pepeW:
14:06 < bridge> <learath2> https://cdn.discordapp.com/attachments/293493549758939136/1301532772749545532/image.png?ex=6724d25a&is=672380da&hm=a918ed6253446f6718f7436e3b7a03091ef07587040b069e96757e4f2cd4d702&
14:40 < bridge> <robyt3> With high workload you mean you want to perform heavy calculations on all CPU cores? I don't really have more experience except the DDNet multi threading, which isn't really doing heavy calculations. According to https://developer.android.com/topic/performance/threads we should lower the thread priority for background jobs so the UI thread can run smoother.
14:40 < bridge> <robyt3> > Generally, threads in the foreground group get about 95% of the total execution time from the device, while the background group gets roughly 5%.
14:42 < bridge> <Jupstar ✪> Yeah, the point is when I tried some stuff with dd-pg on Android I noticed that the main thread got completely blocked if 2 threads consume lot of CPU resource (e.g. during compilation of shaders)
14:42 < bridge> <Jupstar ✪> 
14:42 < bridge> <Jupstar ✪> I even tested it on different devices. So something limits my workload and `available_parallelism` is not really reliable.
14:43 < bridge> <Jupstar ✪> I wonder if that is some energy saving mechanism or smth
14:43 < bridge> <Jupstar ✪> I have 8 cores, so at least 4 should be able to be used
14:43 < bridge> <robyt3> yeah, I guess you need to be careful with threads since they inherit the priority from the thread that spawned them
14:43 < bridge> <robyt3> > Practically speaking, there’s a number of variables responsible for this, but picking a value (like 4, for starters), and testing it with Systrace is as solid a strategy as any other. You can use trial-and-error to discover the minimum number of threads you can use without running into problems.
14:43 < bridge> <robyt3> > Each thread costs a minimum of 64k of memory
14:44 < bridge> <Jupstar ✪> Ok I try to lower the priority
14:44 < bridge> <Jupstar ✪> 
14:44 < bridge> <Jupstar ✪> I use https://github.com/iddm/thread-priority
14:44 < bridge> <Jupstar ✪> for the main thread
15:08 < bridge> <Jupstar ✪> The low prio threads dominate, and for some reason they still block the main thread :c
15:08 < bridge> <Jupstar ✪> https://cdn.discordapp.com/attachments/293493549758939136/1301548410150715413/image.png?ex=6724e0ea&is=67238f6a&hm=64d0a39b1a702565b466a7bb093a8524618b7fac619baecf7b0f516a2ea878df&
15:08 < bridge> <Jupstar ✪> I assume the last 4 cores are the performance cores
15:11 < bridge> <heinrich5991> it doesn't look like they're blocked by CPU usage, the cores aren't at 100%
15:11 < bridge> <heinrich5991> is the main thread waiting for something?
15:11 < bridge> <Jupstar ✪> That doesn't matter if that is the energy management. Especially that they aren't at 100% makes no sense
15:12 < bridge> <Jupstar ✪> No
15:12 < bridge> <heinrich5991> oh, energy management means that the cores could be below 100% and no more CPU time is available?
15:13 < bridge> <Jupstar ✪> I guess good energy management means that CPU cores should clock as low as possible
15:13 < bridge> <Jupstar ✪> I guess running 4 cores at 50% is much more efficient than even one at 100%
15:14 < bridge> <Jupstar ✪> Too bad my desktop is too fast, I can't really see what happens there xD
15:15 < bridge> <Jupstar ✪> The low prio threads are only some noise
15:15 < bridge> <Jupstar ✪> https://cdn.discordapp.com/attachments/293493549758939136/1301550058331701299/image.png?ex=6724e273&is=672390f3&hm=280c7370dd95ed653ec005ad48277193e81838922a8dd38223336c61aa0b0f74&
15:15 < bridge> <Jupstar ✪> Kinda funny how good androids output looks
15:16 < bridge> <Jupstar ✪> Almost as if they have some structure
15:16 < bridge> <Jupstar ✪> And on desktop just completely random mess
15:17 < bridge> <Jupstar ✪> I can defs observe that no thread ever gets above 50% usage
15:18 < bridge> <Jupstar ✪> Everytime that happens the kernel switches the core
15:20 < bridge> <Jupstar ✪> Mh what I find interesting is that once the app is started mostly the first cores are used
15:22 < bridge> <Jupstar ✪> Ah anyway Android is too complicated for me
15:22 < bridge> <Jupstar ✪> I just want to spawn 50000 threads and get the task done
15:22 < bridge> <Jupstar ✪> But if I only use 1 async task it defs works as I expect it
15:22 < bridge> <Jupstar ✪> It does the heavy stuff in parallel without blocking the main thread
15:22 < bridge> <Jupstar ✪> But if I only use 1 async thread it defs works as I expect it
15:29 < bridge> <Jupstar ✪> But dunno it's exactly 2 threads.
15:29 < bridge> <Jupstar ✪> 
15:29 < bridge> <Jupstar ✪> When I use 1 thread - perfect.
15:29 < bridge> <Jupstar ✪> 2 threads => main thread completely blocked
15:29 < bridge> <Jupstar ✪> So weird
15:29 < bridge> <Jupstar ✪> What tools can I use on android to see what happens xd
15:29 < bridge> <Jupstar ✪> Why do I not have a Linux phone xd
15:29 < bridge> <Jupstar ✪> (a proper 😏 )
15:57 < bridge> <Jupstar ✪> @heinrich5991 you once suggested 
15:57 < bridge> <Jupstar ✪> https://github.com/tokio-rs/tracing
15:57 < bridge> <Jupstar ✪> 
15:57 < bridge> <Jupstar ✪> to #[instrument] the code.
15:57 < bridge> <Jupstar ✪> 
15:57 < bridge> <Jupstar ✪> But I dunno, should I put that on every function in my whole code base or what
16:22 < bridge> <teero777> How long can the sounds in tw be? Can I replace the hook sound with an hour long sound of a construction site?
16:22 < bridge> <Jupstar ✪> I think yes
16:23 < bridge> <teero777> Ah finally I can make my dream come true
16:23 < bridge> <Jupstar ✪> If you have too many sounds the thread might hang tho
16:23 < bridge> <Jupstar ✪> since it's to slow to fill buffers
16:23 < bridge> <Jupstar ✪> But knowing ddnet code, it probs has a limit anyway
16:24 < bridge> <Jupstar ✪> this one or what?
16:24 < bridge> <Jupstar ✪> 
16:24 < bridge> <Jupstar ✪> https://www.youtube.com/watch?v=Q7eJg7hRvqE
16:25 < bridge> <teero777> My connection isn't good enough for anything more than text crntly
16:25 < bridge> <teero777> Takes some secs to even send the msg
16:25 < bridge> <Jupstar ✪> Ah sounds like you travel in germany
16:32 < bridge> <0xdeen> Mostly but not always! There is some time-based power usage, so it can be more efficient to run at a higher clock speed and then go to sleep instead of running for longer at a lower clock speed. But the graphs I remembered for that were from 2010 or so
16:33 < bridge> <Jupstar ✪> Ah interesting.
16:33 < bridge> <Jupstar ✪> And are you by chance an Android pro and know the behavior there? xd
16:33 < bridge> <0xdeen> nope
16:33 < bridge> <Jupstar ✪> I let my async task sleep for 10 seconds before they start now. The main thread works for exactly 10 seconds xD
16:33 < bridge> <Jupstar ✪> Then it's rip
16:34 < bridge> <heinrich5991> an android pro would probably ask whether you're on stock android or on some random vendor's rom
16:34 < bridge> <0xdeen> no one is on stock android
16:34 < bridge> <Jupstar ✪> I am on stock, but I also tried on different phones
16:34 < bridge> <Jupstar ✪> Ok but not vendor at least 😄
16:34 < bridge> <heinrich5991> let's say close to stock
16:34 < bridge> <Jupstar ✪> Some open source thing
16:34 < bridge> <0xdeen> even pixel has moved far from stock I thought
16:35 < bridge> <Jupstar ✪> crDroid
16:35 < bridge> <heinrich5991> that should be relatively close to stock, I think
16:35 < bridge> <Jupstar ✪> No google services 😬
16:35 < bridge> <Jupstar ✪> I am savage
16:35 < bridge> <heinrich5991> not a random vendor's "battery optimizations" we're looking at
16:39 < bridge> <reitw> It looks like stock Android, just includes pixel-only features
16:52 < bridge> <jxsl13> I'm on pixel D: what do you mean I'm not on stock android
17:13 < bridge> <learath2> Get the AOSP build and see how far from stock Pixel is now
17:28 < bridge> <reitw> you are on stock android with pixel features
17:28 < bridge> <reitw> technically
17:28 < bridge> <reitw> which pixel
17:37 < bridge> <jxsl13> 7a
17:39 < bridge> <learath2> Go be fair, all android are based on AOSP 😄
17:40 < bridge> <reitw> ye but samsung OneUI overlay makes it feel it's smth completly different
17:40 < bridge> <reitw> ah nice
17:40 < bridge> <reitw> been having fun with the 9 pro
17:40 < bridge> <learath2> I guess pixel is the closest to stock in that sense yeah
17:41 < bridge> <jxsl13> I got a custom launcher, so my phones lioked nearly the same for like the past 8 years
17:42 < bridge> <jxsl13> looked
17:42 < bridge> <jxsl13> :kek:
17:45 < bridge> <Jupstar ✪> I swear these codecov fks are so annoying xdd
17:46 < bridge> <Jupstar ✪> I get so many emails from them the last months
17:46 < bridge> <Jupstar ✪> Why do they update their privacy policy so often
17:46 < bridge> <Jupstar ✪> I didn't even opt-into their emails xD
19:29 < bridge> <milkeeycat> @learath2 modules almost done :clueless:
19:29 < bridge> <milkeeycat> https://cdn.discordapp.com/attachments/293493549758939136/1301613924243865712/image.png?ex=67251dee&is=6723cc6e&hm=9384e73a636e013d8b6e6d887107536fd3982d5dfc1e75a788ac388bf310371f&
20:40 < bridge> <milkeeycat> Oooh yea, it feels nice
20:40 < bridge> <milkeeycat> https://cdn.discordapp.com/attachments/293493549758939136/1301631785972535378/image.png?ex=67252e90&is=6723dd10&hm=0d7fbfaae11857f3d0f9b2bddfdf981fb1a6919da90022dabbf651862c6b7769&
20:44 < bridge> <kollpotato> metal pipe falling sound for hammer
20:44 < bridge> <kollpotato> that would be epic
21:43 < bridge> <ryozuki> @milkeeycat https://github.com/dureuill/nolife/issues/7
21:43 < bridge> <ryozuki> @heinrich5991
21:43 < bridge> <ryozuki> i found a comment related to the issue we had
21:43 < bridge> <ryozuki> &'a Foo<'a> issue
21:44 < bridge> <milkeeycat> pff
21:44 < bridge> <milkeeycat> what a nobo
21:44 < bridge> <milkeeycat> everybody knows `&'a T<'a>` bad :greenthing:
22:19 < bridge> <pathos3005> Would it be best to post feature requests/improvements on Github issues? I just wanted to suggest a smoother camera method for multi-view spectator mode when tees respawn.
22:28 < bridge> <kebscs> yes, or #town-hall
23:19 < zhn> good night guys! is there any way to go around with redirecting from one port to another when registering to master server?
23:19 < zhn> i can connect to the server without any problem with raw ip but it will not register on master because original port is "closed"