00:26 <+bridge_> <learath2> What would you like to talk about? I can make some time tomorrow I think
00:27 <+bridge_> <totar> there's another issue with tele predict pr, idk what to do again
00:27 <+bridge_> <learath2> As in a player disconnects?
00:28 <+bridge_> <totar> no, actually their projectiles get deleted when that happens
00:28 <+bridge_> <learath2> A player kills?
00:28 <+bridge_> <learath2> But then only their character is gone 😄
00:28 <+bridge_> <totar> as far as I'm aware it's possible for a player outside your snap range to fire a projectile into your snap range, and then you have to predict the projectile without their tee data
00:29 <+bridge_> <learath2> Maybe don't predict those? You can't possibly do it correctly since you won't have the tunes, no?
00:29 <+bridge_> <totar> uh
00:29 <+bridge_> <totar> interesting
00:30 <+bridge_> <totar> I forgot about tunes
00:30 <+bridge_> <kebscs> isnt projectile
00:30 <+bridge_> <totar> I thought tunes are seperate from tee
00:30 <+bridge_> <kebscs> isnt projectile's tune based on its spawn pos
00:30 <+bridge_> <totar> that could also be true
00:30 <+bridge_> <totar> I thought tunes are separate from tee
00:30 <+bridge_> <learath2> Do we have tune locks now though? That will change things
00:30 <+bridge_> <kebscs> same as other info, should be contained within projectile if its important
00:30 <+bridge_> <kebscs> projectiles still work same
00:31 <+bridge_> <kebscs> lock isnt merged, but it works only on tunes that affect hte player
00:31 <+bridge_> <totar> what?
00:31 <+bridge_> <learath2> But a player might be tune locked to a different tune zone from the location the projectile started with, no? Or do we only limit tunelock to specific tunes?
00:31 <+bridge_> <kebscs> in tune lock, unrelated to ur pr mine wont get merged for long time i guess
00:31 <+bridge_> <totar> I have the same question as learath
00:32 <+bridge_> <kebscs> yes
00:32 <+bridge_> <learath2> Yes to which question? 😄
00:32 <+bridge_> <kebscs> https://cdn.discordapp.com/attachments/293493549758939136/1465852049198415912/image.png?ex=697a9c9a&is=69794b1a&hm=e41001d2dce3ab612b612d0671e3ce1aaea8f9f167d750a7cc62617568072a24&
00:32 <+bridge_> <kebscs> its unrelated to player's tune/tune lock so it contains all necessary info
00:32 <+bridge_> <learath2> If tunes aren't your issue, then why is there still an issue? That's what I thought you'd need the players info for
00:33 <+bridge_> <totar> let me check before I say something wrong
00:34 <+bridge_> <learath2> Hum, so projectiles only check the tunezone at their spawn position?
00:34 <+bridge_> <kebscs> yes
00:35 <+bridge_> <kebscs> and startpos is sent in snap, so this isnt tele prediction blocker
00:35 <+bridge_> <learath2> Ok, than that part is fine for now, though idk if maybe we'd want to allow that to be tunelock'd too, in that case maybe the players tunezone should be used. Idk different discussion for a different time
00:35 <+bridge_> <kebscs> probably just sending the tune with snap the, instead of guessing it based on spawn pos
00:35 <+bridge_> <kebscs> probably just sending the tune with snap then, instead of guessing it based on spawn pos
00:53 <+bridge_> <totar> I'm adding a m_RngSeed value to extended DDNetCharacter which I mix into the seeding
00:53 <+bridge_> <totar> 
00:53 <+bridge_> <totar> this is technically optional but I thought it would be future proof if we decide we need any other kind of "semi secure random" features and it at least prevents even more replay sharing from happening on ddnet servers.
00:53 <+bridge_> <totar> 
00:53 <+bridge_> <totar> So the options I see are
00:53 <+bridge_> <totar> 
00:53 <+bridge_> <totar> 1. drop the m_RngSeed from the implementation, seed only with client id (which the projectile tells us)
00:53 <+bridge_> <totar> 2. don't predict teleport for projectiles without valid ``CCharacterInfo``, might be annoying on some maps idk
00:53 <+bridge_> <totar> 3. duplicate the m_RngSeed onto projectiles so we always know it (and yes it would need to be duplicated or we can't create projectiles from predicted players during prediction)
00:53 <+bridge_> <totar> 
00:54 <+bridge_> <totar> there's also a more annoying option to save the m_RngSeed whenever we see a player and assume it hasn't changed since they left our snap range, which is kind of optimal but currently m_Snap gets reset between ticks so we don't save any information about the CharacterInfo after we lose sight of them.
00:55 <+bridge_> <totar> I'm adding a m_RngSeed value to extended DDNetCharacter which I mix into the seeding
00:55 <+bridge_> <totar> 
00:55 <+bridge_> <totar> this is technically optional but I thought it would be future proof if we decide we need any other kind of "semi secure random" features and it at least prevents even more replay sharing from happening on ddnet servers.
00:55 <+bridge_> <totar> 
00:55 <+bridge_> <totar> So the options I see are
00:55 <+bridge_> <totar> 
00:55 <+bridge_> <totar> 1. drop the m_RngSeed from the implementation, seed only with client id (which the projectile tells us)
00:55 <+bridge_> <totar> 2. don't predict teleport for projectiles without valid ``CCharacterInfo``, might be annoying on some maps idk
00:55 <+bridge_> <totar> 3. duplicate the m_RngSeed onto projectiles so we always know it (and yes it would need to be duplicated or we can't create projectiles from predicted players during prediction)
00:55 <+bridge_> <totar> 
00:55 <+bridge_> <totar> 4. there's also a more annoying option to save the m_RngSeed whenever we see a player and assume it hasn't changed since they left our snap range, which is kind of optimal but currently m_Snap gets reset between ticks so we don't save any information about the CharacterInfo after we lose sight of them.
00:55 <+bridge_> <totar> we can do 2 now and then either 3 or 4 in the future, but we can't change our mind about 1 without significant backcompat
07:35 <+bridge_> <alw5> @totar could pressing and releasing inputs too fast not registering be a tclient issue?
07:35 <+bridge_> <alw5> i probably should of recorded that video on ddnet client
07:35 <+bridge_> <totar> you should always test bugs on ddnet
07:35 <+bridge_> <alw5> true
07:36 <+bridge_> <alw5> but i dont think its a tclient thing
07:36 <+bridge_> <totar> I test all bugs on ddnet, even if it's not a tclient thing
07:37 <+bridge_> <totar> largely because tclient is usually closer to ddnet nightly which is much less stable in general
07:37 <+bridge_> <totar> I used to selectively merge when I thought ddnet was stable but I don't have time to do that anymore
07:37 <+bridge_> <alw5> no it happens on latest ddnet ver
07:38 <+bridge_> <totar> I also have a folder with every ddnet release version for testing what version something broke in
07:39 <+bridge_> <totar> very useful
07:39 <+bridge_> <alw5> damn
07:39 <+bridge_> <alw5> how much storage does that take
07:39 <+bridge_> <totar> not that much
07:39 <+bridge_> <alw5> several gigabytes?
07:39 <+bridge_> <totar> idk probably
07:39 <+bridge_> <totar> I don't have literally every version
07:39 <+bridge_> <totar> only like ~3 years
07:40 <+bridge_> <totar> super old versions are scary
07:40 <+bridge_> <totar> they mess up new ddnet configs
08:10 <+bridge_> <chillerdragon> Your config is not tracked in git?
08:12 <+bridge_> <chillerdragon> What could he scary about it xd
08:13 <+bridge_> <chillerdragon> I miss Jupstar
08:34 <+bridge_> <totar> cba
09:33 <+bridge_> <Assa> you could automatically wait for 1 month after ddnet switches to a newer version - because maintainers - including me - tend to merge "more unsecure" features on the beginning of a new version in order to have a longer grace period for finding bugs
09:33 <+bridge_> <Assa> idk still probably a lot of work
09:35 <+bridge_> <Assa> 16.0 - 19.latest: ~3.8 GB, 15.0 - 15.latest, 700 MB
09:36 <+bridge_> <Assa> ddnet doesn't have lots of assets by itself - keep in mind that this doesn't inlcude the appdata stuff
10:11 <+bridge_> <milkeeycat> Ah yes, classic 15k lines update readme commit 😬
10:11 <+bridge_> <milkeeycat> https://cdn.discordapp.com/attachments/293493549758939136/1465997601873592423/image.png?ex=697b2429&is=6979d2a9&hm=99c7966e5c6d965bfe9b6b7d31a8eb71c28b4df4c43f40f7db5e7c57e188c260&
10:11 <+bridge_> <milkeeycat> https://cdn.discordapp.com/attachments/293493549758939136/1465997602242957436/image.png?ex=697b2429&is=6979d2a9&hm=598304026551aca31ab451b7e0119674117df857d6899a4417d2382955063082&
10:25 <+bridge_> <manikoo> 👍
10:49 <+ChillerDragon> @robyt3 the default config name space is really cool. Is there a way to get the min and max too?
10:49 <+ChillerDragon> That would come in really handy for compile time bounds checking with static assert for me
10:50 <+ChillerDragon> i do sone `aGoofedArray[g_Config.m_ClMyHackCfg - 2]` and it looks scary xd
10:50 <+ChillerDragon> some*
10:50 <+ChillerDragon> i assume the UI pager code could profit from that too
11:16 <+bridge_> <heinrich5991> sounds reasonable to me, wanna PR it?
11:44 <+ChillerDragon> so it doesnt exist?
11:44 <+ChillerDragon> i rather open an issue i already have a pending pr xd
11:44 <+ChillerDragon> heinrichs comment about an LLM reviwer having caught the server info oopsie made me think
11:44 <+ChillerDragon> that would be really op
11:45 <+ChillerDragon> i would like to have a llm lsp that runs a simple local model and shows warnings like any other linter
11:45 <+bridge_> <heinrich5991> I'm not sure if it would have caught 15 other incorrect bugs as well
11:45 <+ChillerDragon> that could help combat the severe brain damage i got when i was dropped on the head during birth
11:45 <+ChillerDragon> check this out `bool IsDone() const { return m_State != EState::DONE; }
11:45 <+bridge_> <heinrich5991> (and then it'd be useles)
11:45 <+bridge_> <heinrich5991> (and then it'd be useless)
11:45 <+ChillerDragon> aaaa weechat paste ah
11:46 <+ChillerDragon> ye u gotta tweak it to be useful
11:46 <+ChillerDragon> but i think its feasible
11:46 <+bridge_> <heinrich5991> when I debugged my ctf solve script, I pasted it into chatgpt and it told me 15 bugs, 14 of which were made up
11:46 <+bridge_> <heinrich5991> but the one bug that was not was helpful
11:46 <+bridge_> <heinrich5991> but it was only helpful because I already knew that there was a bug
11:46 <+bridge_> <heinrich5991> otherwise I would have just dismissed them
11:46 <+ChillerDragon> ye but u just pasted into gpt
11:47 <+ChillerDragon> u can tune these models to reduce hallucinations and so on
11:47 <+ChillerDragon> im sure there is a way to optimize for no false positives
11:47 <+bridge_> <heinrich5991> I severely doubt it ^^
11:47 <+ChillerDragon> if it misses a bit thats fine
11:47 <+ChillerDragon> woah
11:47 <+ChillerDragon> AI doubter
11:47 <+ChillerDragon> i will report you to sam
11:47 <+bridge_> <heinrich5991> LLMs love false-positives
11:47 <+bridge_> <heinrich5991> "hallucinations"
11:56 <+bridge_> <chillerdragon> Typing git init is 0 effort xd
12:39 <+ChillerDragon> @heinrich5991 can u merge the antibot pr or whats the holdup?
13:08 <+bridge_> <heinrich5991> the holdup is the last comment I sent
13:08 <+bridge_> <heinrich5991> I see you answered it
13:08 <+bridge_> <heinrich5991> you'd like me to drop whatever I'm working on and prioritize your PR right now 🤷‍♀️
13:12 <+bridge_> <heinrich5991> actually, let me not do that
13:13 <+bridge_> <heinrich5991> it's clear what the holdup is. and I haven't had the time to look at your answer yet
13:13 <+bridge_> <heinrich5991> you can answer the question you gave on your own
13:13 <+bridge_> <heinrich5991> if you want to communicate something else, communicate that
13:16 <+bridge_> <Assa> @heinrich5991 if I ping you like that, are you fine if I merge after a while: <https://github.com/ddnet/ddnet/pull/11546#pullrequestreview-3681224432>, because for me it looks like you seem to loose interest or don't get some pings. (I don't want you to drop whatever you are doing and look at this immediately, just a genuine question, I remember another PR where I requested review waiting for months)
13:18 <+bridge_> <Assa> > waiting for months
13:18 <+bridge_> <Assa> 
13:18 <+bridge_> <Assa> #10691
13:18 <+bridge_> <DDNet> https://github.com/ddnet/ddnet/pull/10691
13:49 <+bridge_> <0xdeen> @ldozdng Thanks! ^
13:50 <+bridge_> <chillerdragon> I was told that is the official flow on how to get merges
13:50 <+bridge_> <chillerdragon> I don’t like it either tbh
13:51 <+bridge_> <heinrich5991> where did you get told that?
13:51 <+bridge_> <chillerdragon> Lerato
13:51 <+bridge_> <chillerdragon> In here
13:52 <+bridge_> <heinrich5991> if you have to, I'd prefer something like: "hey, could you take a look at my PR again"
13:52 <+bridge_> <chillerdragon> Isn’t that what I did?
13:52 <+bridge_> <heinrich5991> your message read to me that there are only two possible options: "merge or explain your reasoning now"
13:53 <+bridge_> <chillerdragon> I still see no difference to „hey look“
13:54 <+bridge_> <chillerdragon> I understand that these reminders are annoying and I also would prefer to not ask for merges but the current state seems to require doing exactly that
13:55 <+bridge_> <chillerdragon> And the „how“ it’s being done is not that deep imo
13:55 <+bridge_> <chillerdragon> Same message either way
13:55 <+bridge_> <Assa> thanks heinrich for taking a look at the PRs I mentioned 👍 ❤️
13:55 <+bridge_> <heinrich5991> yea, it's okay. I'll probably get to it eventually, but I currently have a backlog of 600 messages (mostly ddnet), dating back to 2026-01-18, so this one was probably not lost
13:55 <+bridge_> <heinrich5991> the "how" is important in interpersonal communication
13:56 <+bridge_> <Assa> yes I know it's hard to keep up with all of the pings, it's insane, on discord and on github
13:56 <+bridge_> <chillerdragon> Ok
13:56 <+bridge_> <chillerdragon> To me it still reads the same even interpersonally
13:56 <+bridge_> <chillerdragon> You can just respond with „holdup is I’m busy“ ez
13:56 <+bridge_> <heinrich5991> your message is a question, to which the literal answer is something you probably don't want to know
13:57 <+bridge_> <heinrich5991> "the holdup is the thing I said last time, and I haven't read the next message yet"
13:57 <+bridge_> <heinrich5991> what you want isn't in the message, as far as I'd interpret you, namely that you ask/expect me to take a look
13:58 <+bridge_> <chillerdragon> How would it be any different if I ask „could you look“?
13:58 <+bridge_> <chillerdragon> Answer is „yes soon“
13:58 <+bridge_> <chillerdragon> Same thing
13:58 <+bridge_> <chillerdragon> @heinrich5991: I think we need more maintainers to help you with your backlog
14:00 <+bridge_> <heinrich5991> the difference is what you ask is in the message
14:00 <+bridge_> <heinrich5991> more maintainers will not help me with my backlog
14:00 <+bridge_> <heinrich5991> I think we currently have enough maintainers
14:10 <+bridge_> <chillerdragon> @heinrich5991: because only you can merge that pr?
14:11 <+bridge_> <heinrich5991> because the backlog of messages I see from ddnet/ddnet will not decrease when we get more maintainers
14:11 <+bridge_> <chillerdragon> But then I am not blocked by your backlog anymore
14:12 <+bridge_> <chillerdragon> But I see it doesn’t help you if you want to see everything
14:44 <+bridge_> <heinrich5991> anyone going to fosdem this weekend btw?
14:44 <+bridge_> <heinrich5991> https://fosdem.org/2026/
14:44 <+bridge_> <heinrich5991> I'll be there
14:48 <+bridge_> <Assa> are you there on both days? I am thinking about it tbh, it's only 2 hours away from me
14:48 <+bridge_> <Assa> (I like how the website explicitly states belgian beer 😆 )
14:49 <+bridge_> <heinrich5991> I'll be there on both days
15:47 <+bridge_> <melon> :Sadge:
15:47 <+bridge_> <melon> https://cdn.discordapp.com/attachments/293493549758939136/1466082287383875686/image.png?ex=697b7307&is=697a2187&hm=f7b907b5f16233bb50f8659349834dc891e2c886639e7bee268261b304635160&
15:48 <+bridge_> <melon> i would love to attend these things, but my stoopid butt decided "oh yeah, lets move close to austria"
15:51 <+bridge_> <Assa> you decided to life in bavaria, how dare you
15:56 <+bridge_> <heinrich5991> maybe GPN works better for you? https://entropia.de/GPN24
16:00 <+ChillerDragon> @heinrich5991 @kebscs did i understand https://github.com/ddnet/ddnet/pull/11679 correctly?
16:00 <+ChillerDragon> you want to use the existing "team" field for ddrace teams too?
16:00 <+ChillerDragon> this is a bad idea
16:01 <+bridge_> <heinrich5991> I think it's a good idea to remove the concept of two different "teams" in the codebase: https://github.com/ddnet/ddnet/issues/11272
16:02 <+ChillerDragon> but thats a bad idea too
16:06 <+ChillerDragon> @heinrich5991 where do you want your style guide? In the wiki or markdown file in the ddnet repo? https://github.com/ddnet/ddnet/pull/11684
16:07 <+bridge_> <robyt3> The source of truth should be the git repository, either with automatic checks or contributing
16:08 <+ChillerDragon> well i agree on the "either or"
16:08 <+ChillerDragon> heinrich doesnt
16:08 <+ChillerDragon> he wants both
16:08 <+bridge_> <Solly (bun bun)> both but remove the verbosity of contributing.md?
16:08 <+bridge_> <Solly (bun bun)> both but remove the (over) verbosity of contributing.md?
16:08 <+bridge_> <Solly (bun bun)> 😉
16:08 <+ChillerDragon> and i want to not pollute the text that has to be read with the things that are enforced automatically
16:12 <+bridge_> <learath2> huh, Brussels is close enough, I might
16:16 <+bridge_> <robyt3> If the enforcement works well enough *and* makes correct suggestions, we don't need contributing, for example the check for standard C headers. The suggestions regarding variable names made by clang-tidy are often wrong because we abuse the system by overriding everything with regexes.
16:20 <+bridge_> <melon> @robyt3 to not pollute bugs further :kekMonki: 
16:20 <+bridge_> <melon> (and i guess heinrich too, if you're here)
16:20 <+bridge_> <melon> 
16:20 <+bridge_> <melon> ```md
16:20 <+bridge_> <melon> ## Symbolizing a crash dump (.RTP):
16:20 <+bridge_> <melon> Symbolizing converts raw crash data into humanly readable stacktraces.
16:20 <+bridge_> <melon> **Download the crash dump**  
16:20 <+bridge_> <melon>     Make sure it has the original filename because it contains important information.
16:20 <+bridge_> <melon> 
16:20 <+bridge_> <melon> **Download the debug symbols**  
16:20 <+bridge_> <melon>     Get them from https://ddnet.org/downloads/symbols/ with matching git hash and platform (win32/64, steam/non-steam).  
16:20 <+bridge_> <melon>    - The git hash and platform are found in the filename of the crash dump.
16:20 <+bridge_> <melon>    - If the filename does not contain `steam`, then it is the standalone version.
16:20 <+bridge_> <melon> 
16:21 <+bridge_> <melon>  **Install binutils**  
16:21 <+bridge_> <melon>    Install `addr2line` and `objdump` using your system's package manager or from msys2 for Windows:
16:21 <+bridge_> <melon>    - **[MSYS2 binutils](https://packages.msys2.org/packages/mingw-w64-x86_64-binutils)**
16:21 <+bridge_> <melon>    - **Ubuntu/Debian**: `sudo apt install binutils`
16:21 <+bridge_> <melon> 
16:21 <+bridge_> <melon>  **Run the script**  
16:21 <+bridge_> <melon> ```sh
16:21 <+bridge_> <melon>    bash scripts/parse_drmingw.sh <path to server/client symbols exe> <path to server/client crash dump>
16:21 <+bridge_> <melon> ```
16:21 <+bridge_> <melon>    - The script is found under `./scripts/`
16:21 <+bridge_> <melon>    - On Windows, you must explicitly run with `bash` and not in a Windows terminal.
16:21 <+bridge_> <melon> ```
16:21 <+bridge_> <melon> 
16:21 <+bridge_> <melon> i'd put this right before the DDNet DB stuff on the readme, thoughts before i PR?
16:21 <+bridge_> <melon> ah damn it, readme formatting broke discord's formatting
16:21 <+bridge_> <melon> :kek:
16:21 <+bridge_> <learath2> Yeah, the 50 different flavors of markdown interact together brilliantly
16:21 <+bridge_> <learath2> Especially the half ass one Discord ended up with
16:22 <+bridge_> <melon> @robyt3 to not pollute bugs further :kekMonki: 
16:22 <+bridge_> <melon> (and i guess heinrich too, if you're here)
16:22 <+bridge_> <melon> 
16:22 <+bridge_> <melon> 
16:22 <+bridge_> <melon> ## Symbolizing a crash dump (.RTP):
16:22 <+bridge_> <melon> Symbolizing converts raw crash data into humanly readable stacktraces.
16:22 <+bridge_> <melon> **Download the crash dump**  
16:22 <+bridge_> <melon>     Make sure it has the original filename because it contains important information.
16:22 <+bridge_> <melon> 
16:22 <+bridge_> <melon> **Download the debug symbols**  
16:22 <+bridge_> <melon>     Get them from https://ddnet.org/downloads/symbols/ with matching git hash and platform (win32/64, steam/non-steam).  
16:22 <+bridge_> <melon>    - The git hash and platform are found in the filename of the crash dump.
16:22 <+bridge_> <melon>    - If the filename does not contain `steam`, then it is the standalone version.
16:22 <+bridge_> <melon> 
16:22 <+bridge_> <melon>  **Install binutils**  
16:22 <+bridge_> <melon>    Install `addr2line` and `objdump` using your system's package manager or from msys2 for Windows:
16:22 <+bridge_> <melon>    - **[MSYS2 binutils](https://packages.msys2.org/packages/mingw-w64-x86_64-binutils)**
16:22 <+bridge_> <melon>    - **Ubuntu/Debian**: `sudo apt install binutils`
16:22 <+bridge_> <melon> 
16:22 <+bridge_> <melon>  **Run the script**  
16:23 <+bridge_> <melon> ```sh
16:23 <+bridge_> <melon>    bash scripts/parse_drmingw.sh <path to server/client symbols exe> <path to server/client crash dump>
16:23 <+bridge_> <melon> ```
16:23 <+bridge_> <melon>    - The script is found under `./scripts/`
16:23 <+bridge_> <melon>    - On Windows, you must explicitly run with `bash` and not in a Windows terminal.
16:23 <+bridge_> <melon> 
16:23 <+bridge_> <melon> 
16:23 <+bridge_> <melon> i'd put this right before the DDNet DB stuff on the readme, thoughts before i PR?
16:25 <+bridge_> <kebscs> I don't want to "use it too"
16:25 <+bridge_> <kebscs> It's already like that
16:25 <+bridge_> <kebscs> I'm just adding client site handling
16:26 <+bridge_> <melon> is it `stacktraces` or `stack traces`
16:26 <+bridge_> <melon> is it `stacktraces` or `stack traces` :omo:
16:26 <+bridge_> <heinrich5991> https://en.wiktionary.org/wiki/stack_trace
16:26 <+bridge_> <heinrich5991> according to wiktionary 🤷‍♀️
16:27 <+bridge_> <robyt3> Seems good, I was never really sure where to put it though. I feel like the readme has too many chapters and we should split it at some point.
16:28 <+bridge_> <kebscs> I'm just adding client side handling
16:35 <+bridge_> <melon> hmm Roby, given your comment on #10839
16:35 <+bridge_> <melon> 
16:35 <+bridge_> <melon> maybe we should, at this point go with a  `docs/` setup or something similar? we could split up the readme nicely with that. (and not have random documentation across the codebase)
16:35 <+bridge_> <melon> 
16:35 <+bridge_> <melon> > I think it belongs in a separate document, e.g. src/engine/client/README-sound.md
16:35 <+bridge_> <melon> 
16:35 <+bridge_> <melon> i.e
16:35 <+bridge_> <melon> 
16:35 <+bridge_> <melon> readme.md
16:35 <+bridge_> <melon> docs/debugging.md
16:35 <+bridge_> <melon> docs/contributing.md
16:36 <+bridge_> <DDNet> https://github.com/ddnet/ddnet/pull/10839
16:36 <+bridge_> <melon> hmm Roby, given your comment on #10839
16:36 <+bridge_> <melon> >  I think it belongs in a separate document, e.g. src/engine/client/README-sound.md
16:36 <+bridge_> <melon> maybe we should, at this point go with a  `docs/` setup or something similar? we could split up the readme nicely with that. (and not have random documentation across the codebase)
16:36 <+bridge_> <melon> i.e
16:36 <+bridge_> <melon> 
16:36 <+bridge_> <melon> readme.md
16:36 <+bridge_> <melon> docs/debugging.md
16:36 <+bridge_> <melon> docs/contributing.md
16:36 <+bridge_> <melon> hmm Roby, given your comment on #10839
16:36 <+bridge_> <melon> >  I think it belongs in a separate document, e.g. src/engine/client/README-sound.md
16:36 <+bridge_> <melon> maybe we should, at this point go with a  `docs/` setup or something similar? we could split up the readme nicely with that. (and not have random documentation across the codebase)
16:36 <+bridge_> <melon> e.g
16:36 <+bridge_> <melon> 
16:36 <+bridge_> <melon> readme.md
16:36 <+bridge_> <melon> docs/debugging.md
16:36 <+bridge_> <melon> docs/contributing.md
16:37 <+bridge_> <melon> hmm Roby, given your comment on #10839
16:37 <+bridge_> <melon> >  I think it belongs in a separate document, e.g. src/engine/client/README-sound.md
16:37 <+bridge_> <melon> maybe we should, at this point go with a  `docs/` setup or something similar? we could split up the readme nicely with that. (and not have "random" documentation across the codebase)
16:37 <+bridge_> <melon> e.g
16:37 <+bridge_> <melon> 
16:37 <+bridge_> <melon> readme.md
16:37 <+bridge_> <melon> docs/debugging.md
16:37 <+bridge_> <melon> docs/contributing.md
16:38 <+bridge_> <robyt3> Yeah, I would also go with a `docs/` approach. It's in .gitignore though because that's also the output folder for Doxygen
16:38 <+bridge_> <robyt3> https://github.com/libsdl-org/SDL/tree/main/docs
16:42 <+bridge_> <melon> hmm :/
16:42 <+bridge_> <robyt3> Just ignore `docs/html` and `docs/warn.log` instead I guess
16:48 <+ChillerDragon> can we ban melon?
16:50 <+bridge_> <melon> what did i do now
16:52 <+bridge_> <byfox> Hehe
16:52 <+bridge_> <byfox> https://cdn.discordapp.com/attachments/293493549758939136/1466098545512943739/image.png?ex=697b822c&is=697a30ac&hm=9864b574fd26fc8398899825e11df025a9a83d3760f244f007f505f61399d71b&
16:52 <+bridge_> <byfox> :)
16:52 <+bridge_> <byfox> https://cdn.discordapp.com/attachments/293493549758939136/1466098545512943739/image.png?ex=697b822c&is=697a30ac&hm=9864b574fd26fc8398899825e11df025a9a83d3760f244f007f505f61399d71b&
16:53 <+bridge_> <byfox> I'm still working on multithreading somehow :)
16:53 <+bridge_> <byfox> 
16:53 <+bridge_> <byfox> The main problem was with the network, I did it, thanks 0xfaulty
16:53 <+bridge_> <byfox> https://cdn.discordapp.com/attachments/293493549758939136/1466098545512943739/image.png?ex=697b822c&is=697a30ac&hm=9864b574fd26fc8398899825e11df025a9a83d3760f244f007f505f61399d71b&
16:53 <+ChillerDragon> @melon u is spammer
16:55 <+ChillerDragon> hmmm..
16:56 <+ChillerDragon> i want to have the same oder in header and source file and a clean public api and yet still have the source file read cronologically smh
16:56 <+ChillerDragon> https://zillyhuhn.com/cs/.3624459c-894a-40a9-b7d4-11359f579a02.png
16:56 <+ChillerDragon> aaaaa xd
17:23 <+bridge_> <melon> Chiller gon hate me for that one
17:29 <+bridge_> <alw5> why isnt 11685 not merged ?
17:29 <+bridge_> <alw5> #11685
17:29 <+bridge_> <DDNet> https://github.com/ddnet/ddnet/pull/11685
17:29 <+bridge_> <alw5> what
17:29 <+bridge_> <qxdfox> it is
17:29 <+bridge_> <alw5> wrong one xd
17:29 <+bridge_> <alw5> #11665
17:29 <+bridge_> <DDNet> https://github.com/ddnet/ddnet/pull/11665
17:29 <+bridge_> <alw5> why isnt this merged
17:29 <+bridge_> <melon> i mean, you practically said "why is 11685" merged
17:29 <+bridge_> <melon> :WICKED:
17:29 <+bridge_> <alw5> why isnt 11685 merged ?
17:30 <+bridge_> <alw5> omg my brain isnt braining
17:30 <+bridge_> <alw5> why isnt #11665 merged
17:30 <+bridge_> <DDNet> https://github.com/ddnet/ddnet/pull/11665
18:24 <+bridge_> <heinrich5991> without looking at it: probably nobody reviewed it yet
18:24 <+bridge_> <heinrich5991> what are you looking for?
18:27 <+bridge_> <alw5> im looking for it to be merged
18:31 <+bridge_> <alw5> because i find the new change annoying
18:36 <+bridge_> <louis> todo: add new map requirements for tiles in contributing.md
18:38 <+bridge_> <nyrrcly> а
19:03 <+ChillerDragon> yo @louis someone just reported that the comfort entities do not work on fng
19:03 <+ChillerDragon> wanna add a fng version of it?
19:04 <+ChillerDragon> i already got this report twice
19:05 <+bridge_> <manikoo> yo
19:08 <+bridge_> <louis> ChillerDragon: i am not proud of those entities anymore so probably not. do custom assets even support fng modes?
19:08 <+bridge_> <louis> i can maybe edit them so they're compatible
19:17 <+ChillerDragon> yes
19:17 <+ChillerDragon> just create a fng.png
19:17 <+ChillerDragon> copy a existing fng.png as reference and copy paste ur stuff in there and turn urs into a folder i think
21:23 <+bridge_> <byfox> > copy a existing fng.png as reference and copy paste ur stuff in there and turn urs into a folder i think
21:23 <+bridge_> <byfox> ```
21:23 <+bridge_> <byfox> 2026-01-29 01:22:32 E png: failed to open file for reading. filename='assets/entities/comfort/fng.png'
21:23 <+bridge_> <byfox> ```
21:23 <+bridge_> <byfox> You think correctly :)
21:24 <+bridge_> <byfox> > copy a existing fng.png as reference and copy paste ur stuff in there and turn urs into a folder i think
21:24 <+bridge_> <byfox> ```
21:24 <+bridge_> <byfox> 2026-01-29 01:22:32 E png: failed to open file for reading. filename='assets/entities/comfort/fng.png'
21:24 <+bridge_> <byfox> ```
21:24 <+bridge_> <byfox> You think correctly :) // I don't have this file, but this is just an example of the path
21:29 <+bridge_> <byfox> Another prediction error caused by bug #11360
21:29 <+bridge_> <byfox> 
21:29 <+bridge_> <byfox> If you press ESC and receive a weapon at that moment, the weapon acquisition sound will play continuously, as well as when a player spawns.
21:29 <+bridge_> <byfox> 
21:29 <+bridge_> <byfox> Is there already an issue created on this topic?
21:29 <+bridge_> <DDNet> https://github.com/ddnet/ddnet/pull/11360
21:36 <+bridge_> <robyt3> No, there's no issue for this yet
21:37 <+bridge_> <byfox> Then I'll make a video now and create an issue.
21:42 <+bridge_> <kebscs> pick up weapon same time as pressing esc?
21:45 <+bridge_> <byfox> I can't tell you exactly, I always catch this thing in the nightly build
21:45 <+bridge_> <byfox> 
21:45 <+bridge_> <byfox> Now I want to catch it again and see if it hit 19.7.
21:49 <+bridge_> <art0007i> Maybe it's similar to a bug I found with unhooking and opening chat breaking prediction (#10463)
21:49 <+bridge_> <DDNet> https://github.com/ddnet/ddnet/issues/10463
21:51 <+bridge_> <byfox> In the last "nightly", I can’t reproduce it in any way, if I come across it again in the near future, I’ll tell you about it
21:51 <+bridge_> <robyt3> Sound prediction with #11360 is not included in the 19.7 release
21:51 <+bridge_> <DDNet> https://github.com/ddnet/ddnet/pull/11360
21:56 <+bridge_> <learath2> Guys I learned a lot about LLVM, can we do something ddnet x LLVM? 😄
22:11 <+bridge_> <milkeeycat> I want to do something with WASM but I have no idea what :feelsbadman:
22:15 <+bridge_> <melon> Cross language ddnet modding :poggers2:
22:18 <+bridge_> <melon> I wonder if you could utilize llvm IR for stuff like procedual map generation in real time, so you could actually create geometry dash type maps :kek:
22:24 <+bridge_> <louis> geometry dash has procedural map generation?
22:24 <+bridge_> <louis> any typst fans here? So much better than latex in my experience
22:50 <+bridge_> <learath2> It is quite nice how well made the graph stuff is in LLVM ngl
22:51 <+bridge_> <learath2> I wrote a couple very rudimentary optimization passes and it's a breeze with all the support you have. SCCs, DominatorTrees, Loop simplification
22:53 <+bridge_> <chairn> https://unit42.paloaltonetworks.com/real-time-malicious-javascript-through-llms/
23:51 <+bridge_> <patiga> ingame, players showed me that if the hold laser, the fire rate is slower when you enter spec o.o
23:52 <+bridge_> <Assa> this is true, it's one tick slower due to somebody not understanding direct input
23:52 <+bridge_> <Assa> tater checked that