00:00 < bridge> time to slep 00:04 < bridge> lol 00:08 < bridge> You would not have thought of what? 00:08 < bridge> (@heinrich5991) 00:14 < bridge> what is this dude making anyway? 00:15 < bridge> chillerdragon: of the email change reset procedure 04:53 < bridge> Vulkan has dynamic rendering support, which simplifies the Renderpass stuff. Also ofc this has overhead. Additionally he can use a library to simplify it if he wants 05:11 < bridge> https://registry.khronos.org/vulkan/specs/1.3-extensions/man/html/VK_KHR_dynamic_rendering.html 07:47 < bridge> some ffxiv plugin related to rendering, i think related to reshade but with support for stuff like better normals? i forgot 07:47 < bridge> Brainwashed by Microsoft xd 07:48 < bridge> xd 07:49 < bridge> ye happens 07:49 < bridge> @Jupeyy_Keks this is my grip, most games use c#, and lot of m$$ fanatics like c# 07:50 < bridge> and ofc modding for them is c# 07:50 < bridge> ans tooling 07:50 < bridge> but i am madness, i use rudt 07:50 < bridge> rust 07:51 < bridge> Yeah quite possible^^ 07:51 < bridge> i am motivated to learn vulkan again 07:52 < bridge> Good luck 07:52 < bridge> @Jupeyy_Keks what rust lib resembles more the c api, or should i use a safe and correct one that avoisa mistakes 07:52 < bridge> for learning 07:53 < bridge> i remember there is ash and vulkano 07:54 < bridge> Mh. Depends on what exactly u want to achieve.. do u want to understand the concepts of creating pipelines and render pass etc. 07:54 < bridge> Or do u want to learn how to optimize something to fit exactly to your needs 07:55 < bridge> Purely rendering wise, VK is probably not much better than opengl 07:55 < bridge> But for emulators it's the perfect match. BCS if it's control 07:55 < bridge> i wanna understand vulkan 07:55 < bridge> so i dont need to search what a PSP is or ask u what i asked 07:56 < bridge> PSO* 07:56 < bridge> my god mobile 07:58 < bridge> I guess I'd start with pure Vulkan. And if u want to do something greater u can still switch to higher level abstractions again 07:58 < bridge> U'll probably find more tutorials anyway 07:58 < bridge> Because CPP dominated field 08:00 < bridge> ye 08:00 < bridge> Invalidating the revert link sent per mail I mean 08:00 < bridge> Sending a notification email to the old seems obvious doesn’t it? But yea properly invalidating it when it happens multiple times is something I would not have implemented if I were to build it. 08:00 < bridge> (@heinrich5991) 08:47 < bridge> @Jupeyy_Keks 08:47 < bridge> https://cdn.discordapp.com/attachments/293493549758939136/1110821264634945597/image.png 08:47 < bridge> u did it 08:47 < bridge> owned 08:47 < bridge> another internet discussion won today 08:47 < bridge> time to go to leddit 08:49 < bridge> i know it sucks and you could have easily done this in 10 seconds but 08:49 < bridge> https://cdn.discordapp.com/attachments/293493549758939136/1110821824176062524/teecord.png 08:49 < bridge> !c 08:49 < chillerbot> ChillerDragon 08:50 < bridge> also what is teecord_bot.png 08:50 < bridge> image for funny chiller irc client 08:51 < bridge> no i mean 08:51 < bridge> oh 08:51 < bridge> so its also an icon? 08:52 < bridge> what 08:53 < bridge> lmao 08:53 < bridge> xd 09:04 < bridge> this also sucks 09:04 < bridge> https://cdn.discordapp.com/attachments/293493549758939136/1110825512907055144/image.png 09:16 <+ChillerDragon> nice teeish! 09:17 <+ChillerDragon> @Voxel did you just made those or what is that? looks nice 09:17 < bridge> i made these 09:17 < bridge> idk 09:17 <+ChillerDragon> pog 09:17 <+ChillerDragon> can i buy? 09:17 < bridge> i j 09:17 < bridge> you dont have to buy these i can just, give you the file LOL 09:17 <+ChillerDragon> nice can you give me 6 files so i dont have to gimp split them 09:18 < bridge> yeah 09:18 <+ChillerDragon> pog pog 09:18 < bridge> giving file is not enough 09:18 < bridge> put a license 09:18 < bridge> cc-by-sa for example 09:18 < bridge> :justatest: 09:19 < bridge> yeayea i know 09:20 < bridge> how do i send? 09:21 < bridge> do i just drop a ZIP file in here 09:22 < bridge> that works 09:25 < bridge> !c please tell me if anything looks weird 09:25 < chillerbot> ChillerDragon 09:25 < bridge> also yes cc_by_sa 09:25 < bridge> https://cdn.discordapp.com/attachments/293493549758939136/1110830851865985085/teecord_new_icons.zip 09:25 < bridge> wait does irc client even support file uploads 09:26 < bridge> should work, it's just a download link to discords website that get's displayed differently in discord 09:26 < bridge> ah 09:26 < bridge> should work, it's just a download link to discord's website that gets displayed differently in discord 09:31 < ws-client> .pdf 09:31 < ws-client> trol voxel 09:31 < ws-client> thanks a lot worked fine looks good on my machine 09:33 < bridge> common practice for graphic designs to send .pdfs to their clients 09:33 < bridge> *designers 09:33 < ws-client> ok mr professional 09:37 < bridge> How dare you use the jupstar argument to defeat him? 09:37 < bridge> thats the internet 09:38 < bridge> ddnet is like a company, i just consulted the local vulkaner 09:38 < bridge> Btw vulkan docs look so amazing wtf 09:38 < bridge> but it loads so slow xD 09:38 < bridge> @Learath2 they even have a xml with the api 09:38 < bridge> so stuff like ash crate 09:38 < bridge> is generated from it 09:38 < bridge> correct bindings 09:38 < bridge> ez 09:39 < bridge> So comprehensive it's like better than a whole standard 09:39 < bridge> Can't you grab a local copy of it? 09:39 < bridge> probably, but i am fine with it, on desktop its ok. but on my old mobile it takes minimum 10 seconds xD 09:41 < bridge> https://registry.khronos.org/vulkan/specs/1.3/html/vkspec.html#_working_group_contributors_to_vulkan 09:41 < bridge> 09:41 < bridge> it's interesting how much contributions even come from google 09:41 < bridge> "that aint work on android bra" 09:53 < bridge> ChillerDragon: is there an equivalent to python wheels for ruby? 09:53 < ws-client> wat was wheels about again? 09:53 < ws-client> opsi that was my dev tab xd 09:54 < bridge> precompiled binaries so my libtw2 huffman users don't have to have rust installed 09:54 < ws-client> ahm 09:55 < ws-client> probably 09:56 < ws-client> my ruby huffman gem requires a c++ compiler 09:57 < ws-client> which is super common for ruby gems so maybe rust would be fine too 09:59 < bridge> @Ryozuki if u understand this diagram while coding vk, i'd say u got the idea 09:59 < bridge> 09:59 < bridge> it basically shows all stuff u also do when coding: 09:59 < bridge> define the pipeline layout 09:59 < bridge> allocate descriptor sets 09:59 < bridge> allocate resources(memory, etc.) 10:00 < bridge> 10:00 < bridge> the programmable shader stages 10:00 < bridge> 10:00 < bridge> and how exactly you have to use memory fences, semaphores etc. with this pipeline "diagram" 10:00 < bridge> https://cdn.discordapp.com/attachments/293493549758939136/1110839573338144808/image.png 10:01 < bridge> https://github.com/luislavena/gem-compiler something like this maybe? @heinrich5991 10:01 < ws-client> @heinrich5991 this seems helpful https://github.com/flavorjones/ruby-c-extensions-explained/tree/main/precompiled 10:01 < bridge> It's not quite wheels tbf but you can generate a gem that just downloads a binary 10:01 < ws-client> virus 10:03 < bridge> You 10:04 < bridge> ye, both links seem helpful. thanks! 10:04 < ws-client> the one i sent was used for sqlite2-ruby https://github.com/sparklemotion/sqlite3-ruby/releases/tag/v1.5.0.rc1 10:04 < ws-client> 3* 10:05 < ws-client> Make sure to fallback to compiling if no architecure is matching 10:08 < ws-client> https://zillyhuhn.com/cs/.1684915691.png 10:08 < ws-client> thanks a lot @Voxel 10:09 < bridge> no problem 10:10 < ws-client> Could someone leak some discord emote images preferably in high res? 10:10 < ws-client> :pepeH: would be nice 10:12 < ws-client> Dang. Website isn't updated yet. 10:12 < ws-client> nah 10:14 < bridge> https://twemoji.twitter.com/ 10:15 < bridge> https://blogs.windows.com/windowsdeveloper/2023/05/23/bringing-the-power-of-ai-to-windows-11-unlocking-a-new-era-of-productivity-for-customers-and-developers-with-windows-copilot-and-dev-home/ 10:15 < bridge> > We have added native support for additional archive formats, including tar, 7-zip, rar, gz and many others using the libarchive open-source project. You now can get improved performance of archive functionality during compression on Windows. 10:15 < bridge> Why are you making this bizarre irc discord hybrid anyway chiller? 10:16 < bridge> it does seem a little unnecessary ^^ 10:16 < bridge> there's definitely too little chat apps out there 10:16 < ws-client> idk i thought its funni was meant as a 20min trol project to see if i can do it 10:16 < ws-client> but now i kinda like it 10:16 < ws-client> i do not like the other irc frontends 10:16 < bridge> windows now supports more archive formats 😮 10:16 < ws-client> and having a web client has some perks like image display etc 10:17 < ws-client> notifications 10:18 < bridge> now that we have the full range of archive formats 10:18 < bridge> what would be the preferred archive format for custom file formats? 10:18 < bridge> still zip, probably? 10:18 < bridge> maybe with better compression 10:19 < bridge> They only had zip and rar, right? 10:19 < bridge> only zip 10:19 < ws-client> omg i just bought winrar license 10:20 < bridge> oh god now in the entire os?? 10:21 < bridge> not only is that bloat it gives off the vibe my pc's going to take over my hardware and become sentient 10:21 < bridge> nice 10:21 < bridge> "ads in settings? windows, install gentoo" 10:21 < bridge> "im sorry, user. im afraid i cant do that" 10:23 < bridge> Perfect time to jump ship from windows 10:24 < bridge> and only limited zip support. I wonder if they have better support now 10:27 < bridge> @Ryozuki I think an out-of-date list of releases on github is worse than none: https://github.com/ddnet/ddnet/releases Do you mind if I delete it again? As I said, I don't want to maintain it manually 10:27 < bridge> ok 10:31 < bridge> windows 10 is safe.... for now.... 10:35 < bridge> Better safe than sorry. Migrate to linux now 10:39 < bridge> no 10:43 < ws-client> @Voxel your images are live on https://chat.zillyhuhn.com/ :rocket: 10:43 < ws-client> ok 10:45 < ws-client> No they arent 10:45 < ws-client> One second 10:45 < ws-client> reload the page? 10:47 < ws-client> Still not live 10:51 < bridge> This webclient with no authentication that lets you pick any nick is definitely not going to end up abused 😄 10:51 < ws-client> https://zillyhuhn.com/cs/.1684918259.png 10:51 < ws-client> works fine for me 10:51 < bridge> oh so that IS an icon 10:52 < ws-client> those are the ones you made 10:52 < ws-client> dont u recognize em? xd 10:52 < bridge> no the bot one at the bottom 10:52 < bridge> You need to size them better tho, that rocket is massive 10:52 < ws-client> Maybe in comarison to your rocket its massive 10:52 < ws-client> :p 10:53 < bridge> I hope your webclient breaks 10:53 < ws-client> :c why 10:53 < bridge> Mean ! 11:44 < bridge> @Robyt3 when i scroll the server browser, i always see the hovered element from the previous frame above/under where my cursor is 11:44 < bridge> 11:44 < bridge> basically the hovered element moved but is still considered hovered 11:44 < bridge> ez to see with mices with fast scroll 11:44 < bridge> also with vsync, so not tearing related 11:45 < bridge> (I have smooth scroll off) 11:47 < bridge> might be related to immediate mode UI always lagging one frame behind? 11:47 < bridge> not sure 11:48 < bridge> should be instant normally 11:49 < bridge> maybe its some optimization for smooth scroll to not recalculate the hover element 12:08 < bridge> only seems to happen specifically when smooth scrolling is off 13:28 < bridge> fun that it was exactly a year ago 😄 13:28 < bridge> yeah 😄 13:28 < ws-client> what is this :0 13:59 < bridge> Hm, I don't get why that breaks 14:02 < bridge> @Ryozuki what was the async handler crate again u send a few days ago 14:02 < bridge> `std::bitset` too small 14:02 < bridge> i cant find 😄 14:02 < bridge> for `TEAM_SUPER == 64` 14:03 < bridge> But then so was out integer from before the bitset. What was happening then? 14:03 < bridge> probably shift outside of the integer? 14:03 < bridge> i.e. wrong team selected 14:03 < bridge> pollster 14:03 < bridge> ty 14:03 < bridge> Ah, I guess that's possible 15:01 < bridge> @Jupeyy_Keks did u rly try vulkano 15:01 < bridge> apparently they have the same versions ofg the functions with a _unchecked, and they claim that if the normal version works, its safe and u can swap to the unchecked 15:02 < bridge> https://cdn.discordapp.com/attachments/293493549758939136/1110915663855370311/image.png 15:02 < bridge> they dont appear in the docs intentionally btw 15:02 < bridge> so u may have missed em 15:03 < bridge> i didnt try it 15:03 < bridge> u can also opt out of the builtin sync mechanism they use 15:03 < bridge> and use ur own 15:03 < bridge> but it's obvious that u can't do the same, at least not in what u'd call safe 15:03 < bridge> idk, i think vulkano has a nice goal, only allow u to do safe correct stuff according to the spec 15:03 < bridge> how do u want to make a memory fence safe without making everything "more" complex? 15:04 < bridge> yes that's nice indeed 15:04 < bridge> anyway 15:04 < bridge> i think im gonna learn it with vulkano 15:04 < bridge> its nice to know errors early 15:04 < bridge> and i wont do obscure things probs 15:04 < bridge> xd 15:04 < bridge> and i dont know what i am doing 15:05 < bridge> u can also enable these layers with normal vulkan 15:05 < bridge> anyway 15:05 < bridge> yeah 15:05 < bridge> > Can handle synchronization on the GPU side for you (unless you choose to do that yourself), as this aspect of Vulkan is both annoying to handle and error-prone. Dependencies between submissions are automatically detected, and semaphores are managed automatically. The behavior of the library can be customized thanks to unsafe trait implementations. 15:05 < bridge> yeah 15:22 < bridge> biggest troll in history 15:22 < bridge> @Learath2 15:22 < bridge> @Ryozuki 15:22 < bridge> learath the biggest troll, do u agree? 15:23 < bridge> Also an easy review opportunity, you can go "goto, declined, wontfix" 15:23 < bridge> i'd be surprised if we dont have a clang-tidy check 15:25 < bridge> what u actually want is a lambda 15:26 < bridge> @Learath2 wait doesnt rcon auth happen after joining 15:26 < bridge> return false if goto is not called 15:26 < bridge> else true 15:26 < bridge> it's just reusing the rcon passwords for reserved slots 15:26 < bridge> oh goto 15:26 < bridge> It doesn't auth, I thought about using it to pre-auth too but mods might want to sneak on to the server 15:26 < bridge> ur evil learath 15:26 < bridge> as a witch 15:26 < bridge> we should burn u 15:27 < bridge> xDDDDDDD 15:27 < bridge> in rust its ok cuz its not spelled goto but break 'label' 15:27 < bridge> so its fine 15:27 < bridge> in rust its ok cuz its not spelled goto but break 'label 15:27 < bridge> this also isnt a loop 15:27 < bridge> true 15:27 < bridge> I could have made an extremely ugly extra one use function or duplicated branches or inverted branches but I thought this would be an amazing place to demonstrate the power of goto and how pretty it can make code 15:27 < bridge> lambda 15:27 < bridge> Bask in it's glorious power to break out of blocks 15:28 < bridge> @Jupeyy_Keks u can use it here anyway i think 15:28 < bridge> to return to a outer scope 15:28 < bridge> he is using goto to quit 2 scopes 15:28 < bridge> learath has no math brain 15:28 < bridge> he uses weird jumps 15:28 < bridge> pls don't fail my clang-tidy 15:28 < bridge> pls don't fail my clang-tidy 15:28 < bridge> pls don't fail my clang-tidy 15:28 < bridge> pls don't fail my clang-tidy 15:28 < bridge> pls don't fail my clang-tidy 15:29 < bridge> pls don't fail my clang-tidy 15:29 < bridge> pls don't fail my clang-tidy 15:29 < bridge> I could have used an ugly lambda, but I decided to demonstrate the glorious demonized goto 15:29 < bridge> ```rust 15:29 < bridge> fn main() { 15:29 < bridge> 'outer: loop { 15:29 < bridge> println!("Entered the outer loop"); 15:29 < bridge> 15:29 < bridge> 'inner: loop { 15:29 < bridge> println!("Entered the inner loop"); 15:29 < bridge> 15:29 < bridge> // This would break only the inner loop 15:29 < bridge> //break; 15:29 < bridge> 15:29 < bridge> // This breaks the outer loop 15:29 < bridge> break 'outer; 15:29 < bridge> } 15:29 < bridge> 15:29 < bridge> println!("This point will never be reached"); 15:29 < bridge> } 15:29 < bridge> 15:29 < bridge> println!("Exited the outer loop"); 15:29 < bridge> } 15:29 < bridge> ``` 15:29 < bridge> in rust its fine cuz its not goto 15:29 < bridge> its break 15:29 < bridge> kek 15:30 < bridge> imho still not always the best solution, really depends on complexity of the loop 15:32 < bridge> tension increases as clang-tidy comes to compile the server code 15:32 < bridge> comes close 15:34 < bridge> @Learath2 do u rq, if goto fails? 15:36 < bridge> I doubt it will, but I'd be disappointed in whoever wrote the rules 15:37 < bridge> i'd be surprised if it doesn't 15:37 < bridge> Ah, actually these will be the only 2 gotos in the code it might be there then 15:37 < bridge> EPIC 15:37 < bridge> `cppcoreguidelines-avoid-goto` yeah disappointed in all of you 15:37 < bridge> ``` 15:37 < bridge> FAILED: CMakeFiles/game-server.dir/src/engine/server/server.cpp.o 15:37 < bridge> /usr/local/bin/cmake -E __run_co_compile --tidy="clang-tidy;-warnings-as-errors=*;--extra-arg-before=--driver-mode=g++" --source=/home/runner/work/ddnet/ddnet/src/engine/server/server.cpp -- /usr/bin/c++ -DCONF_DEBUG -DCONF_INFORM_UPDATE -DCONF_OPENSSL -DCONF_VIDEORECORDER -DGAME_RELEASE_VERSION=\"16.9\" -DGLEW_STATIC -I/home/runner/work/ddnet/ddnet/clang-tidy/src -I/home/runner/work/ddnet/ddnet/src -I/home/runner/work/ddnet/ddnet/src/rust-brid 15:37 < bridge> /home/runner/work/ddnet/ddnet/src/engine/server/server.cpp:1473:7: error: avoid using 'goto' for flow control [cppcoreguidelines-avoid-goto,-warnings-as-errors] 15:37 < bridge> goto join; 15:37 < bridge> ^~~~~~~~~ 15:38 < bridge> /home/runner/work/ddnet/ddnet/src/engine/server/server.cpp:1492:4: note: label defined here 15:38 < bridge> join: 15:38 < bridge> ^ 15:38 < bridge> /home/runner/work/ddnet/ddnet/src/engine/server/server.cpp:1484:8: error: avoid using 'goto' for flow control [cppcoreguidelines-avoid-goto,-warnings-as-errors] 15:38 < bridge> goto join; 15:38 < bridge> ^~~~~~~~~ 15:38 < bridge> /home/runner/work/ddnet/ddnet/src/engine/server/server.cpp:1492:4: note: label defined here 15:38 < bridge> join: 15:38 < bridge> ^ 15:38 < bridge> 514 warnings generated. 15:38 < bridge> ``` 15:38 < bridge> Yeah so cool, go pray to djikstra or sth 15:38 < bridge> 🤩 🤩 🤩 🤩 🤩 🤩 🤩 🤩 15:38 < bridge> thanks djikstra 15:38 < bridge> thanks for taking pathes so quickly 15:38 < bridge> thanks for taking paths so quickly 15:39 < bridge> Too much education is as bad as no education it seems 15:39 < bridge> 😂 😂 😂 15:39 < bridge> https://github.com/ddnet/ddnet/pull/5235 15:39 < bridge> B.S., M.S., Ph.D. yet unable to follow a simple `goto` 15:40 < bridge> one thing less to care 15:40 < bridge> 😉 15:40 < ws-client> lerato fellow goto enjoyer pog 15:40 < bridge> u can always say, but in these 10 lines of code its easy to follow 15:40 < bridge> !c >.< 15:40 < chillerbot> ChillerDragon 15:41 < bridge> Exactly, use your common sense instead of some old dudes guidelines that he stole from math people 15:41 < bridge> In these 10 lines of code, in this exact usecase it's the better pattern 15:41 < bridge> then the code is a design flaw 😂 15:42 < bridge> No 15:42 < ws-client> ur flawed jopsti 15:43 < ws-client> goto is top10 underrated keywords 15:43 < bridge> goto school 15:43 < bridge> true, indeed a good keyword 15:43 < ws-client> xd 15:44 < bridge> If I was any less confident in being right about this I might be concerned about how against it you are 15:44 < bridge> Anyway, I'll flip it around and make it ugly just as you like, dw 15:44 < ws-client> If the devil would encrypt the hell? would it be shell? 15:45 < ws-client> as in secure hell? 15:45 < ws-client> like sftp etc 15:45 < ws-client> okok wasnt funi i go 15:46 < bridge> i can just repeat what i said last time. 15:46 < bridge> 15:46 < bridge> I generally am not the biggest fan of uncontrolled jumps, this includes: 15:46 < bridge> return 15:46 < bridge> break 15:46 < bridge> 15:46 < bridge> and all the other stuff my brain forgot now 15:46 < bridge> with return i ofc mean return mid func 15:48 < bridge> I think an early return is more readable than an if branch covering the whole function 15:50 < bridge> an early return is different: it's basically an if else 15:50 < bridge> around the whole function 15:50 < bridge> if u mean with early, directly at the start of the function 15:51 < bridge> ah, you're okay with `return`s if they're easily translatable intto huge if-else blocks? 15:51 < bridge> i am ok with returns that don't destroy the flow of a function 15:52 < bridge> so if the if else is mid function 15:52 < bridge> no 15:52 < bridge> nono 15:52 < bridge> return is evil 15:52 < bridge> we must use out pointers 15:53 < bridge> wat 15:53 < bridge> fn(a,b,c, *result) 15:53 < bridge> ah, ez 15:53 < bridge> xd 15:53 < bridge> @Jupeyy_Keks i think csharp has a keyword 15:53 < bridge> out x 15:54 < bridge> instead of reference 15:54 < bridge> https://learn.microsoft.com/en-us/dotnet/csharp/language-reference/keywords/out-parameter-modifier 15:54 < bridge> cool shit 15:54 < bridge> ```csharp 15:54 < bridge> void Method(out int answer, out string message, out string? stillNull) 15:54 < bridge> { 15:54 < bridge> answer = 44; 15:54 < bridge> message = "I've been returned"; 15:54 < bridge> stillNull = null; 15:54 < bridge> } 15:54 < bridge> ``` 15:54 < bridge> is it write only? 15:54 < bridge> imho its better rust 15:54 < bridge> just use tuples 15:54 < bridge> i think c# simply doesnt have first class tuples 15:55 < bridge> idk 15:55 < bridge> yeah in rust for example u can code without using `return` 15:55 < bridge> isnt that epic 15:55 < bridge> xdd 15:55 < bridge> u just omitting it written 15:55 < bridge> no 15:55 < bridge> its implicit at end of scopes 15:55 < bridge> yes 15:55 < bridge> that's what i mean 15:56 < bridge> it guarantees that the flow isn't destroyed 15:56 < bridge>