00:10 < bridge> But shouldn't it be already on Steam if it's intended? 00:10 < bridge> At least have Steam players a chance to utilize it 09:02 < bridge> https://blog.cloudflare.com/introducing-oxy/ 09:03 < bridge> We can say rust runs half the internet now kek 09:03 < bridge> Since cloudflare is used by half the internet 10:19 <+ChillerDragon> im sure cloudflare also uses shellscripts 10:20 <+ChillerDragon> i guess rust runs half the internet since its integrated in the linux kernel 10:22 < bridge> I like your constant arbitrary estimates 11:18 < bridge> https://www.fastcompany.com/90846903/most-innovative-companies-consumer-electronics-2023 11:18 < bridge> https://cdn.discordapp.com/attachments/293493549758939136/1081158574509654086/image.png 11:18 < bridge> gigachad 11:18 < bridge> > By trusting its users instead of locking them in, Valve has produced one of the most exciting consumer tech products in years. 11:18 < bridge> wish more did this 12:02 < bridge> closed source 12:03 < bridge> sadge 12:06 < bridge> Can someone tell me how i can display text with the cursor without the Outline or with a modified Outline? Mb what kind of flags do need? I'm tired, and I don't want to rack my brain reading code and debugging it step by step. The code there is confusing for me, need rack my brains for about an hour. For example the transfer modifiable lvalue ref without staging of but with default colors. 12:11 < bridge> Just make the outline transparent 12:11 < bridge> Setoutlinecolor 12:13 < bridge> setter textcolor work finy, but setter outline color does not work, bindings CTextCursor, and TextEx 12:13 < bridge> Mhh they should work 12:14 < bridge> Show your code then it's easier 12:15 < bridge> Ah you need to use text containers. Apparently we set to default I dunno why 12:16 < bridge> https://cdn.discordapp.com/attachments/293493549758939136/1081173263314599956/image.png 12:16 < bridge> Text containers allow to overwrite the build text color as u do. And additionally allow to set the render color 12:16 < bridge> The problem is. Text has creation color and runtime color 12:17 < bridge> Use text containers then u can control both 12:17 < bridge> Textex will overwrite the runtime color 12:18 < bridge> All right, thanks, I'll give it a try 12:47 < bridge> Murpi best python dev 12:48 < bridge> no, heinrich is the best python dev 12:48 < bridge> Then u are his ghost writer 15:45 < bridge> @Learath2 what do you think about allowing map download via ipfs? 15:45 < bridge> probably a bad idea because it's a little bit p2p and might leak your IP address to someone else than just the server 15:46 < bridge> That'd be my only concern with it. Otherwise sounds like a good idea actually 15:48 < bridge> is there a way to make it safe-ish? 15:48 < bridge> or is there no way to fix it (and it's a dealbreaker)? 15:49 < bridge> what if a map contains illegal material 15:50 < bridge> ipfs download would allow the client to download the map from anyone who offers it 15:50 < bridge> yeah so that guy is a upload in terms of law 15:50 < bridge> I don't understand how it relates to illegal material 15:50 < bridge> ye 15:51 < bridge> uploading illegal material is sued more heavily than downloading 15:51 < bridge> ah, you mean if the client also provides the downloaded maps via ipfs 15:52 < bridge> i thought that was ur idea 15:52 < bridge> another disadvantage would maybe be cable internet suffers a lot from uploading 15:52 < bridge> my idea was first that we could allow the client to download maps via ipfs 15:53 < bridge> not upload 15:53 < bridge> but upload is also interesting, though even more involved as you pointed out 15:53 < bridge> oh i actually thought that players host the map pool 15:53 < bridge> yea, that would make it a lot more resilient at least 15:54 < bridge> in that context i meant that p2p download is also an upload at the same time 15:54 < bridge> <คุณเป็นคนปัญญาอ่อน❤> MODER PLEASE BAN @moriss10_25Zz 15:55 < bridge> #reports 15:57 < bridge> yea, makes sense 16:00 < bridge> @Learath2 mind muting @moriss10_25Zz? he's apparently an ass in voice chat 16:01 < bridge> aren't you a discord mod? 16:01 < bridge> I don't see the option to do it 16:01 < bridge> phone uses.. ^^ 16:01 < bridge> Mods can do it iirc I remeber giving that perm 16:01 < bridge> I'm on a computer 16:01 < bridge> I guess I just don't have the rights 16:03 < bridge> I gave him a timeout instead. Couldn't see the mute button on mobile 😛 16:03 < bridge> ^^ 16:03 < bridge> ah, interesting 16:03 < bridge> didn't know timeout would work for that 16:03 < bridge> but apparently that also prevents joining voice channels 16:04 < bridge> timeout prevents communication with the server 16:04 < bridge> you could troll people by joining a voice channel and leaving it repeatedly 16:05 < bridge> it makes a sound that most people don't have muted 16:08 < bridge> we have server mute, yea 16:47 < bridge> probs mitigated by having lot of seeders? 16:51 < bridge> @Ryozuki what do you say to the new ryzens? 16:51 < bridge> the cpus? 16:51 < bridge> i mean they look pog 16:51 < bridge> i havent tried any 16:52 < bridge> i am a bit disappointed. it feels like AMD purposly made the clock lower so their next generation is again without x3d cache 16:52 < bridge> similar to what they done with the ryzen 7000 series 16:52 < bridge> clock? 16:52 < bridge> isnt the clock the ghz 16:52 < bridge> arent they at 5.5 16:52 < bridge> or smth 16:53 < bridge> i think they made it lower than the 7000 without x3d 16:53 < bridge> oh u mean 16:53 < bridge> another generation than 7000? 16:53 < bridge> where 16:53 < bridge> i didnt know 16:53 < bridge> thought u were talking about 7000 16:53 < bridge> i mean there is 7000 and 16:53 < bridge> 7000 x3d 16:53 < bridge> ahh 16:53 < bridge> xd 16:54 < bridge> yeah x3d is not that worth ithink 16:54 < bridge> and the performance gain is relatively small. and i assume amd did that on purpose 16:54 < bridge> https://cdn.discordapp.com/attachments/293493549758939136/1081243239358672936/image.png 16:54 < bridge> https://www.phoronix.com/benchmark/result/amd-ryzen-9-7950x3d-linux-review-benchmarks/geometric-mean-of-all-test-results-result-composite-ar97lrb.svgz 16:54 < bridge> https://www.phoronix.com/benchmark/result/amd-ryzen-9-7950x3d-linux-review-benchmarks/cpu-power-consumption-monitor-ptssm.svgz 16:54 < bridge> but they are pretty energy efficient like they are 16:55 < bridge> well boost clock is only single core i guess 16:55 < bridge> when all cores are at 100% they probs at base clock 16:55 < bridge> no 16:55 < bridge> on mine 16:55 < bridge> they all bost at like 0.10 less 16:56 < bridge> but they all boost 16:56 < bridge> when 100% 16:56 < bridge> 5800x boost is 4.7 16:56 < bridge> single core it reaches 4.75 on mine 16:56 < bridge> then why do the x3d suck so hard xD 16:56 < bridge> when all cores are full they are at 4.6x 16:56 < bridge> and why would they lower the base clock 16:57 < bridge> i think the 3dx is faster due to cache? 16:57 < bridge> not all is base clock i guess 16:57 < bridge> yeah, but then is the question if they purposly made them slower 16:57 < bridge> the 3dx has 120mb l3 cache 16:57 < bridge> vs 64 16:57 < bridge> to not be much faster 16:57 < bridge> maybe its a limitation idk 16:57 < bridge> afaik, AMD also disallows overclocking on them 16:58 < bridge> kinda weird xD 16:58 < bridge> i smell a conspiracy 16:58 < bridge> <судный день.> maybe they are already at their maximal 17:43 < bridge> @Ryozuki you are biggest rust fan, I need an async db library, is sqlx still the goto? 17:46 < bridge> if u dont want a ORM yes 17:46 < bridge> i kinda like sqlx a lot 17:47 < bridge> also use the sqlx-cli 17:47 < bridge> I wonder if the `Any` driver improved any from the last I used it 17:47 < bridge> its a nice tool for migrations 17:47 < bridge> Yeah migrations are cool 17:47 < bridge> just dont use the any thing 17:47 < bridge> sql is not that generic 17:47 < bridge> its better to use postgresql 17:47 < bridge> :pepeW: 17:47 < bridge> and use its full potential 17:47 < bridge> but idk about that, since i only use psql 17:47 < bridge> psql is glory 17:47 < bridge> psql is now my rust for dbs 17:48 < bridge> I'll just do sqlite 17:49 < bridge> https://market.xivhub.org/ 17:49 < bridge> powered by rust and sqlx btw 17:49 < bridge> and psql 17:50 < bridge> @Learath2 if u want reference code: 17:50 < bridge> ofc my website is open source 17:50 < bridge> AGPL 17:50 < bridge> nothing less 17:50 < bridge> 😄 17:50 < bridge> im such a maximalist 17:50 < bridge> look at main.rs 17:50 < bridge> the denys 17:51 < bridge> ``` 17:51 < bridge> #![forbid(unsafe_code)] 17:51 < bridge> #![deny(warnings)] 17:51 < bridge> #![deny(clippy::missing_const_for_fn)] 17:51 < bridge> #![deny(clippy::nursery)] 17:51 < bridge> #![deny(clippy::pedantic)] 17:51 < bridge> #![allow(clippy::missing_errors_doc)] 17:51 < bridge> #![allow(clippy::module_name_repetitions)] 17:51 < bridge> #![allow(clippy::missing_panics_doc)] 17:51 < bridge> ``` 17:51 < bridge> pedantic kek 17:51 < bridge> https://tenor.com/view/ferris-rust-rustlang-crab-cute-gif-26396486 17:52 < bridge> @Learath2 btw i learnt that if u want to do a count 17:52 < bridge> in psql there is no looose index 17:52 < bridge> but u can do this 17:52 < bridge> https://wiki.postgresql.org/wiki/Loose_indexscan 17:52 < bridge> recursive CTE 17:52 < bridge> my query went from 500ms to 60ms 17:54 < bridge> Loose indexscans are cool :wechat_cool_guy: 17:56 < bridge> That's a cool trick, didn't know you could do that with psql 17:56 < bridge> db people are just magicians 18:40 < bridge> indeed 19:09 < bridge> This is the most rust behaviour I've ever seen 😛 19:09 < bridge> https://cdn.discordapp.com/attachments/293493549758939136/1081277265175912528/image.png 19:20 < bridge> @Learath2 i mean 19:20 < bridge> clap is rly gud 19:20 < bridge> and the most easy to use rn 19:20 < bridge> where did u see this btw? 19:20 < bridge> I was just googling around for what is available nowadays found it on some guys blog 19:20 < bridge> ah well 19:20 < bridge> its a blog 19:21 < bridge> @Learath2 check this out 19:21 < bridge> https://blessed.rs/crates 19:21 < bridge> It's classic rust user behaviour though. Everything is prescriptive 😄 19:21 < bridge> > An unofficial guide to the Rust ecosystem 19:21 < bridge> smh 19:21 < bridge> this is like judging an entire community cuz a guy blog post 19:22 < bridge> altho i agree with clap 19:22 < bridge> its just so useful 19:22 < bridge> u get manpages, help, etc 19:22 < bridge> and bash autocomplete 19:22 < bridge> It's just an example I found in the wild, it's very very common 19:22 < bridge> Didn't you tear heinrich a new one because he didn't agree with clippy on some things? 😄 19:22 < bridge> haha 19:22 < bridge> well i dont like to generalize 19:22 < bridge> thats just how i am 19:23 < bridge> im sure python ppl will tell u should use argparse 19:23 < bridge> or smth 19:23 < bridge> i havent touched much python 19:23 < bridge> <<~{Barsik}~>> oh wow 19:23 < bridge> The only problem I'd ever have with clap is it's insane binary inflation 19:23 < bridge> <<~{Barsik}~>> you last used sqlx in 2021 19:24 < bridge> An entire extra 500kb or so just to parse command line arguments 19:24 < bridge> This is very cool. I was looking for something like this 19:24 < bridge> u should disable default features 19:24 < bridge> and enable what uwant 19:24 < bridge> lexopt looks about what I like in my argparse 19:26 < bridge> oh ye 19:26 < bridge> @Learath2 if u are logging 19:26 < bridge> use tracing 19:26 < bridge> its overtaking log 19:26 < bridge> Yeah, that's what I went for 19:27 < bridge> log with async wouldn't be very nice anyway 19:27 < bridge> tracing works without async too 19:27 < bridge> and its just superior 19:27 < bridge> objectively 19:27 < bridge> if u use all the features its rly nice 19:27 < bridge> spans etc 19:27 < bridge> and instrumentation 19:27 < bridge> Yeah spans are cool 19:28 < bridge> u put #instrument 19:28 < bridge> on top of a function 19:28 < bridge> it logs all calls with the args 19:28 < bridge> and u can skip verbose args 19:28 < bridge> like the db connection xd 19:29 < bridge> also use once_cell over lazy static 19:29 < bridge> im sure u will use regex 19:29 < bridge> ur doing the bot right 19:29 < bridge> Yeah, I'm finally giving it another go now that I have a little free time 19:29 < bridge> There are some very annoying bugs in it 19:29 < bridge> @Ryozuki what are your top5 crates? xd 19:30 < bridge> tracing, tokio, rand, sqlx, itertools 19:30 < bridge> oh 19:30 < bridge> the obvious 1st is 19:30 < bridge> serde 19:30 < bridge> serde is a engineering miracle 19:30 < bridge> cant say it enough 19:31 < bridge> its stupid fast and convenient xd 19:31 < bridge> if u dont need as much options u can also use mini serde 19:31 < bridge> it doesnt support renaming and such 19:31 < bridge> serde really feels like magic 19:31 < bridge> it rly is godlike 19:31 < bridge> there is no match even on corpo langs like java 19:31 < bridge> its superior 19:31 < bridge> why does it feel like magic to you? xd 19:31 < bridge> did u ever use it? 19:32 < bridge> it supports renaming, renaming in X convention, struct flattening, enum tagging, etc 19:32 < bridge> yeah, but i mean what makes it magic to you? 19:32 < bridge> 19:32 < bridge> to me its rust macros that enables serde's power 19:32 < bridge> All the proc macro magic makes it insanely easy to use 19:32 < bridge> yeah 19:32 < bridge> its just rly easy to use 19:32 < bridge> e.g in the js json world they like their stupid namingSense 19:32 < bridge> so renaming to rust naming_sense is rly easy 19:33 < bridge> Well, if I hand you just proc macros, it'd be pretty hard to implement something like serde, that's why it's impressive I guess 19:33 < bridge> its also blazing fast 19:33 < bridge> it uses a visitor pattern iirc 19:34 < bridge> i have to say if tokio would have condition variables 19:34 < bridge> 19:34 < bridge> i like: 19:34 < bridge> 1. tokio 19:34 < bridge> 2. rayon 19:34 < bridge> 3. bincode (uses serde internally) 19:34 < bridge> 4. flate2 (bcs really ez to use) 19:34 < bridge> 5. ash (bcs vulkan, no other reason xd) 19:34 < bridge> ``` 19:34 < bridge> DOM STRUCT 19:34 < bridge> ======= serde_json ======= parse|stringify ===== parse|stringify ==== 19:34 < bridge> data/canada.json 320 MB/s 430 MB/s 580 MB/s 310 MB/s 19:34 < bridge> data/citm_catalog.json 420 MB/s 560 MB/s 710 MB/s 880 MB/s 19:34 < bridge> data/twitter.json 300 MB/s 910 MB/s 550 MB/s 1060 MB/s 19:34 < bridge> 19:34 < bridge> ==== rustc_serialize ===== parse|stringify ===== parse|stringify ==== 19:34 < bridge> data/canada.json 150 MB/s 67 MB/s 120 MB/s 46 MB/s 19:34 < bridge> data/citm_catalog.json 210 MB/s 180 MB/s 140 MB/s 210 MB/s 19:34 < bridge> data/twitter.json 120 MB/s 330 MB/s 87 MB/s 350 MB/s 19:34 < bridge> 19:34 < bridge> ======= simd-json ======== parse|stringify ===== parse|stringify ==== 19:34 < bridge> data/canada.json 380 MB/s 470 MB/s 580 MB/s 19:34 < bridge> data/citm_catalog.json 720 MB/s 760 MB/s 1220 MB/s 19:34 < bridge> data/twitter.json 810 MB/s 890 MB/s 1050 MB/s 19:34 < bridge> 19:34 < bridge> c++ 19:34 < bridge> ==== rapidjson-clang ===================== parse|stringify ==== 19:35 < bridge> data/canada.json 380 MB/s 200 MB/s 19:35 < bridge> data/citm_catalog.json 680 MB/s 1090 MB/s 19:35 < bridge> data/twitter.json 320 MB/s 600 MB/s 19:35 < bridge> 19:35 < bridge> ===== rapidjson-gcc ====================== parse|stringify ==== 19:35 < bridge> data/canada.json 390 MB/s 250 MB/s 19:35 < bridge> data/citm_catalog.json 890 MB/s 1650 MB/s 19:35 < bridge> data/twitter.json 440 MB/s 870 MB/s 19:35 < bridge> 19:35 < bridge> ``` 19:35 < bridge> but i also dont know too many tbf 19:35 < bridge> those are nice 19:35 < bridge> bincode is rly nice 19:35 < bridge> i use it to pack ffxiv item data 19:35 < bridge> rayon is also good 19:35 < bridge> thanks to rust 19:35 < bridge> egui is also nice u want a gui that is ez to use in a custom engine. 19:35 < bridge> 19:35 < bridge> But generally i woudnt say its the greates UI lib 19:35 < bridge> ye egui is nice 19:36 < bridge> immediate mode 19:37 < bridge> oh and cli tools 19:37 < bridge> fd > find 19:37 < bridge> exa > ls 19:37 < bridge> ripgrep > grep 19:37 < bridge> dust > du 19:37 < bridge> delta > whathever diff tool u use 19:37 < bridge> sd > sed 19:38 < bridge> bat is a nice cat with highlighting 19:38 < bridge> u can even use it in vim to have preview code with highlight 19:39 < bridge> thanks for coming to my ted talk 19:44 < bridge> <<~{Barsik}~>> alkahest > bincode change ma mind xd 21:25 < bridge> i configured with -DVULKAN=OFF but am still getting this error from cmake in the generating step 21:25 < bridge> ``` 21:25 < bridge> CMake Error: The following variables are used in this project, but they are set to NOTFOUND. 21:25 < bridge> Please set them or make sure they are set and tested correctly in the CMake files: 21:25 < bridge> /home/mint/Downloads/TaterClient-ddnet-master/VULKAN_INCLUDE_DIRS 21:25 < bridge> used as include directory in directory /home/mint/Downloads/TaterClient-ddnet-master 21:25 < bridge> VULKAN_LIBRARIES 21:25 < bridge> linked by target "DDNet" in directory /home/mint/Downloads/TaterClient-ddnet-master 21:25 < bridge> ``` 21:25 < bridge> seems like someone oops'd on the cmake 21:26 < bridge> can u try on ddnet github 21:26 < bridge> i wonder if its simply bcs its outdated 21:26 < bridge> could be 21:26 < bridge> our 32bit build for example does not use vulkan 21:27 < bridge> and when u reconfigure always try a clean build dir 21:27 < bridge> yeah 21:27 < bridge> sometimes its weird af 21:27 < bridge> it's the cmakecache that's the issue 21:27 < bridge> which will only be created if the makefile generation step succeeds 21:27 < bridge> it's usually the cmakecache that's the issue when you have reconfiguration issues on a dirty build dir 21:28 < bridge> works on latest 21:29 < bridge> but who'd want to turn off vk anyway 😉 21:29 < bridge> probably me who oops'd on the cmake 21:29 < bridge> i like vulkan and use it normally 21:29 < bridge> ;9 21:29 < bridge> but im on a linux live cd rn where using linux would require getting official nvidia drivers & rebooting... which i can't doo 21:29 < bridge> but im on a linux live cd rn where using linux would require getting official nvidia drivers & rebooting... which i can't do 21:29 < bridge> but im on a linux live cd rn where using vlukan would require getting official nvidia drivers & rebooting... which i can't do 21:29 < bridge> rip xd 21:29 < bridge> but im on a linux live cd rn where using vulkan would require getting official nvidia drivers & rebooting... which i can't do 21:30 < bridge> im waiting on the creation of 2 500GB disk images and 1 1TB disk image 21:30 < bridge> https://cdn.discordapp.com/attachments/293493549758939136/1081312641940738148/image.png 21:30 < bridge> need something to keep me occupied 21:38 < bridge> oh hey apparently nouveau can do vulkan now... not performing well though 21:40 < bridge> generally dont expect good perf on nvidia if u dont use the real drivers 21:40 < bridge> even on ogl 21:40 < bridge> yeah 21:40 < bridge> it's chugging pretty hard 21:40 < bridge> this is pretty much worse than having no driver tbh 21:42 < bridge> theoretically getting 45 fps but that feels generous 21:43 < bridge> i can squeeze out 150 at 1080p. meh