12:06 <+bridge> [ddnet] would it be possible to add a NetStruct to data types? (e.g. similar to NetIntAny, NetArray, ...) 12:21 <+bridge> [ddnet] or, perhaps not a good idea to use anything recursive 14:43 <+bridge> [ddnet] We seem to have an issue with how we linked libnotify 14:43 <+bridge> [ddnet] cmake succeeds when it's not installed 14:43 <+bridge> [ddnet] yes 14:43 <+bridge> [ddnet] we dont check it 14:43 <+bridge> [ddnet] We should, right? 14:43 <+bridge> [ddnet] would be nicer yes 15:11 <+bridge> [ddnet] @Not Keks the given error for vulkan things not being found is also very meh 15:11 <+bridge> [ddnet] what does it say? 15:12 <+bridge> [ddnet] https://paste.pr0.tips/t7 15:14 <+bridge> [ddnet] ahh i guess https://github.com/ddnet/ddnet/pull/5076/commits/f595318b830653ca13f462e6949fbf80eb7b6860 fixes this 15:19 <+bridge> [ddnet] I can't believe I've been using my computer at 800mhz, things compile so quickly now :3 15:34 <+bridge> [ddnet] in 20 years 15:34 <+bridge> [ddnet] 15:34 <+bridge> [ddnet] i cant believe ive been using my computer at 800ghz, things compile so quickly now xd 15:40 <+bridge> [ddnet] too bad frequency isn't increasing anymore 15:40 <+bridge> [ddnet] https://cdn.discordapp.com/attachments/293493549758939136/972855639594373180/unknown.png 15:41 <+bridge> [ddnet] power wall hits hard 15:42 <+bridge> [ddnet] true story 15:42 <+bridge> [ddnet] i want more single thread performance xd 15:43 <+bridge> [ddnet] now they throw more l1 cache to it and it works xxd 15:43 <+bridge> [ddnet] where is that from, i want more stats like that 15:44 <+bridge> [ddnet] performance related 15:44 <+bridge> [ddnet] also for gpus xd 15:44 <+bridge> [ddnet] hum, Hennessy & Patterson is a good source 15:44 <+bridge> [ddnet] its the biblio for system architect 15:44 <+bridge> [ddnet] its the bible for system architect 15:45 <+bridge> [ddnet] lol, i typed biblio instead of bible, phd really starts to consume me 15:46 <+bridge> [ddnet] too bad they stats always end in 2010 15:46 <+bridge> [ddnet] https://cdn.discordapp.com/attachments/293493549758939136/972857044115808256/unknown.png 15:46 <+bridge> [ddnet] https://cdn.discordapp.com/attachments/293493549758939136/972857072137957486/unknown.png 15:47 <+bridge> [ddnet] modern cpu are nowadays more dense in terms of power than a nuclear reactor 15:47 <+bridge> [ddnet] nice 15:47 <+bridge> [ddnet] and intel probs already beats the sun at this point xd 15:47 <+bridge> [ddnet] hence the power wall, you can't cool them down efficiently 15:47 <+bridge> [ddnet] also dark silicon problem if you have ever heard of it 15:48 <+bridge> [ddnet] intel cryo cooler 15:48 <+bridge> [ddnet] https://cdn.discordapp.com/attachments/293493549758939136/972857503115251742/unknown.png 15:48 <+bridge> [ddnet] cool by activly using electric energy 15:49 <+bridge> [ddnet] https://cdn.discordapp.com/attachments/293493549758939136/972857796687175740/unknown.png 15:49 <+bridge> [ddnet] https://cdn.discordapp.com/attachments/293493549758939136/972857797454729216/unknown.png 15:49 <+bridge> [ddnet] https://cdn.discordapp.com/attachments/293493549758939136/972857796687175740/unknown.png 15:49 <+bridge> [ddnet] are you also facing such problems in your memory designs? 15:49 <+bridge> [ddnet] partly, but not completely 15:50 <+bridge> [ddnet] hence my phd about computing in memory directly because there is way less power density for the moment 15:50 <+bridge> [ddnet] xd 15:50 <+bridge> [ddnet] so obviously, this is not possible 15:50 <+bridge> [ddnet] faster memory = faster effective CPU, i am fine with that 15:51 <+bridge> [ddnet] actually, we would need to slow memory a lil bit, depend on the technology 15:51 <+bridge> [ddnet] Well each cycle is still becoming more efficient, so comparing clock speeds isn't always really fair 15:51 <+bridge> [ddnet] if you get close to super conduction? 15:51 <+bridge> [ddnet] and we can't compute anything there, mostly parallel stuff for AI or BLAS stuff 15:52 <+bridge> [ddnet] it is because power scales with fΒ² 15:52 <+bridge> [ddnet] heat must go somewhere, superconduction don't apply to semiconductor 15:52 <+bridge> [ddnet] with other materials? 15:52 <+bridge> [ddnet] then it's a drastic technology change, we have nothing close in that matter 15:53 <+bridge> [ddnet] maybe optronic might be a game changer 15:53 <+bridge> [ddnet] but we still have a long way to go 15:53 <+bridge> [ddnet] Well yes, so it might actually be good that f isn't increasing anymore 15:53 <+bridge> [ddnet] yes i want my alien hardware 15:53 <+bridge> [ddnet] spintronic is also interesting in the way it works 15:54 <+bridge> [ddnet] in spintronic, the path (copper wire) is actually the gate 15:54 <+bridge> [ddnet] its quite weird, fucking quantum physic 15:54 <+bridge> [ddnet] If we can extract more performance while not having to increase the frequency it's a win, and if raw performance starts to stagnate we might finally get back to the golden age of optimization when people actually cared about writing nice code 15:55 <+bridge> [ddnet] win win πŸ˜› 15:55 <+bridge> [ddnet] tru, it already happens a lot 15:55 <+bridge> [ddnet] script languages get c bindings for slow paths πŸ˜„ 15:55 <+bridge> [ddnet] well, you cna't have more performance without more frequency or more parallelism 15:56 <+bridge> [ddnet] hardware operators have already been quite optimized 15:56 <+bridge> [ddnet] but there is still room 15:56 <+bridge> [ddnet] so we really need a technology shift 15:56 <+bridge> [ddnet] well you can pipeline more, we most definitely can parallelise more 15:56 <+bridge> [ddnet] pipeline more --> more stalls when branching 15:56 <+bridge> [ddnet] 15 stage is around the norm 15:57 <+bridge> [ddnet] --> more people care about writing nice branchless code --> I'm happy πŸ˜› 15:57 <+bridge> [ddnet] in 2005/2006 they had up to 40 stages and reverted because it was highly inefficient 15:57 <+bridge> [ddnet] you cant write branchless code πŸ˜„ 15:58 <+bridge> [ddnet] well, you can with predicated/conditional hardware operator 15:58 <+bridge> [ddnet] not the norm in x86 15:58 <+bridge> [ddnet] but that's what is used for GPU 15:59 <+bridge> [ddnet] but we still have a bit of headroom for transistor sizes and for the overall gaps between the transistors(for logical units) & other components(memory controler etc) 15:59 <+bridge> [ddnet] dont we? 15:59 <+bridge> [ddnet] interesting tab about simd extension: 15:59 <+bridge> [ddnet] https://cdn.discordapp.com/attachments/293493549758939136/972860260433920020/unknown.png 15:59 <+bridge> [ddnet] well, intel is still at 10nm, so they can go down up to 2 or 3nm, but they already took 10 years to industrialize their 10nm process 16:00 <+bridge> [ddnet] 3 silicon atoms would be around 0.6-1nanometer 16:00 <+bridge> [ddnet] ez 16:00 <+bridge> [ddnet] all we need 16:00 <+bridge> [ddnet] TSMC does sell 5 and 7nm now, but technology is slightly different and less efficient (performance wise) than intel's finfet technology 16:00 <+bridge> [ddnet] I think quantum fuckiness prevents us from going that low πŸ˜„ 16:00 <+bridge> [ddnet] silicon is just the base die 16:01 <+bridge> [ddnet] i dunno 16:01 <+bridge> [ddnet] 16:01 <+bridge> [ddnet] i think the quantum effects are only the space between transistors not the transistors itself? 16:01 <+bridge> [ddnet] you need germanium, carbon and some more as well 16:01 <+bridge> [ddnet] but tbh, who the fuck knows xd 16:01 <+bridge> [ddnet] I'm not especially well read in semi conductor technology, but if you go that small the uncertainty in the position of the electron starts being an issue 16:01 <+bridge> [ddnet] and the fabrication process would be a mess 16:02 <+bridge> [ddnet] they already have huge problems to get a good yield from their wafer 16:02 <+bridge> [ddnet] like 50% are thrown out due to defects 16:02 <+bridge> [ddnet] i pay my 2million bucks for 1 trillion fps 16:02 <+bridge> [ddnet] damn, 1ps per frame πŸ˜„ 16:03 <+bridge> [ddnet] Fwiw, I'm not especially concerned about not getting more performance, I think we have plenty enough we are just wasting right now, if we don't get improvements for a while we'll just have to get smarter about our use of cycles 16:04 <+bridge> [ddnet] i bet as soon as x86 is the reason for no more performance 16:04 <+bridge> [ddnet] 16:04 <+bridge> [ddnet] it dies faster than the industry can imagine rn xd 16:05 <+bridge> [ddnet] risc-v ftw !! 16:05 <+bridge> [ddnet] but is risc really good in your argumentation? 16:05 <+bridge> [ddnet] more pipelined instructions = more effective performance 16:05 <+bridge> [ddnet] so would it make sense to make more ccomplex instructions 16:05 <+bridge> [ddnet] wouldn't* 16:06 <+bridge> [ddnet] or better 16:07 <+bridge> [ddnet] the longer the code runs inside the logic unit the better 16:08 <+bridge> [ddnet] you can add extension to it, so not really a problem 16:09 <+bridge> [ddnet] that's what makes it interesting, you can specialize it following your needs 16:09 <+bridge> [ddnet] yes 16:09 <+bridge> [ddnet] i mean it kinda works out for apple 16:09 <+bridge> [ddnet] but not as much as it should tbh 16:09 <+bridge> [ddnet] considering intel beats it with 10nm 16:09 <+bridge> [ddnet] and m1 is a soc 16:10 <+bridge> [ddnet] do you think a RISC processor can scale up in high power regions? 16:10 <+bridge> [ddnet] (signle threaded ofc) 16:11 <+bridge> [ddnet] what do you mean by "high power regions"? 16:11 <+bridge> [ddnet] throw 100 watts against it and it runs better with like ok efficiency curve 16:11 <+bridge> [ddnet] basically what intel does 16:11 <+bridge> [ddnet] nah 16:11 <+bridge> [ddnet] give it 5ghz instead of 4 and u have a bit more performance πŸ˜„ 16:12 <+bridge> [ddnet] because risc is risc, you'll have linear scale up, but most complex computing would not benefit from it directly 16:12 <+bridge> [ddnet] so another cisc design? or still prefer risc? D: 16:13 <+bridge> [ddnet] i still prefer risc, it's easier to design πŸ˜„ 16:13 <+bridge> [ddnet] hehe 16:13 <+bridge> [ddnet] easier to manage as well for the state machine 16:13 <+bridge> [ddnet] x86 is a real mess 16:13 <+bridge> [ddnet] i believe that πŸ˜„ 16:13 <+bridge> [ddnet] variable length instructions is a bad start 16:13 <+bridge> [ddnet] 40 years of backwards compatibility is a design error 16:14 <+bridge> [ddnet] yeah but i defs want to see another cisc 16:14 <+bridge> [ddnet] also x86_64 is not 40 years old πŸ˜„ 16:14 <+bridge> [ddnet] x86 is 16:14 <+bridge> [ddnet] yes but the fuck uses 32bit 16:14 <+bridge> [ddnet] code compiled today (which most use the same x86 prefix) still runs on the good old 8086 16:15 <+bridge> [ddnet] code compiled today (which most use the same x86 prefix instructions) still runs on the good old 8086 16:15 <+bridge> [ddnet] i mean the industry that uses such programs also doesnt need more performance anyway, isnt that kinda true? 16:15 <+bridge> [ddnet] 16:15 <+bridge> [ddnet] i often think the industry argument isnt really the greatest 16:15 <+bridge> [ddnet] well, now they indeed have 64 bits registers which would not work back then 16:15 <+bridge> [ddnet] just drop x86 and emulate it, and in 3 years bcs hardware gets faster nobody cares xd 16:16 <+bridge> [ddnet] but the decode unit of intel's cpu is a bloat 16:16 <+bridge> [ddnet] i mean i often heard the decoding is expensive 16:16 <+bridge> [ddnet] i mixed myself, 40 years old code still run on today's hardware 16:16 <+bridge> [ddnet] that's a mistake 16:16 <+bridge> [ddnet] and the power consumption doesnt speak for amd or intel 16:16 <+bridge> [ddnet] 16:16 <+bridge> [ddnet] but the effective speed is still good XD 16:16 <+bridge> [ddnet] would a modern cisc just use 32bit word length? 16:16 <+bridge> [ddnet] and not decode? 16:17 <+bridge> [ddnet] power consumption is important for servers 16:17 <+bridge> [ddnet] or 16bit 16:17 <+bridge> [ddnet] and for the planet as well 16:17 <+bridge> [ddnet] 32 is maybe overkill xD 16:17 <+bridge> [ddnet] dont fly or ride a car would be more 16:17 <+bridge> [ddnet] wdym "32bit word length and not decode"? 16:17 <+bridge> [ddnet] and video calls are an easy way to illuminate most of that πŸ˜„ 16:18 <+bridge> [ddnet] i mean x86 without a decoder basiucally 16:18 <+bridge> [ddnet] 16:18 <+bridge> [ddnet] would require fixed since OP code sizes 16:18 <+bridge> [ddnet] you always need a decode stage 16:18 <+bridge> [ddnet] so e.g. 16 bit for the OPcode and i dunno how much more for whatever it needs on informations 16:18 <+bridge> [ddnet] so e.g. 16 bit for the OPcode and i dunno how much more for whatever it needs on information 16:18 <+bridge> [ddnet] risc-v has 4 bytes instructions 16:19 <+bridge> [ddnet] there's a compressed extension with 2 bytes instructions 16:19 <+bridge> [ddnet] you need 3x5 bits for 3 registers opcodes 16:19 <+bridge> [ddnet] mh ok 16:19 <+bridge> [ddnet] plus, you don't use 16 bits opcode, but like 8 is well enough 16:19 <+bridge> [ddnet] for risc? 16:19 <+bridge> [ddnet] but for cisc? 16:20 <+bridge> [ddnet] then you add some bits that depends on each opcodes for variation 16:20 <+bridge> [ddnet] to ease the decode and minimize hardware 16:20 <+bridge> [ddnet] i don't know for cisc 16:20 <+bridge> [ddnet] in all cases, you don't need much opcodes bits 16:20 <+bridge> [ddnet] yeah but we need cisc, bcs i need more performance 16:21 <+bridge> [ddnet] pls study cisc 16:21 <+bridge> [ddnet] the more bits, the more hardware (more muxes in decode stage) 16:21 <+bridge> [ddnet] gib time & money pls 16:21 <+bridge> [ddnet] mostly time πŸ˜„ 16:22 <+bridge> [ddnet] yeah ok, time is an issue accepted 16:22 <+bridge> [ddnet] if you're interested, i can send you my manuscript once i have finished it 16:22 <+bridge> [ddnet] chapter one will discuss global context in semiconductor industry 16:23 <+bridge> [ddnet] power wall, energy wall, dark silicon 16:23 <+bridge> [ddnet] memory problems 16:23 <+bridge> [ddnet] if there are nice graphs i take it 16:23 <+bridge> [ddnet] the picture i sent you are there. I just need to redo them in latex to fit the other figures style 16:25 <+bridge> [ddnet] https://cdn.discordapp.com/attachments/293493549758939136/972866739379523684/Computer_Architecture_Sixth_Edition_A_Quantitative_Approach_by_John_L._Hennessy_David_A._Patterson_z-lib.org.pdf 16:25 <+bridge> [ddnet] enjoy chapter 1 16:25 <+bridge> [ddnet] but there are lots of interesting thing there 16:25 <+bridge> [ddnet] 1300 pages though ^^ 16:34 <+bridge> [ddnet] nice 16:34 <+bridge> [ddnet] even relativly up to date 16:35 <+bridge> [ddnet] will look a bit into it later, but now football xd 16:36 <+bridge> [ddnet] tv or outside πŸ˜„ ? 17:45 <+bridge> [ddnet] outside πŸ˜„ 17:50 <+bridge> [ddnet] xd 17:50 <+bridge> [ddnet] better 18:17 <+bridge> [ddnet] Should I use `LOCK` or `std::mutex` in new code? 18:28 <+bridge> [ddnet] std::mutex xd 18:28 <+bridge> [ddnet] unique_lock 18:28 <+bridge> [ddnet] or lock_guard 18:28 <+bridge> [ddnet] or scoped_lock for the c++17 gangsters 18:31 <+bridge> [ddnet] but scoped_lock apparently allows to use multiple mutex but also no mutex 18:31 <+bridge> [ddnet] that kinda sucks 18:58 <+bridge> [ddnet] only skipped through it, but reads pretty nice, kinda mix of an article and some more specific information πŸ˜„ 18:58 <+bridge> [ddnet] It's sad that `curl_multi_poll` doesn't have a block forever version 19:00 <+bridge> [ddnet] My alternative is to use select with `curl_multi_fdset` but then I have to deal with file descriptors larger than `FD_SETSIZE` myself, or I could use the newer event based API but then I need to wrap `epoll` in system.cpp. Annoying 19:01 <+bridge> [ddnet] Is it maybe time for us to move to libevent to handle this stuff for us? 19:02 <+bridge> [ddnet] wow curls manual really sucks 19:02 <+bridge> [ddnet] it doesnt even mention valid usages for the timeout parameter 19:02 <+bridge> [ddnet] Yeah, pretty annoying, I had to look at the source to make sure 19:15 <+bridge> [ddnet] I guess I could pass `INT_MAX` as timeout 19:46 <+bridge> [ddnet] https://cdn.discordapp.com/attachments/293493549758939136/972917324023742474/RDT_20220508_1945491390148576957613074.jpg 20:12 <+bridge> [ddnet] Why on GER2 my dummy ping 50, and dummy 69? 20:12 <+bridge> [ddnet] Why on GER2 my dummy ping 50, and main 69? 20:14 <+bridge> [ddnet] anyone using dwm there ? 20:19 <+bridge> [ddnet] does anyone know how to build acl? 20:19 <+bridge> [ddnet] does anyone know how to build acl for networking? 20:22 <+bridge> [ddnet] suckless software usually sucks 20:22 <+bridge> [ddnet] software is not that easy and unless u rly want simple stuff its meh 20:22 <+bridge> [ddnet] also if u didnt know, some suckless authors email use a reference to german nazis xd 20:23 <+bridge> [ddnet] > official suckless.org mails originating from a host using a nazi reference (wolfsschanze, hitler’s summer residence) as hostname. 20:25 <+bridge> [ddnet] if u are on x11 i recommend i3-gaps on wayland sway 20:25 <+bridge> [ddnet] mfw getting framed as nazis just because they don't subscribe to the systemd cult 20:26 <+bridge> [ddnet] no, its because using direct references to nazi stuff lol 20:26 <+bridge> [ddnet] https://twitter.com/kuschku/status/1156488420413362177 20:27 <+bridge> [ddnet] hilarious 20:27 <+bridge> [ddnet] konsti would be with them probs 20:27 <+bridge> [ddnet] xddd 20:32 <+bridge> [ddnet] https://news.ycombinator.com/item?id=25753389 20:32 <+bridge> [ddnet] juicy old drama 20:33 <+bridge> [ddnet] wttf is going on 20:33 <+bridge> [ddnet] xd 20:34 <+bridge> [ddnet] nothing i guess 20:34 <+bridge> [ddnet] are u against simple software now bcs some weirdos are for it? 20:34 <+bridge> [ddnet] i dont get it 20:34 <+bridge> [ddnet] no i just pointed it out 20:34 <+bridge> [ddnet] i actually used dwm before i3 20:35 <+bridge> [ddnet] it didnt like it 20:35 <+bridge> [ddnet] i already used i3 20:35 <+bridge> [ddnet] use kwin and ur games just run ℒ️ 20:35 <+bridge> [ddnet] and u dont like i3? 20:35 <+bridge> [ddnet] if u know haskell there is xmoand 20:35 <+bridge> [ddnet] xmonad 20:35 <+bridge> [ddnet] y ik xmonad 20:35 <+bridge> [ddnet] (also buggy af) 20:35 <+bridge> [ddnet] just saiyan 20:35 <+bridge> [ddnet] i3 was cool but i just wanted to test dwm 20:36 <+bridge> [ddnet] dwm is cool and very customizable 20:36 <+bridge> [ddnet] just use terminal lets face it 20:36 <+bridge> [ddnet] x) 20:36 <+bridge> [ddnet] true 20:37 <+bridge> [ddnet] tty only 20:37 <+bridge> [ddnet] why need wm 20:37 <+bridge> [ddnet] text is suckless 20:44 <+bridge> [ddnet] firefox sucks 20:44 <+bridge> [ddnet] use suckless browser 20:44 <+bridge> [ddnet] linux is bloated 20:44 <+bridge> [ddnet] remove drivers 20:44 <+bridge> [ddnet] tru 20:44 <+bridge> [ddnet] package manager is bloat too 20:44 <+bridge> [ddnet] host your own server 20:44 <+bridge> [ddnet] dont use google its bloated af 20:44 <+bridge> [ddnet] own kernel 20:44 <+bridge> [ddnet] use library 20:44 <+bridge> [ddnet] the http protocol is bloated 20:44 <+bridge> [ddnet] dont use the web 20:44 <+bridge> [ddnet] public libraries 20:44 <+bridge> [ddnet] tcp only 20:45 <+bridge> [ddnet] udp 20:45 <+bridge> [ddnet] no tcp 20:45 <+bridge> [ddnet] reliability is bloat 20:45 <+bridge> [ddnet] i miss the old times 20:45 <+bridge> [ddnet] windows xp starts in less than 1 second on my vm xd 20:45 <+bridge> [ddnet] the good old days 20:46 <+bridge> [ddnet] πŸ˜„ 20:46 <+bridge> [ddnet] @Not Keks in the old times u didnt have a ssd 20:46 <+bridge> [ddnet] so its fake 20:46 <+bridge> [ddnet] yes but software wasnt bloated xd 20:46 <+bridge> [ddnet] xd 20:47 <+bridge> [ddnet] it was 20:47 <+bridge> [ddnet] do u remember the browsers 20:47 <+bridge> [ddnet] witht 5 layers of extensions 20:47 <+bridge> [ddnet] on the search bar 20:47 <+bridge> [ddnet] i miss firefox 3.6 20:47 <+bridge> [ddnet] that was stable for like 5years no real updates 20:47 <+bridge> [ddnet] now its firefox 101 20:47 <+bridge> [ddnet] and it sucks xd 20:47 <+bridge> [ddnet] why it sucks 20:47 <+bridge> [ddnet] xd 20:48 <+bridge> [ddnet] the only bad thing is webgl perf 20:48 <+bridge> [ddnet] what does not suck on modern browsers 20:48 <+bridge> [ddnet] ok i actually like css 20:48 <+bridge> [ddnet] i need this for our client 20:48 <+bridge> [ddnet] modern css is way better 20:48 <+bridge> [ddnet] than old css 20:48 <+bridge> [ddnet] old css was spamming tables 20:48 <+bridge> [ddnet] yeah 20:48 <+bridge> [ddnet] to get stuff to work 20:48 <+bridge> [ddnet] xd 20:48 <+bridge> [ddnet] flexbox is godsent 20:49 <+bridge> [ddnet] email templates still use tables btw 20:49 <+bridge> [ddnet] its a nightmare 20:49 <+bridge> [ddnet] well i want to be honest with u 20:49 <+bridge> [ddnet] firefox for devs, best browser(bcs the tools and the f12 menu and stuff) 20:49 <+bridge> [ddnet] but firefox is so much behind webkit 20:49 <+bridge> [ddnet] webgl performance, stuff like backdrop filter still not default 20:50 <+bridge> [ddnet] no scrollbar customizations 20:50 <+bridge> [ddnet] svg is buggy 20:50 <+bridge> [ddnet] xd 20:50 <+bridge> [ddnet] sadge 20:50 <+bridge> [ddnet] imagine not being backed by a multi billion dollar company 20:50 <+bridge> [ddnet] and being not for profit 20:50 <+bridge> [ddnet] web browser development is insanely costly 20:50 <+bridge> [ddnet] yeah but its still kinda mozillas fault 20:50 <+bridge> [ddnet] way more than kernel development 20:50 <+bridge> [ddnet] before chrome was there all used firefox 20:50 <+bridge> [ddnet] its insane 20:51 <+bridge> [ddnet] but they slept and slept 20:51 <+bridge> [ddnet] and chrome modernized the web 20:51 <+bridge> [ddnet] or google 20:51 <+bridge> [ddnet] and then firefox just ran behind 20:51 <+bridge> [ddnet] ye 20:51 <+bridge> [ddnet] netscape best 20:52 <+bridge> [ddnet] i mostly use firefox on mobile tho 20:52 <+bridge> [ddnet] bcs adblock xD 20:52 <+bridge> [ddnet] and videos in background 20:53 <+bridge> [ddnet] oh ye 20:53 <+bridge> [ddnet] i have ublock on firefox android 20:53 <+bridge> [ddnet] i use fenec tho 20:53 <+bridge> [ddnet] bcs its in fdroid 20:53 <+bridge> [ddnet] web dev is rly mobile focused because adblocks there are even rarer 20:53 <+bridge> [ddnet] yes trui 20:54 <+bridge> [ddnet] even on desktop 20:54 <+bridge> [ddnet] i think only 30% ppl use adblock 20:54 <+bridge> [ddnet] ppl are dumb af 20:54 <+bridge> [ddnet] well its probs more ignorance 20:54 <+bridge> [ddnet] if they knew they could hide them they probs would do it 20:54 <+bridge> [ddnet] tbh i am suprised so many ppl arent annoyed enough to activly search for using adblocks 20:55 <+bridge> [ddnet] i mean when u buy a samsung TV nowadays u get fucking ads on them 20:55 <+bridge> [ddnet] like wtf 20:55 <+bridge> [ddnet] xd 20:56 <+bridge> [ddnet] i use a dns based adblock inside our router nowadays 20:56 <+bridge> [ddnet] bcs its too annoying to deal with all platforms individually 20:57 <+bridge> [ddnet] I use bspwm now, it's pretty good 21:00 <+bridge> [ddnet] i have to switch to https://github.com/leftwm/leftwm 21:00 <+bridge> [ddnet] cuz its made in rust 21:00 <+bridge> [ddnet] 21:01 <+bridge> [ddnet] > LeftWM was built from the very beginning to support multiple screens and ultrawide monitors. The default keybindings support ultrawide monitors and multiple screens. 21:01 <+bridge> [ddnet] nice feature 21:01 <+bridge> [ddnet] https://raw.githubusercontent.com/leftwm/leftwm/main/screenshots/4.jpg 21:01 <+bridge> [ddnet] lol 21:01 <+bridge> [ddnet] nobos 21:02 <+bridge> [ddnet] where is glassy blur 21:02 <+bridge> [ddnet] backdrop-filter: blur(3px) 21:02 <+bridge> [ddnet] use that css xd 21:02 <+bridge> [ddnet] but not on firefox xddd 21:03 <+bridge> [ddnet] i want this for ddnet too 21:04 <+bridge> [ddnet] when do we drop all other than vk and gl3.3 xd 21:13 <+bridge> [ddnet] looks sheesh 21:14 <+bridge> [ddnet] nobo 21:14 <+bridge> [ddnet] i have 30$ font 21:14 <+bridge> [ddnet] k ? 21:14 <+bridge> [ddnet] road to 1k $ font 21:15 <+bridge> [ddnet] + where waifu wallpaper 21:16 <+bridge> [ddnet] its the leftwm ryo posted xd 21:17 <+bridge> [ddnet] :( 21:17 <+bridge> [ddnet] Sad 21:17 <+bridge> [ddnet] Where are emojis on status bar? 23:19 <+bridge> [ddnet] hi can anyone guide me where to start if i want to generate ddnet maps with python/scripting? 23:25 <+bridge> [ddnet] eh 23:25 <+bridge> [ddnet] if ur new to programming its gonna be a ride 23:25 <+bridge> [ddnet] but for starters you need to use patigas library 23:25 <+bridge> [ddnet] twmap 23:27 <+bridge> [ddnet] that's why I started using xmonad, it's written and configures in haskell πŸ˜„ 23:27 <+bridge> [ddnet] :greenthing: 23:28 <+bridge> [ddnet] I think someone had some maze examples, would be nice if they could share them, probably easiest to start from a working generator 23:28 <+bridge> [ddnet] in rust there is a popular config format that blends really well with rust types 23:28 <+bridge> [ddnet] its this one 23:28 <+bridge> [ddnet] https://github.com/ron-rs/ron 23:28 <+bridge> [ddnet] Rust object notation 23:29 <+bridge> [ddnet] > RON is a simple readable data serialization format that looks similar to Rust syntax. It's designed to support all of Serde's data model, so structs, enums, tuples, arrays, generic maps, and primitive values. 23:29 <+bridge> [ddnet] https://cdn.discordapp.com/attachments/293493549758939136/972973517643579502/unknown.png 23:29 <+bridge> [ddnet] xD 23:31 <+bridge> [ddnet] using dwm 23:39 <+bridge> [ddnet] gg 23:39 <+bridge> [ddnet] anyone has installed spotify without fking snap ? 23:41 <+bridge> [ddnet] this `https://aur.archlinux.org/spotify.git` is so unstable