00:15 <+bridge> [ddnet] @louis is 1024x1024 00:15 <+bridge> [ddnet] @louis its 1024x1024 00:18 <+bridge> [ddnet] @Cøke then they downloaded wrong 03:30 <+bridge> [ddnet] `Failed to fetch https://packagecloud.io/github/git-lfs/ubuntu/dists/focal/InRelease 402 Payment Required [IP: 54.193.34.251 443]` 03:30 <+bridge> [ddnet] give them the payment 05:52 <+bridge> [ddnet] love ya 12:20 <+bridge> [ddnet] @deen so it fixes the different sizes? 12:20 <+bridge> [ddnet] https://cdn.discordapp.com/attachments/293493549758939136/773869354688708618/Screen_Shot_2020-11-05_at_5.19.59_AM.png 12:21 <+bridge> [ddnet] if u still need to 12:21 <+bridge> [ddnet] > - [ ] Tested the change ingame 12:21 <+bridge> [ddnet] i can do the tee command 12:21 <+bridge> [ddnet] Yes, couldn't test it 12:21 <+bridge> [ddnet] ah ok 12:23 <+bridge> [ddnet] worked? 12:23 <+bridge> [ddnet] yes, added screenshot 12:23 <+bridge> [ddnet] :D 12:23 <+bridge> [ddnet] cool 12:23 <+bridge> [ddnet] its good change 12:23 <+bridge> [ddnet] it's originally by Banana090, I just fixed it up 12:24 <+bridge> [ddnet] 👍 12:52 <+bridge> [ddnet] 👍 13:08 <+bridge> [ddnet] 👍 15:20 <+bridge> [ddnet] is 15.1.3 the latest version? 15:34 <+bridge> [ddnet] <Дядя Женя> there is ddnet-master version which is newer, and there is my forked repo which is way ahead ddnet official client xD 16:32 <+bridge> [ddnet] Are there really codebases which mix tabs and spaces in this way? 16:32 <+bridge> [ddnet] ``` 16:32 <+bridge> [ddnet] def function(x, y): 16:32 <+bridge> [ddnet] __if x: 16:32 <+bridge> [ddnet] >>>>for z in range(y): 16:32 <+bridge> [ddnet] >>>>__wobblify(z) 16:32 <+bridge> [ddnet] ``` 16:32 <+bridge> [ddnet] Has anyone here encountered this indentation style? 16:34 <+bridge> [ddnet] @timakro sometimes i see spaces to fit stuff for example in long function signatures 16:34 <+bridge> [ddnet] but not for that 16:35 <+bridge> [ddnet] Are there really codebases which mix tabs and spaces in this way? 16:35 <+bridge> [ddnet] ``` 16:35 <+bridge> [ddnet] function crazy() { 16:35 <+bridge> [ddnet] __if (true) { 16:35 <+bridge> [ddnet] >>>>if (true) { 16:35 <+bridge> [ddnet] >>>>__if (true) { 16:35 <+bridge> [ddnet] >>>>>>>>if (true) { 16:35 <+bridge> [ddnet] >>>>>>>>__if (true) { 16:35 <+bridge> [ddnet] >>>>>>>>>>>>console.log('indent'); 16:35 <+bridge> [ddnet] >>>>>>>>__} 16:35 <+bridge> [ddnet] >>>>>>>>} 16:35 <+bridge> [ddnet] >>>>__} 16:36 <+bridge> [ddnet] >>>>} 16:36 <+bridge> [ddnet] __} 16:36 <+bridge> [ddnet] } 16:36 <+bridge> [ddnet] ``` 16:36 <+bridge> [ddnet] Has anyone here encountered this indentation style? 16:36 <+bridge> [ddnet] I changed the example to make it more clear ^^ 16:37 <+bridge> [ddnet] Because I wanna write another vim plugin to detect indentation style and configure the tab key to produce the same indent. There are only 5 of those plugins currently but they are all doing it wrong :) 16:38 <+bridge> [ddnet] And vim supports this crazy indent style and some of those detect indent plugins do too. But I'm saying there is no code like that ^^ 16:58 <+bridge> [ddnet] Why would you need to do that tho 16:58 <+bridge> [ddnet] I mean if the tab size is configured differently it would probably looks horrible 17:33 <+bridge> [ddnet] @louis hey once again, can you quickly lookup the broken driver version you are using rn? its in the device manager, when u select settings of the intel GPU probably 17:37 <+bridge> [ddnet] @timakro it's possible to configure vim to do that at least. to indent by 2 spaces and replace every 4 spaces indentation with a tab. But that always seemed like the worst of both worlds to me 17:40 <+bridge> [ddnet] https://cdn.discordapp.com/attachments/293493549758939136/773949815935074365/unknown.png 17:40 <+bridge> [ddnet] oh such a new driver 17:40 <+bridge> [ddnet] rip 17:40 <+bridge> [ddnet] then we have to block like a whole year of intel driver 17:41 <+bridge> [ddnet] or is it 09.03.2020 17:41 <+bridge> [ddnet] i am always confused by english dates 17:41 <+bridge> [ddnet] I think americans do mdy 17:42 <+bridge> [ddnet] so this is 9th of march? 17:42 <+bridge> [ddnet] If it’s an american date, yes 17:42 <+bridge> [ddnet] Why is the US so hard headed on this stuff? 17:42 <+bridge> [ddnet] xD 17:43 <+bridge> [ddnet] Would it really kill them to use metric and write dates the reasonable way? 17:46 <+bridge> [ddnet] @Ravie if u online, can you quickly loopup your driver version aswell? 17:46 <+bridge> [ddnet] > Why is the US so hard headed on this stuff? 17:46 <+bridge> [ddnet] @Learath2 17:46 <+bridge> [ddnet] @Ravie if u online, can you quickly lookup your driver version aswell? 17:46 <+bridge> [ddnet] 2 feet tall 17:46 <+bridge> [ddnet] 🐾 17:46 <+bridge> [ddnet] does intel not know about year month date format :justatest: 17:47 <+bridge> [ddnet] i love how the status in the screen says "This device is working properly" 17:47 <+bridge> [ddnet] while it isnt xd 17:51 <+bridge> [ddnet] @Jupstar ✪ 17:51 <+bridge> [ddnet] https://cdn.discordapp.com/attachments/293493549758939136/773952559592767488/unknown.png 17:51 <+bridge> [ddnet] thanks 18:01 <+bridge> [ddnet] I considered for a second to get a notebook with i3 8130u to see if I could reproduce it but they are all around 500€ 18:01 <+bridge> [ddnet] the funny thing is 18:01 <+bridge> [ddnet] deen has the exact needed intel 18:02 <+bridge> [ddnet] he could just debug xD 18:02 <+bridge> [ddnet] Doesn’t happen for him though I thought 18:02 <+bridge> [ddnet] he even has windows on it xD 18:02 <+bridge> [ddnet] well he could just use louis driver version 18:02 <+bridge> [ddnet] it's my work laptop, so can't help much with that 18:03 <+bridge> [ddnet] No non-work on work laptop? 18:03 <+bridge> [ddnet] kind of 18:04 <+bridge> [ddnet] I’ll check a couple friends laptops maybe one of them has an intel with 620uhd 18:05 <+bridge> [ddnet] Also, I thought there is no good way to reproduce 18:05 <+bridge> [ddnet] ravie found one 18:05 <+bridge> [ddnet] multimap easy non halloween, enable entities jump into the freeze at one spot 18:06 <+bridge> [ddnet] with gl 3.0 18:06 <+bridge> [ddnet] preinit vbo on? 18:06 <+bridge> [ddnet] without gl 3.3 18:07 <+bridge> [ddnet] when u spawn its the second freeze from right 18:07 <+bridge> [ddnet] https://cdn.discordapp.com/attachments/293493549758939136/773956791910006794/screenshot_2020-11-05_18-07-53.png 18:08 <+bridge> [ddnet] huh, reproduced 18:08 <+bridge> [ddnet] that was easy 18:08 <+bridge> [ddnet] worked on halloween version too 18:08 <+bridge> [ddnet] oh ok 18:08 <+bridge> [ddnet] How did we make a crash intel tile even? ;D 18:09 <+bridge> [ddnet] well i wonder too what is actually crashing inside the driver 18:09 <+bridge> [ddnet] bcs while the map is loaded nothing is reloaded afterwards 18:09 <+bridge> [ddnet] CPU usage is at 0% now, so no infinite loop? I still hear sounds. When I alt-tab back into it it just flickers 18:09 <+bridge> [ddnet] It happening on a tile is what’s very novel to me. These sorts of driver bugs should be random 18:09 <+bridge> [ddnet] Something about the render thread seems to be broken somehow 18:10 <+bridge> [ddnet] tho asan didnt show anything 18:10 <+bridge> [ddnet] We send a command to the thread, it does something, which breaks the screen, after which the thread obv can’t recover because it doesn’t even know sth went wrong 18:10 <+bridge> [ddnet] the best would be some bug with the map border 18:10 <+bridge> [ddnet] that would be easiest to fix, but i could only imagine a overflow there 18:11 <+bridge> [ddnet] bcs the renderloop does continue 18:11 <+bridge> [ddnet] maybe its just a "0" instead of a value 18:11 <+bridge> [ddnet] thats causing it 18:11 <+bridge> [ddnet] any hints for how to debug it? 18:12 <+bridge> [ddnet] well only things i could imagine are invalid rendercounts 18:12 <+bridge> [ddnet] so <= 0 and >= 1000 18:12 <+bridge> [ddnet] in CCommandProcessorFragment_OpenGL::Cmd_Render 18:12 <+bridge> [ddnet] the m_PrimCount should be very high 18:13 <+bridge> [ddnet] Since it doesn’t crash and our code just keeps running your only bet would be logging the stuff we do for sanity checking or a gpu debugging tool that could report insane values to us 18:13 <+bridge> [ddnet] same for 18:13 <+bridge> [ddnet] CCommandProcessorFragment_OpenGL2::Cmd_RenderTex3D 18:13 <+bridge> [ddnet] Whether such a tool exists idk. Nvidia had sth iirc maybe intel does too 18:14 <+bridge> [ddnet] Can we do something with Windows dump files? 18:14 <+bridge> [ddnet] Doesn’t windows only dump on crash? 18:15 <+bridge> [ddnet] In task manager I can create a dump 18:15 <+bridge> [ddnet] Hm, cdb can open dumps 18:16 <+bridge> [ddnet] @Jupstar ✪ do you want to add some logging? I don't really know where to start. Then I can build and run it 18:17 <+bridge> [ddnet] Yeah cdb can open these dumps with the executable and the symbol file 18:18 <+bridge> [ddnet] Though I don’t think it will be of much help, since it seems transient 18:18 <+bridge> [ddnet] A value/command/call we send/make just breaks the driver 18:19 <+bridge> [ddnet] i can try, but honestly, if the logging doesnt trigger for me, why should it for u xD 18:19 <+bridge> [ddnet] outside the driver we use the same code 18:21 <+bridge> [ddnet] Oh was there anything in the new opengl debugging stuff we added? 18:22 <+bridge> [ddnet] if that thing would work it should show errors in the stdout 18:22 <+bridge> [ddnet] https://cdn.discordapp.com/attachments/293493549758939136/773960475695906857/backend_sdl.cpp 18:22 <+bridge> [ddnet] here some logging for high values 18:22 <+bridge> [ddnet] There also seems to be a project called glintercept, if I could reproduce it I would also try that 18:22 <+bridge> [ddnet] if you dont zoom out or smth they shouldnt trigger 18:23 <+bridge> [ddnet] if you prefer github i'll upload it there @deen 18:23 <+bridge> [ddnet] @Jupstar ✪ so the debugging doesn’t work or it doesn’t show anything? 18:23 <+bridge> [ddnet] @Learath2 not for me at least 18:23 <+bridge> [ddnet] tried with zink, mesa software and amdgpu 18:24 <+bridge> [ddnet] dbg_gfx 1 shows nothing 18:24 <+bridge> [ddnet] Again, which one? Does it not show any errors or are we failing to enable debugging? 18:24 <+bridge> [ddnet] @Learath2 i quickly did an obvious error and that worked 18:24 <+bridge> [ddnet] @Jupstar ✪ yeah, if you could make a branch that'd be great 18:24 <+bridge> [ddnet] amdgpu even shows performance warnings xD 18:25 <+bridge> [ddnet] @deen pr_TESTING 18:28 <+bridge> [ddnet] sad that zink is really ready yet :D 18:28 <+bridge> [ddnet] i already get over 1k FPS with it with gl 3.3 18:28 <+bridge> [ddnet] but mipmaps are really broken 18:29 <+bridge> [ddnet] cool 18:30 <+bridge> [ddnet] Ah, you know maybe a vogl trace could let us try to debug this outside 18:30 <+bridge> [ddnet] If it gets opengl itself into a state that’s broken, a vogl trace should reproduce it on our computers too. If not we can still inspect the calls, see if anything odd happens 18:31 <+bridge> [ddnet] but that doesnt explain why my 10 years old intel works fine xd 18:32 <+bridge> [ddnet] It’s obviously an odd bug. Just rolling over and saying fuck you people who use this specific generation of cpus is hardly a solution 18:33 <+bridge> [ddnet] I’m just trying to find a way to debug it, if we figure out which call is going wrong, maybe we can work around it 18:33 <+bridge> [ddnet] @Learath2 tbh nobody cares alot about GL anymore, renderdoc is the best but it only supports GL 3.3 18:33 <+bridge> [ddnet] also a software renderer should easily find the error 18:33 <+bridge> [ddnet] @Learath2 tbh nobody cares alot about GL anymore, renderdoc is the best but it only supports GL 3.2+ 18:33 <+bridge> [ddnet] also a software renderer should easily find the error 18:33 <+bridge> [ddnet] Not if the software renderer doesn’t have the same bug as the intel driver 18:33 <+bridge> [ddnet] xD 18:34 <+bridge> [ddnet] I opened a bug at Intel inside of the driver tool btw 18:34 <+bridge> [ddnet] ok 18:34 <+bridge> [ddnet] wdym nobody cares about gl? 18:34 <+bridge> [ddnet] Vulkan made people jump ship 18:34 <+bridge> [ddnet] well its natural but there should be still plenty using opengl 18:34 <+bridge> [ddnet] @Ryozuki as soon as zink is good enough it will probably adapt to windows too 18:34 <+bridge> [ddnet] Valve had a lot of opengl tools planned they just dropped because vulkan makes it easier to port games 18:35 <+bridge> [ddnet] Especially with dxvk around 18:35 <+bridge> [ddnet] bcs opengl support sucks xd 18:35 <+bridge> [ddnet] > Zink is an OpenGL implementation on top of Vulkan. Or to be a bit more specific, Zink is a Mesa Gallium driver that leverages the existing OpenGL implementation in Mesa to provide hardware accelerated OpenGL when only a Vulkan driver is available. 18:35 <+bridge> [ddnet] is it rly a real world case that there is a vulkan driver and not a opengl one? 18:35 <+bridge> [ddnet] xd 18:35 <+bridge> [ddnet] thats not the point 18:36 <+bridge> [ddnet] there is always a vulkan driver 18:36 <+bridge> [ddnet] so opengl isnt required anymore 18:36 <+bridge> [ddnet] even on old gpus? 18:36 <+bridge> [ddnet] well nobody supports old GPUs xD 18:36 <+bridge> [ddnet] xd 18:36 <+bridge> [ddnet] what about android 18:36 <+bridge> [ddnet] i obviously mean that zink will be replaced inside the driver 18:36 <+bridge> [ddnet] Ofc not on older gpus, jupstar loves to pretend they don’t exist. Like all other gfx people I know 18:36 <+bridge> [ddnet] do mobile gpus support vulkan 18:36 <+bridge> [ddnet] @Learath2 nah, but they will work just as now 18:37 <+bridge> [ddnet] and everything that has vulkan will be updated some day 18:37 <+bridge> [ddnet] well old gpus could support vulkan with a driver right? there shouldnt be a hardware limitation 18:37 <+bridge> [ddnet] or is there 18:37 <+bridge> [ddnet] the average lifetime support for GPUs that isnt NVidia is like 2-3 years 18:37 <+bridge> [ddnet] after that time only some rare patches come 18:37 <+bridge> [ddnet] i have my r9 390 since 2016 i think 18:37 <+bridge> [ddnet] or the architecture didnt change 18:38 <+bridge> [ddnet] its true its getting high temps, i want to update it 18:38 <+bridge> [ddnet] It’s not financially justified to bring vulkan to older gpus 18:38 <+bridge> [ddnet] @Ryozuki there probs are hardware limitations, but not in the past 10 years 18:39 <+bridge> [ddnet] well as long as "no gpus" such as intel gpu thingies support vulkan then its fine i guess 18:39 <+bridge> [ddnet] integrated gpus* 18:39 <+bridge> [ddnet] It’s much more profitable to move the industry to vulkan so people have to buy new gpus so the ceos can eat at nice fancy steakhouses 18:39 <+bridge> [ddnet] is it rly common for ppl to buy gpus? 18:39 <+bridge> [ddnet] i dont know 18:39 <+bridge> [ddnet] ofc 18:39 <+bridge> [ddnet] every 4 years the kids have high end shit xd 18:39 <+bridge> [ddnet] well i guess ppl that play yes 18:39 <+bridge> [ddnet] Sure, there are even fanboys lining up for nvidia ones, like the apple ppl 18:39 <+bridge> [ddnet] including me xd 18:40 <+bridge> [ddnet] the marketshare of the 1000 series from NVIDIA is insane 18:40 <+bridge> [ddnet] i remember me playing on windows xp some weeb game slow af 18:40 <+bridge> [ddnet] sad times 18:40 <+bridge> [ddnet] @Learath2 have u played this legend 18:40 <+bridge> [ddnet] :BASED: 18:40 <+bridge> [ddnet] I might move to an rx6000 in a year to move to amdgpu 18:41 <+bridge> [ddnet] Tho somehow nvidias proprietary driver still works the best, even though linus said fuck nvidia 😛 18:41 <+bridge> [ddnet] definitly stay in amd if ur not hyped about this RTX marketing stuff 18:41 <+bridge> [ddnet] the performance in linux is really good with AMD now 18:41 <+bridge> [ddnet] @Learath2 buy a ryzen 5000 and a new gpu 18:41 <+bridge> [ddnet] and u have smart memory access 18:41 <+bridge> [ddnet] easily beats most windows stuff with DXVK 😄 18:41 <+bridge> [ddnet] its dope 18:41 <+bridge> [ddnet] I’m on a 1080 and honestly it works great everywhere 18:41 <+bridge> [ddnet] I'm curious how many fps you'll get with ryzen 5000 and one of the new gpus 18:41 <+bridge> [ddnet] https://www.amd.com/es/technologies/smart-access-memory 18:41 <+bridge> [ddnet] 12-13k 18:42 <+bridge> [ddnet] without the HUD and entities i get max FPS of around 12k 18:42 <+bridge> [ddnet] :poggers: 18:42 <+bridge> [ddnet] https://cdn.discordapp.com/attachments/293493549758939136/773965374685315082/unknown.png 18:42 <+bridge> [ddnet] I will get a ryzen 5000 in a month or so. My 4770k will be promoted to a NAS 18:42 <+bridge> [ddnet] @Jupstar ✪ do i need to enable something or is it logged automatically? 18:43 <+bridge> [ddnet] @Ryozuki I’m not sure if I’m onboard with this feature yet. I don’t really like proprietary stuff 18:43 <+bridge> [ddnet] > every 4 years the kids have high end shit xd 18:43 <+bridge> [ddnet] @Jupstar ✪ gotta play fortnite 4k 18:43 <+bridge> [ddnet] no output :/ 18:43 <+bridge> [ddnet] > @Ryozuki I’m not sure if I’m onboard with this feature yet. I don’t really like proprietary stuff 18:43 <+bridge> [ddnet] @Learath2 im onboard this stuff cuz its good competition against propietary rtx nvidia stuff 18:43 <+bridge> [ddnet] They are kinda locking you into the amd ecosystem if you want full access to the memory 18:44 <+bridge> [ddnet] Amd has always been very open about the tech. Hope they let linux have this benefit aswell without binary blobs 18:44 <+bridge> [ddnet] i hope yea 18:44 <+bridge> [ddnet] i use amd cuz it has good linux support 18:44 <+bridge> [ddnet] Oh and I did play grand chase 18:44 <+bridge> [ddnet] :BASED: 18:44 <+bridge> [ddnet] a man of culture 18:44 <+bridge> [ddnet] @Learath2 @Jupstar ✪ further ideas for debugging? 18:45 <+bridge> [ddnet] Nvidia also has decent linux support now, I didn’t tinker with anything in ages 18:45 <+bridge> [ddnet] https://cdn.discordapp.com/attachments/293493549758939136/773966184248639569/IMG_20201105_184503.jpg 18:45 <+bridge> [ddnet] @deen idk, I would look into glintercept and vogl as I said 18:45 <+bridge> [ddnet] thzats my fps graph 18:45 <+bridge> [ddnet] sadly cant take screenshot 18:45 <+bridge> [ddnet] seems to be some overlay 😄 18:46 <+bridge> [ddnet] @deen i honestly don't know, i just saw the one weird value on louis screen where it rendered 10k triagnles 18:47 <+bridge> [ddnet] @Jupstar ✪ do u have a new ryzen? 18:47 <+bridge> [ddnet] 3800x 18:47 <+bridge> [ddnet] so pretty new yes 18:47 <+bridge> [ddnet] here i am with 2600 18:47 <+bridge> [ddnet] 18:47 <+bridge> [ddnet] did u know ryzen 5 3600 is one of the most sold cpus 18:47 <+bridge> [ddnet] its insane 18:47 <+bridge> [ddnet] didnt know 😄 18:48 <+bridge> [ddnet] https://cdn.discordapp.com/attachments/293493549758939136/773966936727355443/unknown.png 18:48 <+bridge> [ddnet] 13k reviews 18:48 <+bridge> [ddnet] on the spanish one 18:48 <+bridge> [ddnet] its in the sweet spot 18:48 <+bridge> [ddnet] not to pricy and rly good 18:48 <+bridge> [ddnet] 101410 wow 18:48 <+bridge> [ddnet] https://cdn.discordapp.com/attachments/293493549758939136/773967081620111380/unknown.png 18:49 <+bridge> [ddnet] over 100k sold 18:49 <+bridge> [ddnet] yep 18:49 <+bridge> [ddnet] i mean 179€ for 12 threads 18:49 <+bridge> [ddnet] 6 years ago i'd have need to sell my liver for that 18:50 <+bridge> [ddnet] Your liver would buy you many of those 😛 18:50 <+bridge> [ddnet] xd 18:53 <+bridge> [ddnet] well i guess for the intel thing we can agree, its either the hardest to reproduce thing, or its an intel bug, that affects around 3-4 month of driver versions 18:53 <+bridge> [ddnet] @Jupstar ✪ ryzen can put so many threads thanks to their awesome architecture 18:53 <+bridge> [ddnet] or so i read 18:53 <+bridge> [ddnet] thats why they can also fit 64 in threadripper 18:53 <+bridge> [ddnet] @Ryozuki yeah the infinite thing 18:53 <+bridge> [ddnet] awesome stuff 18:53 <+bridge> [ddnet] Oh it definitely is an intel bug. I wonder what it is so we can work around it 18:54 <+bridge> [ddnet] i get around 300fps with software renderer on 2k screen, thats how good the CPUs are xD 18:54 <+bridge> [ddnet] well cpus are better at other stuff 18:54 <+bridge> [ddnet] Updating the driver fixing it confirms that it’s a bug in their code 18:54 <+bridge> [ddnet] > i get around 300fps with software renderer on 2k screen, thats how good the CPUs are xD 18:54 <+bridge> [ddnet] @Jupstar ✪ can u compile a program with ur gpu 18:54 <+bridge> [ddnet] xd 18:54 <+bridge> [ddnet] i remember when that nvidia guy said, that GPUs will replace CPUs xDDD 18:55 <+bridge> [ddnet] Or like 0.0001% chance that it’s undefined behaviour in our code behaving sane on one version 18:55 <+bridge> [ddnet] 18:55 <+bridge> [ddnet] https://ddnet.tw/gliInterceptLog.txt.xz 18:56 <+bridge> [ddnet] how do i open it? xD 18:56 <+bridge> [ddnet] xz -d 18:56 <+bridge> [ddnet] it only shows random stuff 18:56 <+bridge> [ddnet] oh ok 18:56 <+bridge> [ddnet] seems like the KDE program cant handle it 18:57 <+bridge> [ddnet] so these are invalid calls? 18:57 <+bridge> [ddnet] all calls 18:58 <+bridge> [ddnet] until the hang? 19:00 <+bridge> [ddnet] i don't know where the hang is, but it should be in there 19:00 <+bridge> [ddnet] it also created an Images and Shader dir, I guess we don't need that 19:00 <+bridge> [ddnet] Anything interesting in images? 19:01 <+bridge> [ddnet] @deen ah could you try gl major 2, and gl minor 0 19:01 <+bridge> [ddnet] so gl 2.0 19:01 <+bridge> [ddnet] that uses 3d textures instead of 2d array textures 19:02 <+bridge> [ddnet] i wonder if its just the 2d array textures that break 😄 19:02 <+bridge> [ddnet] (which i still won't fix), but then we'd have a clue atleast 19:02 <+bridge> [ddnet] i need to rerun it 19:03 <+bridge> [ddnet] Ah Intel GPA seems to also do tracing so I would probably try that too. This is all just what I would try not sure if they would give any result 19:04 <+bridge> [ddnet] yeah, i'd assume that the debug output form opengl is most useful 19:04 <+bridge> [ddnet] atleast i'd assume a driver that has alot of error checking at specific spots 19:06 <+bridge> [ddnet] huh. with glintercept fulldebug on it doesn't crash anymore... 19:06 <+bridge> [ddnet] is it a software renderer? 19:06 <+bridge> [ddnet] no, a dll that intercepts all opengl calls and prints them out 19:06 <+bridge> [ddnet] before passing them on 19:06 <+bridge> [ddnet] ok 19:06 <+bridge> [ddnet] that's strange 19:07 <+bridge> [ddnet] might be timing related and it adds just enough time... 19:07 <+bridge> [ddnet] true 19:07 <+bridge> [ddnet] because fulldebug calls a filesystem flush after each logging 19:14 <+bridge> [ddnet] uhm, is it normal that glintercept logs have ??? as arguments 19:15 <+bridge> [ddnet] oh @deen u could also try gfx_finish, which always waits before swapping buffers 19:15 <+bridge> [ddnet] random thoughts but who knows 19:15 <+bridge> [ddnet] if its really just timing 😄 19:16 <+bridge> [ddnet] https://ddnet.tw/glintercept_fulldebug_nohang.tar.gz 19:16 <+bridge> [ddnet] https://ddnet.tw/glintercept_fulldebug_nohang.tar.xz 19:16 <+bridge> [ddnet] @Learath2 no, that was because I was missing headers 19:16 <+bridge> [ddnet] i fixed that and reran it, but it doesn't crash with full debug tracing.. 19:16 <+bridge> [ddnet] doesn't hang* 19:17 <+bridge> [ddnet] @deen btw anything in gliLog.txt? 19:17 <+bridge> [ddnet] ok, it's the flushing that prevents hanging 19:17 <+bridge> [ddnet] Now the question is howto force it to flush in the end 19:18 <+bridge> [ddnet] doesn't register an exit handler.. 19:20 <+bridge> [ddnet] hm, it keeps logging opengl calls even after the hang 19:22 <+bridge> [ddnet] Yeah, we already knew the graphics thread kept going as if nothing happened 19:29 <+bridge> [ddnet] https://ddnet.tw/glintercept_hang.tar.xz 19:35 <+bridge> [ddnet] gfx_finish 1 didn't help 19:37 <+bridge> [ddnet] OpenGL 2.0 works @Jupstar ✪ 19:37 <+bridge> [ddnet] ok 19:37 <+bridge> [ddnet] then its either shader or 2d array textures 19:41 <+bridge> [ddnet] does ddnet have rest api for db? 19:41 <+bridge> [ddnet] no 19:41 <+bridge> [ddnet] bad 19:41 <+bridge> [ddnet] but ok 19:47 <+bridge> [ddnet] all we have is https://info2.ddnet.tw/info?name=deen and you can get the full database: https://ddnet.tw/stats/ 19:48 <+bridge> [ddnet] @Jupstar ✪ after using timeout protection it doesn't hang. without timeout protection it hangs 19:49 <+bridge> [ddnet] when u disable it, it doesnt hang? 19:49 <+bridge> [ddnet] I mean after using timeout protection it doesn't hang 19:49 <+bridge> [ddnet] ok 19:49 <+bridge> [ddnet] strange randomness 19:49 <+bridge> [ddnet] I'm trying Intel GPA now, but can't get it to work yet 19:59 <+bridge> [ddnet] Intel GPA doesn't seem to support OpenGL on Windows 19:59 <+bridge> [ddnet] https://software.intel.com/content/www/us/en/develop/documentation/gpa-user-guide-for-windows-host/top/supported-platforms-and-applications.html 20:00 <+bridge> [ddnet] and the performance recording doesn't look so useful, didn't see any warning logging 20:03 <+bridge> [ddnet] sadly once they are at gl 1.x they probably stay at that version 20:03 <+bridge> [ddnet] who? 20:04 <+bridge> [ddnet] the blacklisted ppl 20:04 <+bridge> [ddnet] no, just make it the default for them, but don't persist it 20:04 <+bridge> [ddnet] yeah, but like they know how to change the config 20:10 <+bridge> [ddnet] if the configuration struct would support changing the default values on runtime 20:10 <+bridge> [ddnet] that would be nice 20:10 <+bridge> [ddnet] then it would "think" its the default value and wouldnt save it to config 20:20 <+bridge> [ddnet] we can also warn about high ping? 20:20 <+bridge> [ddnet] well maybe its not an issue with the new filter 20:20 <+bridge> [ddnet] it's still an issue 20:20 <+bridge> [ddnet] I see people with ping > 250 all the time 20:21 <+bridge> [ddnet] :Sadge: 20:23 <+bridge> [ddnet] So I'm out of things to try :/ 21:03 <+bridge> [ddnet] Is there a reason why we can't see other peoples key presses while they are frozen? 21:03 <+bridge> [ddnet] @Ryozuki blocklist instead of blacklist, doesnt that sound good to u xd+ 21:04 <+bridge> [ddnet] or prefer denylist? XD 21:04 <+bridge> [ddnet] blocklist or blacklist is good 21:04 <+bridge> [ddnet] denylist sounds horrible xd 21:05 <+bridge> [ddnet] XDD 21:05 <+bridge> [ddnet] @Jupstar ✪ 21:05 <+bridge> [ddnet] https://cdn.discordapp.com/attachments/293493549758939136/774001455602466856/unknown.png 21:05 <+bridge> [ddnet] nice one 21:05 <+bridge> [ddnet] mixing both 21:05 <+bridge> [ddnet] XD 21:05 <+bridge> [ddnet] oh shit xD 21:05 <+bridge> [ddnet] didnt even notice 21:05 <+bridge> [ddnet] 21:06 <+bridge> [ddnet] first time i see this ptrdiff_t 21:06 <+bridge> [ddnet] i dunno, i am just not used to use int as a datatype, so i probs didnt even notice that iam typing it xd 21:21 <+bridge> [ddnet] @Jupstar ✪ `sizeof(aVersionStrHelper) / sizeof(aVersionStrHelper[0])` 21:21 <+bridge> [ddnet] is this the best way 21:21 <+bridge> [ddnet] char aVersionStrHelper[MAX_PATH_LENGTH]; 21:21 <+bridge> [ddnet] dont u actually know the size 21:21 <+bridge> [ddnet] sizeof(aVersionStrHelper) / sizeof(char) would be the same too 21:21 <+bridge> [ddnet] yes, but then its easier to change the size 21:22 <+bridge> [ddnet] u can also remove the sizeof(char) bcs its 1 21:22 <+bridge> [ddnet] but who cares xD 21:22 <+bridge> [ddnet] well just to be sure 21:22 <+bridge> [ddnet] its the safest 21:22 <+bridge> [ddnet] xd 21:23 <+bridge> [ddnet] sizeof(aVersionStrHelper) would be enough 21:23 <+bridge> [ddnet] i dont get it 21:23 <+bridge> [ddnet] xd 21:23 <+bridge> [ddnet] yes, but the above is the safest 21:23 <+bridge> [ddnet] so whats wrong with it 21:23 <+bridge> [ddnet] @Jupstar ✪ idk most code just uses sizeof(aVersionStrHelper) without dividing 21:24 <+bridge> [ddnet] its constant expression anyway 21:24 <+bridge> [ddnet] :monkalaugh: 21:24 <+bridge> [ddnet] it still feels antipattern to me tho 21:24 <+bridge> [ddnet] well maybe others dont care 21:25 <+bridge> [ddnet] i dunno, i always code like i don't know the datatype 21:25 <+bridge> [ddnet] but u know it 21:25 <+bridge> [ddnet] (the implemenation of the datatype) 21:26 <+bridge> [ddnet] @Learath2 watcha think 21:27 <+bridge> [ddnet] okkkkkkkkkkkkkkkkkk 21:27 <+bridge> [ddnet] then how about this: in 2000 years ppl only use int for strings, bcs unicode uses all 4 billion chars 21:27 <+bridge> [ddnet] and the alien of the future will say "this guy from 2000 years ago... What a genius" 21:29 <+bridge> [ddnet] 21:30 <+bridge> [ddnet] @Jupstar ✪ hey ur more pro than me, if u think its ok then im fine 21:30 <+bridge> [ddnet] xd 21:30 <+bridge> [ddnet] 21:30 <+bridge> [ddnet] i just dont want to make a religion out of this xD 21:30 <+bridge> [ddnet] @Ryozuki what am I thinking about what? 21:30 <+bridge> [ddnet] https://github.com/ddnet/ddnet/pull/3255/files 21:30 <+bridge> [ddnet] `sizeof(char)` is an oxymoron 21:30 <+bridge> [ddnet] @Learath2 why would u use `sizeof(aVersionStrHelper) / sizeof(aVersionStrHelper[0]` for a stack allocated array 21:30 <+bridge> [ddnet] xDDD discord xDD 21:30 <+bridge> [ddnet] YO LETS MAKE A FULLSCREEN PREVIEW 21:31 <+bridge> [ddnet] ye 21:31 <+bridge> [ddnet] discord gone crazy 21:31 <+bridge> [ddnet] nice pfp btw 21:31 <+bridge> [ddnet] @Jupstar ✪ this is funnier with deen face 21:31 <+bridge> [ddnet] Don't bother with dividing by [0] for char arrays 21:31 <+bridge> [ddnet] it always appears out of nowhere 21:31 <+bridge> [ddnet] super big 21:31 <+bridge> [ddnet] We don't do it anywhere in the code either 21:31 <+bridge> [ddnet] thats what i told him 21:31 <+bridge> [ddnet] i always do it 21:31 <+bridge> [ddnet] it must be somewhere 21:31 <+bridge> [ddnet] but he said smth about 2k years in future 21:32 <+bridge> [ddnet] but shouldnt you adapt to the codebase? 21:32 <+bridge> [ddnet] isnt that the normal sane thing 21:32 <+bridge> [ddnet] 21:32 <+bridge> [ddnet] as said, i won't make a religion out of this 21:32 <+bridge> [ddnet] we must all follow teecode religion 21:32 <+bridge> [ddnet] if u want it, force push to my repo 21:32 <+bridge> [ddnet] 21:32 <+bridge> [ddnet] Well even 2k years in the future sizeof(char) has to be 1 21:32 <+bridge> [ddnet] C++ and C standards both require it 21:33 <+bridge> [ddnet] in 2k c wont exist anyway 21:33 <+bridge> [ddnet] and ddnet less 21:33 <+bridge> [ddnet] not even this driver list 21:33 <+bridge> [ddnet] but maybe they wont use chars anymore for strings 21:33 <+bridge> [ddnet] also standard can change in 2k years xD 21:33 <+bridge> [ddnet] doesnt c always keep backwards compat 21:33 <+bridge> [ddnet] register will be removed for example 21:34 <+bridge> [ddnet] from c++ 21:34 <+bridge> [ddnet] `register` will be removed for example 21:34 <+bridge> [ddnet] they can't change sizeof(char) without breaking a whole lot of assumptions 21:34 <+bridge> [ddnet] The only way I see it happening is you joining the commitee and changing it to prove me wrong 21:34 <+bridge> [ddnet] yeah, but we are 40 years into coding. 21:34 <+bridge> [ddnet] so who knows what happens in 2k years xD 21:35 <+bridge> [ddnet] rust takeover 21:35 <+bridge> [ddnet] 21:36 <+bridge> [ddnet] if rust gets better syntax and oop, not unlikely xdd 21:36 <+bridge> [ddnet] honestly the syntax isn't cancer for people just learning it without learning any other language first 21:37 <+bridge> [ddnet] i know 21:37 <+bridge> [ddnet] just like python 21:37 <+bridge> [ddnet] the worst syntax xD 21:37 <+bridge> [ddnet] it's just designed to throw off the intuition of every single developer from the past 21:37 <+bridge> [ddnet] u dont have to fight against rust 21:37 <+bridge> [ddnet] embrace it 21:37 <+bridge> [ddnet] @Learath2 which may not be a bad thing 21:37 <+bridge> [ddnet] lot of ppl hate rust at the start 21:37 <+bridge> [ddnet] i see ito n reddit 21:37 <+bridge> [ddnet] but they like it after getting the hang of it 21:37 <+bridge> [ddnet] Rust is completely designed to make you feel inadequate and bad so that you give in to letting the compiler do everything for you 21:38 <+bridge> [ddnet] `!printf("ryozoozki dumb");` 21:38 <+bridge> [ddnet] well the compiler just ensures u dont do anything stupid 21:38 <+bridge> [ddnet] its more likely we get a new language in the next 2k years xD 21:38 <+bridge> [ddnet] one that is less logic error prone too 21:38 <+bridge> [ddnet] well there are new languages everyday 21:38 <+bridge> [ddnet] yes 21:39 <+bridge> [ddnet] salong they dont have python syntax 21:39 <+bridge> [ddnet] its true tho rust still isnt mature library wise 21:39 <+bridge> [ddnet] cancerous 21:39 <+bridge> [ddnet] sadly blender uses it for plugins xD 21:39 <+bridge> [ddnet] almost died 21:39 <+bridge> [ddnet] i dont like weak typed languages 21:39 <+bridge> [ddnet] they can be scripting but pls dont make it weak typed 21:39 <+bridge> [ddnet] i dont like script lanuages 21:39 <+bridge> [ddnet] nothing is more glorious than anticompletion 21:39 <+bridge> [ddnet] auto 21:39 <+bridge> [ddnet] completiong 21:39 <+bridge> [ddnet] lmao 21:39 <+bridge> [ddnet] my iq lowering 21:40 <+bridge> [ddnet] cant type 21:40 <+bridge> [ddnet] python is love 21:41 <+bridge> [ddnet] @Jupstar ✪ java 21:41 <+bridge> [ddnet] 21:41 <+bridge> [ddnet] i hear java, i think of modded minecraft killing my pc with 20FPS 21:42 <+bridge> [ddnet] class VerySmallJavaClass { 21:42 <+bridge> [ddnet] ups 21:42 <+bridge> [ddnet] nono 21:42 <+bridge> [ddnet] class VerySmallJavaClass { 21:42 <+bridge> [ddnet] int a; 21:42 <+bridge> [ddnet] }; 21:42 <+bridge> [ddnet] 21:42 <+bridge> [ddnet] VerySmallJavaClass pTest = new VerySmallJavaClass(); 21:42 <+bridge> [ddnet] how much bytes has this xD 21:43 <+bridge> [ddnet] 10GB 21:43 <+bridge> [ddnet] private final static volatile String MY_HOUSE_ADDRESS = "HOOOO"; 21:43 <+bridge> [ddnet] 😮 21:44 <+bridge> [ddnet] StringFactory 21:44 <+bridge> [ddnet] xD 21:44 <+bridge> [ddnet] isnt java where there are lot of factories 21:44 <+bridge> [ddnet] now the fun begins 21:44 <+bridge> [ddnet] java is like Rome, infinite ways and all lead to the same goal 21:45 <+bridge> [ddnet] i think dependency injection removes the need of factories 21:45 <+bridge> [ddnet] 21:45 <+bridge> [ddnet] JNI 21:45 <+bridge> [ddnet] and u back to a good language xd 21:45 <+bridge> [ddnet] https://dagger.dev/ 21:45 <+bridge> [ddnet] 21:45 <+bridge> [ddnet] Oh yeah kotlin 21:46 <+bridge> [ddnet] almost as cancerous as python xd 21:46 <+bridge> [ddnet] for me lua is worse 21:46 <+bridge> [ddnet] lua looks ugly af 21:46 <+bridge> [ddnet] Mhh let me see 21:47 <+bridge> [ddnet] oh yeah 21:47 <+bridge> [ddnet] bit strange 21:47 <+bridge> [ddnet] they took the worst from all languages 21:47 <+bridge> [ddnet] including pythonx xD 21:47 <+bridge> [ddnet] Hello 21:47 <+bridge> [ddnet] my friend trying to make an server but when he trying to join " there's no server " " on LAN " 21:48 <+bridge> [ddnet] my friend trying to make server but when he trying to join " there's no server " " on LAN " 21:48 <+bridge> [ddnet] then he probably isnt in the same router 21:48 <+bridge> [ddnet] if he is trying to join your server he should look on the internet tab 21:48 <+bridge> [ddnet] > if he is trying to join your server he should look on the internet tab 21:48 <+bridge> [ddnet] @plsplsplslol nah , his server 21:49 <+bridge> [ddnet] is he filtering out servers that have no players on them 21:49 <+bridge> [ddnet] its either a filter, or its the server that doesnt start correctly 21:50 <+bridge> [ddnet] good now 21:50 <+bridge> [ddnet] ty 21:52 <+bridge> [ddnet] @Jupstar ✪ @Learath2 21:52 <+bridge> [ddnet] https://cdn.discordapp.com/attachments/293493549758939136/774013202941870110/uc0v6po1qgx51.png 21:52 <+bridge> [ddnet] :gachiGASM: 21:52 <+bridge> [ddnet] looks cool 😄 21:52 <+bridge> [ddnet] https://www.reddit.com/r/cableporn/ 21:52 <+bridge> [ddnet] :BASED: 21:52 <+bridge> [ddnet] some day when im rich, i make a pc that looks better than it performs 21:53 <+bridge> [ddnet] so basically an Apple PC 21:55 <+bridge> [ddnet] liquid cooling makes the pc prettier 21:55 <+bridge> [ddnet] but i think it doesnt perform better than air 21:55 <+bridge> [ddnet] just the same 21:55 <+bridge> [ddnet] put pc in fish tank :greenthing: 21:55 <+bridge> [ddnet] i saw a pc 21:55 <+bridge> [ddnet] inside a special liquid 21:55 <+bridge> [ddnet] no electrical conductivity but high thermal conductivity 21:56 <+bridge> [ddnet] well pyshically its performs better.. imagine u have heat and just move it somewhere else 😄 21:56 <+bridge> [ddnet] u could see the liquid boiling at the cpu 21:56 <+bridge> [ddnet] but it probs isnt worth if u arent completly overclocking 21:56 <+bridge> [ddnet] do gamers in antarctica get more fps 21:56 <+bridge> [ddnet] probably 21:57 <+bridge> [ddnet] if u get at -270°C u around the level of supraconductor 21:57 <+bridge> [ddnet] dunno if thats the english word but its an effect of quantons 21:58 <+bridge> [ddnet] dunno if thats the english word but its an effect of quantens 21:58 <+bridge> [ddnet] dunno if thats the english word but its an effect of quantum 22:03 <+bridge> [ddnet] superconductor? 22:03 <+bridge> [ddnet] yeah