00:04 <+bridge> [ddnet] __int128 seems to optimize even worse 00:07 <+bridge> [ddnet] https://godbolt.org/z/iL8fVm 00:08 <+bridge> [ddnet] dunno what you're talking about, unsigned __int128 optimizes slightly better for `check_flag` and is the same optimum for `add` 00:11 <+bridge> [ddnet] it's been a while since I checked it, I think it was optimized less consistently across compilers 00:11 <+bridge> [ddnet] (and isn't supported at all in MSVC) 00:14 <+bridge> [ddnet] besides https://godbolt.org/z/BSAXzu std::bitset optimizes better in all compilers, if all you want is checking and setting bits 00:15 <+bridge> [ddnet] oops 00:15 <+bridge> [ddnet] hm 00:17 <+bridge> [ddnet] (atleast it optimizes to less instructions, I know that doesn't always mean better performance and it'd probably be impossible to predict with the insane pipelines of new cpus) 00:17 <+bridge> [ddnet] s\/optimizes/generates/ 00:17 <+bridge> [ddnet] nah, should be better 00:17 <+bridge> [ddnet] no branches 00:19 <+bridge> [ddnet] If you want to do arithmetic though, _int128 does work if it exists 00:19 <+bridge> [ddnet] arithmetic optimized fine in my own impl too 00:20 <+bridge> [ddnet] + and - are trivial, / and * are the ones that are hard to get right 00:20 <+bridge> [ddnet] <ᶰ°Konͧsti> Most important DDNet dev change my mind 00:20 <+bridge> [ddnet] <ᶰ°Konͧsti> https://cdn.discordapp.com/attachments/293493549758939136/729824133499846727/robot.png 00:21 <+bridge> [ddnet] most important ddnet bug tester* 00:24 <+bridge> [ddnet] <ᶰ°Konͧsti> :troll: 00:24 <+bridge> [ddnet] Most important DDNet dev**,** change my mind 00:24 <+bridge> [ddnet] the comma really helps 00:24 <+bridge> [ddnet] > Most important DDNet dev**,** change my mind 00:24 <+bridge> [ddnet] @heinrich5991 and funnily enough, clang optimizes both your add and the native add to the exact same instructions, which is shorter than the gcc one 00:25 <+bridge> [ddnet] hum 00:27 <+bridge> [ddnet] and without using compiler intrinsics, the msvc one just looks abysmal 😄 00:30 <+bridge> [ddnet] https://godbolt.org/z/EgMHWm 00:30 <+bridge> [ddnet] even with the intrinsics the code generated is just not as clean as clang and gcc, but atleast it's readable 00:30 <+bridge> [ddnet] it sets up a stack 00:31 <+bridge> [ddnet] ah 00:31 <+bridge> [ddnet] maybe you need that due to SSEH 00:31 <+bridge> [ddnet] SEH* 00:31 <+bridge> [ddnet] It also actually calls the constructor 00:32 <+bridge> [ddnet] it should be a trivial constructor to inline, I wonder why it won't 00:35 <+bridge> [ddnet] lol there is no /O3 in msvc that's why 00:37 <+bridge> [ddnet] at /O2 it looks better, still not as cute as clang or gcc 01:58 <+bridge> [ddnet] found a bug where the score hud at the bottom right corner doesnt show 01:58 <+bridge> [ddnet] i downloaded ddnet source with `git clone --recursive https://github.com/ddnet/ddnet` and ran it with the cmake gui for windows on visual studio 2017 10:44 <+bridge> [ddnet] https://sqlite.org/testing.html#coverage_testing_of_defensive_code 10:45 <+bridge> [ddnet] interesting 10:46 <+bridge> [ddnet] https://docs.microsoft.com/en-us/cpp/build/reference/rtc-run-time-error-checks?view=vs-2019 could enable /RTC1 on ci 11:07 <+ChillerDragon> yea sqlite is crazy well tested 11:07 <+ChillerDragon> the lite name might be confusing its acually very pro imo :D 11:10 <+bridge> [ddnet] sqlpro? 😉 11:45 <+bridge> [ddnet] Should we get rid of the very meh file score backend for an sqlite one? 11:52 <+bridge> [freenode] I actually did that over the weekend. I'll try to make the PR today. 11:57 <+bridge> [ddnet] \o/ 12:00 <+bridge> [ddnet] https://github.com/teeworlds/teeworlds/pull/2671 12:00 <+bridge> [ddnet] Can we add smth like that in ddnet 😂 12:02 <+bridge> [ddnet] Would require quite a lot of work to get into an acceptable form 12:06 <+bridge> [ddnet] @Trafalgar Law didn't you say you tested #2433 and it worked mostly fine? 12:06 <+bridge> [ddnet] https://github.com/ddnet/ddnet/pull/2433 12:06 <+bridge> [ddnet] deen seems to have had a very different experience 😄 12:07 <+bridge> [ddnet] Zwelf: Very cool! 12:07 <+bridge> [ddnet] @Learath2 I tested it yea, as I said the most problem was the double msges 12:09 <+bridge> [ddnet] @deen you sure the same ip check wasn't messing with you? 12:09 <+bridge> [ddnet] And the vote was not abored, after the vote caller left the srv 12:09 <+bridge> [ddnet] they were connected with 3 different ips, checked in `status` 12:13 <+bridge> [ddnet] Ah I know what went wrong there 12:15 <+bridge> [ddnet] Talking about code testing in SQL databases, I just got my first patent in that area. Should be granted in about a week. But I should really write a paper about it because patents in lawyer-speak are hard to understand. 12:16 <+bridge> [ddnet] Are there lawyers that specialize in software patents? I'd be surprised if most lawyers could really adequately put this stuff in legalese 12:17 <+bridge> [ddnet] @deen the last commit should fix the voting issue 12:27 <+bridge> [ddnet] @Learath2 maybe ddnet has the same problem with 0.7 register as the normal 0.7 server 12:28 <+bridge> [ddnet] does it? I didn't notice losing master connection 12:28 <+bridge> [ddnet] i have no idea, you said something about not registering 12:28 <+bridge> [ddnet] some messages up 12:28 <+bridge> [ddnet] Votes 12:28 <+bridge> [ddnet] no 12:28 <+bridge> [ddnet] He mean votes 12:29 <+bridge> [ddnet] Look up :P 12:29 <+bridge> [ddnet] no 12:29 <+bridge> [ddnet] ok 12:29 <+bridge> [ddnet] lol 12:41 <+bridge> [ddnet] @Learath2 yeah, the one we had was a former engineer who then became a lawyer on top, so he's really good at understanding the technical descriptions we gave him 12:44 <+bridge> [ddnet] @Learath2 just tried, the vote for 0.7 player still doesn't register on kick vote 12:45 <+bridge> [ddnet] it also shows 50% voted already, so I think the 0.7 player is totally getting ignored there 12:45 <+bridge> [ddnet] but for map votes it works 12:45 <+bridge> [ddnet] Okay, I fixed what I saw by looking at the code, I'll take a better look tonight 12:46 <+bridge> [ddnet] It'd be nice if you could take a look where the vote disappears, but if you don't have the time I'll fix it sooner or later 13:28 <+bridge> [ddnet] when RTX support :3Head: 15:20 <+bridge> [ddnet] :tee_thinking: 15:37 <+bridge> [ddnet] Teeworlds RTX :kek: 15:37 <+bridge> [ddnet] I'm having trouble building with `cmake -DCLIENT=OFF -DMYSQL=ON ..` 15:37 <+bridge> [ddnet] ``` 15:37 <+bridge> [ddnet] -- Could NOT find Notify (missing: NOTIFY_LIBRARIES NOTIFY_INCLUDE_DIRS) 15:38 <+bridge> [ddnet] -- Could NOT find OpenGL (missing: OPENGL_opengl_LIBRARY OPENGL_glx_LIBRARY OPENGL_INCLUDE_DIR) 15:38 <+bridge> [ddnet] 15:38 <+bridge> [ddnet] ``` 15:39 <+bridge> [ddnet] Shouldn't depend on opengl with `-DCLIENT=OFF`, right? 15:46 <+bridge> [ddnet] @timakro does it not build as a result of that or does it just report that those are not found? 15:46 <+bridge> [ddnet] Doesn't build 15:46 <+bridge> [ddnet] error message? 15:46 <+bridge> [ddnet] that 15:46 <+bridge> [ddnet] That's not an error message 15:47 <+bridge> [ddnet] can you give the complete output pelase? 15:47 <+bridge> [ddnet] It wouldn't stop it from building 15:47 <+bridge> [ddnet] can you give the complete output please? 15:47 <+bridge> [ddnet] oh the real error is above sry 15:47 <+bridge> [ddnet] says 15:47 <+bridge> [ddnet] ``` 15:47 <+bridge> [ddnet] CMake Error at CMakeLists.txt:433 (message): 15:47 <+bridge> [ddnet] You must install Python to compile DDNet 15:48 <+bridge> [ddnet] ``` 15:48 <+bridge> [ddnet] is the problem fixed? 15:48 <+bridge> [ddnet] or do you already have python installed? 15:48 <+bridge> [ddnet] i have python installed xD 15:49 <+bridge> [ddnet] can you give us the complete output then? ^^ 15:49 <+bridge> [ddnet] ``` 15:49 <+bridge> [ddnet] $ cmake -DCLIENT=OFF -DMYSQL=ON .. ~/src/ddnet/build 15:49 <+bridge> [ddnet] -- Could NOT find PkgConfig (missing: PKG_CONFIG_EXECUTABLE) 15:49 <+bridge> [ddnet] -- Could NOT find Freetype (missing: FREETYPE_LIBRARY FREETYPE_INCLUDEDIR) 15:49 <+bridge> [ddnet] -- Could NOT find GLEW (missing: GLEW_INCLUDE_DIR GLEW_LIBRARY) 15:49 <+bridge> [ddnet] -- Could NOT find GTest (missing: GTEST_LIBRARY GTEST_INCLUDE_DIR GTEST_MAIN_LIBRARY) 15:49 <+bridge> [ddnet] -- Could NOT find Ogg (missing: OGG_INCLUDEDIR) 15:49 <+bridge> [ddnet] -- Could NOT find Opus (missing: OPUS_INCLUDEDIR) 15:49 <+bridge> [ddnet] -- Could NOT find Opusfile (missing: OPUSFILE_LIBRARY OPUSFILE_INCLUDEDIR) 15:49 <+bridge> [ddnet] -- Could NOT find PythonInterp: Found unsuitable version "2.7.16", but required is at least "3" (found /usr/bin/python) 15:49 <+bridge> [ddnet] -- Could NOT find SDL2 (missing: SDL2_LIBRARY SDL2_INCLUDEDIR) 15:49 <+bridge> [ddnet] -- ******** DDNet ******** 15:49 <+bridge> [ddnet] -- Target OS: linux 64bit 15:50 <+bridge> [ddnet] -- Compiler: /usr/bin/c++ 15:50 <+bridge> [ddnet] -- Build type: Release 15:50 <+bridge> [ddnet] -- Dependencies: 15:50 <+bridge> [ddnet] -- * Curl found 15:50 <+bridge> [ddnet] -- * Freetype not found 15:50 <+bridge> [ddnet] -- * Glew not found (using bundled version) 15:50 <+bridge> [ddnet] -- * GTest not found 15:50 <+bridge> [ddnet] -- * MySQL found 15:50 <+bridge> [ddnet] -- * Ogg not found 15:50 <+bridge> [ddnet] -- * OpenSSL Crypto found 15:50 <+bridge> [ddnet] -- * Opus not found 15:50 <+bridge> [ddnet] -- * Opusfile not found 15:50 <+bridge> [ddnet] -- * Pnglite not found (using bundled version) 15:50 <+bridge> [ddnet] -- * PythonInterp not found 15:50 <+bridge> [ddnet] -- * SDL2 not found 15:50 <+bridge> [ddnet] -- * Wavpack not found (using bundled version) 15:50 <+bridge> [ddnet] -- * Zlib found 15:50 <+bridge> [ddnet] CMake Error at CMakeLists.txt:433 (message): 15:50 <+bridge> [ddnet] On debian /usr/bin/python is python 2.7 i guess it expects python 3 there 15:50 <+bridge> [ddnet] `python3 --version` please? 15:50 <+bridge> [ddnet] should look for py3 first, I think 15:50 <+bridge> [ddnet] ``` 15:50 <+bridge> [ddnet] $ python3 --version 15:50 <+bridge> [ddnet] Python 3.7.3 15:50 <+bridge> [ddnet] ``` 15:50 <+bridge> [ddnet] hm 15:51 <+bridge> [ddnet] I think I got a report that you have to delete the build directory and try again after some recent change 15:51 <+bridge> [ddnet] can you try that (or create a new build directory) 15:51 <+bridge> [ddnet] sure 15:52 <+bridge> [ddnet] Works, thx 15:52 <+bridge> [ddnet] np 🙂 15:52 <+bridge> [ddnet] Why isn't the error message at the bottom -.- 15:53 <+bridge> [ddnet] I wonder why the error doesn't immediately stop cmake 15:53 <+bridge> [ddnet] It's not a warning 15:53 <+bridge> [ddnet] we can do that (FATAL_ERROR), but then it won't report all errors 15:54 <+bridge> [ddnet] Hm, true 16:00 <+bridge> [ddnet] @Learath2 How do I get started with the 0.7 support? Is there a setting I need to enable? Do I need to convert maps myself and where to put them? 16:00 <+bridge> [ddnet] I'm adding it to trashmap 16:02 <+bridge> [ddnet] You only need maps in the maps7 folder 16:02 <+bridge> [ddnet] That's it, it's enabled by default 16:02 <+bridge> [ddnet] Okay great 17:45 <+bridge> [ddnet] So when ddnet rtx in vr? 17:46 <+bridge> [ddnet] :kek: 18:04 <+bridge> [ddnet] @deen vacotion already over 🤔 ? 18:13 <+bridge> [ddnet] when teeworlds on nintendo ds 18:14 <+bridge> [ddnet] I tried to compile it for switch but the reverse engineered libraries just weren't far enough along 18:14 <+bridge> [ddnet] oh lol 18:18 <+bridge> [ddnet] @timakro and for the conversion map_convert_07 is what we use 18:44 <+bridge> [ddnet] https://cdn.discordapp.com/attachments/293493549758939136/730101906277662748/IMG_20200707_123934.jpg 18:44 <+bridge> [ddnet] That is okay? 19:07 <+bridge> [ddnet] Is that okay? 19:28 <+bridge> [ddnet] ok in what sense? 19:34 <+bridge> [ddnet] <ᶰ°Konͧsti> I think how he placed translation 19:38 <+bridge> [ddnet] https://cdn.discordapp.com/attachments/293493549758939136/730115509864890428/Untitled.png 19:38 <+bridge> [ddnet] did hook collision always work like this on solo servers? 19:42 <+bridge> [ddnet] I don't think it should 19:42 <+bridge> [ddnet] what client version? 19:42 <+bridge> [ddnet] works for me on current client 19:54 <+bridge> [ddnet] <ᶰ°Konͧsti> Wasnt there a bug too where hookcolls colored when they crossed at their ends? 19:56 <+bridge> [ddnet] newest ddnet 19:56 <+bridge> [ddnet] 14.1 19:56 <+bridge> [ddnet] I think it was cause by the unsolo tiles that are in the solo map xD 19:58 <+bridge> [ddnet] Is it possible to reconnect with the master srv? 19:58 <+bridge> [ddnet] Because my server lost the connection to the master srv from 0.7 20:01 <+bridge> [ddnet] @deen Ok in the sense that if it is well written to translate it 20:04 <+ChillerDragon> restart server @Trafalgar Law 20:04 <+ChillerDragon> btw check wire 20:05 <+ChillerDragon> maybe you have a similar issue like @fokkonaut has see #teeworlds for details 20:05 <+bridge> [ddnet] After the restart the srv has still no connection to the master srv? 20:05 <+bridge> [ddnet] I can't see the srv 20:05 <+bridge> [ddnet] @TearlessAbyss46 yeah, unsolo tiles shouldn't even exist in solo map. do they work? 20:06 <+bridge> [ddnet] maybe the mapper thought it would be cool to be able to play the map without the solo map setting 20:22 <+bridge> [ddnet] <ᶰ°Konͧsti> but u dont need unsolo tiles for that because u dont even need solotiles 20:26 <+bridge> [ddnet] <ᶰ°Konͧsti> @deen they work like they should but by server setting sv_soloserver 1 everx interacting with other tees is impossible so u cant hook them but the hookcoll thinks u can because its not like a "u cant hook other tees" tile 20:40 <+bridge> [ddnet] cool bug 21:41 <+bridge> [ddnet] @Learath2 works, can you take a look at the code? https://github.com/fokkonaut/F-DDrace/commit/4e0abd5e430781e0f206a934fbee6b40e88fd96f 22:35 <+bridge> [ddnet] hi, the new smooth zoom is annoying me and for some reason the config flag is not working for me. am i doing something wrong? 22:35 <+bridge> [ddnet] https://cdn.discordapp.com/attachments/293493549758939136/730160030988369931/ddnet_zoom.png 22:35 <+bridge> [ddnet] https://cdn.discordapp.com/attachments/293493549758939136/730160069957648515/ddnet_zoom2.png 22:48 <+bridge> [ddnet] @nils there is no flag to disable the smooth zoom 22:50 <+bridge> [ddnet] but the second picture is from the commit the smooth zoom was added. it shows the flag 22:50 <+bridge> [ddnet] in which way is it annoying you? 22:52 <+bridge> [ddnet] @nils no, that screenshot is from https://github.com/ddnet/ddnet/pull/2420/files, which we did *not* merge. Instead we took https://github.com/ddnet/ddnet/pull/2429/files 22:52 <+bridge> [ddnet] @Patiga i am zooming quite alot while playing and the animation is just braking my focus 22:52 <+bridge> [ddnet] @deen he is right, it annoys me too 22:52 <+bridge> [ddnet] thats why i was against the removal of a config option by @heinrich5991 22:53 <+bridge> [ddnet] oh i'm sorry @deen didn't see that it wasn't merged. but the setting would be a blessing 22:55 <+bridge> [ddnet] @fokkonaut if that was why, why didn't you say sth? smh 22:56 <+bridge> [ddnet] We should add the config option back, and add a special case for 0 22:59 <+bridge> [ddnet] done 22:59 <+bridge> [ddnet] @Learath2 ^ 23:02 <+bridge> [ddnet] but i need a setting to turn it off, not to tweak the time. or should i set cl_smooth_zoom_time to 0 now? 23:04 <+bridge> [ddnet] i heavily agree with nils 23:04 <+bridge> [ddnet] i dont undersand u can implement such stuff with no off option 23:04 <+bridge> [ddnet] 0 is the same as off 23:04 <+bridge> [ddnet] no need to have two options imo 23:05 <+bridge> [ddnet] well ok, thanks alot :) 23:05 <+bridge> [ddnet] fair enough then 23:05 <+bridge> [ddnet] @snail our combination of options grows exponentially with each added options, we can't possible test all, thus we end up having more bugs 23:05 <+bridge> [ddnet] yeah ik 23:05 <+bridge> [ddnet] so I can understand the motivation not to introduce more options 23:06 <+bridge> [ddnet] but that option literally changes the gameplay when using zoom 23:06 <+bridge> [ddnet] :justatest: 23:07 <+bridge> [ddnet] @Patiga i am zooming quite alot while playing and the animation is just breaking my focus 23:18 <+bridge> [ddnet] @deen doesn't this cause a SIGFPE? 23:19 <+bridge> [ddnet] `return (CurrentTime - m_ZoomSmoothingStart) / (m_ZoomSmoothingEnd - m_ZoomSmoothingStart);` would be `(Now - Now) / (Now - Now)` which is not only division by 0 but also 0/0 😄 23:22 <+bridge> [ddnet] locally it worked, I didn't even look at the calculation 😄 23:22 <+bridge> [ddnet] maybe I got lucky and the floats are not exact enough? 23:29 <+bridge> [ddnet] hi can someone fact check some facts for a video im making 23:29 <+bridge> [ddnet] 1. strong and week hook existed from the beginning of tw 23:29 <+bridge> [ddnet] 2. whoever was spawned first gets strong 23:29 <+bridge> [ddnet] @Learath2 ZoomProgress is never called since line 71 checks that Time >= m_ZoomSmoothingEnd 23:30 <+bridge> [ddnet] 2 is correct 23:30 <+bridge> [ddnet] for 1 you should grab an old teewars build 23:30 <+bridge> [ddnet] @louis stop using incorrect skin 23:31 <+bridge> [ddnet] @deen oh, that makes sense 23:32 <+bridge> [ddnet] @deen do you know anywhere i can find them? 23:33 <+bridge> [ddnet] ask zombietoad he has them 23:33 <+bridge> [ddnet] Teewars :kek: 23:33 <+bridge> [ddnet] sry i dont have anything for louisers 23:34 <+bridge> [ddnet] trol 23:34 <+bridge> [ddnet] Zombietroll 23:35 <+bridge> [ddnet] who is marcell 23:35 <+bridge> [ddnet] https://www.teeworlds.com/forum/viewtopic.php?id=8016 for example 23:36 <+bridge> [ddnet] I am your father 23:40 <+bridge> [ddnet] heh funny how the server and the client was just one executable 23:49 <+bridge> [ddnet] seems like strong and weak does not exist or appear to exist on teewars 0.1.1 23:49 <+bridge> [ddnet] the hook physics are also really crazy