Config
Log for #openttd on 26th February 2021:
Times are UTC Toggle Colours
00:00:20  <Xaroth> just need to figure out when to exactly sync the copy of the tile map so it doesn't spaz out.
00:04:56  <Xaroth> also figure out why it's constantly seeing type changes on every chunk of the map on bigger maps
00:07:30  *** daspork has quit IRC
00:07:31  *** daspork has joined #openttd
00:09:33  <_dp_> hm, server regularly checks this->HasClientQuit() but does it ever set it?
00:12:09  <Xaroth> hm, frequently changes between MP_TREES and MP_CLEAR
00:12:55  *** ChanServ sets mode: +v orudge
00:12:55  *** ChanServ sets mode: +v Terkhen
00:12:55  *** ChanServ sets mode: +v planetmaker
00:12:55  *** ChanServ sets mode: +v peter1138
00:12:55  *** ChanServ sets mode: +v DorpsGek
00:13:21  <Xaroth> guess we should ignore that
00:13:44  <_dp_> Xaroth, yeah, that sounds normal with tree growth
00:17:21  *** Progman has quit IRC
00:19:28  *** APTX has quit IRC
00:19:30  *** APTX has joined #openttd
00:20:59  *** gelignite has quit IRC
00:22:31  *** jinks has quit IRC
00:22:34  *** jinks has joined #openttd
00:23:02  *** Beerbelott has quit IRC
00:25:42  *** jottyfan has joined #openttd
00:26:19  *** jottyfan has joined #openttd
00:26:47  <Xaroth> heh, there's also a switch between MP_WATER and MP_TREES
00:27:13  <Xaroth> dbg: [map] [UpdateShadowTile] Tile type changed: 8627 6 4
00:27:14  <Xaroth> dbg: [map] [UpdateShadowTile] Tile type changed: 8627 4 6
00:27:19  <Xaroth> water to trees, trees back to water
00:27:51  *** k-man has quit IRC
00:28:13  *** k-man has joined #openttd
00:28:36  *** Sacro has quit IRC
00:28:36  <Xaroth> or rather, other way around
00:28:47  *** Sacro has joined #openttd
00:29:56  <_dp_> that's probably shore tiles
00:36:17  *** roadt_ has quit IRC
00:36:33  *** roadt_ has joined #openttd
00:37:56  *** nnyby has quit IRC
00:37:57  *** nnyby has joined #openttd
00:49:56  *** orudge has quit IRC
00:49:56  *** orudge has joined #openttd
00:52:31  *** mindlesstux has quit IRC
00:52:42  *** mindlesstux has joined #openttd
00:52:57  *** LordArtoo has joined #openttd
00:54:16  *** LordAro has quit IRC
01:03:20  *** glx has quit IRC
01:06:11  <DorpsGek> [OpenTTD/OpenSFX] Wuzzy2 updated pull request #22: Shorten some description texts https://git.io/JtF05
01:06:34  <DorpsGek> [OpenTTD/OpenSFX] Wuzzy2 commented on pull request #22: Shorten some description texts https://git.io/JtbYM
01:07:42  *** luaduck has quit IRC
01:07:42  *** luaduck has joined #openttd
01:08:20  *** innocenat_ has quit IRC
01:08:31  *** innocenat_ has joined #openttd
01:09:56  <DorpsGek> [OpenTTD/OpenSFX] Wuzzy2 commented on pull request #19: Replace Mercurial code with Git code https://git.io/JtbYQ
01:16:31  *** fnutt has quit IRC
01:16:38  *** fnutt has joined #openttd
01:21:08  *** skrzyp has quit IRC
01:21:21  *** skrzyp has joined #openttd
01:22:17  *** iSoSyS has joined #openttd
01:23:00  *** Flygon has joined #openttd
01:24:20  *** m1cr0man has quit IRC
01:24:44  *** m1cr0man has joined #openttd
01:39:54  *** osvaldo[m] has quit IRC
01:39:58  *** osvaldo[m] has joined #openttd
01:42:28  *** murr4y has quit IRC
01:42:41  *** murr4y has joined #openttd
01:46:44  *** JamesRoss[m] has quit IRC
01:46:49  *** JamesRoss[m] has joined #openttd
01:51:19  *** khavik[m] has quit IRC
01:51:27  *** khavik[m] has joined #openttd
01:53:43  *** Tirili has quit IRC
02:00:42  *** shedidthedog[m] has quit IRC
02:00:48  *** shedidthedog[m] has joined #openttd
02:10:02  *** yur3shmukcik[m] has quit IRC
02:10:11  *** yur3shmukcik[m] has joined #openttd
02:10:50  *** karl[m]5 has quit IRC
02:10:55  *** karl[m]5 has joined #openttd
02:11:06  *** Eddi|zuHause has quit IRC
02:12:10  *** natalie[m] has quit IRC
02:12:12  *** natalie[m] has joined #openttd
02:15:08  *** Eddi|zuHause has joined #openttd
02:15:08  *** snail_UES_ has quit IRC
02:15:18  *** snail_UES_ has joined #openttd
02:18:12  *** CornsMcGowan[m] has quit IRC
02:18:17  *** CornsMcGowan[m] has joined #openttd
02:38:40  *** iSoSyS has quit IRC
02:46:55  *** Ttech has quit IRC
02:48:35  *** Ttech has joined #openttd
02:58:00  *** Tirili has joined #openttd
03:03:31  *** Wormnest has quit IRC
03:14:50  *** Heiki[m] has quit IRC
03:14:54  *** Heiki[m] has joined #openttd
03:15:06  *** EmeraldSnorlax[m] has quit IRC
03:15:09  *** EmeraldSnorlax[m] has joined #openttd
03:20:09  *** debdog has joined #openttd
03:23:32  *** D-HUND has quit IRC
03:26:55  *** Tirili has quit IRC
03:29:08  *** Speeder has quit IRC
03:29:31  *** Speeder has joined #openttd
03:32:58  *** muffindrake has joined #openttd
03:34:45  *** muffindrake4 has quit IRC
03:56:55  *** heffer has quit IRC
03:56:56  *** heffer has joined #openttd
04:16:07  *** Gustavo6046 has quit IRC
04:16:09  *** Gustavo6046 has joined #openttd
04:20:04  *** Extrems has quit IRC
04:20:10  *** Extrems has joined #openttd
04:27:43  *** WormnestAndroid has quit IRC
04:27:59  *** WormnestAndroid has joined #openttd
04:56:09  <DorpsGek> [OpenTTD/nml] Andrew350 commented on pull request #173: Fix: Access to persistent storage of towns https://git.io/JtbcM
05:07:48  *** didac has quit IRC
05:08:36  *** lobster has quit IRC
05:12:07  *** y2kboy23 has quit IRC
05:12:10  *** Cursarion has quit IRC
05:12:13  *** Cursarion has joined #openttd
05:12:18  *** y2kboy23 has joined #openttd
05:14:36  *** snail_UES_ is now known as Guest56
05:14:36  *** snail_UES_ has joined #openttd
05:26:34  *** argoneus has quit IRC
05:26:42  *** argoneus has joined #openttd
05:27:00  *** didac has joined #openttd
05:47:46  *** Wuzzy2 has quit IRC
05:58:07  *** andythenorth has joined #openttd
06:01:46  *** blathijs has quit IRC
06:02:20  *** blathijs has joined #openttd
06:17:14  *** Yexo has quit IRC
06:17:30  *** tneo has quit IRC
06:17:46  *** ^Spike^ has quit IRC
06:17:59  *** Yexo has joined #openttd
06:17:59  *** tneo has joined #openttd
06:18:17  *** ^Spike^ has joined #openttd
06:18:20  *** andythenorth has quit IRC
06:24:43  *** andythenorth has joined #openttd
06:36:42  *** cawal[m] has quit IRC
06:36:50  *** cawal[m] has joined #openttd
06:43:06  *** nartir[m] has quit IRC
06:43:08  *** nartir[m] has joined #openttd
06:44:27  *** HerzogDeXtEr has joined #openttd
07:03:43  *** colde has quit IRC
07:03:44  *** colde has joined #openttd
07:22:02  *** jeremy[m] has quit IRC
07:22:07  *** jeremy[m] has joined #openttd
07:22:50  *** einar[m] has quit IRC
07:22:52  *** einar[m] has joined #openttd
07:26:32  *** sla_ro|master has joined #openttd
07:37:34  *** twpol has quit IRC
07:37:45  *** twpol has joined #openttd
08:07:38  *** grag[m] has quit IRC
08:07:47  *** grag[m] has joined #openttd
08:08:08  *** Maarten has quit IRC
08:09:47  *** Wolf01 has joined #openttd
08:11:36  *** crem2 has quit IRC
08:11:47  *** crem2 has joined #openttd
08:13:45  *** LordArtoo is now known as LordAro
08:15:20  *** pothyurf[m] has quit IRC
08:15:25  *** pothyurf[m] has joined #openttd
08:15:36  *** bkilm[m] has quit IRC
08:15:39  *** bkilm[m] has joined #openttd
08:18:24  *** Maarten has joined #openttd
08:27:24  <DorpsGek> [OpenTTD/OpenTTD] TrueBrain approved pull request #8743: Change: Default to a 32bpp blitter. https://git.io/JtbuT
08:29:27  <Wolf01> Ohhh, bigger font is more readable from the bed, I can't undrstand why I didn't figure it out before
08:35:35  <andythenorth> get bigger eyes
08:35:48  <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on pull request #8715: Fix: Vehicle list windows did not update when this year's profit changed https://git.io/JtbuE
08:35:56  <Wolf01> Lol
08:41:30  *** Tulitomaatti has quit IRC
08:41:43  *** Tulitomaatti has joined #openttd
08:46:32  * andythenorth considers wrapping python around the CHIPS C-pre-processor templating
08:47:50  <Wolf01> Why don't you develop an IDE (web based) to create grfs in an user friendly way?
08:47:51  <andythenorth> make -> instantiate python classes for station tiles from config files -> python template .cpp defines -> run the cpp -> nfo -> grfcodec -> grf
08:48:23  <andythenorth> Wolf01 I did have a web-based grf configurator, but I deleted it
08:48:53  <Wolf01> :(
08:49:25  <andythenorth> no big loss, it was quite horrible
09:00:04  <TrueBrain> hmm ... DrawViewport is called 14 times, where I expected it to be called once
09:01:17  *** supermop_Home has quit IRC
09:03:36  *** greeter has quit IRC
09:03:51  *** greeter has joined #openttd
09:03:59  <DorpsGek> [OpenTTD/OpenTTD] LordAro commented on pull request #8715: Fix: Vehicle list windows did not update when this year's profit changed https://git.io/Jtbgt
09:10:07  <DorpsGek> [OpenTTD/OpenSFX] orudge commented on pull request #19: Replace Mercurial code with Git code https://git.io/Jtbgz
09:11:06  <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on pull request #8715: Fix: Vehicle list windows did not update when this year's profit changed https://git.io/Jtbgw
09:11:12  <DorpsGek> [OpenTTD/OpenSFX] orudge approved pull request #22: Shorten some description texts https://git.io/Jtbgo
09:11:28  <DorpsGek> [OpenTTD/OpenSFX] orudge merged pull request #22: Shorten some description texts https://git.io/JtF05
09:18:32  <TrueBrain> sorry LordAro ; I simply do not understand enough of modern C++ to judge that :) It seems you are absolutely correct, but *shrug* :)
09:20:37  <LordAro> TrueBrain: me either :p
09:21:39  <TrueBrain> :D
09:33:08  <TrueBrain> so our window dirty-rect system is pretty clever, minimizes what needs to be drawn
09:33:13  *** Wolf01 has quit IRC
09:33:14  <TrueBrain> but it makes optimizing Viewport a lot harder :P
09:33:47  *** Wolf01 has joined #openttd
09:33:49  <TrueBrain> but okay, it seems that depending on the map we can offload between 20% and 50% of the drawing time
09:34:27  <TrueBrain> it only does require keeping around sorting data longer than currently :)
09:36:43  <DorpsGek> [OpenTTD/OpenTTD] btzy commented on pull request #8715: Fix: Vehicle list windows did not update when this year's profit changed https://git.io/Jtbaa
09:37:57  <TrueBrain> LordAro: so the first parameter should have a & is what he is saying?
09:39:09  <LordAro> i'm really not sure
09:39:20  <LordAro> i can't find any change that makes any difference to the asm output, anyway
09:39:33  <TrueBrain> indeed
09:39:56  <TrueBrain> it is optimized to fully use registers
09:40:28  <LordAro> again, might be a consequence of the minimal code, but i don't think so
09:44:14  <TrueBrain> gcc is even worse
09:44:24  <TrueBrain> it just makes GetDisplayProfitThisYear2 jump to GetDisplayProfitThisYear
09:44:27  <TrueBrain> like IN YOUR FACE
09:44:28  <TrueBrain> :P
09:44:52  <LordAro> haha
09:45:14  <TrueBrain> so somehow it picks up the Money class is a primitive, I guess
09:45:26  <TrueBrain> well, more likely, that it fits in a register
09:45:36  <TrueBrain> after that, all this std::move() becomes irrelevant
09:46:32  *** gregdek has quit IRC
09:46:33  *** gregdek has joined #openttd
09:52:58  *** Exec has quit IRC
09:53:00  *** Exec has joined #openttd
09:58:16  *** dvim has quit IRC
09:58:27  *** dvim has joined #openttd
10:01:45  <TrueBrain> okay, so there really is a lot to gain for big games if we can split ViewportDoDraw() in a part that influences the game state and a part that does not
10:01:52  <TrueBrain> about 50% on most games is spend in either one
10:02:21  <TrueBrain> the heavy NewGRF games, like one with Timberwolf's, are less impacted (as most time is spent in parts that requires a lock on the game-state)
10:03:04  *** robert[m]2 has quit IRC
10:03:07  *** robert[m]2 has joined #openttd
10:03:12  <TrueBrain> but it will cost more memory, mostly vectors
10:05:00  <TrueBrain> is there an easy way to get the memory size of a vector, hmm
10:05:02  <TrueBrain> or a class
10:11:15  <LordAro> sizeof()
10:11:15  <LordAro> :p
10:12:29  <TrueBrain> yeah
10:12:33  <TrueBrain> I was hoping for something more clever :D
10:13:19  <TrueBrain> okay, it is about 500KB of data, the vectors combined
10:13:34  <TrueBrain> well, 50KB in normal zoom
10:13:42  <TrueBrain> 750KB in max zoomout
10:14:04  <TrueBrain> and 3MB in extreme games like ProGame5
10:14:51  <TrueBrain> 15MB if I go 2k
10:17:18  <peter1138> Are we still trying to keep within the CPU memory cache? :p
10:17:23  <peter1138> FUCK
10:17:25  <TrueBrain> so yeah, in high resolutions with huge games, memory goes up from 1KB to 15MB in that way
10:17:28  <peter1138> Is OpenTTD still trying to keep within the CPU memory cache? :p
10:17:53  <TrueBrain> I can imagine for 4k resolution this goes up to 60MB
10:17:57  <peter1138> Thats only 15000%, nothing to worry about.
10:18:01  <TrueBrain> and for HighRes 4k to .. euh .. 120MB
10:19:24  <TrueBrain> currently the game uses 220MB for me, so on my resolution it would go from 220MB to 235MB
10:19:32  <TrueBrain> I am curious how much memory a 4k resolution takes
10:19:56  <_dp_> what data are you looking at? I don't think gui memory usage should be affected by screen resolution
10:20:03  <_dp_> except for viewport drawing ofc
10:20:29  <TrueBrain> ... so yes, it is influenced by it ;)
10:22:21  <LordAro> TrueBrain: how about with something like zbase?
10:22:28  <TrueBrain> owh, do I dare ....
10:22:30  <TrueBrain> good question :D
10:24:08  <TrueBrain> owh, wait, it currently uses 40KB, not 1KB .. so from 40KB to 15MB for ProGame5 zoomed out :P
10:25:11  <TrueBrain> zbase consumes 400MB of RAM for me
10:25:17  <TrueBrain> so an increase of 15MB is even less there :P
10:25:27  <TrueBrain> either way, I don't like this solution as it scales really poorly with resolution :D
10:26:07  <_dp_> to draw the screen you basically just need some strings and a list of sprites
10:26:10  <_dp_> can't take that much
10:26:27  *** gelignite has joined #openttd
10:26:28  <TrueBrain> I was just watching a youtube movie about armchair developers
10:27:13  <andythenorth> were you in an armchair?
10:28:38  <TrueBrain> its funny to see how overly optimized OpenTTD's drawing is :)
10:28:48  <TrueBrain> clearly build in a time where people still cared about that stuff
10:30:20  <TrueBrain> okay, the current code can already burst to the same values, when the full screen is redrawn in a single iteration
10:30:43  <TrueBrain> that basically only happens during zoom-out
10:31:01  <TrueBrain> possibly also the reason that is so sluggish, as the sorter will have a hard time sorting that all at once
10:31:55  <peter1138> Yeah, see, if there was a *proper* OpenGL driver... *runs away very fast*
10:33:15  <_dp_> hey, my sorter is quite fast :p
10:37:49  *** rptr has quit IRC
10:49:44  *** virtualrandomnumber has joined #openttd
10:49:55  *** virtualrandomnumber has quit IRC
10:57:39  <TrueBrain> okay, made a very crude implementation of splitting the Viewport code
10:57:52  <TrueBrain> this game that was running on 13 simulation fps, is now doing 15
10:58:09  <TrueBrain> with 10ms for drawing now done in parallel with the gametick
11:00:24  <TrueBrain> funny, simulation can now run at, for this game, 20fps, but drawing is down to 5fps
11:00:32  <TrueBrain> simply because I am fully zoomed out and it cannot keep up :D
11:00:37  <TrueBrain> (instead of them both being at 5fps)
11:06:16  *** reldred has quit IRC
11:06:27  *** reldred has joined #openttd
11:07:13  <TrueBrain> hitting 9999 simulation fps on a fully zoomed-out map is weird
11:07:50  <TrueBrain> https://pasteboard.co/JQ8bmwZ.png
11:09:55  <TrueBrain> owh, and this is a debug build
11:09:57  <TrueBrain> lol
11:11:30  <TrueBrain> and frosch123 said it couldn't be done in a week, pffft
11:11:41  <TrueBrain> well, it breaks sometimes, as too many things assume what is happening shouldn't happen
11:11:47  <TrueBrain> mostly, fonts sometimes crash the game :P
11:12:13  <TrueBrain> https://github.com/OpenTTD/OpenTTD/compare/master...TrueBrain:split-viewport-drawing <- last patch in the series .. it is ugly code :D
11:13:52  <TrueBrain> it can be done smarter by using a pool of threads
11:14:02  <TrueBrain> and assigning each a "vd" to be drawn
11:14:25  <TrueBrain> but that would make OpenTTD really become multithreaded, I am not ready for that
11:15:06  <peter1138> So the graphics framerate is dropped because so much is updating, but the sim can still run flat out...
11:15:14  <TrueBrain> exactly
11:15:27  <TrueBrain> so what is on your screen has 50% less influence on the simulation speed
11:15:34  <peter1138> That is ... interesting.
11:16:35  <TrueBrain> slowly ... removing what needs a lock on the game-state, and what can run without, basically :)
11:17:05  *** Samu has joined #openttd
11:17:15  <peter1138> What's the frame rate if it's not running flat out?
11:17:21  <TrueBrain> but the more proper solution here is moving a lot of these Viewport _vd functions into a single class, so it is more clear it is isolated
11:17:41  <TrueBrain> 33.33fps for simulation and 60fps for graphics
11:18:17  <peter1138> Yeah, ideally you'd throttle the sim rate to not run so fast as to cause the framerate to drop. Heurrghghghghgghistics..
11:18:53  <TrueBrain> the balancing act there is a matter of opinion, I found out
11:19:00  <TrueBrain> some want the simulation to be as quick, at the cost of drawing
11:19:11  <TrueBrain> others want drawing to hit 60fps as much as possible, at the cost of simulation
11:19:30  <Eddi|zuHause> speaking of framerate, i was looking for the setting yesterday, but couldn't find it
11:19:31  <peter1138> I think that's reasonable if you're hitting framerate and simrate limits < 1x.
11:19:41  <TrueBrain> Eddi|zuHause: its an ini-file-only setting
11:20:04  <Eddi|zuHause> i would have expected it in game setting, where resolution and stuff is
11:20:27  <TrueBrain> currently, I think nobody should change it unless they know what they are doing :)
11:22:18  <peter1138> Can Windows explorer mix directories and files together? I should know this...
11:22:33  <TrueBrain> yes, it can
11:23:00  <TrueBrain> hmm .. if I put this work in another thread, it can already start with sorting and drawing while the next rect on the screen is started .. making this even faster
11:23:50  <Eddi|zuHause> well, i don't think i know what's happening, but my fps seemed to be 30, not 60
11:24:15  <TrueBrain> did you change your openttd.cfg to test a while back? :D
11:24:30  <Eddi|zuHause> i don't think so
11:26:33  <Eddi|zuHause> "display_hz = 0" <-- this?
11:26:41  <TrueBrain> no, that is an old setting and removed now
11:26:45  <TrueBrain> refresh_rate is what you are looking for
11:28:02  <Eddi|zuHause> yeah, that's set to 30. but i don't think i ever touched that
11:28:17  <TrueBrain> well, the game sets it to 60 by default shrug
11:28:56  *** jgx has quit IRC
11:29:18  *** jgx has joined #openttd
11:41:26  <Eddi|zuHause> i don't think i'll find out when this was set. maybe some earlier test version set it to 30 by default
11:46:12  <LordAro> it would be nice to add it to the options window, i think
12:11:47  <dihedral> hey hey
12:19:48  <TrueBrain> LordAro: my main issue is, and the reason for not being in favour of adding it to the options window, is that increasing the refresh_rate has an impact on simulation speed, something you wouldn't expect (and most games don't have :P)
12:19:59  <TrueBrain> so I think we should first make sure that the impact of that is far less than it currently is
12:20:05  <TrueBrain> before adding it for anyone to change :)
12:20:30  <TrueBrain> it was why I originally wanted to keep it a const, but I can understand that some people will have to change it to 30 or so in order to play the games they can play now
12:20:33  <TrueBrain> for the same reason :)
12:20:54  <TrueBrain> but a dropdown with <30> <60> as only values, would result in more questions (and complaints) of users than simply not mentioning it
12:21:03  <TrueBrain> at least, that was my reasoning for doing it this way :)
12:21:52  <TrueBrain> sometimes you have to protect the user from himself, basically :D
12:22:48  <Eddi|zuHause> well, you can try reducing fps when simulation speed dips below x1?
12:30:43  <Eddi|zuHause> so the behaviour would be ideal conditions: simulation = x1, fps = user setting, under stress: simulation = x1, fps = reduced, and under high stress (fps < 33), also drop simulation speed, so simulation ticks and refresh rate are roughly equal
12:45:58  <TrueBrain> there are many things we can do to improve a lot of this, but it requires time and testing :)
12:46:24  <TrueBrain> and as I mentioned earlier, it also turns out people expect different things from the game when there is a resource shortage
12:46:34  <TrueBrain> so this takes more time to figure out
12:47:52  <TrueBrain> owh boy, we have 1 place in the code that does misused _realtime_tick
12:48:04  <TrueBrain> it uses it to check if there has been a draw-frame
12:49:21  <FLHerne> dihedral: Ahey! Seems this is not-seen-for-ages people week.
12:49:38  <dihedral> :-)
12:50:00  <TrueBrain> FLHerne: be careful, before you now it ludde drops by :)
12:50:05  <TrueBrain> but indeed: hi :)
12:50:31  <dihedral> the nice thing must be, that one cannot stay away for too long ;-)
12:50:46  <TrueBrain> some try more than others :P
12:51:01  <dihedral> i recently sent a message to darkvater - wow was he surprised :-D
12:51:29  <dihedral> "recently" .... in relation to my "recent" visits to this channel may not habe been all that recent
12:51:42  <TrueBrain> so 4 years ago, you mean :P
12:51:56  <dihedral> nah - i believe it was late summer 2020
12:52:05  <TrueBrain> recent enough :)
12:53:14  <dihedral> uh - december even
12:53:29  <dihedral> (and summer 2020) :-P
12:53:41  <TrueBrain> stalker :P
12:54:09  <dihedral> na - that's what I concider "keeping in touch"
12:56:37  <TrueBrain> tomato tomato :P
12:56:42  <TrueBrain> what have you been up to lately?
12:58:19  <dihedral> busy - quote busy
12:58:44  <dihedral> have a company in karlsruhe and am working on my ppl
12:59:37  <dihedral> how about yourself?
13:02:02  <DorpsGek> [OpenTTD/OpenTTD] TrueBrain updated pull request #8747: Codechange: replace _realtime_tick with std::chrono https://git.io/JtFpt
13:02:13  <TrueBrain> was bored with my last job, quit, and have been working on OpenTTD instead :P
13:02:26  <dihedral> financed how? :-P
13:02:36  <TrueBrain> unrelated: w00p! No more _realtime_tick, no more race-condition with my other PR \o/
13:02:37  <dihedral> i did notice that you have been busy with openttd alot
13:02:42  <TrueBrain> dihedral: a bank accont, ofc :)
13:02:53  <dihedral> Ha! that's what they all say :-P
13:02:59  <LordAro> TrueBrain: nice
13:03:28  <TrueBrain> cannot believe it was defined in debug.h ..
13:05:17  <LordAro> TrueBrain: newgrf_debug_sprite_picker stuff feels real hacky
13:05:25  <LordAro> oh, it was in window.cpp already
13:05:37  <LordAro> no more hacky than it already was then :p
13:05:38  <TrueBrain> that doesn't change it is a hack :)
13:05:40  <TrueBrain> :D
13:05:51  <TrueBrain> it is SLIGHTLY better, as it no longer abuses a variable for something unrelated
13:05:59  <LordAro> mm
13:06:15  *** supermop_Home has joined #openttd
13:06:45  <LordAro> i suppose i could argue that creating lots of static variables is just creating lots of "global" variables rather than just using a single one
13:06:47  <supermop_Home> hello
13:06:48  <LordAro> but eh
13:07:34  <TrueBrain> LordAro: did I add any static variables?
13:08:03  <TrueBrain> I should have only removed a few
13:08:26  <LordAro> oh, i can't read properly
13:08:37  <LordAro> yes, on the whole you've removed a couple
13:08:46  <TrueBrain> pfew :D Otherwise I screwed up :P
13:09:03  <dihedral> @TrueBrain, since when?
13:09:22  <dihedral> i mean - when did you quit?
13:09:23  <TrueBrain> that I screwed up? According to the forums, a few weeks ago :P
13:09:25  <TrueBrain> :D
13:09:28  <TrueBrain> somewhere last year
13:09:41  <dihedral> and what did you not like about your job?
13:09:46  <TrueBrain> the working part
13:09:49  <dihedral> ...
13:09:52  <TrueBrain> :D
13:09:54  <dihedral> har har
13:09:56  <LordAro> lol
13:10:01  <TrueBrain> I am not even kidding :D
13:10:32  <dihedral> yeah, can understand - but more specifically? :-P
13:10:36  <andythenorth> so is Belguas visiting next?
13:10:44  <andythenorth> Belugas *
13:10:48  <LordAro> @seen Bjarni
13:10:48  <DorpsGek> LordAro: Bjarni was last seen in #openttd 9 years, 20 weeks, 5 days, 12 hours, 51 minutes, and 41 seconds ago: <Bjarni> heh
13:10:59  <TrueBrain> @seen ludde
13:11:00  <DorpsGek> TrueBrain: I have not seen ludde.
13:11:01  <dihedral> uh - that was the guz
13:11:10  <TrueBrain> bad DorpsGek , bad bad DorpsGek
13:11:11  <TrueBrain> @seen Yexo
13:11:12  <DorpsGek> TrueBrain: I have not seen Yexo.
13:11:12  <andythenorth> TrueBrain we could just patreon you :P
13:11:16  <TrueBrain> he keeps forgetting people
13:11:26  <TrueBrain> andythenorth: meh; means people have expectations
13:11:30  <andythenorth> yes
13:11:33  <andythenorth> I would expect reports
13:11:35  <LordAro> :D
13:11:40  <andythenorth> and we would want to review you quarterly
13:11:44  <dihedral> patreon can be unconditional
13:11:48  <andythenorth> and we'd have to have a HR policy
13:11:56  <dihedral> as it is voluntary anyway
13:12:07  <andythenorth> TrueBrain we could construct it so you have to review yourself!
13:12:13  <andythenorth> on livestream
13:12:13  <dihedral> andythenorth, you want to patreon TrueBrain not openttd
13:12:16  <TrueBrain> https://gist.github.com/TrueBrain/25a5a6f98f761ef5be27b493b1b6ada5 there you go andythenorth
13:12:26  <andythenorth> yes
13:12:31  <andythenorth> then you can sign it off, yourself
13:12:40  * andythenorth has one sponsor, andythenorth 
13:12:54  <andythenorth> I just rely on doing work to fund it
13:13:09  <andythenorth> interestingly, getting paid for newgrf commissions is now a thing
13:13:13  <andythenorth> not for me, but for other people
13:13:16  <TrueBrain> I am pretty sure only a very few people can make a living out of Open Source work via patreon :)
13:13:34  <TrueBrain> haha, who is doing that andythenorth ? :D That is pretty funny :)
13:15:04  <dihedral> TrueBrain, Next report could be a link to itself :-D
13:15:28  <TrueBrain> dihedral: fixed
13:15:48  <dihedral> perfect
13:16:45  <TrueBrain> okay, with the _realtime_tick removed, I can thread drawing and gameloop for all platforms safely
13:16:47  <TrueBrain> so that is nice
13:17:18  <TrueBrain> less platform-specific code \o/
13:17:49  <TrueBrain> next would be to refactor viewport.cpp into more class-based system
13:21:11  <TrueBrain> ugh, still avoiding working on the GameSpeed window ... lets pick an issue that has been unresolved for a while ..
13:32:13  <supermop_Home> oof all that messing around to get pretty vehicle cargoes and i forgot to add plant fibers
13:32:27  <TrueBrain> tssk
13:33:36  <FLHerne> Yexo reappeared pretty recently for a while?
13:35:07  *** dvim has quit IRC
13:35:07  *** gregdek has quit IRC
13:35:08  *** greeter has quit IRC
13:35:08  *** Maarten has quit IRC
13:35:08  *** crem2 has quit IRC
13:35:08  *** y2kboy23 has quit IRC
13:35:08  *** WormnestAndroid has quit IRC
13:35:08  *** Extrems has quit IRC
13:35:08  *** Speeder has quit IRC
13:35:08  *** EmeraldSnorlax[m] has quit IRC
13:35:08  *** Ttech has quit IRC
13:35:08  *** fnutt has quit IRC
13:35:08  *** innocenat_ has quit IRC
13:35:08  *** mindlesstux has quit IRC
13:35:08  *** k-man has quit IRC
13:35:08  *** Smedles has quit IRC
13:35:08  *** Vadtec has quit IRC
13:35:08  *** dwfreed has quit IRC
13:35:08  *** mikegrb has quit IRC
13:35:13  *** mikegrb has joined #openttd
13:35:19  <TrueBrain> yeah, and DorpsGek didn't track that, which annoys me :P
13:35:38  <andythenorth> I suspect yexo got another job :P
13:35:42  <andythenorth> oof
13:35:44  *** k-man has joined #openttd
13:36:01  <LordAro> did you look to see if the dorpsgek issue was an irc bot issue, or a database issue?
13:36:09  *** dvim has joined #openttd
13:36:09  *** gregdek has joined #openttd
13:36:09  *** greeter has joined #openttd
13:36:09  *** crem2 has joined #openttd
13:36:09  *** y2kboy23 has joined #openttd
13:36:09  *** WormnestAndroid has joined #openttd
13:36:09  *** Extrems has joined #openttd
13:36:09  *** Speeder has joined #openttd
13:36:09  *** EmeraldSnorlax[m] has joined #openttd
13:36:09  *** fnutt has joined #openttd
13:36:09  *** innocenat_ has joined #openttd
13:36:09  *** mindlesstux has joined #openttd
13:36:09  *** Smedles has joined #openttd
13:36:09  *** Vadtec has joined #openttd
13:36:09  *** dwfreed has joined #openttd
13:36:34  <TrueBrain> I did not
13:36:40  <TrueBrain> as I did not care enough, I have to admit :)
13:36:49  <LordAro> fair enough :p
13:37:09  <TrueBrain> it is somewhere between too-lazy and if-I-find-the-problem-it-means-I-have-to-fix-it :P
13:37:12  <TrueBrain> btw: certbot!
13:37:24  <LordAro> aaah
13:38:17  *** Ttech has joined #openttd
13:39:08  *** Maarten has joined #openttd
13:44:02  <TrueBrain> I have this ticket, that is invalid, because it talks about a solution, not the real problem
13:44:09  <TrueBrain> but the real problem, might still be a thing
13:44:14  <TrueBrain> tempted to close the ticket
13:44:19  <TrueBrain> as I am too lazy to check if the real problem is a thing :D
13:44:26  <TrueBrain> (or also, possibly, invalid)
14:00:02  *** nielsm has joined #openttd
14:04:11  * andythenorth looks on reddit
14:05:02  <andythenorth> WHY
14:05:04  <andythenorth> WHY
14:05:05  <andythenorth> WHY
14:05:05  <andythenorth> https://www.reddit.com/r/openttd/comments/lstczw/why/
14:05:42  <TrueBrain> best reply evah
14:05:48  <TrueBrain> (the one reply)
14:05:51  <supermop_Home> those ginzu a4s giving me some real 1995 vibes
14:06:03  <andythenorth> diesel A4?
14:06:07  <andythenorth> A4 with pans?
14:06:11  <andythenorth> hmm
14:09:33  <DorpsGek> [OpenTTD/OpenTTD] EratoNysiad commented on pull request #8604: Feature: allow limiting zoom level of NewGRF-provided sprites https://git.io/Jtb7A
14:12:11  *** gelignite has quit IRC
14:18:15  <TrueBrain> so if 5 commands use an ugly hack 1 way, and 1 command an ugly hack the other
14:18:24  <TrueBrain> but the 1 command is slightly less ugly
14:18:28  <TrueBrain> should I change all 5, or the 1 :P
14:18:30  <TrueBrain> :D
14:24:51  *** Samu has quit IRC
14:25:49  <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on issue #8045: AI estimated costs are 0 for commands that fail while returning a cost https://git.io/JvSaL
14:26:40  <DorpsGek> [OpenTTD/OpenTTD] TrueBrain opened pull request #8748: Fix 02e770ff: allow estimating CloneVehicle if short on money https://git.io/Jtbdq
14:26:41  <TrueBrain> 14 year old bug ^^ :D
14:27:29  <TrueBrain> and a nice example of: the issue starter might say A, but means B, and tried to solve it in a way that breaks more than it solves :D
14:29:35  <Eddi|zuHause> that's a typical samu
14:29:47  <TrueBrain> "reading between the lines" skill leveled up \o/
14:29:50  <TrueBrain> Eddi|zuHause: yup
14:30:20  <TrueBrain> changing something for ALL commands while ONE command was at fault :D
14:30:50  <Eddi|zuHause> and then he gets mad if someone proposes an alternate solution
14:32:56  <TrueBrain> I spend 50% of the time working on this finding out if and which commands return a partial cost when erroring out :P
14:33:36  <Eddi|zuHause> command return values are a fractal on their own :p
14:39:11  <_dp_> vehicle cloning may not be the only command to behave that way...
14:39:49  <TrueBrain> I checked; it is
14:39:56  <TrueBrain> I wrote that in my comment even
14:43:48  <DorpsGek> [OpenTTD/OpenTTD] 2TallTyler commented on pull request #8604: Feature: allow limiting zoom level of NewGRF-provided sprites https://git.io/JtbFB
14:47:09  <_dp_> well, mb not for money but terra limit seems to affect estimations as well
14:47:15  <_dp_> which is kinda questionable
14:49:07  <_dp_> pretty much a bug I'd say even as it's not that different from money
14:52:20  <LordAro> TrueBrain: "... hopefully for good this time"
14:53:00  <TrueBrain> LordAro: sorry, context? :D
14:53:12  <nielsm> I really want to get JGR's route restriction signals and realistic braking in
14:53:17  <LordAro> 02e770f commit message
14:53:32  <nielsm> haven't tried them but in principle they sound really good
14:53:34  <TrueBrain> LordAro: ghehe, yeah :D
14:54:29  <DorpsGek> [OpenTTD/OpenTTD] LordAro commented on pull request #8748: Fix 02e770ff: allow estimating CloneVehicle if short on money https://git.io/JtbFp
14:55:08  <_dp_> oh, and terraform estimations are wrong anyway so it's not really unaffected there :p
14:57:05  <TrueBrain> LordAro: "AIVehicle::CloneVehicle now correctly returns estimate when short on cash" <- good enough?
14:57:13  <LordAro> yeah
14:57:26  <_dp_> nielsm, doesn't realistic braking only work with pbs?
14:57:42  <DorpsGek> [OpenTTD/OpenTTD] TrueBrain updated pull request #8748: Fix 02e770ff: allow estimating CloneVehicle if short on money https://git.io/Jtbdq
14:57:50  <TrueBrain> tnx LordAro :)
14:58:02  <nielsm> _dp_ yes, I don't see any problem with that
14:59:44  <_dp_> well, I guess that would be another option never to enable in a goal game :p
15:08:22  <_dp_> interesting way to implement realistic braking would be to have fake head of the train always a braking distance ahead of the real one
15:08:33  <_dp_> but probably would be quite hard to do in openttd
15:09:30  <LordAro> how very "shadows"
15:10:30  <TrueBrain> OpenTTD wouldn't be OpenTTD if trains cannot stop instantly
15:10:38  <TrueBrain> makes me giggle with Factorio too
15:10:40  <TrueBrain> so much fun :D
15:11:27  <Eddi|zuHause> _dp_: the main problem with that idea is that braking distance changes with speed, so you need this pseudo-vehicle to be of variable length
15:11:54  <_dp_> Eddi|zuHause, I think variable length is the least problem here :p
15:12:27  <Eddi|zuHause> _dp_: what i would rather do is check the length of the path reservation
15:12:47  <Eddi|zuHause> _dp_: if less than braking distance, try extending it
15:12:54  <Eddi|zuHause> _dp_: if extending fails, brake
15:14:36  <_dp_> well, reservation isn't that different from a fake head anyway
15:17:35  <_dp_> btw, I don't think #6503 was ever fixed :p
15:20:04  <LordAro> clearly not
15:22:19  <DorpsGek> [OpenTTD/OpenTTD] LordAro approved pull request #8748: Fix 02e770ff: allow estimating CloneVehicle if short on money https://git.io/JtbNX
15:22:47  <DorpsGek> [OpenTTD/OpenTTD] LordAro reopened issue #6503: Abnormalities in train subtile coordinates when reversing at the end of line https://git.io/fhZIc
15:22:49  <LordAro> mwhaha
15:23:01  <DorpsGek> [OpenTTD/OpenTTD] TrueBrain closed issue #8045: AI estimated costs are 0 for commands that fail while returning a cost https://git.io/JvSaL
15:23:04  <DorpsGek> [OpenTTD/OpenTTD] TrueBrain merged pull request #8748: Fix 02e770ff: allow estimating CloneVehicle if short on money https://git.io/Jtbdq
15:23:47  <Eddi|zuHause> LordAro: so you're volunteering? :p
15:24:25  <LordAro> nope!
15:25:52  <_dp_> and I don't think I ever reported similar stuff for road vehicles because it's such a mess that I couldn't figure out a logical way to plot subtile movement for them xD
15:26:39  <Eddi|zuHause> movement is also a fractal...
15:26:55  <Eddi|zuHause> we have this length and speed distortion on diagonals
15:27:06  <Eddi|zuHause> we have jumpy movement on x4 zoom
15:27:46  <_dp_> diagonals aren't that much of a problem for movement
15:28:30  <_dp_> but sorting out subtile stuff will help the 4x cause for sure
15:28:39  <Eddi|zuHause> https://github.com/OpenTTD/OpenTTD/issues/1063
15:30:32  <_dp_> Eddi|zuHause, yeah, but it's more of a ui/gamplay issue, movement code doesn't really care about it
15:32:09  <_dp_> it's unreasonable to expect smooth train graphics when rails do 45 deg corner :p
15:32:32  <Eddi|zuHause> "smooth" and "inconsistent" are very different
15:35:11  <supermop_Home> maybe instead of worrying about bounding boxes, it would be best to just always force rear catenary sprites to be drawn before any vehicle etc?
15:35:26  <_dp_> that inconsistency comes from the grid-based nature of the game
15:37:03  <_dp_> you can't change graphics without changing speed but changing speed raises balance issues
15:38:43  <Eddi|zuHause> _dp_: i disagree. you can not *completely* fix it with integers, but you could make it more consistent. the main problem (as discussed in the old ticket) is that this breaks all kinds of backwards compatibility with graphics
15:45:30  <_dp_> integers are kinda irrelevant here but, yeah, graphics compatibility is a significant issue here as well
15:45:49  <_dp_> but what I'm saying that inconsistency didn't come out of nowhere, it has a reason to exist
15:46:18  <_dp_> which cased graphics to be made the way they are
15:46:27  <Eddi|zuHause> the inconsistency comes exactly from a quick-and-dirty integer implementation that was never properly fixed
15:47:01  <Eddi|zuHause> like rounding 1/sqrt(2) to 0.5
15:47:11  <Eddi|zuHause> or to 1
15:47:23  <_dp_> but game isn't rounding it to 0 or 1
15:47:27  <Eddi|zuHause> in different parts of the code
15:47:28  <_dp_> it's rounding to 128 and 196
15:47:41  <_dp_> which could easily be more precise
15:48:02  <_dp_> *192
15:49:04  <Eddi|zuHause> the main problem isn't how exact the rounding is. the problem is that it's rounded differently for speed and length
15:49:42  <_dp_> if it was different trains would have gaps :p
15:49:56  <Eddi|zuHause> so on making a turn, the train changes length, so for a while, front and tail move at different speeds
15:50:33  <_dp_> yeah, it's not your typical euclidean space :P
15:53:13  *** Progman has joined #openttd
15:58:09  <_dp_> but I think it is a metric space nonetheless
15:58:58  *** Wuzzy has joined #openttd
15:59:43  <_dp_> so in a way it doesn't change length and speed, it just uses non-euclidean length
16:00:32  <Wuzzy> so is there any way i can help with work on <https://github.com/OpenTTD/OpenSFX/pull/19> now? i am a bit overwhelmed atm
16:00:35  <Eddi|zuHause> i don't think you understood what i was trying to say
16:07:50  *** Flygon has quit IRC
16:10:04  <_dp_> neither do you :p
16:10:38  <_dp_> basically, if game used euclidean space there would be no point in ever using diagonal tracks
16:11:06  <Eddi|zuHause> what?
16:11:11  <Eddi|zuHause> you're making no sense
16:16:57  *** gelignite has joined #openttd
16:22:46  <_dp_> euclidean distance makes all directions equal. but on a grid building diagonal stuff is much harder
16:23:00  <_dp_> there would be no point going diagonal if there was no benefit to it
16:23:38  <_dp_> so game has to compromise
16:28:54  *** Samu has joined #openttd
16:32:17  <Wolf01> _dp_: we should switch to hex grid
16:33:39  <_dp_> if anything it would be better to have no grid at all
16:33:51  <_dp_> but that would be a completely different game
16:34:03  <_dp_> like nimby for example
17:04:01  *** supermop_Home has quit IRC
17:06:11  *** Wormnest has joined #openttd
17:30:27  *** supermop_Home has joined #openttd
17:41:23  *** rptr has joined #openttd
17:59:15  *** Wormnest has quit IRC
17:59:15  *** supermop_Home has quit IRC
17:59:15  *** Ttech has quit IRC
17:59:15  *** Vadtec has quit IRC
17:59:15  *** Smedles has quit IRC
17:59:15  *** fnutt has quit IRC
17:59:15  *** EmeraldSnorlax[m] has quit IRC
17:59:15  *** Speeder has quit IRC
17:59:15  *** WormnestAndroid has quit IRC
17:59:15  *** y2kboy23 has quit IRC
17:59:15  *** crem2 has quit IRC
17:59:15  *** gregdek has quit IRC
17:59:15  *** mindlesstux has quit IRC
17:59:15  *** Extrems has quit IRC
17:59:15  *** dwfreed has quit IRC
17:59:15  *** innocenat_ has quit IRC
17:59:15  *** greeter has quit IRC
17:59:15  *** dvim has quit IRC
17:59:15  *** Maarten has quit IRC
17:59:15  *** mikegrb has quit IRC
17:59:15  *** robert[m]2 has quit IRC
17:59:15  *** roadt_ has quit IRC
17:59:15  *** APTX has quit IRC
17:59:15  *** daspork has quit IRC
17:59:15  *** k-man has quit IRC
17:59:15  *** jgx has quit IRC
17:59:15  *** reldred has quit IRC
17:59:15  *** twpol has quit IRC
17:59:15  *** colde has quit IRC
17:59:15  *** snail_UES_ has quit IRC
17:59:15  *** Gustavo6046 has quit IRC
17:59:16  *** heffer has quit IRC
17:59:16  *** muffindrake has quit IRC
17:59:16  *** murr4y has quit IRC
17:59:16  *** nnyby has quit IRC
17:59:16  *** jinks has quit IRC
17:59:16  *** DorpsGek has quit IRC
17:59:16  *** ST2 has quit IRC
17:59:22  *** supermop_Home has joined #openttd
17:59:22  *** Wormnest has joined #openttd
17:59:22  *** Maarten has joined #openttd
17:59:22  *** Ttech has joined #openttd
17:59:22  *** dwfreed has joined #openttd
17:59:22  *** Vadtec has joined #openttd
17:59:22  *** Smedles has joined #openttd
17:59:22  *** mindlesstux has joined #openttd
17:59:22  *** innocenat_ has joined #openttd
17:59:22  *** fnutt has joined #openttd
17:59:22  *** EmeraldSnorlax[m] has joined #openttd
17:59:22  *** Speeder has joined #openttd
17:59:22  *** Extrems has joined #openttd
17:59:22  *** WormnestAndroid has joined #openttd
17:59:22  *** y2kboy23 has joined #openttd
17:59:22  *** crem2 has joined #openttd
17:59:22  *** greeter has joined #openttd
17:59:22  *** gregdek has joined #openttd
17:59:22  *** dvim has joined #openttd
17:59:22  *** k-man has joined #openttd
17:59:22  *** mikegrb has joined #openttd
17:59:22  *** jgx has joined #openttd
17:59:22  *** reldred has joined #openttd
17:59:22  *** robert[m]2 has joined #openttd
17:59:22  *** twpol has joined #openttd
17:59:22  *** murr4y has joined #openttd
17:59:22  *** nnyby has joined #openttd
17:59:22  *** roadt_ has joined #openttd
17:59:22  *** jinks has joined #openttd
17:59:22  *** APTX has joined #openttd
17:59:22  *** daspork has joined #openttd
17:59:22  *** ST2 has joined #openttd
17:59:22  *** DorpsGek has joined #openttd
17:59:22  *** muffindrake has joined #openttd
17:59:22  *** heffer has joined #openttd
17:59:22  *** Gustavo6046 has joined #openttd
17:59:22  *** snail_UES_ has joined #openttd
17:59:22  *** colde has joined #openttd
17:59:22  *** synthon.oftc.net sets mode: +ov DorpsGek DorpsGek
18:01:53  *** frosch123 has joined #openttd
18:10:25  *** Wuzzy2 has joined #openttd
18:11:47  *** Wuzzy has quit IRC
18:22:20  *** dvim has quit IRC
18:22:20  *** gregdek has quit IRC
18:22:20  *** greeter has quit IRC
18:22:20  *** crem2 has quit IRC
18:22:20  *** y2kboy23 has quit IRC
18:22:20  *** WormnestAndroid has quit IRC
18:22:20  *** Extrems has quit IRC
18:22:20  *** Speeder has quit IRC
18:22:20  *** EmeraldSnorlax[m] has quit IRC
18:22:20  *** fnutt has quit IRC
18:22:20  *** innocenat_ has quit IRC
18:22:20  *** mindlesstux has quit IRC
18:22:20  *** Smedles has quit IRC
18:22:20  *** Vadtec has quit IRC
18:22:20  *** Wormnest has quit IRC
18:22:20  *** supermop_Home has quit IRC
18:22:20  *** dwfreed has quit IRC
18:22:20  *** Ttech has quit IRC
18:22:35  *** supermop_Home has joined #openttd
18:22:35  *** Wormnest has joined #openttd
18:22:35  *** Ttech has joined #openttd
18:22:35  *** dvim has joined #openttd
18:22:35  *** gregdek has joined #openttd
18:22:35  *** greeter has joined #openttd
18:22:35  *** crem2 has joined #openttd
18:22:35  *** y2kboy23 has joined #openttd
18:22:35  *** WormnestAndroid has joined #openttd
18:22:35  *** Extrems has joined #openttd
18:22:35  *** Speeder has joined #openttd
18:22:35  *** EmeraldSnorlax[m] has joined #openttd
18:22:35  *** fnutt has joined #openttd
18:22:35  *** innocenat_ has joined #openttd
18:22:35  *** mindlesstux has joined #openttd
18:22:35  *** Smedles has joined #openttd
18:22:35  *** Vadtec has joined #openttd
18:22:35  *** dwfreed has joined #openttd
18:26:58  <DorpsGek> [OpenTTD/OpenTTD] michicc merged pull request #8743: Change: Default to a 32bpp blitter. https://git.io/Jtdoj
18:27:27  <DorpsGek> [OpenTTD/OpenTTD] 2TallTyler opened pull request #8749: WIP Feature: Setting to loop the same year forever https://git.io/JtNkN
18:28:46  <andythenorth> :o
18:28:47  <andythenorth> :)
18:32:21  *** Strakaty has joined #openttd
18:37:00  <DorpsGek> [OpenTTD/OpenTTD] James103 commented on pull request #8749: WIP Feature: Setting to loop the same year forever https://git.io/JtNI2
18:38:20  *** robert[m]2 has quit IRC
18:38:20  *** Maarten has quit IRC
18:38:20  *** mikegrb has quit IRC
18:38:50  *** Maarten has joined #openttd
18:38:50  *** mikegrb has joined #openttd
18:38:50  *** robert[m]2 has joined #openttd
18:41:05  *** colde has quit IRC
18:41:05  *** snail_UES_ has quit IRC
18:41:05  *** Gustavo6046 has quit IRC
18:41:05  *** heffer has quit IRC
18:41:05  *** muffindrake has quit IRC
18:41:05  *** jinks has quit IRC
18:41:05  *** nnyby has quit IRC
18:41:05  *** murr4y has quit IRC
18:41:05  *** reldred has quit IRC
18:41:05  *** jgx has quit IRC
18:41:05  *** ST2 has quit IRC
18:41:05  *** twpol has quit IRC
18:41:05  *** DorpsGek has quit IRC
18:41:19  *** jgx has joined #openttd
18:41:19  *** reldred has joined #openttd
18:41:19  *** twpol has joined #openttd
18:41:19  *** colde has joined #openttd
18:41:19  *** snail_UES_ has joined #openttd
18:41:19  *** Gustavo6046 has joined #openttd
18:41:19  *** heffer has joined #openttd
18:41:19  *** muffindrake has joined #openttd
18:41:19  *** murr4y has joined #openttd
18:41:19  *** nnyby has joined #openttd
18:41:19  *** jinks has joined #openttd
18:41:19  *** DorpsGek has joined #openttd
18:41:19  *** ST2 has joined #openttd
18:41:19  *** reflection.oftc.net sets mode: +ov DorpsGek DorpsGek
18:43:33  *** rptr_ has joined #openttd
18:44:10  <frosch123> so did dih*dral succeed in hiring Tru*Brain? that part is unclear from the log
18:44:29  <TrueBrain> was he attempted to do so? Oops .. might have missed that :P
18:45:58  <frosch123> the meetup is some years ago, but back then he had an IT company
18:46:34  <andythenorth> we could all hire each other?
18:46:46  <TrueBrain> pyramid scheme?
18:46:54  <andythenorth> if we can just keep the money moving around on 30 day invoicing, I think it works
18:47:24  <frosch123> we could make some VAT triangle trade
18:47:36  <andythenorth> there is some illegal form of that
18:47:46  <TrueBrain> UK is now out of the EU; not sure if business there still is against 0% VAT now :P
18:47:46  * andythenorth won't be indulging
18:48:15  <TrueBrain> annnyyywwaaayyy, hello to you too frosch123 :)
18:48:38  <frosch123> i read you prepared a challenge for me to break something
18:48:49  <TrueBrain> breaking it is easy, honestly
18:49:09  <TrueBrain> as I missed at least one place that changes _vd out of the scope I thought it had
18:49:17  <TrueBrain> but what do you think about such solution?
18:49:27  *** rptr has quit IRC
18:49:28  <TrueBrain> a thread should constantly be reading _vd_queue of course
18:49:36  <frosch123> i did not look at the code yet
18:50:05  <TrueBrain> its funny how the NewGRFs you have loaded make such change from worth the time to barely useful :P
18:50:48  <TrueBrain> but viewport.cpp is such ugly code :P It needs some refactoring before such patch would be doable to get accepted :)
18:51:01  <TrueBrain> mainly _vd's access needs to be more .. in contorl
18:51:26  <frosch123> well, when you are done with thread-pooling the viewport drawing, i can hint you how to thread-pool the newgrf stuff :p
18:51:48  <frosch123> i will set the estimate for multi-thread newgrf to two week, just so you take the challenge bait :)
18:52:27  <TrueBrain> that would be safe to do so? Now that would make it interesting :)
18:52:27  <TrueBrain> frosch123: hahaha :D
18:52:33  <TrueBrain> well, this was a few hours work instead of a week, so ... :P
18:52:45  <TrueBrain> nah, a real PR would take a week or so
18:53:27  <frosch123> resolving vehicle sprites in paralell is only limited by using too many global variables
18:53:43  <frosch123> but they access the gamestate read-only
18:53:47  <TrueBrain> OpenTTD does have that issues, yes ... too many global state :D
18:53:58  <TrueBrain> cool :D
18:54:17  <frosch123> about the game-state itself. you can shuffle the vehicle ticks around, so they do the acceleration-computation in paralle, and then advance sequentially
18:54:19  <TrueBrain> I found another optimization that would remove some of the stutter when doing things quickly ... never draw the whole screen all at once :P :P
18:54:46  <TrueBrain> normally the buffer to sort is like 10k items
18:54:50  <TrueBrain> but in peek it is 8M items
18:54:54  <TrueBrain> for some reason, that takes longer
18:55:53  <TrueBrain> but the dirty-rect in the viewport is crazy complex
18:55:57  <TrueBrain> a line-scanner of some sorts
18:56:04  <TrueBrain> couldn't figure it out within 10 minutes, so just ignored it
18:56:04  <frosch123> yes, the spritesorter is quadratic or cubic
19:01:40  <DorpsGek> [OpenTTD/OpenTTD] DorpsGek pushed 1 commits to master https://git.io/JtNtR
19:01:41  <DorpsGek>   - Update: Translations from eints (by translators)
19:02:32  <TrueBrain> also a funny thing, viewport locations are updated after they are drawn
19:02:34  <TrueBrain> which feels funky
19:02:55  <TrueBrain> (also means they always lag behind 1 tick)
19:04:14  *** Alkel_U3 has quit IRC
19:04:28  *** Alkel_U3 has joined #openttd
19:07:22  <_dp_> sprite sorter is not that quadratic anymore, it's nearing qsort territory now
19:07:23  <peter1138> Does game time backwards yet?
19:08:02  <_dp_> you sure can make a quadratic worst case, but should be near-linear for a real game
19:08:15  <TrueBrain> https://github.com/OpenTTD/OpenTTD/commit/c693463c29f98f763cc4dd6346527262b34ff6ea <- what is this for anti-feature? :P
19:08:17  <andythenorth> peter1138 if we go fast enough, does it overflow into signed?
19:08:34  <_dp_> well, loglinear I guess since it uses std::sort
19:08:42  <peter1138> Don't look at me...
19:10:12  <peter1138> Have you undone it?
19:10:26  <TrueBrain> no, it is rightfully there; just the commit message is funny :D
19:10:33  <peter1138> Oh.
19:10:49  <peter1138> Yeah, it describes the what, not the why.
19:10:56  <peter1138> Well, sort of.
19:11:29  <TrueBrain> it is part of the PR that made my work possible, so yeah ;)
19:11:47  <peter1138> It is amusing to watch vehicles teleport across the map because the game didn't bother drawing the inbetween states.
19:12:10  <peter1138> I was useful after all? Damn.
19:12:25  <TrueBrain> yeah, I do apologize for pointing that out
19:12:47  <TrueBrain> okay, in r1 it was already that first the screen was redrawn, than the viewport locations were updated
19:12:48  <TrueBrain> very odd
19:12:56  <TrueBrain> than? then
19:13:00  <TrueBrain> lol ... typing hard
19:13:27  <TrueBrain> GitHub interface has a lovely addition to blame functionality .. no clue how long it is in there
19:13:37  <TrueBrain> but on each line you can click: show me the file from before this commit
19:13:41  <TrueBrain> so it makes delving down really easy
19:14:06  <TrueBrain> also: https://github.blog/changelog/2021-02-23-github-discussions-graphql-api-public-beta/
19:14:13  <TrueBrain> means we can make the bot announce those events I guess
19:15:02  <TrueBrain> no webhook event yet, bah
19:17:40  <TrueBrain> ah, viewport is moved after drawing, as otherwise it doesn't know what pixels to redraw ..
19:17:42  <TrueBrain> lazy :P
19:18:34  <supermop_Home> is it worth suing my landlord?
19:23:20  <TrueBrain> holy crap, I enabled the overlay for cargo flow
19:32:06  *** ericnoan has quit IRC
19:32:27  <peter1138> uh oh
19:32:40  *** ericnoan has joined #openttd
19:36:56  *** jgr has joined #openttd
19:43:12  *** rptr_ has quit IRC
19:49:17  *** otetede has joined #openttd
19:50:59  <michi_cc> And with that TrueBrain vanished and was never to be seen again :D
19:51:12  *** Artea has quit IRC
19:51:16  *** Artea has joined #openttd
20:00:50  <TrueBrain> in horror, at best :P
20:00:51  <TrueBrain> :D
20:01:40  <TrueBrain> does anyone mind if I mention the Discord we are on for OpenTTD on our webpage for Support question?
20:02:10  <TrueBrain> no doubt Steam will give an influx of people with stupid questions, and they are kinda setup of dealing with it, and doing a good job at it
20:02:29  <TrueBrain> would take that burden away from us :P
20:03:11  <TrueBrain> and very specific, I mean mentioning here: https://www.openttd.org/contact.html
20:03:19  <TrueBrain> (and on the Steam page)
20:04:16  <frosch123> i would vote for removing all personal emails from that page, add links to forums, reddit, discord
20:04:31  <frosch123> and make sure that info@ is only used for serious stuff
20:04:33  <TrueBrain> I was working on something similar :P
20:04:39  <frosch123> never figure out a wording for that though
20:04:55  <TrueBrain> but it does require making the Discord (and possibly reddit) "official"
20:05:22  <TrueBrain> I will cook something up, so we can see a bit what is going on :)
20:07:52  <TrueBrain> in return, I am guessing the Discord owner will ask if we can make the Discord official, by going through the validation process that we, openttd.org, have at least 1 owner in that Discord. I assume that isn't a real problem for anyone?
20:08:15  <TrueBrain> I would make DorpsGek at info@ that owner
20:08:44  <frosch123> nah, they don't need to be official
20:08:54  <TrueBrain> quid-pro-quo
20:09:18  <frosch123> we just need to find a wording, so that people understand to ask other people :)
20:09:47  <TrueBrain> I mean, is there a reason not to make it a validated Discord server?
20:09:59  <TrueBrain> wording is easy, I think :D
20:13:31  <_dp_> is there any good way for slowing down map download on localhost (for testing)?
20:14:02  <frosch123> wasn't that asked just yesterday?
20:14:28  <_dp_> idk, I may have missed that
20:16:30  <frosch123> 20:27:21 <LordAro> https://stackoverflow.com/q/3536249/995325
20:16:31  <_dp_> oh, nvm I forgot to make 4k map
20:16:40  <_dp_> that solves it nicely xD
20:17:55  <michi_cc> It's a well run Discord, I wouldn't mind if anyone put a stamp on it.
20:18:24  *** Speedy` has quit IRC
20:18:36  *** Speedy` has joined #openttd
20:19:10  *** glx has joined #openttd
20:19:10  *** ChanServ sets mode: +v glx
20:21:36  *** Markk has quit IRC
20:21:41  *** Markk has joined #openttd
20:35:41  <_dp_> I finally got the desync I've been hunting for a year
20:35:50  <_dp_> but a repro involves patched server and a python script
20:35:56  <_dp_> fan-freaking-tastic :/
20:37:19  <andythenorth> ha
20:41:16  <_dp_> lol, I have progress, unpatched server but 2 python scripts now xD
20:42:08  <LordAro> it counts!
20:42:17  <_dp_> and 2 bugs in network gui as a bonus xD
20:44:23  <_dp_> there is openttd server called mashinky oO
20:44:59  <frosch123> call your's trainfever?
20:45:19  *** _2TallTyler has joined #openttd
20:45:20  <TrueBrain> and for a moment, enjoy the moment of missing certain other servers from the list :)
20:45:23  <TrueBrain> I should still PR that :P
20:46:26  <frosch123> i don't know what you mean :p
20:50:48  <TrueBrain> working on the website reminds me how much the layout sucks for 2021 :P
20:51:19  <TrueBrain> https://pasteboard.co/JQc0gBR.png
20:51:32  <TrueBrain> some text still needs work, but how about something like this on our Contact page?
20:52:00  <frosch123> not feature requests on gh please :)
20:52:26  <TrueBrain> that shouldn't be there, no
20:52:39  <TrueBrain> notice how info@ is now at the bottom? :D
20:53:22  <_2TallTyler> Maybe provide a link to the manual instead of sending people to ask signal questions on Discord/Reddit?
20:53:39  <_2TallTyler> I don't know what you have in the menu above, maybe you already have a link
20:53:42  <TrueBrain> there is already a Manual button, yes
20:53:50  <frosch123> and a community button
20:53:57  <TrueBrain> so if people got here, they already went past that :P
20:54:00  <frosch123> it's just to avoid adding a captcha to the emails
20:54:37  <TrueBrain> so, improvement?
20:55:04  <frosch123> some category "how to complain about other players on the server i am playing on"
20:55:11  <frosch123> not sure where to link to though :p
20:55:14  <TrueBrain> exactly :P
20:55:27  <TrueBrain> reddit seems the current target :P
20:55:29  <TrueBrain> but that is not nice :D
20:55:38  <_2TallTyler> An improvement though, definitely
20:55:49  <_dp_> frosch123, /dev/null :p
20:56:27  <frosch123> can we display a category "cat pictures" on april 1st?
20:56:48  *** michi_cc has quit IRC
20:56:48  <frosch123> by forum pm to andy, he'll never figure it out
20:56:51  *** michi_cc has joined #openttd
20:56:51  *** ChanServ sets mode: +v michi_cc
20:57:03  *** supermop_Home has quit IRC
20:58:24  <frosch123> TrueBrain: do the forum links link to the specific subforums?
20:58:39  <frosch123> problems / feature requests
20:58:39  <TrueBrain> not yet, will do
20:59:14  <_2TallTyler> Question: If I have a CompanyID, how can I get the Company associated with it? I know how to go the other way with company->index, just not the reverse.
20:59:32  <TrueBrain> GetCompanyById or something? There is some function like that
20:59:32  <frosch123> Company::Get or Company::GetIfValid
20:59:33  <glx> Company::Get
20:59:46  <TrueBrain> bah, I was waaayyyy off :P
20:59:53  <LordAro> Get::Company
21:00:00  <frosch123> one asserts for invaild ids, the other returns nullptr
21:00:17  <TrueBrain> frosch123: do you have a problem with it if we validate the Discord server? (to approach it from the different angle)
21:00:51  <frosch123> i have no idea what that means :) but as long as i do not have to be there, i don't care :)
21:00:57  <TrueBrain> you don't
21:00:58  <glx> we don't own it (same as forum), but it can be helpful
21:01:08  <TrueBrain> it just adds a green checkmark, indicating openttd.org validated the Discord as OpenTTD's ;)
21:01:23  <TrueBrain> that is literally everything :P
21:01:46  <glx> and I think it's the same with reddit
21:01:50  <TrueBrain> I guess I should also ask the reddit owners if they don't mind if we link to them .. I doubt they will
21:01:53  <TrueBrain> but always better to ask
21:02:08  *** gnu_jj has quit IRC
21:02:13  *** gnu_jj has joined #openttd
21:02:45  <frosch123> i thought they were the same
21:03:01  <TrueBrain> I believe the owners are 2 different people, but I will check :)
21:05:43  <_2TallTyler> `Company::GetIfValid(company)` says incomplete type is not allowed. Am I doing something wrong?
21:06:12  <frosch123> that means you miss the #include
21:06:33  <frosch123> in general: when c++ gives you a weird error message, it's always a missing #include
21:07:09  <frosch123> msvc usually says "missing type specifier, assuming int"
21:07:15  <frosch123> other compilers are more clever
21:08:17  <_dp_> there are plenty of weird compiler errors
21:08:18  <_2TallTyler> So I'm missing the #include for the while which contains GetIfValid? Or Company?
21:08:24  <_dp_> try forgetting } somewhere
21:08:27  <_2TallTyler> *file, not while
21:08:29  <_dp_> or messing with templates :p
21:09:02  <DorpsGek> [OpenTTD/website] TrueBrain opened pull request #192: Add: rework our "Contact" page https://git.io/JtNn3
21:09:05  <glx> company_type.hpp ?
21:09:14  <_dp_> _2TallTyler, company_base.h i think
21:10:02  <_2TallTyler> Aha, that did it. Thanks!
21:12:08  <DorpsGek> [OpenTTD/website] TrueBrain updated pull request #192: Add: rework our "Contact" page https://git.io/JtNn3
21:12:56  <_dp_> well, that's one weird error... https://i.imgur.com/buGBREo.png
21:13:11  <TrueBrain> especially that HUGE cursor :P
21:13:26  <TrueBrain> and I see my PR is still not merged :D
21:13:43  <TrueBrain> owh, it is not reviewed yet :P
21:13:46  <TrueBrain> such slackers here :D
21:14:10  <TrueBrain> frosch123 / michi_cc : C++ wizards, what is your verdict on https://github.com/OpenTTD/OpenTTD/pull/8715#discussion_r583467886 ?
21:16:10  <michi_cc> Move semantics are only relevant if copying is expensive. For a type that is not larger than a pointer and has no constructor/destructor, it has no practical meaning.
21:16:39  <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on pull request #8715: Fix: Vehicle list windows did not update when this year's profit changed https://git.io/JtNnF
21:16:46  <DorpsGek> [OpenTTD/OpenTTD] TrueBrain approved pull request #8715: Fix: Vehicle list windows did not update when this year's profit changed https://git.io/JtNnN
21:16:51  <TrueBrain> cheers :)
21:16:58  <frosch123> yes, no std::move for trivial types. it only slows down non-inlined debug builds
21:17:26  *** supermop_Home has joined #openttd
21:18:29  <frosch123> but ther GetOldestVehicleAge is broken
21:18:44  <LordAro> oh no
21:18:58  <TrueBrain> do tell?
21:19:05  <frosch123> i need to login
21:19:11  <frosch123> but can the list be empty?
21:19:13  <TrueBrain> you are not by default? :o
21:22:42  <DorpsGek> [OpenTTD/OpenTTD] frosch123 commented on pull request #8715: Fix: Vehicle list windows did not update when this year's profit changed https://git.io/JtNcu
21:23:05  <TrueBrain> lol .. lovely details :D
21:24:01  <frosch123> btw. you can also use std::accumulate with std::max
21:25:30  <frosch123> just in case you want to become a haskell programmer in the future
21:33:40  <andythenorth> that's probably the most epic gravedig ever https://www.tt-forums.net/viewtopic.php?f=29&t=6566
21:34:06  <_dp_> you can do functional programming in python!
21:34:22  <_dp_> coworkers were very happy to see my 2 screen long function call :p
21:35:14  <frosch123> andythenorth: what's that game? i don't recognise the title in the window bar
21:35:23  <andythenorth> me neither
21:35:31  <andythenorth> 17 years though
21:36:28  <_dp_> haha, BUSTED! :p
21:36:44  <frosch123> _dp_: every modern language can do functional programming, lambdas and coroutines
21:36:54  <frosch123> they like to talk about that all the time
21:38:49  <frosch123> you can recognise the real functional programmers when they complain that ohter languages are not self-modifying...
21:40:46  <DorpsGek> [OpenTTD/OpenSFX] glx22 commented on pull request #19: Replace Mercurial code with Git code https://git.io/JtNCA
21:41:47  <glx> if you want to do functional go Caml and O'Caml
21:41:52  *** ookfof[m] has quit IRC
21:41:56  *** ookfof[m] has joined #openttd
21:42:04  <DorpsGek> [OpenTTD/OpenTTD] 2TallTyler opened issue #8750: Graph point connector lines are transparent https://git.io/JtNWv
21:43:41  <frosch123> i only hear about clojure
21:43:54  <frosch123> is caml newer?
21:43:56  <_dp_> looks like opengl has thinner lines
21:44:02  <glx> caml is old
21:44:17  <glx> learned it around 2000
21:46:47  <LordAro> looks like the lines are transparent
21:48:12  <frosch123> anti-aliasing?
21:49:49  <TrueBrain> I blame _2TallTyler , he worked last on the graphs!
21:50:12  <_dp_> TrueBrain, it's fine without opengl :p
21:50:16  <frosch123> oh, silly me. we draw our lines in our own blitters... i thought we had opengl lines :p
21:50:20  <michi_cc> Without having checked this issue, in general there are some corner cases where the 8bpp only looked at the palette value and the 32bpp blitters only look at rgb, while the 40bpp blitter keeps both.
21:50:23  <TrueBrain> irrelevant _dp_ , we need someone to blame, so that is it
21:50:37  <_2TallTyler> I thought I had broken it at first!
21:51:02  *** Tirili has joined #openttd
21:51:21  <_2TallTyler> But you reviewed and approved my code so it would be partially your fault too ;)
21:51:28  <TrueBrain> darnit!
21:51:31  <TrueBrain> remind me to stop reviewing :P
21:51:32  <TrueBrain> :D :D
21:52:08  <_dp_> blame CS, everything is his fault.... partially :p
21:55:34  *** Samu has quit IRC
21:59:21  *** frosch123 has quit IRC
21:59:28  *** ircer[m] has quit IRC
21:59:39  *** ircer[m] has joined #openttd
22:00:27  *** otetede has quit IRC
22:02:18  <nielsm> https://0x0.st/-KQY.webm  I got a thing kinda working, and needed a test file
22:02:24  *** iarp[m] has quit IRC
22:02:29  *** iarp[m] has joined #openttd
22:03:19  <glx> haha encoding issues ;)
22:06:36  <andythenorth> nielsm is livestream? :P
22:07:00  <nielsm> I don't have a proper driver for it at hand, but I know it's supposed to be capable of grapgics
22:07:14  <nielsm> because I can see the shadow of a windows 98 logo on the paper roller
22:07:26  <nielsm> (printer test page printed without paper)
22:07:42  <glx> I think our amstrad dmp 2160 should still work (just needs new ribbon probably)
22:08:22  <nielsm> you're more likely to rejuvenate the old ribbon than find a new one that fits
22:08:29  <glx> and it used to work on at least win95 (I'm remember I tried it)
22:08:55  <glx> compatible with epson driver
22:09:24  *** sla_ro|master has quit IRC
22:17:06  <DorpsGek> [OpenTTD/OpenTTD] ldpl opened issue #8751: Disconnecting clients affect other clients in the queue https://git.io/JtN8c
22:23:23  <DorpsGek> [OpenTTD/website] nielsmh commented on pull request #192: Add: rework our "Contact" page https://git.io/JtN8P
22:24:18  <TrueBrain> nielsm: we follow the forums? :D
22:24:25  <TrueBrain> I will add something in that vibe, good idea :)
22:25:28  <nielsm> because I know I've seen talk in here before mentioning long threads on reddit/discord about bugs and inconveniences that then never reach anyone here until much much later, after everyone is already angry nobody is doing anything about it
22:26:10  <glx> well bugs should go on github first
22:26:30  <nielsm> yes exactly, that's the point :)
22:26:45  <nielsm> if you discuss it in a thread on reddit, chance is nobody here will see it
22:26:49  <_dp_> some feature requests are better stay on reddit though :p
22:34:02  <_dp_> oh, got desync repro without python as well, updated the bug
22:34:40  *** _2TallTyler has quit IRC
22:35:07  <_dp_> now I'm pretty sure that accounts for at least some of the desyncs plaguing the servers
22:35:38  <_dp_> and mb some other weird errors as well
22:48:07  <DorpsGek> [OpenTTD/website] TrueBrain updated pull request #192: Add: rework our "Contact" page https://git.io/JtNn3
22:48:35  <DorpsGek> [OpenTTD/website] TrueBrain commented on pull request #192: Add: rework our "Contact" page https://git.io/JtNBC
22:51:15  <DorpsGek> [OpenTTD/website] TrueBrain updated pull request #192: Add: rework our "Contact" page https://git.io/JtNn3
22:52:43  *** gelignite has quit IRC
23:04:10  <DorpsGek> [OpenTTD/OpenTTD] JGRennison commented on issue #8751: Disconnecting clients affect other clients in the queue https://git.io/JtN8c
23:04:58  *** jottyfan has joined #openttd
23:05:43  *** jottyfan has quit IRC
23:07:01  <_dp_> oh, and desync is probably because clients are not queued
23:07:23  <_dp_> so when second one tries to catch up client is already gone
23:07:25  <andythenorth> bed
23:07:26  *** andythenorth has quit IRC
23:07:46  <_dp_> kinda makes it a separate bug though if that's true
23:08:37  *** nielsm has quit IRC
23:15:30  *** Tirili has quit IRC
23:18:40  <DorpsGek> [OpenTTD/OpenTTD] ldpl commented on issue #8751: Disconnecting clients affect other clients in the queue https://git.io/JtN8c
23:25:52  *** Wolf01 has quit IRC
23:27:07  <TrueBrain> lol, a bug in our state machine .. another 14 year old bug fix incoming tomorrow? :D
23:29:57  <LordAro> tis the season for it
23:31:57  <TrueBrain> one still needs a review :D
23:32:25  <DorpsGek> [OpenTTD/OpenTTD] LordAro approved pull request #8746: Fix bddfcaef: don't tell twice that a client left because of a timeout etc https://git.io/JtNE8
23:32:42  <TrueBrain> \o/
23:32:55  <DorpsGek> [OpenTTD/OpenTTD] TrueBrain merged pull request #8746: Fix bddfcaef: don't tell twice that a client left because of a timeout etc https://git.io/JtFxq
23:32:59  <TrueBrain> cheers
23:33:16  <DorpsGek> [OpenTTD/OpenTTD] LordAro approved pull request #8747: Codechange: replace _realtime_tick with std::chrono https://git.io/JtNE0
23:33:17  <LordAro> oh, i did already look through this one
23:33:32  <LordAro> let's find out if chrono is stable enough
23:33:44  <DorpsGek> [OpenTTD/OpenTTD] JGRennison commented on issue #8751: Disconnecting clients affect other clients in the queue https://git.io/JtN8c
23:33:54  <DorpsGek> [OpenTTD/OpenTTD] LordAro approved pull request #8745: Fix: don't link to OpenGL with SDL2 as backend; SDL2 dynamically loads it https://git.io/JtNEu
23:34:45  <DorpsGek> [OpenTTD/OpenTTD] TrueBrain merged pull request #8745: Fix: don't link to OpenGL with SDL2 as backend; SDL2 dynamically loads it https://git.io/JtFQ4
23:36:09  <TrueBrain> LordAro: our video drivers depend on it; I sure hope it is :D
23:36:19  <DorpsGek> [OpenTTD/OpenTTD] TrueBrain merged pull request #8747: Codechange: replace _realtime_tick with std::chrono https://git.io/JtFpt
23:36:40  <LordAro> true enough :)
23:38:58  <_dp_> feels like overkill to create a bug for it but can someone with better english fix this message? https://github.com/OpenTTD/OpenTTD/blob/master/src/network/network_server.cpp#L1108
23:39:23  <_dp_> smth along the lines "spectator ({client_id}) issued non-spectator command {cmd}"
23:39:59  <Xaroth> then create a PR for it? :P
23:40:19  <LordAro> _dp_: as you say would be fine
23:40:48  <LordAro> s/issuing command/issued non-spectator command {cmd}/
23:44:15  <DorpsGek> [OpenTTD/OpenTTD] TrueBrain updated pull request #8741: Move GameLoop into a thread (and no longer run Paint in a thread) https://git.io/JtdUz
23:44:25  <TrueBrain> I think ^^ is now ready for review, but I need some sleep first to be sure :P
23:44:32  <_dp_> hm... and what's the correct format for command id? xD
23:46:22  <_dp_> though I guess those messages don't particularly care about correct format strings
23:47:07  <_dp_> pretty sure %d isn't quite the right one for uint32
23:48:56  <LordAro> %u
23:49:33  <_dp_> not %"PRIu32" ?
23:49:41  <LordAro> well, technically speaking, yes
23:49:45  <LordAro> but urgh
23:49:52  <LordAro> can you imagine having to write that every time? :p
23:50:20  <_dp_> no, so I use fmt::* :p
23:50:40  <_dp_> or %d and cast everything to int xDD
23:50:52  <TrueBrain> or just, you know, %u :)
23:52:30  <_dp_> does %u actually work fine on 64 bit archs?
23:53:06  <DorpsGek> [OpenTTD/OpenTTD] TrueBrain updated pull request #8741: Move GameLoop into a thread (and no longer run Paint in a thread) https://git.io/JtdUz
23:53:09  <TrueBrain> okay, so this is ready for review ^^ :)
23:53:25  <TrueBrain> _dp_: grep for it in our code, and get your answer
23:53:28  <LordAro> dejavu
23:53:45  <TrueBrain> nah, now I am sure :)
23:54:00  <TrueBrain> would really appreciate it if other people could test it for a bit
23:54:02  <TrueBrain> see if they can break it
23:54:15  <_dp_> TrueBrain, the first thing I did is I grepped PRIu32, 0 hits :p
23:54:19  <TrueBrain> looking at the code, I see no way it can deadlock, but ..
23:54:25  <TrueBrain> _dp_: and now grep %u ...
23:54:36  <TrueBrain> some questions are already answered by just looking at how other code does it, really :)
23:54:45  <_dp_> I don't need to grep to see %d used for uint32 :p
23:55:05  <TrueBrain> or, of course, you can create problems where there are now :) That is fine too :)
23:55:08  <TrueBrain> now = none
23:55:09  <_dp_> question is should I try to make it correct or just do like other code does :p
23:55:34  <TrueBrain> lot of code does %u, so I am not sure what you hope to gain from this reasoning :)
23:55:55  <TrueBrain> I am just waiting for the "OpenTTD sucks" sentence :D
23:56:16  <_dp_> mostly I'm just curious how %u manages to work fine even on 64 architectures xD
23:56:31  <TrueBrain> what do you expect that breaks?
23:56:38  <_dp_> I'd expect it to crash or print garbage
23:56:46  <TrueBrain> why?
23:56:50  <_dp_> since it should expect 64 value
23:56:55  <_dp_> but gets 32
23:57:01  <_dp_> * bit
23:57:14  <TrueBrain> pretty sure %lu is for 64bit?
23:57:29  <TrueBrain> %u does unsigned int
23:57:43  <TrueBrain> (by specs)
23:58:23  <_dp_> but is int 4 bytes?
23:59:03  <_dp_> oh, int is <= 32
23:59:05  <TrueBrain> src/stdafx.h:        typedef unsigned int uint;
23:59:08  <_dp_> so it only breaks on 16 bit :p
23:59:12  <TrueBrain> for OpenTTD, "uint" == %u
23:59:35  <_dp_> it's lu that will break on 32/64 :p

Powered by YARRSTE version: svn-trunk