Config
Log for #openttd on 25th January 2021:
Times are UTC Toggle Colours
00:13:29  *** iSoSyS has quit IRC
00:17:55  *** gregdek_ has joined #openttd
00:18:17  *** gregdek has quit IRC
00:18:17  *** gregdek_ is now known as gregdek
00:19:39  *** reldred_ has joined #openttd
00:19:44  *** reldred has quit IRC
00:19:44  *** reldred_ is now known as reldred
01:53:47  <DorpsGek> [OpenTTD/OpenTTD] maaaaaaaaaad opened issue #8606: Assertion failed at line 582 https://git.io/JtZwW
02:17:35  *** Flygon has joined #openttd
03:34:19  *** Wormnest has quit IRC
03:55:20  *** debdog has joined #openttd
03:56:02  *** glx has quit IRC
03:58:42  *** D-HUND has quit IRC
04:52:55  *** keith_ has joined #openttd
05:32:57  <gregdek> Trying to load scenarios / savefiles in dedicated server mode, and for some reason a new map is loaded anyway. Any hints?
05:34:16  <gregdek> Logfile indicates that it sees the savefile, but then it just creates a new map.
05:34:54  <gregdek> Two consecutive lines in commands-out.log:
05:35:07  <gregdek> (well, three):
05:35:17  <gregdek> load: /home/openttd/scenario/simple-scenario.scn
05:35:17  <gregdek> load: /home/openttd/scenario/simple-scenario.scn
05:35:17  <gregdek> new_map: da57ddbb
06:02:59  <gregdek> Ah. And the answer is "spurious newgame command in on_dedicated.scr". Never mind!
06:06:49  *** keith_ has quit IRC
06:21:48  *** keith__ has joined #openttd
07:05:53  *** andythenorth has joined #openttd
07:09:41  *** sla_ro|master has joined #openttd
07:24:42  *** andythenorth has quit IRC
07:25:40  *** snail_UES_ has quit IRC
07:51:20  *** WormnestAndroid has quit IRC
07:51:33  *** WormnestAndroid has joined #openttd
07:58:44  *** supermop_Home_ has quit IRC
08:09:00  *** Wolf01 has joined #openttd
08:24:05  *** heffer has quit IRC
08:24:19  *** heffer has joined #openttd
08:55:39  *** HerzogDeXtEr has joined #openttd
09:25:14  <Wolf01> So, I might be a special thing of stupid, but I'm not able to compile even with vs2019: RC2135	file not found: C:/Users/Wolfolo/CMakeBuilds/7d3d6e5d-10cb-414a-a4b9-147a8ea1645d/build/x64-Debug/CMakeFiles/openttd.dir/embed.manifest	E:\progetti\OpenTTD\master\master	E:\progetti\OpenTTD\master\CMakeFiles\openttd.dir\manifest.rc	2
09:36:57  <LordAro> rerun cmakw?
09:46:24  *** qwebirc43687 has quit IRC
09:56:58  <Wolf01> Already cleared cache and recreated it, that was the first thing I did
10:28:24  <TrueBrain> wauw, 8606 savegame has 0.5GB of NewGRFs .. lol
10:28:45  <TrueBrain> 2 of those are not on BaNaNaS :(
10:29:51  <LordAro> i find it interesting that they clearly wrote the assertion message out manually
10:29:55  <LordAro> due to its inaccuracies
10:30:34  <TrueBrain> it is sweet :)
10:30:42  <TrueBrain> newshipsw.grf .. euhh
10:30:52  <LordAro> that's an MB thing, isn't it?
10:31:02  <_dp_> it's not that uncommon, just yeti alone is like 250m, rawr 360
10:34:12  <TrueBrain> hmm, and I cannot find the right WAS.grf
10:34:19  <TrueBrain> it is not the one from openttdcoop space
10:34:48  <TrueBrain> is there a way to load the savegame anyway?
10:36:17  <LordAro> newgrf_developer setting, iirc
10:36:29  <LordAro> but looking at the gamelog, there's an awful lot of newgrf changes in there
10:38:02  <LordAro> stacktrace makes no sense either, but that might be macos being weird
10:40:59  <TrueBrain> checked all the WAS nightlies .. none match
10:42:20  <_dp_> it does assert on linux on the same line
10:42:30  <_dp_> but considering all the newgrf issues...
10:43:42  <TrueBrain> owh,found WAS.grf .. now newshipsw.grf :P
10:44:34  <TrueBrain> not on grfcrawler either
10:49:47  <TrueBrain> right .. I did press "load" ... it is taking its sweet time :)
10:51:10  <TrueBrain> many GRFs in there are not multiplayer safe :P
10:52:45  <TrueBrain> tile=6031629
10:52:54  <_dp_> btw, just thought that COMPILING.md should probably show how to make release build instead of sending to cmake docs
10:53:29  <TrueBrain> we had that talk recently .. if you are compiling manually, aren't you most likely to be a developer? And as such, a debug build is of more use to you?
10:53:29  <_dp_> apparently not everyone using linux can navigate dev documentations easily ;)
10:54:10  <_dp_> I've seen a lot of people compiling manually on linux who didn't even compile anything before
10:54:13  <TrueBrain> We used to explain a lot of things in a lot of detail ... it often no longer worked :P
10:54:27  <TrueBrain> which was even more horrible
10:54:38  <TrueBrain> I think someone should write a very nice and clear written wiki page, honestly
10:55:08  <_dp_> e.g. just some random players who want a dedicated server
10:55:19  <TrueBrain> COMPILING.md seems like the wrong place to explain those use-cases
10:55:25  <TrueBrain> as they most likely need a bit more help :D
10:55:45  <TrueBrain> and we can point to the wiki from COMPILING.md btw
10:56:44  <_dp_> well, it's not that hard to compile stuff on linux when you can find right commands to paste ;)
10:56:58  <TrueBrain> exactly what a wiki is for, not ? :)
10:57:08  <TrueBrain> means people can also explain how to install the dependencies depending on the OS used
10:57:17  <_dp_> also any dev that wants debug can probably figure out what part of "cmake -DCMAKE_BUILD_TYPE=Release .." he needs to change :p
10:57:33  <TrueBrain> lately I have seen a few people who don't have LZMA installed, and wonder why things don't work correctly :P
10:58:07  <TrueBrain> what we talked about a while ago, is: who is the public of COMPILING.md
10:58:10  <TrueBrain> the answer was: developers
10:58:18  <TrueBrain> I agree there is another group of people who need help
10:58:23  <TrueBrain> but I really think a wiki page is more suited for that
10:58:36  <TrueBrain> as that means the community can maintain what works for these group of people
10:58:59  <TrueBrain> and we can simply link to that from COMPILING.md, in case people are a bit lost :)
10:59:16  <TrueBrain> possibly it is even already on the wiki somewhere .. just .. not sure what the state is there :)
11:00:05  <TrueBrain> https://wiki.openttd.org/en/Archive/Compilation%20and%20Ports/Compiling%20on%20GNU%E2%88%95Linux%20and%20%F0%9F%9F%89BSD
11:00:09  <TrueBrain> someone made an attempt already
11:00:14  <TrueBrain> not sure .. it is quality I am reading there
11:00:54  <TrueBrain> the crash of 8606 is because of this: DoTriggerHouse(TILE_ADDXY(tile, 0, 1), trigger, random_bits, false);
11:00:59  <TrueBrain> where tile is now out-of-bounds
11:01:35  <_dp_> TrueBrain, that attempt doesn't even seem to mention release build :p
11:02:13  <TrueBrain> _dp_: and contains a lot of text with no clear guidance to the user
11:02:18  <TrueBrain> so I fully understand people fail in the exercise
11:02:54  <TrueBrain> we really need someone to clean up (content-wise) our wiki, honestly
11:03:29  <TrueBrain> my issue is, sure, we can mention "release" part in COMPILING.md, but that is not really the problem of the user. He also needs to understand dependencies, for example. And if you never used CMake, I am pretty sure you don't really understand dev-dependencies too
11:03:54  <TrueBrain> and before you know it, COMPILING.md is just a wiki page with all kind of well-intended advise how to run on OS-flavour-X
11:04:49  <TrueBrain> this is what other CMake projects do too btw; at least, those I could find and knew of :)
11:05:07  <TrueBrain> but we severely lack people willing to maintain the wiki .. which tells you something too :)
11:06:11  <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on issue #8606: Assertion failed at line 582 https://git.io/JtZwW
11:07:34  <TrueBrain> anyway, _dp_ , I always wondered why CMake doesn't do RelWithDebInfo by default .. they are pretty clear they will never change that, but I never found a good reason why Debug is their default. A bit on the same argumentation: it is meant for developers, I guess
11:08:58  <_dp_> personaly my main issue is that I always forget how to make release and have to google xD
11:08:59  <TrueBrain> backtrace of 8606 suggests there is a 2x2 house on the border of the map, with tiles extending past it, or something
11:09:11  <TrueBrain> _dp_: I absolutely have the same issue :P
11:09:18  <TrueBrain> that is why I linked the CMake docs
11:09:26  <TrueBrain> the only hit on release is how to make a release :)
11:09:59  <TrueBrain> but once you have a build folder, you can rerun cmake to use the same arguments .. that helps :)
11:10:04  <TrueBrain> and normal people use cmake-gui, I have been told
11:10:55  <_dp_> I'm clearly not normal since I don't understand why would it even have a gui...
11:11:06  <TrueBrain> there is also .. ccmake I believe
11:11:08  <TrueBrain> which does the same in CLI
11:11:25  <TrueBrain> ha, I remembered correctly :)
11:11:38  <LordAro> _dp_: nice way of selecting all the options, iirc
11:11:41  <TrueBrain> and no, I am also not normal :P
11:12:52  <_dp_> though I guess when your options need googling every damn time making a gui suddenly seems useful xD
11:14:12  <TrueBrain> hmm, no, that 2x2 house is in the middle of the map
11:14:15  <TrueBrain> so it is not a border issue
11:14:31  <TrueBrain> it is a house marked as a 2x2 which really is not a 2x2 or something
11:15:21  <TrueBrain> airport is build on top of a 2x2 house
11:16:22  <_dp_> yeah, that save looks very messed up
11:16:41  <_dp_> may not be reproducable with a clean start
11:17:22  <TrueBrain> how on earth did that ever happen, I wonder :)
11:17:34  <_dp_> newgrf changed mid-game?
11:17:36  <LordAro> house became not a house
11:18:13  <_dp_> I wouldn't be surprised if this asserts in 1.10 as well if run in debug build
11:18:21  <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on issue #8606: Assertion failed at line 582 https://git.io/JtZwW
11:18:30  <TrueBrain> _dp_: it should, yes
11:18:50  <TrueBrain> disabling asserts makes the game faster, but also means people miss errors :)
11:18:56  <TrueBrain> but how did this situation came to be?
11:19:13  <TrueBrain> we check that there is space for a 2x2 before it is build, not?
11:19:32  <LordAro> ideally the game wouldn't crash
11:19:40  <LordAro> regardless of newgrf shenanigans
11:19:42  <_dp_> can't it be that 2x2 house stole id from 1x1?
11:19:52  <LordAro> yeah, that's what i suspect
11:19:58  <TrueBrain> at least from a 2x1 in that case
11:20:03  <TrueBrain> as 2 of the tiles really are of the same building
11:20:09  <TrueBrain> just on the other 2 is an airport :P
11:20:15  <TrueBrain> and it seems to be in construction
11:20:23  <TrueBrain> and there are no recent NewGRF changes
11:22:24  <_dp_> there was this change in 1.11 https://github.com/OpenTTD/OpenTTD/commit/a82572d0f5a55a33df4163b429f891ea76ed92a7
11:22:28  <_dp_> may or may not be related
11:23:08  <LordAro> there was also TB's no-house-walking change
11:23:31  <_dp_> yeah, but that's even less likely to be relaled
11:24:32  <TrueBrain> it does check if the tiles below are available, even when rebuilding
11:24:42  <TrueBrain> LordAro: yeah, I considered it too :D
11:24:44  <TrueBrain> but it seems unlikely
11:25:25  <_dp_> this like looks quite suspicious https://github.com/OpenTTD/OpenTTD/commit/a82572d0f5a55a33df4163b429f891ea76ed92a7#diff-88e5b6c0a7c202f09d40e5a721b54aeec678086c7116459241d4fab02f30a33cL2529
11:26:16  <_dp_> I've no idea what _loaded_newgrf_features.has_newhouses did but it suddenly enables a lot of code for that case
11:27:29  <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on issue #8606: Assertion failed at line 582 https://git.io/JtZwW
11:27:51  <TrueBrain> it is only 1 building that is the issue, it seems
11:30:49  <_dp_> TrueBrain, do you know what newgrf that building comes from?
11:31:04  <TrueBrain> I love how both the PR and commit message do not define AT ALL why it would be save to remove that flag btw :P (https://github.com/OpenTTD/OpenTTD/pull/8178 )
11:31:11  <TrueBrain> euh, I can look that up
11:32:36  <TrueBrain> Japan Set3: Buildings
11:32:57  <TrueBrain> game runs at 15 frames/s for me, lol
11:34:09  <_dp_> ty
11:34:13  <_dp_> TrueBrain, in release? ;)
11:34:20  <TrueBrain> yes, in release :)
11:34:31  <_dp_> oof
11:34:37  <TrueBrain> I have more than one build-folder .. I don't have to remember settings anymore :P
11:34:47  <TrueBrain> I just jump in one which is preconfigured to do a certain combination of options :D
11:34:51  <TrueBrain> really useful, I can tell you :)
11:35:18  <_dp_> oh, that's a nice trick
11:35:19  <TrueBrain> well, it runs at 15 fps, and I killed the AIs :P
11:35:21  <_dp_> should do that too
11:35:25  <TrueBrain> with the AIs it is .... even worse
11:35:50  <TrueBrain> CMake just reloads when ever the code changed in a way it should; so the build-folders survive any "git checkout" :)
11:36:17  <TrueBrain> funny enough, my main build folder is called "build4"
11:36:24  <TrueBrain> not sure why that happened ....
11:36:27  <TrueBrain> "build" is emscripten ..
11:37:23  <_dp_> lol, that no growth nrt is such a noob trap now :p
11:37:35  <TrueBrain> no growth nrt?
11:37:44  <_dp_> roads that don't allow towns to grow
11:38:25  <TrueBrain> oef
11:38:36  <_dp_> like in that save https://i.imgur.com/ADZ2Tv5.png
11:38:55  <_dp_> he clearly didn't mean that
11:39:11  *** skrzyp2 has joined #openttd
11:39:17  <TrueBrain> nasty ...
11:39:26  <TrueBrain> okay, let the game run for a few months, no other crashes
11:39:30  <TrueBrain> so it really is that one building
11:40:36  <TrueBrain> so if I can find the correct version of newshipsw, I can fix the savegame for him
11:41:09  <TrueBrain> initially it looked like it was a fresh building
11:41:10  <TrueBrain> it is not
11:41:20  <TrueBrain> just most of the graphics are on the other 2 tiles
11:41:46  <TrueBrain> so who knows how long this building has been there :P
11:42:23  <TrueBrain> @base 16 10 4a
11:42:23  <DorpsGek> TrueBrain: 74
11:42:26  <TrueBrain> 74 years old
11:47:27  <TrueBrain> right, now for that grf ...
11:49:30  *** virtualrandomnumber has joined #openttd
11:50:13  *** virtualrandomnumber has quit IRC
11:53:11  <TrueBrain> found it :D
11:54:08  <Timberwolf> I learnt this from gamescripts: the more reliable your code, the wilder the savegames you get with bug reports.
11:54:58  <Timberwolf> There was one with Villages Is Villages that was a JGR 8192*8192 with towns and industry density set to "high"
11:55:39  <Timberwolf> I think it took me a good 30 minutes of staring at it going, "how do you ever plan to connect all of this stuff?" before I could start actually investigating and fixing the problem.
11:59:01  <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on issue #8606: Assertion failed at line 582 https://git.io/JtZwW
11:59:04  <DorpsGek> [OpenTTD/OpenTTD] TrueBrain closed issue #8606: Assertion failed at line 582 https://git.io/JtZwW
11:59:43  <TrueBrain> yeah ... this map too .. really big ... no clue how anyone would ever keep sanity in it .. but given by the amount of vehicle errors, I think he is not :P
12:00:14  <TrueBrain> owh, I did not use the beta1 tag to fix the savegame .. owh well, hope that isn't an issue :D
12:01:03  <TrueBrain> well, that was an interesting way to spend my morning
12:02:05  <LordAro> such a helpful TrueBrain
12:02:16  <LordAro> i'd have just closed it with "games with changed newgrfs are not supported"
12:05:04  <Timberwolf> I feel that way when I get a "so I combined your roads with 7 other roadtype sets, and now buses only work on canals..."
12:06:34  <Timberwolf> Mostly because roadtypes is like a modding community from the early 2000s, and the thought of having to visit 3 different forums and then decompile the sets to figure out what labels and compatibility tables they used doesn't sound like fun.
12:16:17  <_dp_> nah, there might be more to that bug
12:16:34  <_dp_> there was a similar bug btw https://github.com/OpenTTD/OpenTTD/issues/2613
12:16:46  <_dp_> not the same cause but apparently same effect
12:18:20  <Eddi|zuHause> that doesn't sound related at all
12:18:32  *** Samu has joined #openttd
12:18:58  <_dp_> Eddi|zuHause, "Also it results in half houses, if the newgrf overrides houses by houses of different size"
12:19:06  <Samu> this conversion isn't as easy as it looks
12:19:18  <_dp_> it also messes with the same part of code #8178 did
12:19:25  <Timberwolf> I do like with these games where you'll open them up and they'll have CZTR combined with the most flat-coloured, jagged 1x road vehicle set imaginable. Even I'm not that much of a monster :)
12:27:13  <Samu> im testing timberwolfs articulated trams, they're easier to pixel count
12:27:21  <Samu> matt kimber or something
12:47:46  <TrueBrain> LordAro: looking at it from another angle, the dude has been playing ... months? on this map .. small effort to make sure he can keep on playing :D
12:50:35  <_dp_> ok, if I understood it correctly, if #8178 changed anything it was to the good :)
12:50:46  <_dp_> so should be unrelated
13:05:20  <LordAro> TrueBrain: sure, but they've explicitly bypassed all the "changing newgrfs will break things" warnings
13:05:36  <TrueBrain> effort vs reward :)
13:06:22  <Timberwolf> Samu: Yes, my immense efforts to keep my "online name" separate from my real name ;)
13:07:42  <Timberwolf> The only risk with my trams is they don't use the correct vehicle templates, so it's possible to create something that works for them and breaks for everything else.
13:14:14  <Eddi|zuHause> _dp_: but even if the two are related, it needs a reproducible case, which the report doesn't (and can't) provide
13:16:36  *** glx has joined #openttd
13:16:36  *** ChanServ sets mode: +v glx
13:18:52  <glx> nice detective work on #8606 TrueBrain, when I tried to load the savegame (in debug build) I gave up after 1h and it was still in AfterLoad()
13:19:10  <TrueBrain> yeah ... it is a bit of a big game :P
13:19:20  <TrueBrain> it helps I have a CPU with a very high single-core clock speed, I noticed :D
13:20:19  <glx> but yeah base on crash log I expected it was related to the many grf changes during the life of the savegame
13:20:42  <glx> anyway MacOS stack trace is so unuseful ;)
13:20:43  <TrueBrain> I couldn't find a code-path that causes an airport to be build on top of a 2x2 house :P
13:20:50  <TrueBrain> yeah ... that is really totally not helpful at all
13:20:51  <TrueBrain> weird
13:21:21  <_dp_> mac trace looks like it is from another thread or smth
13:21:46  <TrueBrain> if I had a Mac and the willpower, I would triage it, but I have neither :D
13:30:39  *** snail_UES_ has joined #openttd
13:31:49  <TrueBrain> LordAro: do you think it is worth backporting my SDL2 PR to SDL1? Or will we be dropping SDL1 any time soon?
13:33:38  <LordAro> honestly not sure
13:33:53  <TrueBrain> I had the same issue :P
13:33:54  <LordAro> given there are no codechanges in SDL1, probably not a particular problem not to backport them anyway
13:33:59  <LordAro> er
13:34:08  <LordAro> s/no codechanges/no actual changes/
13:34:23  <TrueBrain> I was wondering if we should port OpenGL to SDL1
13:34:32  <TrueBrain> but I don't think many people still use SDL1
13:34:40  <LordAro> i'm not sure if there's any platform that would still actively use it
13:34:49  <LordAro> android, perhaps?
13:34:55  <TrueBrain> maybe we should add a deprecation warning to SDL1
13:34:59  <TrueBrain> just so people report that if they still use it
13:35:25  <glx> so yesterday Wolf01 found an issue with VS2017 (I could confirm it https://github.com/glx22/OpenTTD/actions/runs/507639860)
13:35:40  <TrueBrain> "You are using the SDL1 video-driver; this is set to be removed for 1.12. If you feel this would be an error, please create an issue on our issue tracker"
13:35:56  <glx> the error message is kind of weird, but I think it's right and we are lucky other compilers accept our code
13:36:36  <Eddi|zuHause> TrueBrain: why would that be any more of an issue than suddenly dropping XP support?
13:36:40  <glx> so I tried https://github.com/glx22/OpenTTD/commit/b8dcecdbf4840cce2137c621e457ed6713826030 that seems safer and the result https://github.com/glx22/OpenTTD/runs/1762162432
13:39:13  <glx> string literals are const char[] and an aggregate initialisation does copy initialisation and we ask to copy const char[] to char[] which is not very clean
13:39:27  <TrueBrain> Eddi|zuHause: because we can estimate how many people (should be) using XP .. we have no clue about SDL1 usage
13:41:00  <Eddi|zuHause> right, introduce phone-home telemetry :p
13:42:41  <TrueBrain> no
13:45:52  <glx> and today I noticed discussion are now visible in mobile app
13:46:36  <TrueBrain> glx: a few days ago, yes :) https://github.blog/changelog/ shows 12th of Jan :)
13:47:14  <glx> yeah but there was no new discussion notifications until today :)
13:47:22  <TrueBrain> haha
13:47:58  <glx> https://github.blog/changelog/2021-01-21-github-actions-short-sha-deprecation/ <-- are we concerned ?
13:48:10  <TrueBrain> we use either master of branch-names
13:48:11  <TrueBrain> so not really
13:48:18  <TrueBrain> at least .. we shouldn't been using sha-hashes
13:48:21  <TrueBrain> so that would be a bug :)
13:48:52  <Eddi|zuHause> TrueBrain: so there's two questions about that deprecation warning: a) who is the target audience? (players, package maintainers, people who compile for random weird platforms?), and b) where should it be displayed? (configure, console, ingame?)
13:51:56  <Eddi|zuHause> also, there should be a longer explanation somewhere (public, e.g. readme, news, whatever) which explains why it's being deprecated and what replaces it, so we can refer people to it
13:57:03  <Samu> wow, there can be 3 articulated parts in the same inner curve
13:57:10  <Samu> that's a feat!
14:01:47  <TrueBrain> As there is no .dorpsgek.yml: [OpenTTD/OpenSFX] Wuzzy2 opened issue #2: Relicense OpenSFX under an actual FOSS license https://git.io/JtnIG
14:02:55  <LordAro> didn't realise opensfx didn't have readme/changelog/licence in the right place
14:04:04  <milek7_> I think sdl2 on haiku was glitched
14:05:33  <Eddi|zuHause> TrueBrain: so i read the issue, and the first post of the forum thread, and came to the conclusion "ah whatever. it's some technicality bikeshedding, why even bother?"
14:07:07  <Samu> this conversion code keeps surprising me
14:07:51  <Samu> if all vehicle parts are told to advance one frame, then they don't advance at all. that's smart
14:13:11  <Samu> whoever came with SLV_188, was a genius, that code is impecabl
14:14:55  <Samu> it was frosch123
14:18:00  *** supermop_Home_ has joined #openttd
14:21:45  <TrueBrain> just because I enjoy playing bot: [OpenTTD/OpenSFX] Wuzzy2 opened issue #3: Documenting original license of sounds https://git.io/JtnIN
14:22:20  *** iSoSyS has joined #openttd
14:26:26  <Samu> i can't find more issues, for now
14:27:39  <Samu> https://github.com/OpenTTD/OpenTTD/compare/master...SamuXarick:roadveh-movement
14:27:47  <Samu> should I PR?
14:29:34  *** Wuzzy has joined #openttd
14:30:40  <Samu> (after cleaning up the commit history to something more readable)
14:33:24  <Samu> no more crashes, no more misaligned articulated vehicles
14:33:29  <Samu> drive on left works
14:33:32  <Samu> drive on right works
14:35:27  <_dp_> that savegame conversion...
14:35:50  <_dp_> why not just decouple movement logic from graphics? should be simpler and much better in a long run
14:36:56  <Eddi|zuHause> context?
14:38:31  <Samu> i'm not sure I understand your question
14:40:39  <Timberwolf> _dp_: Might also have a chance of solving the age-old Gold Rush problem (vehicles at 4x zoom still move according to a 1x movement scheme, making the slow ones appear a little jerky)
14:40:51  <_dp_> yeah, that too
14:42:40  <Eddi|zuHause> well, my solution to that problem was rejected :)
14:43:10  <DorpsGek> [OpenTTD/OpenTTD] mattkimber commented on pull request #8604: Feature: allow limiting zoom level of NewGRF-provided sprites https://git.io/JtnmG
14:45:05  <DorpsGek> [OpenTTD/OpenTTD] mattkimber commented on pull request #8604: Feature: allow limiting zoom level of NewGRF-provided sprites https://git.io/JtnmV
14:46:22  <supermop_Home_> oops my game was unpaused for like 12 hours last night
14:46:58  <DorpsGek> [OpenTTD/OpenTTD] mattkimber commented on pull request #8604: Feature: allow limiting zoom level of NewGRF-provided sprites https://git.io/Jtnmy
14:50:15  <_dp_> Eddi|zuHause, what was your solution?
14:52:32  <Eddi|zuHause> _dp_: i was exposing some "substep" value to the GRF https://github.com/OpenTTD/OpenTTD/issues/5300
14:56:44  <_dp_> yeah, that doesn't look like a good solution :p
14:56:54  <_dp_> how would newgrf even use it, add sprite offsets?
14:57:41  <Eddi|zuHause> yeah
15:01:40  <Eddi|zuHause> i think bad_brett actually coded a test newgrf. i vaguely remember he made a video
15:02:32  <_dp_> yeah, I saw some ponies, question is did he ever publish it :p
15:03:12  <Eddi|zuHause> https://www.youtube.com/watch?v=mGUnjnNEfGE
15:04:40  <_dp_> yeah, horned ponies xD
15:05:31  <LordAro> oh god the cpu requirements for such a thing
15:06:10  <Eddi|zuHause> can't imagine this being much worse than whatever NewGRFs are already doing
15:06:52  <Eddi|zuHause> from OpenTTD's perspective, this is a one-liner
15:07:04  <LordAro> fair
15:07:08  <_dp_> yeah, it's just changing sprites
15:07:17  <_dp_> only difference is computing which to use
15:08:31  <Eddi|zuHause> well, trying to cache sprites between multiple ticks is going to be hopeless
15:08:46  <_dp_> what's funnier is that after such a fine movement they'll just jump 45 degrees on a turn :p
15:10:30  <Eddi|zuHause> well, you can sorta try to remedy that with var62
15:10:49  <Eddi|zuHause> but it's going to be a nightmare to handle all combination
15:11:42  <Eddi|zuHause> but there's probably a reason he made this preview only with a straight road :p
15:12:10  <DorpsGek> [OpenTTD/OpenTTD] glx22 opened pull request #8607: Codechange: Implement a constructor for CurrencySpec (fixes VS2017 compile failure) https://git.io/JtnOX
15:12:44  <Eddi|zuHause> https://www.youtube.com/watch?v=PBinroFjza8
15:13:20  <_dp_> Eddi|zuHause, what is var62?
15:13:37  <Eddi|zuHause> checking angles between vehicles
15:14:11  <DorpsGek> [OpenTTD/OpenTTD] LordAro approved pull request #8607: Codechange: Implement a constructor for CurrencySpec (fixes VS2017 compile failure) https://git.io/JtnOQ
15:14:14  <_dp_> oh, that one
15:16:24  <_dp_> jeez, so much internal stuff exposed to newgrfs
15:16:31  <_dp_> vehicle movement is doomed :/
15:16:38  <LordAro> var62 considered harmful
15:17:28  <Eddi|zuHause> that one is also my fault, based on a request by George
15:18:26  <Samu> overdose of information https://i.imgur.com/Z54DSNj.png
15:19:34  <Eddi|zuHause> https://github.com/OpenTTD/OpenTTD/issues/2521
15:22:52  <Eddi|zuHause> (note that var45 was there before, so angles were already exposed anyway)
15:27:05  <DorpsGek> [OpenTTD/OpenTTD] Eddi-z commented on pull request #8604: Feature: allow limiting zoom level of NewGRF-provided sprites https://git.io/Jtn3i
15:32:09  <_dp_> I don't get newgrf docs at all, where is the spec for these? https://github.com/OpenTTD/OpenTTD/blob/master/src/newgrf_engine.cpp#L808
15:33:57  <Eddi|zuHause> those are the 0x80+ variables
15:35:08  <glx> http://marcin.ttdpatch.net/sv1codec/TTD-locations.html
15:35:12  <glx> hard to find
15:35:15  <Eddi|zuHause> "For other 80+x variables confer the TTD vehicle structure." http://marcin.ttdpatch.net/sv1codec/TTD-locations.html#_VehicleArray
15:35:37  <Eddi|zuHause> it's all really buried and ancient
15:37:55  <Eddi|zuHause> was there ever a sentence there about "80+ variables that are not described in this page might not be available in OpenTTD"?
15:41:05  <_dp_> I wonder if that makes _road_drive_data a part of newgrf spec...
15:43:38  <Eddi|zuHause> what's that?
15:44:07  *** Flygon has quit IRC
15:45:28  <_dp_> Eddi|zuHause, https://github.com/OpenTTD/OpenTTD/blob/master/src/table/roadveh_movement.h
15:45:41  <_dp_> subtile part of x_pos and y_pos basically
15:46:03  <_dp_> like, in theory, there could be a newgrf that would break if any of those are changed
15:46:21  <Eddi|zuHause> ah, basically the "state machine"
15:46:56  <Eddi|zuHause> i don't see how that would break anything. as long as you make sure that the "steps" don't change in size
15:47:38  <_dp_> Eddi|zuHause, newgrf gets some unexpected values that aren't handled by its switches properly?
15:47:55  <_dp_> so choses wrong sprite or smth
15:49:32  <Eddi|zuHause> _dp_: the closest i see to that is 0x80+0x62, which apparently OpenTTD doesn't even expose
15:49:58  <Eddi|zuHause> oh, it does
15:50:22  <Eddi|zuHause> (scrolled down a bit)
15:57:02  *** tokai has joined #openttd
15:57:02  *** ChanServ sets mode: +v tokai
16:00:13  <Samu> Success!!! https://i.imgur.com/BGocZ8c.png
16:00:24  *** keith__ has quit IRC
16:00:41  <Samu> it is equalised!
16:01:53  <Samu> 592 days for every direction
16:03:59  *** tokai|noir has quit IRC
16:06:01  <Eddi|zuHause> _dp_: i'm not 100% sure of that, but i think i was once told that if it's not explicitly listed in the NewGRF specs, then it's not reliable to use. if you're using something from the TTD vehicle structure in your GRF, you should push to add it to the specs
16:08:20  <Eddi|zuHause> so treat this entire section of the newgrf code as a "best effort" to emulate compatibility. but you can see that some parts are simply omitted already
16:14:31  *** blathijs has quit IRC
16:17:43  <Samu> did someone mentioned roadveh_movement.h?
16:18:39  <Samu> _dp_ which newgrfs are supposed to break if the "state machine" changes?
16:19:03  <Samu> ('cus I changed the state machine)
16:24:14  *** gelignite has joined #openttd
16:30:11  *** blathijs has joined #openttd
16:32:29  <DorpsGek> [OpenTTD/OpenTTD] Eddi-z opened pull request #8608: Codechange: [NewGRF] Make it more explicit which parts of the TTD veh… https://git.io/JtncP
16:33:07  <Eddi|zuHause> should this have been "Cleanup" instead of "Codechange"?
16:35:13  <glx> quickly looking at the diff, I wonder why would a newgrf need access to orders
16:35:44  *** Wormnest has joined #openttd
16:36:42  <LordAro> someone failed to say "no" at some point in the past?
16:36:44  <Eddi|zuHause> dunno, mostly just "it was there before, so why remove it?"
16:39:57  <DorpsGek> [OpenTTD/OpenTTD] Eddi-z updated pull request #8608: Codechange: [NewGRF] Make it more explicit which parts of the TTD veh… https://git.io/JtncP
16:40:35  <DorpsGek> [OpenTTD/OpenTTD] Eddi-z updated pull request #8608: Codechange: [NewGRF] Make it more explicit which parts of the TTD veh… https://git.io/JtncP
17:04:47  <Wuzzy> Hello. I like to talk about OpenSFX.
17:05:16  <Wuzzy> sadly, OpenSFX is under Sampling Plus, which is an ancient legacy non-FOSS license.
17:05:36  <Wuzzy> Today I have continued my work on researching the license status of the source files.
17:06:00  <Wuzzy> I don't know of any1 actually is watching the OpenSFX repo, but I have posted some issues there, and also a PR with a README update
17:10:00  <Eddi|zuHause> yeah, we noticed that the bot isn't actually announcing the OpenSFX repo :)
17:26:22  *** frosch123 has joined #openttd
17:26:23  *** Wuzzy has quit IRC
17:27:13  *** WuzzyBot has joined #openttd
17:27:46  <WuzzyBot> Beep. Bop. I am a bot, and totally not Wuzzy.
17:28:12  <WuzzyBot> Beep bop. Wuzzy, who is totally not me, just posted new issues on the OpenSFX repo: https://github.com/OpenTTD/OpenSFX/issues
17:28:34  <WuzzyBot> Beep bop. Wuzzy, who is absolutely, positively not me, just posted a PR as well: https://github.com/OpenTTD/OpenSFX/pull/5
17:28:46  <WuzzyBot> Beep bop. and bye
17:28:52  *** WuzzyBot has quit IRC
17:29:02  <LordAro> ...
17:29:12  *** Wuzzy has joined #openttd
17:29:19  <LordAro> "the bot"
17:29:23  <LordAro> not "any bot"
17:36:16  <_dp_> there is not a single beta server...
17:36:47  <_dp_> also idk why but new "Search internet" feels weird
17:37:13  <_dp_> * button
17:54:55  <DorpsGek> [OpenTTD/OpenTTD] LordAro commented on pull request #8607: Codechange: Implement a constructor for CurrencySpec (fixes VS2017 compile failure) https://git.io/Jtn4j
18:01:21  *** Gustavo6046 has quit IRC
18:01:57  <DorpsGek> [OpenTTD/OpenTTD] mattkimber commented on pull request #8604: Feature: allow limiting zoom level of NewGRF-provided sprites https://git.io/JtnBC
18:12:43  <DorpsGek> [OpenTTD/OpenTTD] 2TallTyler commented on pull request #8604: Feature: allow limiting zoom level of NewGRF-provided sprites https://git.io/JtnBQ
18:14:06  <DorpsGek> [OpenTTD/OpenTTD] glx22 commented on pull request #8607: Codechange: Implement a constructor for CurrencySpec (fixes VS2017 compile failure) https://git.io/JtnBb
18:15:43  *** Progman has joined #openttd
18:23:26  <DorpsGek> [OpenTTD/OpenTTD] LordAro commented on pull request #8607: Codechange: Implement a constructor for CurrencySpec (fixes VS2017 compile failure) https://git.io/JtnRE
18:29:08  <DorpsGek> [OpenTTD/OpenTTD] glx22 dismissed a review for pull request #8607: Codechange: Implement a constructor for CurrencySpec (fixes VS2017 compile failure) https://git.io/JtnOQ
18:29:11  <DorpsGek> [OpenTTD/OpenTTD] glx22 updated pull request #8607: Codechange: Implement a constructor for CurrencySpec (fixes VS2017 compile failure) https://git.io/JtnOX
18:32:28  <glx> hmm but I'll need to update PR merge requirements
18:37:16  <Wuzzy> Speaking of merge requirements ...
18:37:33  <Wuzzy> I see that https://github.com/OpenTTD/OpenSFX/pull/5 has a "merge blocked"
18:37:57  <glx> yes it needs a review :)
18:40:00  <Wuzzy> it seems like i'm the first one doing anything (issue, PR, literally anything) on the OpenSFX repo for years. am i correct?
18:41:19  *** Speeder has joined #openttd
18:42:24  <DorpsGek> [OpenTTD/OpenTTD] DorpsGek pushed 1 commits to master https://git.io/Jtn0j
18:42:25  <DorpsGek>   - Update: Translations from eints (by translators)
18:45:58  <DorpsGek> [OpenTTD/OpenTTD] Ohrer commented on issue #8601: Female CEOs might get the title of “Businessman” or “Chairman” https://git.io/Jt365
18:50:17  <DorpsGek> [OpenTTD/OpenTTD] SamuXarick opened pull request #8609: Fix: Equalise the number of frames needed for road vehicles to traverse different radius curves https://git.io/JtnEF
18:50:30  <Samu> yes!!! it is done!
18:50:32  <Samu> now dinner
18:50:42  <Samu> oh, i might add the savegame
18:51:03  *** ericnoan has quit IRC
18:52:49  *** andythenorth has joined #openttd
19:03:36  *** Gustavo6046 has joined #openttd
19:12:32  *** ericnoan has joined #openttd
19:22:55  <_dp_> shouldn't language strings have been removed as well here? https://github.com/OpenTTD/OpenTTD/commit/188bf0fbc92b26cbbd5bcd67c5352ab7f6ad769b
19:26:50  <LordAro> _dp_: we let eints do removals now
19:27:05  <_dp_> ah, ok
19:27:13  <LordAro> oh, hang on
19:27:18  <_dp_> it didn't remove them in beta btw
19:27:23  <LordAro> i thought you were talking about non-english strings
19:27:26  <LordAro> yes, probably should have
19:59:18  <Samu> oh noes, i found overtaking vehicles in curves just now
19:59:58  <andythenorth> GG
20:00:16  <Samu> my conversion is thus incomplete
20:00:44  <Samu> time to shift to draft
20:01:40  *** WormnestAndroid has quit IRC
20:01:55  *** WormnestAndroid has joined #openttd
20:09:43  <andythenorth> so can we url2png the WASMs?
20:09:48  <andythenorth> for screenshots of them :P
20:09:55  * andythenorth doesn't know why, just seems cool
20:11:29  *** Gister has joined #openttd
20:13:12  *** Gister has quit IRC
20:20:06  <Samu> wow, what an easy fix
20:20:41  <Samu> the road simply becomes drive on left, which already has a conversion
20:20:51  <Samu> wow!!
20:21:01  <Samu> so much magic
20:32:16  <DorpsGek> [OpenTTD/team] gummipunkt opened issue #133: [de_DE] Translator access request https://git.io/JtnVn
20:32:32  *** frosch123 has quit IRC
20:34:34  <supermop_Home_> nrt now allows multiple town built road types, correct?
20:35:38  <Wolf01> Really?
20:36:57  <supermop_Home_> idk
20:37:16  <supermop_Home_> i thought it chooses one based on fastest or something
20:37:31  <supermop_Home_> and i would need to update unspooled accordingly
20:42:40  *** iSoSyS has joined #openttd
20:47:20  <DorpsGek> [OpenTTD/OpenTTD] SamuXarick updated pull request #8609: Fix: Equalise the number of frames needed for road vehicles to traverse different radius curves https://git.io/JtnEF
20:49:22  <Samu> ready for review again
20:57:31  <DorpsGek> [OpenTTD/OpenTTD] nielsmh commented on pull request #8604: Feature: allow limiting zoom level of NewGRF-provided sprites https://git.io/JtnwR
21:04:39  <supermop_Home_> i guess if such a feature was not added i don't really have to update it
21:04:54  <supermop_Home_> other that make it look prettier
21:14:06  *** gelignite has quit IRC
21:23:08  <Samu> overdose of information on that PR...
21:23:15  <Samu> hope TrueBrain likes it
21:31:08  <Samu> https://github.com/OpenTTD/OpenTTD/pull/8609/checks?check_run_id=1764874676#step:6:600 what is this warning :(
21:32:31  <Samu> where is the parenthesis?
21:33:44  <Samu> ah, i see
21:36:06  <DorpsGek> [OpenTTD/OpenTTD] SamuXarick updated pull request #8609: Fix: Equalise the number of frames needed for road vehicles to traverse different radius curves https://git.io/JtnEF
21:36:47  *** dvim has quit IRC
21:37:11  *** dvim has joined #openttd
21:38:01  *** reldred has quit IRC
21:39:12  *** reldred has joined #openttd
21:48:44  *** Gustavo6046 has quit IRC
21:49:21  <DorpsGek> [OpenTTD/OpenTTD] perezdidac commented on pull request #8597: Fix: Prevent showing company is in trouble again after about 260 months due to overflow https://git.io/JtnK3
21:49:44  *** Gustavo6046 has joined #openttd
21:52:22  <_dp_> lol, just found out there is misc.rightclick_emulate setting that kills all the ctrl magic xD
21:55:23  <_dp_> also, any particular reason openttd doesn't use middle mouse button?
21:56:05  <TrueBrain> 1994? :)
21:56:16  *** Samu has quit IRC
21:56:53  <_dp_> TrueBrain, I though openttd was a bit more modern than that :p
21:57:09  <TrueBrain> you asked for any reason
21:57:10  <TrueBrain> :)
21:57:41  <_dp_> also I bet apple doesn't have middle button
21:57:53  <_dp_> only middle finger :/
22:04:38  <supermop_Home_> maybe 1915 is too early for asphalt road with stripes
22:12:24  <DorpsGek> [OpenTTD/OpenTTD] embeddedt commented on pull request #8480: Multitile depots https://git.io/Jtn6Q
22:18:50  <supermop_Home_> ooh roadtyiles have random bits
22:19:13  <glx> for random dirt ;)
22:20:04  <supermop_Home_> i feel like i have to add new candy to my roads
22:20:29  <supermop_Home_> other than signs/greeble on catenary poles idk
22:22:48  <supermop_Home_> TOWN_BUILD
22:22:48  <supermop_Home_> Enables the roadtype to be built by towns, picked by highest speed (defaults to ROAD if no speed limits?)
22:24:08  <supermop_Home_> i guess its just whatever is fastest
22:38:01  <_dp_> wtf is this enum %) https://github.com/OpenTTD/OpenTTD/blob/master/src/window.cpp#L2820
22:38:34  <TrueBrain> old-style constants :)
22:38:39  *** Wuzzy has quit IRC
22:40:44  <supermop_Home_> would be neat if there was a number like the sort order or whatever
22:41:31  <supermop_Home_> not entirely sure why id want the preferred town road to be slower,
22:42:00  <andythenorth> _dp_ gesture support or go home :P
22:42:54  <supermop_Home_> would fund road reconstruction upgrade roads?
22:44:18  <supermop_Home_> i wonder if i can abuse the town zone variable to make town streets in zone 4-5 lower speed limit
22:44:50  <supermop_Home_> i got to draw some less crappy sidewalks
22:49:48  *** andythenorth has quit IRC
22:53:03  <Eddi|zuHause> supermop_Home_: that sounds like a job for a gamescript
22:59:07  *** jottyfan has joined #openttd
23:01:55  *** sla_ro|master has quit IRC
23:03:54  <supermop_Home_> Eddi|zuHause but the spec is tempting me!
23:03:55  *** Progman has quit IRC
23:07:09  *** innocenat__ has joined #openttd
23:09:17  *** innocenat_ has quit IRC
23:09:17  *** innocenat__ is now known as innocenat_
23:31:50  <supermop_Home_> can roads have fences?
23:36:42  *** HerzogDeXtEr has quit IRC
23:40:00  *** snail_UES_ has quit IRC
23:43:06  *** Wolf01 has quit IRC
23:43:54  *** nielsm has quit IRC
23:47:28  <DorpsGek> [OpenTTD/OpenTTD] mattkimber updated pull request #8604: Feature: allow limiting zoom level of NewGRF-provided sprites https://git.io/JtsF7
23:49:12  <DorpsGek> [OpenTTD/OpenTTD] mattkimber updated pull request #8604: Feature: allow limiting zoom level of NewGRF-provided sprites https://git.io/JtsF7
23:52:03  <DorpsGek> [OpenTTD/OpenTTD] mattkimber commented on pull request #8604: Feature: allow limiting zoom level of NewGRF-provided sprites https://git.io/JtnMI
23:53:11  <DorpsGek> [OpenTTD/OpenTTD] mattkimber commented on pull request #8604: Feature: allow limiting zoom level of NewGRF-provided sprites https://git.io/JtnMt
23:56:52  <DorpsGek> [OpenTTD/OpenTTD] mattkimber updated pull request #8604: Feature: allow limiting zoom level of NewGRF-provided sprites https://git.io/JtsF7

Powered by YARRSTE version: svn-trunk