IRC logs for #openttd on OFTC at 2021-08-30
⏴ go to previous day
00:19:45 *** ChanServ sets mode: +v tokai
00:26:40 *** tokai|noir has quit IRC (Ping timeout: 480 seconds)
00:55:22 <DorpsGek> [OpenTTD/OpenTTD] James103 commented on issue #9523: [Bug]: "ship is lost" is only reported via news https://git.io/JEimw
02:38:41 *** Tirili has quit IRC (Quit: Leaving)
02:39:06 *** debdog has quit IRC (Ping timeout: 480 seconds)
02:56:49 *** Wormnest has quit IRC (Quit: Leaving)
06:08:06 *** sla_ro|master has joined #openttd
06:27:30 *** andythenorth has joined #openttd
07:17:41 <andythenorth> today: adventures in...something OpenTTD
07:17:49 <andythenorth> hopefully not adventures in solitaire
07:18:56 <TrueBrain> I would say: just install solitaire
07:19:02 <TrueBrain> so Microsoft is to blame :P
07:19:46 <andythenorth> I chose to install if from app store
07:25:53 * andythenorth now reading spider solitaire theory blogs
07:26:05 <andythenorth> there's a whole class of analysis about it
07:33:42 <TrueBrain> stop doing that! :P
07:47:18 <andythenorth> that bot isn't working so well
07:54:16 * andythenorth closes solitaire
07:55:50 <andythenorth> right, what am I doing?
07:56:00 <andythenorth> automate publishing of grf docs via AWS and GHA?
07:56:08 <andythenorth> or making FIRS compile a GS from python?
07:56:13 <andythenorth> or painting snow?
07:58:30 <TrueBrain> I dunno, roll a dice! :P
07:58:35 <TrueBrain> I am also rather unsure what to do with this day
08:13:01 *** jottyfan has joined #openttd
08:16:28 *** WormnestAndroid has quit IRC (Remote host closed the connection)
08:16:41 *** WormnestAndroid has joined #openttd
08:18:35 <andythenorth> let's make a GS in the FIRS compile
08:18:46 <andythenorth> probably just template a constants.nut or something
08:18:54 * andythenorth doesn't know how squirrel uses .nut
08:22:53 * andythenorth reading about superlib
08:27:34 <andythenorth> hmm obligations :P
08:27:52 *** andythenorth has quit IRC (Quit: andythenorth)
08:54:49 *** tokai|noir has joined #openttd
08:54:49 *** ChanServ sets mode: +v tokai|noir
09:01:40 *** tokai has quit IRC (Ping timeout: 480 seconds)
09:38:40 *** D-HUND is now known as debdog
09:46:56 *** ChanServ sets mode: +v tokai
09:48:31 *** andythenorth has joined #openttd
09:48:54 <andythenorth> ok so how is a GS?
09:49:07 <andythenorth> main.nut is the entry point?
09:50:25 <andythenorth> AI docs seem to explain
09:53:51 *** tokai|noir has quit IRC (Ping timeout: 480 seconds)
10:03:13 <TrueBrain> maybe explain what you want to do, instead of how to do it :D
10:04:19 <andythenorth> but I did not know it existed until someone (gl* maybe) gave me the link
10:04:54 <andythenorth> you might say, 'andythenorth wtf? how do you not know that page existed, read the wiki edit history it's all andythenorth\
10:05:12 <andythenorth> but these are the things
10:06:41 <andythenorth> I reorganised all the development section of the wiki, but I didn't read it :P
10:07:11 <TrueBrain> if we want such link, best to do that via Doxygen, but I don't know what is possible there
10:07:37 <andythenorth> I am learning what Doxyfile_GS.in is
10:10:25 * andythenorth looks for local docs output
10:10:37 <TrueBrain> I believe you can influence the header or something
10:11:18 <TrueBrain> no clue if it is a good idea btw, to link to there from the API docs
10:11:40 <andythenorth> but it's something to learn :)
10:11:58 <andythenorth> let's see if compiling.MD tells how to actually build docs
10:16:44 <andythenorth> `make package` seems to do something
10:17:55 <andythenorth> I can see that call doxygen, but can't find the output :)
10:47:29 *** roadt__ has quit IRC (Ping timeout: 480 seconds)
10:47:54 <andythenorth> `make all` maybe? :P
10:48:02 * andythenorth reading a lot of makefiles :P
10:54:39 <TrueBrain> did you try 'make docs'? I would expect someone made that entry ;)
10:56:30 <glx> When you don't know just check workflows
10:57:01 <glx> If a target exist it's most likely because a workflow needed it
11:01:39 <andythenorth> I did try `make docs`
11:01:47 * andythenorth looks in workflows
11:01:53 <andythenorth> this is better than solitaire
11:07:22 * andythenorth finds -DOPTION_DOCS_ONLY=ON
11:07:38 <andythenorth> ha I don't have doxygen :D
11:09:12 <andythenorth> definitely more fun than solitaire
11:14:04 <andythenorth> docs take a while to compile eh :)
11:14:14 * andythenorth reassess what 'slow' means compared to grf
11:14:48 <andythenorth> FIRS docs are 2s, I spent time trying to optimise that :P
11:15:59 *** WormnestAndroid has quit IRC (Remote host closed the connection)
11:18:05 *** WormnestAndroid has joined #openttd
11:19:56 *** virtualrandomnumber has joined #openttd
11:20:45 *** virtualrandomnumber has quit IRC ()
11:28:12 <andythenorth> hmm I've taught doxygen about a main page
11:28:28 * andythenorth out of depth here, but what's new?
11:28:32 <andythenorth> swimming in the deep end :P
11:29:28 <andythenorth> but I had to comment
11:29:29 <andythenorth> #INPUT_FILTER = "./doxygen_filter.sh GS"
11:29:32 <andythenorth> which creates other problems
11:40:00 *** roadt_ has quit IRC (Ping timeout: 480 seconds)
11:40:17 <andythenorth> well that was fun
11:55:59 *** Etua has quit IRC (Quit: Etua)
12:01:39 <FLHerne> andythenorth: neat, though
12:02:22 <FLHerne> Yes, GS docs could definitely use more "how to use this" documentation in addition to the per-class API description
12:02:30 <FLHerne> whether that's in-line or links to the wiki
12:08:40 <andythenorth> I don't know if most people even use the NoGo docs
12:08:48 <andythenorth> but then again...there are almost no GS
12:08:53 <andythenorth> so hard to figure out :)
12:32:02 *** jottyfan has joined #openttd
12:33:10 *** glx has quit IRC (Quit: Quit)
12:51:27 *** jottyfan has quit IRC (Quit: jottyfan)
13:22:09 <andythenorth> is it officially GS, gamescript, or script?
13:22:14 * andythenorth needs to name some FIRS stuff
13:22:22 <andythenorth> OpenTTD seems to use both script and GS
13:33:44 <TrueBrain> I am trying to import a lot of historical data of how my house has been doing ... Influx doesn't like me very much :D
13:37:27 <Samu> IM Fixing 9521, with some ugly code repetition, let's see
13:46:32 * andythenorth generates a GS from FIRS compile
13:56:30 <andythenorth> .nutt for nut templates?
13:59:59 *** Wormnest has joined #openttd
14:17:47 *** gelignite has joined #openttd
14:25:55 <andythenorth> compiling a GS from a grf compile is...interesting
14:26:05 <andythenorth> both need lang dirs compiled
14:26:10 <andythenorth> my script names get confusing
14:31:17 <andythenorth> ok I need to replicate FIRS grf structure inside GS next
14:31:28 <andythenorth> some tree of all industries, production algorithms etc
14:31:35 <andythenorth> just templating probably?
15:23:11 <DorpsGek> [OpenTTD/OpenTTD] SamuXarick opened pull request #9524: Fix #9521: Don't load at just removed docks that were part of a multi-dock station https://git.io/JEMNJ
15:25:33 <debdog> Is it just me or is the grafics on master very juddery compared to 1.11.2? not just the vehicle's movements but also dragging windows and such. might only occur with 'many' vehicles on the map. here is a savegame to comapre the two versions: https://beggabaur.rocks/daten/OTTD_1.11.2.sav
15:28:40 *** jottyfan has joined #openttd
15:29:15 <TrueBrain> debdog: betas are not the same as full releases .. betas have all kind of additional verifications active on runtime to ensure we didn't create a regression etc
15:29:25 <TrueBrain> In result, it is fully expected them to be slower
15:30:06 <TrueBrain> Especially noticeable when you have many vehicles active
15:30:26 <TrueBrain> Beta -> master, same thing btw
15:30:52 <debdog> TrueBrain: nothing wrong with that. but my question, reading between the lines, was whether this is know and been worked on or can I expect this behaviour even after the next release
15:33:57 <debdog> I've updeted to master so I can give your PR #9522 a try. the juddery grafic immediately cought my attention, it was so obvious
15:35:33 <debdog> ... and if it isn't known I'd be volunteering to track the commit down which caused it. (now that am I am familiar with git bisect :) )
15:36:19 <TrueBrain> not sure where it goes wrong in communication, so let me be more blunt: yes, master (and especially debug builds) are slower than full releases
15:37:10 <TrueBrain> not sure how that got lost in translation exactly tbh :P
15:37:30 <TrueBrain> build a full release (Release + no-asserts) to compare them
15:37:36 <TrueBrain> you will see they match up ;)
15:38:26 <debdog> so maybe -DCMAKE_BUILD_TYPE=Release would be fine for me if I just intend to test some stuff?
15:38:44 <TrueBrain> RelWithDebInfo, I would suggest, and yeah, that is faster than Debug at least
15:38:49 <TrueBrain> still slower than a full release
15:39:08 <andythenorth> can I declare arbitrary classes (with methods and properties) in GS, or is it lists of lists all the way down? :)
15:39:09 <TrueBrain> basically, nothing to worry about
15:39:11 * andythenorth looks in other GS
15:40:11 <andythenorth> BusyBee seems to have arbitrary classes
15:40:15 * andythenorth feels like such a noob
15:43:27 <debdog> oh, CMAKE_BUILD_TYPE is empty by default
15:59:46 *** Progman has joined #openttd
16:03:26 <FLHerne> (tl;dr git.io will give you a shortened version of any github.com URL if you ask it)
16:04:33 <Samu> you have the right to reject #9524 on the conditions that I am assuming tile3 is the docking tile. Im trying to figure out how a docking tile is marked
16:06:16 <Samu> there's only a UpdateStationDockingTiles(st);
16:14:04 <andythenorth> squirrel `compilestring` looks..interesting
16:14:14 <andythenorth> seems to be `eval` equivalent
16:14:17 <andythenorth> let's not use that
16:16:54 <andythenorth> super, I can now template squirrel from python
16:23:00 <Samu_> well, IsShipDestinationTile also does the IsValidDockingDirectionForDock check that is also present on UpdateStationDockingTiles
16:24:14 <Samu_> so, my assumption is for now, valid. If one day it changes, then there's trouble. I dunno how to make my fix future proof
16:27:21 *** Samu has quit IRC (Ping timeout: 480 seconds)
16:28:41 <Samu_> on placing a dock, a rectangle around the dock is searched for a "possible docking tile", a 3x4 or 4x3 area where the inner tiles are the dock
16:29:27 <Samu_> someone already anticipated the future might have 1 dock with multiple docking tiles around it
16:30:40 <andythenorth> squirrel seems to have some root table thing
16:30:58 <andythenorth> when I create constants I have to use <- to put them in slots
16:32:04 <andythenorth> "maybe nmlc could also create a GS, for those things that grf can't do"
16:32:08 <andythenorth> said nobody sane
16:35:35 *** WormnestAndroid has quit IRC (Ping timeout: 480 seconds)
16:35:42 *** WormnestAndroid has joined #openttd
16:42:37 *** HerzogDeXtEr has joined #openttd
16:43:43 *** WormnestAndroid has quit IRC (Ping timeout: 480 seconds)
16:44:12 *** WormnestAndroid has joined #openttd
16:45:33 <andythenorth> can't find string codes for AI/GS
16:45:37 <andythenorth> guess I read src
16:50:35 <andythenorth> not sure how to display a literal value "4.4.0-62-gab15ba62"
16:50:50 <andythenorth> it's passed to the string as a param
16:53:11 <andythenorth> probably have to provide that literal in the lang files?
16:54:26 <andythenorth> using {NUM} does get a parsed result, which is bizarre, I expected a crash of some kind
16:54:33 <andythenorth> it's not a useful result :P
17:27:08 *** jottyfan has quit IRC (Quit: jottyfan)
18:10:03 *** frosch123 has joined #openttd
18:11:49 <andythenorth> should I be using GSList?
18:11:55 <andythenorth> or squirrel tables?
18:12:15 * andythenorth failing to understand squirrel data structures
18:13:01 <nielsm> AFAIK the GSList is just an interface used for the game to supply data to the script
18:14:48 <nielsm> if you want to support savegames properly and restore your state, maintaining all runtime data (that can't be easily and quickly refetched/recalculated) in a structure that can be fed as-in to the save-load routines is the optimal
18:14:56 *** Flygon has quit IRC (Quit: A toaster's basically a soldering iron designed to toast bread)
18:15:05 <andythenorth> I am hoping I can create some tree
18:15:14 <nielsm> you basically don't want to be doing any transformations or expensive collection of data for saving or loading
18:15:45 <andythenorth> firs.industries[industry_key].accepted_cargos
18:16:04 <andythenorth> I'm puzzled about tables vs objects
18:16:16 <andythenorth> I don't want to just project python onto squirrel, in case I'm wrong
18:16:28 <andythenorth> table seems dict-like
18:16:51 <frosch123> TrueBrain: is github's change of authentication token format important for us?
18:16:58 <nielsm> yeah there's an invisible (hash) in front of table
18:17:04 <frosch123> i think noone is reading info@ anymore :p
18:17:08 <TrueBrain> frosch123: no, but I should fix it some day .. just a shitty job
18:17:30 <andythenorth> not sure why this doesn't construct a table
18:17:31 <andythenorth> INDUSTRIES <- {"foo":"bar"};
18:18:11 <nielsm> but tbh I've never done enough work in squirrel to have real experience
18:18:54 <andythenorth> I don't really understand creating entities
18:18:59 <andythenorth> there's some kind of root table
18:19:12 <andythenorth> I assume I have to use that to refer to entities across modules
18:19:37 <andythenorth> GS, it's really for proper programmers eh :)
18:20:24 <andythenorth> INDUSTRIES <- {}; works
18:20:45 <andythenorth> the docs are mostly squirrel 3 I think, we have squirrel 2 I believe
18:22:46 <nielsm> (the PDF version is readable, the HTML version has terrible styling)
18:23:32 <andythenorth> why do the docs not use { } for conditional s and loops? But we do?
18:24:08 <nielsm> because they mostly have single-statement bodies
18:24:17 <nielsm> and then you don't need to create a block
18:25:05 <andythenorth> hmm foreach has odd quirk on tables
18:25:35 <andythenorth> have to unpack the values in the foreach(a, b in table)
18:25:44 <andythenorth> otherwise it does a weird auto-increment thing
18:27:32 * andythenorth wonders if global names should be all caps
18:27:38 <andythenorth> or if that's just constants
18:27:45 <andythenorth> probably just constants
18:28:07 <andythenorth> is a singleton immutable object created at init a constant?
18:28:34 <nielsm> a singleton is just a fancy name for a global
18:28:43 <nielsm> it's a global variable wrapped as an object
18:29:06 <andythenorth> so "firs" or "FIRS" :P
18:29:25 <andythenorth> in python it would probably be a module imported, and it would be firs
18:30:36 <nielsm> I'd suggest doing something like this, perhaps:
18:32:41 <nielsm> I wanted to suggest making a FirsManager class as a singleton that holds the data
18:36:02 <andythenorth> I would like to be able to do dotted access to props
18:36:09 <andythenorth> not always by key
18:37:14 <nielsm> sometable.foo and sometable["foo"] should be referencing the same thing
18:38:53 <andythenorth> it does thanks :)
18:43:05 <andythenorth> hmm this is kinda fun
18:46:55 <andythenorth> lol if the compile fails I have to restart OpenTTD
18:47:04 <andythenorth> as the makefile deletes the GS prior to installing
18:47:33 <andythenorth> there's no 'rescan' :)
18:51:23 <DorpsGek> - Update: Translations from eints (by translators)
18:51:42 <_aD> Citizens celebrate as first compiler arrives at andythenorthville! (for the 500th time)
18:59:33 *** jottyfan has joined #openttd
19:05:52 <andythenorth> you should see how often I'm doing 'newgame' :(
19:07:04 <andythenorth> 'constant too long'
19:07:09 <andythenorth> have I exceeded memory?
19:08:55 <frosch123> unclosed string constant
19:12:36 <andythenorth> hmm I have to do `'"' + foo '"'` for injecting strings to squirrel
19:12:41 <andythenorth> but black is going to reformat those
19:13:17 <andythenorth> that's interesting
19:13:27 <frosch123> are you using black to format sq code?
19:13:56 <glx> he generates nut with python
19:14:14 <frosch123> just what i expected :)
19:15:38 <andythenorth> I am copying the firs logic into GS
19:16:07 <andythenorth> I could do a json parser or something, but it's easier to just template a firs.pynut file
19:17:33 <andythenorth> oof I exceeded the GS log length trivially :D
19:17:51 <Eddi|zuHause> have i said yet (this week) that this path is pure madness?
19:17:59 <frosch123> i think yuo can redirect it to stdout
19:42:43 <Samu_> oh, someone invented Spectator string, nice!
19:43:13 <andythenorth> Eddi|zuHause but why? o_O
19:43:39 <andythenorth> so how does GS versioning work?
19:44:00 <andythenorth> e.g. I will need to release a new GS version when I release the grf
19:44:11 <andythenorth> but if players have the old grf, will they be able to use the old GS?
19:44:46 <andythenorth> grf we handle that with grfid
19:45:01 <andythenorth> looks like we have GetShortName
19:45:21 <glx> not sure short name is actually used
19:45:43 <frosch123> it is meant to work exactly the same as newgrf
19:45:49 <frosch123> but it may be buggy
19:46:10 <andythenorth> guess we find out :)
19:46:27 <andythenorth> do I want point versions? :P
19:46:41 <frosch123> noone worked on it for 8 years :p
19:48:02 <andythenorth> so now we have GSNewGRF I can get versions and so on
19:48:11 <andythenorth> so I can disable the GS if wrong FIRS I think
19:48:20 <andythenorth> can I disabled FIRS if wrong GS?
19:48:46 <glx> and yet you keep asking ;)
19:49:00 <glx> newgrf are handled before script start
19:49:02 <andythenorth> I have a limited memory
19:49:30 <andythenorth> also typing I cannot accurately
19:49:47 <glx> because you need to know newgrf before generating map
19:49:55 <andythenorth> I try to remember
19:49:59 <andythenorth> until next time I forget :)
19:50:03 <glx> while script start is part of map generation
19:50:15 <andythenorth> ok, so `GSNewGRF.GetParameter(grfid, parameter_num)`
19:50:24 *** nielsm has quit IRC (Ping timeout: 480 seconds)
19:50:26 * andythenorth needs to get the FIRS settings
19:50:42 <glx> I don't think settings are supported
19:52:26 <frosch123> you could probably extend action14 to define a string identifier for settings
19:52:45 <frosch123> and then gs could get a table {string-identifier -> setting value}
19:52:57 <andythenorth> there's probably...a wiki spec for that from 2012? :)
19:54:27 <andythenorth> do we think just use the setting num is too unstable?
19:54:39 <andythenorth> FIRS tends to preserve them where it can for savegame reasons
19:54:52 <Samu_> damn savegame upgrades!
19:55:12 <Samu_> gotta update some old prs
20:00:39 <andythenorth> hmm how hard is action 14? :)
20:03:05 <Samu_> funny, a conflict with something that was mine! hilarious
20:03:59 <andythenorth> hmm maybe for now I can brute force detect the FIRS economy
20:04:07 <andythenorth> using an industry cargo map
20:04:31 <andythenorth> oh, the enabled industry ids will be unique combination per economy
20:04:35 <andythenorth> can I just hash those somehow?
20:04:47 * andythenorth hasn't done hashing since about 2004
20:04:58 <andythenorth> flash game high score 'anti cheating protection'
20:05:57 <DorpsGek> [OpenTTD/OpenTTD] SamuXarick updated pull request #8390: Add: [AI/GS] Missing water related functions and objects https://git.io/JLnqJ
20:07:50 <andythenorth> can I compare 2 arrays?
20:08:26 <Samu_> a conflict in landscape_grid.html is so boring to fix
20:09:46 *** gelignite has quit IRC (Ping timeout: 480 seconds)
20:10:06 <andythenorth> has to be done ;)
20:10:18 *** geli has quit IRC (Quit: Stay safe!)
20:18:06 <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
20:18:52 <andythenorth> hmm sorted arrays :( :P
20:19:34 <andythenorth> sorting str(int) is lolz
20:22:12 *** frosch123 has quit IRC (Quit: be yourself, except: if you have the opportunity to be a unicorn, then be a unicorn)
20:25:22 *** jottyfan has quit IRC (Quit: jottyfan)
20:28:17 <Samu_> I have a new goal, now that my AI isn't that intensive on the cpu: (5000 rvs + 5000 ships + 5000 aircraft) * 15 companies
20:29:17 <andythenorth> be handy if squirrel had a repr equivalent to log arrays to console
20:29:23 <andythenorth> maybe it does in superlib somewhere
20:29:51 <DorpsGek> andythenorth: zuu was last seen in #openttd 5 weeks, 6 days, 11 hours, 40 minutes, and 48 seconds ago: <Zuu> anyway I see the point and taking something down or break an API is when you get to know if someone really care about the service provided. :-)
20:33:41 <andythenorth> can't trivially compare 2 arrays in squirrel
20:36:05 *** roadt_ has quit IRC (Ping timeout: 480 seconds)
20:40:10 <andythenorth> lol can't sum an array either
20:40:27 <andythenorth> I should twitch stream this
20:40:39 <andythenorth> increasingly aged man rages at programming language to make train game
20:40:56 <glx> you can sum an array, but you need to write a function to do it :)
20:41:07 <andythenorth> this is so crippled :)
20:41:25 <andythenorth> it looks like a programming language...but it's more like a starter kit
20:41:38 <andythenorth> I guess it needs to be small to be embeddable
20:42:08 <andythenorth> can squirrel even string compare?
20:42:13 <andythenorth> or do I just get instance refs?
20:42:51 <andythenorth> ok at least string compare works
20:47:49 <andythenorth> the GSIndustryList is all the industries in the game
20:48:03 <andythenorth> but the grf uses skip sprites to turn them on or off
20:48:29 * andythenorth trying to detect which FIRS economy it is
20:49:24 <glx> GSIndustryList is existing (=built) industries
20:53:54 <andythenorth> hmm think it remaps the IDs
20:54:09 <glx> it uses openttd internal IDs
20:54:39 <andythenorth> ok that's not going to work then :)
20:55:06 <andythenorth> I could compare the length of 2 arrays, but it risks colliding if 2 economies have same number of industries
20:55:22 <andythenorth> lots of this won't work if the IDs can't be mapped
20:55:28 <andythenorth> I can't map the production logic over
20:55:33 <andythenorth> that's....a thing
20:56:45 <andythenorth> maybe OpenTTD builds lowest ID first from the grf?
20:56:56 <andythenorth> which means they might be mappable in the compile
20:57:26 <glx> you can analyse the industry type (produced and accepted cargos)
20:57:45 <andythenorth> I just can't think of a tidy way to compare
20:57:55 <andythenorth> I will need to iterate all the industries and all will need to match
20:58:07 <andythenorth> maybe a loop I can break on if result is found
20:58:38 * andythenorth never uses loop breaks
20:59:04 <glx> if an industry exists only in one economy it should not be too hard to detect the economy
20:59:48 <andythenorth> might need manual configuration though
21:00:06 <andythenorth> nah, I can do that in python somehow automatically
21:00:13 <andythenorth> assuming that there is always a unique
21:00:23 <andythenorth> sort uniques, take first
21:00:54 <andythenorth> who said this would never work? :P
21:00:58 <andythenorth> was it Eddi|zuHause ?
21:01:37 <Eddi|zuHause> no, i said you can't manage keeping versions in sync
21:01:38 <glx> and with cargo label of accepted/produced it can't be difficult to infer an industry
21:01:58 <Eddi|zuHause> ... across all potential users
21:02:16 <andythenorth> oh yeah, that's never going to work
21:02:49 <andythenorth> but when I delegate industry construction 100% to the GS
21:02:53 <andythenorth> players will get the message :P
21:03:02 <andythenorth> "I installed FIRS grf, why are there no industries"
21:03:43 <Eddi|zuHause> never underestimate the stupidity of users
21:04:07 <Eddi|zuHause> if the default configuration doesn't work, you lose 99% of them
21:04:43 <Eddi|zuHause> which is why people constantly request shipping AIs with openttd
21:04:58 <andythenorth> they can have XIS
21:05:27 <andythenorth> hmm detecting unique industries is beyond my brain tonight
21:05:42 <andythenorth> it's just some check if industry has only one economy
21:05:52 <andythenorth> and a heuristic to prevent that failing somehow
21:06:05 <andythenorth> there might be economies without any unique industry
21:06:30 <andythenorth> guaranteeing uniqueness in arbitrary combinations :|
21:07:57 <andythenorth> how does bayesian probability work?
21:08:00 <andythenorth> maybe I can use that
21:08:16 <andythenorth> "if x industries are matched, it's probably Steeltown"
21:08:25 *** Progman has quit IRC (Remote host closed the connection)
21:12:13 <andythenorth> hmm Port is in 4 out of 5 economies, with unique cargo map
21:12:18 <andythenorth> Wharf is in the other 1
21:13:25 <Eddi|zuHause> andythenorth: you should play regex golf with the industries :)
21:13:40 <andythenorth> I have used many things I don't understand
21:13:47 <andythenorth> but regex is not one of them
21:13:57 <andythenorth> trying to get to 'andythenorth is dead' without learning it
21:22:04 *** virtualrandomnumber has joined #openttd
21:22:27 *** virtualrandomnumber has quit IRC ()
21:25:56 <DorpsGek> [OpenTTD/OpenTTD] LordAro approved pull request #9522: Fix: pathfinders always tried to avoid docking tiles (even if nothing was on them) https://git.io/JEDpM
21:33:50 *** Samu_ has quit IRC (Ping timeout: 480 seconds)
21:37:42 *** andythenorth has quit IRC (Quit: andythenorth)
21:49:41 *** sla_ro|master has quit IRC ()
21:56:51 *** Wolf01 has quit IRC (Quit: Once again the world is quick to bury me.)
22:15:31 *** Gustavo6046_ has joined #openttd
22:17:58 *** Gustavo6046 has quit IRC (Ping timeout: 480 seconds)
22:17:58 *** Gustavo6046_ is now known as Gustavo6046
22:27:33 *** Gustavo6046 has quit IRC (Remote host closed the connection)
22:30:45 *** Gustavo6046 has joined #openttd
22:35:30 *** Gustavo6046_ has joined #openttd
22:38:50 *** Gustavo6046 has quit IRC (Ping timeout: 480 seconds)
22:38:59 *** Gustavo6046_ is now known as Gustavo6046
22:46:10 <Gustavo6046> metro tracks load and unload instantly, right?
22:46:37 <Gustavo6046> if I wanna do a big passenger line where one end feeds from multiple cities around
22:47:03 <Gustavo6046> do the feeder lines use metro and the main line use eletric rail, or the other way around?
22:48:07 <Gustavo6046> probably the former, since loading and unloading fast means better station ratings, right?
22:48:16 <Gustavo6046> and station ratings at the supplying places are what matter most, right?
23:02:03 *** Gustavo6046 has quit IRC (Remote host closed the connection)
23:05:20 *** Gustavo6046 has joined #openttd
continue to next day ⏵
