IRC logs for #openttd on OFTC at 2019-02-02
⏴ go to previous day
00:03:04 <DorpsGek_II> [OpenTTD/OpenTTD] PeterN commented on pull request #7148: Switch saveload versions from literal numbers to enum values. https://git.io/fhSIL
00:03:56 <DorpsGek_II> [OpenTTD/OpenTTD] JGRennison opened issue #7159: Reverse at signal timeouts occur unexpectedly quickly, affects title game https://git.io/fhSIt
00:04:47 <DorpsGek_II> [OpenTTD/OpenTTD] glx22 commented on pull request #7158: Add: Client setting gui.start_spectator https://git.io/fhSIm
00:05:02 <DorpsGek_II> [OpenTTD/OpenTTD] J0anJosep commented on pull request #7148: Switch saveload versions from literal numbers to enum values. https://git.io/fhSIY
00:06:15 <peter1138> Such conversations.
00:11:33 <DorpsGek_II> [OpenTTD/OpenTTD] PeterN commented on issue #7159: Reverse at signal timeouts occur unexpectedly quickly, affects title game https://git.io/fhSIC
00:11:54 <DorpsGek_II> [OpenTTD/OpenTTD] J0anJosep commented on pull request #7148: Switch saveload versions from literal numbers to enum values. https://git.io/fhSIW
00:12:53 <DorpsGek_II> [OpenTTD/OpenTTD] glx22 commented on pull request #7148: Switch saveload versions from literal numbers to enum values. https://git.io/fhSIl
00:19:13 <peter1138> Seems like keeping the number in the SLV enum would kinda defeat the point of it to me :p
00:20:07 <glx> number or not, some conflicts will still be present :)
00:20:17 <DorpsGek_II> [OpenTTD/OpenTTD] J0anJosep commented on issue #7159: Reverse at signal timeouts occur unexpectedly quickly, affects title game https://git.io/fhSIu
00:20:20 <nielsm> what would SLV numbers translate to, previous release version with some suffix on?
00:20:22 <DorpsGek_II> [OpenTTD/OpenTTD] PeterN commented on pull request #7148: Switch saveload versions from literal numbers to enum values. https://git.io/fhSIz
00:20:24 <glx> but with the enum number are optionnal
00:20:54 <peter1138> glx, yeah, that's just it. Switching to the enum at all will conflict with anything that saveloads.
00:21:21 <peter1138> I think he is suggesting to keep using raw numbers until SLV_200.
00:21:30 <peter1138> Which to me is just no!
00:21:44 *** Gustavo6046 has joined #openttd
00:22:12 <glx> yeah if the enum is there, use it for all revisions
00:22:47 <glx> doesn't make sense to use it for 200+ only
00:23:24 <peter1138> It's a little pain now for much benefit later.
00:23:50 <peter1138> And I don't want to keep including the nubmer either
00:23:58 <peter1138> Not for new entries certainly.
00:24:21 <glx> new entries can use PR number
00:24:57 <LordAro> would lose the "obvious" sequential ordering though
00:25:16 <LordAro> same arguments as keeping svn, i imagine :)
00:25:41 <peter1138> e.g. SLV_PR7108_GROUP_LIVERIES?
00:25:55 <peter1138> I don't know how relevant the PR is
00:26:01 <LordAro> PR num + something short about the name would be nice
00:26:22 <LordAro> i think some sort of number is important so you can easily tell at a glance the approximate order of things
00:26:27 <glx> having the PR number can help in case there something wrong unspotted before the merge
00:26:37 <peter1138> LordAro, the order is exactly the order in the enum :-)
00:27:17 <glx> but it can be in the comment
00:27:18 <peter1138> PR doesn't help with order, and including the number means you have to keep updating numbers in your external changes.
00:27:24 <LordAro> well yes, but i can tell that 203 < 205 without looking it up, but i can't tell about SLV_GROUP_LIVERIES < SLV_SOME_OTHER_FEATURE
00:27:41 <peter1138> If the number if in the comment, then that is the only place where the number is.
00:27:54 <LordAro> whereas SLV_PR1234_SOMETHING < SLV_PR2345_ELSE
00:28:07 <peter1138> PRs aren't in order.
00:28:22 <LordAro> i know, but they're approximately in order, just how things naturally work
00:28:31 <peter1138> PR6867 was committed after PR6885.
00:28:35 <peter1138> s/committed/merged/
00:28:42 <peter1138> Therefoore it's not useful.
00:29:04 <LordAro> as long as it's in the doxy comment, i'll be happy
00:29:31 <glx> yes PR in comment as it is now is good enough indeed
00:31:36 <Samu> those STR_NETWORK... strings being used in single player...
00:34:53 <glx> oh reused string happened before
00:35:53 <glx> if translation is not context dependant it's ok
00:37:04 <dwfreed> rename the identifier to not reference multiplayer
00:37:46 <peter1138> Or don't worry about it.
00:37:59 <peter1138> Hmm, remember when we still use numbers in the langfiles?
00:40:02 <LordAro> ok, so they are strings that are network related
00:40:12 <LordAro> unless your name is Samu
00:40:33 <Samu> now they're also in single player !~
00:40:53 <glx> not important, the usage will be the same
00:40:57 <LordAro> only as much as they were before
00:41:59 <DorpsGek_II> [OpenTTD/OpenTTD] JGRennison commented on issue #7159: Reverse at signal timeouts occur unexpectedly quickly, affects title game https://git.io/fhSIF
00:42:20 <Samu> oh, and the tooltip for Join
00:42:47 <glx> btw I think you can work around all the copy/paste stuff using more #ifdef and finetuning their position
00:44:03 <glx> as you add extra _networking checks it should be safe if done properly
00:45:10 <glx> and stop writing unreadable assignation using a lot of ?: operators :)
00:45:32 <glx> especially without parenthesis
00:46:07 <Samu> that is to show or hide the "Join" button
00:46:18 <glx> I don't see how you can even understand what the assignation does
00:47:58 <glx> yes it was already not very clear it seems :)
00:48:11 <glx> but you are making it worse ;)
00:56:32 <nielsm> rule of thumb: every commit should increase code quality and documentation quality
00:57:13 <nielsm> or at the very least not make things worse
00:59:10 <DorpsGek_II> [OpenTTD/OpenTTD] PeterN commented on issue #7159: Reverse at signal timeouts occur unexpectedly quickly, affects title game https://git.io/fhSLT
01:00:35 <DorpsGek_II> [OpenTTD/OpenTTD] PeterN updated pull request #7148: Switch saveload versions from literal numbers to enum values. https://git.io/fhDIU
01:01:53 <DorpsGek_II> [OpenTTD/OpenTTD] PeterN commented on pull request #7148: Switch saveload versions from literal numbers to enum values. https://git.io/fhSLI
01:06:24 <DorpsGek_II> [OpenTTD/OpenTTD] PeterN commented on pull request #7150: Change: Make ships stop in locks to move up/down instead of following the slope. https://git.io/fhSLL
01:07:58 <glx> || _local_company == COMPANY_SPECTATOR is useless
01:08:32 <glx> can't be true without the setting
01:10:05 <glx> and can be simplified with _settings_client.gui.start_spectator && _local_company == COMPANY_SPECTATOR
01:13:14 <Samu> it can be disabled and i'm still being a spectator
01:13:30 <Samu> must keep the join button yet
01:15:19 <Samu> hmm looks like i still have a problem t.t
01:15:24 <glx> but spectator mode should be valid only with the setting
01:16:05 <Samu> if i disable the setting while i'm still a spectator, i can't start a new company, must fix
01:16:58 <glx> I think you should prevent setting disabling if you are a spectator
01:24:29 <Samu> if (!c->is_ai && (_settings_client.gui.start_spectator || _local_company == COMPANY_SPECTATOR)) {
01:24:44 <Samu> it's pointless to show the join greyed out
01:28:03 <Samu> i guess it's also pointless to show it greyed out when i'm already at the company
01:29:47 <Samu> testing if (!c->is_ai && !local && (_settings_client.gui.start_spectator || _local_company == COMPANY_SPECTATOR)) {
01:30:25 <glx> yes just show it when you are spectator, if you are in a company and want to change you'll just need to switch to specator first
01:30:46 <glx> it's not a network game anyway
01:32:17 <Samu> i see, that's what the cheat is for, good idea
01:33:45 <Samu> something like this if (!c->is_ai && !local && _local_company == COMPANY_SPECTATOR) {
01:34:06 <glx> and yes indeed if you start as spectator just use the cheat to join a company
01:34:36 <Samu> what if i dont wanna cheat?
01:34:37 <glx> you just need an option to create a company, but you can forget join
01:35:46 <glx> or start as spectator and create or join a company
01:36:01 <glx> but switching later means cheating
01:36:59 <glx> allow to load a network save and continue to play in your previous company
01:38:16 <Samu> i couldn't go back to spectating mode
01:38:37 <Samu> i could cheat back to spectator mode i guess
01:38:49 <Samu> but then it ticks "you have cheated"
01:39:16 <glx> well you can save, and restart
01:40:22 <glx> if you want to see you AI playing you won't join their company anyway
01:41:07 <glx> and if you want to play you won't need to switch company after join
01:42:05 <Samu> problem is when I actually cheat myself into an AI company and then decide to go back to spectator, oh... right, nevermind, i have already cheated
01:42:37 <Samu> ok, removing spectator line
01:43:42 <glx> of course you can still cheat without being detected using reload
01:46:48 <peter1138> And... who'd care anyway?
01:48:04 <Samu> if (!_networking && _local_company == COMPANY_SPECTATOR && include_spectator && widget == WID_TN_COMPANIES) {
01:48:17 <Samu> *list->Append() = new DropDownListStringItem(STR_NETWORK_COMPANY_LIST_NEW_COMPANY, CTMN_NEW_COMPANY, Company::GetNumItems() >= MAX_COMPANIES);
01:48:24 <Samu> can only create new company, once
01:48:39 <Samu> can't move back to spectator
01:51:08 <glx> the setting is start as spectator, not emulate network in single player mode ;)
01:52:03 <Samu> hmm it's not entirely right yet
01:52:14 <Samu> i can still create a new company if i load a game
01:53:11 <Samu> so i loaded a game, it has a human company
01:53:23 <Samu> i am currently spectator
01:53:25 <glx> as a spectator you can create a new company if you want, but then you're in it and no longer a spectator
01:53:32 <Samu> new company is enabled :| shouldn't be
01:54:30 <glx> unless you want to allow it only if not company exist
01:54:37 <Samu> i could create a 2nd human company this way
01:55:57 <Samu> i have the option to join one of the human companies, or create a new one
01:56:05 <Samu> it's not that bad of an idea, but...
01:56:53 <glx> maybe join a human company or create one if no human company
01:57:46 <glx> being able to create one for new games is required I think
02:06:36 <Samu> nop, it's not here, or i failed
02:07:40 <glx> hmm you are adding it disabled it seems
02:10:21 <Samu> if (!human) *list->Append() = new DropDownListStringItem(STR_NETWORK_COMPANY_LIST_NEW_COMPANY, CTMN_NEW_COMPANY, Company::GetNumItems() >= MAX_COMPANIES);
02:11:58 <Samu> ehm, it looks really weird, with no indication anywhere that I'm a spectator
02:13:03 <glx> except a lot of disabled tools
02:13:22 <Samu> well, as long as the user knows that setting was left on, he should know
02:13:57 <glx> or you could add a message box
02:14:33 <DorpsGek_II> [OpenTTD/OpenTTD] Eddi-z commented on pull request #7148: Switch saveload versions from literal numbers to enum values. https://git.io/fhSLH
02:15:09 <Samu> the message box is already used for failed industries or no engines
02:17:10 <Samu> that bottom right corner of the screen
02:17:27 <Samu> could be saying Spectating
02:17:54 <glx> the money display section ?
02:18:25 <glx> could be nice for network games too
02:19:39 <Samu> working with displayplanes is something that I wanted to understand better
02:20:40 <glx> basic drawing with some check around
02:21:16 <glx> probably just need an else
02:21:44 <Samu> let me search for a suitable string
02:23:50 <Samu> meh, must create new string
02:24:12 <glx> there's STR_NETWORK_TOOLBAR_LIST_SPECTATOR :{BLACK}Spectator
02:24:31 <glx> but black text seems wrong
02:25:34 <glx> I think something like "*** SPECTATOR ***" in red could work
02:31:24 <glx> but size can be adapted around line 123
02:36:06 <Samu> RED is quite hmm... alarming
02:37:48 <Samu> was also condireing displaying it in the middle
02:38:11 <glx> in the middle there are the news
02:38:28 <Samu> yes, but also the name of the company
02:38:46 <glx> ha just replace company name then
02:39:02 <Samu> the two bars are simply emtpy
02:41:15 <Samu> eww... i guess i prefer the right side, the middle is already being used for many things
02:42:21 <glx> but showing spectator at the same place as company name doesn't sound wrong
02:42:35 <glx> and you have plenty of space
02:44:30 <Samu> which color? Silver it seems?
02:45:17 <glx> not yellow nor red at least
02:46:01 <Samu> STR_STATUSBAR_SPECTATOR :{SILVER}- - Spectator - -
02:53:37 <Samu> yeah spectator, I like it
02:59:31 <Samu> hmm the interaction between cheat and spectator
02:59:37 <Samu> may need a bit more thought
03:17:56 *** WWacko1976-work has joined #openttd
03:56:19 *** Thedarkb-X40 has joined #openttd
03:59:06 <Samu> white spectator is maybe better
04:00:56 <Samu> still unsure what to do about cheat_gui.cpp
04:24:09 <DorpsGek_II> [OpenTTD/OpenTTD] SamuXarick updated pull request #7158: Add: Client setting gui.start_spectator https://git.io/fhSk4
04:38:47 <DorpsGek_II> [OpenTTD/OpenTTD] SamuXarick updated pull request #7158: Add: Client setting gui.start_spectator https://git.io/fhSk4
04:43:25 <Samu> still need to change all those comments :|
05:59:07 *** ChanServ sets mode: +v tokai
08:28:05 *** andythenorth has joined #openttd
08:31:02 <peter1138> I'm outta duck eggs now.
08:57:49 <Eddi|zuHause> <andythenorth> love a duck <- that sounds illegal
08:58:12 <andythenorth> eh no PR_RUNNING_WAGON
08:58:40 <andythenorth> wagon running cost base factor
08:58:43 <peter1138> Oh, wagon running costs.
08:58:54 <andythenorth> there is PR_BUILD_VEHICLE_WAGON
08:59:18 <peter1138> Are there even wagon running costs?
08:59:32 <Eddi|zuHause> i'm pretty sure that exists
09:00:43 <Eddi|zuHause> after playing TF for a bit, they have a weird balancing where wagons are insanely expensive
09:01:23 <Wolf01> A wagon could cost a lot more than an engine
09:02:56 <nielsm> I think TF pricing is based on something like power*max_speed + capacity*max_speed
09:03:36 <Wolf01> Another thing which baffles me is a 4 seat "bus" which can carry 16 passengers
09:03:38 <andythenorth> that sounds like all of my yesterday :P
09:03:42 <nielsm> and it's all algorithmmic, if you check its data files every cost is specified as -1
09:03:51 <andythenorth> horse is now algorithmic
09:04:06 <andythenorth> the algorithm is pure evil and will explode every time I add a new vehicle type
09:04:19 <Wolf01> You should use colours in the algorithm too :P
09:04:21 <andythenorth> I've tuned it to give the results I want for this specific set of vehicles :P
09:05:28 <Wolf01> Multiply each spec by the colour component, speed*red, running cost*(1/green)
09:05:38 <peter1138> Squashing rgb makes it easier to rebase, but then it's squashed.
09:07:06 <peter1138> Also the way I pass colour index + rgb around is annoying :/
09:09:38 *** sla_ro|master has joined #openttd
09:15:07 *** Progman has joined #openttd
09:15:34 <peter1138> Hmm, maybe I should prepare for MTB riding.
09:21:43 <andythenorth> I should delete all this faffy cost code
09:26:07 <Pikka> maybe I should go prepare dindins
09:37:55 <andythenorth> so this mouse stuff
09:42:44 <andythenorth> so I need to enable scroll eh
09:42:51 <andythenorth> I never use that because it's weird and broken
09:46:06 <andythenorth> now I need USB-A->USB-C adapter
09:47:48 <andythenorth> what's the expected behaviour of a mouse scroll wheel?
09:48:18 <andythenorth> specific to map scroll
09:50:41 <andythenorth> well the setting says 'Function of scrollwheel: Scroll map'
09:51:02 <andythenorth> this is consistent with Apple maps apparently (I checked)
09:51:08 <andythenorth> but it makes no sense as a user
09:51:34 <Eddi|zuHause> quite possibly the whole feature is broken
09:51:56 <andythenorth> every time I test it I think that :P
09:53:22 <andythenorth> well github have hidden the 'add your review' link
09:53:26 <andythenorth> so I can't review it
09:54:27 <Eddi|zuHause> i found one at the top of the page when i didn't look for it and just went "so what does this button do?"
09:54:44 <andythenorth> it's usually near 'Review Required'
09:55:18 <andythenorth> can someone find it and give me the url?
09:55:42 <andythenorth> I reverse engineered it from another ticket :P
09:57:25 <DorpsGek_II> [OpenTTD/OpenTTD] andythenorth requested changes for pull request #7109: [OSX] Use high-precision scrolling properties for 2D scrolling https://git.io/fhSYn
10:00:08 <DorpsGek_II> [OpenTTD/OpenTTD] J0anJosep commented on pull request #7148: Switch saveload versions from literal numbers to enum values. https://git.io/fhSYc
10:01:01 <andythenorth> I rewrote my comment on 7109 a bit
10:09:34 <DorpsGek_II> [OpenTTD/OpenTTD] nielsmh commented on pull request #7109: [OSX] Use high-precision scrolling properties for 2D scrolling https://git.io/fhSYB
10:10:22 <DorpsGek_II> [OpenTTD/OpenTTD] J0anJosep commented on pull request #7150: Change: Make ships stop in locks to move up/down instead of following the slope. https://git.io/fhSYR
10:29:06 <andythenorth> could we add RUNNING_COST_WAGON?
10:29:14 <andythenorth> it's not necessary, but it's simpler
10:29:33 <andythenorth> currently I am abusing RUNNING_COST_ROADVEH for wagons
10:29:43 <andythenorth> this will confuse base cost mod newgrfs
10:30:25 <andythenorth> alternatively I nerf down RUNNING_COST_STEAM, so it can be used for wagons, but that is also confusing
10:32:01 *** HerzogDeXtEr has joined #openttd
10:34:51 <DorpsGek_II> [OpenTTD/OpenTTD] flitzpiepe closed pull request #7145: TBTR 2.0 (Template-based Train replacement) https://git.io/fhMov
10:34:52 <DorpsGek_II> [OpenTTD/OpenTTD] flitzpiepe updated pull request #7145: TBTR 2.0 (Template-based Train replacement) https://git.io/fhMov
10:40:36 <DorpsGek_II> [OpenTTD/OpenTTD] flitzpiepe commented on pull request #7145: TBTR 2.0 (Template-based Train replacement) https://git.io/fhSY6
10:46:19 <DorpsGek_II> [OpenTTD/OpenTTD] planetmaker commented on pull request #7145: TBTR 2.0 (Template-based Train replacement) https://git.io/fhSY1
11:08:40 <DorpsGek_II> [OpenTTD/OpenTTD] flitzpiepe commented on pull request #7145: TBTR 2.0 (Template-based Train replacement) https://git.io/fhSY7
11:14:55 *** Smedles_ has joined #openttd
11:19:53 <DorpsGek_II> [OpenTTD/OpenTTD] planetmaker commented on pull request #7145: TBTR 2.0 (Template-based Train replacement) https://git.io/fhSYA
11:21:32 <nielsm> y'know what I want? an "intro" loading/splash screen before the main menu, with a nice big logo
11:22:09 <andythenorth> could we also have a title game that doesn't muller my CPU and battery? o_O
11:25:45 <planetmaker> hm... splash ... wasn't there once one?
11:26:08 <planetmaker> or was it an osx-only thing?
11:26:34 <andythenorth> I do get a nice big logo :D
11:26:50 * andythenorth wonders if nielsm is sarcasm :)
11:31:45 <andythenorth> meh, if I nerf down steam run cost to suit wagons, it's too low for engines
11:32:06 <andythenorth> so either I use RV run cost and break base cost mod grfs
11:32:22 <andythenorth> or small wagons have wrong costs
11:32:34 <TrueBrain> meh; you remove 1 mirror from rotation, and the load on the others double :P
11:33:34 <Pikka> andy: why would you use the different cost bases for different traction types?
11:34:07 <Pikka> just use one for locos and one for wagons. using different, almost identical bases for steam, diesel and electric like the original is pointless
11:34:31 <andythenorth> because base cost mod grfs
11:34:49 <andythenorth> that is literally the only reason
11:35:06 <andythenorth> sometimes newgrf is impossible to make nice things :P
11:35:14 <Pikka> why would anyone want different traction types to have orders of magnitude different costs though?
11:35:39 <andythenorth> file under "don't care"?
11:35:57 <andythenorth> I'm already adjusting costs myself anyway
11:36:02 <andythenorth> for electric etc
11:37:03 <Pikka> it makes more sense imo to use, eg, "steam running cost" for all locos, and "diesel running cost" for all wagons. I think that's what I do. Then at least they're all in the train costs. and you can set the base multipliers different to give yourself more granularity for the wagons
11:37:22 <andythenorth> I just copy-paste that :)
11:37:47 <andythenorth> obviously I am never going to use a base cost mod grf with Horse :P
11:37:50 *** Speedy` has joined #openttd
11:38:28 <andythenorth> ever had a bug report about it? o_O
11:39:12 <Pikka> but it feels less likely to give grf-mashers unexpexted results than using the road vehicle cost
11:39:39 <andythenorth> and I have wagons that cost £300 and engines that cost £100000 to run
11:41:27 *** Speedyn has joined #openttd
11:43:28 <Pikka> I have my base multiplier at 11 for loco running costs, and 5 for wagon running costs.
11:45:21 <nielsm> bah, VS profiler is dumb, for some reason it can't seem to find the PDB file so I don't get any browse information in the profiling output? just raw addresses
11:53:51 <andythenorth> Pikka: yeah those number sound similar to mine
11:53:59 <andythenorth> different calculation, same result
11:54:47 <andythenorth> yeah Horse run costs very similar to NARS 2
12:04:56 <nielsm> okay well, I did get confirmed that when ottd is taking 10+ seconds before drawing anything at all, when running a debug build, most of that time is spent md5summing music files
12:17:46 <andythenorth> read the colour of the lead engine, and remap CC for coaches accordingly
12:17:57 <andythenorth> or rely on player setting coach colour to match? :P
12:22:32 <Pikka> if you livery override the coaches, doesn't it autoset the colour too?
12:23:46 <andythenorth> livery override eh?
12:24:02 <andythenorth> never tried it :)
12:24:20 <TrueBrain> so ... who pissed off SilverSurfer? :P
12:24:36 <andythenorth> TrueBrain: you gravedigging old forums?
12:26:38 <andythenorth> how do I livery override?
12:26:44 <andythenorth> I was kind of advised not to :P
12:27:09 <Pikka> I don't know how in nml
12:27:47 <andythenorth> it's an action 0 prop? Or action 3 result?
12:27:58 <Pikka> it's a special type of action 3
12:28:16 <andythenorth> I don't know anything about nml either, I just think in actions
12:34:43 <andythenorth> maybe I can just return same graphics, but it will pick up the cc
12:35:05 <Pikka> yep... the simplest way to use it is to put the MU after the passenger cars, then just hook back into the normal action 2 chain with the override. then the only effect is that it will match colours
12:37:59 * andythenorth will try it after cost nonsense :)
12:38:09 <andythenorth> maybe I should use "wagons add power" instead of CB 36 :P
12:38:49 <Pikka> you definitely should not do that
12:43:47 <andythenorth> livery override is on the engine? :o
12:44:15 <andythenorth> but I have to chain to the wagon action 3?
12:44:45 <Pikka> you don't have to, you could have a completely separate action 2 chain
12:44:56 <Pikka> but yes, that's the easiest way
12:45:11 <Pikka> the override is effectively an alternative action 3 for the wagon, only when it's behind that loco
12:45:12 <andythenorth> fantastically odd :)
12:46:56 <andythenorth> ouch, it requires a different order for the vehicle declarations
12:47:07 <andythenorth> the coach has to come before the engine in the nml file
12:48:18 <andythenorth> I love these type specific livery settings
12:48:58 <andythenorth> I could just declare these to be EMUs :P
12:56:38 <andythenorth> train_flag_mu doesn't work here, nvm
14:11:35 *** oskari89 has joined #openttd
14:15:39 <peter1138> Aww, nobody merged anything.
14:16:56 <peter1138> Oh, that TBTR guy was using master to develop in o_O
14:18:08 <peter1138> andythenorth, don't override colours based on the head.
14:18:34 <peter1138> andythenorth, either set the MU flag and it should work, or just tell the player to put it in a group livery :p
14:18:59 <peter1138> setting colour remap yourself is a bad feature :/
14:19:07 <peter1138> (and will upset my rgb colours patch)
14:21:21 *** synchris has joined #openttd
14:26:24 <andythenorth> MU flag doesn't work
14:28:00 <andythenorth> I am going to add livery_override crap for this case
14:30:38 <Pikka> is it a bug that the MU flag doesn't give wagons the same livery colour? Is that what it's supposed to do?
14:31:58 <Pikka> I know that it was originally added for TTDP, which had MUs as a type you could livery seperately from non-MU locos
14:32:25 <andythenorth> I am doing evil with these wagons, so it might be me
14:32:44 <Pikka> eh, my impression has always been that that flag did nothing in OpenTTD
14:33:58 <peter1138> It worked for NewGRF engines.
14:34:13 <andythenorth> dunno what the game thinks these vehicles are
14:34:40 <andythenorth> ENGINE_CLASS_ELECTRIC, power 0 in buy menu, changes power if attached to right engine
14:35:35 <peter1138> The fact it shows as electric in your screenshot means the MU flag isn't set
14:36:00 <andythenorth> nah that's prior to applying flag
14:36:43 *** frosch123 has joined #openttd
14:36:46 <Pikka> I missed that peter had added MUs to the livery scheme :)
14:36:59 <peter1138> I hadn't. They were always there.
14:37:05 <peter1138> However, you are right, it doesn't work properly.
14:37:30 <Pikka> do they only show up if there's a vehicle with the flag set?
14:37:31 <LordAro> peter1138: how was cycling?
14:38:34 <peter1138> Pikka, that was the plan, it was kinda broken though.
14:39:29 *** Thedarkb2-T60 has joined #openttd
14:40:04 <peter1138> If you haven't set the MU flag, then it won't show up.
14:40:06 <Samu> oh you talking about wagon running costs?
14:40:12 <peter1138> It's separate from the multi-head flag, btw
14:40:37 <Samu> well, "I had a patch that..."
14:41:27 <Pikka> guess I'll set some MU flags then!
14:49:23 <peter1138> andythenorth, hm, so has it never worked? o_O
14:51:21 <peter1138> Shall I find an old ttdpatch to see what that does? :p
14:52:37 <Pikka> MU flag for livery selection, and livery override for matching cars, sounds like how it should be
14:53:13 <Pikka> it is *possible* that someone might want a loco that's recoloured with the MUs, but doesn't get matching coaches, and the current spec gives that flexibility
14:56:02 <peter1138> Maybe we should just drop them and let everyone use group liveries.
14:56:15 <andythenorth> I am +0.5 to that
14:56:24 <andythenorth> I like being able to colour engines + wagons differently
14:56:58 <peter1138> I did consider having the original set of livery schemes... per group :p
14:57:13 <andythenorth> engine / wagon would be enough :P
14:57:22 <andythenorth> even 4 options is...a lot to look at
14:57:32 <andythenorth> be quite cool though
14:57:49 <andythenorth> also yeah, mu flag definitely works for engines, but not for wagons
14:58:20 <Pikka> do you get an "SMU" option if you set the MU flag on a steam engine?
14:58:46 <peter1138> Also monorail/maglev
14:58:56 <andythenorth> engines and wagons then? :P
14:59:14 <andythenorth> I dunno, it's nice to have colours in buy menu for electric vs diesel etc
14:59:35 * andythenorth invents a CC offset property for vehicles
14:59:39 <andythenorth> it shifts by n from CC
14:59:44 <peter1138> Ballsacks, found a bug :/
15:04:00 <peter1138> Yeah, twitch is shit :(
15:04:45 <peter1138> So yeah that's with the MU flag applying to wagons
15:05:11 <peter1138> blue dmu and yellow emu is how it is mean to be, I'm sure.
15:05:47 <peter1138> I'm just wonder if there is a reason we shouldn't just add SMU. Maybe the others.
15:05:58 <peter1138> What was the other issue?
15:06:25 <peter1138> Cabeese should be freight, right?
15:06:59 <peter1138> What about brake vans on a passenger train?
15:08:04 <peter1138> Ah, it's only messed up if you change the gui size while it's open. How... urgh.
15:09:20 <peter1138> Hmm, you used to be able to place wagons after the last multihead part, heh.
15:09:42 <Pikka> three options for cabeese currently in the hands of set coders: they can have 0 pax cargo and get coach livery, they can have 0 freight cargo and get wagon livery, or they can have an override and get the same livery as the loco
15:10:21 <Pikka> not sure they need any consideration in the spec, especially since they serve no gameplay purpose outside whatever BF newgrf authors dream up
15:10:22 <peter1138> 0 freight cargo gives them wagon? I heard that didn't work.
15:12:17 <Pikka> give me a sec and I'll check
15:12:48 <peter1138> If it does, then yeah that's fine.
15:13:13 <Pikka> you're right, it doesn't seem to work
15:13:23 <Pikka> but if it did, I think that's all they need
15:21:30 <andythenorth> Pikka: you tested with nfo?
15:21:50 <andythenorth> I assumed it was because nml is using cb, not prop, but I didn't investigate further
15:24:56 <Pikka> for the brakevan colour, andy? I just changed the cargo from pax to coal in the nfo props, yes
15:25:44 <andythenorth> not just me then
15:28:49 <andythenorth> we need reference.grf :P
15:28:58 <andythenorth> with all the test cases for all the things
15:29:08 <peter1138> Too many test cases.
15:29:21 <andythenorth> where are all those nml coders when we need them? :P
15:34:15 <Pikka> it's the refittability... I added goods to the "always refit" mask and it works
15:34:47 <Pikka> without it, it reverts to the passenger livery, even if the default cargo is set to goods
15:34:58 <peter1138> Refit shouldn't be related :/
15:35:22 <peter1138> cargo_type = e->GetDefaultCargoType();
15:36:02 <peter1138> And that simply returns the cargo type.
15:37:06 <Pikka> I'll add a "broken" one, send you the nfo and the grf
15:39:41 <Pikka> you're right, the broken one isn't broken... so why did my first test fail
15:39:58 <Eddi|zuHause> reverse heisenbug
15:40:31 <Eddi|zuHause> a bug that only appears if you're looking for it
15:42:17 <Pikka> hmmm... and now my first test seems to be working too. maybe it worked and I didn't reload the newgrf properly :S
15:42:31 <Pikka> maybe cabeese are fine :D
15:44:58 <Samu> looking at the bankrupt code
15:45:21 <peter1138> So why doesn't it work for andy?
15:45:27 <Samu> if i bankrupt in single player and start_spectator is on
15:45:43 <peter1138> And should we add SMU? I don't think it breaks anything at all.
15:45:50 <peter1138> I needs a savegame bump, stupid me.
15:45:57 <peter1138> Let's not do that then.
15:46:17 <peter1138> Samu, you can't bankrupt in single player, afaik.
15:46:49 <Samu> you could now with this spectator stuff
15:46:51 <Pikka> I don't know. I'll send you both the test grf
15:47:15 <andythenorth> then I will fix horse :P
15:47:33 <andythenorth> then I will make depot flip alternate between pax and freight livery
15:48:07 <peter1138> Okay, so ther is another complaint, that mail is "not freight"
15:48:38 <peter1138> That's only from Eddi|zuHause though, heh
15:49:07 <peter1138> Oh, was it someone else?
15:49:25 <Pikka> prop 15 cargo type is the only difference between the two brakevans
15:49:50 <andythenorth> now I have to figure out the why
15:50:06 <andythenorth> or I just leave them as pax? :P
15:50:14 <andythenorth> livery_override? o_O
15:50:18 <andythenorth> match them to engine?
15:51:42 <Eddi|zuHause> i don't think i've even used livery colours for trains in the last 10 years
15:51:55 <peter1138> Sorry, was someone else.
15:52:35 <Pikka> usually you want mail vans matching your passenger trains
15:52:55 <Eddi|zuHause> but mail trucks should match freight?
15:53:03 <Eddi|zuHause> or separate mail
15:53:14 <Eddi|zuHause> or make mail depend on whether there's a passenger wagon?
15:53:31 <peter1138> I wondered about that, but... didn't.
15:53:38 <Pikka> the one corner-argument for doing something here is that sometimes, they're refit-in-stationable to other cargos, and can change liveries if they change cargo. but it's too much faff to make everyone happy.
15:54:22 <peter1138> Oh god don't being up autorefit :/
15:54:31 <peter1138> Maybe they paint them at the station too :D
15:54:39 <peter1138> Might explain why they're always clean.
15:55:13 <Pikka> refit-in-station-which-I-thought-we-didn't-call-autorefit-any-more is the worst feature ever featured ;)
15:55:21 <Eddi|zuHause> or it's like those animals that change colour depending on what they eat
15:55:23 <andythenorth> you don't even want to ask
15:55:34 <andythenorth> about what Horse does with mail van liveries
15:55:48 <andythenorth> also station refit ftw, it's great
15:55:48 <Pikka> what does horse does with mail van liveries, andy?
15:56:02 <andythenorth> if the consist is mostly pax, it shows livery 1 or 2, depending on engine
15:56:10 <andythenorth> player can swap 1 & 2 by depot flip
15:56:28 <andythenorth> if the consist is mostly mail or freight, it shows livery 3 or 4, depending on engine
15:56:34 <andythenorth> again player can swap by flip
15:56:54 <andythenorth> if it's mail, it will also vary sprites, adding in post office cars with windows and stuff
15:57:25 <Eddi|zuHause> andythenorth: nobody besides you will ever use the flip mechanic
15:57:56 <Pikka> scuddles probably would
15:57:59 <andythenorth> I am at least 50% of the target audience
15:58:14 <andythenorth> never make something 'just for the players'
15:58:28 <andythenorth> is scuddles real?
15:58:36 <andythenorth> or is it like fight club, alternative pikka?
15:58:59 <Pikka> in UKRS3, I have a "mail carriage" for passenger-like trains and a "utility van" for grungier mail
15:59:18 <Pikka> I don't know, I haven't talked to him for a while either way
15:59:24 <andythenorth> I have more mail vans than anything else :P
15:59:42 *** Wormnest has joined #openttd
16:00:28 <Pikka> I think that's more than I have total wagons :)
16:01:40 <andythenorth> so what am I doing with brake van livery? o_O
16:01:53 <Samu> it's trying to move me from the bankrupting company to another company, but there is no other, there's the spectator slot, though
16:02:04 <Eddi|zuHause> if you were in a fight club situation, would you even notice?
16:02:08 <Samu> that's where I am supposed to be moved to
16:02:10 <andythenorth> and am I using livery_override for EMUs? Or is it being fixed? :)
16:02:53 <Pikka> use the MU flag to get separate MU livery, and livery_override to get matching coaches?
16:03:13 <Pikka> and brake vans.. I dunno. if you have multiple brake vans, you could do different things with them ;)
16:03:58 <andythenorth> I am hoping peter1138 fixes MU flag for coaches :P
16:05:25 <Eddi|zuHause> so why is youtube suggesting darude-sandstorm to me?
16:05:39 <Pikka> because it knows you so well?
16:06:35 <Pikka> uhoh... I've run into a classic ambiguity. Is the HST a DMU?
16:07:45 <andythenorth> Eddi|zuHause: watch the floppotron version
16:08:22 <Eddi|zuHause> so apparently Darude got nominated for finnish representation at next ESC. if i'm reading these comments right
16:08:39 <Eddi|zuHause> andythenorth: pretty sure i did that
16:08:46 <Samu> what is Backup<CompanyByte> cur_company2(_current_company, FILE_LINE); blabla; cur_company2.Restore();
16:08:56 <Eddi|zuHause> andythenorth: can't remember
16:10:52 <peter1138> Yeah, I think I was always testing it with NewGRFs, hence why DMU/EMU didn't appear by default, and default DMU/EMU don't work.
16:11:17 <peter1138> It was really common for NewGRFs to use livery overrides back then.
16:11:53 <SpComb^> why wouldn't youtube recommend Darude - Sandstorm? It's a perfect fit for all situations
16:13:20 <Eddi|zuHause> SpComb^: in that case, why wouldn't it have suggested it before?
16:13:55 <peter1138> andythenorth, Pikka what's the concensus, passenger / frieght cabeese do work?
16:15:01 <Samu> im failing to bankrupt this without asserting, t.t
16:15:07 <Eddi|zuHause> must always think of "Kapuze" with that word
16:15:13 <andythenorth> my Horse is fail
16:19:24 <peter1138> I've forgotten what I wanted to work on :(
16:20:44 <Samu> i think there's a bug with network messages
16:21:14 <andythenorth> peter1138: buy menu groups
16:21:15 <TrueBrain> @calc 272000 / 20 / 3600
16:21:15 <DorpsGek> TrueBrain: 3.77777777778
16:21:21 <Samu> *** Xarick has joined company #1
16:21:31 <Samu> doesn't disappear for a month already
16:21:38 <peter1138> window tick refactor, eh?
16:22:11 <TrueBrain> @calc 400 / 4 / 60 / 60
16:22:11 <DorpsGek> TrueBrain: 0.0277777777778
16:22:17 <TrueBrain> @calc 400 / 4 / 60 / 60 * 1000
16:22:17 <DorpsGek> TrueBrain: 27.7777777778
16:22:22 <TrueBrain> sorry :P Could hav ePMd him
16:22:40 <Samu> it's been 2 months, it's still displaying it
16:23:13 <andythenorth> boom boom, EMU livery works peter1138
16:23:39 <andythenorth> dunno what I'll do about buy menu
16:24:10 <peter1138> Do you need to do anything?
16:25:19 <andythenorth> "Brennder - Middle" has MU flag set, but is a wagon
16:26:19 <peter1138> MU flag doesn't apply to wagons, no.
16:26:41 <andythenorth> this might be a "won't fix"
16:26:47 <Pikka> does it give the MU livery to any wagons between the heads?
16:27:02 <Pikka> the wagon isn't MU coloured in the buy menu I notice
16:27:07 <peter1138> only passenger wagons.
16:27:21 <peter1138> In the buy menu there is no parent engine
16:27:47 <Pikka> right. so having the MU flag set on the wagon is irrelevent, it's being coloured because it's a passenger wagon in an MU consist
16:27:50 <peter1138> And wagons don't have an engine class, afaik.
16:28:35 <peter1138> Maybe I actually should dig out ttdpatch :/
16:28:40 <DorpsGek> peter1138: lakie was last seen in #openttd 5 years, 18 weeks, 4 days, 23 hours, 50 minutes, and 11 seconds ago: <Lakie> Hehe
16:29:11 <planetmaker> maybe we should just do a sensible thing [TM] :P
16:29:45 <Samu> network messages do not disappear, do i need to create issue?
16:29:55 <peter1138> Samu, I'm looking into it.
16:31:47 <peter1138> Don't have to. I might forget though.
16:32:04 <andythenorth> I think the answer for Horse is 'keep EMU and coach livery same'
16:32:05 <peter1138> Urgh, ottd go stuck.
16:32:19 <andythenorth> this is too edge-case to fix
16:32:26 <andythenorth> enough string already here
16:32:56 <nielsm> should train vehicles have a CB the game can ask for its "livery class"?
16:33:40 <andythenorth> seems like a step too far
16:34:28 <andythenorth> what about steam ships vs. diesel eh? :P
16:34:45 <andythenorth> I was going to suggest small and large aeroplanes, as a troll idea
16:34:47 <andythenorth> but that exists :P
16:35:56 <Samu> that's funny, screenshot doesn't capture the network message
16:36:43 <Pikka> larger aeroplanes, can only use the intergalactic-size airports :D
16:36:46 <peter1138> #ifdef ENABLE_NETWORKING
16:36:49 <peter1138> #ifdef ENABLE_NETWORK
16:37:12 <peter1138> Can we please drop this silly "without network" bollocks?
16:37:20 <nielsm> is there a separate livery option for passenger and non-passenger aircraft?
16:37:47 <LordAro> peter1138: but Win9x!
16:37:57 <peter1138> Win9x has networking?
16:38:44 <DorpsGek_II> [OpenTTD/OpenTTD] SamuXarick opened issue #7161: Network messages do not disappear after a while https://git.io/fhSGi
16:39:15 <Pikka> there is not a separate livery option for non-passenger aircraft
16:39:20 <Pikka> but group liveries will fix all :D
16:40:08 <peter1138> Well, unless everyone finds the bugs in that.
16:40:34 <DorpsGek_II> [OpenTTD/OpenTTD] PeterN opened pull request #7162: Fix #7058, #7161: Network chat messages did not expire. https://git.io/fhSGP
16:41:06 <nielsm> yeah 9x has winsock IP networking, and it's also available for windows 3.1x
16:41:22 <nielsm> you can even get IP networking drivers for DOS, just not much that supports it
16:41:29 <andythenorth> group buy menu liveries :P
16:41:54 <DorpsGek_II> [OpenTTD/OpenTTD] nielsmh approved pull request #7162: Fix #7058, #7161: Network chat messages did not expire. https://git.io/fhSG1
16:42:15 <planetmaker> do we have the network version stuff already approved?
16:42:52 <nielsm> I think everything that needs to be done with that is done now
16:42:58 <nielsm> so ready for review/approve
16:43:00 <Eddi|zuHause> probably should review all the situations of "#ifdef ENABLE_NETWORK" and see if they can be abstracted somehow, so it all ends up in some network.h
16:45:56 <nielsm> something like stubbing out socket functions so they pretend there is networking except it can't connect anywhere?
16:47:08 <peter1138> nielsm, yeah, that'd be better.
16:47:19 <peter1138> It's pretty ridiculous how widespread this is :/
16:51:44 <Samu> what's the difference between _current_company and _local_company and what is Backup doing?
16:52:07 <Samu> until i understand what's this, i'm getting asserts
16:52:33 <Samu> bankrupt code is full of these
16:52:46 <Samu> void ChangeOwnershipOfCompanyItems(Owner old_owner, Owner new_owner)
16:52:52 <peter1138> _local_company is the company the local player is.
16:53:02 <peter1138> _current_company is the company a command is executing as.
16:54:03 <peter1138> Oh, grr, it needs another Fix :(
17:04:15 <planetmaker> why not one style which worked for... decades?
17:05:18 <frosch123> TrueBrain: i see no benefit. keep a single style
17:06:16 <TrueBrain> no benefit? Awh :( Well, I can point you out to a few :D
17:06:26 <TrueBrain> Sementic Releases are used by many projects, and solve a tons of annoying issues
17:06:30 <TrueBrain> like bumping the versions
17:06:30 <frosch123> openttd just switched to the style all the tools use
17:06:32 <TrueBrain> creating the changelogs
17:06:44 <TrueBrain> especially, it automates releases
17:06:52 <TrueBrain> so you don't have to do that yourself, which is kinda annoying as ...
17:07:10 <TrueBrain> additionally, tons of bots out there help you on GitHub with this, if you follow the simple rules put out there
17:07:18 <TrueBrain> so it would really help getting things like a new version of the website online
17:07:23 <TrueBrain> otherwise we have to write our own tooling
17:07:27 <TrueBrain> which is a bit ... counterproductive :)
17:08:58 * andythenorth lost in semver github repos
17:09:15 <Samu> assert(IsLocalCompany()); and asserts :(
17:09:18 <andythenorth> TrueBrain: I am +/-0 as long as the rules are clear
17:10:07 <TrueBrain> main difference with the current style is 'Feature' -> 'feat', I guess
17:10:32 <Samu> funny that it only asserts a whole lot after, in the next StateGameLoop
17:11:01 <milek7> maybe bot rules could be changed?
17:11:23 <milek7> 'By default semantic-release uses Angular Commit Message Conventions. The commit message format can be changed with the preset or config options of the @semantic-release/commit-analyzer and @semantic-release/release-notes-generator plugins.'
17:11:53 <TrueBrain> seems 'fix' and 'feat' are by spec
17:12:01 <TrueBrain> and it uses () instead of [] for categories
17:12:55 <planetmaker> you need the logic for releases anyway. The only difference it will make is: the rules are our own. Or those someone else wrote at some point
17:13:02 <DorpsGek_II> [OpenTTD/OpenTTD] nikolas updated pull request #7086: Change #6173: Update SDL driver to use SDL 2.0 https://git.io/fhamZ
17:13:32 <andythenorth> 'feat' is ugly, it smells of some javascript coders and their stupid variables like 'a' and 'b'
17:13:44 <andythenorth> but give me rules to follow, I'll follow :P
17:13:58 <andythenorth> but think of the bytes saved andythenorth!
17:14:06 <andythenorth> no 'ure' characters
17:14:11 <andythenorth> time saved typing!
17:14:13 <TrueBrain> yeah .. I dont like 'feat' either .. but I am not fancying writing our own release tool too
17:14:18 <andythenorth> but it's ugly, like everything to do with JS :P
17:14:24 <TrueBrain> (again, for non-main-OpenTTD stuff)
17:14:29 <frosch123> switching just creates lots of work: change commit checker, change changelog, break prs, inconsistent with all projects in our subuniverse, ...
17:14:47 <TrueBrain> frosch123: I am explicitly excluding OpenTTD itself :) I mean things like 'website' etc
17:14:51 <TrueBrain> which do not have a commit checker yet
17:14:57 <planetmaker> simple tool: check for "Release" at the start of the commit message. Done
17:15:09 <TrueBrain> planetmaker: I expect that tool tomorrow then :)
17:16:13 <planetmaker> the commit message checker exists, does it?
17:17:08 <TrueBrain> it automates the release process of, for example, the website
17:25:15 <Samu> hmm... network is cheating somehow
17:25:28 <Samu> i am already spectator before the bankrupt company is executed
17:25:32 <TrueBrain> milek7: seems most bots don't fancy an alternative style :D That is sad
17:26:16 <Samu> in single player, i am not executing the bankrupt command with a delay, i'm executing it "right now"
17:26:35 <TrueBrain> but okay, main difference is only 'feature' -> 'feat' and [] -> ()
17:26:45 <TrueBrain> otherwise we already use semantic commit messages, it seems
17:27:38 <peter1138> TrueBrain, yeah, I think we all missed the non-OpenTTD part.
17:27:49 <andythenorth> ship it, see what breaks
17:27:54 <peter1138> But otoh, why change what we have.
17:28:01 <peter1138> I prefer our "Fix" over "fix" anyway.
17:28:17 <TrueBrain> Fix vs fix are both fine, it seems
17:28:24 <planetmaker> proper punctiation and capitalization
17:28:31 <TrueBrain> and yeah, I can understand highlighting doesnt work on IRC :D
17:28:54 <andythenorth> so what was I working on?
17:29:02 <andythenorth> brake van liveries? :P
17:29:18 <peter1138> I'm gonna have turkey burgers tonight.
17:29:33 <peter1138> There is one left over actually, it was a packet of ... 3.
17:29:35 <TrueBrain> (either lasagne or burgers)
17:29:43 <zuzak> why not a turkey burger lasagne
17:30:08 <peter1138> Wouldn't that just be a turkey mince lasagne?
17:30:29 <DorpsGek_II> [OpenTTD/OpenTTD] PeterN updated pull request #7162: Fix #7058, #7161: Network chat messages did not expire. https://git.io/fhSGP
17:30:30 <DorpsGek_II> [OpenTTD/OpenTTD] PeterN dismissed a review for pull request #7162: Fix #7058, #7161: Network chat messages did not expire. https://git.io/fhSG1
17:30:33 <andythenorth> can't remember when Horse is finished
17:30:36 <TrueBrain> okay, now I just wanna try semantic-releases; see what it does for 'website' ...
17:30:49 <TrueBrain> nice thing about GitHub, it has all these apps that do things for you
17:30:51 <andythenorth> there's always revert
17:30:54 <TrueBrain> instead that you have to invent them yourself :)
17:30:58 <peter1138> Try it, as it's not for OpenTTD itself I don't really mind :)
17:31:04 <zuzak> idk peter1138 i was expecting turkey burger in place of the lasagne sheets maybe
17:31:10 <andythenorth> we tried the facebook thing for a while
17:31:12 <TrueBrain> peter1138: I figured most would think that way :P
17:31:14 <andythenorth> 'move fast and break stuff'
17:31:23 <andythenorth> then 'move fast, with stable infra'
17:31:29 <peter1138> zuzak, that... wouldn't be lasagne.
17:31:41 <andythenorth> now we have our own: 'just at least move, and suck up the consequences'
17:31:59 <TrueBrain> I say that often at work: "moving is better than standing still, so go for it"
17:32:14 <peter1138> At some point I need to be brave and defrost that meat at the bottom of the freezer.
17:32:19 <peter1138> It may be another life-form by now.
17:32:29 <TrueBrain> just throw it away ffs
17:32:44 <peter1138> Dogs'll eat it anyway.
17:34:40 <DorpsGek_II> [OpenTTD/OpenTTD] PeterN approved pull request #7151: Change: Allow AI companies to start immediately. https://git.io/fhSZ4
17:34:59 <DorpsGek_II> [OpenTTD/OpenTTD] andythenorth commented on pull request #7160: Fix a couple of livery related issues. https://git.io/fhSZB
17:35:14 <andythenorth> I could approve, but eh C++
17:35:28 <peter1138> Testing is helpful, thank you.
17:35:37 <peter1138> Especially OS X issues ;)
17:37:14 <DorpsGek_II> [OpenTTD/OpenTTD] PeterN merged pull request #7151: Change: Allow AI companies to start immediately. https://git.io/fhyUY
17:37:25 <Samu> yes, it's simulating to be in the old company only to re-put it as spectator
17:37:41 <Samu> because... multiplayer has a delay for do command stuff, bah
17:38:08 <andythenorth> 3rd company colour? o_O
17:38:30 <Samu> in single player i don't need to simulate
17:38:31 <andythenorth> hmm, maybe I can fake a 3rd company colour, with a 'dependent' remap
17:38:32 <peter1138> Also, that'll be 4096 recolourmaps.
17:38:44 <Samu> but i'm still not spectator yet
17:38:53 <andythenorth> I could use recolour sprites depending on CC set
17:38:54 <peter1138> Although actually we have the ability to just pre-calculate recolour maps as needed.
17:39:03 <peter1138> And doing that would, you know, allow us to have more than 16 colours.
17:39:19 <peter1138> Might just break a few NewGRF assumptions though.
17:39:41 <peter1138> I recommend not using custom recolour sprites.
17:39:55 *** techmagus has joined #openttd
17:40:34 <peter1138> Interestingly if you defined a custom recolour map for a 32bpp sprite you could have way more than the 8 shades...
17:40:42 <peter1138> It would go horribly wrong for 8bpp though.
17:40:51 <andythenorth> but the black is painted on, so eh
17:41:13 <peter1138> Just draw it as a variant.
17:41:13 <andythenorth> vehicle groups? o_O
17:41:32 <peter1138> Variant is a better now.
17:41:35 <peter1138> Cos we already have groups.
17:41:55 <andythenorth> variants is actually what I've called them in newgrfs :P
17:42:18 <andythenorth> it's clear that it's variations on a single vehicle then
17:42:21 <andythenorth> and not a group for 'buses'
17:42:23 <peter1138> It also emphasizes it'l... yeah
17:42:25 <andythenorth> or 'BR trains 1990'
17:42:46 <andythenorth> we don't need 'folders' for the buy menu :P
17:43:11 <andythenorth> eh, we should allow buy menu separator grfs :P
17:45:45 <DorpsGek_II> [OpenTTD/OpenTTD] michicc approved pull request #7162: Fix #7058, #7161: Network chat messages did not expire. https://git.io/fhSZr
17:55:07 <andythenorth> also nml 16 cargos? :P
17:57:43 <andythenorth> industries 16 cargo in/out
17:57:47 <andythenorth> doesn't have nml support
17:58:07 <DorpsGek_II> [OpenTTD/OpenTTD] PeterN merged pull request #7162: Fix #7058, #7161: Network chat messages did not expire. https://git.io/fhSGP
17:59:26 <DorpsGek_II> [OpenTTD/OpenTTD] PeterN commented on issue #7161: Network messages do not disappear after a while https://git.io/fhSZ1
18:00:59 *** gelignite has joined #openttd
18:05:12 <peter1138> Urgh, setting up servers is a pain :(
18:05:17 <peter1138> All that command line stuff!
18:05:44 <andythenorth> puts me off 1 hour silly MP games
18:05:51 <andythenorth> 30 minutes of set up :P
18:06:03 <andythenorth> rcon and all that jazz
18:06:35 <peter1138> We should have a GUI "rcon"
18:06:53 <andythenorth> but with 15 not 14
18:07:36 <peter1138> Why am I listening to Les Mis?
18:07:50 <andythenorth> accidental random play?
18:08:06 * andythenorth wonders what works and what doesn't
18:08:47 <andythenorth> frosch123 planetmaker ^^ 16 cargo support, is that going in as one PR, n commits, or multiple PRs?
18:09:12 <andythenorth> it doesn't make sense if it's too split up, but I can't keep track of what's done and what isn't
18:10:11 <peter1138> 40% water, that's a bit better.
18:10:31 <peter1138> 16 cargo in/out support is one new feature.
18:10:35 <planetmaker> 16 cargoes does sound like one commit in nml. Should not be a huge change?
18:10:50 <peter1138> Unless you need infrastructure changes first.
18:10:58 <peter1138> Hmm, I might play a game.
18:11:48 <andythenorth> ouch, I've got merge commits in my branch
18:11:58 <andythenorth> we going to just squash it all down?
18:12:11 <andythenorth> Pikka: I just want to draw little pixel trains :P
18:12:17 <andythenorth> there is all this other stuff to make it possible
18:14:36 <peter1138> Merge commits just to increase a limit to 16?
18:16:14 <andythenorth> looks like I screwed up
18:16:24 <andythenorth> merged upstream instead of rebase
18:16:31 <andythenorth> can't see how to fix that in rebase -i
18:18:13 <peter1138> dbg: [net] Client: Performing emergency save (netsave.sav)
18:18:21 <peter1138> Is that normal on a disconnect?
18:18:33 <glx> drop merge commit, and edit following commits maybe
18:19:59 <peter1138> Weird, when I switch to Firefox, Netflix minimizes.
18:20:07 <peter1138> This putty window, and my VNC, are fine though.
18:20:09 <glx> peter1138: that's a feature I thing
18:20:10 <Gabda> in scenario editor, when I delete a town, CmdDeleteTown gets called twice
18:20:26 <glx> yes, test then exec Gabda
18:21:18 <Gabda> what is the difference? because it goes through the whole function both times
18:21:20 <glx> so in case something fails you are not in half removed state
18:22:10 <glx> second time uses DC_EXEC flags
18:22:30 <peter1138> It's important for network clients as well, of course.
18:22:50 <andythenorth> oh wow, fixing merge commits is like landing on the moon
18:22:59 <peter1138> Client tests, so you get instant feedback, and then sends command to server, where it actually happens.
18:23:07 <andythenorth> it's a flowchart and multiple worfklows
18:23:29 * andythenorth might just abandon 16 cargos for now
18:23:38 <glx> ah, I would just drop the merge and edit others
18:23:41 <peter1138> Is it that complex? :s
18:23:47 <andythenorth> merge commits don't show in rebase
18:23:49 <andythenorth> it's a git feature
18:24:05 <andythenorth> the workflow is to manually rebuild with cherrypick
18:24:30 <peter1138> Hmm, is cherry-pick useful?
18:24:43 <peter1138> Sorry I hadn't read your line.
18:24:50 <peter1138> I meant is it useful for you to use now.
18:24:53 <Gabda> oh, I can see now, there is a DoCommandP as well
18:25:06 <Gabda> these one letter differences in the function names are hard to spot
18:26:17 <peter1138> That isn't the difference between test and EXEC.
18:26:51 <andythenorth> possibly I just make a new branch, cp out what I want and nuke the rest from orbit
18:27:08 <glx> DoCommand and DoCommandP both call the CmdXXX function twice
18:27:18 <andythenorth> or I just diff against master, and commit the diff to a new branch
18:27:33 * andythenorth doesn't enjoy being the nml maintainer :P
18:27:52 <peter1138> So basically, you already added lots of different stuff to your branch?
18:28:10 <peter1138> If it's different, create separate branches for each feature, cherry-pick into them.
18:28:23 <glx> hey one branch per feature is the way to go
18:28:30 <andythenorth> nah it's all related to 16 industry cargos
18:28:34 <andythenorth> it's just fricking big
18:28:41 <peter1138> How can it be so big?
18:28:46 <peter1138> It's just 2/3 -> 16
18:28:47 <andythenorth> oops, have I turned into samu?
18:28:54 <andythenorth> nope, it's a boatload of new properties
18:29:03 <glx> but you can create a new branch and pick each commit of the old branche
18:29:06 <Gabda> ok, now I can see the part with the EXEC flag as well
18:30:05 <andythenorth> can I cherrypick out-of-order? o_O
18:30:22 <andythenorth> I want to group related commits which are currently out of sequence, then squash them
18:30:30 <peter1138> Yup, that's doable.
18:30:35 <andythenorth> this should be about 4 commits by the end
18:31:00 <Gabda> town deletion can be called from two places, I tracked only the first one, and that was with DoCommand (without P), and I couldn't find the double calling there
18:31:44 <peter1138> I do those commits, but I never do merges with master.
18:31:49 <peter1138> You should've learned from NRT!
18:32:08 <glx> yeah only rebase with master :)
18:32:38 <glx> merge is for when you merge another branch in your branch
18:32:49 <andythenorth> in OpenTTD land yes
18:33:07 <peter1138> In other land, you also merge feature branches into master, when they are complete.
18:33:13 <peter1138> But you never merge master into feature branches!
18:33:30 <andythenorth> otherwise bit rot and surprise conflicts
18:33:59 <glx> but rebase is a problem if you share the branch
18:34:05 <Samu> I still can't bankrupt to spectator :(
18:34:13 <andythenorth> rebase is a problem if force push is banned :)
18:34:33 <andythenorth> anyway, nml mess needs un-messed
18:34:37 <glx> Samu: single player, you bankrupt, game over
18:35:09 <andythenorth> there's bankrupty in single player now?
18:35:15 *** Thedarkb-T60 has joined #openttd
18:35:25 <glx> there always was andythenorth
18:36:39 <glx> and in single player mode bankrupt should stay game over, not spectate
18:37:07 <Samu> it doesn't game over, it keeps the company alive
18:37:23 <Samu> i want to move to spectator now if the start_spectator setting is still on
18:40:31 <peter1138> Yeah, SP doesn't ever finish. It's probably a bug but it's been like that for years.
18:40:32 <glx> I really think the setting should only have an effect on game start
18:40:51 <Samu> that description is my problem
18:41:20 <glx> if on, start as spectator, then join an existing human company (or create one if none), then the game works as before
18:43:10 <glx> once started as spectator, you only need to check _networking and COMPANY_SPECTATOR to determine you are in single SP mode and enabling the initial join/create
18:43:31 <glx> after that it's a standard game with no spectator
18:45:14 <peter1138> Bank Balance -£11,744,883
18:45:17 <peter1138> Not bankrupt yet ;(
18:46:18 <Samu> I believe it was done that way for the ability to spectating AIs
18:46:36 <Samu> but now i'm really making a spectator slot
18:47:38 <Samu> no need to keep a company alive
18:51:20 <andythenorth> now just to squash the features :P
18:51:25 <andythenorth> still ain't done
18:51:36 <andythenorth> but this current mess hurts my brain
18:52:22 <TrueBrain> want some ice on it?
18:53:07 <andythenorth> ice on my house right now
18:53:30 * andythenorth advertises for nml maintainer :)
18:53:44 <glx> personnaly I would keep the SP bankrupt as it is
18:56:05 <glx> yes and no move to spectator
18:56:48 <glx> the goal of the setting is start as spectator, not change how SP game works :)
18:57:53 <DorpsGek_II> [OpenTTD/OpenTTD] TrueBrain opened pull request #7163: Add: [AzurePipelines] split the CI in two parts: building and commit checking https://git.io/fhSns
18:58:37 <DorpsGek_II> [OpenTTD/OpenTTD] PeterN updated pull request #7150: Change: Make ships stop in locks to move up/down instead of following the slope. https://git.io/fhDyC
18:59:21 <DorpsGek_II> [OpenTTD/OpenTTD] TrueBrain updated pull request #7163: Add: [AzurePipelines] split the CI in two parts: building and commit checking https://git.io/fhSns
19:00:07 <TrueBrain> peter1138: something like this; would that help?
19:00:34 <glx> so 2 lines in PR status ?
19:01:36 <peter1138> Although it's done the CI first, heh.
19:01:48 <TrueBrain> yeah; also because I manually started this, to show how it looks
19:01:55 <TrueBrain> but agents ran out, so it will be a while :D
19:02:02 <glx> good, it's better to know commit checker failed without having to open the azure log
19:02:20 <peter1138> If it's beneficial, go for it?
19:02:32 <TrueBrain> you were asking for it; so I am checking if this would have helped you :)
19:02:47 <glx> because often you get CI failed when it was just the commit
19:02:51 <peter1138> Hmm, to be honest, having the commit checker locally helps more :)
19:02:58 <TrueBrain> peter1138: fair enough :)
19:03:28 <glx> and sometimes CI failed for connection issues ;)
19:04:47 <glx> and commit checker could also run the scripts and detect missing project updates
19:05:20 <glx> (usually MSVC fails to build in this case)
19:05:59 <glx> is it possible to not start build check if commit checker failed ?
19:10:16 <TrueBrain> glx: not in this setup; that is only possible by moving it into a single build
19:10:33 <TrueBrain> not sure if that fixes anything
19:10:52 <TrueBrain> takes 50 seconds to fetch the sources; lol
19:11:00 <andythenorth> how does a cp have conflicts with files that aren't changed in the commit?
19:11:14 <andythenorth> is the changeset not isolated?
19:11:20 <andythenorth> does it bring in other stuff from the tree?
19:11:21 <TrueBrain> glx: ideally I wish every job would report to GitHub, instead of a single entry .. but .. yeah ..
19:11:59 <DorpsGek_II> [OpenTTD/OpenTTD] TrueBrain commented on pull request #7163: Add: [AzurePipelines] split the CI in two parts: building and commit checking https://git.io/fhSnW
19:12:03 * andythenorth thought this would be possible just tedious :(
19:12:14 <TrueBrain> no clue what you are trying andythenorth :D
19:12:39 <andythenorth> but I think it would be easier to just diff master > big_patch.diff
19:12:48 <TrueBrain> anyway; I put the PR up there, feel free to approve or decline it. I dont have a favour; just wanted to show the other way of doing this :)
19:13:38 <TrueBrain> git checkout 16-industry-cargos-in-out-rebase
19:13:46 <TrueBrain> git cherrypick 239c5990a2322f67c47fdd7e658f8b92835cc8cc
19:13:52 <TrueBrain> if that fails, you are screwed :P
19:14:22 <TrueBrain> I assume with a merge conflict
19:14:26 <TrueBrain> that is just annoying to fix :)
19:14:30 <andythenorth> massive merge conflict
19:14:48 <glx> you can get a diff for each commit I think
19:15:17 <glx> then do a manual cherrypick
19:15:57 <andythenorth> I think it's time for just starting from scratch tbh
19:16:20 <andythenorth> did we include 16 cargos for 1.9.0?
19:16:37 <glx> I think 16 cargo is in master
19:17:44 <andythenorth> is it worth shipping without nml support?
19:18:10 <andythenorth> my concern is that it is pretty much untested
19:18:28 <glx> nml is an independant project anyway ;)
19:18:28 <andythenorth> it introduces quite a lot of new spec
19:18:43 <andythenorth> but nobody knows if the new industry grf spec makes sense
19:18:55 <andythenorth> and once it's out there, it's hard to change it
19:28:04 * andythenorth might be trying to solve a non-problem
19:34:37 <Gabda> anyone interested in trying out the closest town visualization in the scenario editor?
19:35:40 <Gabda> (it is not the purpose of the PR, just a method to test and play with the town voronoi diagram)
19:40:21 <Samu> if I cheated to an AI company, the other part of the code prevents the company i'm currently at, to bankrupt
19:42:50 <Samu> for a network game, the code just before line 299 moves local company to spectator
19:46:30 <glx> ChangeOwnershipOfCompanyItems() is not called as the company is not deleted
19:49:52 <Samu> I'm apparently making that part to run
19:50:53 <Samu> will assert later on IsLocalCompany() in StateGameLoop()
19:51:00 <glx> yes your active company can't disappear
19:51:33 <glx> the command for network games say something about this assert
19:52:53 <glx> "It is done in this way as we are called from the StateGameLoop which can't change the current company, and thus updating the local company triggers an assert later on."
19:58:58 <peter1138> andythenorth, expected outcome is a set of changes that can be reviewed as PR :p
19:59:34 <peter1138> Hmm, I didn't know nml had been moved to us.
19:59:44 <peter1138> Shall I clone it and check out your changes?
19:59:56 <andythenorth> but it's a mess :)
20:00:09 <peter1138> There's a theme here.
20:02:00 <peter1138> So, er, where does the set of changes start?
20:02:55 <peter1138> Codechange: Rename CargotypeListProp... ?
20:03:37 <DorpsGek_II> [OpenTTD/OpenTTD] alexanderweiss commented on pull request #7109: [OSX] Use high-precision scrolling properties for 2D scrolling https://git.io/fhScv
20:03:45 <andythenorth> peter1138: where even is that commit?
20:04:00 <peter1138> b13ede709fc9a1cc0abb1550689e5a531475b1da
20:05:00 <andythenorth> ok everything makes sense now
20:05:18 <andythenorth> I should have been rebasing much further down the tree
20:05:43 <andythenorth> yes that's where it starts
20:05:58 <Samu> I think I have an idea for a "fix"... just set current company equal to local company when it returns
20:06:07 <DorpsGek_II> [OpenTTD/OpenTTD] PeterN commented on pull request #7109: [OSX] Use high-precision scrolling properties for 2D scrolling https://git.io/fhSct
20:06:20 <Samu> unless this breaks something more
20:06:50 <andythenorth> yeah clown shoes
20:06:58 <andythenorth> that's why I had unexpected changesets
20:07:09 <Gabda> at last my PR passed the checks :)
20:07:28 <peter1138> What me to try sorting it out still or are you good now?
20:08:15 <andythenorth> I'm still pecking one key at a time with git rebase
20:08:19 <andythenorth> and I can't use nano properly
20:08:23 <andythenorth> high chance of more mess
20:08:49 <andythenorth> this should come out as 1 docs commit, and about 3 feature commits (props, cbs, codechanges)
20:09:40 <andythenorth> the spec kept...evolving
20:11:56 <DorpsGek_II> [OpenTTD/OpenTTD] alexanderweiss commented on pull request #7109: [OSX] Use high-precision scrolling properties for 2D scrolling https://git.io/fhScO
20:12:15 <peter1138> CargotypeListProp -> ByteListProp seems to be a separate change.
20:12:32 <peter1138> Like a necessary precursor, but not necessarily part of the 16-in-out PR.
20:13:07 <DorpsGek_II> [OpenTTD/OpenTTD] PeterN commented on pull request #7109: [OSX] Use high-precision scrolling properties for 2D scrolling https://git.io/fhScn
20:13:36 <Samu> if (!IsLocalCompany()) SetLocalCompany(_local_company);
20:13:55 <Samu> probably gonna assert because...
20:14:25 <Samu> because SetLocalCompany asserts IsLocalCompany
20:15:05 <Samu> if (!IsLocalCompany()) _current_company = _local_company;
20:16:33 <peter1138> Yeah, I might just cherry pick changes as well.
20:16:52 <andythenorth> peter1138: b13ede709fc9a1cc0abb1550689e5a531475b1da I can probably move to master
20:17:02 <peter1138> andythenorth, it needs changes.
20:18:41 <Gabda> is there an easy and globally reachable method/flag that tells if we are in "build" mode, and if so, what is the selected build (e.g. railway station, bus station, industry, town)?
20:19:22 <Gabda> (the only one I found so far is to get the mouse sprite, but that is...)
20:25:28 <peter1138> Something like _thd
20:25:58 <andythenorth> Samu: less of that
20:31:32 <DorpsGek_II> [OpenTTD/OpenTTD] alexanderweiss commented on pull request #7109: [OSX] Use high-precision scrolling properties for 2D scrolling https://git.io/fhScu
20:36:55 <DorpsGek_II> [OpenTTD/OpenTTD] PeterN commented on pull request #7109: [OSX] Use high-precision scrolling properties for 2D scrolling https://git.io/fhScV
20:38:48 <peter1138> Hmm, git cherry-pick is weird :s
20:39:05 <peter1138> I still end up with a load of other changes.
20:40:09 <peter1138> -m parent-number sounds relevant.
20:41:32 <andythenorth> I thought cherry-pick was isolated
20:41:45 <andythenorth> I use it a lot at work, it's fine
20:42:39 <DorpsGek_II> [OpenTTD/OpenTTD] alexanderweiss commented on pull request #7109: [OSX] Use high-precision scrolling properties for 2D scrolling https://git.io/fhSci
20:47:34 <DorpsGek_II> [OpenTTD/OpenTTD] michicc commented on pull request #7109: [OSX] Use high-precision scrolling properties for 2D scrolling https://git.io/fhScM
20:51:49 <DorpsGek_II> [OpenTTD/OpenTTD] andythenorth commented on pull request #7109: [OSX] Use high-precision scrolling properties for 2D scrolling https://git.io/fhScS
20:54:26 <Samu> this can't be done :| or I don't know how to
20:54:57 <Samu> StateGameLoop wants everything done as OWNER_NONE
20:57:30 <Samu> but how could network mode work?
21:06:28 <peter1138> Ye olde Viglen mod.
21:06:44 <andythenorth> I have BeebEm for free :P
21:33:56 <peter1138> Well that cheese & biscuits was a whole other meal.
21:38:42 <peter1138> Right well it's rebased to master, now.
21:39:31 <peter1138> I haven't done any further squashing.
21:42:02 * peter1138 ponders squashing locks and calling it done.
21:43:35 <peter1138> Any further word on the saveload enum?
21:43:40 * andythenorth ponders squashing more nml commits
21:43:58 <peter1138> andythenorth, yeah, reorder and squash what is relevant, i guess.
21:44:49 <andythenorth> might wait for nielsm to be back :)
21:46:32 <andythenorth> looks like we had a few versions of prop 28
21:49:01 <DorpsGek_II> [OpenTTD/OpenTTD] PeterN updated pull request #7150: Change: Make ships stop in locks to move up/down instead of following the slope. https://git.io/fhDyC
21:50:19 <peter1138> I should fix this ship thing.
22:04:12 <Gabda> peter1138: thanks, _thd contains the window class where the tile highlight was set from, which is mostly enough for me
22:04:30 <peter1138> There may be functions around that already test for this stuff.
22:31:18 <DorpsGek_II> [OpenTTD/OpenTTD] michicc approved pull request #7148: Switch saveload versions from literal numbers to enum values. https://git.io/fhSWV
22:39:10 <DorpsGek_II> [OpenTTD/OpenTTD] PeterN merged pull request #7148: Switch saveload versions from literal numbers to enum values. https://git.io/fhDIU
22:40:14 <Eddi|zuHause> gah, now i actually need to do this rebase...
22:41:54 <peter1138> Time to rebase NRT I think :-)
22:43:43 <Samu> StateGameLoop cannot beat me
22:44:26 <Samu> it does not like bankrupting to spectator
22:47:39 <Samu> i just have no idea if I broke something else though, but at least I can finally bankrupt to Spectator in single player
22:48:01 <Samu> so, I'm gonna have to ask
22:49:21 <Samu> what could possibly go wrong?
22:49:55 <Samu> IncreaseDate is where bankruptcy checks are done
22:51:30 <Samu> i made sure current company and local company are restored to OWNER_NONE for every company bankrupt check
22:52:32 <Samu> when IncreaseDate() gets its job done, the others that follow are now handed OWNER_NONE to them
22:52:43 <Samu> the assert does not trigger
22:53:04 <Samu> line 72 restores the local company
22:53:15 <Samu> which could now no longer exist due to bankruptcy
22:53:31 <Samu> that's why I move to spectator with the checks 76 77 78
22:53:57 <Samu> success?! or did I broke something else?
23:10:52 <peter1138> Hmm, NRT actually doesn't have any saveload changes.
23:11:07 <peter1138> Well, no SLE stuff.
23:11:15 <peter1138> Obviously the map changes :)
23:12:48 <andythenorth> yay I fixed run costs
23:12:57 <andythenorth> at least for this test map with this FIRS cargo rates :P
23:13:02 <andythenorth> for narrow test cases :P
23:13:06 <andythenorth> that's 2 days work :P
23:17:19 <Eddi|zuHause> one day i figure out what andy means with "broken" and "fixed
23:17:20 <DorpsGek_II> [OpenTTD/OpenTTD] PeterN updated pull request #7150: Change: Make ships stop in locks to move up/down instead of following the slope. https://git.io/fhDyC
23:17:43 <andythenorth> Eddi|zuHause: it's some kind of cycle
23:17:48 <peter1138> Hmm, I can remove my group-livery branches now, I guess.
23:17:51 <andythenorth> like that diagram of a snake eating its own tail
23:17:51 <Eddi|zuHause> there seems to be little overlap with my usage of those words
23:27:33 <peter1138> Hmm, I apparently have a patch to make ships go fast.
23:27:56 <peter1138> Why did I open that? Hmm.
23:31:36 <andythenorth> or just for consistency?
23:32:48 <Wolf01> Meh, I started with a robotic arm, now I have factorio
23:33:40 <andythenorth> articulated boats for 3.0
23:52:13 *** Thedarkb-T60 has joined #openttd
continue to next day ⏵
