IRC logs for #openttd on OFTC at 2020-12-25
⏴ go to previous day
00:40:58 <DorpsGek> [OpenTTD/OpenTTD] floodious commented on pull request #7745: Feature: setting for more flexible town spacing https://git.io/JL1Zw
00:54:41 <DorpsGek> [OpenTTD/OpenTTD] floodious commented on pull request #7745: Feature: setting for more flexible town spacing https://git.io/JL1ZH
00:58:23 <LordAro> what an interesting view of the game
00:58:40 <_dp_> yeah, what a surprise, it's not a simcity :p
01:08:55 <_dp_> btw, it's likely possible to write a gs that will grow town like that
01:09:17 <_dp_> after all, it only needs to place roads...
01:09:58 <_dp_> though it wouldn't hurt to have a method to place houses precisely as well.
01:12:28 *** gelignite has quit IRC (Quit: Stay safe! Stay at home! Stop the chain reaction!)
01:54:48 <DorpsGek> [OpenTTD/OpenTTD] Eddi-z commented on pull request #7745: Feature: setting for more flexible town spacing https://git.io/JL1cP
02:07:28 <DorpsGek> [OpenTTD/OpenTTD] Eddi-z commented on pull request #7745: Feature: setting for more flexible town spacing https://git.io/JL1Ct
02:11:05 <DorpsGek> [OpenTTD/OpenTTD] JGRennison commented on pull request #7745: Feature: setting for more flexible town spacing https://git.io/JL1CO
02:13:08 *** Progman has quit IRC (Remote host closed the connection)
02:14:46 <DorpsGek> [OpenTTD/OpenTTD] floodious commented on pull request #7745: Feature: setting for more flexible town spacing https://git.io/JL1Cn
02:15:31 <DorpsGek> [OpenTTD/OpenTTD] Eddi-z commented on pull request #7745: Feature: setting for more flexible town spacing https://git.io/JL1C8
02:19:37 <DorpsGek> [OpenTTD/OpenTTD] floodious commented on pull request #7745: Feature: setting for more flexible town spacing https://git.io/JL1C0
02:32:54 <DorpsGek> [OpenTTD/OpenTTD] floodious commented on pull request #7745: Feature: setting for more flexible town spacing https://git.io/JL1CP
02:34:26 <DorpsGek> [OpenTTD/OpenTTD] Eddi-z commented on pull request #7745: Feature: setting for more flexible town spacing https://git.io/JL1CX
02:35:32 <DorpsGek> [OpenTTD/OpenTTD] floodious commented on pull request #7745: Feature: setting for more flexible town spacing https://git.io/JL1C1
02:36:12 <DorpsGek> [OpenTTD/OpenTTD] Eddi-z commented on pull request #7745: Feature: setting for more flexible town spacing https://git.io/JL1CD
02:38:23 <DorpsGek> [OpenTTD/OpenTTD] floodious commented on pull request #7745: Feature: setting for more flexible town spacing https://git.io/JL1CH
03:24:33 *** heffer has quit IRC (Quit: heffer)
03:33:58 *** debdog has quit IRC (Ping timeout: 480 seconds)
03:36:09 <DorpsGek> [OpenTTD/OpenTTD] floodious commented on pull request #7745: Feature: setting for more flexible town spacing https://git.io/JL1lO
04:34:52 *** WormnestAndroid has quit IRC (Remote host closed the connection)
04:37:24 *** WormnestAndroid has joined #openttd
06:32:36 *** HerzogDeXtEr1 has quit IRC (Read error: Connection reset by peer)
07:17:00 *** andythenorth has joined #openttd
07:30:39 *** andythenorth has quit IRC (Quit: andythenorth)
07:49:13 *** andythenorth has joined #openttd
08:14:47 *** sla_ro|master has joined #openttd
08:36:00 *** andythenorth has quit IRC (Quit: andythenorth)
08:40:52 *** andythenorth has joined #openttd
08:42:30 *** andythenorth has quit IRC ()
08:48:00 *** keith__ has joined #openttd
08:48:20 *** andythenorth has joined #openttd
08:59:59 <andythenorth> lol the M1 is doing 500fps on ffwd
09:00:59 <andythenorth> this is via Rosetta
09:01:11 <andythenorth> the i9 does 300fps
09:13:42 <andythenorth> the i9 actually drops to 150fps after a short period
09:13:53 <andythenorth> until a news message is opened, then it goes back to 300fps for a while
09:14:06 * _dp_ just noticed someone pinged me a year ago ...
09:17:48 * andythenorth FIRS 4 changelog time
09:17:59 <andythenorth> kids are plugged in to new devices, christmas under control
09:54:57 *** WormnestAndroid has quit IRC (Remote host closed the connection)
09:55:10 *** WormnestAndroid has joined #openttd
10:08:41 *** andythenorth has quit IRC (Quit: andythenorth)
10:09:03 *** andythenorth has joined #openttd
10:09:07 <TrueBrain> is it just me, or are some of these replies impossible to read? Like, I see that they are english, and that they make sentence, but I cannot make heads nor tails what it is trying to tell me .. I find it very difficult to read :(
10:09:11 * andythenorth has been watching Blade Runner clips
10:09:24 <andythenorth> someone asked in the stream which actor frosch was
10:09:36 <andythenorth> I am sure it's Rutger Hauer at end of Blade Runner
10:09:52 <TrueBrain> Blade Runner, a true xmas movie :)
10:10:53 <andythenorth> original screening in the UK right now :P
10:13:01 <TrueBrain> easiest ticket evah :P
10:13:05 <LordAro> TrueBrain: floodius definitely uses very verbose english :)
10:13:22 <LordAro> also merry christmas etc
10:13:22 <TrueBrain> it is not just that, it is also full of contradictions
10:13:29 <TrueBrain> I am often confused if he means left or right
10:13:36 <TrueBrain> and I don't mean this in a bad way, as I like understanding users
10:13:44 <TrueBrain> but .. this is very hard for me .. it might be the verbose english btw :P
10:13:51 <TrueBrain> "I would argue that yes, such small improvements are exactly the right solution for OpenTTD unless a specific objection can be made to some part of this implementation. Outright rejection of any improvement whatsoever is also possible."
10:14:06 <TrueBrain> I mean ... it states both arguments in 2 sentences ..
10:14:20 <TrueBrain> I guess a lot of hyperbolic use of english or something?
10:14:26 <TrueBrain> <- totally confused :)
10:15:38 <michi_cc> "You are monsters and want OTTD to die!" (very liberal translation :P)
10:16:09 *** Progman has joined #openttd
10:16:10 <michi_cc> Evil devs don't care about users and always reject :)
10:16:42 <TrueBrain> thank you for that translation :)
10:18:08 <michi_cc> But, yes, comments are not always helpfull :) See that comment by J... on the inflation PR
10:19:29 <TrueBrain> what were we going to do with the inflation PR .. I forgot .. I am very tempted to just approve it :)
10:20:09 <TrueBrain> the idea of the livestream was to have answers on these kinds of things :D
10:23:04 <michi_cc> It has no flashing fancy graphics, you don't get feedback for that.
10:23:37 <TrueBrain> basically, we really need a rating system for devs :)
10:23:48 <TrueBrain> just so I feel less alone when I approve or reject something :P
10:23:56 <TrueBrain> TOGETHER WE STAND STRONG :)
10:24:56 <michi_cc> The only feedback that oocasionally pops up about infrastructure maintenance is that airport maintenance values are completely broken. But apparently completely broken is still not enough for somebody to ever suggest new values for like 5 years or so.
10:25:29 <_dp_> signals are broken, airports are fine :p
10:25:58 <michi_cc> Don't have signal spacing 2 and you are fine :p
10:26:40 <_dp_> not much better and > spacing means less trains :P
10:27:23 <TrueBrain> make smaller trains
10:27:46 <_dp_> TrueBrain, and how's that an improvement
10:27:50 <TrueBrain> I love how we think in solutions :D
10:28:01 <TrueBrain> how should I know, you are playing the game :P
10:28:03 <michi_cc> _dp_: Joking aside, each and every value is broken. I asked for feedback and suggestions multiple times before and after merging it, but I never every got any suggestions for better values. Thus they are broken because I just pulled them out of my arse.
10:28:27 <TrueBrain> michi_cc: hahahaha; what a nice description of events :D
10:28:31 <_dp_> it was funny when one of Hellish viewer games ended in a "who killed us with signals" rage without them realizing they just build unprofitable crap the whole game xD
10:30:13 <TrueBrain> "this cannot be correct, as I made it up", I have to remember that defense :)
10:30:28 <Samu> hello, merry christhmas everyone, if you're into that
10:30:43 <TrueBrain> I have a tree and everything!
10:30:47 <TrueBrain> IT EVEN HAS LIGHTS IN THEM
10:33:13 <_dp_> well, to me signal maintenance /10 would be a solution :p
10:33:17 <_dp_> or mb even remove it at all
10:33:56 <_dp_> they aren't exactly irl signals but more like inseparable part of track
10:35:08 <_dp_> 99,9% of my signals just exist to let trains follow each other
10:35:28 <TrueBrain> yeah, signals in OpenTTD has a slightly different meaning from signals in real life :P
10:35:41 <TrueBrain> they carry other functions :D
10:35:53 <TrueBrain> I guess no network n the world operates at such high efficiency :P
10:36:06 <TrueBrain> at least, I always pack trains wayyyyy to close on the mainbus :P
10:36:16 <DorpsGek> [OpenTTD/OpenTTD] SamuXarick updated pull request #7193: Fix #6468: Load correct version of AI as specified during the time of its save. https://git.io/fhHI1
10:36:59 <_dp_> from gamplay perspective it's also quite annoying to control signal density
10:37:15 <_dp_> you can't just increase or decrease it as needed, you need to resignal the whole line
10:38:00 <TrueBrain> "magic" signals is what we need :D
10:38:06 <TrueBrain> (I rejected a PR because it was too magic :P)
10:38:24 <DorpsGek> [OpenTTD/OpenTTD] SamuXarick updated pull request #7193: Fix #6468: Load correct version of AI as specified during the time of its save. https://git.io/fhHI1
10:39:31 <_dp_> TrueBrain, yeah, like inbuilt one-way pbs on every tile
10:39:39 <_dp_> so you just have one way track and two way track
10:41:56 <TrueBrain> _dp_: well, I was thinking, if we would add yellow signals, you can also reduce signal density a bit: that would prevent trains from slowing down and speeding up constantly
10:42:07 <TrueBrain> it does not solve anything if your density is like 90%
10:42:27 <_dp_> TrueBrain, I've no idea what you mean by "yellow signals"
10:42:59 <TrueBrain> I do not know if they do it in other countries, but here a signal can be red, green or yellow
10:43:10 <TrueBrain> red: STOP, green: you are good for at least 2 more signals, yellow: the next signal is red
10:43:27 <TrueBrain> so trains can slow down when they hit a yellow signal knowing they are going to hit another yellow by the time they get to the next
10:43:39 <TrueBrain> so they are constantly driving to hit yellow signals
10:43:54 <_dp_> sounds like smth completely unnecessary for the game
10:44:09 <_dp_> if anything it should be smth like atc control patch
10:44:30 <TrueBrain> depends on how you define atc, but this is a form of atc :)
10:44:42 <andythenorth> I went through PRs a few days ago
10:44:59 <andythenorth> there are only about 4 that I think relate to gameplay that should be moddable
10:45:06 <_dp_> well, basicaly not add any signals just make trains smart enough to slow down if needed
10:45:29 <andythenorth> but gamplay issues occupy disproportionate effort
10:45:29 <TrueBrain> that is on the end of a spectrum; I personally don't like that magic :)
10:46:07 <TrueBrain> I like watching at how countries solved these things in the real world
10:46:32 <_dp_> opentdd isn't a rail network simulator :p
10:46:59 <TrueBrain> I never said it was
10:47:21 <TrueBrain> you heavily suggested it ;)
10:48:41 <TrueBrain> btw, we already have ATC in-game .. PBS is also a form of ATC :)
10:48:56 <TrueBrain> (that is mostly how I know it .. preventing trains to run red lights :P)
10:49:02 <_dp_> anyway, a lot of fun in openttd lies in those ridiculously dense networks, so irl logic stops applying to it at some point
10:49:18 <TrueBrain> I agree if you hit 90% density, other solutions are needed
10:49:31 <TrueBrain> I still wonder how/if we can tune the game to allow both playstyles
10:49:33 <_dp_> and if you have 80% it's already fine as it is
10:49:57 <TrueBrain> as I can fully understand a group of people wanting a non-signal game where trains just do the right thing
10:50:04 <_dp_> well, I guess you can't quite mix trains of different speed but with economy as it is it's not rly needed
10:50:07 <TrueBrain> I am also very much aware another group won't want that :D
10:50:53 <TrueBrain> this game is so many games in one :D
10:51:02 <_dp_> well, "pbs everywhere" isn't that different from "non-signal" :p
10:51:40 <TrueBrain> I guess you hit this moment of building fatigue, where you are like: I did this 10 times now, now I want something to do it for me :)
10:52:00 <TrueBrain> I like how some games solve that by constantly adjusting the gameplay, where you first have to do a job 3 times, after that, it automates and you can worry about the next level
10:52:03 <TrueBrain> progressive gameplay
10:52:08 <TrueBrain> really difficult to tune well btw
10:52:11 <TrueBrain> really awesome to play
10:52:27 <andythenorth> shall I close 7000?
10:52:38 <TrueBrain> andythenorth: we do not close anything on xmas
11:00:04 * andythenorth back to tech support
11:00:18 <TrueBrain> "pappa, how do I build a coal train?"
11:00:30 <TrueBrain> "pappa, why can I only play OpenTTD and not any other game?"
11:00:51 <andythenorth> more like 'pappa how do I install this Android emulator on macOS so I can get games from Google Play Store"
11:01:13 <andythenorth> which involves an adventure into 'really is this malware?'
11:01:35 <andythenorth> and "why don't you just play it on iPad where you already have it"
11:01:39 <andythenorth> instead of wasting my time
11:01:50 * andythenorth is a grumpy parent
11:02:15 <michi_cc> Isn't the new ARM macOS supposed to allow iOS apps?
11:02:27 <andythenorth> let's do more tech support! :)
11:02:29 <andythenorth> also I need him to get off the mac so I can put clang on it and try a native M1 compile
11:06:41 * andythenorth tries to make sense of PR list :)
11:06:51 <andythenorth> finding it really hard to make sense of :)
11:07:05 <andythenorth> dunno if it's the titles, the green labels, github UI or what :D
11:07:27 <TrueBrain> I would suspect it is the fact it is a C++ PR most of the time has something to do with :P :P
11:08:37 <andythenorth> draft:false helps
11:10:35 <andythenorth> where's the vote button? :P
11:10:58 <DorpsGek> [OpenTTD/OpenTTD] andythenorth commented on pull request #7441: Feature: Per-group wagon removal flag https://git.io/JL1rI
11:11:19 <andythenorth> chances Peter is going to return and fix it are low I think
11:11:27 <TrueBrain> you are aware there is just an emoticon button in the top post, right? :)
11:12:23 <TrueBrain> well, there is on all comments too :P But that is often less useful :D
11:17:26 <DorpsGek> [OpenTTD/OpenTTD] andythenorth commented on pull request #7589: Change: Always apply inflation from 1920 to 2090, no matter the game start year. https://git.io/JL1rR
11:21:48 <TrueBrain> ha, finally found my argument why having things in mods is better than experimenting in the codebase: now we, as devs, have to judge if something is an improvement. with mods, our players do (really quick) :D
11:22:04 <DorpsGek> [OpenTTD/OpenTTD] andythenorth commented on pull request #7729: Add: Allow autoreplace with same model vehicle https://git.io/JL1ry
11:22:39 <andythenorth> TrueBrain yes massive +99
11:22:55 <andythenorth> review time is the most scarce commodity
11:23:15 <andythenorth> and what some projects do 'let anything in if it passes CI and nobody blocks it'
11:23:29 <TrueBrain> maybe more importantly: we are not players :)
11:23:45 <TrueBrain> (and I don't mean, because we don't play or anything, but: we are the minority of the minority
11:23:49 <andythenorth> I am, but I have a distinct play style far away from some of the others
11:24:10 <TrueBrain> we are not the "average" player, I should have said I guess
11:24:27 * andythenorth playing a meta-game, called 'can we extend the game?'
11:24:43 <andythenorth> the meta-game used to be mostly about 'problems in nfo'
11:24:49 <andythenorth> now it seems to be 'people'
11:26:34 <_dp_> with the diversity of openttd it's quite interesting what "average" player even does
11:26:42 <TrueBrain> haha, yeah, hence the quotes :)
11:26:43 <_dp_> he does struggle with signals I'd bet :p
11:27:16 <TrueBrain> but how ever we play the game, we are most likely a very small minority :)
11:27:19 <TrueBrain> nothing wrong with that btw :P
11:27:53 <andythenorth> I haven't dared to suggest it yet, but maybe we're not really making a game :P
11:28:17 <andythenorth> maybe we're making a framework for modding and sharing mods, with multiplayer support
11:28:22 <TrueBrain> this is all a simulation anyway
11:28:40 <andythenorth> I am curious how we make choices so that in THIS simulation we get a happy result :)
11:28:47 <andythenorth> I don't want to be in the boring simulation
11:29:01 * andythenorth worryingly actually believes some of this shit
11:29:26 <TrueBrain> the fun part is, your believe in it is irrelevant :)
11:29:33 <TrueBrain> how it influence your life, that is the question ;)
11:29:48 <andythenorth> well my belief is just code in the simulator :P
11:31:29 <TrueBrain> to me it seems you are just linking random PRs and go: OOOEEEHHH, pretty
11:32:49 <_dp_> as much as I like the "framework" idea it gets quite confusing to players
11:33:08 <andythenorth> it's the wrong language
11:33:15 <_dp_> like there is already this thing with town grow that's impossible to explaing because every damn gs does it differently
11:33:26 <andythenorth> I used to worry about that
11:33:34 <andythenorth> but then again...in base game...how do towns grow?
11:33:52 <andythenorth> also I watch my kids play all these roblox games and minecraft mods and stuf
11:34:29 <TrueBrain> _dp_: haha, yes, supplying information to the player in a sane way always has been a bit of an issue :P :D
11:34:34 <_dp_> and it's not just a difference in requirements, it's a difference in actual growth mechanic
11:34:43 <andythenorth> loads of mods on other games have poor explanations
11:34:45 <_dp_> but they all look the same except for requirements
11:34:51 <andythenorth> so they just go in fandom or youtube
11:35:42 <andythenorth> this would be a completely unacceptable attitude in a paid game or an application :)
11:35:51 <Samu> I just realised my AI saveload PR is actually fixing more than I initially thought. Now I understand why TrueBrain wants me to have everything explained in detail
11:36:00 <TrueBrain> andythenorth: that is not true .. many paid games do exactly the same :)
11:36:22 <andythenorth> I don't really play many games except Blitz
11:36:27 <_dp_> GS has hould building command that actually works different from how town builds it naturally
11:36:31 <Samu> as time goes by, things gets forgotten, and a PR with nearly no detail, also makes myself forget what it does
11:36:34 <_dp_> and that difference is well-hidden
11:36:51 <andythenorth> I am going to compete with GS for towns :D
11:37:56 <TrueBrain> you don't use t9 on your pc?!
11:39:02 <_dp_> yeah, probably should start so I don't blame it unfairly :p
11:41:55 <Samu> requesting to start version 15 of an AI, when the only installed version is 16, would actually start version 16 of the AI. However, when storing this information on a savegame, both version 15 and 16 are saved.
11:42:39 <Samu> i fix this by correcting the name of the ai, by discarting version 15 from the name
11:42:47 <DorpsGek> [OpenTTD/OpenTTD] michicc opened pull request #8427: Add: [NewGRF] Patch flag to test if inflation is on or off. https://git.io/JL1o6
11:43:38 <TrueBrain> Samu: now the question is, is that the correct solution for the problem?
11:44:43 <TrueBrain> michi_cc: looks trivial to me, but brrrrrrrr, I don't dare to approve it :D
11:44:48 <DorpsGek> [OpenTTD/OpenTTD] glx22 requested changes for pull request #7193: Fix #6468: Load correct version of AI as specified during the time of its save. https://git.io/JL1oX
11:52:44 <Samu> i know nothing about memory management :(
11:53:01 <TrueBrain> ignore that for now; first focus to see if your solution is the correct one
11:53:31 <DorpsGek> [OpenTTD/OpenTTD] James103 commented on pull request #7589: Change: Always apply inflation from 1920 to 2090, no matter the game start year. https://git.io/JL1KI
11:55:12 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on pull request #7589: Change: Always apply inflation from 1920 to 2090, no matter the game start year. https://git.io/JL1KO
12:06:10 <DorpsGek> [OpenTTD/OpenTTD] andythenorth commented on pull request #7589: Change: Always apply inflation from 1920 to 2090, no matter the game start year. https://git.io/JL1KK
12:07:08 <andythenorth> but we can rebuild him!
12:07:32 <TrueBrain> too bad you cannot move a comment to a discussion
12:07:34 <TrueBrain> that would be useful honestly
12:07:44 <andythenorth> I mean, I could copy-paste them around
12:09:07 <andythenorth> hmm who was writing FIRS 4 changelog?
12:09:18 * andythenorth looks at the other contributors
12:09:23 <andythenorth> THE VOID STARES BACK
12:12:47 <DorpsGek> [OpenTTD/OpenTTD] SamuXarick commented on pull request #7193: Fix #6468: Load correct version of AI as specified during the time of its save. https://git.io/JL1Kb
12:13:01 *** keith__ has quit IRC (Read error: No route to host)
12:13:59 <andythenorth> is a "I was drunk at the time" FIRS 4 changelog an improvement?
12:15:16 <TrueBrain> oef, forgot about lunch
12:15:20 <TrueBrain> lets fix that realllll quick
12:18:52 <Samu> this->name is not nullptr
12:20:14 <Samu> it has a name, which is the wrong name, and my intention here is to rename it
12:20:48 <Samu> how do I achieve this without creating a memory leak?
12:22:01 <LordAro> Samu: so this->name is currently pointing to a particular block of memory (the old wrong name)
12:22:32 <LordAro> to release this memory (before reassigning it to the new name), you need to call a particular function...
12:26:43 *** Progman_ has joined #openttd
12:34:07 *** Progman has quit IRC (Ping timeout: 480 seconds)
12:55:22 <andythenorth> oof making changelogs drunk...not such a good idea
12:55:44 <TrueBrain> being drunk at 1300 is not such a good idea, I would propose
12:56:23 <andythenorth> it will entail a nap later
12:56:39 <TrueBrain> are you so sad that your kids have a better machine than you have? :P
12:56:46 <TrueBrain> just rubbing it in, nothing to see here :D
12:57:27 <andythenorth> my sadness is without end
12:57:47 <andythenorth> maybe he wants a 16" screen
13:01:12 <TrueBrain> time to run eints for a bit .. I don't like all these warnings :P
13:02:17 <andythenorth> or maybe some lateral solution
13:02:46 <andythenorth> roads are part of the landscape in a way that rails aren't
13:05:08 <andythenorth> divide the list into two?
13:07:24 <DorpsGek> [OpenTTD/OpenTTD] andythenorth commented on pull request #8063: Fix #8055: Always allow building any available roadtypes, even if there are no road vehicles available https://git.io/JL1iF
13:07:52 <DorpsGek> - Update: Translations from eints (by translators)
13:09:49 <TrueBrain> funny, cannot reproduce a bug I had anymore :D Interesting .... :)
13:12:17 * andythenorth considers map storage registers
13:12:26 <andythenorth> we have industries and towns, why not map?
13:12:46 <TrueBrain> hahaha, okay ... seems problems disappear when I look at them ..
13:12:58 <TrueBrain> I was testing why the content service was downloading the wrong version of pathfinder.road
13:13:05 <TrueBrain> turns out, the author selected the wrong dependency
13:13:18 <TrueBrain> so yeah ... that is not an OpenTTD issue :)
13:13:56 <TrueBrain> we should extend BaNaNaS to help people with that
13:14:04 <andythenorth> improved edition
13:15:00 <andythenorth> we have 256 storages per grfid per town, so another 256 for the map won't kill us? :)
13:15:11 <andythenorth> the things I could do with storage...
13:15:56 <LordAro> newgrf in squirrel when
13:17:47 * andythenorth considers reduced FIRS 4 changelog
13:17:50 <Samu> LordAro, is it delete this->name?
13:17:54 <andythenorth> "Changes: changes"
13:18:33 <andythenorth> seems I did a crapload of non-player-visible refactoring
13:18:46 <andythenorth> I have no memory of doing that, but I'm sure it was....aces
13:20:20 <DorpsGek> [OpenTTD/OpenTTD] LordAro commented on pull request #8063: Fix #8055: Always allow building any available roadtypes, even if there are no road vehicles available https://git.io/JL1PI
13:21:01 <LordAro> Samu: close! if it were allocated with "new" (C++), that would be the case
13:21:16 <LordAro> but this is allocated with the C-style functions instead
13:21:32 *** frosch123 has joined #openttd
13:22:26 <DorpsGek> [OpenTTD/OpenTTD] andythenorth commented on pull request #8063: Fix #8055: Always allow building any available roadtypes, even if there are no road vehicles available https://git.io/JL1Pq
13:26:26 <andythenorth> that one definitely has the smell of 'valid as player setting'
13:26:42 <andythenorth> it's another variant of filter / hide / show / sort
13:26:50 <andythenorth> but in a dropdown list, which is clunky
13:26:55 <_dp_> btw, dunno if it's related to any storage but chunk dumps 64000 * 6(7) zeros to the savegame
13:27:06 <TrueBrain> LordAro: @TrueBrain Other language modifications, sure, but removals too? Leaving them in would cause compile warnings until the next eints run, which I'm not a particular fan of <- we talked about this a few times now already :) We can run eints when-ever we want :)
13:27:21 <DorpsGek> [OpenTTD/OpenTTD] James103 commented on pull request #8063: Fix #8055: Always allow building any available roadtypes, even if there are no road vehicles available https://git.io/JL1Ps
13:27:23 <TrueBrain> it is better to have an automated process do these things, than a manual .. as I have no way to validate if your changes are correct ;)
13:27:29 <andythenorth> do the apple thing, extra options in the dropdown if a modifier key is held down :P
13:27:46 <andythenorth> alt on Apple dropdowns makes all kind of weird shit possible
13:28:28 * andythenorth exaggerates maybe 90%
13:31:51 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on pull request #8063: Fix #8055: Always allow building any available roadtypes, even if there are no road vehicles available https://git.io/JL1P8
13:33:23 <LordAro> TrueBrain: isn't the lack of any (new) compile/build warnings enough?
13:34:10 * andythenorth going socialising
13:34:19 *** andythenorth has quit IRC (Quit: andythenorth)
13:34:32 <TrueBrain> LordAro: strgen warnings are NOT shown by the CI problem matcher
13:34:43 <TrueBrain> but removing additional strings would be completely unnoticeable
13:34:52 <TrueBrain> and history has shown people have issues removing things like cases
13:37:13 <TrueBrain> and most of all: why do a bot's job :P :D
13:38:14 <TrueBrain> anyway, to be clear, this is what I was told to do too for these cases; just following instructions :)
13:38:29 <TrueBrain> sorry frosch123 :) Feel free to add it :P
13:39:15 <frosch123> i guess it could just be a matcher. if someone says "bot", dorpsgek would reply with "what? me again?"
13:46:08 <frosch123> i created a file "todo202012.txt" with things i may do this holiday. in the place i put it, i found a file "todo201912.txt"
13:46:12 <TrueBrain> I forgot how complex some script functions are, damn
13:46:25 <TrueBrain> frosch123: how much of the 2019 did you do? :)
13:46:38 <frosch123> no idea, i remove things when they are done
13:46:43 <frosch123> there were two things left
13:46:51 <TrueBrain> that doesn't sound too bad :)
13:47:29 <frosch123> i am not sure whether i did one of them actually
13:50:45 <TrueBrain> I sometimes love reading code that is both old, and most likely would have made sense one day
13:50:51 <TrueBrain> how AIs are loaded is just funny
13:50:56 <TrueBrain> there is a parameter "force_exact_match"
13:51:13 <TrueBrain> which ... is ignored if no exact mach is found
13:51:21 <TrueBrain> so ... "prefer_exact_match" is what it does
13:53:21 <TrueBrain> ScriptInfoList::iterator it = this->info_list.begin();
13:53:21 <TrueBrain> for (; it != this->info_list.end(); it++) {
13:53:26 <TrueBrain> I am guessing this has a better C++ variant these days?
13:55:48 <LordAro> depends what's done with 'it' :p
13:55:57 <TrueBrain> AIInfo *i = static_cast<AIInfo *>((*it).second);
13:56:09 <LordAro> but probably a range for loop
13:57:37 <TrueBrain> "* @param force_exact_match Only match name+version, never latest." <- this is a lie .. not sure if the comment is wrong, or the code :P
13:57:48 <TrueBrain> LordAro: sadly, that is not really helping me :)
14:03:44 <frosch123> TrueBrain: "for (const auto& item : this->info_list)" whenever you are not modifying the container while iterating
14:04:03 <frosch123> oh, & right-aligned for ottd of course
14:04:09 <TrueBrain> cheers frosch123 :) I might add some C++ love to the script stuff ...
14:04:15 <frosch123> ottd was written in C, so it follows C formatting for * and & :)
14:04:26 <TrueBrain> yeah ... for C++ it makes a lot less sense :P
14:06:41 <TrueBrain> the only use of the force_exact_match is from the console ..
14:06:59 <TrueBrain> I vaguely remember that the requirements for launching an AI from console were weird at best
14:09:12 <TrueBrain> (svn r18944) -Change [FS#3232]: use the highest version of an AI that can load the AI data from a savegame instead of the exact same version
14:09:16 <TrueBrain> seems things went wrong there :D
14:11:28 <michi_cc> Even in C++ operator * and & bind the the right and not the left, meaning the formatting makes a lot of sense everywhere you can multiple right things.
14:13:25 <TrueBrain> from what people have told me (again, C programmer), this is more uncommon in C++ not?
14:13:49 <frosch123> yes, declaring multiple things on the same line is "very uncool" in C++
14:14:01 <frosch123> it's probably also uncool in C99
14:14:09 <TrueBrain> I always lost the debate on that argument :P For C I win it easily, but in C++ they just tell me: you shouldn't do that :D
14:14:11 <frosch123> it only makes sense when you declare everything at the top
14:14:30 <TrueBrain> ANSI-C or go bust :P
14:14:36 <TrueBrain> (hahaha, sorry, that was a bad joke)
14:14:44 <frosch123> but with initialise-on-definition there is little point in putting multple things on the same line
14:15:21 <frosch123> TrueBrain: at least you did not say k&r c :)
14:15:42 <frosch123> at least you know prototypes and other things than int
14:15:44 <TrueBrain> fuck k&r, really, that coding style is horrible :P
14:16:10 <TrueBrain> but having only /* */ as a way to comment is also silly (ANSI-C)
14:16:27 <TrueBrain> int a; /* this is a comment */
14:16:29 <TrueBrain> is annoying REAL quick
14:16:54 <frosch123> does ansi c allow names longer than 8 chars?
14:17:06 <TrueBrain> it is not fortran :)
14:17:19 <TrueBrain> OpenDUNE was ansi-c (C89 with pedantic) for a long long time
14:17:37 <TrueBrain> this was mostly as it used TCC to do static decompilation in the background, which had a very limited set of C implemented :)
14:20:38 <DorpsGek> [OpenTTD/OpenTTD] SamuXarick updated pull request #7193: Fix #6468: Load correct version of AI as specified during the time of its save. https://git.io/fhHI1
14:21:42 <frosch123> also, sorry for turning the dev stream into a hovercraft meme :p
14:21:55 <DorpsGek> [OpenTTD/OpenTTD] SamuXarick updated pull request #7193: Fix #6468: Load correct version of AI as specified during the time of its save. https://git.io/fhHI1
14:22:00 <TrueBrain> nothing to be sorry about frosch123 :D It is epic :)
14:24:27 <Samu> There is a problem with the DEBUG messages presenting the version it is trying to load
14:25:47 <Samu> mainly... when it's working with possible 3 different version reports: -1, the version in the name, and the actual version of the script
14:27:20 <Samu> version being -1 is caused by my workaround to pass the correct variables to FindInfo
14:27:44 <Samu> so, there's that being my PR's fault
14:28:06 <Samu> but then there's the "version in the name" vs "actual version saved"
14:28:20 <Samu> which one to report for the DEBUG message
14:44:09 <DorpsGek> [OpenTTD/OpenTTD] SamuXarick commented on pull request #7193: Fix #6468: Load correct version of AI as specified during the time of its save. https://git.io/JL1Xj
14:44:50 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on pull request #7193: Fix #6468: Load correct version of AI as specified during the time of its save. https://git.io/JL11v
14:45:14 <TrueBrain> sorry Samu , but you analysis did not make sense to me, so I dived into it .. my analysis comes back with a different result; hopefully my story makes a bit sense to you
14:45:54 <glx> yeah the issue is a wrong this->name
14:46:26 <glx> the fix should be in console command area
14:46:54 <TrueBrain> I wonder why we ever added that you can load a specific AI version
14:46:59 <TrueBrain> it sure sounded as a good idea
14:47:00 <TrueBrain> it really is not :P
14:47:22 <glx> for ai writer doing tests
14:47:42 <TrueBrain> guess we should add a third flag to indicate if it should use an exact match or not
14:48:18 <TrueBrain> or indeed change the name earlier, and simply let it auto-upgrade on next load
14:48:29 <TrueBrain> making starting a specific version just a temporary thing till next load
14:50:57 <TrueBrain> the code is buggy too .. "startai pr7193.1abc" also starts version 1
14:51:12 <TrueBrain> it just does an atoi() on what-ever is behind the dot
14:52:27 <TrueBrain> so I guess the question is: do we want it to be possible that you can lock the AI version (which is only possible when loading via the console atm)
14:52:32 <TrueBrain> does it make sense? Opinions?
14:52:40 <TrueBrain> (similar for GS, of course)
14:58:08 <Samu> i like to test different versions of my AI sometimes
14:58:55 <Samu> so, yes, for me it's helpful
14:59:04 <TrueBrain> funny, this code also made it into game_scanner, but it can never be used there :D
14:59:28 <TrueBrain> Samu: the question is, does it need to survive a save/load sequence?
14:59:33 <TrueBrain> I would guess you only use it for testing
15:00:40 <Samu> per documentation, it should
15:00:55 <TrueBrain> 1) documentation can be changed, 2) without any reference that is not helpful
15:01:28 <glx> anyway once loaded name should not have ".version" attached
15:02:14 <Samu> if it is to survive future saves, it has to still keep the version in the name
15:02:20 <TrueBrain> glx: indeed; the question I have, should we store the version should be kept "exact", or the other way around: should we keep a flag that an AI can be auto-upgraded
15:02:34 <glx> Samu: version is stored in the save already
15:02:39 <TrueBrain> Samu: I am waiting for the reference to documentation that says what you claim :)
15:03:40 <Samu> but glx, what happens next time you load it, it will try to get the latest available version then
15:04:02 <glx> it first tries the saved version
15:04:09 <TrueBrain> glx: it does not :)
15:04:16 <TrueBrain> AIs and GSes are always auto-upgraded
15:04:27 <TrueBrain> it will pick the latest version that is compatible
15:05:14 <glx> ok so that may require a fix, anyway no need to store version in the name
15:05:23 <TrueBrain> no, that does not require a fix :) That is fully intended :)
15:05:25 <glx> as it's already in the save
15:05:27 <TrueBrain> that is 100% as it should work :P
15:06:22 <glx> but yeah usually if you need to start different version of ai it's probably to check savegame compatibility
15:06:40 <TrueBrain> so a save/load should load the latest compatible again
15:06:46 <TrueBrain> I think that is the most sane thing to do
15:06:59 *** DasPoseidon has joined #openttd
15:07:56 <Samu> doc says: "load the exact same version of the same script", it doesn't document saving behaviour
15:08:45 <Samu> under Detailed Description
15:09:00 <TrueBrain> lolz, pretty sure it never worked like that :D Lets dig into the history :)
15:12:09 <TrueBrain> in Jan 2009 it did work like described .. now let's find when that changed
15:13:10 <TrueBrain> you cannot reverse-blame in Git: when did this line get removed, I guess? :D
15:13:57 *** Flygon has quit IRC (Quit: A toaster's basically a soldering iron designed to toast bread)
15:14:13 <TrueBrain> Jan 2010 that got removed, and from there it always picked the latest
15:14:19 <TrueBrain> so the documentation is only out-of-date for 10 years :D
15:14:45 <glx> oh maybe when force_exact_version was added
15:15:04 <TrueBrain> it is the same commit, yes
15:17:50 <glx> now the question is should we restore old behaviour
15:18:08 <TrueBrain> this was a well defended choice
15:18:13 <TrueBrain> which nobody has complained about in 10 years
15:18:56 <glx> anyway there's still the issue with wrong this->name ;)
15:19:20 <TrueBrain> long fixed locally in my branch :P
15:19:52 <glx> when trying to fix libs stuff ?
15:21:59 <glx> hmm maybe it's possible to add a console only dev settings that is never saved and always off when openttd is started
15:22:17 <TrueBrain> what are you trying to solve, sorry?
15:22:49 <glx> but really even when developping AI I don't see the need for exact match
15:27:27 *** andythenorth has joined #openttd
15:28:28 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain opened pull request #8431: Doc: for over 10 years, we do not load the exact AI version first https://git.io/JL1M8
15:29:08 <DorpsGek> [OpenTTD/OpenTTD] michicc approved pull request #8431: Doc: for over 10 years, we do not load the exact AI version first https://git.io/JL1MB
15:30:05 <TrueBrain> Samu: if you can test my PR and let me know if that fixes the issue for you too, that would be nice
15:30:11 <TrueBrain> do note, it does NOT fix old savegames
15:30:52 <michi_cc> TrueBrain: Your PR has a fail :(
15:31:35 <TrueBrain> last minute cleanup hurting me :D
15:33:05 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain updated pull request #8430: Fix #6468: don't store version of AIs-started-via-console in name https://git.io/JL1MC
15:33:15 <Samu> 'e': undeclared identifier
15:33:55 <TrueBrain> cannot believe the documentation states a false for 10 years :D And nobody noticed :D That is just funny :)
15:34:01 <TrueBrain> shows that Zuu made a good call with his request :)
15:35:15 <michi_cc> It probably more that nobody every reads that comment :)
15:35:34 <TrueBrain> I wonder if we should keep code to fix old savegames ... it feels like it is only an issue for 0.000001% of the savegames
15:42:42 <TrueBrain> for (const auto &item : *this->GetConfigList()) { <- that * feels a bit icky .. guess I should change the value the function returns instead :D
15:45:19 *** Progman_ has quit IRC (Remote host closed the connection)
15:46:02 <glx> oh I think I know why the .version code was duplicated in ai and game, it's probably for manual version forcing in cfg
15:46:16 <TrueBrain> or .. it was just a copy/paste :P :D
15:47:17 <Samu> oh, the fix in the startai is good, I like it, and agree with it
15:48:37 <Samu> can no longer retain the exact version
15:48:43 <Samu> when loading savegame back
15:49:22 <Samu> erm, yes, you're right, though I actually thought it was what the documentation implied
15:50:26 <TrueBrain> SettingValueList::const_iterator it = this->settings.find(name);
15:50:26 <TrueBrain> if (it == this->settings.end())
15:50:29 <glx> it was possible at some point, but it then has been modified to auto update on load
15:50:32 <TrueBrain> is there a C++ way to do that better?
15:51:04 <TrueBrain> is there a better C++ way to do that better?
15:51:04 <glx> .end() is the not found return value
15:51:18 <TrueBrain> I can add more "better" if you like :D
15:51:27 <TrueBrain> it makes for really annoying-to-read statements
15:51:40 <TrueBrain> I wouldn't be surprised there can be used some "auto" or things here
15:52:24 <glx> well auto can replace SettingValueList::const_iterator I think
15:53:26 <TrueBrain> guess it should be a "const auto" to match the current form
15:55:07 <Samu> so the "feature" to start specific AI versions and surviving save/load is what my PR tried to "fix according to the (outdated) documentation",
15:55:28 <Samu> too bad, I actually liked the feature
15:55:45 <glx> you still can start specifi version
15:56:21 <glx> exact load is easy to add, but I don't see that happening
15:56:36 <glx> as auto upgrading looks like the rigth thing to do
15:59:07 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain opened pull request #8432: Codechange: use more C++11 in script-related functions https://git.io/JL1Dz
15:59:09 <TrueBrain> draft, looking for input ^^
15:59:27 <DorpsGek> [OpenTTD/OpenTTD] glx22 approved pull request #8430: Fix #6468: don't store version of AIs-started-via-console in name https://git.io/JL1Dg
15:59:47 <TrueBrain> I made a lot of iterators "const" which weren't before; no clue why they were not
16:00:05 <glx> if there's an issue with people setting version in cfg they'll probably complain ;)
16:01:08 <glx> (if it was possible to do it)
16:01:39 <TrueBrain> it possibly was, but it would also be horribly broken after saving and loading :D
16:02:54 <TrueBrain> what an absolute boring PR is it to read, that codechange one of mine, lolz
16:03:35 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain merged pull request #8431: Doc: for over 10 years, we do not load the exact AI version first https://git.io/JL1M8
16:03:48 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain closed issue #6468: Loading savegame with AI loaded as specific version of that AI fails to load correct version of AI https://git.io/JL1DS
16:03:51 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain merged pull request #8430: Fix #6468: don't store version of AIs-started-via-console in name https://git.io/JL1MC
16:04:10 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on pull request #7193: Fix #6468: Load correct version of AI as specified during the time of its save. https://git.io/JL1D9
16:04:13 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain closed pull request #7193: Fix #6468: Load correct version of AI as specified during the time of its save. https://git.io/fhHI1
16:04:51 <TrueBrain> right, time for another xmas dinner :D
16:04:59 <glx> for me the most annoying stuff is the FOR_XXX macro still present in the source :)
16:12:36 *** Tirili has quit IRC (Remote host closed the connection)
16:22:26 <DorpsGek> [OpenTTD/OpenTTD] frosch123 commented on pull request #8432: Codechange: use more C++11 in script-related functions https://git.io/JL1yz
16:28:12 <DorpsGek> [OpenTTD/OpenTTD] frosch123 commented on pull request #8427: Add: [NewGRF] Patch flag to test if inflation is on or off. https://git.io/JL1yP
16:29:29 <TrueBrain> this->config_list = new ScriptConfigItemList();
16:29:30 <TrueBrain> return *this->config_list;
16:29:39 <TrueBrain> am I doing that correct? puzzled face
16:30:47 <glx> you can change this->config_list type, it's a vector anyway
16:31:08 <frosch123> one change at a time :)
16:31:47 <TrueBrain> I always feel a bit weird if I dereference a "new" element
16:31:53 <TrueBrain> dunno, somewhere in my head that kinda crashes :D
16:31:54 <frosch123> today tb learns range-based loops. tomorrow he learns the STL containers, then unique_ptr and optional :)
16:32:33 <DorpsGek> [OpenTTD/OpenTTD] LordAro commented on pull request #8432: Codechange: use more C++11 in script-related functions https://git.io/JL1y7
16:32:37 <TrueBrain> and should I include that in this PR, or rather in another?
16:32:40 <frosch123> then he upgrades to c++20, and learns that range-based loops are uncool and cool kids use std::for_each
16:33:40 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on pull request #8432: Codechange: use more C++11 in script-related functions https://git.io/JL1yF
16:33:44 <frosch123> TrueBrain: same pr, but different commit?
16:34:07 <glx> will be squashed later I guess
16:34:16 <TrueBrain> it is also about: what is easiest to review
16:34:20 <TrueBrain> and: do we want this, etc :)
16:34:35 <TrueBrain> I am not looking for stuff to do, basically :P But reading these endless ::iterator was annoying me :D
16:36:34 *** Wormnest has joined #openttd
16:37:25 <DorpsGek> [OpenTTD/OpenTTD] andythenorth commented on pull request #8427: Add: [NewGRF] Patch flag to test if inflation is on or off. https://git.io/JL1yp
16:44:24 <DorpsGek> [OpenTTD/OpenTTD] SamuXarick updated pull request #7890: Fix #6452: Reset only editable and visible settings from GUI https://git.io/Jexqi
16:52:44 <DorpsGek> [OpenTTD/OpenTTD] LordAro commented on pull request #8432: Codechange: use more C++11 in script-related functions https://git.io/JL1Sn
17:07:27 <glx> oh visual studio supports clang-format
17:11:22 * andythenorth wonders how many water tiles need to be connected to differentiate 'lake' and 'sea'
17:11:30 <andythenorth> in some kind of reasonable world
17:12:06 * andythenorth counts with finger
17:13:11 <andythenorth> 512 isn't enough :)
17:13:12 <TrueBrain> LordAro: how much of a diff does that create? :D
17:13:46 <LordAro> but less than the others :p
17:24:51 <Wolf01> andythenorth: what about if is considered sea if connected with map border?
17:29:19 <frosch123> main problem i have. there is no format option for enums
17:30:24 <frosch123> a lot of enums are a lot better if the "=" are aligned
17:30:39 <frosch123> clang has a single "align assignments"
17:30:51 <frosch123> which is fine in many cases, where ottd already does that
17:30:56 <DorpsGek> [OpenTTD/OpenTTD] floodious commented on pull request #7745: Feature: setting for more flexible town spacing https://git.io/JL19M
17:31:08 <frosch123> but it does not stop at extremes, where 3 vars are 8 chars, and 1 is 40
17:41:26 <andythenorth> Wolf01 pathological on larger maps :)
17:42:00 <Samu> just finished rebaseing every of my branches to upstream/master
17:43:12 <Samu> the most complicated was lifetime profit conflicting with group by shared orders commit
17:43:26 <Samu> oh well, nobody cares :p
17:54:13 <TrueBrain> So rename the var! :p
17:59:51 <frosch123> you can insert empty lines and no-format comments
17:59:58 <frosch123> but it's a lot of work :)
18:00:47 <frosch123> also i used a setting that debian stable's clang-format does not know yet :p
18:01:02 <frosch123> but that will fix itself in half a year
18:03:42 <TrueBrain> it only needs to run on a single CI target, so that is not an issue :P
18:13:55 <TrueBrain> holy crap, the errors you get if a for() fails, are absurd
18:14:02 <TrueBrain> like 100+ lines of errors, where only the first line mattes :)
18:16:38 <DorpsGek> - Update: Translations from eints (by translators)
18:17:34 <frosch123> TrueBrain: there are compiler fights, which produces the best errors :)
18:18:16 <frosch123> i have no experience with clang, but msvc is way worse than gcc
18:22:19 <TrueBrain> okay, MultiMap is weird .. going to skip that :D
18:24:19 <TrueBrain> and I see we call the iterator in this constrct "i", "iter", or "item" ... :D
18:29:03 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain updated pull request #8432: Codechange: use more C++11 in script-related functions https://git.io/JL1Dz
18:30:35 <TrueBrain> only 7 files, that was easier than expected :)
18:31:20 <DorpsGek> [OpenTTD/OpenTTD] LordAro approved pull request #8432: Codechange: use more C++11 in script-related functions https://git.io/JL1Q2
18:31:51 <TrueBrain> that was ... unexpected quick :o
18:32:10 <LordAro> i get email notifications
18:32:35 <frosch123> you read email faster than irc?
18:32:38 <LordAro> sometimes i pay attention to them
18:33:27 <LordAro> phone notifications :p
18:33:37 <LordAro> i reviewed it from my phone too
18:33:52 <TrueBrain> but it is appreciated :D
18:34:08 <frosch123> TrueBrain: just think about where he was while reviewing
18:34:12 <glx> I find reading diffs on phone not that easy
18:34:26 <glx> indentation is all broken in github app
18:34:27 <TrueBrain> frosch123: no; just: no
18:35:12 <andythenorth> ipad viewing angle is optimised for toilet
18:35:29 <andythenorth> that was the original design brief
18:35:44 <TrueBrain> okay .. next challenge ... emscripten ... it had bugs ... lets find out why and how to solve them :D
18:38:21 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain merged pull request #8432: Codechange: use more C++11 in script-related functions https://git.io/JL1Dz
18:42:01 <TrueBrain> sometimes I do see them in my shadow channel
18:42:03 <TrueBrain> sometimes I do not :P
18:43:25 <TrueBrain> 39 locations where we catch errors in OpenTTD
18:48:44 <TrueBrain> a for with "int i = 1" is just confusing my brain, as it turns out
18:51:38 <TrueBrain> I blame the original writer of the code btw :P
18:54:32 <TrueBrain> will I be in time for the nightly, now that is the question :D
18:54:55 <DorpsGek> [OpenTTD/OpenTTD] SamuXarick commented on pull request #8398: Change: Distribute left over cargo to stations according to the highest remainder https://git.io/JL17c
18:55:47 *** HerzogDeXtEr has joined #openttd
18:56:39 *** Samu has quit IRC (Quit: Leaving)
18:58:21 *** HerzogDeXtEr has quit IRC ()
18:58:38 *** HerzogDeXtEr has joined #openttd
18:58:42 <TrueBrain> running AIs in emscripten is kinda hurtful for performance :D That is not a real surprise, I guess
18:59:54 <TrueBrain> pretty okay on Chrome, so what-ever
19:00:15 <TrueBrain> why is left-shift on Linux fast-forward, but on other OSes it is tab?
19:01:46 <TrueBrain> I did not expect that :)
19:03:01 <andythenorth> catches most of us now and then
19:03:56 <TrueBrain> dead-keys are funny .. when I press `, the console opens, but the next character I type sends a "`<key>" pair to OpenTTD, making it close the console again
19:03:59 <TrueBrain> so I cannot open the console :D
19:04:26 <TrueBrain> (well, via the menu ofc)
19:08:58 <frosch123> i had the opposite problem
19:09:04 <frosch123> i can open the console, but not close it
19:10:31 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain opened pull request #8434: Fix: [Emscripten] compile with exceptions enabled, as our AIs depend on it https://git.io/JL17w
19:10:42 <TrueBrain> dead-keys are annoying
19:10:46 <TrueBrain> not sure how we want to deal with it
19:10:56 <frosch123> oh wait, i figured it out
19:11:17 <frosch123> german layout, no dead keys: i open the console with ^, but i close it with `
19:11:21 <TrueBrain> I do not think there is a way to detect it ...
19:12:03 <TrueBrain> at least fixed most of the other issues with emscripten, I think :)
19:14:43 <milek7> how much performance penalty is DISABLE_EXCEPTION_CATCHING=0?
19:14:48 <TrueBrain> really tempted to just approve https://github.com/OpenTTD/OpenTTD/pull/7589 .. even if we remove inflation completely later on, I think the only way we are going to get useful feedback about that PR is to simply merge it, and revert it if we don't want it
19:14:55 <TrueBrain> milek7: check the PR, I mention that
19:15:18 <TrueBrain> frosch123: how do you rate #7589 .. yah or nah?
19:15:33 <TrueBrain> milek7: in the PR description and in the commit message itself
19:15:45 <TrueBrain> lol, emscripten build fails .. now that is interesting
19:15:49 <frosch123> TrueBrain: i don't think anyone will notice a difference. so if it makes some people happy, it is fine
19:16:19 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain updated pull request #8434: Fix: [Emscripten] compile with exceptions enabled, as our AIs depend on it https://git.io/JL17w
19:19:17 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain approved pull request #7589: Change: Always apply inflation from 1920 to 2090, no matter the game start year. https://git.io/JL15U
19:21:49 <andythenorth> I wanted to troll this by stealing a comment from 7745
19:21:53 <andythenorth> but I can't find a good one
19:22:49 <milek7> TrueBrain: I wonder if it's possible to limit that setting to only script/sl code
19:22:57 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain updated pull request #8434: Fix: [Emscripten] compile with exceptions enabled, as our AIs depend on it https://git.io/JL17w
19:23:27 <TrueBrain> milek7: as mentioned earlier, there are 39 places where we catch .. pretty sure that would be hard-to-maintain code :)
19:24:09 <TrueBrain> so I did not opt for that route :)
19:27:01 *** Wormnest has quit IRC (Quit: Leaving)
19:29:44 <frosch123> if you can profile stuff, maybe it makes sense to add some "noexcept" to drawing and pathfinding code
19:30:24 <TrueBrain> you can only indicate where it should do exception handling
19:30:37 <TrueBrain> would be nice if it could be done the other way around :D
19:31:12 <frosch123> i would expect your c++->js compiler to understand "noexcept"
19:31:49 <TrueBrain> and sorry, then I misunderstood, wtf is "noexcept"?
19:32:03 <frosch123> a keyword you can add to any method/function prototype
19:32:31 <frosch123> says, the compiler is allowed to assume the function will never throw
19:32:52 <TrueBrain> but wouldn't that be a lot of definitions?
19:33:17 <LordAro> that's why most people don't bother
19:33:21 <frosch123> TrueBrain: exactly, hence only for pathfinders and drawing
19:33:35 <LordAro> i've never heard of it making any actual difference
19:33:42 <TrueBrain> can't you mark whole files?
19:34:26 <TrueBrain> anyway, I couldn't measure any difference, despite me using the slowest browser (FF) and build (Debug) I could find
19:34:47 <TrueBrain> from their page, a library that is burning the CPU, it makes everything 15% slower
19:35:12 <TrueBrain> so I am not sure any of those efforts are worth it :)
19:35:54 <LordAro> TrueBrain: i tagged your PR :)
19:36:18 <DorpsGek> [OpenTTD/OpenTTD] frosch123 merged pull request #7589: Change: Always apply inflation from 1920 to 2090, no matter the game start year. https://git.io/fjWc7
19:37:53 <DorpsGek> [OpenTTD/OpenTTD] SamuXarick updated pull request #8398: Change: Distribute left over cargo to stations according to the highest remainder https://git.io/JLuoN
19:45:10 <TrueBrain> I really should also publish a "master" version of emscripten
19:45:15 *** WormnestAndroid has quit IRC (Ping timeout: 480 seconds)
19:45:15 <TrueBrain> easier to check what is "normal" etc
19:45:18 <TrueBrain> something for another day :)
19:45:29 *** WormnestAndroid has joined #openttd
19:47:44 <frosch123> can you start bith in on window and replicate input events? :p
19:50:34 <TrueBrain> LordAro: it is done! And AIs work :D \o/ AroAI is a goooooooo
19:51:16 <frosch123> does that also fix loading the savegames? or is there another savegame incompatibility?
19:51:55 <TrueBrain> it should fix them, but .. I forgot to test :D Lets fix that!
19:54:11 <TrueBrain> works flawness now :D
19:54:59 <TrueBrain> still need to add a way for you to upload and download savegames
19:55:02 <TrueBrain> but that is for another day :)
19:55:44 <TrueBrain> so: works as advertised :D
19:56:10 <TrueBrain> I have some ideas to improve performance, but that needs more experimentation :D Not for now at least :)
19:56:13 <DorpsGek> [OpenTTD/OpenTTD] LordAro approved pull request #8434: Fix: [Emscripten] compile with exceptions enabled, as our AIs depend on it https://git.io/JL1dV
19:56:29 <DorpsGek> [OpenTTD/OpenTTD] TrueBrain merged pull request #8434: Fix: [Emscripten] compile with exceptions enabled, as our AIs depend on it https://git.io/JL17w
20:03:27 * andythenorth is not in the Ballmer zone
20:08:07 *** Tirili has quit IRC (Ping timeout: 480 seconds)
20:12:55 *** HerzogDeXtEr has quit IRC (Read error: Connection reset by peer)
20:14:02 <Xaroth> not enough train stations.
20:17:02 <andythenorth> needs more state machines
20:22:08 *** WormnestAndroid has quit IRC (Ping timeout: 480 seconds)
20:47:30 <DorpsGek> [OpenTTD/OpenTTD] michicc commented on pull request #8427: Add: [NewGRF] Patch flag to test if inflation is on or off. https://git.io/JL1bi
20:48:45 *** Xaroth7 has joined #openttd
20:49:24 *** Xaroth is now known as Guest9085
20:49:24 *** Xaroth7 is now known as Xaroth
20:51:28 *** Guest9085 has quit IRC (Ping timeout: 480 seconds)
20:55:56 <DorpsGek> [OpenTTD/OpenTTD] michicc updated pull request #8427: Add: [NewGRF] Patch flag to test if inflation is on or off. https://git.io/JL1o6
20:56:29 <DorpsGek> [OpenTTD/OpenTTD] michicc commented on pull request #8427: Add: [NewGRF] Patch flag to test if inflation is on or off. https://git.io/JL1NT
20:58:32 *** gelignite has joined #openttd
21:04:48 <DorpsGek> [OpenTTD/OpenTTD] JGRennison commented on pull request #8427: Add: [NewGRF] Patch flag to test if inflation is on or off. https://git.io/JL1N4
21:15:40 *** supermop_Home_ has quit IRC (Remote host closed the connection)
21:15:49 *** WormnestAndroid has joined #openttd
21:44:58 <nielsm> ah phew good, looks like nobody's tried to get hold of me today
21:51:00 *** Samu has quit IRC (Quit: Leaving)
21:56:33 *** WormnestAndroid has quit IRC (Ping timeout: 480 seconds)
21:56:51 *** WormnestAndroid has joined #openttd
22:33:57 <Naruni> Hello, I'm working on a patch. In source file rail_gui.cpp:351 (function DoRailroadTrack) there is a function DoCommandP called with two TileVirtXY arguments. I'm trying to offset the location of the track being built but can't get it to work.
22:34:44 <Naruni> So far I've tried changing TileVirtXY(_thd.selstart.x, _thd.selstart.y) to TileVirtXY(_thd.selstart.x, _thd.selstart.y)+1 and TileVirtXY(_thd.selstart.x, _thd.selstart.y) << 4 but neither work.
22:35:32 <Naruni> it tries to place the track at the cursor location instead of offset
22:37:53 <frosch123> no idea what you want to do. but try putting you offsets inside the TileVirtXY instead of outside?
22:39:02 <frosch123> or use TileAdd, TileAddWrap or similar
22:40:14 <frosch123> so many functions :)
22:43:29 <Naruni> frosch123, thanks! I'm not familiar with the code at all, I worked on other projects but just wanted to take a gander at this game's code
22:44:07 <Naruni> and yes I also tried (_thd.selstart.x+1, _thd.selstart.y+1), my mistake in typing that incorrectly
22:44:27 <Naruni> ADDXY looks good, I'll go look through that
22:46:03 <Naruni> I'm just trying to patch it so whenever I build track it will build a second track of the same orientation
22:53:41 *** Progman has joined #openttd
23:02:49 *** Wolf01 has quit IRC (Quit: Once again the world is quick to bury me.)
23:06:28 *** frosch123 has quit IRC (Quit: be yourself, except: if you have the opportunity to be a unicorn, then be a unicorn)
23:08:41 *** andythenorth has quit IRC (Quit: andythenorth)
23:16:32 *** nielsm has quit IRC (Ping timeout: 480 seconds)
23:27:54 *** sla_ro|master has quit IRC ()
continue to next day ⏵
