Log for #openttd on 10th January 2021:
Times are UTC Toggle Colours
00:19:29  *** nielsm has quit IRC
00:21:45  <DorpsGek> [OpenTTD/OpenTTD] mattkimber updated pull request #8540: Fix eeb88e8: Trains reversed while paused do not correctly update sprite bounds
00:23:24  <Timberwolf> I'm getting too used to Go refusing to compile code with an unused variable.
00:27:54  *** WormnestAndroid has quit IRC
00:29:01  *** WormnestAndroid has joined #openttd
00:33:34  *** rptr_ has joined #openttd
00:33:34  *** rptr has quit IRC
00:53:05  *** Flygon has joined #openttd
01:21:42  *** gelignite has quit IRC
01:28:26  *** Progman has quit IRC
01:58:07  *** rptr_ has quit IRC
02:58:44  *** snail_UES_ has quit IRC
03:04:35  *** snail_UES_ has joined #openttd
03:13:46  *** heffer has joined #openttd
03:13:55  *** D-HUND has joined #openttd
03:17:18  *** debdog has quit IRC
03:45:25  *** ekangmonyet_ has quit IRC
03:55:39  *** glx has quit IRC
06:39:10  <DorpsGek> [OpenTTD/OpenTTD] perezdidac opened pull request #8552: Feature: Prevent vehicles to unload cargo in implicit stops in cargodist mode if the cargo is not accepted in such station
06:46:49  <DorpsGek> [OpenTTD/OpenTTD] perezdidac updated pull request #8552: Feature: Prevent vehicles to unload cargo in implicit stops in cargodist mode if the cargo is not accepted in such station
07:09:47  <DorpsGek> [OpenTTD/OpenTTD] perezdidac updated pull request #8552: Feature: Prevent vehicles to unload cargo in implicit stops in cargodist mode if the cargo is not accepted in such station
07:21:56  <DorpsGek> [OpenTTD/OpenTTD] perezdidac updated pull request #8552: Feature: Prevent vehicles to unload cargo in implicit stops in cargodist mode if the cargo is not accepted in such station
07:35:17  *** nielsm has joined #openttd
07:39:59  *** andythenorth has joined #openttd
07:43:05  *** sla_ro|master has joined #openttd
07:53:06  <andythenorth> yo
08:09:19  *** Wolf01 has joined #openttd
08:10:12  *** snail_UES_ has quit IRC
08:35:36  *** Gustavo6046 has quit IRC
08:36:01  *** Gustavo6046 has joined #openttd
08:36:30  *** jottyfan has joined #openttd
08:43:53  <TrueBrain> wow, what a PR. Good motivations, description, and code looks good to me .. lol, what is happening :D
08:45:12  <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on pull request #8552: Feature: Prevent vehicles to unload cargo in implicit stops in cargodist mode if the cargo is not accepted in such station
08:57:29  *** Tirili has joined #openttd
09:12:28  <andythenorth> livery_types then with labels and name strings
09:12:33  <andythenorth> such lol
09:13:04  <andythenorth> liivery_type subtypes also
09:15:03  <andythenorth> common livery_type label scheme
09:27:02  *** Tirili has quit IRC
09:37:30  *** Tirili has joined #openttd
09:54:15  <DorpsGek> [OpenTTD/OpenTTD] Kuhnovic commented on pull request #8550: Feature: "Remove all industries" button in scenario editor
09:54:28  <Eddi|zuHause> i'm pretty sure that's the main reason we SHOULDN'T have such a property
09:57:52  <andythenorth> it wouldn't be just a property at that point
09:57:56  <andythenorth> it would be a newgrf feature
09:58:00  <andythenorth> with action 0 and so forth
09:58:13  * andythenorth is just following one route to logical conclusion
09:58:54  <Eddi|zuHause> i'm not awake enogh to follow that conclusion
09:59:24  <DorpsGek> [OpenTTD/OpenTTD] Kuhnovic updated pull request #8550: Feature: "Remove all industries" button in scenario editor
10:00:24  <andythenorth> if liveries are a concept that exists beyond a specific vehicle
10:00:32  <andythenorth> then they need an entity to define them
10:01:06  <andythenorth> 'if'
10:09:29  *** jottyfan has quit IRC
10:14:23  *** Samu has joined #openttd
10:14:30  <DorpsGek> [OpenTTD/OpenTTD] Kuhnovic commented on pull request #8550: Feature: "Remove all industries" button in scenario editor
10:45:33  *** Progman has joined #openttd
10:51:26  <DorpsGek> [OpenTTD/OpenTTD] SamuXarick updated pull request #8390: Add: [AI/GS] Missing water related functions and objects
11:08:44  <Samu> weird diff
11:12:10  <Samu> glx isn't around t.t
11:12:53  <Samu> the generated log file is adding extra lines
11:13:04  <Samu> on stuff that wasn't changed, because of ";"
11:16:03  <LordAro> Samu: aha, that's a newline character (ish) in cmake
11:16:13  <LordAro> convert your stuff to use ':' like everything else
11:16:20  <LordAro> or that stuff
11:16:22  <LordAro> whatever
11:16:44  <Samu> you mean, a new PR changing all ; into :
11:17:05  <LordAro> well, in that output, yes
11:17:16  <LordAro> in both the regression AIs and the expected output
11:22:36  *** frosch123 has joined #openttd
11:26:08  <frosch123> i don't get 8552... doesn't that completely break cargodist?
11:26:43  <frosch123> the whole point of feeder services is to have intermediate stations that do not accept a cargo
11:28:13  <TrueBrain> I think this is one of these things that confuses new players mostly, that implicit all of a sudden can get cargo, from how I understand it
11:28:21  <TrueBrain> but I never played with cargodist, so what do I know :D :)
11:28:37  <frosch123> well, that is already solved now, by changing the default settings
11:28:57  <TrueBrain> non-stop is default now? Yeah, I guess that solves the issue too ;)
11:29:49  <frosch123> i think 8552 tries to turn non-non-stop to behave like non-stop, which is entirely pointless?
11:30:10  <TrueBrain> depends on the reason why implicits were created I guess
11:30:12  <michi_cc> Apparently cargodist will create a link if e.g. a coal train auto-stops at a pax-only station and will forcefully use it to transfer cargo between trains (that what I understood at least, never use no-non-stop orders myself).
11:30:17  <TrueBrain> is it done for cargodist?
11:30:57  <frosch123> implicit orders are for cargodist, yes
11:30:59  <TrueBrain> (I was honestly confused by the implicit orders, what they do and why they show up :D But I am a newbie player in those regards :P)
11:31:03  <TrueBrain> so yeah, that invalidates the PR :)
11:31:14  <andythenorth> everyone is confused by implicit orders
11:31:19  <andythenorth> but no better way was found :)
11:31:19  <frosch123> TrueBrain: without cargodist, the player decides where vehicles load, unload or transfer cargos
11:31:31  <frosch123> with cargodist, cargodist decides where to load, unload or transfer
11:31:36  <TrueBrain> I guess that is why I was confused what they did, the implicit orders :D
11:31:43  <frosch123> for that cargodist has to know where vehicles stop, so it needs implicit orders
11:31:49  <frosch123> to track intermediate stations
11:40:31  <Samu> is this a good commit name? Fix: No semicolon in regression output
11:40:51  <Samu> i'm unsure how to describe
11:41:00  <Samu> CMake uses ; to create newline
11:41:14  <LordAro> Samu: "Fix: Remove semicolons from regression output"
11:41:30  <LordAro> technically speaking, the cmake script should be escaping the ';'
11:41:34  <LordAro> but that's a separate problem
11:41:59  <Samu> i think we should wait for glx
11:42:05  <Samu> see what he has to say
11:42:46  <Samu> but ill make the PR anyway, just in case it's needed
11:43:01  <Samu> just dont merge asap before glx has a say
11:43:18  <LordAro> lol, sure
11:43:33  <TrueBrain> LordAro < glx, good to know :)
11:43:40  <LordAro> :c
11:44:36  <Samu> no, it's because it's due to a change yesterday night
11:44:41  <Samu> glx created
11:44:54  <TrueBrain> I am totally fine with LordAro < glx, it is not a complain, just a mere observation :D
11:45:05  <LordAro> wow.
11:45:09  <TrueBrain> <3
11:45:15  <LordAro> :p
11:45:29  <TrueBrain> sorry, that was just too easy :P
11:45:36  <LordAro> oh, i quite agree
11:45:40  <TrueBrain> :D
11:45:41  <LordAro> anyway, what needs reviewing/rebasing?
11:45:49  <TrueBrain> scroll patch of yours!
11:46:07  <LordAro> ah yes
11:46:10  <TrueBrain> <- basically, that :)
11:47:34  <DorpsGek> [OpenTTD/OpenTTD] frosch123 commented on pull request #8552: Feature: Prevent vehicles to unload cargo in implicit stops in cargodist mode if the cargo is not accepted in such station
11:48:09  <frosch123> do we have any cargodist expert left?
11:48:19  <TrueBrain> I promote michi_cc
11:48:34  <frosch123> i misread 8552 as changing the station acceptance rule. but it actually changes the input to the cargoflow planner
11:48:46  <michi_cc> I'm not a cargodIst expert at all.
11:49:08  <frosch123> so, my initial confusion was wrong, but i still do not know enough about that code :)
11:49:29  <TrueBrain> I never understood cargodist, sadly enough
11:49:35  <TrueBrain> same with cargodest
11:50:19  <michi_cc> I can tell you about cargodest; how I envision it to work when I finish it in 100 years :P
11:50:23  <DorpsGek> [OpenTTD/OpenTTD] SamuXarick opened pull request #8553: Fix: Remove semicolons from regression output
11:50:29  <TrueBrain> michi_cc: haha :D
11:50:38  * andythenorth votes don't touch cargodist, back away
11:50:42  <TrueBrain> is that before or after the economy revamp Celestar is working on?
11:50:58  <michi_cc> Hey, YACD is a patch, it's not vaporware :)
11:51:11  <frosch123> there was a svn branch for economy :p
11:51:22  <TrueBrain> there are PDFs for economy
11:51:24  <TrueBrain> like .. real papers
11:51:34  <LordAro> this->vscroll->GetCount() - this->vscroll->GetCapacity()
11:51:46  <LordAro> have i forgotten how scrollbars work, or is this always negative?
11:52:00  <TrueBrain> what does GetCount do?
11:52:03  <TrueBrain> total items in the list?
11:52:13  <TrueBrain> as in that case, no
11:52:25  <LordAro> ah right
11:52:26  <TrueBrain> well, it might
11:52:30  <frosch123> TrueBrain: yeah, sadly noone of them notices the fundamental problem: if you do not have enough money in ottd, you have to wait for money to pile up. so "trying to make the game harder" usually ends up with "making the game more boring by adding waiting times"
11:52:35  <TrueBrain> but these functions really confused me LordAro
11:52:59  * andythenorth never understood 'game needs to be harder'
11:53:04  * andythenorth is quite stupid though
11:53:30  <TrueBrain> frosch123: for me it always read that they confused "fun to play" with "has to be realistic"
11:53:30  <frosch123> andythenorth: do you remember that multiplayer ncg game with pikka
11:53:45  <frosch123> pikka was so interested to see people play with his set
11:53:55  <andythenorth> I remember a few
11:54:02  <frosch123> but everyone was just frustrated that there was barely enough money for one player to build something
11:54:07  <frosch123> while the rest was idling for two hours
11:54:11  <andythenorth> oh yes
11:54:15  *** WormnestAndroid has quit IRC
11:54:24  <andythenorth> those MP GS games massively influenced my design choices
11:54:28  *** WormnestAndroid has joined #openttd
11:54:34  <andythenorth> FIRS got redesigned to scale cargo output easily
11:54:57  <TrueBrain> LordAro: if GetCount does return the amount of items in the list, maybe a GetItemCount would be a better name, tbh :P
11:55:08  <frosch123> iirc pikka added a setting to his set after that: divide all costs by 8 or so
11:55:11  <andythenorth> and I tried to keep the needle of vehicle cost + power pointing at 'you can make progress without waiting'
11:55:21  <michi_cc> Good multi-player design goals are probably very orthogonal to good single-player design goals.
11:55:27  <andythenorth> game needs to be richer, not harder
11:55:34  <andythenorth> just my £0.02
11:55:57  <andythenorth> I just played a 160 years of FIRS Steeltown because there's enough depth to it
11:56:06  <andythenorth> being 'harder' wouldn't have kept me playing
11:56:25  * andythenorth needs to start a new game
11:57:15  <andythenorth> the next 2 weeks will contain a lot of questions about JGRPP
11:59:01  <reldred> There's ways to make a game more difficult, beyond just making everything more expensive. For instance I like the depth that can be added by well designed train sets; deciding what branch of a tech tree to go down, I like super tall maps with heighmaps that introduce varieties of regions with different topology, the chalenges you then faced by trying to traverse that (One of the reasons I really want to see cliffs, more object
11:59:01  <reldred> obstacles)
11:59:34  <reldred> God what's JGR up to now andythenorth? I hadn't really seen anything have I missed some news?
11:59:41  <andythenorth> I don't know
12:00:04  <andythenorth> I have tried JGRPP before, and I find it quite overwhelming
12:00:14  <andythenorth> but there are too many reasons to switch
12:00:50  <reldred> Yeah JGRPP is my regular, I'm particularly fond of custom bridgeheads and things like the tweaked alpine tree growth.
12:00:51  <andythenorth> playing it requires pretty much constant questions
12:01:03  <andythenorth> but fortunately the discord is almost entirely JGRPP
12:01:43  <andythenorth> I keep asking newgrf design questions in discord and the answers are constantly 'solved in JGRPP'
12:02:09  <andythenorth> apparently all the livery subtypes crap is fixed
12:02:10  <reldred> Yeah that doesn't sound constructive.
12:02:23  <andythenorth> I think it's fine no?
12:02:41  <reldred> No I mean just saying 'jgr fixed it' and not really helping any further.
12:02:54  <Timberwolf> The problem with "harder" is most of the obvious options add linear difficulty to a game with an exponential growth curve.
12:03:23  <DorpsGek> [OpenTTD/OpenTTD] frosch123 commented on pull request #8552: Feature: Prevent vehicles to unload cargo in implicit stops in cargodist mode if the cargo is not accepted in such station
12:03:40  <reldred> Yeah see I'm not interested in 'harder', I'm interested in challenges and solving them
12:03:47  <reldred> Rather than just a race against the time/money clock.
12:03:59  <Timberwolf> You still eventually hit the elbow where everything you do brings in >1 times what it cost. Making that 1.0001 instead of 1.1 just means more time spent in that "waiting for trains to arrive" stage of the game.
12:05:10  <andythenorth> it's unlikely to go well, asking "how do I do X in vanilla" to people who are like "wtf are you talking about, that is solved?" :)
12:05:13  <andythenorth> anyway
12:05:16  * andythenorth tries it
12:05:18  <Timberwolf> In original TTD and Locomotion, the real difficulty is getting overwhelmed by busywork on a large network.
12:05:31  <andythenorth> +1
12:05:38  <reldred> I mean I geuss it depends how you play it? I mean you can setup a low cost low risk pax route between two cities and put it on fast forward and hit 2050 with millions in the bank
12:05:55  <reldred> that's a pretty boring way to play
12:06:22  <reldred> I mean I consider anything other than alpinist/maxheight255/very rough terrain to be basically cheating. But that's me,
12:06:25  <reldred> It's a sandbox game.
12:06:26  <Timberwolf> For me, the difficulty is not the money but the inevitable "I need 100 trains to arrive here every minute and... er... HOW"
12:06:57  <andythenorth> the scarce commodity is 'throughput'
12:07:00  <andythenorth> not 'money'
12:07:14  <frosch123> i liked playing settlers 1 in the 90s. the main difficulty was to make the weird transport mechanics deal with the vast amount of cargo. but i also knew people, who just left the game running for 5 hours unattended, until they had enough stuff produced
12:07:24  <Timberwolf> Which gets more challenging (and "fun") on rugose terrain with many obstacles.
12:07:58  <andythenorth> frosch123 apparently
12:08:18  <Timberwolf> frosch123: I have a friend who plays Factorio like that sometimes.
12:08:37  <Timberwolf> "I need 18 million green circuits, I'll just leave it running overnight"
12:11:30  <reldred> I mean you're never going to stop someone playing that way if they want. Even if you have some sort of random disasters system to keep people engaged people will demand the ability to turn it off (I mean I probably would, I make life hard enough for myself in other areas).
12:14:55  <DorpsGek> [OpenTTD/team] frosch123 commented on issue #123: [de_DE] Translator access request
12:18:33  *** jottyfan has joined #openttd
12:20:35  <Samu> is this CMake error that kind of error that shouldn't happen?
12:22:45  <LordAro> that is a bit weird
12:23:26  <Eddi|zuHause> you're also never going to stop people building point-to-point routes in openttd, instead of complex networks
12:28:18  <DorpsGek> [OpenTTD/OpenTTD] LordAro updated pull request #8006: Codechange: Increase scrollbar length limit to UINT_MAX and make their length properly unsigned
12:28:20  <reldred> Yup
12:28:22  <LordAro> pretty sure that's incorrect in places
12:28:53  <reldred> it's a sandbox game, people are going to play it however they want and however hard they want.
12:29:23  <reldred> some people play almost entirely for aesthetics and efficiency or profits are entirely on the backseat
12:31:17  <TrueBrain> LordAro: am I correct to assume you did not process all feedback in #8006 yet?
12:31:34  <LordAro> ah, no
12:31:36  <LordAro> that was just a rebase
12:31:53  <TrueBrain> k :)
12:46:19  <LordAro> TrueBrain: i'm thinking that a simple "underflow safe unsigned subtract" function might be a good idea
12:46:52  <LordAro> given the amount of times i've written it (probably incorrectly in places)
12:47:07  <TrueBrain> that would be even better, yes :)
12:47:18  <TrueBrain> generalizing hard-to-read-statements is always a good thing :D
12:47:43  <TrueBrain> frosch123: <- with this latest change, do you see any additional issues with this PR?
12:49:55  <LordAro> TrueBrain: now the hard bit is finding all of them
12:50:04  <TrueBrain> haha, yes
12:50:08  <TrueBrain> ": 0" is a give-away :P
12:53:25  <Eddi|zuHause> i think technically "0-1" in unsigned is still considered an "overflow". and "underflow" is used in floating point contexts where very small numbers are rounded to 0
12:54:00  <TrueBrain> you remember when we talked about this channel being often overly pedantic with no real added value? :P :D
12:54:26  <Eddi|zuHause> that's an occupational hazard :p
12:54:36  <DorpsGek> [OpenTTD/OpenTTD] frosch123 approved pull request #8032: Fix #7904: Don't use a timer for hundredth tick determination
12:54:49  <TrueBrain> tnx frosch123 !
12:56:05  <frosch123> <- no need to be lonely, there is also stuff for you :)
12:56:16  <andythenorth> TrueBrain we talked about that?! Please, give me exact times and dates
12:56:28  *** DasPoseidon has joined #openttd
12:56:43  <TrueBrain> frosch123: :D
12:56:45  <TrueBrain> andythenorth: FU
12:56:53  <TrueBrain> <3
12:58:38  <andythenorth> I had a bet on how you would answer
12:58:43  <andythenorth> I was off by a single character
12:59:10  <TrueBrain> happy I did not disapoint
13:04:21  *** gelignite has joined #openttd
13:06:28  * andythenorth takes all the questions to discord
13:06:31  <frosch123> yay, i sent my first mail from the new mail account :p
13:06:32  <andythenorth> so many questions
13:06:50  <TrueBrain> frosch123: gratz!
13:06:56  <TrueBrain> how does that make you feel?
13:07:01  <frosch123> (it's so weird that credentialy of receiving and sending are independent)
13:07:22  <DorpsGek> [OpenTTD/OpenTTD] TrueBrain merged pull request #8032: Fix #7904: Don't use a timer for hundredth tick determination
13:07:27  <TrueBrain> I had that too the other day :P
13:08:13  <Eddi|zuHause> a propos, when i pushed the first time in a while, github sent me a "deprecation warning" email, should i worry?
13:08:26  <Eddi|zuHause> about logging in with password
13:08:48  <reldred> i think they're converting over to key based auth?
13:09:24  <reldred> which tbh kinda concerns me considering there's whole classes of malware now that focus entirely on compromising keys after AWS did similar
13:10:33  <TrueBrain> a key has a selected amount of permissions; so the malware no longer gets full access if it succeeds ;)
13:11:12  <TrueBrain> security is never about completely removing any risk; it is merely about reducing it to an acceptable and/or manageable level :)
13:11:27  <reldred> dunno, I memorize long passwords and would prefer to keep doing that.
13:11:48  <DorpsGek> [OpenTTD/OpenTTD] glx22 approved pull request #8553: Fix: Remove semicolons from regression output
13:20:32  <andythenorth> fuck me
13:20:42  <andythenorth> I am a big fan of having patch packs as experiments
13:20:46  <TrueBrain> we talked about, and we kinda have to decline your offer
13:20:47  <andythenorth> it really is the way forward
13:21:04  <andythenorth> but this template stuff, is absolutely non-working
13:21:11  <andythenorth> it destroys existing trains
13:24:20  *** emeraldsnorlax[m] is now known as EmeraldSnorlax[m]
13:26:23  <andythenorth> it has 'use depot'
13:26:34  <andythenorth> oops wrong
13:27:50  <LordAro> reldred: etc
13:28:29  <reldred> Lol
13:31:15  <andythenorth> this would be the worst livestream ever
13:31:20  <andythenorth> andythenorth tries to template-replace trains
13:31:30  <andythenorth> currently I am at 'wtf does it delete other trains for?'
13:31:34  <andythenorth> how is that useful?
13:31:38  <andythenorth> my face :P
13:34:04  <Timberwolf> This reminds me of the best excuse I ever heard for broken workflows, about 15 years ago when I was integrating some code which had been generated by an enthusiastic (but slightly lax on testing) developer in the Frankfurt office of place I worked at the time.
13:34:18  <Timberwolf> Sent an e-mail "I clicked on A then B, and got this crash".
13:34:29  <Timberwolf> Reply: "Germans would never click on things in that order."
13:37:05  <reldred> andythenorth: from my use of it you make templates, assign the templates to groups, and it auto replaces the trains in the group with the template. I didn’t really have too many problems with it.
13:37:20  <andythenorth> it absolutely just deletes trains
13:37:25  <andythenorth> I could video it doing it :)
13:37:43  <andythenorth> you have 4 trains in depot, set the templater going, now you have 3
13:37:50  <reldred> Huh, I dunno. I mean I use your sets almost exclusively.
13:47:11  <Samu> the regression savegame map doesn't have rivers, there's no way to test positive for IsRiverTile
13:47:19  <Samu> what do I do?
13:47:25  *** Progman has quit IRC
13:53:22  *** glx has joined #openttd
13:53:22  *** ChanServ sets mode: +v glx
13:56:53  <LordAro> Samu: checking that it tests negative in various places will do half the work
13:57:14  <LordAro> difficult to add rivers to such an old map really, we'd need to create a new map
13:58:06  *** sla_ro|master has quit IRC
13:59:36  <Samu> i iterated all tiles, just for the sake of it, returned false for every tile
13:59:59  <Samu> now im reducing to just a few tiles
14:00:12  <glx> yeah, but a valid test should return true in some cases
14:00:30  <glx> as in really checking the function works
14:17:44  <Samu> do you mind if i change the TileArea rectangle where the water tests are done in the map?
14:17:59  <Samu> i found a better spot
14:18:21  <Samu> it tests over a ship depot, a dock, a canal tile, then ground and sea
14:18:47  <Samu> the current rectangle is just ground and sea
14:19:06  <glx> if there is more different stuff it's better I think
14:19:19  <glx> and you can test all functions on the same area
14:20:16  <glx> but you need to carefully check the result is correct
14:21:34  <Samu> oh, good idea
14:30:28  <Samu> testing this area [img][/img]
14:30:49  <Samu> let me get the current area
14:32:56  <Eddi|zuHause> am i blind? i can't find a .h file that declares SndPlayTileFx
14:33:11  <Eddi|zuHause> ah, right. i'm blind
14:36:05  <Samu> current area: [img][/img]
14:37:12  <glx> indeed the "new" area seems a better test case
14:37:47  <Samu> too bad the buoy is deleted by the time it tests
14:38:44  <Samu> there's always that care IsWaterTile on the buoy returning false
14:38:47  <Samu> case*
14:39:52  <glx> you can update the buoy stuff
14:40:24  <DorpsGek> [OpenTTD/OpenTTD] Eddi-z commented on pull request #8444: Feature: Close adjacent level crossings as if they were one large crossing
14:41:42  <glx> like depot and dock are removed then rebuilt
14:41:57  <glx> you can rebuild the buoy too
14:43:19  <glx> and the lock
14:45:46  <DorpsGek> [OpenTTD/OpenTTD] Eddi-z updated pull request #8444: Feature: Close adjacent level crossings as if they were one large crossing
14:46:30  <andythenorth> is it naptime?
14:47:25  <DorpsGek> [OpenTTD/OpenTTD] glx22 merged pull request #8553: Fix: Remove semicolons from regression output
14:48:44  <DorpsGek> [OpenTTD/OpenTTD] Eddi-z updated pull request #8444: Feature: Close adjacent level crossings as if they were one large crossing
14:48:58  <Eddi|zuHause> andythenorth: is that before or after lunch?
14:49:04  <andythenorth> uncertain
14:49:42  <glx> ahah rebuilding buoy and dock changes some timings and affect vehicle age output
14:50:32  <Eddi|zuHause> so... crossing dingdingding. currently happens in two ways: a) a crossing with no PBS signals nearby dings when it is activated, b) a crossing with PBS signals dings when the train approaches
14:50:51  <Eddi|zuHause> should it always be case b)?
14:51:07  <glx> I vote b)
14:51:12  <glx> but maybe not easy
14:52:08  <glx> but dingding means closing for me
14:52:16  *** iSoSyS has joined #openttd
14:52:55  <Eddi|zuHause> well, currently, a PBS reservation closes the crossing, but doesn't ding
14:53:21  <glx> for me if it's closed it should ding
14:54:22  *** openttd_1 has joined #openttd
14:54:30  <openttd_1> hi!
14:54:33  <glx> ding from start of closing until fully open
14:55:11  <Eddi|zuHause> that sounds... annoying
14:55:31  <Eddi|zuHause> also, i don't think the current sound effect can loop
14:55:44  <glx> but realistic ;)
14:56:28  <openttd_1> I would like to do some data analysis on the ongoing game, I'm looking for some API of the game which could let me access the data of my busses , trains, other stuff which i costing money
14:56:47  <openttd_1> in the end I would like to create a sort of statistical dashboard
14:56:52  <openttd_1> ( just for fun )
14:57:13  <Eddi|zuHause> openttd_1: a GameScript that gathers the data and outputs it through either the debug console or adminport
14:58:17  <openttd_1> cool, so it's doable!
15:00:38  <openttd_1> I thought that GameScripts was only for AI
15:01:56  <glx> no they are similar, but very different
15:12:23  *** openttd_1 has quit IRC
15:14:39  <DorpsGek> [OpenTTD/OpenTTD] Eddi-z updated pull request #8444: Feature: Close adjacent level crossings as if they were one large crossing
15:29:48  <Samu> was wondering if I could include some aqueduct building into the area
15:30:18  <Samu> guess i rather not
15:30:43  <glx> will be a "big" bridge
15:30:58  <Samu> yes, then the area becomes too large, too many tiles printing
15:31:35  <glx> and bridge should not have effect on tile tests I think
15:32:43  <Samu> the IsWaterTile test
15:33:27  <DorpsGek> [OpenTTD/OpenTTD] SamuXarick updated pull request #8390: Add: [AI/GS] Missing water related functions and objects
15:35:05  <Samu> those age tests changed
15:37:26  <Samu> i verified the water, coast, canal tests, and yeah... they are "correct"
15:37:39  <Samu> at least, for what AI makers have to deal with
15:38:04  <Samu> ship depot is a Water tile, lock is a water tile
15:38:15  <Samu> but buoy is not
15:38:39  <Samu> coasts aren't water tiles
15:39:19  <Samu> even if they're halfcoasts
15:42:22  <Samu> lock isn't a canal
15:42:35  <Samu> which is another issue
15:42:40  <Samu> we have to deal with
15:43:30  <Samu> it's not straight forward what is what when dealing with water
15:52:55  <DorpsGek> [OpenTTD/OpenTTD] SamuXarick commented on pull request #8390: Add: [AI/GS] Missing water related functions and objects
15:59:12  *** Flygon has quit IRC
16:02:45  <DorpsGek> [OpenTTD/OpenTTD] Eddi-z updated pull request #8444: Feature: Close adjacent level crossings as if they were one large crossing
16:05:45  <DorpsGek> [OpenTTD/OpenTTD] Eddi-z commented on pull request #8444: Feature: Close adjacent level crossings as if they were one large crossing
16:10:29  <DorpsGek> [OpenTTD/OpenTTD] frosch123 opened pull request #8554: Allow NewGRF vehicles to query the track-type NewGRF about the current track-type
16:11:09  <andythenorth> o_O
16:12:39  <DorpsGek> [OpenTTD/OpenTTD] frosch123 commented on pull request #7000: Some NewGRF variables concerning railtypes
16:12:42  <DorpsGek> [OpenTTD/OpenTTD] frosch123 closed pull request #7000: Some NewGRF variables concerning railtypes
16:12:43  *** openttd_1 has joined #openttd
16:14:42  <openttd_1> is there a way to send an httprequest with a game script?
16:15:24  <frosch123> the only connection from GS to the outside world is the admin port.
16:15:44  <frosch123> you can write a "proxy" that connects to the admin port and provides the stuff you need
16:15:54  <Xaroth> i was about to type: inb4 somebody suggests a proxy
16:15:58  <Xaroth> but you beat me to it, frosch123
16:16:07  <frosch123> i believe there is even a library to execute GS commands that were sent from the admin port
16:16:31  <Xaroth> yeah
16:17:18  <frosch123> openttd_1:
16:20:08  *** Wormnest has joined #openttd
16:21:32  <DorpsGek> [OpenTTD/OpenTTD] Eddi-z commented on pull request #8554: Allow NewGRF vehicles to query the track-type NewGRF about the current track-type
16:22:24  <DorpsGek> [OpenTTD/OpenTTD] frosch123 updated pull request #8554: Allow NewGRF vehicles to query the track-type NewGRF about the current track-type
16:22:30  <openttd_1> frosch123 that's interesting, I didn't know about the admin port
16:22:39  <frosch123> i am no longer user to ottd coding style :)
16:23:53  <frosch123> *used
16:24:34  <andythenorth> so I need something like getbits(var[0x63, 8, 0xFF, ELRL]), 2, 1)
16:24:54  <glx> oh your PR adds an unrelated fix ;)
16:25:08  <frosch123> openttd_1: <- there are plenty of libraries to interface it
16:27:18  <andythenorth> hmm ELRL like that isn't a valid param
16:27:52  <glx> there should be a label thing
16:28:19  <andythenorth> getbits(var[0x63, 8, 0xFF, railtype(ELRL)], 2, 1) might be better
16:28:25  <andythenorth> with fewer syntax errors also
16:28:52  *** iSoSyS has quit IRC
16:28:54  <openttd_1> frosch123: thanks a lot, that's exactly what I was looking for
16:29:03  <Eddi|zuHause> andythenorth: ELRL is the index, railtype(ELRL) is the label, or something
16:29:39  <Eddi|zuHause> andythenorth: also, you don't need getbits, you adjust the "8" and "0xFF"
16:29:40  <glx> railtype is the index in translation table
16:29:40  <andythenorth> thanks, the problem was the syntax error
16:32:09  <andythenorth> Eddi|zuHause so to pick bit 2?
16:33:25  <Eddi|zuHause> var[0x63, 2, 1, ...] i believe
16:33:47  *** openttd_1 has quit IRC
16:35:27  <glx> var[<num>, <shift>, <mask>, <param>]
16:35:43  <Eddi|zuHause> yes. you shift 2 bits, and mask 1 bit
16:36:15  <andythenorth> confirmed bit 2 works as expected with nutracks
16:36:22  <andythenorth> I'll do a more complete test
16:36:38  <andythenorth> not sure how I'll do backwards compatibility with older openttd
16:36:45  <andythenorth> maybe branch switches?
16:36:47  *** Speedy` has quit IRC
16:38:52  <Eddi|zuHause> does old openttd care about unknown variables?
16:39:06  <Eddi|zuHause> also, why?
16:39:23  <Eddi|zuHause> old openttd can still use old newgrf version
16:39:47  <andythenorth> just use bananas min. compatible?
16:39:51  <andythenorth> that's what FIRS does
16:40:20  <glx> openttd returns 0xFFFFFFFF for unknown
16:40:40  <Eddi|zuHause> yes, then they won't be able to update unless they also update openttd.
16:41:52  *** Speedy` has joined #openttd
16:43:07  <andythenorth> that might coincide nicely
16:43:36  <andythenorth> I am going to use the inflation warning, and put railtypes into Horse that rely on the curve speed fix
16:43:43  <andythenorth> so 1.11 min. then
16:44:50  <andythenorth> more UK royal engine liveries for frosch123, who is now interested in them :P
16:44:56  <andythenorth> now / not /s probably
16:45:05  <andythenorth> 'pls send grf with that livery'
16:47:21  <Eddi|zuHause> frosch123: right, i get that #7000 didn't age well, and mixed too much stuff, and didn't handle NRT on account of not having existed yet. ... but you dropped all the Action7 stuff, and complain that there's no usecase when there was...
16:50:05  <glx> can be added later I guess
17:08:02  *** virtualrandomnumber has joined #openttd
17:08:02  *** virtualrandomnumber has quit IRC
17:08:38  *** qwebirc63162 has joined #openttd
17:10:29  <michi_cc> andythenorth: Could I bother you for an OSX test whenever convenient? I noticed some strange things which might or might not be caused by me using only a virtual machine.
17:10:35  <Eddi|zuHause> also RE: "please consider all cases before submitting a PR" ... this isn't ever going to happen. i have a silly idea, i code it, submit it publicly for comments, and then refine it until it is either considered "finished" or i lose interest. but if i don't act on this first impulse, nothing is ever going to happen
17:11:25  <Eddi|zuHause> and i don't think draft PRs existed back then
17:15:04  *** qwebirc63162 has quit IRC
17:15:26  <DorpsGek> [OpenTTD/OpenTTD] saveinthesky commented on issue #8501: OTTD crashes on start macosx . Log
17:21:13  *** WormnestAndroid has quit IRC
17:21:28  *** WormnestAndroid has joined #openttd
17:25:17  <andythenorth> michi_cc is there a PR? :)
17:25:33  <michi_cc> andythenorth:
17:25:44  <andythenorth> ok
17:25:55  <michi_cc> There is no PR, I've just noticed strange things for me. But a VM might mess with timing.
17:26:37  * andythenorth always forgets p1 instead of p0
17:26:43  <michi_cc> Things to watch for: frame rate (in the framerate window) right now and after applying the patch, both in normal and in fast forward.
17:27:05  <andythenorth> for current master, or will a recent binary do?
17:27:07  <michi_cc> I get a totally inconsistent behaviour for something that should be absolutely identical.
17:27:15  <michi_cc> Recent will do
17:29:04  <michi_cc> With the patch I get a lot more FPS in fast forward, but also a higher variance in FPS in normal speed. Which is very illogical.
17:29:34  <frosch123> andythenorth: using "unknown variables" will pick the first switch option
17:29:48  <andythenorth> thanks
17:29:50  <frosch123> not sure whether you can control what becomes the first option in nml
17:30:07  <andythenorth> I'll just bump openttd version, assuming this PR makes 1.11
17:31:08  <Eddi|zuHause> you used to be able to check for 1.11-alpha vs release, but i don't know if that survived the github move
17:32:28  *** jottyfan has joined #openttd
17:32:33  <andythenorth> did we add a chart to FPS?
17:32:40  <andythenorth> or was that the mac profiler I used?
17:32:44  * andythenorth can't remember
17:32:48  <michi_cc> Click on the respective time.
17:33:46  <glx> <Eddi|zuHause> you used to be able to check for 1.11-alpha vs release, but i don't know if that survived the github move <-- still available, only svn rev stuff is no longer updated
17:33:52  <andythenorth> thanks
17:36:53  <michi_cc> andythenorth: I don't need any scientific study, just saying same or different is quite enough for now :P
17:37:20  <michi_cc> I.e. just new game, empty map.
17:45:08  *** snail_UES_ has joined #openttd
17:45:34  * andythenorth BRB - cooking :)
17:56:50  <andythenorth> michi_cc nothing obviously different between the gist and 1.10.2
17:57:25  <DorpsGek> [OpenTTD/OpenTTD] frosch123 updated pull request #8554: Allow NewGRF vehicles to query the track-type NewGRF about the current track-type
17:59:14  <andythenorth> 1.10.2 pushes a higher graphics FPS generally, but there are other variables like map etc
17:59:28  <andythenorth> I have a release build of the gist
17:59:42  <michi_cc> It's probably some virtual machine thing. But I get a fastforward (without full anim) of 200 fps normally and like 1000 fps with the patch.
18:00:04  <michi_cc> Might be differences in virtualized CPU tick timing though.
18:01:36  <andythenorth> oh this is weird
18:01:41  <andythenorth> with full anim off
18:01:44  * andythenorth restarts the game
18:02:22  <andythenorth> michi_cc with full anim off, it runs so fast it's in permanent autosave :P
18:02:29  <andythenorth> and graphics FPS shows 50
18:02:43  <michi_cc> Okay,  that matches my observations.
18:03:02  <michi_cc> I have more fps fluctuations in normal speed though.
18:03:24  <andythenorth> ok with autosave off, it's running insanely fast
18:03:31  <andythenorth> but mostly failing to update the UI
18:03:48  <andythenorth> yeah it will do about 1 year / s
18:03:59  <andythenorth> but graphics FPS is 10
18:04:25  <andythenorth> (this is with ffwd)
18:04:26  <michi_cc> Okay, 10 is a bit low.
18:04:40  <andythenorth> it fails to track mouse etc as well
18:04:46  <andythenorth> it's like UI is desynced
18:05:33  <andythenorth> FPS 0.98 :)
18:05:38  <andythenorth> after a sustained FFWD
18:07:31  *** DasPoseidon has quit IRC
18:08:49  <michi_cc> Okay. One more test. I but a second diff on, should not conflict with the first.
18:09:59  <michi_cc> Use it in addition, not instead of.
18:10:31  <andythenorth> 'assume -R' ?
18:10:53  <michi_cc> Hmm, no???
18:11:02  <andythenorth> thought not
18:11:46  <andythenorth> these diffs look the same in raw
18:12:00  <michi_cc>
18:12:02  <andythenorth> oh I see
18:12:06  <michi_cc> The first on is completely unchanged.
18:12:23  <andythenorth> I missed the second, raw gives each diff, not the whole gist , my mistake
18:13:54  <TrueBrain> holy crap, reading that new PR makes me realise I really really do not understand NewGRFs :P Well, code-wise it looks good, best I can tell :D
18:15:05  <TrueBrain> some mad respect to anyone who understands this :D
18:16:22  <Eddi|zuHause> i think nobody completely understands it, which is partly why #7000 went so out of hand. everyone understood a different thing, or the same thing differently...
18:18:27  <andythenorth> michi_cc behaviour with second diff is similar
18:20:13  *** sla_ro|master has joined #openttd
18:20:56  <andythenorth> is it just concidence that full anim makes such a difference to this specific behaviour?
18:21:03  <andythenorth> coincidence *
18:22:36  *** qwebirc37973 has joined #openttd
18:24:05  <DorpsGek> [OpenTTD/OpenTTD] SamuXarick updated pull request #8468: Fix #8316: Make sort industries by production and transported with a cargo filter possible
18:24:28  <DorpsGek> [OpenTTD/OpenTTD] TrueBrain approved pull request #8554: Allow NewGRF vehicles to query the track-type NewGRF about the current track-type
18:25:42  <TrueBrain> the | for return values confused me for a bit .. kept reading || :D
18:25:50  <TrueBrain> funny, how a mind fills in things that are not there :)
18:25:52  <Samu> TrueBrain	::SetDParam(0, 1 << cargo_type); visual studio complains with  warning C4334: '<<': result of 32-bit shift implicitly converted to 64 bits (was 64-bit shift intended?)
18:25:56  * andythenorth wonders what to call the nml vars
18:26:53  <Eddi|zuHause> TrueBrain: i also don't always manage to separate | vs ||
18:27:12  <andythenorth> maybe "var[0x63, x, y, LABL]" is good enough
18:27:17  <andythenorth> :P
18:27:30  <TrueBrain> call it the andy-var? :D
18:27:34  <TrueBrain> I just like calling things silly
18:28:01  <glx> Samu: it should be a 64bit value, probably needs explicit
18:28:23  <andythenorth> nml authors have it too easy, they're not learning enough hex
18:28:24  <frosch123> TrueBrain: feel free to add "provide a newgrf that proofs that this is usable at all" to the PR template :p
18:28:43  <TrueBrain> haha, that is not a bad idea honestly :P
18:28:55  <frosch123> 7000 contained so much stuff, that made so little sense, that if i would have to write a newgrf to use it, i would rather quit trying
18:28:59  <Eddi|zuHause> andythenorth: adding nml vars is easy, just remember the "x" and "y" and give it a name
18:29:01  <TrueBrain> but I am afraid people really than give a .grf, instead of anything I can read :)
18:29:02  *** qwebirc37973 has quit IRC
18:30:31  <Eddi|zuHause> frosch123: andy had a test grf, before it grew out of hand
18:30:39  <michi_cc> andythenorth: Okay, then I will shelve that for now. Fast fast-forward is nice, but a broken GUI is not.
18:30:49  <andythenorth> it's very fast :)
18:31:17  <andythenorth> 8554 needs added to
18:31:24  <andythenorth> I can do it later if that helps
18:31:42  *** Progman has joined #openttd
18:32:00  <DorpsGek> [OpenTTD/OpenTTD] DorpsGek pushed 1 commits to master
18:32:01  <DorpsGek>   - Update: Translations from eints (by translators)
18:32:15  <TrueBrain> you think frosch123 is going to forget that andythenorth ? Thy of little faith :D
18:32:29  <frosch123> andythenorth: this time i use strike-through in the review-TODO list, like i saw in other PR. not sure whether that makes it more readable though
18:32:48  <TrueBrain> we should use "- [ ]" I guess
18:32:51  <TrueBrain> so it can be checked off :D
18:33:04  <Samu> "show potential fixes" tells me to do this: ::SetDParam(0, static_cast<uint64>(1) << cargo_type);
18:33:06  <andythenorth> I misread approved and merged, my silly
18:34:04  <TrueBrain> :D
18:34:25  <Eddi|zuHause> frosch123: the wording made it sound like that striking should be done by the reviewers instead of the pr authors
18:35:01  <andythenorth> I am glad I don't have shoelaces in my shoes, I'm not sure I could tie them some days
18:35:22  <TrueBrain> I'm not sure others think I tied them, you meant :P
18:35:23  <TrueBrain> :D
18:36:09  <frosch123> i learnt that there are two methods to tie shows 2 years ago
18:36:21  <frosch123> and i actually switched the method
18:36:30  <TrueBrain> its crazy, isn't it? :D
18:36:40  <frosch123> didn't expect people at my age would be capable to switch the habit of such a basic thing
18:37:20  <Eddi|zuHause> i've used 2 main methods to tie shoe laces in my life. in the beginning, i sat there smiling, until someone came around to tie them for me, and later i tied them once and never untied them
18:38:51  <frosch123> my niece was scolded in 1st grade in school because she helped others close the zipper of their jackets, so they would never learn to do it themself
18:40:05  <Eddi|zuHause> my mom always tells this story how she came to pick me up from kindergarten, and said "tie your shoes", and i did that. then the teachers there were suprised "wait? he can do that?" :p
18:42:05  <andythenorth> I think 11 years later (almost to the day)
18:42:07  <andythenorth> is a good time
18:42:12  <andythenorth> to revisit this
18:48:05  <frosch123> you edited it in 2017 :p
18:49:11  <andythenorth> so I did
18:49:12  <frosch123> "ford tiles" sounds like the reverse of hovercraft
18:49:13  <andythenorth> "multi-mode vehicles, e.g. road-rail; hovercraft on land, amphibious trucks etc."
18:49:29  <andythenorth> there are so many 'done' ponies to add though
18:49:48  <frosch123> did you watch some western movies at the time? or do you know roadvehicles which commonly drive through water?
18:50:10  <glx> we have hovercraft on land ;)
18:50:30  <Eddi|zuHause> "ford-tiles" would be a special case of "objects with road state machine"
18:50:38  <andythenorth> I honestly can't remember frosch123 :)
18:50:56  <andythenorth> it was probably an excuse to justify bulldozers or something
18:53:56  <andythenorth> I suspect this video caused it, date is close (turn sound off)
18:54:00  <DorpsGek> [OpenTTD/OpenTTD] SamuXarick opened pull request #8555: Fix 218f40e: Warning about 32-bit shift implicitly converted to 64 bits
18:54:06  * andythenorth awaits SpinTires mac build
18:54:43  <Samu> visual studio added a newline at the end of file
18:54:50  <Samu> can't make it not do it
18:58:06  *** DasPoseidon has joined #openttd
18:59:14  <FLHerne> andythenorth: I think that one would just underscore how incompatible OpenTTD's vehicle scales are
18:59:25  <FLHerne> tram-trains might just about work
18:59:52  <FLHerne> Hovercraft or amphibious trucks would look silly
19:00:11  <andythenorth> FLHerne Hovercraft would never happen
19:00:16  <andythenorth> game engine won't support it
19:00:30  <FLHerne> I thought frosch did it?
19:00:35  <FLHerne> For a given value of 'it
19:00:37  <FLHerne> '
19:00:41  <andythenorth> I might be lying, for effect
19:01:12  <FLHerne> And that demonstration proved that it does indeed look really silly :p
19:02:06  <andythenorth> we should be more adventurous
19:02:15  <andythenorth> belts for the next livestream, or go home
19:07:50  <TrueBrain> inserts!
19:10:07  <andythenorth> or I could just make a pipe grf with NRT
19:10:37  * andythenorth wonders about 'no junctions' somehow, but how to attach depots then?
19:15:08  <frosch123> "no junctions" is fine for towns, but silly for humans
19:15:37  <frosch123> AIs are somewhere inbetween
19:19:50  <andythenorth> I wonder if RL pipelines have junctions
19:19:53  <andythenorth> I presume so
19:19:58  <andythenorth> just some valves I guess
19:21:35  <frosch123> of course they have junctions
19:22:08  <frosch123> gas pipelines from russia do not daisy-chain through all of europe
19:22:25  <frosch123> there are different types of valves
19:22:45  <frosch123> proper one which can be remote-controlled and which can be opened in percentages to regulate flow
19:23:05  <frosch123> and cheap ones all over the place to isolate area that need maintainance
19:23:48  <frosch123> though you keep the gas pressure on the pipe while working on the outside
19:24:01  <frosch123> it's better to have pressured gas inside, that leaks at some points
19:24:28  <frosch123> than insufficient pressure inside that may make oxygen leak inside
19:25:30  <frosch123> though if needed, yuo can also fill the inside with pressured nitrogen. complicated but with both advantages
19:25:44  <frosch123> does it show i wrote software for that once?
19:29:37  <andythenorth> maybe you just like pipelines :)
19:30:06  * andythenorth contemplates one pipe from Russia per destination, TTO style
19:37:32  <DorpsGek> [OpenTTD/OpenTTD] TrueBrain commented on issue #8501: OTTD crashes on start macosx . Log
19:56:28  <andythenorth> Update!
19:56:39  <andythenorth> I snuck a made one in an at the end
19:56:41  <andythenorth> mad *
19:57:46  <FLHerne> "...and use it to offer an additional cargo"
19:58:04  <andythenorth> that's not the mad one :)
19:58:12  <FLHerne> Can't we just *not* use absurd hacks that made sense when binary-patching a game written in 1990s asm?
19:58:38  <andythenorth> you have no respect for tradition :D
20:00:54  <Timberwolf> Bring back the creating a sign named "cht: x y" interface.
20:01:23  <andythenorth> is my pony list open for contributions
20:01:31  <andythenorth> should I have made a discussion first? :P
20:02:45  <andythenorth> FLHerne you think 16 separate holds might be better? o_O
20:02:56  * andythenorth dreads to imagine the cargo sprite generation for that :P
20:04:33  <Timberwolf> Good news is I made about 2 hours of videos using #8540 today without anything untoward happening sprite-wise.
20:06:04  <andythenorth> \o/
20:06:17  <andythenorth> hmm I probably don't to rerun cmake when I pull remote?
20:06:19  <andythenorth> in most cases?
20:10:06  <Timberwolf> There is an entirely unprompted "that looks good and industrial doesn't it? Good job Andy on that, that's very well done. Let's just admire the industrialness of that" in one of them after I scroll to a carbon black plant for the first time :)
20:10:23  <glx> <Samu> visual studio added a newline at the end of file <-- visual studio is right, we used to check/enforce newline at end of file, dunno if it's still the case
20:10:41  <Timberwolf> It is a very good sprite!
20:13:36  <DorpsGek> [OpenTTD/OpenTTD] glx22 approved pull request #8555: Fix 218f40e: Warning about 32-bit shift implicitly converted to 64 bits
20:14:01  <andythenorth> Timberwolf :)
20:14:20  <andythenorth> I got bored of all FIRS industries being copy-paste-o same textures
20:14:31  <andythenorth> I pushed FIRS 4 to be more diverse
20:15:05  <andythenorth> TrueBrain so JGR official builds when? (or no?) :)
20:16:10  <frosch123> did you figure out the train templates?
20:17:10  <andythenorth> somewhat
20:17:15  <andythenorth> JGR has fixed a few things
20:17:26  <glx> andythenorth: he already knows we can provide builds if he wants
20:17:26  <andythenorth> the templates are *not* something we could PR
20:17:38  <andythenorth> the templates are quite interesting though
20:18:02  <andythenorth> the functionality of "I don't have to yak-shave to change my trains" is nice
20:18:22  <frosch123> i thought yak-shaving was your favorite hobby
20:18:31  <andythenorth> yak-shaving grfs is my favourite hobby yes
20:18:37  <andythenorth> the implementation gets in the way, and the UI design fails to follow any of the 19 or so conventions we already have
20:18:45  <andythenorth> which is remarkable, but no surprise I guess
20:19:12  <andythenorth> I think templates would be really nice
20:19:27  <andythenorth> I also think that the hard part of any version of newgrf liveries is auto-replace
20:19:36  <andythenorth> and the hard part of templates is auto-replace :P
20:19:49  <andythenorth> if there are templates...does auto-replace stay or go?
20:20:24  <glx> auto replace is a hard part by itself ;)
20:21:00  <glx> but it "works"
20:21:08  <andythenorth> it's really remarkable
20:21:21  <andythenorth> auto-replace is a really good part of openttd
20:21:53  <glx> minus the confusion between renew and replace
20:22:08  <andythenorth> I asked the discord realism grf liveries-are-cargo-subtypes crowd how they manage the horror of subtypes UI
20:22:25  <andythenorth> and they are all just like "JGR templates, not an issue mate, just move on"
20:23:00  <andythenorth> OTOH, FIRS is pretty much going to end up banning every realism train grf
20:23:19  <andythenorth> because they are using cargos for liveries, which breaks one or the other
20:23:35  <andythenorth> CZTR is already banned, which is going to get me grief
20:23:52  <andythenorth> RUKTS is getting banned next, then I need to find the others
20:24:17  <andythenorth> breaks XIS and AuzInd too, so I might help those authors downstream my changes
20:24:34  <TrueBrain> glx: static cast around 1? Isn't that just 1LL or something?
20:24:43  <DorpsGek> [OpenTTD/OpenTTD] frosch123 merged pull request #8554: Allow NewGRF vehicles to query the track-type NewGRF about the current track-type
20:24:49  <andythenorth> it's been a while since a proper newgrf author drama :(
20:25:04  <glx> maybe just need a new engine property with some kind of list of supported livery to choose from, and a variable to access the choice from action 2
20:25:15  <frosch123> andythenorth: you could just skip slot 31
20:25:32  <Timberwolf> andythenorth: Want to ban Trains? If you do it on the basis of zoom compatibility you have a good chance of me actually being angry about it :)
20:25:36  <frosch123> or we could make ottd support multiple cargo grfs
20:25:55  <andythenorth> Timberwolf you said you don't abuse cargo subtypes though :D
20:25:58  <andythenorth> zoom is pff
20:26:00  <frosch123> but the latter only makes sense if not every industry set maxes every cargo type limit with 3 days
20:26:11  <Timberwolf> Indeed. FIRS compatibility is a deliberate objective.
20:26:26  <glx> TrueBrain: ha yes maybe a cast is not needed
20:26:29  <andythenorth> 2x FIRS Timberwolf fork when?
20:26:36  <andythenorth> can't be hard
20:26:45  <TrueBrain> glx: can't test why msvc is being stupid, but a static cast over a number feel very weird :p could you see what the proper syntax is?
20:26:47  <andythenorth> you don't have to design anything, and all of FIRS looks same
20:27:03  <andythenorth> TimberFIRS
20:27:11  <andythenorth> singularity happens
20:27:23  <TrueBrain> Tnx :D
20:27:30  <Timberwolf> We already have TimberFIRS, it's the one which relentlessly chucks out 6400 units from secondary industries no matter what.
20:27:40  <andythenorth> TimberFIRS 2
20:28:29  <andythenorth> glx yeah, pretty much, or do it on engine IDs and not a property, but provide UI to group them.  Either way, auto-replace is the hard part
20:29:07  <andythenorth> like, are liveries [some word I don't know like extrinsic or transcendent]?
20:29:27  <andythenorth> if I have Super Engine in DB colours, can I replace it with Mega Engine and it automatically goes to DB colours?
20:29:28  <glx> well auto replace can try the same livery first, and if failing it will use the first available
20:29:43  <andythenorth> and then do we need strings for each cargo and crap?
20:29:48  <andythenorth> livery / cargo /s
20:31:11  <frosch123> TrueBrain: 1ull ?
20:31:58  <andythenorth> I really like the 'group vehicles in the buy menu' route, but I think it's a bit unix worse-is-better thinking
20:32:07  <andythenorth> grf authors seem really hostile to it when discussed
20:32:17  <andythenorth> but auto-replace would
20:32:41  <andythenorth> it's one property and some UI crap
20:33:11  <TrueBrain> Unsigned makes more sense yes :D
20:33:15  <andythenorth> and pooling certain stats like reliability and availability
20:33:25  <glx> I think they are hostile because it means "rewrite" their grfs ;)
20:34:04  <glx> current livery stuff is just abusing around limitations
20:34:22  <frosch123> andythenorth: do you know whether silicon valley works with 64 cargos?
20:34:32  <andythenorth> frosch123 no but I need to play a test game anyway
20:34:37  <andythenorth> might be time
20:34:45  * andythenorth considers a fork
20:34:51  <frosch123> i think i have the same "1 << cargo" in sv. and i have no idea whether our squirrel does 64bit ints
20:35:20  <frosch123> and i have really surprised there is no other case of it in ottd
20:35:25  <frosch123> *am
20:35:41  <glx> tried 1LL and 1ULL, both work
20:35:46  <andythenorth> I'll start a SV map and see
20:35:51  *** jottyfan has quit IRC
20:36:03  <glx> let's check gcc quickly
20:36:09  <glx> (but should work too)
20:36:33  <frosch123> ah, upper-case is a good hint :)
20:36:44  <frosch123> so yes, we have multiple (1LL << foobar) in ottd
20:37:06  <andythenorth> how many openttd checkouts do others have?
20:37:16  *** Wormnest has quit IRC
20:37:24  <frosch123> one git, 2 old hg, 5 old svn
20:37:30  <frosch123> the latter with whatever modifications in them
20:37:51  <frosch123> but there were 5 hg and 12 svn once, so cleaned up some of them :p
20:37:54  <glx> I don't remember how many hg and svn I still have
20:38:30  <glx> git is so easier, no need to duplicate checkout to work on something else
20:39:04  <andythenorth> I have vanilla and jgr
20:39:08  <frosch123> oh, i forgot the checkouts of the release branches. but meh
20:39:32  <andythenorth> I keep pulling PRs to test and breaking my savegames due to forgetting :P
20:39:48  <andythenorth> but I don't think 2 checkouts will improve my memory
20:40:41  <glx> haha there's even         SQInteger mask = 0x00000001LL << i;
20:42:22  <andythenorth> frosch123 if SV didn't work, what might I see? :P
20:42:23  <glx> and we also have 39 1ULL occurences
20:43:03  <frosch123> andythenorth: it would fail to name the cargo that you have to produce
20:43:06  <frosch123> just blank
20:43:14  <andythenorth> if ID was above 32?
20:43:32  <frosch123> or maybe it would name the wrong one, shifted by 32
20:43:41  <michi_cc> SetDParam takes uint64, so ULL I'd think.
20:43:48  <frosch123> so if you have to produce cargo 34, it would tell you cargo 2, or so :p
20:44:13  <glx> I think it would tell 0
20:44:16  <andythenorth> seems to work afaict
20:44:49  <michi_cc> typedef __int64 SQInteger
20:44:59  <andythenorth> it's given me Vehicle Parts with ID 45
20:45:42  <frosch123> michi_cc: cool, i was worried we switched to int32 for everyone, to make 32bit and 64bit behave the same
20:45:50  <andythenorth> it's detected an industry I built correctly producing VPTS
20:46:08  <DorpsGek> [OpenTTD/OpenTTD] glx22 requested changes for pull request #8555: Fix 218f40e: Warning about 32-bit shift implicitly converted to 64 bits
20:46:54  * andythenorth considers an nml to m4nfo code generator for stations, 
20:47:04  <andythenorth> oh I think timberwolf already suggested that?
20:47:29  <Timberwolf> Heh. No, I wrote my own weird sidetrack for stations.
20:47:37  <michi_cc> I think 32-bit compilers that don't have a 64-bit type are not thing for some time now :)
20:49:38  *** tokai has joined #openttd
20:49:38  *** ChanServ sets mode: +v tokai
20:50:06  <DorpsGek> [OpenTTD/OpenTTD] SamuXarick updated pull request #8555: Fix 218f40e: Warning about 32-bit shift implicitly converted to 64 bits
20:50:55  <Samu> hi
20:51:12  <DorpsGek> [OpenTTD/OpenTTD] glx22 approved pull request #8555: Fix 218f40e: Warning about 32-bit shift implicitly converted to 64 bits
20:51:21  <andythenorth> glx fancy trying a rough prototype of this to see if it's rubbish?  :P
20:51:24  <andythenorth> I could provide a grf
20:53:11  <andythenorth> prototype could just group vehicles where the name string matches
20:54:58  <glx> I hate touching GUI ;)
20:56:18  * andythenorth wonders where GUI even is
20:56:26  <andythenorth> must be some widgets.cpp or something
20:56:44  <andythenorth> lol there is widget.cpp
20:58:38  <glx> I think it's in build_vehicle_gui.cpp
20:58:56  <andythenorth> tree thing is not widgetised I think
20:59:28  <andythenorth> the existing subtypes tree drawing is in vehicle_gui.cpp L454 and so
20:59:43  <andythenorth> tree is not essential, just hide / show variants on click
20:59:57  * andythenorth "my first C++ patch" time
21:00:04  *** WormnestAndroid has quit IRC
21:00:10  <andythenorth> this always goes...great
21:01:13  *** WormnestAndroid has joined #openttd
21:01:24  <glx> DrawEngineList() handles the drawing
21:01:50  <andythenorth> thanks
21:02:04  <andythenorth> guess it needs fed a list of engines
21:02:22  <andythenorth> hmm, the existing hide/show might give clues
21:02:24  * andythenorth new samu
21:03:57  <glx> and this draw function is shared between autoreplace and buy
21:04:21  <andythenorth> yup, I saw the 'crashed' palette use :)
21:04:41  <andythenorth> no idea how to associate engines in eng_list
21:05:02  <andythenorth> does that get split up so it contains a list of lists?
21:05:16  <andythenorth> or is grouping a property on each engine, read when drawing?
21:05:50  <andythenorth> list of lists means rebuilding all sorting
21:05:53  <andythenorth> so 'nope'
21:06:40  <glx> the list is sorted (by criteria selected in GUI)
21:07:06  <andythenorth> maybe something like engine has parent ID / none
21:10:37  <glx> but adding "group by" may need to create a list of lists
21:12:03  <glx> that's where a livery property would help a lot ;)
21:12:11  <andythenorth> yes
21:12:22  <andythenorth> I am +/-0 on the IDs idea
21:12:40  <glx> 1 engine ID, many liveries, naturally grouped
21:13:26  <andythenorth> it does seem the simplest overall
21:13:35  <andythenorth> overwhelmingly, grf authors want to set liveries
21:13:50  <andythenorth> the abuse of subtypes for stats changes is minimal (I do it)
21:13:54  <glx> and maybe a new feature to define liveries specs, like name
21:14:06  <andythenorth> livery labels :P
21:14:10  <andythenorth> common livery scheme :P
21:14:12  <andythenorth> nope
21:14:57  <glx> stats can still be modified via CB36 depending of selected livery
21:16:13  <glx> I think it's the simplest solution for both newgrf authors, and openttd devs
21:16:53  <andythenorth> I was trying to avoid combinatorial crap like "BR Blue 87mph, air brakes"
21:16:56  <andythenorth> but yes
21:16:59  <andythenorth> agree
21:18:14  <glx> in most cases I guess all stats are the same, just the visual changes
21:19:43  <andythenorth> the changing stats idea...rarely works IMHO
21:19:47  <andythenorth> I've done it 2 or 3 times now
21:20:26  <andythenorth> it's nice idea for short buy menu, but bad gameplay
21:20:29  <glx> probably more effective to just create a new engine for different stats
21:21:14  <andythenorth> yup
21:21:23  <andythenorth> I'm removing refittable capacities from ships currently
21:21:45  <andythenorth> no station refit support, no auto-replace support, just kind of fail :)
21:23:28  <TrueBrain>
21:23:35  <TrueBrain> Pretty decent
21:24:24  <glx> haha but if it's not merged, it stays open for years ;)
21:24:41  <TrueBrain> No no, read the fine print on that page
21:24:52  <TrueBrain> They take that into account t pretty nicely
21:25:13  <glx> based on 161 most recent
21:25:58  <TrueBrain> 161 non-members
21:26:04  <TrueBrain> Which is a lot
21:26:14  <glx> oh I see, they're still open for us, but considered closed by the analyser
21:26:51  <andythenorth> shall we get a badge? o_O
21:27:18  <TrueBrain> It doesn't feel like we accept 80%, but stats don't lie
21:27:41  <TrueBrain>
21:27:50  <TrueBrain> They do the inverse :p
21:28:27  <glx> but faster ;)
21:28:46  <TrueBrain> Not sure that means better :p
21:28:54  <andythenorth> so a train would need a list of the liveries it supports?
21:29:11  <andythenorth> like refittable cargos or something?
21:29:28  <glx> yeah similar to refittable
21:29:37  <andythenorth> static prop, or CB?
21:30:37  <glx> I guess usual way is both
21:31:06  <DorpsGek> [OpenTTD/OpenTTD] michicc merged pull request #8555: Fix 218f40e: Warning about 32-bit shift implicitly converted to 64 bits
21:31:23  <glx> so it's possible to "remove" liveries based on current year
21:31:54  <andythenorth> hmm can I change cargo refits in cb36?
21:31:59  * andythenorth imagines mischief
21:32:06  <glx> remove/add from the list, but still available to draw current running vehicles
21:32:18  <glx> just not for new ones
21:32:53  <andythenorth> I think for the full range of madness, authors will want to check by dates and stuff yes
21:35:47  <frosch123> andythenorth: the mistake of cargo subtypes to make the callback define the list
21:36:04  <frosch123> it's easier if you have a static list, and a callback that tells available/not available
21:36:30  <frosch123> that way it's clearly known which items exist, even if they are not available now
21:36:32  <andythenorth> so similar to cargos
21:36:36  <andythenorth> but not
21:37:10  <andythenorth> there will need to be some new feature type 'livery' to provide a label and name string?
21:37:25  <andythenorth> or can it be jammed into action 14? :P
21:37:38  <glx> new feature is probably better
21:37:57  <frosch123> andythenorth: action14 is for static stuff.
21:38:15  <frosch123> just add a text property? like for object and station classes?
21:38:16  * andythenorth wonders if availability of a livery is better handled per livery
21:38:19  <andythenorth> not per vehicle
21:38:38  <glx> per livery makes sense
21:38:46  <andythenorth> I can see authors arguing it both ways
21:39:01  <andythenorth> e.g. SNCF Fret is only after 199x
21:39:20  <andythenorth> but blablah engine has special realism based rules due to it's exact history
21:41:18  <andythenorth> I have 'liveries' in Iron Horse already, and the definitions started per engine, and pretty quickly moved to global
21:41:27  <andythenorth> (company colour rules stupidity)
21:43:34  <andythenorth> looking at Objects: text ID, Introduction date, End of life date seem stealable
21:43:38  <andythenorth>
21:44:35  <andythenorth> the 'this engine is special' case is solved by defining a different livery, same string, different dates, only used for that engine
21:44:38  <andythenorth> or something
21:44:55  <frosch123> TrueBrain: how does it do the "non member" thing? most members are not public
21:47:15  <TrueBrain> I guess by the member label on PRs, but I didn't write it, so I don't know
21:48:50  *** arikover has joined #openttd
21:50:53  <frosch123> when i log out, glx is listed as "contributor"
21:51:10  <frosch123> so it does not spoil member secrets
21:52:45  <frosch123> ah, the filter-by-author dropdown seems to spoil it :p
21:54:17  <Samu> isn't my solution in #8009 similar somewhat to #7979 ?
21:55:01  <Samu> :( sad face because it was closed
21:57:15  <andythenorth> livery UI?  Similar to subtypes, but less crap?
21:57:59  <frosch123> those issue/pr filters spoil the whole member structure
22:05:34  *** Wolf01 is now known as Guest10183
22:05:36  *** Wolf01 has joined #openttd
22:07:54  <Samu> cyas
22:08:02  *** Samu has quit IRC
22:09:06  *** frosch123 has quit IRC
22:11:43  *** Guest10183 has quit IRC
22:14:26  <DorpsGek> [OpenTTD/OpenTTD] Milek7 commented on pull request #8391: RFC: Feature: Public key company authorization
22:26:40  *** arikover` has joined #openttd
22:31:56  *** arikover has quit IRC
22:39:46  *** Wolf01 has quit IRC
22:42:46  *** sla_ro|master has quit IRC
22:53:24  *** arikover` has quit IRC
22:55:32  *** gelignite has quit IRC
23:08:20  *** Progman has quit IRC
23:11:49  *** andythenorth has quit IRC
23:13:57  *** jottyfan has joined #openttd
23:14:03  *** jottyfan has quit IRC
23:30:29  *** DasPoseidon has quit IRC
23:30:30  <DorpsGek> [OpenTTD/OpenTTD] mattkimber updated pull request #8540: Fix eeb88e8: Trains reversed while paused do not correctly update sprite bounds
23:30:51  <Timberwolf> Spoke too soon. I broke ships.
23:31:41  <Timberwolf> Noticed it while reviewing videos, there was a one-frame glitch when a ship turned.
23:32:05  <Timberwolf> We definitely need more people playing nightlies!
23:37:16  <DorpsGek> [OpenTTD/OpenTTD] LC-Zorg commented on issue #8551: Flat maps of arctic and tropical climates are completely inaccessible
23:50:33  *** roadt__ has joined #openttd
23:50:51  *** roadt_ has quit IRC

Powered by YARRSTE version: svn-trunk