IRC logs for #openttd on OFTC at 2017-08-19
⏴ go to previous day
00:00:11 <andythenorth> hope is not a strategy :)
00:00:13 <perk11> Do what you think is right
00:00:41 <LordAro> perk11: i'd imagine it's the sort of thing where if you can demonstrate that it would benefit more than one language, it might get done
00:00:45 <LordAro> but otherwise... probably not
00:01:00 <Eddi|zuHause> andythenorth: tbh, rubidiums statement sounds more like an excuse than a reason
00:01:15 <andythenorth> I don’t find fragmenting translations a big deal
00:01:28 <andythenorth> but then again for OpenTTD, I’m privileged :P
00:01:31 <andythenorth> I have the base lang
00:01:33 <perk11> LordAro: I see, don't know other languages that it would benefit, but there should be some
00:01:39 <Eddi|zuHause> the issue certainly came up in german language as well
00:01:53 <andythenorth> but I write UI code for translated web apps, and I often deliberately split strings
00:02:03 <andythenorth> even though it costs more money when we get them translated
00:02:08 <andythenorth> we get a better result
00:02:25 <_dp_> LordAro, likely also true for belarusian and ukranian, though I'm not entirely sure
00:02:37 <andythenorth> sometimes the translation varies by context, even though word is same in English
00:03:04 <Eddi|zuHause> problem gets worse by english being very simplified grammar
00:03:25 <Eddi|zuHause> (terrible grammar in that sentence)
00:06:45 <Wolf01> <Eddi|zuHause> (terrible grammar in that sentence) <- it's so simple you can throw here words like that and still understand it :P
00:07:14 <Eddi|zuHause> Wolf01: but that is sort of true for any language
00:07:22 <Eddi|zuHause> Wolf01: lots of redundance in language
00:09:04 <_dp_> andythenorth, yeah, GS the way it is now is not suited for callbacks at all
00:09:16 <_dp_> andythenorth, it's too slow even for its own commands
00:09:44 <LordAro> andythenorth: apparently the UFO can pick a vehicle stopped (or is currently) in a depot
00:09:48 <andythenorth> Zuu is trying to suggest alternatives
00:10:00 <andythenorth> but the OP’s request is not implementable
00:10:03 <LordAro> i'd probably call 5868 a bug, actually
00:10:25 <Eddi|zuHause> Wolf01: in the early days of computing, Shannon (who basically invented the "compression" field) ran an experiment and wrote a sentence on the wall, abruptly ending it halfway through and check whether the people could finish it for him. by evaluating whether that finishing was correct, he concluded that for every letter, there's about 1 bit of information in the english language
00:10:34 <Eddi|zuHause> Wolf01: so all the rest is redundancy
00:12:53 <LordAro> 5309 - won't implement?
00:12:53 <Wolf01> He was right, since every letter could be or couldn't be right for that position
00:12:55 <Eddi|zuHause> (that is basically why text files can be compressed really well)
00:13:20 <Eddi|zuHause> Wolf01: that sounds so fake-news-y :p
00:14:33 <andythenorth> I’ve dm-ed george asking him if all his issues are still valid / wanted
00:15:57 *** Tharbakim has joined #openttd
00:17:09 <_dp_> 5697 reminded me it's a super easy to do patch that allows/denies vehicles for specific companies
00:17:20 <_dp_> bitmask is already there so all it needs is GS api
00:18:04 <Eddi|zuHause> i think i suggested giving GS control over that bitmask
00:18:36 <_dp_> Eddi|zuHause, I'm sure I suggested it too xD
00:18:44 <LordAro> andythenorth: omg, 501
00:19:30 <andythenorth> which seems to descend into argument
00:19:37 <LordAro> andythenorth: no, the number of issues :p
00:19:51 <andythenorth> 19 minutes past bedtime
00:19:55 <LordAro> andythenorth: close, i think
00:20:29 <LordAro> any solution would be very platform specific, and SDL doesn't really do it (even on v2, iirc)
00:20:32 * andythenorth thought window/canvas management was handled by OS
00:21:50 <andythenorth> those should probably be closed as ‘not under OpenTTD control’ :P
00:21:59 <andythenorth> anyone who argues differently can provide patches
00:22:01 <glx> it's also driver dependant I think
00:22:02 <LordAro> and "no sane cross-platform solution"
00:22:04 <Eddi|zuHause> re 4637: i think the solution is to have "detachable" viewports
00:22:42 <Eddi|zuHause> so you have two separate windows (of possibly different sizes)
00:22:50 <andythenorth> 2733 can be closed I think
00:23:10 <Eddi|zuHause> and possibly move ingame windows (like graphs) to the other window as well
00:23:30 <LordAro> andythenorth: probably, yes
00:24:28 <_dp_> 2733 was quite valid tho imo
00:24:33 <andythenorth> I don’t know enough facts to say that stuff is delegated to the OS
00:24:40 <andythenorth> _dp_: have you verified the bug?
00:24:55 <LordAro> andythenorth: close both and open a "detachable viewports" feature request?
00:24:55 <Eddi|zuHause> andythenorth: i'd leave 4637 open
00:25:35 <_dp_> andythenorth, no but I always knew one-way roads are useless
00:25:47 <_dp_> andythenorth, unless for hacky things like trapping vehicles
00:26:17 <andythenorth> they work exactly as expected
00:26:31 <andythenorth> but that issue won’t get them changed
00:26:40 <_dp_> andythenorth, eh.. but they look like two lane :p
00:26:47 <Wolf01> It could be solved by overpainting the unused lane :P
00:26:56 <andythenorth> it’s an overtaking lane
00:27:16 <Wolf01> Too bad vehicles don't even overtake
00:27:31 <Eddi|zuHause> andythenorth: and you didn't find a "duplicate" as reason to close it?
00:27:32 <andythenorth> or is that the bug? o_O
00:27:46 <_dp_> they do but very slow :p
00:28:00 <Wolf01> No, because you flooded the game with articulated vehicles which can't overtake :P
00:28:06 <andythenorth> they all go same speed
00:28:12 <andythenorth> so it’s not an issue
00:28:17 <andythenorth> I fixed it in newgrf
00:28:19 <Eddi|zuHause> overtaking works terribly in the game
00:28:20 <_dp_> andythenorth, breakdons? ;)
00:28:26 <andythenorth> I’ve fixed loads of these stupid issues in newgrf
00:28:43 <andythenorth> breakdowns are turned off
00:28:51 <andythenorth> fixed that issue too
00:28:55 <Eddi|zuHause> because not only with articulated vehicles, but also with crowded traffic, vehicles won't overtake
00:29:00 <Wolf01> The problem raise when you still have different generations running
00:29:09 <_dp_> basically there is no difference in throughput (in one directino) between one way and two way roads currently
00:29:14 <_dp_> which is kinda silly imo
00:29:15 <andythenorth> I introduce most of the generations at the same time Wolf01 :)
00:29:26 <andythenorth> _dp_: one-way was just added for deliberate griefing no?
00:29:27 <Eddi|zuHause> andythenorth: all in 1600?
00:29:34 <Wolf01> But they don't replace at the same time
00:29:34 <andythenorth> Eddi|zuHause: nice idea
00:29:49 <_dp_> andythenorth, yeah, works well for that xD
00:30:06 <andythenorth> Wolf01: when overtaking is fixed, we can all have cake
00:30:23 <andythenorth> LordAro: stop at 500?
00:30:41 <andythenorth> Wolf01 and _dp_ can add more for their new patches
00:31:07 <LordAro> since my browser doesn't let me look at paste.ottdc.o, can someone tell me whether there's a "raw" url i can get to easily?
00:31:09 * _dp_ secretly planning that whole evening
00:33:27 <_dp_> i recall someone suggesting bitcoin with a random exchange rate
00:33:53 <andythenorth> surely that’s just ‘losing’
00:33:58 <andythenorth> or ‘being crap at the game'
00:34:18 <Eddi|zuHause> andythenorth: having fractional exchange rates might be useful, but constantly adjusting the exchange rates to real world values is surely a non-goal
00:34:45 <_dp_> andythenorth, or cargodist
00:34:54 <Eddi|zuHause> andythenorth: i'd even go so far and request the original 1994 game exchange rates be reestablished, undoing any adjustments done in the past
00:35:05 <andythenorth> Eddi|zuHause: add a feature request :)
00:35:14 <andythenorth> can we also undo inflation, and the smooth economy?
00:35:35 <Eddi|zuHause> like, at some point DM:GBP exchange rate was changed from original 4:1 to 3:1
00:35:37 <andythenorth> _dp_: 499, you have a quota of 1 patch :P
00:35:37 <_dp_> andythenorth, leave smooth, undo original one :p
00:35:50 <andythenorth> frosch dislike smooth strongly for some reason
00:36:12 <Eddi|zuHause> smooth economy is incompatible with most newgrfs
00:36:33 <Eddi|zuHause> particularly the ones which use production callback
00:36:41 <_dp_> andythenorth, I guess from implementation point it's done quite awfully
00:36:44 <andythenorth> most newgrfs do what they want, ignoring economy
00:37:08 <_dp_> but for gameplay it's like "the only right way to play" imo :p
00:37:42 <Eddi|zuHause> smooth economy is probably fine, but it should be automatically disabled if newgrfs using production callback is used
00:37:56 <Eddi|zuHause> not just for the industries that use the callback, but for all industries
00:38:24 * andythenorth closed a “Shunting” feature request
00:38:43 <_dp_> Eddi|zuHause, it does too much different thigs, I'd rather split it in a few independent settings
00:39:35 <_dp_> well, I guess, some of them will still need to interact with grfs somehow
00:40:28 <andythenorth> economy script :P
00:40:32 <Eddi|zuHause> _dp_: honestly, touching smooth economy will probably open a bottomless pit of new feature requests
00:40:47 <andythenorth> just throw them down to the bottom then
00:41:10 <Supercheese> yeah, There's a Patch for That™
00:41:14 <Supercheese> and that's good enough for me
00:42:14 <_dp_> Eddi|zuHause, yeah, we have andy to deal with them :p
00:42:57 <_dp_> though adding some more sensible options for those settings will probably cover some requests
00:43:02 <andythenorth> thanks Supercheese
00:43:24 <andythenorth> _dp_: just make economy scriptable
00:43:33 <_dp_> like I rly want completely stagnant economy where production never changes
00:43:33 <andythenorth> be done by Tuesday
00:44:06 <andythenorth> are signals for bridges and tunnels ever or never?
00:44:16 <andythenorth> apaprently there’s a super-awesome patch for it
00:44:24 <andythenorth> maintained since 2009, seems to still be working
00:44:25 <_dp_> andythenorth, yeah, cloning all industry grfs
00:44:33 <andythenorth> _dp_: sounds crap
00:44:46 <andythenorth> but industry grfs ignore economy anyway
00:44:51 <andythenorth> so what to do what to do :)
00:45:39 <_dp_> andythenorth, yeah, that's one of reasons i still haven't done a patch
00:45:57 <_dp_> andythenorth, adding "stable god dammit" is probably a bad option...
00:46:02 <andythenorth> if there was economy script, I would make FIRS primaries listen to it
00:46:26 <andythenorth> I would want it to set a production level byte either per tile, or per town
00:46:37 <andythenorth> that then hooks into standard production behaviour
00:46:47 <andythenorth> and player can boost with supplies
00:47:20 <LordAro> there do seem to be quite a lot of little patches just sitting around that, to my eyes, look perfectly valid for merging
00:47:30 <andythenorth> alternatively, economy could set the production level directly in the industry
00:47:51 <andythenorth> LordAro: I can’t judge quality, but yes, there seemed to be a lot
00:48:32 <andythenorth> one option is fork on github, merge them in one at a time, get TB to hook it up to a build (or you Circle build)
00:48:37 <andythenorth> then get players to test them
00:49:14 <andythenorth> or just get them reviewed by someone who can review, and put them in trunk :P
00:49:33 <andythenorth> corrupt a few savegames :)
00:49:51 <_dp_> if something fits cm client I can merge it there first
00:50:01 <_dp_> there seem to be some people still using it)
00:55:55 <Eddi|zuHause> andythenorth: the general idea of signals on bridges is a yes, the hackalittlebit approach is definitely a no
00:56:37 *** Supercheese has joined #openttd
00:57:11 <Eddi|zuHause> andythenorth: the hackalittlebit approach is too narrow, and would block a broader solution
01:00:20 <LordAro> doesn't the "proper approach" require that mythical new map array though?
01:01:04 <Eddi|zuHause> which brings us to the topic of cirdan...
01:01:27 <LordAro> didn't michi start one?
01:06:00 <LordAro> urgh, all these lovely patches tied up in patchpacks
01:06:27 <LordAro> just need more people to go through and separate out features and tidy up patches
01:06:33 <LordAro> and people that can actually merge
01:06:44 <LordAro> but we've had this conversation already
01:09:58 <Eddi|zuHause> we've had that discussion multiple times before. often patch authors think having their patch included in a patchpack is a jumping ground for making it into trunk, when it's often the opposite
01:11:09 <LordAro> in a perfect world, it probably would be
01:11:16 <Eddi|zuHause> lots of patches died in patchpacks because no effort was done to maintain it separately, and actually clean it up for trunk merging
01:12:14 <LordAro> there's also the issue where many patches are split up into the nice sections the devs like, but they're just too large for the devs to properly review them
01:12:15 <Eddi|zuHause> while the patchpack version was buried under layers and layers of further contamination from ultimately unmaintainable merging
01:12:19 <LordAro> so they just sit there
01:12:47 <LordAro> Eddi|zuHause: yeah, which is why i like the look of juanjo's stuff, given they've got things in fairly nice separate commits already
01:14:17 <Eddi|zuHause> i've really not followed any patchpacks in like 5 years
01:26:34 *** supermop has joined #openttd
01:31:01 *** chomwitt has joined #openttd
01:43:53 <Etua> Hello, why OpenTTD doesn't have it's own repository for backport updates?
01:53:49 <Eddi|zuHause> Etua: openttd has a release branch, and bugfixes are backported to that branch
01:54:44 <Eddi|zuHause> (currently branches/1.7)
01:57:24 <Etua> Eddi|zuHause: I mean if there is a repo with deb files containing newer version than that in my distribution's LTS repo. Did you mean something like that or dev svn branch?
01:57:55 <Eddi|zuHause> Etua: i meant dev branch
01:59:16 <Eddi|zuHause> Etua: i don't think you can expect a project like this providing repositories for all the distributions out there
02:01:25 <Etua> Eddi|zuHause: For me deb repo would be enough ;-) I know a few projects that offer them in order to skip distribution queues.
02:01:48 <Eddi|zuHause> Etua: there are some .deb files on the download page, if that's what you're asking
02:02:21 <Eddi|zuHause> Etua: and a generic bundle that should run on any distribution
02:09:19 *** supermop_home has joined #openttd
02:09:50 <Etua> Eddi|zuHause: I know but installing standalone deb files is a fast way to make your system a mess and it does not offer automatic updates. For example I use Prosody repo for their single package in newest version.
02:09:53 <Etua> Eddi|zuHause: So I suppose that OpenTTD does not offer deb repo/ppa?
02:12:02 <Eddi|zuHause> Etua: no, but if your distribution has anywhere beyond 100 users, it probably has someone that offers a repository containing an updated version
02:18:02 <Etua> Eddi|zuHause: I'd like to install it on LM so probably anything from the Debian family would work. The problem is not even that I need most recent features but to be server-compatible with Windows user that presumably will have newest stable version.
02:18:34 *** supermop_home has joined #openttd
02:20:27 <Etua> Eddi|zuHause: Do you know how big is the gap needed to loose compatibility between OpenTTD versions? Is that any difference is a blocker or compatibility is lost only after some significant changes?
02:21:01 <Eddi|zuHause> Etua: i (sorta) understand the problem, but really, this is a problem you should talk with distribution-maintainers about, you probably won't get a satisfying answer here
02:21:39 <Eddi|zuHause> and i'm certain within "anything from the debian family" you will find hundreds of suitible repos
02:23:17 <Etua> Eddi|zuHause: Ok, I just wondered if there is some official OpenTTD repository. LM uses mainly Ubuntu LTS packages base so I probably won't do anything about the release cycle. Anyway, thank you for your help!
02:25:16 *** orudge` has joined #openttd
02:25:17 *** ChanServ sets mode: +o orudge`
02:32:52 *** supermop_home has joined #openttd
04:01:04 *** orudge` has joined #openttd
04:01:04 *** ChanServ sets mode: +o orudge`
05:03:10 *** HerzogDeXtEr has joined #openttd
05:09:13 *** supermop_home has joined #openttd
05:20:29 *** supermop_home has joined #openttd
05:31:12 *** HerzogDeXtEr1 has joined #openttd
05:41:44 *** orudge` has joined #openttd
05:41:44 *** ChanServ sets mode: +o orudge`
05:42:38 *** supermop has joined #openttd
06:11:13 *** supermop has joined #openttd
06:17:28 *** orudge` has joined #openttd
06:17:28 *** ChanServ sets mode: +o orudge`
06:51:38 *** supermop has joined #openttd
07:15:46 *** orudge` has joined #openttd
07:15:46 *** ChanServ sets mode: +o orudge`
07:25:55 *** supermop has joined #openttd
07:35:01 *** andythenorth has joined #openttd
07:38:01 *** orudge` has joined #openttd
07:38:01 *** ChanServ sets mode: +o orudge`
07:50:20 *** andythenorth has joined #openttd
07:52:43 *** sla_ro|master has joined #openttd
07:54:38 *** Alberth has joined #openttd
07:54:38 *** ChanServ sets mode: +o Alberth
08:09:07 <Alberth> looks highly complicated
08:09:31 <Alberth> more something for a planning kind of view
08:20:45 <Alberth> company value is the sum of the prices of the shares, so if it's negative, you'd get money if you buy stocks from it
08:21:40 <Alberth> the OP probably reasoned that a negative value of a company should be an option, ie if you invest a lot without a proper profit
08:23:05 <Alberth> it's not going to be changed, at least not soon, and likely never
08:24:23 *** orudge` has joined #openttd
08:24:23 *** ChanServ sets mode: +o orudge`
08:31:29 * andythenorth will close that one
08:34:05 <andythenorth> isn’t this just ‘losing at OpenTTD’?
08:34:26 <andythenorth> at first glance, I’d say that town is terribly served by transport
08:34:51 <andythenorth> although I can see the logical fallacy of 3000+ pax in an isolated town with 1000 popn. :D
08:47:39 <Alberth> stations are a bit too good at collecting cargo :p
08:48:18 <andythenorth> I was just intrigued by the fallacy :
08:49:32 <andythenorth> we need a FS category for scenario editor
08:50:57 *** supermop_home has joined #openttd
08:52:03 <SpComb> needs induced demand for pax
08:55:14 <andythenorth> edits all relevant issue titles to start “Scenario Editor: "
08:56:28 *** Progman has joined #openttd
08:56:52 <LordAro> andythenorth: not easier to wait for frosch to reappear and add a category? :p
08:57:22 <andythenorth> such lack of patience
08:58:25 <andythenorth> if I use ‘restart’ in console, I get the current map reset
08:58:57 <andythenorth> doesn’t matter if scenario or random game
08:59:25 <andythenorth> no actually, that’s user error
08:59:35 <andythenorth> yeah, it bins the scenario and starts a new game
09:00:48 <Alberth> restart generates a new world, which doesn't do what you expect when you start from a loaded game :p
09:01:11 <andythenorth> I have left it open low priority
09:01:15 <andythenorth> should it actually be closed?
09:01:32 <Alberth> it differs between game versions, since the places where random is used changes
09:01:46 <Alberth> I don't think we want to stay compatible there
09:02:17 <Alberth> so yeah, close would be fine, imho
09:02:24 <andythenorth> do you have close rights? :)
09:17:19 <Alberth> Not sure I agree with your conclusion of 6525
09:17:37 <Alberth> I think it makes sense to use the same rules in both cases
09:18:14 <andythenorth> does that let us consolidate anything?
09:18:38 <andythenorth> I am with frosch - the addition of a 2nd economy was an unfortunate mistake :)
09:19:42 * andythenorth wonders how to withdraw a closure request :P
09:20:54 <andythenorth> there a quite a few where I requested closure instead of just closing
09:21:00 <andythenorth> because I’m really not certain
09:21:42 <andythenorth> what is clear: the majority of the 255 feature requests are (1) fair and valid (2) never going to get done :)
09:21:58 <Alberth> If the OP had fixed the patch, I would have tried getting it in, but apparently even that is not discussable
09:21:59 <andythenorth> I am starting to priortise to ‘high’ and ‘low’ to try and make sense of them without closing
09:22:40 <andythenorth> even just this search for ‘timetables’ shows how half-assed new features spawn more work :)
09:22:43 <andythenorth> even just this search for ‘timetables’ shows how half-assed new features spawn more work :)
09:22:44 <andythenorth> even just this search for ‘timetables’ shows how half-assed new features spawn more work :)
09:23:02 <andythenorth> even just this search for ‘timetables’ shows how half-assed new features spawn more work :)
09:23:30 <LordAro> i think that's unfair though, there's a difference between "half assed" and "use case no one thought of"
09:23:37 <Alberth> of course, you get bug reports about stuff that doesn't work :p
09:23:57 <andythenorth> Alberth: I would still reject 6525 on basis of “OP has abandoned work, and was slightly rude about it"
09:25:28 <Alberth> "rude" is perhaps not the right word, but we fundamentally differ in opinion
09:26:12 *** supermop_home has joined #openttd
09:27:30 <andythenorth> so now it’s basically “to move this forward anywhere, dev must rewrite patch for issue they never really cared about” o_O
09:28:43 <LordAro> jesus guys, it's not a huge amount of work
09:33:48 * andythenorth found some nice duplicates
09:33:55 <andythenorth> almost word for word
09:34:13 *** gelignite has joined #openttd
09:36:07 <Alberth> LordAro: yep, I know, but I scribble down to do it, then just forget it until I run into it again
09:36:28 <andythenorth> like my FIRS ‘to do later’ list :)
09:36:36 <andythenorth> which is in no way going in tickets
09:36:53 * andythenorth opens that list to check
09:37:00 * andythenorth closes that list quickly
09:41:12 *** supermop_home has joined #openttd
09:43:52 <andythenorth> I think that isn’t
09:44:39 <andythenorth> I think OP confuses ‘cargo is assigned a destination’ (false) with ‘cdist controls loading of vehicles at stations’ (true)
09:46:34 <LordAro> ok, this code is actually just a mess
09:48:35 *** supermop has joined #openttd
09:49:42 <LordAro> smooth & original are just smashed together
09:50:03 <andythenorth> there’s no abstraction
09:50:09 <andythenorth> it’s very much JFDI spaghetti
09:50:39 * andythenorth is very familiar with industry_cmd.cpp and related
09:51:27 <andythenorth> compared to other bits of src I’ve read, it seems to be one of the messiest areas
09:51:49 <andythenorth> but somehow newgrf industries are one of the most successful areas of modding :)
09:52:53 <andythenorth> nice houses are sometimes messy :)
09:53:14 <Alberth> economies is likely close to original decoded source :)
09:53:29 <andythenorth> there are probably technically pure bits of newgrf support that are never used
09:53:58 * andythenorth would burn the economies, and replace with economy script
09:54:16 <andythenorth> which could manipulate industry and town
09:54:25 <andythenorth> and in turn, be issued events by GS
09:54:37 <andythenorth> economy script would be pretty dumb and limited
09:54:55 <andythenorth> callbacks on some frequency, with control over map regions or tiles
09:55:50 <andythenorth> GS controls goals
09:56:21 <andythenorth> so in this idea, you could have city builders or busy bee or whatever
09:56:30 <andythenorth> independent of, e.g. industry production
09:56:48 <Eddi|zuHause> haven't we discussed yesterday hat squirrel is terrible for callbacks?
09:56:57 * andythenorth didn’t specify an implementation
09:57:14 <andythenorth> economy script could be nfo, possibly
09:57:26 <andythenorth> or something else
09:57:33 <Eddi|zuHause> new nfo feature?
09:58:54 <andythenorth> my idea is to literally dump a couple of economy vars on tiles or towns
09:59:05 <andythenorth> and let ES adjust them on some frequency
09:59:13 <andythenorth> or response to limited number of events maybe
09:59:31 <andythenorth> don’t go around on a loop trying to micro-manage industries and towns
10:01:24 <Eddi|zuHause> andythenorth: the first thing i would do would be to implement some kind of slowly growing world population, which moves between towns and cities (urbanisation), when provicded with serviced industries
10:02:11 <Alberth> you need some system that can carry out long term orders, imho
10:02:39 <Alberth> and something of a planner
10:03:07 <andythenorth> it needs some kind of state
10:03:19 <Alberth> we have that, the entire game
10:03:23 <andythenorth> not just stateless response to cb, which limits you to pretty much random response to local vars
10:03:33 <andythenorth> I mean state outside the map array
10:03:41 <andythenorth> but maybe that’s over-complicated
10:03:47 <andythenorth> and maybe that’s for GS
10:04:00 <andythenorth> yeah, ES should be simpler than that
10:04:11 <Alberth> stateless should be included, if only for backward compatibility
10:04:26 <Alberth> and it's a valid form of planning :p
10:04:35 <Eddi|zuHause> the GS has state, but not a lot of context (used GRF), whereas the GRF has lots of context but not enough state
10:04:44 <Alberth> not very effective though :p
10:06:07 <Eddi|zuHause> so things like house construction or production callback need access to this economy state, but also in a way that makes combining grfs from different authors possible
10:06:12 <Alberth> GRF has context at too low leve
10:07:13 <Alberth> so randomisation stays in GRF, higher long term direction more in GS
10:07:46 <andythenorth> newgrf is a walled garden for production
10:07:57 <andythenorth> all we can do is indicate to the newgrf what the local map conditions are
10:08:05 <_dp_> 6525 is an example of what should economy be splitted into imo
10:08:08 <Alberth> perhaps something related tp position (this position should grow/stay the same/shrink) ?
10:08:52 <Alberth> wrt to some property (population, industry production, ..)
10:09:04 <andythenorth> indicating current economic health
10:09:08 <andythenorth> that would be enough
10:09:20 <andythenorth> I want to keep it really simple
10:09:35 <andythenorth> if we do it simple and blunt, it can be used not just by houses and industries
10:09:41 <andythenorth> but also towns can use it to grow or shrink
10:10:14 <andythenorth> and it _could_ be used to localised cargo payment rates (custom profit calculation exists already)
10:10:47 <Eddi|zuHause> there should be a global level of coordination (like world population), and a regional level coordination (like town has access to a source of electricity) additionally to the local "coordination" that newgrfs can already do
10:10:59 <_dp_> your ES looks pretty much the same as client-side GS I was talking about recently
10:11:39 <andythenorth> Eddi|zuHause: coordination could be ES or GS
10:11:47 <Alberth> Eddi|zuHause: I'd see that as something internal to the global planner
10:11:55 <andythenorth> I would give GS commands to ES, on a per-region basis
10:12:24 <andythenorth> hmm, that might be a disaster though :)
10:13:18 <Alberth> ES == economy script?
10:13:19 <Eddi|zuHause> i think what is missing is a communication layer between GS and GRF
10:13:56 <Eddi|zuHause> so that the GS can understand what each industry is currently doing, and can indicate to the industry what it wants to be controlling
10:14:07 <andythenorth> ES = Economy Script
10:14:21 <andythenorth> I think it’s possibly an Economy Layer as much as a script
10:14:25 <andythenorth> it would have logic, but not much
10:14:29 <Alberth> one way communication GS -> GRF would be sufficient
10:14:36 <andythenorth> it _could_ have been done with newgrf towns
10:14:53 <andythenorth> Alberth: the communication has to be semaphore/telegraph, not direct control
10:15:20 <Eddi|zuHause> there is this weird town storage
10:15:44 <andythenorth> I considered using it, but it’s a hack
10:15:51 <andythenorth> and again, the towns don’t have global overview
10:16:03 <andythenorth> consistent theme
10:16:28 <Eddi|zuHause> it could, if the GS also had access to that town storage, and could organize the data
10:17:16 <Eddi|zuHause> but then you still need a common interface, so you could plug in and out different grfs and gamescripts
10:17:19 <andythenorth> at that point, my ES idea is ~identical, except it provides a dedicated storage, not dependent on convention
10:17:37 <andythenorth> using town storage is totally dependent on everyone agreeing which register to use
10:18:03 <andythenorth> also I would like to see the smooth economy eliminated from openttd src
10:18:07 <andythenorth> and moved into a script layer
10:18:57 <andythenorth> NoIndustries would also be good :P
10:19:10 <andythenorth> re-implement them all in clean NML, ship it with openttd
10:19:10 <_dp_> why wasn't GS called NoScript?
10:19:17 <_dp_> would be a very fitting name :p
10:19:40 <Alberth> let's make NoOpenTTD :p
10:19:44 <andythenorth> move more ‘vanilla’ stuff to newgrf, ship source clean with the game, let people mod
10:19:55 <andythenorth> eliminate > 'some' feature requests
10:20:02 <Eddi|zuHause> _dp_: it would not. because NoAI was named that when the first step was ripping out the old AI. with GS there was nothing existing to rip out
10:20:02 <_dp_> Alberth, fork it and rename trunk :p
10:20:24 <andythenorth> ship a cross-platform nmlc, create a ‘compile’ button right in the game
10:30:03 <V453000> hmmm the colour offset logic I have for darkening semi-transparent edges is an absolute nightmare :D
10:30:08 <V453000> it has so many exceptions and special cases
10:30:30 <_dp_> LordAro, what file is it from?
10:30:55 <andythenorth> V453000: edges are hard :)
10:32:59 <_dp_> LordAro, don't think they are, looks like while only closes when it tries to go below PRODLEVEL_MINIMUM
10:33:09 <_dp_> LordAro, and if closes when it reaches it
10:33:46 <andythenorth> fricking airports
10:35:01 <andythenorth> specifically town limitations on them
10:35:05 <andythenorth> should have been newgrf
10:35:22 <andythenorth> n airports per town, noise limits, blah blah
10:35:36 <andythenorth> as it is I have to use OGFX+ Airports just to fix it
10:35:54 <andythenorth> that has a patch, but should be closed imho
10:36:17 <andythenorth> solved: use noise limits, and OGFX+ Airports
10:39:45 <V453000> andythenorth: iz thing, would be cute, but it's just another nice idea that doesn't really need doing
10:39:54 <V453000> probably more problems than worth
10:40:15 <_dp_> 5968 should probably go along with splitting town attitude setting
10:40:34 <andythenorth> just delegate it to mods
10:40:38 <andythenorth> noise limit is regret
10:40:59 <_dp_> as economy attitude also controls several independent things
10:42:39 <andythenorth> and it’s a trivial patch if yes
10:44:17 <Wolf01> No, as Alberth said, change your hotkeys
10:45:07 <Wolf01> As I might already use CTRL+S for other things :P
10:45:47 <Wolf01> Also one might like ctrl+shift+F12 to savem because office use that XD
10:45:59 <LordAro> andythenorth: kill it
10:46:15 <andythenorth> LordAro: it’s nice evidence for scripted town control :)
10:46:47 <Wolf01> 3951 is "close, reason: disable road building for towns"
10:47:01 <_dp_> andythenorth, and that one may already be possible actually
10:47:18 <_dp_> andythenorth, iirc GS is allowed to bridge as deity
10:47:33 <andythenorth> GS is too slow to properly control trowns
10:47:40 <Wolf01> Oh, I didn't read the title XD
10:47:58 <Wolf01> It was about building bridges, don't town already build bridges over rails?
10:48:55 <Wolf01> I didn't play a full game lately, the only one was a friend, maybe he built them
10:49:07 <_dp_> though in this case I'd say speed isn't the issue, but lack of knowlege when town is going to expand over that rail
10:49:21 <_dp_> Wolf01, it builds level crossings if allowed
10:49:48 <andythenorth> _dp_: that’s still speed
10:49:53 <Wolf01> I would like them to build bridges if level crossing is not allowed
10:49:55 <andythenorth> “GS can’t respond to callbacks” :)
10:50:07 <_dp_> andythenorth, well, if you put it like that :)
10:50:09 <andythenorth> some new script layer, or newgrf could do this
10:50:18 <andythenorth> cb -> town is trying to build road -> check tiles
10:52:11 <andythenorth> SE is *UI* pretty much independent of savegames, right?
10:52:47 <Wolf01> Could I add my patches too?
10:52:54 <_dp_> oh, actually, if town roads are disabled there could be a GS that takes that part completely. Then it will decide itself when to expand and where
10:53:34 <andythenorth> also NoTownLayouts
10:53:47 <andythenorth> core should do less and less
10:54:02 <_dp_> andythenorth, those are just imaginary things, I'm talking of what could be done already :p
10:54:04 <andythenorth> the less core does, the more that patchpacks can roam around
10:54:18 <Alberth> scenario editor needs lots of work
10:54:48 * andythenorth wonders what the separation between SE and world-gen should be
10:54:53 <andythenorth> world-gen is pretty crap
10:55:06 <andythenorth> tropic doesn’t work at all
10:55:33 <Wolf01> BTW, I usually try to build embanked or trenched rails, for noise... you know... so towns can build tunnels or bridges, depends how I build them
10:55:52 <andythenorth> why do we have the combination of (1) climate specific topography and
10:55:59 <andythenorth> (2) also player options for ‘mountainous’, ‘variety distribution’ etc
10:56:05 <andythenorth> as most of (2) doesn’t really work
10:56:13 <andythenorth> it’s false configuration options
10:56:22 *** supermop_home has joined #openttd
10:56:26 <andythenorth> because the climate topography just ignores it
10:56:39 * andythenorth may be over-stating the case here :)
10:57:32 <andythenorth> what if I starting a game and building a scenario were same pipeline, used differently
10:57:51 <Alberth> it's surely trivial to spend a decade full time reworking all the parts
10:58:38 <andythenorth> generate height terrain / use heightmap -> apply climate -> generate landscape greeble / manually apply -> generate towns / place towns -> generate industries / place industries -> play
10:58:43 <Alberth> starting a game would have much less control, I think
10:59:21 <Alberth> "editing" also implies you can change your mind without starting from scratch again :)
10:59:59 <andythenorth> drop into SE at any point, edit your map, continue playing
11:00:18 <andythenorth> like creative/survival mode switch in minecraft
11:00:56 <andythenorth> 7 year old just found the “why can’t aRVs go in drive-in stops” issue :(
11:01:01 * andythenorth having to do user support here
11:01:18 <LordAro> i think it's as simple as i can make a vaguely sane "split the two economies"
11:01:24 <andythenorth> message could be better “Articulated vehicles can’t use this road stop"
11:04:35 <andythenorth> LordAro: I could test that gist, but I can’t review it :|
11:05:06 <andythenorth> if we had discord, I could just put a mic on my 7 year old, and we could answer all his ‘why did you make it like this?’ questions :P
11:05:23 <andythenorth> apparently ‘smooth is not very smooth'
11:05:33 <andythenorth> and ‘Engineering Supplies are very annoying'
11:05:52 <LordAro> andythenorth: i think that second one is on you
11:06:06 <Wolf01> andythenorth: you can ask him and just report here the answers
11:06:27 <andythenorth> he could just join irc
11:06:32 <andythenorth> but we’d have to not swear
11:06:40 * andythenorth would have to ban V453000
11:07:39 <Wolf01> He is like the Deadpool of this channel
11:11:19 <LordAro> andythenorth: how many open tasks did you start with?
11:11:50 <andythenorth> there are 475 open now
11:12:11 <andythenorth> of those, 71 are patches that might be worth reviewing
11:12:47 <andythenorth> ~30 are crashes that might want investigated (boring)
11:13:06 <Alberth> LordAro: I had a try, but your economy patch moves too much stuff around to understand
11:13:32 <andythenorth> there are another 130 or so bugs, which are probably mostly TMWFTLB, but ought to be triaged
11:13:40 <andythenorth> so that’s 230 issues that could go
11:14:00 <andythenorth> the rest are feature requests, which look increasingly hard to reject
11:14:07 <_dp_> diff itself is kind of weird, it mixes removes and adds from different functions
11:14:11 <andythenorth> 40 of them are george asking for specific newgrf things
11:14:14 <andythenorth> the other 190…blah
11:14:15 <_dp_> I'm trying to do a split view atm
11:14:18 <LordAro> Alberth: yeah, it's probably easier to compare the files themselves
11:14:22 <LordAro> rather than the raw diff
11:15:51 <andythenorth> if I close a lot of these feature requests…it’s heading into drama land
11:15:59 <andythenorth> even though, empirically nothing changed
11:16:07 <andythenorth> before closure: feature will not be done
11:16:14 <andythenorth> after closure: feature will not be done
11:16:20 * andythenorth collapsed a waveform though
11:16:28 <Wolf01> Added a new task, for the glory of satan
11:16:47 <andythenorth> think that overlaps another one somewhere
11:18:05 <_dp_> though now that I look at it I like github split veiw more
11:19:08 *** HerzogDeXtEr has joined #openttd
11:20:24 <andythenorth> can I close 1738 as duplicate now?
11:20:33 *** Stimrol has joined #openttd
11:20:33 <Wolf01> Mmmh, I use hotkeys for fast switching
11:20:57 <LordAro> Alberth: _dp_: it'd now be fairly trivial to get rid of Industry::prod_level as well
11:21:19 <Wolf01> I don't remember if there's a hotkey to increase and decrease the area size
11:21:48 <andythenorth> multi-item wish-list tickets always need to die :P
11:21:49 <_dp_> ... ok, mb I don't like github one more xD
11:22:09 <Wolf01> andythenorth: yes, it addresses "2"
11:23:16 <Wolf01> Isn't that already implemented?
11:23:41 <Wolf01> I used the level land :P
11:24:19 <Wolf01> BTW, yes, could be easily addressed
11:27:42 <Wolf01> Does TTDP works on Win10? I would like to try and see how the SE worked there
11:28:38 <andythenorth> if only $some people would stop adding new issues ;)
11:28:58 <Eddi|zuHause> although, who really has the dos version of ttd?
11:29:05 <Wolf01> I warned you yesterday I wouldn't stop to add new issues XD
11:30:01 <Wolf01> Time to get my old TTDx game disk
11:30:04 <LordAro> andythenorth: i think they're talking about how it "bounces"
11:30:19 <andythenorth> is it actually broken, or just weird? :P
11:30:28 <andythenorth> yeah it’s probably a bug eh
11:32:35 <LordAro> andythenorth: it's enough of a bug that adf thought it was worth patching :p
11:32:49 <andythenorth> hmm 6284 - not a current goal
11:32:53 <andythenorth> goals are pretty clear
11:33:03 <andythenorth> cleaning up alleged MP exploits is pretty definitely not one
11:33:36 <_dp_> andythenorth, idk, let me thing a bit
11:33:44 <andythenorth> I am about to close it
11:33:55 <_dp_> andythenorth, I'm fine with leaving things as they are but it may be a valid request
11:33:58 <andythenorth> closing alleged exploits always pisses off as many people as it pleases
11:34:27 *** Wormnest has joined #openttd
11:34:37 <Eddi|zuHause> yeah, i think that request is invalid
11:34:53 <Wolf01> Ok, win x64 doesn't have anymore wow16, so I can't straignt install TTDx, dosbox might help
11:35:16 <andythenorth> the pattern of ‘change established behaviour because some limited number of people running goal servers don’t like it'
11:35:21 <andythenorth> is not a winning strategy
11:35:26 <LordAro> think they removed that in win8
11:35:37 <Wolf01> They removed that with x64
11:37:17 <andythenorth> FWIW, I don’t want station signs moving around
11:37:29 <_dp_> andythenorth, I'd say 6284 raises up a valid problem but no solution
11:37:37 <Wolf01> Oh, I already had some D-Fend config for TTDx
11:37:43 <_dp_> andythenorth, just calling UpdateStationSignCoord will do nearly nothing
11:37:53 <andythenorth> 7 year old is now having a one-sided conversation with Convoy AI :P
11:38:03 <andythenorth> he is critiquing AI choices :P
11:39:27 <LordAro> andythenorth: i did always find it a bit odd that the sign only moved when deleting bits of the station, rather than adding to it
11:39:34 *** Maraxus has joined #openttd
11:40:03 <Eddi|zuHause> it used to be that the sign never moved
11:40:09 <_dp_> one possible solution mb to find closest parts of two stations and use distance between them in payment calculations
11:40:10 <andythenorth> +1 LordAro….but of the things we could spend time on :)
11:40:18 <Eddi|zuHause> then it was changed so the sign would be moved if it ended up outside of the station
11:40:24 * andythenorth would choose ones that are fun
11:40:44 <andythenorth> my goal here is ‘make developing great again'
11:45:16 <andythenorth> nobody needs to do that
11:47:01 * andythenorth considered it briefly
11:47:07 <Wolf01> Main viewport becomes minimap when unzoomed
11:47:11 <Eddi|zuHause> is there no button to focus the minimap on the current viewport?
11:47:26 <_dp_> isn't there an easy way to do it already?
11:47:28 <andythenorth> that would actually be super useful
11:47:45 <andythenorth> Eddi|zuHause: add an issue for that :D
11:48:32 <_dp_> minimap is good for jumping around, zooming viewports isn't quite the same
11:48:52 <Wolf01> Tooltip says "centre the smallmap in the current location"
11:49:05 <andythenorth> also 5658, just click the eye icon
11:49:41 <andythenorth> hmm, no it does have to be paused for that to work
11:50:11 <V453000> hm shit my super smart system of automatically detecting index brightness doesn't work for shit :D
11:50:14 <V453000> need to do it all manually
11:51:05 <Wolf01> Oooooh the TTDx installer
11:51:18 <andythenorth> V453000: why not working?
11:52:12 * andythenorth still wonders why it matters where a vehicle is on minimap
11:52:45 <_dp_> 5658 is funny. minimap centers if same location is requested
11:53:05 <_dp_> so if you doubleclick fast enough on moving vehicle it centers xD
11:53:06 <andythenorth> I would like to just be able to press the hotkey again to center minimap (mine is ‘m’)
11:53:24 <andythenorth> that would actually be super not-shit
11:53:55 <_dp_> andythenorth, actually just hotkey to open it is enough
11:54:05 <_dp_> andythenorth, it centers on viewport when opened
11:54:34 *** supermop has joined #openttd
11:54:43 <andythenorth> yeah, but I have to use ‘del’ to close all windows, then press ‘m’ again
11:54:49 <andythenorth> that’s actually my serious workaround
11:55:01 <V453000> andythenorth: because many indexes have rather bright colours at the 2nd darkest index already etc
11:55:03 <_dp_> still good enough for that bug as you already have 2 options
11:55:36 <V453000> plus all the wtf colours like 3-index range of yellow, super bright green or cyan being reversed in order of darkness :D
11:55:42 <_dp_> andythenorth, I don't have map key binded, does it not close minimap if it's already open?
11:56:02 <_dp_> andythenorth, probably should if it doesn't
11:56:10 <andythenorth> nah it doesn’t toggle it
11:59:49 <andythenorth> has a patch attached :P
12:04:40 <andythenorth> vague request, links me to a 2 page thread of people arguing about how to build multiple unit trains in newgrf
12:04:50 <andythenorth> arguing / debating /s
12:05:00 <andythenorth> wtf is it supposed to be
12:05:32 * andythenorth pasted a vanilla reply on and closed it
12:05:36 <andythenorth> no gain by being grumpy :)
12:09:08 <andythenorth> globally though, not per town :P
12:11:08 *** supermop_home has joined #openttd
12:14:45 <Alberth> inflation with equally increasing costs and profits is useless, just turn it off then
12:15:45 <andythenorth> why not have date-based variation of base costs?
12:15:52 <andythenorth> so recessions, booms etc
12:16:02 <andythenorth> inflation seems a waste of time to me
12:16:14 <andythenorth> I’d just rm it, and migrate savegames
12:17:23 * _dp_ never understood inflation
12:18:04 <_dp_> though I don't play more than 5-15 years usually
12:18:44 <LordAro> could take some sort of rolling average
12:19:26 <andythenorth> also are we going to send pings from the client to hundreds of servers?
12:19:34 <andythenorth> seems like a ping-flood :P
12:19:43 <_dp_> I think I've never seen a server that is unblayble due to ping
12:20:12 <Wolf01> Just let me start mine XD
12:20:47 <_dp_> andythenorth, iirc it already pings all servers
12:21:43 *** supermop_home_ has joined #openttd
12:22:46 <_dp_> well, techically it's not a ping but udp request but whatever
12:22:48 <Wolf01> _dp_: ...unblayble... <- Oh, I noticed you have bdpq problems too, I wrote p in place of b for 3 times today (luckily I noticed them before sending) XD
12:23:25 <_dp_> Wolf01, i noticed I have a lot of problems with typos xD
12:23:39 *** supermop_home has joined #openttd
12:24:07 <Wolf01> My fingers usually forget the "h", some days ago I wrote "i" in place of "y" everywhere
12:25:26 <_dp_> My weirdest typo is that I sometimes write "your" instead of "our"
12:26:07 <Eddi|zuHause> recently i find myself typing a lot of "sounds" in the wrong way
12:26:24 <Eddi|zuHause> i can't come up with an example right now
12:26:47 <Eddi|zuHause> but if you said it outloud, it would sound the same, but written down it's complete jibberish
12:26:58 <LordAro> andythenorth: many other game lobbys have the feature, it can't be that difficult to achieve
12:28:45 <Eddi|zuHause> andythenorth: wrt ping time, only use the timing of the initial response packet, don't constantly update it?
12:29:00 <Eddi|zuHause> would put no additional stress on the servers, because that packet is sent anyways
12:29:44 *** supermop_home_ has quit IRC
12:31:34 <_dp_> nice and relatively easy feature imo but hardly useful
12:31:42 <_dp_> but I may be too spoiled by a good internet :)
12:51:30 <Alberth> andythenorth: is it correct that bulk terminal does not show "(supplied)" but does show "Gung ho" instead?
12:51:39 <andythenorth> bit weird though eh?
12:51:54 <andythenorth> port-type industries are quite non-standard
12:52:26 <andythenorth> mechanic seems to work, but words…dunno
12:53:04 <Alberth> words sometimes work, sometimes they don't
12:54:08 <Pikka> someone's been closing all my flysprays
12:54:33 <andythenorth> got any more? o_O
12:54:46 <Pikka> I don't know, haven't checked
12:55:37 <Pikka> there's some good ones there
12:55:54 *** mescalito has joined #openttd
13:01:47 <ic111> Does anyone know, wether the days_in_transit (in cargo packets) are actually in unit "2.5 days" like the comment of CargoPacket::DaysInTransit() states?
13:02:34 <ic111> I ask, because I started inspecting why I earn a surprisingly low amount of money with transporting passengers sometimes
13:02:57 <andythenorth> Pikka: so can you get elected as a federal politician? o_O
13:03:36 <ic111> So I added a logger to PayFinalDelivery, PayTransfer, and AgeCargo
13:03:50 <_dp_> ic111, sounds likely but not sure
13:04:49 <ic111> ... and in a small test game starting at 1st January 1905, my train loaded around 3rd January, and unloaded 19th January, and the days_in_transit variable contained the value 20!
13:05:30 <_dp_> ic111, may also be some accumulated value for all cargo packets
13:05:42 <Wolf01> Days in transit iirc is from when they appear at station
13:05:57 <ic111> Yes, I am aware of that
13:06:01 <Wolf01> It's the life of the cargo packet
13:06:35 <ic111> As I have said, the game started at 1st January, so at 19th January, following the comment, no cargo packet can have a value > 19 / 2.5 = 7 or 8
13:06:54 <_dp_> it's only icreased when cargo is in train, I definitely checked that recently
13:07:30 <_dp_> I guess it's easier to just ask grep :p
13:08:09 <ic111> Yes, but my logging output looks like it´s increased much more often than the comment indicates.
13:09:29 <_dp_> ic111, are you sure you're not confusing days_in_transit and cargo_days_in_transit?
13:12:20 <ic111> cargopacket.h has an implementation for DaysInTransit(), implemented as "return this->days_in_transit;". I output cp->DaysInTransit(). So, not impossible that I confuse something, but currently I don´t see a mistake in my logging.
13:13:30 <_dp_> looks like vehicle newgrf can speed up cargo aging as well
13:13:55 <ic111> Ah, ok. Then I indeed should test this without NewGRF
13:14:02 <ic111> ... and maybe that comment is misleading
13:16:05 <andythenorth> vehicle newgrf can control decay rate
13:18:52 <_dp_> cat it change it dynamically?)
13:20:08 *** orudge` has joined #openttd
13:20:08 *** ChanServ sets mode: +o orudge`
13:20:09 <_dp_> looks like it can't, so no reducing decay in snow :(
13:21:55 <ic111> Seems that those early SBB set vehicles increase cargo aging by (in my example) a factor three
13:22:35 <ic111> ... now at least I know why earning money with a larger passenger network is so difficult with those vehicles, something which I never understood so far
13:23:17 <andythenorth> there is no indication to player about decay rates
13:23:23 <andythenorth> nor loading speed
13:23:41 * andythenorth thinks vehicle window should show those somehow
13:23:48 <andythenorth> or players should all use text cb :P
13:23:59 <Wolf01> Maybe cargo payment rate graph window too
13:25:34 <_dp_> Wolf01, rate graph doesn't know about vehicles
13:25:44 <Wolf01> Put rate graph in vehicles
13:26:55 <V453000> yeah loading speed is good to know
13:27:07 <V453000> all of nuts vehicles add that in text cb manually ._.
13:27:32 <_dp_> for a single wagon it's better to show it as a value
13:27:39 <_dp_> graphs make more sense for a whole train
13:27:49 <_dp_> not so easy to calculate though
13:28:48 <ic111> ... although I don´t understand the idea behind that - what´s the idea behind faster cargo aging (specifically?) in a situation where vehicles are slow, and need more time for going from A to B, regardless how efficient you build your network?
13:29:55 <V453000> it's dumb really ic111
13:29:58 <andythenorth> what chart library? o_O
13:30:05 <V453000> nuts tries to simulate "losing cargo" by that, though it's pointless
13:30:09 <_dp_> people get sick faster in old vehicles :)
13:30:25 * andythenorth makes some vehicles age cargo faster
13:30:59 <Wolf01> andythenorth: highcharts
13:31:59 <andythenorth> hmm costs money eh
13:32:50 *** FLHerne has joined #openttd
13:32:55 <_dp_> yeah, for non-commercial it's the best library I know
13:32:59 <ic111> Yes, the idea that people get sick (or rather bored, or whatever you want to call it) faster in old wagons is valid.
13:33:09 <andythenorth> err…per developer licensing :|
13:33:19 <andythenorth> back to open source charting for andythenorth
13:33:20 <ic111> And I admit, that I use the date cheat, to build a network with those old vehicles
13:34:00 <ic111> But, nevertheless, it blocks longer distance trains with those vehicles
13:34:38 <ic111> Maybe time for the next private sourcecode hack ;-)
13:35:01 <_dp_> openttd doesn't seem to be a good economy simulator anyway
13:35:19 <_dp_> just make good moneyline and do whatever you want afterwards)
13:36:16 <ic111> It was just a new expericence, that with a large passenger networks, that transports quite a lot of passengers, earning money became so difficult.
13:37:06 <V453000> well being able to transport furhter more efficiently isn't that much of a bad thing, gives some more progression
13:37:14 <V453000> but yeah, money in openttd :)
13:37:24 *** supermop_home has joined #openttd
13:39:55 <_dp_> btw, speaking of increased passenger "aging", why do they even "age" in a first place? :)
13:40:22 <V453000> I think the idea was to motivate an efficient network which gets there quick
13:40:34 <V453000> and again progression with faster engines
13:40:39 <ic111> Income per distance unit depends on the time spent for that distance
13:40:56 <_dp_> there should probably be some average travel speed expectancy for passengers that increase with time
13:41:08 <Flygon> I do kinda wish Passengers had a different style of economy.
13:41:14 <Flygon> That encouraged using smaller capacity carriages.
13:41:36 <Flygon> ie. high quality carriages for long distance journeys, with small capacities, will generate a lot of profit.
13:41:43 <Flygon> But will be terrible for profit short distance.
13:41:43 <ic111> And implementors of CargoDist chose to not calculating this by storing the timestamp when the cargo entered the vehicle, but by aging cargo periodically.
13:45:20 <ic111> Yes, it would be quite good if the expectation "what´s an average travel speed" would depend on the available vehicles
13:51:05 <Wolf01> Flygon: that's what tourists are for, they should pay more to travel more, like a logarithmic function :P
13:51:58 <Flygon> :P That's it's own Cargo Class.
13:52:00 <andythenorth> Flygon: I’m going to try two types of carriage in Iron Horse
13:52:05 <Flygon> American Tourists are a sub-class.
13:52:06 <andythenorth> it works for bus/coach in Road Hog
13:55:54 <andythenorth> Tropic map gen is junk
13:56:04 <andythenorth> but that issue is not getting it fixed
14:03:11 *** supermop_home has joined #openttd
14:03:15 <Wolf01> Tropic gen is broken, yes
14:03:55 <LordAro> andythenorth: defo still valid
14:04:20 <andythenorth> Wolf01: would you review my tropic patch? o_O
14:05:05 <_dp_> does anyone know if it's even worth to constatly call fontconfig's FcFini to save up some memory?
14:05:07 <_dp_> because time-wise it definitely doesn't
14:05:08 <_dp_> actually I couldn't even confirm there is any memory gain by doing it
14:05:09 <_dp_> and every FcInit takes about 2 seconds on my pc
14:05:09 <_dp_> I could probably patch it not to reinitialize fontconfig 12 times on every load but it somewhat feels that just keeping it initialized constantly would be even better.
14:07:02 <LordAro> _dp_: doesn't FcInit do that anyway? not like it returns anything
14:07:25 <_dp_> damn, I think i jinxed my internet :(
14:10:48 <Wolf01> andythenorth: wtf you coded it like that? :D
14:11:12 <andythenorth> you think I should apply DRY? o_O
14:11:17 <andythenorth> rather than copy-paste? o_O
14:11:46 <Wolf01> No, just remove the code from the case and move the tropic case up with temperate and toyland
14:12:13 <andythenorth> I had a problem with that idea
14:12:22 <andythenorth> …why bother checking cases at all? :P
14:12:28 <andythenorth> just use one terrain generator
14:13:03 <Wolf01> Mmmh, it works better for the terrain, but still broken as fuck
14:13:58 * andythenorth should try the temperate generator for arctic too
14:14:25 <V453000> tropic has hard time making hills
14:14:53 <andythenorth> it’s totally fucked :)
14:15:17 <andythenorth> not fixing it is one of the few things that makes me sad about current OpenTTD dev :)
14:15:31 <andythenorth> everything else is pretty awesome
14:18:15 <andythenorth> how do I generate same seed?
14:19:57 <LordAro> i'm gonna be miserable now. until we can actually get a dev to look at this stuff, say yes, and merge it; or give someone else commit rights, this is all pointless
14:20:09 <andythenorth> more mountains Wolf01 ?
14:20:23 <andythenorth> LordAro: we can fork!
14:20:27 *** supermop_home has joined #openttd
14:20:33 <andythenorth> but we could have a github fork
14:20:38 <andythenorth> with feature branches
14:20:38 <_dp_> LordAro, what do you mean? FcInit doesn't reitialize when called in succession. but openttd also calls FcFini all the time
14:20:41 <andythenorth> grouped around topics
14:21:05 <LordAro> _dp_: are we talking about the same thing? afaict, FcInit only needs to be called once per program run
14:21:58 *** JacobD88 has joined #openttd
14:22:02 <_dp_> LordAro, yeah, but it doesn't matter how many times it is called unless there is a FcFini in between them
14:22:09 <andythenorth> I’m totally not convinced that the sub-arctic map gen is doing anything worth having
14:22:25 <andythenorth> the higher mountains are better achieved by selecting 'alpinist'
14:24:01 <Pikka> andythenorth, no I can't. Although I've had several people in the last couple of weeks tell me I should be running at state level
14:24:05 <LordAro> i've been misreading FcFini as FcInit this whole time
14:24:30 <Wolf01> andythenorth: updated imgur with a bigger map
14:25:31 <andythenorth> Wolf01: so what changes then? :)
14:25:39 <andythenorth> I can read the gist, but what is it doing now?
14:25:45 <LordAro> _dp_: yeah, call FcInit/FcFini in some global static data structure and be done with it
14:26:04 <Wolf01> andythenorth: First 4 levels are more flat
14:26:36 <andythenorth> does it still flatten mountains?
14:28:33 <_dp_> LordAro, or just never call FcFini
14:28:57 <LordAro> _dp_: it wouldn't be good form to not cleanup after ourselves
14:29:36 <_dp_> LordAro, but main question here is whether is it necessary to do FcFini before entering the game
14:29:44 <andythenorth> let people write their own terrain manglers
14:29:53 <andythenorth> put em on bananaramas
14:30:09 <_dp_> LordAro, theoretically it can save up some memory by the expence of reintializing fontconfig when it's going to be needed again
14:30:14 <andythenorth> probably not even the hardest mod, terrain mangling
14:30:17 <LordAro> andythenorth: like i said, NoOTTD
14:30:18 <andythenorth> it’s only maths, no vars
14:30:26 <andythenorth> LordAro: it’s a catchy phrase
14:30:49 <LordAro> whatever happened to YA<thing>
14:30:58 <andythenorth> YA<thing> is now a proven bad joke
14:31:09 <LordAro> bad jokes can still be funny
14:31:10 <Wolf01> I'll edit a seed box in the new landscape on SE
14:31:15 <andythenorth> should have been YA<thing>deleting<oldthing>
14:32:14 <andythenorth> Terrain is just a tile loop?
14:32:26 <andythenorth> do it in some functional programming language :D
14:32:30 <andythenorth> for extra cool points :P
14:33:26 <Wolf01> Create a scenario via GS
14:34:52 <_dp_> LordAro, it's kind of hard to move fonconfig initialization to global because it's stateless and you never know if it going to be needed or whether it was initialized
14:35:10 <andythenorth> ^ sounds like he’d need a very specific network to get that to work ever
14:35:34 <_dp_> LordAro, solvable with extra bool though
14:35:54 <andythenorth> is there anyone here who uses timetables properly?
14:36:16 <Wolf01> Timetables are way too obscure to be used properly
14:36:32 <FLHerne> andythenorth: Eh, it makes sense, that's how real railways work :P
14:36:56 <FLHerne> Why would you manually skip a station (without changing the TT) if not to avoid delays?
14:37:14 <FLHerne> I'm not convinced it's possible to use timetables "properly"
14:37:37 <Wolf01> I'm not convinced a vehicle in R-world skips stations
14:38:02 <andythenorth> I can’t comment on timetable requests, because I have zero use for a broken feature
14:38:05 <andythenorth> but someone ought to
14:38:15 <FLHerne> They're far too complicated for general top-down management, but the smaller details like start dates and that bug are almost designed to make micromanagement impractical
14:38:33 <FLHerne> Wolf01: The trains here certainly do
14:39:46 <Wolf01> FLHerne: Trains here don't even arrive when they are late, and when they arrive they stop at the station because passengers already found another way to get to destination (like the next train)
14:39:56 <LordAro> ah, good ol' FS search urls
14:40:11 <ic111> I played with the native timetables in exactly one game - then I realized that I needed an external spreadsheet to keep track about them, and rewrote from scratch...
14:40:14 <FLHerne> Wolf01: If they leave Kings Cross late, they often skip some/all the stops before Stevenage (I think to make sure they cross the bottleneck at Welwyn in their booked slot)
14:40:57 <Wolf01> Maybe your lines are better organized
14:41:53 <Wolf01> Here you have 1 train every 40 minutes (average), so if one gets >40' late, then you just take the next one and the late one stops at the first big station
14:42:11 <FLHerne> Well, that's what the buses do
14:42:33 <FLHerne> The ECML is a bit busier than that :P
14:42:40 <andythenorth> FLHerne: what order should it skip? o_O
14:42:52 <andythenorth> what if it’s a servicing order, and it breaks down?
14:43:06 <andythenorth> what if cdist needs the order to calculate link stats?
14:43:27 <andythenorth> what if it’s a conditional order and the vehicle gets stuck in a loop?
14:43:39 <andythenorth> ok ‘what if’ is silly :)
14:43:40 <FLHerne> andythenorth: That's not what the bug asks for
14:44:29 <andythenorth> so that one’s valid?
14:44:32 <FLHerne> andythenorth: The problem is that if you manually skip an order, it also skips the timetabled time for that order
14:44:39 <andythenorth> sounds a bit crap
14:44:49 <FLHerne> Which is completely insane
14:45:15 <FLHerne> I mean, it makes no sense and helps make timetables impossible to manage
14:45:26 <FLHerne> But that's every aspect of the timetable system
14:45:54 <ic111> IMHO, the design flaw in the present system is that you configure a relative time ("travel 6 days") at all
14:46:06 <ic111> IMHO, timetables need to be specified in an absolute manner
14:46:16 <ic111> E.g., depart "3rd August 1925"
14:47:10 <ic111> Then the next order can specify "Arrive 17th August, Depart 21st August", and so on...
14:47:34 * andythenorth reading the TIP thread
14:47:35 <_dp_> LordAro, I genuinely hate adding global variables :p
14:47:37 <ic111> And once the last order is done, you increase all dates by a given timetable length, and start at the beginning
14:48:29 <andythenorth> I’ve only scan read, but it makes more sense than ‘travel n days'
14:48:39 <andythenorth> I never understood that
14:48:45 <andythenorth> vehicle has a speed
14:48:46 <LordAro> _dp_: i mean, you're not wrong
14:49:06 <andythenorth> with timetables, if a vehicle travels 5 days, but needs 10 to do the route, wtf happens?
14:49:08 <andythenorth> does it just stop>?
14:49:23 <FLHerne> ic111: How does your patch do old-game compatibility? Does it?
14:49:33 <andythenorth> specifying how many days a vehicle takes violates s=d/t physics
14:49:39 <FLHerne> andythenorth: It just gets increasingly late
14:49:57 <andythenorth> I have had timetables explained to me about 5 times
14:49:59 <ic111> I´ve need to look into the code myself ;-) (implemented some years ago already...)
14:50:13 <FLHerne> andythenorth: You can specify buffer time at key points to counteract that (which is why drive-in stops are useful)
14:50:17 <andythenorth> they conceptually make absolutely no sense
14:50:22 <andythenorth> vehicles have a speed
14:50:29 <andythenorth> does the timetable slow the vehicle down>?
14:50:48 <andythenorth> ok, so the concept is just a lie :)
14:51:01 <FLHerne> andythenorth: If the vehicle arrives earlier than it was supposed to, it waits at that station until it's time to leave
14:51:18 * andythenorth isn’t actually being difficult here, I really don’t understand timetables at all
14:51:23 <FLHerne> andythenorth: If it arrives late, it leaves as soon as possible
14:51:31 <andythenorth> but after loading?
14:51:34 <andythenorth> or without loading?
14:51:51 <FLHerne> With whatever the orders say
14:51:56 <andythenorth> what’s the objective?
14:52:04 <andythenorth> is it modelling realism?
14:52:16 <andythenorth> there are people out there with the National Rail website?
14:52:18 <FLHerne> Essentially, the only behaviour of timetables is to force vehicles to wait at stations for longer than they normally would
14:52:22 <ic111> The objective is that you can make your vehicle pass certain stations etc. at defined dates.
14:52:36 <ic111> E.g., I can make a long distance train overtake a short distance train at a defined station
14:52:56 <FLHerne> More straightforwardly, it's the only in-game way to prevent vehicles bunching up
14:53:09 <andythenorth> if they want to track trains
14:53:22 <FLHerne> (because you can enforce a fixed round-trip time, and then set start dates at fractions of that time)
14:53:33 <ic111> Or I can make two trains meet at a station at a defined date
14:53:33 <ic111> where they can exchange passengers
14:53:37 <ic111> Plus I can make vehicles travel in a defined interval, i.e. one train every two or four months, instead of a random order
14:53:44 <andythenorth> oh so you could have a train meet a ferry?
14:54:11 <andythenorth> so you could synchronise feeders?
14:54:38 <andythenorth> and doing that with conditional orders would be crap
14:54:44 <FLHerne> But at the moment, they're such a complete pain to set up that even I don't bother usually :-/
14:54:49 <andythenorth> also defining some kind of ‘routes’ system would be crap
14:54:51 <LordAro> ic111: hrm, i cannot get 270 to apply
14:55:33 <andythenorth> it’s the ‘travel’ thing that puts me right off
14:55:55 <FLHerne> Yeah, ic111's thing is vastly better
14:56:11 <ic111> did you use the version 20 attached to the first post?
14:56:38 *** supermop_home has joined #openttd
14:56:39 <ic111> Hm, I last checked in April that it applies against trunk
14:57:01 <ic111> So maybe a change in trunk since then, I´d have to look at it
14:57:22 *** frosch123 has joined #openttd
14:57:32 <LordAro> it was the last commit that broke it
14:58:50 <ic111> andythenorth: I removed the routes in a later version
14:59:26 <andythenorth> ic111 I’d offer to test it, it looks better, but honestly I am the wrong person to give opinions on timetables :)
15:00:19 <andythenorth> this is news to me
15:00:22 <ic111> Thank you, maybe a person that never dealt with timetables before is not that bad at all...
15:01:01 <andythenorth> FIRS doesn’t define any tile output
15:01:05 <andythenorth> how come it works?
15:01:23 <ic111> Note: I will be totally offline Sunday to Tuesday or Wednesday, so if you ask some question etc. and I don´t answer, this is the reason
15:04:41 <frosch123> andythenorth: there is no tile output, that's what the task asks about
15:05:00 <frosch123> andythenorth: you disabled tile input because it sucks, the task asks to make output as bad as input
15:07:18 <ic111> andythenorth: Just one hint: I implemented specifying timetable lengths etc. in Days, Months, or Years, but experience shows, that you usually want to specify it in Months. Days are useful if you really need a train every 15 days (but have the problem that months have variable length, which the Month variant shadows).
15:07:46 <andythenorth> does this mean I can deliver FIRS supplies guaranteed once per month? o_O
15:08:44 <ic111> If a timetable has e.g. length 4 months, and you have an order "Depart 15 January", then the next departure will be "15th May", regardless of the month length in between.
15:11:46 <_dp_> lagging heavily here, sorry
15:20:20 *** Hiddenfunstuff has joined #openttd
15:23:30 <andythenorth> Wolf01: I am re-prioritising feature requests: high -> current goals; low -> maybe a goal in future
15:23:40 <andythenorth> is SE stuff high or low?
15:24:03 <andythenorth> LordAro is a troll :)
15:25:31 <_dp_> andythenorth, what conclusion do you want? it useful feature imo, but not an easy one and nobody seems to be interested in implementing it.
15:25:41 <_dp_> andythenorth, also I'm not very interested in implementing it since I'm already pretty close to having similar stuff on citymania (which is a pile of hacks ofc so not reusable sadly).
15:25:57 <LordAro> i don't think you can classify things based on how interested people are in implementing it
15:31:01 <Wolf01> We don't have a roadmap since 1.3
15:31:19 <ic111> Beside the aforementioned reject in patch 270 (which has a quite trivial fix) the timetables patch applies for me.
15:31:29 <andythenorth> roadmaps are bollocks imho :)
15:31:37 <LordAro> ic111: i've had 3/4 so far
15:31:42 <andythenorth> except as far as setting ‘ current goals'
15:31:43 <LordAro> none are difficult to fix
15:31:57 <LordAro> it might be git being more picky
15:32:37 <ic111> Conversion of old timetable information indeed isn´t done, patch 280 does something with the start dates and lengths, but zeros the arrivals and departures out.
15:32:43 <andythenorth> I can’t adjust FS, probably wise
15:32:51 <ic111> What do you mean with "3/4"?
15:33:02 <andythenorth> but I would categorise things ‘interesting’, ‘might be interesting later’, ‘dull’
15:34:08 <ic111> I downloaded trunk just a moment ago, applied the stip_v20.zip, and all I got was this conflict in 270, where you did something in a code block my patch removes
15:34:22 <Wolf01> andythenorth: what are the current goals?
15:34:52 <LordAro> ic111: i'm using git am to (very slowly) apply the patch queue, all of other issues have been with the diff's "context", rather than the code itself
15:35:13 <andythenorth> the goals were set by the only person actively committing
15:35:16 <andythenorth> which seems fair tbh
15:35:22 <Wolf01> andythenorth: that are the game lifetime goals, not current ones
15:35:34 <andythenorth> I am trying to figure out current ones :)
15:35:39 <ic111> ok, I checked out using hg, and seems that it was more friendly against differences...
15:36:12 <andythenorth> current goals are likely to be both (a) things that devs with commit access support (b) things that interest people writing the patch
15:36:51 <andythenorth> SE is interesting, because it’s probably easy to get big UI wins with no worries about savegame, desync, NoGo, NoAI, newgrf etc
15:37:39 <_dp_> I'd it would be more benefitial to focus on small improvements rather than big rewrites for now.
15:37:56 <Wolf01> SE always got a background role
15:37:57 <_dp_> Unless there is someone who is ready to do such rewrites ;)
15:39:38 <andythenorth> also, if .scn format stays same
15:39:57 <andythenorth> we could get an official build of an SE fork, and players could try it as a binary
15:40:07 <andythenorth> whilst being able to use the results in any compatible ottd
15:41:20 *** supermop_home has joined #openttd
15:42:18 <Wolf01> I think a current goal could be to focus on daylength, people really enjoy it much, I would too, and I think we can get some compromises
15:43:00 <andythenorth> I can’t see that getting into trunk tbh :)
15:43:19 <andythenorth> it’s probably confined to patchpacks
15:43:34 <Wolf01> Then everything is confined to patchpacks
15:44:35 <LordAro> in a perfect world, uncoupling how fast the date changes from the game's tick rate should be easy
15:44:37 <Wolf01> DL is not a big change, only scatters some multiplications around, the problem is to multiply the right things
15:45:00 <andythenorth> if you want to change that, you might likely need approval from MB and george
15:45:31 <andythenorth> I proposed before just lying about the date
15:45:38 <Wolf01> We could provide a different value for tick counter, accordingly to daylength
15:45:59 <andythenorth> but the train grfs are relying on ticks to do date-related things
15:46:13 <andythenorth> DL pisses me off tbh :)
15:46:27 <andythenorth> nobody makes any sense about it so far
15:46:37 <andythenorth> if the only purpose is to display a different date
15:46:42 <andythenorth> why don’t we just lie about the date?
15:46:45 <Wolf01> Instead they should do date stuff related to date, not try to calculate the date
15:46:59 <andythenorth> instead of trying to change the fundamentals of many game loops and tick counters
15:47:28 <Wolf01> What if we need to change to another value of tick counter some day?
15:47:46 <LordAro> andythenorth: doing that effectively would be a worse hack, imo
15:48:12 * andythenorth is out of ideas then
15:52:36 <ic111> DayLength: What I don´t understand is: You have the I would like to call it real-time-part of the game mechanics, i.e. how fast do trains move, how fast are passengers generated, and so on.
15:53:28 <ic111> This part - ok, in some respect I regard it as not 100 percent balanced - but beside this, this part is senseful as it is IMHO
15:53:39 <ic111> At least, it doesn´t need to be changed by a huge factor
15:53:42 <_dp_> DL seems so easy on outside, it's tempting to do a patch :)
15:55:00 <ic111> So, concerning day length I would expect that one talks about (1) the long term game mechanics, e.g., how many years does the game spend, until the next engine becomes available
15:55:52 <ic111> and (2) maybe about pure GUI, e.g. display a time like "13:30" that approximately matches vehicle movement speed, instead of a date
15:56:12 <ic111> Now you tell me that many things are based on ticks, especially with NewGRFs.
15:57:11 <ic111> But for those things, that I called "real time part" above, if you just don´t change the speed of those ticks (relative to real world time) you shouldn´t need to change anything
15:57:12 <_dp_> ic111, for time to match speed there has to be some notion of distance in relation to real world :p
15:57:37 <_dp_> ic111, well, openttd has none afaik
15:58:38 <ic111> But well, a train traveling between two cities between 13:30 and 13:50 looks much more aligned with time, than a city traveling for 20 days.
16:00:34 * _dp_ lives in a country where travelling for days between cities is pretty common
16:00:38 <ic111> So I would expect that you would need to do something about things that happen on a long time scale (like new vehicles showing up), and on GUI.
16:01:09 <ic111> Now, given that, I don´t understand where those many places where factors would have to be adjusted are.
16:02:08 <andythenorth> search ‘tick’ in /src? o_O
16:04:21 <LordAro> ic111: ooh, apparently i just broke it as well
16:04:28 <LordAro> game's completely locked up
16:04:38 <ic111> Yes, but if you just don´t touch the tick vs. real world time ratio, and just (more or less) add extra game years, e.g. not switch from 1910 to 1911, but from 1910 to 1910.A to 1910.B to 1910.C ... to 1911, how many of those occurrences persist?
16:05:22 <LordAro> impressively locked up, i had to kill -9 it
16:05:22 <andythenorth> ic111: are you proposing just lying about the date? o_O
16:06:38 <_dp_> to me lying about date sounds like changing amount of ticks per day
16:07:53 <_dp_> and if I were to write DL patch I'd go that route
16:08:14 <Wolf01> _dp_: yes, that is what I wanted to implement
16:08:31 *** supermop has joined #openttd
16:08:33 <ic111> Just lying about the date is probably not enough, but enlarging the time interval until new vehicles arrive, and similar things I would suspect
16:08:33 <ic111> LordAro: Now I am surprised.
16:08:37 <ic111> andythenorth: Or, if people add more changes to their dayLength patches, what is their goal in doing so?
16:08:57 <LordAro> ic111: i am trying to work that out
16:09:50 <_dp_> Wolf01, I know for sure there will be problems with town gui and GS api :)
16:10:17 <andythenorth> ic111: I have no idea what people do with DL :)
16:10:22 <_dp_> because it kind of assumes TOWN_GROWTH_TICKS ~= TICKS_PER_DAY
16:10:30 <andythenorth> Eddi|zuHause gave me a 100% working 2 or 3 line DL patch once
16:10:54 <andythenorth> at the end of every year, it reset the year back again
16:11:02 <andythenorth> it might have had a counter to only do it once
16:11:12 <andythenorth> totally 100% bug free
16:11:12 <Wolf01> I have one patch with 3 characters
16:11:15 <ic111> Thats what I effectively do using the Date cheat
16:11:32 <andythenorth> had some interesting effects on things like…date sensitive industry animations
16:11:38 <andythenorth> and vehicle intro dates
16:11:39 <Wolf01> But for example we need to identify cases where daylength must be applied, like this one: (_tick_counter + t->index) % TOWN_GROWTH_TICKS == 0
16:12:09 <ic111> And using the Date cheat in my experience never caused problems.
16:12:44 <_dp_> Wolf01, that part doesn't seem affected by tics per day to me
16:13:19 <Wolf01> _dp_: no, but do you want cities to grog 4,8, 10 times faster?
16:14:00 <_dp_> Wolf01, faster in relation to game date not real time, right?
16:14:17 <_dp_> but I understand if some people don't
16:14:45 <milek7> imo problem with daylength is not that it is hard to implement, but everybody have diffirent opinion on how it should work :p
16:15:12 <Wolf01> The only biggest problem with DL is the economy, you have a train moving 8x faster and making 8x more profit for the same "day time"
16:15:31 <Wolf01> Dividing it doesn't work
16:15:52 <Wolf01> You will slow down the entire game pace
16:16:11 <Wolf01> Making 8x more real time to be able to purchase another train
16:16:55 <Wolf01> So we could compromise on economy, because we keep the same game pace, but just slowing down the time at which new generations will be introduced
16:17:04 <milek7> isn't that the point, to slow down game pace?
16:17:30 <andythenorth> I don’t think there is a single point
16:17:31 <_dp_> Wolf01, yeah, that why I'd leave tick-based stuff intact and suggest using basecost, town speed setting, etc.
16:17:44 <andythenorth> script daylength with mods
16:17:55 <andythenorth> it’s impossible to do in core
16:18:09 <andythenorth> there is no clear definition of what it is
16:18:28 <LordAro> ic111: so i can reproduce it
16:18:58 <andythenorth> segment and group the places where daylength would be applied (places using ticks)
16:19:08 <andythenorth> define static constants for them
16:19:19 <andythenorth> let newgrf modify those constanst, once, on game start
16:19:29 <LordAro> ic111: make bus route (nothing fancy); set "default" timetable start, offset & length; change timetable length down to 0 months, then to days, then to maximum (20)
16:19:31 <andythenorth> there will be about 8 or so probably
16:19:36 <LordAro> wait a bit; hard freeze
16:19:43 <LordAro> not entirely sure how much of that is necessary
16:19:46 <_dp_> andythenorth, newgrf again :(
16:19:49 <LordAro> but it seems to be stuck in a loop somewhere
16:20:06 <andythenorth> nobody has to design a feature in core, just an API
16:20:10 <_dp_> andythenorth, it's such a pain to configure server when everything is in a newgrf
16:20:28 <andythenorth> it’s such a pain to have options where none existed before :)
16:20:30 <LordAro> ic111: seems to happen when the bus finishes loading
16:21:23 <Wolf01> <andythenorth> nobody has to design a feature in core, just an API <- it would be cool if core allowed it
16:21:35 <andythenorth> well the API has to be in core
16:21:44 <andythenorth> but it abdicates having to make choices
16:21:51 <andythenorth> which is better in this case
16:22:14 <LordAro> ic111: gdb output seems to imply a loop somewhere around CorrectTimetableOffset
16:24:07 <_dp_> andythenorth, you still need to choose an API :p
16:26:42 <ic111> Indeed, setting the timetable length to zero is the problem.
16:27:20 <LordAro> yeah, i've just pinpointed the do {} while loop :)
16:27:23 <_dp_> making and maintaining API is much harder than just regular features, only worth it if there are many mods using that api
16:27:28 <Wolf01> andythenorth: And a DL API should decide to slow down the entire game pace even by slowing down vehicles making them travel a tile in a minute?
16:27:29 <LordAro> looks like it's missing a case for 0
16:27:43 <ic111> If you look into the loop in vehicle_base.h / CorrectTimetableOffset, then I shift the timetable by its length, *and* the exit condition depends on length and min_offset there
16:27:53 <andythenorth> Wolf01: if there’s a constant for that, then yes
16:28:05 <ic111> I just should prevent setting the length to zero, having a timetable of length zero is senseless.
16:28:14 <LordAro> ic111: incidentally, why are those functions in the header? shouldn't they be in the cpp?
16:28:21 <andythenorth> there is literally nobody in the community who could get a daylength patch through, socially
16:28:50 <ic111> (and because of that, I probably never triggered the case during playing ;-))
16:28:57 <Wolf01> I'm open to critiques and debates
16:29:21 <LordAro> ic111: see, having people who have no idea what they're doing is useful
16:29:58 <ic111> LordAro: Good question, probably I tried to mimic what I found before
16:30:22 <LordAro> ic111: in terms of some other feedback, the timetable window shouldn't be default, and i think its minimum size is far too wide
16:30:50 <LordAro> but still despite not knowing how to make it work, it looks a lot better than the current one
16:31:22 <ic111> LordAro: Making it work: Just don´t touch the length = zero case, or did I miss some additional problem?
16:32:27 <ic111> And: I´ll try to add comments in the forum thread with your points, but as you find quite a lot of them, please note problems there if I don´t do it.
16:32:49 <ic111> Getting five bugs / change requests in a chat increases the probability that you miss two of them ;-)
16:33:30 <LordAro> ic111: no, i've just not tried anything much beyond that :p
16:40:45 <ic111> Regarding opening the timetable window: There is a setting "Open timetable view (with orders, arrivals, departures) by default"; so this request is just about which value the setting takes by default
16:40:45 *** supermop has joined #openttd
16:40:47 <milek7> for me daylength should: increase ticks per day, lower payment rates
16:41:48 <milek7> and increase cargo production if it is day based (leave unchanged if tick based)
16:42:05 <supermop> ic111 the 24 hour clock works reasonably well for what you describe
16:42:49 <supermop> of course you still end up with thousands of passengers at your station at 04:00
16:44:36 <andythenorth> ic111 do you have a github repo for TIP?
16:45:47 <_dp_> milek7, imo there enough allmighty settings already. one setting should control one thing
16:46:17 <_dp_> milek7, if you want all 3 then do 3, one for daylength, one for payment one for cargo
16:46:25 <andythenorth> if you have a repo, you can ask to get binaries built on the ottd compile farm
16:46:36 <_dp_> but then, again, some if it is newgrfble so nothing gets done :(
16:46:39 <andythenorth> doesn’t have to be github, but eh, why use anything else?
16:47:06 <ic111> There was an attempt to set up an project there some years ago, as far as I remember parts of that project exist.
16:47:14 <andythenorth> I did it for NRT
16:47:18 <andythenorth> just clone openttd git repo
16:47:32 <andythenorth> dunno how you’d get a patch queue in there
16:47:38 <ic111> But things stopped before a repository was actually activated, and at that time, the repository wasn´ t that important as there was someone building windows binaries for me
16:47:56 <supermop> I feel like nrt is so much easier to test and solve problems with because of that
16:48:02 <andythenorth> queues throw away everything good about a repo, because they’re private
16:48:13 <Wolf01> andythenorth: what if we provide DL factor to grfs? One could use it for calculations?
16:48:44 <andythenorth> Alberth: or have I misunderstood queues?
16:49:00 <Alberth> depends on how yo see patch files
16:49:14 <Alberth> I regularly edit and shuffle those files
16:49:22 <Alberth> which is next to impossible in git
16:49:41 <Wolf01> I'll try a game with DL factor set to 1200
16:50:00 <Alberth> there is cherry-pick, but you need the entire commit machinery then
16:50:13 <Alberth> eg rename a variable that you added
16:50:18 <Alberth> trivial in a diff file
16:52:12 <ic111> It´s also a question of, which tools are you used to. I mean, given that I know how to use patch files, and didn´t use git so far (real world software development in my case is svn), switching to git because of a couple of bugfixes doesn´t seem senseful to me.
16:52:21 <Alberth> splitting or editing the change
16:53:35 <Alberth> git branches can't easily be stacked and changed
16:54:24 <Alberth> ie you have much better access to the changes that you make, rather than only to the state of the file at some point in the patch
16:55:19 <Alberth> unfortunately, hg seems to be moving away from patch queues too, afaik
16:56:16 *** orudge` has joined #openttd
16:56:17 *** ChanServ sets mode: +o orudge`
16:56:19 <andythenorth> Wolf01: dunno about providing DL to grfs
16:56:25 <andythenorth> the problem I can forsee there
16:56:30 <andythenorth> is grf authors vs. players
16:56:41 <andythenorth> but eh, am I here to sort out social problems? :P
16:57:21 <andythenorth> ic111 the upside of going to any repo that the compile farm can build…is binaries
16:57:32 <andythenorth> binaries => more players
16:57:42 <milek7> Alberth: interactive rebasing
16:57:52 * andythenorth would like to see 20 or 30 forks building on farm
16:58:01 <_dp_> andythenorth, oh, if only it worked like that...
16:58:19 <andythenorth> which is my sole data point so far
16:58:34 <andythenorth> only way to prove that wrong is try more
16:58:54 <_dp_> andythenorth, idk, feels like it kinda works coz 1 >> 0 (not talking NRT here) but meh...
16:59:14 <Alberth> milek7: I know that's git solution, but try changing something you changed in some commit spread all over the place
16:59:22 <Alberth> diff file gives me 1 file to edit
17:00:00 <Alberth> andI I don't get any match on code I didn't touch
17:00:46 <milek7> commit it, run rebase, move it under original commit you want to modify and change pick to squash
17:01:05 <milek7> but yes, more complicated than patch files ;p
17:01:27 <Alberth> I have also been editing patches higher in the queue to avoid merge conflicts
17:02:01 <Alberth> both git and hg are incredibly stupid in understanding there is nothing else but the sequence patches that they have
17:03:29 <_dp_> andythenorth, in other words there are not that many active players so if we make more binaries some of them are bound to get 0 interest
17:04:14 <andythenorth> and yet people apply patches and test them
17:04:17 <andythenorth> forums tell us that
17:04:18 <_dp_> andythenorth, while having binaries may increase interest in a patch it won't increase overall amount of players
17:04:23 <andythenorth> what’s the loss of having binaries?
17:04:37 * andythenorth is confused by the supposed downsides?
17:04:46 <andythenorth> computer does the work
17:05:11 <_dp_> andythenorth, nah, no loss rly, I'd like to see it too
17:05:49 <andythenorth> ok 461 FS issues left
17:06:02 <andythenorth> I have hit maximum boredom now
17:06:29 <andythenorth> I have a list of 15 more that I totally cannot review alone
17:07:02 <_dp_> andythenorth, just "binaries => more players" sounds a bit too enticing :)
17:07:15 * _dp_ hopes he picked a right word out of dictionary
17:07:21 <andythenorth> oh yeah, that was probably lame
17:07:43 <andythenorth> binaries => fractionally higher chance of useful feedback
17:09:50 <_dp_> andythenorth, it's a hack, but a damn good one :p
17:11:06 <supermop> I only play binaries
17:11:11 <supermop> can't be asked to compile
17:13:45 <_dp_> andythenorth, I didn't get all the details in that particular bug, but in general server feels very handicapped
17:14:15 <_dp_> andythenorth, I remember some check like that one giving me a lot of troubles
17:16:38 <_dp_> there is always something to improve
17:17:49 <_dp_> and any stored password is pretty much bound to be unencrypted
17:17:49 <andythenorth> ^ this is what I think has fucked OpenTTD a bit
17:18:03 <andythenorth> _dp_: well you could always store the key locally
17:18:06 <andythenorth> what could go wrong?
17:18:55 <_dp_> andythenorth, decryption key you mean?
17:19:05 <_dp_> andythenorth, that's not much different from plain text imo
17:19:19 <andythenorth> well you could put the key in a key store
17:19:23 <andythenorth> and keep the key for that locally
17:19:40 <_dp_> andythenorth, then you have a key to uncrypt a key :p
17:19:44 <andythenorth> can we talk about FIRS now?
17:19:48 <andythenorth> at least that can make progress
17:25:55 <_dp_> I think we have two mods vanishing already after I gave them access for configuring new firs server xD
17:28:19 <_dp_> it's a cb so you need to balance it properly
17:28:31 <_dp_> and find right grfs for everything
17:28:53 *** sla_ro|master has joined #openttd
17:29:50 <supermop> what needs to be fixed in unspooled?
17:30:31 <_dp_> also custom industry sets need custom mapgens in cb to fairly place industries for everyone
17:41:45 <andythenorth> supermop: find some FS issues to close, and I’ll tell you? o_O
17:41:51 <andythenorth> actually I don’t know :)
17:53:41 <Wolf01> I'm still reasoning about balancing income and running costs
17:56:11 <Wolf01> Maybe I should think about it while lucid dreaming, so I have a lot more time, but it's difficult and I always get up with headache
18:04:04 <supermop> hmm should I composit on tram bogies?
18:05:37 <supermop> need to redo whole template
18:06:15 <supermop> also... I take 1/8 in _ to be 4px
18:06:58 <supermop> feel like I don't like that somehow
18:07:15 <supermop> also I guess this is why all my bendy buses have such a gap
18:09:50 <supermop> if I build my trams out of 1/8 end cabs, and n*2/8 body segments, some of which may have wheels...
18:09:59 <supermop> I am generally happy
18:10:16 <supermop> but there is no natural place to put doors, espescially on older trams
18:11:22 <supermop> if I make the cabs 2/8 that works better for end doors, and modern aerodynamic cabs
18:11:58 <supermop> but then its unclear where to put the wheels, as the shortest trams will be all cab
18:12:52 <supermop> so can just make different rules for each generation of tram, but then why even bother with a system
18:14:35 <supermop> I guess compositing on doors could work
18:16:58 *** mindlesstux has joined #openttd
18:28:07 <Eddi|zuHause> the trams that i know usually have doors near the cab
18:28:27 <Eddi|zuHause> and sometimes in the middle
18:29:54 *** supermop_home has joined #openttd
18:31:11 <Eddi|zuHause> (that middle door is in a separate bendy section)
18:48:58 <supermop_home> Eddi|zuHause does that segment have wheels?
18:50:25 <supermop_home> the anglo American world doesn't have enough yellow buildings
18:50:55 <supermop_home> SE Asia, latin america, central and eastern Europe all have lots
18:51:03 <supermop_home> color never caught on here
19:03:58 <Eddi|zuHause> make east german houses: all grey
19:04:23 <supermop_home> what about before the war?
19:09:18 <Eddi|zuHause> supermop_home: i don't know about before the war... all relevant pictures are black and white
19:15:09 <Wolf01> supermop_home: with good accuracy you can go for brick-yellow and copper-oxyde for roofs
19:15:35 <Wolf01> But it might vary between different places
19:27:51 <andythenorth> now what shall I do?
19:48:49 <Alberth> 2837 could be done by saving the game and loading it again?
19:52:44 <Alberth> 5390 needs that nogo change, and even then I am not convinced, ie why would you need an event for a destroyed house?
19:59:27 <andythenorth> Alberth: I’ll close 2837
20:01:30 <andythenorth> 5390 - it’s an anti-griefing measure according to forum thread
20:01:42 <andythenorth> prevents vandalism on city-builder goal servers
20:01:51 * andythenorth didn’t read the whole thread :P
20:03:19 <andythenorth> after about 300 ‘no’, it’s getting harder to say ‘no’ :)
20:03:28 <andythenorth> esp. about stuff I domn’t understand
20:05:12 <Eddi|zuHause> whenever a game works like that, i immediately disable it
20:05:32 <Eddi|zuHause> but yes, that code would probably be platform specific
20:10:52 <andythenorth> that request is basically bollocks, no?
20:14:35 <andythenorth> I have NFI how you could have ’10% cargo dist'
20:16:38 <Alberth> 2155 seems a good feature
20:16:54 <Alberth> not used much likely, as there are very few dedicated servers
20:18:10 <frosch123> ottd command line could use a redesign :)
20:18:41 <frosch123> it has many weird things, but lacks close related ones
20:18:44 <Alberth> technically, you can allocate less than 100% capacity for a line to CD
20:19:00 <Alberth> which means you have space left
20:19:01 <frosch123> i believe there are multiple fs tasks about console options
20:19:18 <Alberth> not sure how you get pax to enter such a train though
20:19:31 <Alberth> as CD has to decide which pax not to move itself
20:19:41 <frosch123> like, it's exactly the case where every single fs task is bad, but by combining them you can put together something sane
20:20:37 <frosch123> like, why in the world is there a parameter to set the starting year?
20:21:20 <andythenorth> because we don’t have tags, I have been prefixing stuff in the titles
20:21:26 <andythenorth> I can’t make much use of categories
20:21:47 <andythenorth> currently you can search e.g. “Timetable:”, “Orders:” “Scenario Editor:”
20:21:51 <andythenorth> trying to group stuff
20:22:32 <andythenorth> w.r.t 6466, unless I misunderstood cdist
20:22:43 <andythenorth> most of the magic happens when moving cargo into and out of vehicles
20:23:06 <frosch123> i have no experience with cdist, but i remember that there is cargo with unassigned destination sometimes
20:23:20 <frosch123> possibly only intermediate until the next calculation finishes
20:23:36 <andythenorth> that will be loaded onto vehicles without established links afaik
20:23:45 <andythenorth> actually no real clue :)
20:24:33 <andythenorth> but anyway, I cannot think how to even describe “10% manual and 90% cdist”
20:25:17 <Alberth> "unassigned" it will hop on to any unfilled train
20:26:44 <Alberth> andy, the problem is likely how to point out pasengers that should not be routed by cdist
20:27:27 <Alberth> hmm, likely you can even do that given the capacity on a train
20:27:43 <Alberth> with a few unassigned, that should do the trick
20:28:16 <frosch123> Alberth: i do not see much point in a single seed variable
20:28:29 <frosch123> how much can you randomise with 32bits for everything?
20:29:01 <V453000> I see '32bits' and my eyes get red
20:29:25 <frosch123> so you would rather need something which gives you a new random number every time
20:31:20 <frosch123> hmm, how was it worded in that c++ conference: we are not opposed to the idea, but it requires more thought on the etails
20:32:35 <Alberth> andy, 5464 delivers 1 unit of cargo in round robin fashion to the industries, that will take forever to distribute eg ship deliveries
20:33:57 <frosch123> "EWG found the use case compelling, because (...). Several details still need to be worked out. "
20:33:58 <andythenorth> w.r.t to the seed, is anyone actually planning to use it?
20:34:10 <frosch123> andythenorth: how does sound for a closing reason? :p
20:36:51 <andythenorth> frosch123: add a closing category “no pony today” ?
20:38:55 <frosch123> advanced settings selection is ill-formed, no diagnostics required
20:39:22 <frosch123> did i read too many c++ papers lately?
20:42:49 <andythenorth> Alberth: should I close 5646, or does it need a fairer review?
20:43:03 <andythenorth> it’s only 2013, if it was older I’d bin it
20:43:25 <Alberth> ie cargo to multiple industries
20:43:45 <Alberth> patch is not fast enough
20:43:55 <andythenorth> we could just extend cdist
20:44:03 <andythenorth> create virtual links for move-to-industry
20:44:18 <Alberth> you don't distribute 1000 candies one-by-one to 25 children
20:44:29 <andythenorth> or…use stations near the N tile and stop worrying about it
20:44:54 <andythenorth> ah…it’s lipstick on a pig, but if it was indicated at the station which industry was actually accepting…
20:44:58 <andythenorth> that would help :P
20:45:23 <Alberth> closest to the label afaik
20:46:01 <Alberth> I can see the value in the issue, but the solution just uses the wrong approach
20:46:27 <Alberth> it needs to calculate how to distribute, and then do it in one sweep
20:47:16 <Wolf01> <Alberth> you don't distribute 1000 candies one-by-one to 25 children <- why not? And I also wait they finish the previous one before giving them another
20:47:58 <Wolf01> 10 candies to 25 children could be a challenge
20:48:19 <Alberth> not for speed, most likely :p
20:48:32 <Wolf01> Maybe I should just put one against the other in a battle royale and give all of them to the survivor
20:49:27 <Wolf01> And I think that is what happened with the industries, one won
20:53:04 <andythenorth> that one is just a yes/no even for the principle
20:54:36 <Alberth> does an industry always have a station name?
20:54:50 <Alberth> in particular, does default set have that?
20:55:29 <Alberth> likely you can get away with defining an industry without a station name
20:56:22 <Alberth> if so, something has to produce station names
20:57:10 <Alberth> which would be "player names", as these are the only available names
20:57:53 <Alberth> if stations must always provide a name, then it's a matterof changing the code into assigning
20:58:30 <Wolf01> So it can be defined via grf
20:58:44 <Wolf01> andythenorth would be happy
20:58:48 <Alberth> which would at least be feasible in the latter case
20:59:15 <Alberth> you can have station names in newgrf, I am asking if it's obligatory
20:59:33 <Alberth> ie is there always a name specific for an indsutry?
21:00:30 <Alberth> if not, the entire idea isn't even feasible currently
21:01:26 <Wolf01> With specific name you mean like "Fartbottom Coal mine"?
21:02:06 <andythenorth> it goes to “town coal mine #3” or something
21:02:15 <Wolf01> Industries for sure have a city bound
21:03:30 <Wolf01> andythenorth: NoStationNames spec
21:05:00 <andythenorth> goes as far as “Lower [town name]” then “[town name] Station #16"
21:05:44 <andythenorth> changing original gameplay is against our objectives?
21:05:48 <andythenorth> so 6017 can be closed
21:06:00 <andythenorth> also, players will complain
21:06:19 <andythenorth> that doesn’t mean much, but in this case they will ask why station names for industries are now bugged
21:06:25 <andythenorth> because they all go to numbers
21:07:05 <andythenorth> unless we just patch oil rigs
21:07:23 <andythenorth> but that will likely also get applied to newgrf industries using the station tile
21:12:53 <V453000> I was trying to use lists as templates :D
21:13:27 * andythenorth probably misunderstood
21:14:14 <Alberth> deepcopy is the brute-force approach :p
21:15:01 <V453000> need to convert it to functions or something I guess
21:15:38 <andythenorth> openttdcoop lost it’s cert?
21:15:43 * andythenorth getting warnings
21:16:34 <andythenorth> SSL seems to be broken
21:16:52 <andythenorth> only according to Google
21:17:13 <andythenorth> anyway, V453000 all those [n+1, n+1, n+1] offsets
21:17:24 <andythenorth> is it always n+1, n+2 etc?
21:17:34 <andythenorth> looks like you don’t need to write that out
21:17:41 <V453000> usually, sometimes it's hardcoded number
21:17:49 <V453000> there's a weird pattern
21:17:58 <andythenorth> probably fine written out then
21:18:27 *** chomwitt has joined #openttd
21:18:30 <V453000> it's n, n+1, n+2 for the first index, but for example n+7, n+6, n+5 for the last index
21:18:39 <V453000> which means they go against each other and eventually they meet somehow
21:18:41 <V453000> which I edit manually
21:18:46 <V453000> which is why it's all written out yeah
21:18:58 <andythenorth> which lucky patch shall I test first? :P
21:19:10 <andythenorth> they are mostly unappealing eh?
21:19:22 <andythenorth> hth can we make testing patches more fun? o_O
21:19:37 <andythenorth> it’s totally like eating unwanted vegetables right now
21:20:30 <andythenorth> so it’s not like I asked people to send me this work
21:20:49 <andythenorth> same goes for devs eh?
21:24:22 <_dp_> I think it's only worth testing patches that are somewhat reviewed
21:24:59 <_dp_> or where testing also kind of reviewing, like UI patches
21:26:05 <andythenorth> there’s still 75 to review in that case
21:27:03 <V453000> loading BRIX with NUTS could cause desyncs
21:27:45 <frosch123> when static and non-static grfs try to mess with each other, the static one is dropped
21:27:52 <andythenorth> @seen supercheese
21:27:52 <DorpsGek> andythenorth: supercheese was last seen in #openttd 20 hours, 46 minutes, and 37 seconds ago: <Supercheese> and that's good enough for me
21:27:57 <V453000> I guess that's because NUTS is trying to check for signals
21:28:57 <V453000> I guess that only happens for multiplayer, right?
21:29:25 <Eddi|zuHause> need to load BRIX non-static, or drop the check from NUTS
21:29:46 <V453000> why don't I get the error in single player then ._.
21:30:10 <Eddi|zuHause> hm, then i don't know
21:30:34 <V453000> well if it's desync related protection I would understand it only applying to MP
21:30:49 <Eddi|zuHause> anyway, the thing is, the NUTS check cannot have different result depending on whether static BRIX is loaded or not, so the only valid solution is to always not load BRIX
21:31:23 <V453000> yes, or load it normally without static
21:31:41 <Eddi|zuHause> yes, but that must be done by the server owner
21:36:15 <frosch123> what should i do with all the fish that i accidentially get from destruction robots?
21:36:47 <Eddi|zuHause> give it to the dolphins?
21:37:08 <frosch123> that achievement has no hints how to get it
21:37:19 <frosch123> it's an easter dophin or something
21:37:27 <V453000> put fish into rocket, launch it
21:38:00 <Wolf01> Did you try to put them on the water again?
21:38:35 <frosch123> i actually got excited when i thought about the rocket
21:38:45 <Eddi|zuHause> frosch123: i just randomly said that, didn't know there was an achievement :p
21:38:46 <frosch123> and now you tell me that it should actually work? :o
21:39:21 <Wolf01> Try to put 42 fishes on the rocket
21:40:11 <frosch123> hmm, i only launched one
21:40:17 <frosch123> what is the stack size for fish?
21:45:14 <frosch123> andythenorth: i would like smarter hotkeys
21:45:29 <frosch123> like rotate selected object
21:45:48 <andythenorth> presumably we can’t just give every UI control a UUID?
21:45:59 <andythenorth> and then let players map keys to UUIDs?
21:46:40 * andythenorth looks how it works
21:46:48 <_dp_> andythenorth, an entry in hotkeys.cfg is a nice UUID ;)
21:47:07 <frosch123> andythenorth: not everything has a gui button
21:47:20 <frosch123> like all those weird cycle signal type
21:47:34 <frosch123> or toggle depot orientation
21:48:08 <_dp_> rotate hotkey would be very nice, it's the one probably missing the most right now
21:49:06 <_dp_> or mb more like cycle hotkey indeed, that cycles signals too
21:50:03 <andythenorth> “First join us on irc, and discuss what you are trying to achieve"
21:50:27 <andythenorth> “Patches with no prior discussion are rarely accepted from contributors who aren’t known to us"
21:50:46 <Eddi|zuHause> i always want to reply to suggestion threads like "how did nobody ever think of that before?" and each word is a link to a previous suggestion about the exact same thing. but then i can't be bothered to look for all those threads
21:51:09 <andythenorth> “It can be disheartening to attach a patch to the tracker and then nobody even comments. To avoid this, conversation in advance is best"
21:51:23 *** orudge` has joined #openttd
21:51:23 *** ChanServ sets mode: +o orudge`
21:51:27 <andythenorth> “Discussing in irc is still no guarantee, and some days the channel is quiet."
21:51:36 <andythenorth> “But eh, it’s just more fun this way.”
21:52:49 <andythenorth> “FS ticket comments are a poor way to have a conversation, and can seem very formal, cold, dismissive or rude when not intended to be any of those"
21:53:25 <andythenorth> “Often it turns out that individual patches aren’t much use in isolation, or are fixing part of a bigger issue.”
21:53:41 <andythenorth> “Only when the bigger issue is tackled is it worth the effort to change OpenTTD”
21:53:52 <_dp_> and bundled together they are too big to review :p
21:55:38 <supermop_home> tram segment sprites, assuming each part can have and either end: a cab, a 'blind' end with no cab, or a gangway to next vehicle
21:57:00 <andythenorth> “Sometimes patches get no comments simply because NOBODY LIKES YOUR IDEA, but nobody wants to say so in case it looks rude ;)”
21:57:05 <supermop_home> not all combinations are likely to occur, so I don't need nomeclature that covers everything
21:57:56 <Eddi|zuHause> supermop_home: so what was the question?
21:57:58 <andythenorth> supermop_: got a mockup?
21:58:46 <frosch123> btw. i am interested in learning xslt
21:58:50 <supermop_home> but something like " A is a part with 2 cabs. B is a part with one cab and one gangway. C is a part with one cab and one blind end. D is a part with two gangways. E is a part with two blind ends. F is a part with one gangway and one blind end. etc
21:59:00 <supermop_home> is not very helpful or intuitive
21:59:12 <frosch123> so if you can find a use case to use xslt in ottd / some patch... :p
21:59:57 <supermop_home> other way is something like A= cab, b= gangway etc, and call each one "part_X_AB" etc
22:00:17 <supermop_home> but it seems clunky to use so many digits
22:00:18 * _dp_ tried to understand xslt few times but failed
22:00:35 <Eddi|zuHause> supermop_home: parts with gangway are articulated front/middle/end, parts without gangway are front/end
22:01:13 <Eddi|zuHause> so cab+gangway = articulated front, gangway+gangway = articulated middle, gangway+blind = articulated end
22:01:13 <supermop_home> Eddi|zuHause some older tram might have articulated parts with no gangway
22:01:21 <andythenorth> frosch123: I am trying desperately to get xslt out of my life
22:01:33 <Eddi|zuHause> cab+blind = front, blind+blind = end
22:01:43 <andythenorth> eh the website is django?
22:01:58 * andythenorth wonders if it’s running a wsgi stack
22:02:15 <supermop_home> question is mostly, how to give these sprites short but meaningful systematic names
22:02:21 <andythenorth> actually doesn’t matter, Diazo works outside WSGI
22:02:44 <andythenorth> if you want to learn xslt, we can reskin the website without touching django
22:03:23 <andythenorth> you will then learn why that is a terrible idea, unless you have literally no control over the website code (happens when you’re selling consulting services sometimes)
22:03:49 <supermop_home> ooops forgot wheels
22:04:27 <_dp_> andythenorth, omg, instead of fixing the website let's write another website that will fix our website?
22:04:41 <andythenorth> there are valid use cases
22:04:46 <andythenorth> they’re not common
22:04:47 <frosch123> hmm, this time the fish worked
22:04:56 <frosch123> i wonder what i misclicked last time
22:05:36 <andythenorth> <replace fish=“click” content=“click” />
22:05:56 <andythenorth> so eh, 75 patches to review
22:05:57 <frosch123> i also do not find a way to put them back into the water
22:06:00 <_dp_> andythenorth, spamming identical websites?
22:06:07 <frosch123> blue prints to not capture fish population
22:06:18 <frosch123> and i cannot put them manually either
22:06:35 <andythenorth> of those 75, only 2 interest me
22:06:43 <andythenorth> and 1 just fixes a type in code :P
22:06:49 <frosch123> maybe i can put them into the reactor
22:07:17 <andythenorth> there are maybe 5 more patches that are like “I should look at this to be a good person"
22:07:21 <andythenorth> but eh, it’s boring isn’t it/
22:07:23 * _dp_ not sure if frosch is talking about fish or patches
22:07:32 <andythenorth> patches are my problem today
22:07:38 <andythenorth> frosch123 is fishing
22:07:39 <Eddi|zuHause> you think being a good person is boring?
22:07:52 <andythenorth> when pushed to answer
22:08:07 <V453000> :D foundation sprite 1666 is broken
22:08:46 <Eddi|zuHause> you know, the number 666 being evil has nothing to do with it being a 6 repeated 3 times
22:09:08 <Eddi|zuHause> because back when that story was written, the "arabic numbers" weren't invented yet
22:09:22 <andythenorth> it’s just Nero isn’t it?
22:09:39 <Eddi|zuHause> yes, allegedly, probably
22:09:58 <andythenorth> unless that’s false flag :P
22:10:03 <andythenorth> wheels in wheels in wheels
22:10:11 <Eddi|zuHause> but still, back then the number 666 would be written with greek letters
22:10:31 <Eddi|zuHause> or hebrew letters
22:10:33 <andythenorth> so patches via FS - mostly dead then? o_O
22:10:49 <Eddi|zuHause> and they were different letters for 6, 60 and 600
22:11:07 <frosch123> V453000: so 1666 contains every roman number exactly once, and sorted? MDCLXVI
22:11:34 <Eddi|zuHause> frosch123: that misses the double-M and double-D
22:11:42 <Eddi|zuHause> for 5000 and 10000
22:11:57 <frosch123> Eddi|zuHause: if i accept those, it's infinite
22:14:15 <Eddi|zuHause> frosch123: also, you can make a chain 16/166/1666/... or 6/16/66/166/...
22:14:30 <frosch123> the reactor does not accept fish
22:15:10 <Eddi|zuHause> frosch123: try mr fusion from the far future of 2017?
22:22:07 <V453000> I fixed a BRIX bug \o/
22:22:47 <andythenorth> I avoided finishing FIRS
22:22:53 <andythenorth> or even thinking about newgrf
22:23:06 * andythenorth has had holiday in Flyspray
22:23:15 <V453000> by making sure finishing of other people ideas was avoided? :D
22:23:29 <andythenorth> V453000: you should get ‘close ticket’ rights :)
22:23:39 <andythenorth> would be spree for you
22:24:09 <V453000> haz time to read ticket not
22:24:28 <V453000> my cat language is getting less readable by the minute
22:24:39 <andythenorth> total sense makes to me
22:24:50 *** FLHerne has joined #openttd
22:26:04 <andythenorth> is nice to have break from newgrf
22:26:18 <andythenorth> probly time haz FIRS 3 done
22:26:39 <V453000> I'm just going through BRIX and wondering what do I REALLY want to do for next version
22:26:52 <V453000> is a bunch of smaller fixes which isn't particularly fun to do :D but eh
22:26:56 <V453000> might also rework all trees
22:27:48 <V453000> yeah I should do that first
22:28:01 <V453000> need a break from the python thing for today
22:28:10 <andythenorth> trees not aweome awlready?
22:28:15 <V453000> rewriting that index definition needs to be left for tomorrow
22:28:18 <V453000> trees are super fucked
22:28:23 <V453000> some of them are good
22:28:32 <V453000> but most of them are not and are totally different style from good ones
22:28:46 <V453000> + no trees for arctic, tropic or toyland, and some missing for temperate
22:30:49 <V453000> but yeah otoh 344 models or how many
22:32:00 <andythenorth> which is good? o_O
22:33:45 <V453000> that's kind of insane
22:34:23 <andythenorth> do all same tree? o_O
22:35:18 <V453000> I'd like to avoid that
22:35:39 <V453000> might just do something simplistic
22:36:14 *** supermop has joined #openttd
22:36:32 * andythenorth will await results
22:41:45 *** sim-al2 has joined #openttd
22:45:10 *** tycoondemon has joined #openttd
23:06:13 *** luxtram has joined #openttd
23:30:23 *** mindlesstux has joined #openttd
23:52:56 <Eddi|zuHause> now, is the result of "such sleep" actually "find sleep"? [sorry, joke works better if you know german]
23:54:18 *** Wolf01 is now known as Guest2865
23:54:19 *** Wolf03 is now known as Wolf01
23:55:24 <Wolf01> Eh, ISP got nuked by blackout
23:55:43 <Eddi|zuHause> you're in serious need of a new ISP
continue to next day ⏵
