Back to Home / #openttd / 2015 / 02 / Prev Day | Next Day
#openttd IRC Logs for 2015-02-15

---Logopened Sun Feb 15 00:00:14 2015
00:07-!-smoke_fumus [~smoke_fum@] has quit [Read error: Connection reset by peer]
00:31-!-itsatacoshop247 [] has quit [Read error: Connection reset by peer]
00:32-!-itsatacoshop247 [~itsatacos@2601:9:1180:237:f16d:3c7a:3ff5:b6bc] has joined #openttd
00:38<Diablo-D3>that one popular citybuilder mod
00:38<Diablo-D3>it should require coal for the power plant at some point
00:44-!-Netsplit <-> quits: jonty-comp, Cursarion, glevans2, eQualizer, Extrems2, Eddi|zuHause, raincomplex
00:45-!-Netsplit over, joins: glevans2, Extrems2, Eddi|zuHause, eQualizer, Cursarion, jonty-comp
00:51-!-raincomplex [] has joined #openttd
00:56-!-Eddi|zuHause [] has quit []
00:56-!-Eddi|zuHause [] has joined #openttd
01:27-!-shirish [~quassel@] has joined #openttd
01:34-!-Celestar [] has joined #openttd
01:59-!-KenjiE20 [] has quit [Quit: WeeChat 1.0.1]
02:09-!-shirish_ [~quassel@] has joined #openttd
02:10-!-shirish_ [~quassel@] has quit [Read error: Connection reset by peer]
02:10-!-shirish [] has quit [Read error: Connection reset by peer]
02:15-!-Progman [] has joined #openttd
02:29-!-luaduck is now known as luaduck_zzz
02:48-!-sla_ro|master [slamaster@] has joined #openttd
02:57-!-andythenorth [] has joined #openttd
03:04-!-Alberth [~alberth@2001:981:c6c5:1:be5f:f4ff:feac:e11] has joined #openttd
03:04-!-mode/#openttd [+o Alberth] by ChanServ
03:08-!-DanMacK [] has joined #openttd
03:09-!-roidal [] has joined #openttd
03:18-!-Progman [] has quit [Remote host closed the connection]
03:27<supermop_>yo andy
03:30-!-Pensacola [] has joined #openttd
03:30-!-gk [] has quit [Ping timeout: 480 seconds]
03:40<@Alberth>out secret plot to take over the steel factory production is out!
03:42-!-OsteHovel_ [] has quit []
03:46<andythenorth>a bug :)
03:46<andythenorth>or a badly named industry?
03:46<@Alberth>yeah, I make lots of stupid little typos these days :( Supercheese
03:46<@Alberth>not by the looks of it
03:46<Supercheese>Maybe the ore is both ferriferous and carboniferous
03:47<andythenorth>I can barely type anymore
03:47<andythenorth>must be some kind of global decline :P
03:49<supermop_>so i've been making these fake dieter trams as placeholders for real melbourne trams
03:50<supermop_>but now im tempted to add extra trams that shouldn't be in the melbourne tram set just so i get to draw more variations on these fake trams
03:51<V453000>as soon as realism limits you in any way, it is bad :)
03:52<@Alberth>+1 :D
03:52<supermop_>i like how that guy blurred out his company name as if sensitive information
03:52<@Alberth>yeah :p
03:52<@Alberth>maybe his player name
03:52<Eddi|zuHause>uhm, limiting yourself is the entire point of a game
03:53<supermop_>the extra trams exist in real life, they are just identical to another class except for manufacturer of the brakes
03:54<supermop_>but adding them lets be turn another coffee grinder into a tram!
03:55<@Alberth>Eddi|zuHause: sure, but realism is not by definition a good limit
03:56<Eddi|zuHause>but there's still a difference between "not always" and "never"
03:58*andythenorth offers platitudes
03:58<V453000>OMFG philosophical questions
03:58<andythenorth>choosing to limit by realism is good
03:58<andythenorth>arbitrarily only limiting by realism is not good
03:59<V453000>if supermop is having one tram in his set due to realism, it might be nicer to have 10 trams for variety :P
03:59<V453000>making things roughly believable to create the atmosphere is nice, copying it 1:1 is not
04:00<Eddi|zuHause>tbh, i have not understood supermop_'s approach enough to judge that
04:09<Supercheese>Hurray for easy-to-implement feature requests
04:11-!-DanMacK [] has quit [Quit: Page closed]
04:11<supermop_>well ill model and render for now
04:11<roidal>from when is 'days_in_transit' of a CargoPacket counted on?
04:12<roidal>at the moment of creation, or is it the time being loaded into a vehicle?
04:12<Supercheese>pretty sure it's on loaded
04:13<roidal>so there is no penalty in payment for long waiting cargo at stations?
04:13<Supercheese>I do believe that is the case
04:13<roidal>ok, thx
04:14<Supercheese>other than the station rating being too low and the cargo disappearing
04:14-!-Pol [] has joined #openttd
04:17-!-LadyHawk- [] has joined #openttd
04:19-!-liq4 [] has joined #openttd
04:19-!-TrueBrain_ [] has joined #openttd
04:20-!-zeknurn` [] has joined #openttd
04:20-!-__ln___ [] has joined #openttd
04:21-!-Netsplit <-> quits: Kurimus, liq3, __ln__, dihedral, zeknurn, Stimrol, JGR, TrueBrain, dfox, Diablo-D3, (+4 more, use /NETSPLIT to show all of them)
04:21-!-zeknurn` is now known as zeknurn
04:21-!-LadyHawk- is now known as LadyHawk
04:21-!-Netsplit over, joins: andythenorth
04:22-!-TheMask96 [] has joined #openttd
04:24-!-Netsplit over, joins: Kurimus
04:28-!-Netsplit over, joins: JGR
04:29-!-Netsplit over, joins: Stimrol
04:35-!-dihedral [] has joined #openttd
04:35-!-dfox [~dfox@] has joined #openttd
04:39-!-Diablo-D3 [] has joined #openttd
04:51-!-theholyduck [~theholydu@2602:ffea:1:73b::8c23] has quit [Remote host closed the connection]
04:52-!-theholyduck [~theholydu@2602:ffea:1:73b::8c23] has joined #openttd
05:00<andythenorth>reduce() is gone in python 3 maybe
05:04<@Alberth>what does it do?
05:04<@Alberth>some functional stuff got moved to packages iirc
05:04-!-HerzogDeXtEr [] has joined #openttd
05:06<andythenorth>wondering if I should use that, or try and write a for loop
05:06<andythenorth> years = sorted(reduce(set.union, [(variant.intro_date, variant.end_date) for variant in self.model_variants], set()))
05:06<andythenorth>dunno what that does
05:06<@Alberth>"Removed reduce(). Use functools.reduce() if you really need it; however, 99 percent of the time an explicit for loop is more readable."
05:07<Xaroth|Work>it calls set.union on the generated list
05:08*andythenorth tests
05:09<@Alberth>depending on what you do with years, it might need an extra list( sorted(...) )
05:12*andythenorth experiments
05:13-!-Wolf01 [] has joined #openttd
05:38-!-TrueBrain_ is now known as TrueBrain
05:39-!-oskari89 [] has joined #openttd
05:49-!-KenjiE20 [kenjie20@] has joined #openttd
05:52-!-roidal_ [] has joined #openttd
05:57-!-roidal [] has quit [Ping timeout: 480 seconds]
06:08-!-heffer [] has joined #openttd
06:15-!-gelignite [] has joined #openttd
06:23-!-Quatroking [] has joined #openttd
06:51-!-Yotson [~Yotson@2001:980:6ac8:1:e1e3:89a3:4edd:e9c1] has joined #openttd
07:10-!-itsatacoshop247 [~itsatacos@2601:9:1180:237:f16d:3c7a:3ff5:b6bc] has quit [Ping timeout: 480 seconds]
07:12-!-supermop_ [] has quit [Ping timeout: 480 seconds]
07:17*andythenorth is in habit of ctrl-d for python
07:17<andythenorth>but that also kills my shell
07:17<andythenorth>don’t press it at wrong time
07:21<@Alberth>don't use python interactively :)
07:21<andythenorth>I try not to :)
07:21<andythenorth>except to learn things
07:21<andythenorth>unrelated; I have these clunky args for make, like ‘NO_MP=False’ and ‘COMPILE_FASTER=True’
07:21<andythenorth>should they be —options of some kind?
07:22<andythenorth>or does that likely conflict with make options?
07:22<andythenorth>they look stupid when I type them
07:23<@Alberth>-D X=Y iirc, but that doesn;t help a lot
07:24<@Alberth>you can set them with a default value in the makefile
07:24<andythenorth>I think they are already, maybe
07:24<andythenorth>maybe I should just make them look less stupid
07:24<@Alberth>so you only need to type them when they are changed
07:24<andythenorth>shorter names
07:24<andythenorth>no double negative :P
07:25<@Alberth>hide them in a script :)
07:25<andythenorth>make faster
07:25<andythenorth>make only_a_bit_faster
07:25<@Alberth>that would be feasible too
07:26<andythenorth>make without_mp_so_python_doesn’t_crash_on_error
07:26<@Alberth>yeah, but you have to type the target every time
07:26-!-frosch123 [] has joined #openttd
07:26<andythenorth>and no autocomplete
07:26<@Alberth>./make_faster shell script ?
07:27<@Alberth>I often have a ./mk script with all defaults
07:28<andythenorth>shell script with - options? o_O
07:29*andythenorth ponders
07:29-!-Celestar [] has quit [Quit: Leaving.]
07:29<andythenorth>probably overengineered for the case :)
07:30<@Alberth>just make a few more scripts, with 4 scripts you have every option you ever dreamed about :p
07:31<@Alberth>not not often used cases are probably better handled as Makefile targets
07:32-!-Supercheese is now known as Guest5518
07:32-!-Supercheese [] has joined #openttd
07:34<@Alberth>keep your build logic at one place, eg in the Makefile
07:35<andythenorth>ok after 2 hours of hunting an unrelated bug, I am looking at this reduce() thing again
07:38-!-Myhorta [] has joined #openttd
07:38-!-Quatroking [] has quit [Read error: Connection reset by peer]
07:38-!-Guest5518 [] has quit [Ping timeout: 480 seconds]
07:38-!-chillcore [~chillcore@] has joined #openttd
07:39<chillcore>hello al
07:40-!-Quatroking [] has joined #openttd
07:41<@Alberth>hi hi
07:41<chillcore>going to polish some more ... is there still some other known in MHL prob besides terrain presets not looking good?
07:41<chillcore>hello Alberth
07:42<@Alberth>/me gives wax and a new cloth to chillcore
07:42-!-Romazoon [] has joined #openttd
07:42<Eddi|zuHause>do you also czech?
07:42<chillcore>I will go over all commits int a little bit
07:42*andythenorth wishes he could hold datastructures in his head
07:43<Eddi|zuHause>trees drive you crazy :p
07:43*chillcore accepts wax and new cloth from Alberth but would kindly would request second cloth
07:43<chillcore>shining with dirty cloth does not work very well ;)
07:44<@Alberth>/me gives a shelf full of new cloth
07:44<chillcore>damned typing ... so much better in forums
07:45*chillcore goes WOOT shelf and all ...
07:46<chillcore>trees eddi ... what do you men exactly? too many or ...
07:46<chillcore>sometimes they generate slow I do not know why
07:46<Eddi|zuHause>chillcore: navigating and following algorithms that go through large trees
07:47<Eddi|zuHause>tree (data structure)
07:47<Eddi|zuHause>not tree (plant)
07:47-!-Romazoon [] has quit [Remote host closed the connection]
07:47<chillcore>ah yes ... feel free to ...
07:47<chillcore>in my head I still do BASIC :P
07:48<@Alberth>no recursive data structures in basic?
07:48<chillcore>hmm I alwayds avoided the topics about pointers and such
07:48<chillcore>they confuse me
07:49<chillcore>like how you do guis ... I would love to be able to do that but ...
07:49<chillcore>getting there though
07:49<@Alberth>it's just like a piece of paper with an address to go to, on it
07:49<Eddi|zuHause>tree algorthisms are often laid out so they describe the process that happens on each node. and then dive either into the parent or children and come back with magic values
07:50<chillcore>I know ... alberth ... I really should sit down one day and take a good practise at it
07:50<Eddi|zuHause>pointers for trees/lists are not that bad. it gets crazy when you do pointer arithmetics
07:51<Eddi|zuHause>but BASIC probably couldn't do that
07:51<@Alberth>yeah, like recursion, you don't think in terms of overall structure, you just handle one single node, without caring where things point to
07:51<chillcore>I see Eddi ... if you are refering to querying the values by query box ... I put that into seperate functions for the hotkeys later
07:51<chillcore>alhtough I could simulate a button click maybe
07:52<chillcore>which I will have to for scriptts/AIs
07:52<Eddi|zuHause>chillcore: i don't know what you're talking about
07:52<chillcore>not there yet
07:52<chillcore>keep it like that eddi :P
07:52<chillcore>I mean np ofcourse
07:54<chillcore>Alberth ye that is the thing ... not thinking enough in modern languages yet
07:54<chillcore>I started reading another book but I am at chapter 8 only ... this time skipping nothing
07:55<chillcore>then MHL happened
07:55<chillcore>during my off time I mean
07:55<chillcore>s off/no internet
07:55<@Alberth>language isn't that important, it's a mental step that you can safely ignore everything around the node
07:56<@Alberth>or around the function, in case of recursion
07:56<chillcore>ok ...
07:56<@Alberth>you can make a recursive function?
07:57<chillcore>for loop and such?
07:57<chillcore>if I have to yes
07:57<@Alberth>no, a function calling itself
07:57<chillcore>never tried that untill now .. but saying it like that ye I know about it
07:58<@Alberth>knowing and writing are very different things :)
07:59-!-Romazoon [] has joined #openttd
07:59<@Alberth>if you are wondering about making one, sum(N) of factorial(N) are simple ones
08:00<chillcore>ah ok
08:00<@Alberth>or difficult ones, as you are immediately confronted with recursion
08:00<@Alberth>no way around it :)
08:01<chillcore>I have this tendency to jump in on the deep end and see if I float
08:02<chillcore>what you talk about is on the other end I guess
08:02<Romazoon>is there a relatively up to date known bug list for Iron horse 1.5.0? ( box car have zero capacity when refitted to mail in a game i just launched )
08:02<Eddi|zuHause>Romazoon: bugs go to the devzone
08:03<Romazoon>thanks for the headers, i lll report it asap
08:04<@Alberth>chillcore: yep, with small examples, it's easier to get the core idea, less noise from unrelated parts
08:06<chillcore>I'll have a play around once I get these pacthes working properly
08:06<chillcore>not putting time pressure on myself but ... 1.5 is nearing and there is much to check
08:07<chillcore>for forgotten parts
08:07<chillcore>or parts that are not adjusted while bumping
08:07<chillcore>not that ic111 did this on purpose but he rushes a bit most of the times
08:08<Eddi|zuHause>chillcore: if you want "the deep end" of tree algorithms, you study compiler construction :p
08:08<chillcore>hehe ye eddi ... maybe in a few years
08:09<chillcore>then again is one more needed?
08:09<@Alberth>always :p
08:10<@Alberth>compilers come in many forms, not only things that are called 'compiler'
08:10<chillcore>I see
08:10<@Alberth>ie your BASIC interpreter is also a compiler
08:11-!-blathijs [] has quit [Ping timeout: 480 seconds]
08:11<@Alberth>and mostly theoretical, openttd is also a compiler, it translates realtime user input to displayed pictures :p
08:11<@Alberth>ie basically any program that takes input and produces output :p
08:11<chillcore>ah just like Java VM executes Java
08:12<frosch123> <- somone has a broken compose key :p
08:12<@Alberth>chillcore: yes, and the Java byte code generator on top of it
08:13<Eddi|zuHause>that took me a few tries :p
08:14<@Alberth>grfcodec also counts as compiler, and nml of course
08:15<chillcore>frosh: tt-forums ... I thought it was a safe haven form the idiots over at steam ... I was wrong
08:15<frosch123>it's still the internet
08:15-!-blathijs [] has joined #openttd
08:15<frosch123>though no idea why you would go to steam
08:15<chillcore>play games
08:15<Eddi|zuHause>nothing is ever safe
08:16<Eddi|zuHause>10 years ago it was leetspeek
08:16<Eddi|zuHause>all the kiddies did it.
08:16<frosch123>do you imply V is older than 10?
08:17<Eddi|zuHause>physically or mentally? :p
08:17-!-Myhorta [] has quit [Ping timeout: 480 seconds]
08:17<chillcore>N0 1375 n07 90 7h3r3 eddi
08:17<chillcore>stupid indeed
08:18<chillcore><i never even bothered trying to read it
08:18<Eddi|zuHause>there were also dialects of it, the one which more relies on numbers, and the one which more relies on /\|
08:19<chillcore>ye ... there was this progamming language too ... forgot name
08:19<chillcore>basically you assigned other names to keywords
08:20<Eddi|zuHause>exchanging keywords is the easiest part of compiler construction :p
08:21<Eddi|zuHause>for my professor's professor, they once translated a language to swabian :p
08:21<chillcore>that must have been fun to work with
08:22<chillcore>I once toyed with the idea of making the pc understand spoken language to program with but quickly abandoned
08:23<chillcore>it was still the very early stages of voice recognition
08:23<chillcore>that and well you need to pronounce it everytime in exactly in the same fashion
08:24<Eddi|zuHause>spoken language has two main problems: 1) it's extremely bloated, lots of redundancies, 2) it's often ambiguous
08:24<chillcore>hmm going back and adding stuffs to sentences ... bad while chatting
08:25<Eddi|zuHause>loads of language designers fell into the trap of "if making the language more natural, programming gets easier"
08:25<chillcore>also correcting things ... no idea how to do that sanely
08:25<Eddi|zuHause>just look at COBOL
08:25<chillcore>I read abook about it never used it so ...
08:26<chillcore>my brother does COBOL
08:26<chillcore>unless they changed at Colruyt
08:27-!-tokai|noir [] has joined #openttd
08:27-!-mode/#openttd [+v tokai|noir] by ChanServ
08:27<chillcore>we don't talk much about work in the family ... just how's it going or found a new job yet?
08:28<chillcore>any experts on Firefox here?
08:29<chillcore>I want to make a public PC safe for multiple users somehow
08:29<chillcore>in and idiot proof way
08:29<chillcore>prob is that they keep changing stuffs and enabling things they should not
08:30<chillcore>so that the next user has no access to the previous his data/cookies
08:30<chillcore>prob is where it is at they are not aware and they do not reboot browser nor pc between session
08:30<chillcore>o/ planetmaker
08:31<chillcore>I huess I will need to educate them ... not hte users but the place where it is
08:32<chillcore>many many many public places belonging to the same institution
08:32<chillcore>all over the country
08:32<chillcore>wait I found a job ...
08:32<chillcore>now convince them to pay me ...
08:33<chillcore>the last time I checked they had a virus
08:33<chillcore>I delete cookies and this one site puts it right back
08:33<chillcore>I disable cokies and it keeps doing it
08:33<@Alberth>use a linux or a windows system with different user accounts
08:34<@Alberth>hi hi planetmaker
08:34-!-tokai|mdlx [] has quit [Ping timeout: 480 seconds]
08:34<chillcore>that is exactly the prob Alberth
08:35<chillcore>on boot it resets to satndard setting and ask your name ... you enter whatever so they can monitor you somewhat
08:35<chillcore>but between users they do not boot
08:35<chillcore>granted they have no access to USB or CD but who needs that while online
08:36<@Alberth>why would need to boot? just logout
08:36<chillcore>hmm ... forced logout
08:36<chillcore>that is an idea ...
08:37<chillcore>and forbid FB and the likes ...
08:37<andythenorth>Alberth: so which is more readable? o_O Guido thinks the new one, approximately
08:38<chillcore>ye I'll see if I can make them do that ... sessions of 15 minutes or so and it logs out automatically
08:39<chillcore>these PCs are meant for people wh have no internet at home and can not afford it
08:39<Eddi|zuHause>andythenorth: depends on how fluent you are in functional programming
08:39<andythenorth>no Lisp, no Haskell
08:39<andythenorth>no Erlang
08:39<andythenorth>no Scheme
08:39<chillcore>anyhoo thx for the hint Alberth, I'll fist have to convince them there is a serious privacy issue at hand
08:40<chillcore>then take it from there
08:40<Eddi|zuHause>reduce is often stretching it a bit
08:40<andythenorth>I read a long essay the other day that OO, imperative programming and iterators are killing us
08:40-!-Myhorta [] has joined #openttd
08:40<andythenorth>and that we should embrace functional programming and immutable state
08:40<Eddi|zuHause>loads of things are killing us...
08:41<andythenorth>it was all very appealing until I tried to understand the examples
08:41<chillcore>The biggest prob is that when you speak about that they look at you like you are some kind of madman alberth
08:41<chillcore>even highly educated peeps
08:42<@Alberth>very few people understand it :(
08:42-!-quorzom [] has joined #openttd
08:42<chillcore>when I see a PC with the password stickered to it my hair in the back of my neck stannds up
08:42<chillcore>sadly enough
08:42-!-TELK [~oftc-webi@] has joined #openttd
08:42<chillcore>the other day I said something about that and the reaction was ... oh but that is not my pc
08:42<chillcore>still it contains very sensitive data
08:43<TELK>Hello, is there any way to show non-english in dedicated server command(console) window?
08:43<chillcore>social security , medical , whatever you can think of
08:43<Eddi|zuHause>TELK: no
08:43<@Alberth>quite chillcore
08:43<chillcore>those are not the public pcs bythe way ... thank god
08:43<TELK>oh, I see you chillcre :)
08:44<chillcore>hello TELK, have we met before?
08:44<TELK>many korean uses your branches
08:44<chillcore>your nick does not ring a bell
08:44<TELK>You are famous in korean ottd users
08:44<chillcore>ah like that ...
08:45*chillcore checks bank account
08:45<chillcore>nope empty
08:45<@Alberth>andythenorth: made more new thingies
08:45<chillcore>just kidding ;)
08:45<chillcore>but it is empty ...
08:46<chillcore>I am happy to hear that I reach that far ..; and famous no less
08:46*chillcore blushes
08:46<andythenorth>Alberth: no set? o_O
08:46<@Alberth>hmm, good point
08:46<TELK>It was very nice to meet you, chillcore :)
08:46<chillcore>tell em I said hello ;) TELK
08:47<@Alberth> set-ted
08:47<TELK>byebye and thanks for answer Eddi
08:47-!-TELK [~oftc-webi@] has quit [Quit: Page closed]
08:49<andythenorth>puts out a list of tuples now, instead of a list of ints ;)
08:49<andythenorth>the union() fixes that
08:50<chillcore>that was ... unexpected
08:50<@Alberth>you just need all years, no matter if entry or exit date?
08:50<Eddi|zuHause>it's the internet, what can you really expect?
08:51<Eddi|zuHause>maybe that's just the korean way of saying "thanks for your patchpack, lots of people i know use it"
08:52<chillcore>ye maybe ... very kind though
08:53<chillcore>ego goes throug the roof ... feet still touching ground ? check
08:54<andythenorth>Alberth: yes, it ends up generating simple lists like
08:54<andythenorth>[0, 2010, 5000001]
08:54<Eddi|zuHause>hm, i've got this alarm clock
08:54<frosch123>chillcore: TELK runs the korean forums
08:54<Eddi|zuHause>that goes "miau" at completely random intervals
08:55<chillcore>ok Alberth ..;. equivalent to tt-forums fame, not national tv fame :P
08:55<chillcore>I'd better do some coding
08:56<chillcore>else not much is going to happen today ..
08:57<chillcore>must be fun Eddi ... my little buddy left me a portal gun in minecraft ... I thought I was alone
08:58<chillcore>then this thing starts talking and I almost peed myself
08:58<chillcore>sound was way up
08:58<@Alberth>andythenorth: this seems the simplest solution
08:59<andythenorth>because union() is infrequently used and has to be looked up, and has to be fed a list comprehension? o_O
08:59<@Alberth>no idea how to do what you want in a single line, although it sounds like something you might want to do
09:00<andythenorth>this isn’t a place for golf
09:00<chillcore>oh tgen bugreports coming in ... yay
09:00<@Alberth>hmm, update, probably
09:00<@Alberth>ah, yes years.update(variant.intro_date, variant.end_date)
09:01<@Alberth>x = x.union(y) <--> x.update(y)
09:03-!-Celestar [] has joined #openttd
09:04<andythenorth>update does seem better here
09:04<andythenorth>no new set returned
09:07<andythenorth> years.update((variant.intro_date, variant.end_date))
09:08<@Alberth>ah, no multiple arguments I guess
09:09<andythenorth>seems to accept an iterable as first arg
09:09<andythenorth>anyway, works thanks
09:11-!-samu [] has joined #openttd
09:22<andythenorth>pyflakes hates pixa
09:23<andythenorth>from pixa import *
09:23<andythenorth>is verboten
09:23<chillcore>darn this is going to be one of them hard to fix pacthes ... can not reproduce
09:23<chillcore>either windos or newgrf related
09:24-!-jjavaholic [] has joined #openttd
09:25<andythenorth>I have a module, with no test cases
09:25<chillcore>could be newgrf as I have not yet adjusted the API to the new setting .. which is the same but not
09:25<andythenorth>and to make it valid python 3, I think I have to break the API
09:26<chillcore>hehe andy ... you fix mine and I fix yours? ... but honestly I stink at newgrf related stuffs
09:26<andythenorth>this is pure python
09:26<chillcore>then I smeel even worse ;)
09:27<chillcore>s smeel/smell
09:27<andythenorth>I can’t face it right now :P
09:29<chillcore>I know the feeling andy
09:31<chillcore>question ... when changing API widgets but not the setting behind it, except for adding custom ... Do I need to adjust something else?
09:31-!-jjavaholic [] has left #openttd [Leaving]
09:31<frosch123>gs/ai api?
09:32<chillcore>smoothness puldown
09:32<frosch123>there is a script to generate those files
09:32<chillcore>ok ...
09:33<frosch123> and
09:33<chillcore>and I execute them frm console the same location as compiling yes?
09:33<frosch123>there is no compatbiilty layer in the script widget api, it is supposed to only work as long as windows do not change
09:34<frosch123>since there is no way to make it compatible anyway
09:34<frosch123>i think you run the sripts from the folder they are in
09:34<chillcore>besides changing the widget name it moved gui
09:35<chillcore>I'll give it a go ... I have my backup on forum
09:38<chillcore> sounds like the right one
09:39<chillcore>yep that was it ... thank you frosch ... I would never have found that
09:40<chillcore>but but it seems you forgot to do it yourself in trunk ...
09:40<chillcore>I see WID_GO_GUI_ZOOM_DROPDOWN
09:41<chillcore>^^^ inside my newly created patch
09:42<chillcore>or maybe it is not needed in trunk ... i dunno
09:43<chillcore>you == someone or the compiler
09:46<chillcore>either way I am glad I did not have to type all that manually
09:48<frosch123>likely it got forgotten in trunk :)
09:49<frosch123>we cannot run it automatically on build, due to inferior operating systems
09:50<chillcore>I sees
09:51<chillcore>I left it as a seperate patch so it is just a matter of deleting patch and running script again
09:53<samu>need opinion
09:54<chillcore>it is too cold for my taste <- opinion
09:55<chillcore>context samu?
09:56<samu>ah, right
09:56<@Alberth>except for the block at the lower left, it doesn't make much sense to have such large tracks
09:56<samu>costs for water
09:56<@DorpsGek>Commit by frosch :: r27152 /trunk (3 files in 3 dirs) (2015-02-15 14:56:21 UTC)
09:56<@DorpsGek>-Fix: Generated files.
09:56<samu>all others are default prices, canal prices are being edited
10:03<chillcore>thank you frosch for the quick fix ... again ;)
10:03<samu>I was thinking
10:03<samu>raise maintenance costs for canal
10:04<samu>but lower cost for building it
10:04<samu>how much lower? that is the question
10:06<samu>build price is at 1/8 of default costs
10:07-!-liq4 [] has quit []
10:07<samu>1/1 = 5000, 1/2 = 2500, 1/4 = 1250, 1/8 = 625
10:08<chillcore>hmm ... play test samu and see if you can survive early games at all
10:08<chillcore>or make it configurable
10:08<@Alberth>it is configurable with a basecost newgrf
10:09<chillcore>it should be yes ...
10:09<samu>im using the basecost
10:26<samu>gonna try super imba godly hard settings
10:29<chillcore>k. new version out the door ... where was I ... dissapearing tilebar when dragged while shaded.
10:30<andythenorth>why am I converting to python 3 anyway? o_O
10:30<andythenorth>faster nmlc
10:31-!-Progman [] has joined #openttd
10:31<chillcore>only my gui so that is good
10:33<samu>lol im dead already
10:37<samu>woah, can only build 3 ships with a max loan of 100k
10:38<andythenorth>MP=1 ?
10:38<andythenorth>better, more terse?
10:39<andythenorth>MP=16, and pass that to the number of pool threads?
10:39*andythenorth trying to remove the gak from the developer interface
10:43<andythenorth>this is a python MP pool, *not* the GNU make jobs pool
10:45*chillcore sings "Another bug bites the dust tjsaaaah dun dun dun"
10:46<chillcore>if (this->IsShaded()) return; where did I steal that and why ...
10:46<chillcore>beats me
10:53<samu>maintenance costs need to be higher, but not that much higher after all
10:53<samu>about double or 4x the current
10:54<samu>else ship running costs + construction cost + property maintenance will be higher than ship income
10:55<samu>+ loan interest also
10:55<samu>forgot about that
10:59<Eddi|zuHause>andythenorth: maybe pass make's -j parameter onto python?
10:59<andythenorth>it’s a good suggestion
10:59<Eddi|zuHause>not sure if that can be done
11:02-!-ard_ [~oftc-webi@] has joined #openttd
11:02<ard_>Can some1 help me?
11:03<Eddi|zuHause>not like this.
11:04-!-LadyHawk- [] has joined #openttd
11:04<ard_>Can you help me telling me how to play with AI??
11:04<Eddi|zuHause>what have you tried so far?
11:05*andythenorth wonders if 1 worker thread is too conservative by default :P
11:05<andythenorth>16 is rude
11:07<andythenorth>also COMPILE_FASTER (unrelated to use of worker thread pool) is a stupid arg name
11:08<andythenorth>it strictly enables partial compiles by only recompiling changed vehicles
11:08<ard_>no cab and much more AI and does nothing
11:08*andythenorth is unfamiliar with conventional interfaces to this stuff
11:09-!-LadyHawk [] has quit [Ping timeout: 480 seconds]
11:09-!-LadyHawk- is now known as LadyHawk
11:10<ard_>Some1 tell how to play with AI and works good and do something
11:12<lobster>well, to be fairly honest people will need some more info first
11:12<lobster>what version are you running? which AI are you trying to get working? did you follow the installation guide?
11:12<lobster>and does it give you some sort of error?
11:12<lobster>that kinda stuff
11:15<ard_>and i heave 1.4.4 version
11:18<@planetmaker>ard_: you always get more info than 'crash'. And 'AI'
11:18<ard_>i heave donwload 6 AI and no1 works
11:18<@planetmaker>what crashes? OpenTTD or the AIs?
11:20<@planetmaker>did you use ingame content download to obtain them?
11:20<@planetmaker>and which?
11:24<chillcore>one more for the gui gods ... alignment in _nested_create_scenario_widgets[]
11:28<@Alberth>ard_: you can have a look in the thread of an ai, to check if it is reported and/or fixed already
11:29<@Alberth> <-- has links
11:30-!-Myhorta [] has quit [Ping timeout: 480 seconds]
11:39<@Alberth>haha chillcore :)
11:42<chillcore>? Alberth
11:43<chillcore>sorry was absent for a sec
11:43-!-ard_ [~oftc-webi@] has quit [Quit: Page closed]
11:43<chillcore>s absent/not here <- spelling :P
11:44<@Alberth>your paste with the looooong line
11:44<chillcore>ah ok
11:44<chillcore>happens ... to me too
11:45<chillcore>there are guis where it is aligned like that and most likely got copied from there
11:45<@Alberth>one of the few cases where non-wrapping editors are useful
11:45-!-smoke_fumus [~smoke_fum@] has joined #openttd
11:46<chillcore>I should fix em all when encounterd,
11:46<@Alberth>I mostly get annoyed with any editor that doesn't act like vi/vim :)
11:47<@Alberth>I guess I am too much attached to it :)
11:47<chillcore>yeah usually I go for notepad++ but this gedit is pretty much the same
11:48<@Alberth>in the KDE world, there is kate and kwrite
11:49<chillcore>I have too many choice too mention ... kinda anoying that different types of files open different editors
11:49<chillcore>had to change with "open with ..."
11:49<@Alberth>change your preferences :)
11:50<chillcore>for .ini, .h and cpp ... not that they are bad but i prefer in one gui
11:50<chillcore>I did
11:51<@Alberth>yep, I even have different configurations for files in different projects :)
11:51*LordAro regularly tries to :wq in gedit
11:51<LordAro>or notepad++
11:51<@Alberth>not ESC? :)
11:52<chillcore>ye that is anoying ... one of them in settings.ini two of theses ' makes half the page be pink text
11:52<chillcore>' thes in the comments that is
11:52<@Alberth>:set color=none :)
11:53<chillcore>syntax highlighting is usefull ... fixed it in the meantime
11:54<chillcore>it helps to see structure in the files
11:54<samu>canal maintenance: 4x default - looks fine, very tight for a first transportation route in very hard settings
11:54<samu>canal cost: 1/8 default
11:55<@Alberth>you don't want to start with canal building as first route
11:55<chillcore>but if I see more of them lines should they all be fixed? there are more and I'll do them when encounterd close by... Alberth
11:55<@Alberth>chillcore: hmm, tough one
11:56<samu>i'm comparing the same route vs rail and vs road
11:56<samu>construction costs were similar to rail, road was cheapest
11:57<samu>const costs: 49650 vs 45288 vs 23856
11:58<samu>new vehicles: 36562 vs 32339 vs 73084
11:58<samu>1 ship, 1 train, 11 trucks
12:00<samu>running costs: 4320 vs 3249 vs 7062
12:00<samu>maintenance costs: 2730 vs 2189 vs 419
12:01<chillcore>Alberth: ye ... then again you did break all of our patches too long time ago ... every single on in existance
12:01<chillcore>we can tackle this later maybe
12:01<@Alberth>just collect them, and we'll see
12:02<@Alberth>work in the newworld gen window isn't done yet :)
12:02<samu>income on the first year: 16934 vs 20463 vs 30431
12:03<chillcore>I got 5-ish patches/fixxes left to do
12:03<chillcore>but for everyone I solve 2 emerge so make that 20-ish fixes
12:04<samu>profit graph is lowest for ship, but considering the first route was all canal only
12:04<samu>I think i am satisfied
12:04<chillcore>there are more but those are adons/improvement in functionality
12:04<chillcore>eg sliders
12:05<chillcore>useful but not needed perse
12:06<samu>it can profit :) that's all that I cared to do
12:06<@Alberth>samu: costs should be more varied imho
12:06<samu>tell me
12:06<@Alberth>a player should make a smart choice how to transport cargo, if you make it all the same, you eliminate that need
12:08<@Alberth>that's also why there are more than one train, for example, you should think which one to use at what place
12:08<@Alberth>otherwise, you can have just one train for everything
12:08<samu>this train is size 6
12:08<samu>forgot to mention
12:11<samu>for this test I was only trying to see if it was even possible to start a all-canal route with only 100k loan
12:12<samu>it is, but it's not really getting you loads of cash
12:12-!-gk [] has joined #openttd
12:13<Eddi|zuHause>samu: you can try to make ships more profitable by increasing the cargo aging interval
12:14<Eddi|zuHause>if you double the interval, a ship will make as much money as a train twice the speed
12:14<Eddi|zuHause>if you feel fancy, you can do this only for "non-perishable" cargos (coal, steel, ...)
12:16<samu>ships last 30 years
12:16<samu>i'm okay that they profit half than a train
12:17<samu>it's not exactly how much they profit that worries me, but the building / maintenance costs in relation to profit
12:19<samu>year 1951 profit: 6468 vs 14495 vs 21677
12:20<samu>ship lowest, trucks highest
12:21<@Alberth>ships should be high volume transport imho
12:21<@Alberth>insane high construction costs, and small profits, so you need all those years to get your investment back
12:22<samu>well, yeah, as you said, not best to start with ships and canal
12:22<samu>i could start with ships without needing to build any canal tile
12:23<samu>but i was doing the extreme here
12:23<samu>see if it was possible
12:24<V453000>cause openttd is only about making money :)
12:25<roidal_>maybe someone can explain me the operating profit graph?
12:27<roidal_>even if i have a year with negativ income the graph stays above 0
12:29<roidal_>negativ win*
12:29<roidal_>or loss
12:32-!-Biolunar_ [] has quit [Quit: leaving]
12:32<samu>trying to make my 2nd ship
12:33<samu>going the same route
12:33<samu>i'm 2k short
12:36-!-roidal_ is now known as roidal
12:38<roidal>i think thats a bug
12:40<samu>2 ships, 3 trains, 23 road vehicles, though I build 24, 1 died to a ufo colision
12:40<samu>gonna wait for year 1955
12:42<samu>hmm 23 trucks is excessive for this route, must expand stations
12:44<Taede>roidal, which graph are you looking at? if it is income, it can only be => 0, as that counts only the income vehicles get. expenses (running cost, maintenance cost etc) do not count for that graph (iirc)
12:44<Taede>i think you want the 'operating profit' graph
12:45<roidal>Taede: yes, i mean the 'operating profit'
12:45<roidal>but it seems to be broken in 1.4.4.
12:45<roidal>will make one
12:45<@DorpsGek>Commit by translators :: r27153 /trunk/src/lang (3 files) (2015-02-15 17:45:47 UTC)
12:45<@DorpsGek>-Update from WebTranslator v3.0:
12:45<@DorpsGek>german - 3 changes by mbender
12:45<@DorpsGek>korean - 2 changes by telk5093
12:46<@DorpsGek>malay - 12 changes by im54
12:48<roidal>Taede: where i should upload the screenshot?
12:51<Taede>tinypic or similar service
12:51<Taede>anything thatll let the public see your image
12:52<Taede>and/or upload the savegame to dropbox or similar
12:53<samu>albert is right
12:53<roidal>year 2032
12:53<samu>trains and trucks have already achieved saturation point
12:54<samu>ships have not
12:55<andythenorth>is it acceptable to shorten ‘incremental’ to ‘inc’ as an arg?
12:55<andythenorth>or does it conflate with other things?
12:55<Taede>at a guess, i'd say construction costs (and new vehicle costs) are not used in those graphs at all
12:55<Taede>probably intended behaviour
12:56<@Alberth>andythenorth: seems ok to me, perhaps is_inc (if it is a boolean)
12:56<roidal>Taede: what costs are used for this graph?
12:56<roidal>only running costs?
12:56<Taede>income, running costs, maintenance, interest (in other words, regular outgoings and incoming)
12:57<roidal>this must be "new"?
12:57<Taede>excluded would be one-offs, eg buying vehicles, terraforming, construction
12:57<roidal>because i remember on a graph
12:58<roidal>where this was included?
12:58<roidal>...long time ago :D
12:59*andythenorth names it to the terrible ‘isinc’ because _ is hard to type fast
12:59<andythenorth>or ic
13:00<@Alberth>search&replace :p
13:00<@Alberth>or use an editor with completion
13:01<chillcore><Alberth> search&replace :p <- and coninue past end of file to make sure to catch 'em all
13:02<@Alberth>ugh, a so terribly broken concept
13:02<chillcore>hmm never had probs with it ...
13:02<@Alberth>I have not yet found a use for that behavior, why would you ever not wrap search?
13:03<chillcore>I am rarely at beginning of file ...
13:03<chillcore>compiler will complain though ... just a bit annoying if i patch 2/10 you forgat a few
13:03<chillcore>hmm spelling ...
13:03<@Alberth>note, that I am asking for a "not wrapping" use case :)
13:04<@Alberth>I can't see a use for it
13:04<chillcore>then no ... no use for that
13:05<chillcore>my bad
13:05<@Alberth>which then leads to the problem why an editor has it in the first place :p
13:05<@Alberth>np, I didn't formulate it very clearly
13:05<chillcore>story of my lafe Alberth story of my life
13:06<chillcore>hmm typ,ing while eating ... even badder
13:06<@Alberth>yep, trying that here too :)
13:07<Eddi|zuHause>roidal: i think you're imagining things
13:07<chillcore>bon apetit
13:07<Eddi|zuHause>or maybe you mix it up with another game
13:08<roidal>maybe youre shure
13:09<frosch123>Alberth: when replace "A" with "AB" you certainly don't want to wrap infinitely :p
13:09-!-zeknurn [] has quit [Remote host closed the connection]
13:10-!-zeknurn [] has joined #openttd
13:10<chillcore>hmm we have a winner ;)
13:11<samu>oh alberth looks like... 2 ships on the same route pays off
13:11<@Alberth>frosch123: so you disable wrapping, have to go to the start of the file before you can replace?
13:11<frosch123>well, usualy editors are smart enough to not iterate multiple times over the file when replacing all :p
13:11<@Alberth>why not try all lines exactly once instead, no matter where you are?
13:11<samu>getting the 2nd ship is however, the boring part
13:12<frosch123>if in doubt, use sed :)
13:12<andythenorth>use better ships
13:12<samu>there's only 1 cargo ship
13:13<@Alberth>I rarely doubt in such cases :p
13:13<samu>2 ships are profiting better than 3 trains
13:13<andythenorth>anybody want to sanitise pixa?
13:13<Eddi|zuHause>"continue replacing from beginning of file" usually stops at the point where you started, it doesn't loop
13:13<andythenorth>and make it python 3 compatible?
13:13*andythenorth guesses the answer
13:13<samu>then again, the ship route isn't saturated yet
13:13<samu>the train route is
13:13<@Alberth>samu: ships routes are never satured
13:13<samu>it was faster to get 3 trains
13:13<Eddi|zuHause>at least, i never have seen an editor that does...
13:14<@Alberth>andythenorth: I could have a go at it, but not today
13:14<samu>i'm transporting coal
13:14<andythenorth>Alberth: another day maybe :)
13:14<andythenorth>it’s mostly stupid illegal imports I think
13:15<@Alberth>import * seems to work less good in python 3
13:15<andythenorth>or not at all :)
13:15<andythenorth>also there’s no tests or reference case
13:15<andythenorth>so testing pixa is a chore
13:15<andythenorth>but I could write that
13:16<@Alberth>you're not even using pixa?
13:16<andythenorth>I use it in a couple of places
13:16<Eddi|zuHause>i use it, too
13:16<andythenorth>it’s really convenient
13:16<andythenorth>flipping 2CC colours
13:16<Eddi|zuHause>but CETS is not ported to python3 either
13:16<@Alberth>you have at least 1 other user :)
13:16<andythenorth>recolouring cargos
13:16<andythenorth>bloody users :)
13:17<andythenorth>generating spritesheets programmatically
13:17<Eddi|zuHause>i also know virtually nothing about what changed between 2 and 3
13:17<Eddi|zuHause>i wanted to incorporate pixeltool, but never got anywhere with it
13:18<Eddi|zuHause>i think i stopped at finding a command line interpreter for javascript
13:18<@Alberth>oh, the zephyris thing?
13:18<andythenorth>we should convert the game to voxels :P
13:21<andythenorth>eh, apart from pixa, pyflakes is now happy with Iron Horse
13:21<andythenorth>small victories :P
13:22<andythenorth>are conditional imports a stupid idea?
13:22<Eddi|zuHause>i have the pixa import in a try/catch
13:23<Eddi|zuHause>for when it is not installed, i provide stub functions
13:23<andythenorth>Iron Horse being split into rosters, it would sometimes be useful to only compile one roster
13:24<@Alberth>another reason why imports should not have side effects :)
13:24<samu>gonna try 2x maintenance costs
13:25<andythenorth>imports now mostly don’t have side effects :)
13:25<andythenorth>there are still a few
13:25<andythenorth>but my code is now littered with calls to main()
13:25<Eddi|zuHause>then importing "useless" modules shouldn't hurt
13:25<Eddi|zuHause>put the control flow around the calling functions, not the imports
13:25<@Alberth>compiled byte code is cached anyway
13:27<andythenorth>ok ta
13:41<andythenorth>is flat better than nested?
13:42<frosch123>green is certainly better than yellow
13:43<andythenorth>that’s what I think too
13:43<andythenorth>except on Tuesdays
13:44<chillcore>lol andy
13:48<chillcore>hmm why do I always hesitate to do something when it is two or three lines and when it is a few hundred I don't ...
13:48<chillcore>starnge me
14:00<Eddi|zuHause><frosch123> green is certainly better than yellow <-- so, you are green, but nobody here is yellow.
14:00<Eddi|zuHause>well, maybe samu gets closest to yellow :p
14:07<samu>Build Canal: 1/8, Remove Canal: undecided, Build Aqueduct: 1/8, Remove Aqueduct: undecided, Build Lock: 1/4, Remove Lock: unchanged, Canal Maintenance: x2 or x4 (must decide)
14:07-!-Romazoon [] has quit [Remote host closed the connection]
14:07<chillcore>remove canal 2 * build at least
14:07<samu>maintenance costs quickly go up per each aqueduct tile
14:08<chillcore>cleanup you dirty ...
14:08<Eddi|zuHause>remove should certainly be expensive
14:08<chillcore>but yeah something like that after buildingless cleanup
14:08<samu>maintenance per canal tile: x1, per aqueduct tile: x4, per lock (includes 2 canals): x8, per lock (without canals): x6
14:09<samu>if the mainentance is to be x4, all those values will go up x4
14:09<Eddi|zuHause>you might be able to set how many trackbits an aqueduct is counted as, but not sure
14:10<chillcore>that is why you test and play samu, you do not borrow money and test like a new player would play with it
14:10<chillcore>if you intend to make your modifications available that is
14:11<chillcore>I do that too whenever I tune
14:11<samu>there is some advantages for aqueducts
14:11<chillcore>ofcourse I had the luxury of having a ton of testers
14:11<samu>can build stuff below them
14:11<chillcore>so it took me less time
14:12<chillcore>and later you make configs and do a hard setting and an easy one
14:13<chillcore>when done with that custom or move on
14:13<chillcore>to a new challange
14:14<Eddi|zuHause>auf zum atem!
14:15<samu>i know, seems like aqueducts as first route is a no
14:15<Eddi|zuHause>'s one of those things i have no idea how the english original joke might have been)
14:16<samu>if maintenance is at x4, then it's a no
14:16<samu>at x2, i'm unsure, must try
14:16<@Rubidium>Eddi|zuHause: lmgtfy?
14:16<Eddi|zuHause>Rubidium: probably not.
14:17<samu>locks seem fine
14:17<samu>i had 3 locks in the first test
14:17<Eddi|zuHause>samu: keep in mind that infrastructure costs go up the more you have of that type of infrastructure
14:17<@Rubidium> vs
14:18<samu>there's only "canal" maintenance, everything goes to that
14:20<Eddi|zuHause>Rubidium: maybe that joke actually wins something by the nonsensicalness of the german translation.
14:21<chillcore>I am watching in german
14:23<@Rubidium>yeah, translating jokes is a major PITA
14:26<@Rubidium>although hearing dubbed things when you are very familiar with the original voices and sounds is really annoying as well
14:26*andythenorth wonders if it’s worth trying to do a two factor sort in python
14:26<andythenorth>or tmwftlb
14:27<@Rubidium>what's a two factor sort?
14:27-!-roidal [] has quit [Quit: WeeChat 1.0.1]
14:28<andythenorth>e.g, sort lexically by name, then within ‘a’, sort by intro date
14:28<andythenorth>(made up case, but similar)
14:28<andythenorth>bad example
14:28<@Alberth>make a tuple of values to sort one (name, intro_date, full_data)
14:28<@Rubidium>shouldn't you always do that if you want stable sort results?
14:30<samu>aqueduct costs are so difficult
14:30<samu>to balance
14:31<chillcore>*chillcore still ponders about that max heighlevel setting ...
14:32<chillcore>snowline goes up to heighest level right ...
14:33<chillcore>so why do we need an upperlimit for that ?
14:33<chillcore>255 done
14:33<chillcore>it is the bottom that varies
14:33<chillcore>untill now always below 15
14:34<chillcore>so igf we raise the top levels become snow
14:34<chillcore>again done
14:34<chillcore>what is the diff with now
14:34<chillcore>is say we lower to 3
14:34<chillcore>the top levell I mean
14:34<chillcore>we hate 1 lien of snow
14:34<chillcore>amse if you come dowwn from 255 to 1
14:35<Eddi|zuHause><Rubidium> although hearing dubbed things when you are very familiar with the original voices and sounds is really annoying as well <-- yes, as soon as you compare the translation with the original, the translation basically lost.
14:35<chillcore>no snow
14:35<Eddi|zuHause>but a good translation makes you laugh anyway
14:35<chillcore>I really misss the point
14:35<chillcore>NewGRF or not
14:35<andythenorth>seems complex sorts are just handled in python by sorted()
14:35<frosch123>chillcore: just grep the source for places where max_heightlevel is used
14:35<andythenorth>htf does that work :|
14:35*andythenorth mind blown
14:35<chillcore>grep ...
14:36<andythenorth>surely if I sort once, then sort again, the order of the first sort is destroyed
14:36<@Alberth>andythenorth: ?
14:36<Eddi|zuHause>andythenorth: supply it a cmp function
14:36<samu>lol, i made the route, now i don't have money for the ship
14:36<Eddi|zuHause>andythenorth: CETS does this
14:37<Eddi|zuHause>andythenorth: the function is probably in scripts/
14:37<Eddi|zuHause>and used in scripts/
14:37<andythenorth>apparently it magically just works
14:38<chillcore>thx for the link
14:38<andythenorth>according to those docs
14:38<andythenorth>I haven’t tested it, but that example would seem to violate expected behaviour
14:39<chillcore>I can do that on a folder directly ?
14:39<samu>i have a problem, if i lower the cost factor for the aqueduct one more half, then everything else must also be lowered, but then things won't start to make much sense
14:39<samu>cost wise
14:39<frosch123>chillcore: does your editor not allow searching in all files?
14:41<samu>lock pricing doesn't make much sense already
14:41<samu>thx to 2 canal tiles being in the cost
14:41<@Alberth>andythenorth: it looks like you sort on the 2nd key first
14:42<samu>at such prices, making ships would also be too good for what I had in mind
14:42<samu>that is, if no aqueduct is used
14:43<andythenorth>maybe it depends on that example using triples
14:43<andythenorth>I have a list of objects I wanted to sort
14:44<andythenorth>I might use a dict of lists instead, grouping on the first sort
14:44-!-Romazoon [] has joined #openttd
14:44<chillcore>I was momentarily busy ... going to do it in a bit
14:45<chillcore>never tested really
14:45<Eddi|zuHause>maybe i'm the only one remembering this, but there was a grep.exe in novell dos 7
14:45<samu>aqueduct at 1/16 will make lock pricing make no sense
14:45<chillcore>was just trying to understand why snowline could be a reason
14:45<Eddi|zuHause>i think
14:45<chillcore>I'll will happily accept another reason np
14:46<Eddi|zuHause>man that was a long time ago
14:46<samu>a 6 sized aqueduct would be cheaper, and 2 locks taking 6 tiles, and with the advantage of building routes underneath
14:47<chillcore>its doing something ...
14:47<samu>typoe, would be cheaper than*
14:47<chillcore>should it take long?
14:48<Eddi|zuHause>samu: how is that a problem?
14:48<samu>2 locks in a row, will block
14:48<samu>no route underneath
14:49<Eddi|zuHause>yes. so don't use locks in that situation.
14:49<@Rubidium>use the chunnel patch... problem solved ;)
14:49<samu>they should be cheaper
14:50<samu>precisely because of that disadvantage
14:50<Eddi|zuHause>locks have their strength in moving ships up multiple times
14:50<@peter1138>Locks never block...
14:51<samu>i mean 2 locks together, one going down, one going up versus 1 aqueduct
14:51-!-DanMacK [] has joined #openttd
14:51<Eddi|zuHause>also, you can build bridges over the canal, instead of going under it..
14:51<Eddi|zuHause>samu: i think your case is bogus.
14:51<samu>the 2 locks structure will create a barrier
14:51<samu>no bridge possible
14:51<@Alberth>ho, DanMacK
14:52<samu>but with aqueduct, there's 2 free tiles underneath it, let's me have other routes under it
14:52<Eddi|zuHause>samu: things just don't get cheaper, because they are stupid.
14:55<samu>i want my pricing structure to make sense :P
14:57<chillcore>hmm I need a million USB sticks ...
14:58<Eddi|zuHause>i currently have two.
15:00<chillcore>me 3
15:01<chillcore>one has linux installed for if/when OS should die
15:02<chillcore>could beused as portable pc hehe
15:02<Eddi|zuHause>in those cases i usually dig through the layers on my desk to find that knoppix cd i downloaded 5 years ago.
15:02<chillcore>just attach usb screen and go
15:02<chillcore>would need really small chip
15:02<chillcore>and power ofcourse
15:03<chillcore>maybe in screenie
15:03<Eddi|zuHause>i'm sure there are USB-stick-sized computers nowadays
15:03<chillcore>sure ..; google glass and smaller
15:04<Eddi|zuHause>it won't be very powerful, but these devices usually serve as thin client anyway, so the heavy stuff is on some "cloud" server somewhere
15:06<chillcore>it served me well last year when my win 8 died and the store was closed
15:06<chillcore>my stuff was not on there but I could use the laptop normally
15:06<chillcore>almost just a bit slow loading of progs
15:16-!-oskari89 [] has quit []
15:29<samu>there's a £43 construction cost difference, favoring aqueduct, but a £36 maintenance cost per year difference, favoring 2xlocks
15:31<samu>I guess I'll accept this compromise
15:31<chillcore>sorry eddi got distracted ...
15:31<chillcore>ramdrive for that usb?
15:33*chillcore needs more screenspace
15:34<chillcore>or rather connections to attach screens to
15:36<andythenorth>ternary operator is to weird to use for things like appending to lists
15:36<andythenorth>foo.append(bar) if True else None is weird
15:36<andythenorth>too *
15:41<@Alberth>yep, I don't use it at all
15:43<andythenorth>works well in templates and such
15:43<andythenorth>neater than my weird tuple trick
15:49-!-glx [] has joined #openttd
15:49-!-mode/#openttd [+v glx] by ChanServ
15:54<TrueBrain>talking about Python, here is a nice riddle
15:54<TrueBrain>a = {"a": 1}
15:54<TrueBrain>"a" in a == True
15:54<TrueBrain>what do you think the answer is, and if you would add (), where would they be? :D
15:55<chillcore>1 pair of ()?
15:55<frosch123>well, the answer is easy to guess from the fact that tb asks a question :)
15:56<frosch123>but to track it off-topic
15:56<frosch123>"== True" and "!= False" is silly
15:56<frosch123>in all languages except C
15:56<TrueBrain>well, the answer might surprise you in this case
15:56<TrueBrain>as that statement does not really what you expect
15:56<TrueBrain>like .. it really doesnt :)
15:57<frosch123>though C being the exception just shows to silly C is :p
15:57<TrueBrain>"a" is in a, so you would expect it to return True .. but it doesn't :P
15:57<TrueBrain>("a" in a) == True returns True
15:57<TrueBrain>"a" in (a == True) returns TypeError
15:57<TrueBrain>"a" in a == True returns False
15:57-!-Pol [] has quit [Remote host closed the connection]
15:57<TrueBrain>thought it would be nice to share :D
15:58<chillcore>never done python yet and it is ;)
15:58-!-HerzogDeXtEr1 [] has joined #openttd
15:59<frosch123>hmm, i would have guessed all of them, except the last one
15:59<andythenorth>is there some curve speed penalty property for trains?
15:59*andythenorth ponders dibbling narrow gauge up
15:59<frosch123>why is '"a" in a == True' False ?
16:00<TrueBrain>frosch123: now that is the question ;)
16:00<TrueBrain>not often you see in a language that adding () gives other results than without any :)
16:00<TrueBrain>if I would guess, I would say it does something like type("a" in a) == type(True)
16:00<TrueBrain>which is False
16:01*andythenorth always adds the brackets in (2 * 2) + 4
16:01<andythenorth>because reasons
16:01<TrueBrain>well, the above Python snippet suggests you would be more safe than I would be :P
16:02<TrueBrain>anyway, time for a good night sleep :)
16:02<@Alberth>perhaps the x <= y <= z syntax is interfering?
16:02-!-HerzogDeXtEr1 [] has quit []
16:03-!-HerzogDeXtEr [] has quit [Ping timeout: 480 seconds]
16:04<@Alberth>hmm, nice one to report :p
16:09-!-FLHerne [] has joined #openttd
16:10-!-HerzogDeXtEr [] has joined #openttd
16:13-!-frosch123 [] has quit [Quit: be yourself, except: if you have the opportunity to be a unicorn, then be a unicorn]
16:14<samu>looks like there is a bug in calculating lock price
16:15-!-sla_ro|master [slamaster@] has quit []
16:15<samu>it's cheaper to build it on it's own than building on a rivercliff
16:15<samu>shouldn't it be the opposite?
16:16<chillcore>bug bug ... it has a price yes... needs tuning
16:16<samu>it is adding 10k to the total
16:18<samu>don't add the 10k
16:18<samu>and it's fixed
16:19<@Alberth>don't bother about 10k, and it's a non-issue
16:19<chillcore> chillcore: does your editor not allow searching in all files?
16:19<@Alberth>in particular as you have a zillion, so 10k more or less makes no difference
16:19<chillcore>I was reading ab it ...
16:19<samu>omg alberth :O
16:19<chillcore>workbench allowzed clicking a revision and search that from gui
16:20<@Alberth>you're at linux right?
16:21<chillcore>linux mint debian based
16:21<@Alberth>open a terminal right next to the editor, install 'ack', done
16:21<chillcore>but not much installed and configured yet
16:21<samu>oh, it's 11250 in excess
16:22<chillcore>k alberth ...
16:22<@Alberth>hmm, debian gave it a weird name....
16:22<@Alberth>chillcore: ack-grep
16:22<samu>12304 - 1054 = 11250
16:23<samu>yep, for me this looks like bug
16:23<samu>that rivercliff tile isn't cleared
16:23<samu>when i destroy the lock, it's still there
16:23<samu>shouldn't account those 11250
16:23<@Alberth>it is, there were rocks and stuff, boats break when they hit rock
16:24<samu>rocks don't come back when i destroy lock
16:24<@Alberth>they were removed :)
16:25<@Alberth>chillcore: open terminal, "ack-grep <term>"
16:25<samu>i'm reporting it as bug
16:26<@Alberth>samu: it needs a stronger argument than "I think it's wrong"
16:26<samu>the rivertile is kept wen a lock is built
16:26<Eddi|zuHause>i always felt rocks were "wrong"
16:26<samu>and also when destroyed
16:27<chillcore>naw it works ...
16:27<Eddi|zuHause>they should maybe have a bit like desert, so it comes back when clear
16:27<chillcore>why is this not default?
16:27<chillcore>thx Alberth
16:28-!-jjavaholic [] has joined #openttd
16:28<@Alberth>it's mostly like recursive grep, but does some added magic, like skipping certain files
16:28<@Alberth>option -i does case-insensitive search, -l just dumps filenames
16:28<samu>there is no price for a cliffed canal
16:28<@Alberth>and I added an "ack" alias, as that's how it is called at my Fedora system
16:28<andythenorth>bollocks, /me used multiprocessing, and now has 2 problems
16:29<samu>there's only the price for the lock
16:29*andythenorth wanted to say the cliche
16:29<@Alberth>you succeeded :)
16:29<samu>the price can be based on the lock itself, not on the river
16:29<chillcore>ye If I ghad nknow I had to install it ...silly me
16:30<chillcore>but now I also know I can do the same in my gui
16:30<samu>it's wrong that when I build a lock on 3 dry tiles it is cheaper than building it on river/rivercliff/sea
16:30<chillcore>doh typing
16:30<chillcore>thank you
16:30<@Alberth>there are way too many packages to install all :)
16:31<@Alberth>so standard tactic if you don't have something, look for a package :p
16:31<chillcore>good tip
16:32<@Alberth>though I only install packages from trusted sites, like fedora
16:34<samu>I see your problem alberth
16:35<@Alberth>samu: nobody said you have to build over a river cliff, you can just build next to it, if you want
16:35<samu>that defies common sense
16:35<andythenorth>hmm stumped
16:36<@Rubidium>samu: it's cheaper to build a lock on a "dry" bit of land because otherwise you first need to drain the location where to build the lock *or* use expensive under water construction
16:37<andythenorth>getting different results when using multiprocessing
16:37<andythenorth>I’m relying on python object hashes
16:37-!-jjavaholic [] has quit [Remote host closed the connection]
16:37<andythenorth>maybe they’re not consistent under multiprocessing
16:37<samu>that logic isn't consistent
16:37<samu>if I remove a lock that was built on dry tiles, it won't cost 10k..
16:38-!-andythenorth [] has left #openttd []
16:39-!-Yotson [~Yotson@2001:980:6ac8:1:e1e3:89a3:4edd:e9c1] has quit [Quit: .]
16:40<samu>there could declived canal tiles
16:40-!-gelignite [] has quit [Quit:]
16:40<samu>force the lock to build that declived tile
16:41<samu>damn my english today is horrible
16:41<Eddi|zuHause><Alberth> perhaps the x <= y <= z syntax is interfering? <-- that would be my first thought, too. but that would mean it tests '("a" in a) and (a == True)', which would bring back the typeerror
16:42<samu>when building a canal tile on a river tile, it won't cost me 11250, just the cost of that canal tile, it ignores the river
16:42<@Alberth>Eddi|zuHause: yep, 'False' makes no sense at all, imho
16:42<samu>it should be similar for lock, thought instead of building 2 canal tiles, it builds 3
16:42<samu>and ignore the river
16:43<samu>ships are still unable to go up those cliffed canals on their own without a lock
16:43-!-supermop [] has joined #openttd
16:46<samu>while I'm at it, when destroying a canal tile, it shouldn't destroy the river tile if it was built on top of it
16:46<samu>i was once screwed because of this on a citibuilder game
16:47<samu>that river tile was part of someone else area
16:47<samu>and the script destroyed those tiles
16:50<samu>one other bug, though i don't consider it too critical, is upon drag&drop demolishing a large area when locks are in there
16:51<samu>it leaves 1 canal tile behind
16:51<samu>have to destroy a second time
16:54<Eddi|zuHause>samu: wrt the river, dynamite always clears the entire tile. it's like removing road stations this way also clears the road.
16:55<Eddi|zuHause>the other thing: are you sure the area was covered with the dynamite?
16:55<Eddi|zuHause>that may be an actual bug.
16:58<samu>when using dynamite on a lock that was built on river/rivercliff/sea, it won't clear the river/rivercliff/sea tiles - works as expected
16:59<samu>when using dynamite on a lock that was built on 3 dry tiles, via drag&drop, it leaves 1 canal tile at the upper level
17:00-!-DabuYu [DabuYu@] has joined #openttd
17:01<samu>without drag&drop, it leaves 2 canal tiles behind
17:04<samu>woah, discovered another drag&drop problem
17:05<samu>drag and drop over the 3 tiles that were built on top of river/rivercliff/sea, will clear the rivercliff+sea tiles
17:05<samu>only the river tile survives
17:05<samu>the one at the top
17:08<samu>i have a suggestion for the way locks are built, they could follow the same rule as docks
17:08<samu>it needs a water tile already in place
17:08<samu>locks should also do the same, they would require 3 water tiles already in place
17:09<samu>problem is that there is no canalcliff in the game
17:09<samu>just create this
17:10<Eddi|zuHause>docks only need one water tile in place
17:11<Eddi|zuHause>and the way docks work is ... suboptimal.
17:11<samu>hmm drag and drop clears water under the dock
17:11<samu>i didn't expect this
17:14<samu>which one is optimal then? lock or dock?
17:15<samu>do you want docks to come packaged with a water tile when built?
17:16-!-Quatroking [] has quit [Read error: Connection reset by peer]
17:16-!-Alberth [~alberth@2001:981:c6c5:1:be5f:f4ff:feac:e11] has left #openttd []
17:24<samu>does the game have a demolish table or something similar?
17:25<juzza1>what is demolish table?
17:25<samu>special rules for demolishing for each type of structure
17:26<Supercheese>There should be something like that
17:28<Supercheese>I presume you have the source at hand
17:28<samu>me, not looking at it, but I have it
17:29<Supercheese>Well, src/table/town_land.h has "rating decrease if removed"
17:29<Supercheese>among other things
17:31-!-DanMacK [] has quit [Ping timeout: 480 seconds]
17:32-!-Progman [] has quit [Remote host closed the connection]
17:34-!-Biolunar [] has joined #openttd
17:49<supermop>what do you guys think about showing purchase sprites from a lower angle
17:49<supermop>like, closer to a side elevation instead of overhead dimetric?
17:52<V453000>why not :)
17:52<V453000>I think especially purchase sprites have a lot of freedom that hasnt been used up yet
17:57<supermop>will it break too many things if tram is longer than normal while loading? visually only
17:57<supermop>i want to have coffee bins removed for loading but then where do they go?
17:58<supermop>is the vehicle info windo sprite same as purchase menu?
18:04<V453000>purchase menu is only purchase menu I think
18:04<V453000>vehicle info windows is just -- view of the vehicle
18:04<V453000>e.g. you could have purchase menu sprite of flatbed wagon for 1 thing
18:04<V453000>but in the vehicle info window it always shows the right cargo
18:05<supermop>hmm i messed up this alpha stuff
18:05<supermop>and the chunk of clear plastic is just blue
18:14<Eddi|zuHause>you can also have a separate sprite in list/details/details windows by checking var10 (extra_callback_info1)
18:14<Eddi|zuHause>last one should be /depot
18:15<Eddi|zuHause>the purchase list is special, because the vehicle is not built yet, so some variables cannot be accessed
18:15<Wolf01>'night all
18:15-!-Wolf01 [] has quit [Quit: Once again the world is quick to bury me.]
18:16-!-itsatacoshop247 [] has joined #openttd
18:16<Eddi|zuHause>supermop: there's nothing fundamentally wrong with longer sprites, but it may overlap with other vehicles
18:16<Eddi|zuHause>supermop: there may also be clipping errors
18:17<Eddi|zuHause>e.g. under bridges, and stuff
18:17<Eddi|zuHause>they míght be less prominent with standing vehicles
18:19-!-Speedy` [] has joined #openttd
18:19-!-Speedy` is now known as Speedy
18:27<supermop>is there an easy way in PS to take an area that is grey-ish tinted cyan and make it grey-ish tinted clear alpha?
18:29<V453000>I guess easiest is to use the magic wand
18:29<V453000>with the right settings
18:29<chillcore>night all
18:30-!-chillcore [~chillcore@] has quit [Quit: Only mortals are affected by fame and power.]
18:38<supermop>figured out a way in PS
18:38<supermop>but would be way easier if flamingo added alpha info to areas with transparent materials or had an alpha ground plane
18:39<supermop>which it does, just not in the 7 year old version i am using
18:41<V453000>flamingo is rendering engine for rhino?
18:42<samu>woah, this demolish table is complex after all, thx to lock
18:43-!-smoke_fumus|2 [~smoke_fum@] has joined #openttd
18:43-!-smoke_fumus [~smoke_fum@] has quit [Read error: Connection reset by peer]
18:44<samu>there's too many ways to build a lock
18:45<samu>on a coast, on a river cliff, on a cliff
18:46<samu>the partial parts can be on bare land, on river, on sea, on canal
18:46<Eddi|zuHause>samu: what i think you are seeing with the drag&drop clearing is that the game loops over all tiles you selected, especially all 3 of the lock tiles. on the first tile, it sees the lock, clears it, and leaves the water pieces. on the second and third tile, it doesn't see the lock anymore, because it has been cleared, so it clears the remaining water tiles
18:50<samu>wait, i lie, sea can never be on the upper part
18:50<supermop>V453000: its one of the plugins, and the only one i have now
18:50<V453000>right :)
18:51<supermop>it was the 'best' one back when i was in school, i think there are some more advanced ones now, as well as better versions of flamingo
18:51<samu>rivers can never be on the lower part
18:51<V453000>what does it do though? :P plugin is kind of a broad statement :P
18:51<samu>hmm 3*3*3
18:51<Eddi|zuHause>yes, they can
18:51<samu>they can ?
18:51<Eddi|zuHause>also, the canal may be by you or someone else
18:52<Eddi|zuHause>also, the game probably does not handle this case as "table"
18:52<samu>I see
18:53<samu>so I guess better solution is to create a canalcliff?
18:53-!-kais58 is now known as kais58|AFK
18:54<Eddi|zuHause>solution for what?
18:55<samu>locks have some issues
18:55<Eddi|zuHause>what that "solves" is that people drag&drop a canal, and then never figure out why the ships are not using it
18:55<supermop>its a renderer
18:56<V453000>I think I read about it somewhere before yeah :)
18:56<samu>they can't go up rivercliffs either
18:56<supermop>bu it also handles lighting, materials, texture mapping, environment plants, separate from rhino itself
18:57<samu>another simpler solution could be
18:57<samu>prevent river tiles from being demolished by drag&drop
18:57<samu>but not singleclick demolish
18:59<Eddi|zuHause>samu: but "rivercliffs" (they are actually "slopes" in game terminology, cliffs would be vertical drops between edges) have visual blockings
19:01<samu>river slope
19:02<samu>visual blockings, what's that?
19:03<supermop>i should have signed up for some classes here just to get a student id so i could upgrade my rhino stuff cheaply
19:05-!-Romazoon [] has quit [Quit: Page closed]
19:05<samu>can you attach a canal slope to the lock, exclusive to the lock only, not buildable via canal drag&drop?
19:06<samu>problem might come when destroying that lock
19:06<samu>this is so complicated
19:07<Eddi|zuHause>the lock IS the canal slope.
19:07<Eddi|zuHause>samu: with "visual blockings" i mean "there are rocks in the way"
19:08<Eddi|zuHause>may be difficult to see, depending on which base set you are using
19:08<samu>ah I get it
19:11<samu>trying to come up with a solution for the lock build cost
19:11<Eddi|zuHause>maybe you start by reading what the code actually does currently?
19:15<samu>that's the lock itself without the extra tiles
19:17<samu>building on bare land: 7500+5000+5000
19:17<samu>building on river: 7500+0+0+10000
19:17<samu>problem comes up when the canal tiles have different costs
19:18<samu>building on bare land: 7500+625+625
19:18<samu>building on river: 7500+0+0+10000
19:21<Eddi|zuHause>so, really you want to check the build locks command function, and change the handling of rivers.
19:44<samu>where is it ? can't find
19:48<samu>i find this, but
19:48<samu>what now
19:56<samu>complicated to follow this. It is building command names into other names
19:57<Eddi|zuHause>samu: well CMD_BUILD_LOCK is the internal number that is sent (e.g. over the network) when a lock should be built, and this line registers CmdBuildLock as the function that should be called when this command number is received
19:58<Eddi|zuHause>so the next step would be to look where CmdBuildLock is defined
20:12<samu>i can't find it
20:13<samu>it is not defined, the command is constructed?
20:13<samu>then sent into some recursive thing
20:13<samu>i dont know how to follow this code
20:17<Eddi|zuHause>maybe you're looking in the wrong file?
20:22-!-liq3 [] has joined #openttd
20:32<samu>ah, i think this is it
20:32<samu>static CommandCost DoBuildLock(TileIndex tile, DiagDirection dir, DoCommandFlag flags)
20:35-!-michi_cc_ [] has joined #openttd
20:35-!-michi_cc [] has quit [Read error: No route to host]
20:38-!-Biolunar [] has quit [Quit: leaving]
20:42<samu>it looks correct
20:43<samu>maybe the river slope is missing a property?
20:54-!-luaduck_zzz is now known as luaduck
20:55<samu>im looking at water_cmd.cpp
20:55<samu>what does this mean?
20:57<+glx>that they will be redrawn
21:03<samu>WaterClass wc_middle = IsWaterTile(tile) ? GetWaterClass(tile) : WATER_CLASS_CANAL;
21:07<samu>i'm looking here
21:07<samu>i don't see anything wrong
21:07<samu>i think
21:11<samu> /* Sloped canals are locks and no natural water remains whatever the slope direction */
21:11<samu>am i reading this right?
21:12<samu>sloped canals are locks, but it makes use of the water there is on that slope
21:12<samu>hmm, help
21:13<samu>it's a river
21:15<samu>im confused
21:23<samu>anyone here?
21:25<samu>all I know is that the cost is taking the middle tile clearing cost into the sum, but when destroying the lock, the river slope tile is restored
21:28<samu>I need something in there that goes like "if river tile is present -> add 0 to sum, if river tile not present -> sum the cost of clearing the tile"
21:29<samu>oh gosh, nobody answers, I have to go sleep
21:30-!-samu [] has quit [Quit: Page closed]
21:37-!-Celestar1 [] has joined #openttd
21:42-!-Celestar [] has quit [Ping timeout: 480 seconds]
21:49-!-quorzom [] has quit [Remote host closed the connection]
22:11-!-glx [] has quit [Quit: Bye]
22:36-!-Speedy [] has quit [Remote host closed the connection]
22:36-!-Speedy` [] has joined #openttd
22:36-!-Speedy` is now known as Speedy
23:50-!-luaduck is now known as luaduck_zzz
---Logclosed Mon Feb 16 00:00:15 2015