IRC logs for #openttd on OFTC at 2013-08-17
⏴ go to previous day
00:49:07 *** DarkAceZ has joined #openttd
00:50:33 *** Virtulis has joined #openttd
00:50:38 *** Virtulis is now known as a_sad_dude
00:57:10 <Eddi|zuHause> it's called "night"
00:58:42 <Eddi|zuHause> that's what i always say :p
00:59:22 <Bad_Brett> it's finally working... sort of
01:01:56 <Eddi|zuHause> you have some jumps in there
01:02:26 <Eddi|zuHause> are those because you didn't get the offsets right or because the length distortion in openttd's movement?
01:02:56 <Eddi|zuHause> and now for those curvy rails... :p
01:03:18 <Bad_Brett> i think some of the offsets can be improved
01:04:29 <Bad_Brett> but i'm happy that it kinda works, because it was pure hell coding this
01:05:32 <Bad_Brett> since it consists of 16 parts
01:10:19 *** NeuhNeuh has joined #openttd
03:06:17 *** scshunt has joined #openttd
04:03:58 *** roboboy has joined #openttd
04:37:10 *** a_sad_dude has joined #openttd
04:43:17 *** montalvo has joined #openttd
04:56:17 *** Eddi|zuHause has joined #openttd
04:56:27 *** amiller has joined #openttd
06:18:05 *** amiller has joined #openttd
06:26:04 <Arkabzol> as opposed to ladies of the night
06:27:11 <Arkabzol> In other news, cities grow strangely...
06:28:39 <Arkabzol> I wonder if grid size matters...
06:44:27 *** zeknurn has joined #openttd
06:58:23 <Rubidium> moin maker of planets
07:10:29 *** Pensacola has joined #openttd
07:41:27 *** Ristovski has joined #openttd
08:20:50 *** Alberth has joined #openttd
08:20:50 *** ChanServ sets mode: +o Alberth
08:40:57 *** ntoskrnl has joined #openttd
09:26:06 *** LordAro has joined #openttd
09:32:42 *** valhallasw has joined #openttd
09:34:34 *** DarkAceZ has joined #openttd
09:53:47 <maddy_> how do I show a generic error window (red background)?
09:55:12 <Alberth> there is some call to create an error window
09:57:14 <Alberth> ShowErrorMessage in error.h, I think
09:59:07 *** Progman has joined #openttd
10:03:09 *** frosch123 has joined #openttd
10:11:32 <maddy_> how would I solve this problem: my identifier for a window is TileIndex+Track, but WindowNumber is just int32
10:12:36 <maddy_> I guess I will need to create separate identifier for it
10:13:54 <frosch123> TileIndex is 22 bit, Track is 6 or 3 bit (depending whether you want TrackMask or TrackBit), so where's the problem?
10:14:54 <maddy_> well, nowhere if that is the case ;) I just saw TileIndex defined as uint32 or whatever, so I thought that takes all 32 bits
10:15:07 <frosch123> maximum mapsize is 2kx2k
10:15:17 <frosch123> so 11 bits + 11 bits
10:15:33 <maddy_> ok then it works perfectly
10:50:57 *** Devroush has joined #openttd
11:07:57 <Rubidium> ... except for the huge map patch ofcourse...
11:08:16 <Rubidium> but don't care about them. That's (eventually) the problem of that patch
11:15:31 *** Aristide has joined #openttd
11:31:11 <Eddi|zuHause> maddy_: ti | (track << (map_x + map_y)) should work
11:32:22 <maddy_> Eddi|zuHause: yeah that's what I used
11:33:33 <Eddi|zuHause> that should also work for most of the "huge map" patches
11:34:18 <Eddi|zuHause> at least i have never heard of them supporting 64k x 64k maps :p
11:35:16 <Eddi|zuHause> and all the newgrf stuff etc. breaks way earlier than that
11:38:19 <frosch123> how much newgrf stuff is there that depends on absolute map positions?
11:38:36 <Eddi|zuHause> well, there is one variable :)
11:39:22 <Eddi|zuHause> and the one GRF that could really make use of it (AV8) doesn't...
11:40:11 <Eddi|zuHause> so i guess nobody actually uses it and nothing will thus break ;)
11:40:13 <frosch123> what use should it make of it?
11:41:14 <Eddi|zuHause> scale ranges of the largest planes so they (barely) make it across the whole map
11:41:20 <frosch123> plane range is a relative distance, no absolute one
11:41:21 <Eddi|zuHause> depending on map size
11:42:06 <Eddi|zuHause> i was fairly sure when it was introduced they said it was absolute
11:42:14 <Eddi|zuHause> i don't actually use planes :p
11:43:19 <frosch123> well, yes, it is additive relative, not multiplicative relative :p
11:43:33 <Alberth> you use only low-flying vehicles at tracks :)
11:43:59 <Eddi|zuHause> i don't really use maglev either
11:44:17 <Eddi|zuHause> as in: i rarely get that late in the game.
11:44:17 <frosch123> Eddi|zuHause: what about hoover cats?
11:44:29 <frosch123> ah, they are late game as well
11:44:39 <planetmaker> how do hover cats work? bind marmelade bread on their back?
11:44:55 <frosch123> they only work on water afaik
11:45:14 <frosch123> maybe they are just unicorns in disguise
11:45:14 <MNIM> planetmaker: no, that's infinite energy generators
11:45:34 <planetmaker> true, he should know
11:45:39 <MNIM> Hovercrafts do work on land
11:46:05 <Alberth> that cannot be true, I never saw that happen in OpenTTD
11:46:10 <MNIM> it's just that they're a ^#%$@ to drive in a straight line that isn't a 0.0% grade exact
11:46:14 <planetmaker> but only because someone patched them, MNIM. In OpenTTD they can't yet
11:46:18 <frosch123> Alberth: there is a patch for it .p
11:46:48 <Alberth> frosch123: there is? cool. Ships on land!
11:47:25 <frosch123> Alberth: i changed the gettiletrackstatus function to return valid track for all tiles for ships
11:47:38 <frosch123> though it crashed when ships traversed certain tile types
11:48:02 <frosch123> because they tried to access something that asserts on some tiletypes
11:48:24 <frosch123> but andy is still dreaming about it :p
11:48:29 <Eddi|zuHause> probably should exclude houses and industries :p
11:48:45 <frosch123> Eddi|zuHause: or demolish them while driving :)
11:48:51 *** roboboy has joined #openttd
11:48:59 <planetmaker> godzilla-ships :D
11:49:05 <Eddi|zuHause> i think i had such problems with either the level crossing or fence patch
11:49:08 <planetmaker> the new disaster type. Borrowed from SC
11:49:18 <Eddi|zuHause> probably the fence
11:49:35 <Eddi|zuHause> i had to drop out early for things like houses
11:57:36 <maddy_> why do my UI components go in vertical order even when using NWID_HORIZONTAL
11:57:46 <maddy_> e.g. on top of each other
11:58:54 <frosch123> make sure the horizontal container is th emost inner one
11:59:02 <frosch123> there are some widgets which are containers in disguise
11:59:37 <maddy_> yeah that's probably the problem
12:02:47 *** TrueBrain has joined #openttd
12:30:21 *** Aristide has joined #openttd
12:31:54 <Eddi|zuHause> "Google downtime causes 40% drop in worldwide internet traffic"
12:40:14 <maddy_> looks like I might get my patch somewhat working today
12:59:43 *** Aristide has joined #openttd
13:06:39 <dihedral> Eddi|zuHause, then shut them down, the world can do wit the extra bandwidth
13:21:39 *** Pensacola has joined #openttd
13:23:39 *** brambles has joined #openttd
13:53:44 *** brambles_ has joined #openttd
13:58:35 *** NeuhNeuh has joined #openttd
14:02:30 <maddy_> how does CMD_ get resolved to a command function?
14:03:20 <frosch123> there's a table in command.cpp
14:07:54 *** Bad_Brett has joined #openttd
14:17:23 *** Tom_Soft has joined #openttd
14:25:33 *** ntoskrnl has joined #openttd
14:54:24 *** Aristide has joined #openttd
15:18:36 *** oskari89 has joined #openttd
15:21:18 *** sla_ro|master has joined #openttd
15:21:29 <maddy_> what am I doing wrong: first SB(p1, 8, 3, 1), then GB(p1, 8, 3) I get 0 back, when I expect to get 1?
15:21:53 <dihedral> i am in Germany - i want DE :-P
15:24:16 <maddy_> come on guys, must be something simple, just not seeing it
15:24:21 <Rubidium> maybe something in the (missing) context messes something up?
15:24:55 <maddy_> hmm, oh I think I know...
15:32:24 <maddy_> I had to set the bits in right order in consecutive SB calls
15:37:24 <dihedral> creating an irc connection is a slow process
15:44:25 *** montalvo has joined #openttd
15:45:29 <maddy_> I am using SB somehow wrongly
15:48:30 <maddy_> if I want to set a value to the first 6 bits, I should do SB(p1, 5, 6, ...) right?
15:49:30 <maddy_> what is the different between SB and Extract?
15:52:21 <dihedral> as far as i get the source, there are usually comments
15:53:24 <maddy_> yeah I know, I didn't understand from the comments, do you think I would ask for help if I'd understand it from the comments?
15:55:55 <frosch123> assigning first 6 bits would be SB(bla, 1, 6, ble)
15:56:03 <frosch123> err, SB(bla, 0, 6, ble)
15:56:19 <frosch123> GB is more low level than Extract
15:57:25 <maddy_> frosch123: ok, thanks, that was my error
15:58:05 <frosch123> what interpretation lead you to 5 ?
15:58:50 <maddy_> I thought the start bit is from the other end, e.g. for 6 bits 0 to 5, so I thought 5 is the start
15:59:09 <frosch123> ok, noone uses big endian
15:59:26 <frosch123> big endian is from an era where people thought computer shoud use decimal numbers
15:59:57 <maddy_> right, not sure why I thought that
16:00:39 *** andythenorth has joined #openttd
16:13:32 <V453000> frosch123: people still DO think computers should use decimal numbers :D
16:13:53 <frosch123> people also think you can play sc without keyboard
16:13:57 <frosch123> or program without linux
16:18:41 <Xaroth|Work> programming without linux is doable though
16:20:15 <frosch123> ok, that one is not as clear as the other ones :)
16:20:28 <Xaroth|Work> I blame .net for that :P
16:21:48 <Rubidium> Xaroth|Work: yeah, in the past everyone coded op paper with whitespace (or rather lack thereof)
16:22:51 <Rubidium> V453000: dozenal is nicer though ;)
16:23:40 <Rubidium> Xaroth|Work: if you thought I meant fortran, then you haven't thought back far enough
16:24:52 * TWerkhoven gets out the puncher
16:25:01 <Xaroth|Work> Rubidium: after sitting in the sun for this long my mind refuses to do much work, so I'm glad I made it as far as fortran
16:25:17 <TWerkhoven> when you could get coding equipment from the hardware store
16:41:40 *** Biolunar has joined #openttd
17:00:09 *** oskari892 has joined #openttd
17:14:31 *** Biolunar has joined #openttd
17:37:57 <maddy_> if I have TileIndex and Point, can I convert them to TileIndex and Track?
17:40:04 <Alberth> is there any relation?
17:40:20 <maddy_> not sure, but I think so
17:41:06 <Alberth> what do you express with a TileIndex and Point ?
17:41:40 <Alberth> note that I have no idea what you are coding at all
17:43:01 <maddy_> OnPlaceObject gives them as arguments
17:45:34 <DorpsGek> Commit by translators :: r25725 /trunk/src/lang (3 files) (2013-08-17 17:45:26 UTC)
17:45:35 <DorpsGek> -Update from WebTranslator v3.0:
17:45:36 <DorpsGek> afrikaans - 232 changes by mulderpf
17:45:37 <DorpsGek> indonesian - 31 changes by abdu354
17:45:38 <DorpsGek> luxembourgish - 69 changes by Phreeze
17:45:47 <Alberth> track_func.h looks useful to me for this problem
17:47:48 <Alberth> ah, you get precise coordinates, apparently
17:48:12 <Alberth> you may want to look what the rail construction toolbar window does there
17:49:37 <maddy_> yeah solved it I think
18:10:54 *** oskari89 has joined #openttd
18:20:47 *** HerzogDeXtEr has joined #openttd
18:49:56 *** andythenorth has joined #openttd
19:01:44 <krinn> is it normal to pay again station price to build a station at the same place with the same type (building a station over a station)
19:04:03 <krinn> the question is then why?
19:05:04 <Wolf01> in your country, if one want to renew/change something, can do it for free?
19:08:13 <krinn> well, of course not, but i would have payed the new thing, while here i pay the same thing
19:08:36 <krinn> except station creation date, nothing change (i'm not even sure the creation date change tbh)
19:09:00 <krinn> ah yes, i forgot (that's the main point here), the graphic change ^^
19:11:55 <Wolf01> I doubt the game can recognize it's the very same object, or it wouldn't allow the action (already built)
19:15:44 <krinn> hmmm they speak about newGRF, it's the same kind of issue, but i just speak about non newGRF
19:16:06 <Wolf01> it's to make that work
19:18:46 <krinn> that answer my "why" question, thanks guys
19:26:28 <Alberth> without NewGRF the issue is indeed confusing, but so far we haven't found a nice general rule that prevents this confusion without breaking other cases as well
19:28:20 <krinn> the only purpose (on non newGRF) is to make the graphic change, how about a zero or lower cost and allow an existing station graphic to change with another function instead of the one that create the station?
19:29:17 <andythenorth> I don't think it's worth worrying about
19:29:31 <andythenorth> I agree that it can be annoying to see when playing
19:29:45 <andythenorth> but it's only a problem when you don't have much money
19:30:21 <krinn> the money problem is always easy for human, it love to get complex with an AI
19:34:10 <krinn> and that's just the money problem : those kind of issue change the station tile name position (so the station position from an AI pov), something a human also don't care
19:34:47 *** DarkAceZ has joined #openttd
19:37:20 <krinn> ah no i've just tried, location doesn't change at least
19:37:48 <Eddi|zuHause> so why would your AI overbuild a station?
19:37:50 <Alberth> perhaps only when you remove the tile with the sign
19:37:51 <andythenorth> location only changes if station is rebuilt iirc
19:38:19 <Eddi|zuHause> location only changes if you remove parts so that the sign is outside the station rectangle
19:38:30 <krinn> Eddi|zuHause, i suppose like human, to get the new graphic: built one width station, add one width to it : you get a 2 flat platforms station
19:38:50 <krinn> Eddi|zuHause, overbuild it with a 2 width station, you get the graphic with roof
19:39:08 <Eddi|zuHause> yes, and why do you expect that to be free?
19:39:54 <krinn> because two width station or 2x 1 width station == same price, while now it gave twice
19:40:20 <krinn> so if you don't pay the roof, i don't see why i should pay it ?
19:41:01 <Eddi|zuHause> actually i think building a second 1 width platform is more expensive than the first
19:42:01 <krinn> no exact same price as first (if no tree... on it)
19:42:20 <Eddi|zuHause> i'm not saying the current pricing model makes any sense. but you don't really make a case for why a different method would make more sense
19:43:10 <krinn> told u, if roof cost 0 why i should pay roof the 2xstation price ?
19:44:07 <Eddi|zuHause> there is no roof
19:44:22 <Eddi|zuHause> it's an illusion
19:47:55 <andythenorth> newgrf cost cb? :P
19:48:03 <andythenorth> make yet more work for newgrf authors? :P
19:48:12 <andythenorth> and some vars to read the current tile info
19:48:38 <krinn> that would work for "classic" one too?
19:51:20 <MNIM> So, not even a single comment on how inefficient that station is?
19:51:28 <krinn> but it's ok if design make it hard, let's pay the price or just not make it with roof
19:53:51 <Alberth> MNIM: it's sufficiently efficient, given the number of trains in the image
19:54:25 <MNIM> Hah. In this case it would rather be inefficient, as a waste of infrastructure :P
19:54:59 <Alberth> andy, what does it carry?
19:55:16 <Alberth> it says "schwerlast", ie "heavy load"
19:55:22 <MNIM> Looks like it pulls, rather than carry
19:55:53 <MNIM> I think it's a tractor for heavy terrain, weighted with concrete
19:55:55 <krinn> looks like kind of truck to pull heavy engine
20:30:17 *** HerzogDeXtEr1 has joined #openttd
21:12:37 *** Supercheese has joined #openttd
21:20:09 *** tokai|mdlx has joined #openttd
21:21:44 <Supercheese> I'm having a tough time understanding how OTTD generates station names
21:22:48 <Supercheese> Seemingly a pool of available names is gradually decreased
21:26:53 <planetmaker> Supercheese, the best explanation here is indeed the source
21:27:13 <planetmaker> it basically goes down a list of available names. And choses the first available one
21:27:27 <frosch123> most names have restrictions
21:27:53 <frosch123> like west/east/south/north/heights/center/woods etc. actually have a meaning
21:27:56 <Supercheese> So, when a grf provides a nearby station name, where does that enter into the name generation? It seems somehow the Oilfield and Mines names are disabled, for some reason
21:28:05 <Supercheese> in FindNearIndustryName
21:28:13 <planetmaker> grf-provided names got priority
21:28:19 <planetmaker> but still must be unique
21:28:33 <Supercheese> I cannot find where they are removed from the pool of free names
21:28:42 <Supercheese> if they are, that is
21:30:03 <Eddi|zuHause> what exactly are you reading?
21:30:14 <Supercheese> GenerateStationName
21:34:26 <Eddi|zuHause> uint32 tmp = free_names & _gen_station_name_bits[name_class]; <-- this is the line where oilfield etc. are tested
21:35:09 <Eddi|zuHause> airport/oilfield/dock/heliport like in the beginning of the function
21:35:52 <Supercheese> Some of these operators confuse me, I think they're bitwise operators?
21:36:36 <Eddi|zuHause> yes. 1<<blah means "set the blah-th bit of the bitfield"
21:36:58 <Supercheese> No matter how many times I read about it on wikipedia, it doesn't make much sense
21:38:14 <Eddi|zuHause> in theory one could introduce there one or more names for stations and busstops as well. like the first railway station could get "(main) station" and "central" could be reserved for bus stops, etc.
21:38:38 <Supercheese> Rail and Road don't have any custom strings, it seems
21:38:55 <Eddi|zuHause> but they could have
21:39:03 <Supercheese> that whole station name bits thing is perplexing
21:39:09 <Supercheese> what are the "1U"s?
21:39:30 <Eddi|zuHause> 1U is like a regular 1, only it's "unsigned"
21:39:45 <Eddi|zuHause> which may or may not make a difference in certain situations
21:40:41 <Eddi|zuHause> the M(STR_SV_STNAME_AIRPORT) counts the position of the station name relative to the beginning of the list
21:41:12 <Supercheese> I wonder why setting a nearby station name in a grf disables the Mines name bit
21:41:27 <Eddi|zuHause> so if that is the 5th station name, then the 5th bit will be set, i.e. yout of the 32 bits of an "unsigned int" you get "0000 0000 0001 0000"
21:41:32 <Supercheese> I don't see anything special about the Mine name check
21:41:42 <Eddi|zuHause> forgot a few 0000
21:42:45 <Eddi|zuHause> so this check "free_names & _gen_station_name_bits[name_class]" looks up whether a name is free AND is in the list of special names for this class. if there is one, it is returned in the next line
21:43:04 <frosch123> mine is the prototype of a station name depending on an industry
21:43:31 <frosch123> also the rule for "mine" is not exactly useful for newgrf industries
21:43:44 <frosch123> it would also trigger on stuff like sandpits
21:45:19 <Supercheese> Hmm, where is the property 24 station name returned? I can't seem to find it
21:45:30 <Supercheese> all I see is Oilfield and Mines are disabled
21:45:47 <frosch123> it stores the industry type in the station
21:46:06 <frosch123> and sets the name to some STR_ representing naming via industry
21:46:17 <Eddi|zuHause> that's probably buried in the CircularTileSearch
21:46:18 <frosch123> the actual property is evaluated way later
21:46:30 <frosch123> that is needed because the grf may change and the name may change
21:46:40 <Eddi|zuHause> FindNearIndustryName
21:46:40 <frosch123> so you cannot store the name in the savegame
21:46:46 <frosch123> but only a reference to the industrytype
21:48:19 <Supercheese> I think I need to find the place where the name is evaluated
21:53:20 <frosch123> strings.cpp or string.cpp
21:53:31 <frosch123> never can remember which is which :p
21:53:46 <frosch123> but i have no idea what you want in that place
21:53:53 <frosch123> everything is decided there already
21:54:07 <frosch123> it's only the gui code for drawing
21:54:11 <Supercheese> overriding newgrf-provided nearby station names with station names from nearby industries
21:54:33 <Supercheese> basically I don't like FIRS' new nearby names
21:55:04 <frosch123> you want all stations to have the exact same name as the industry :p
21:55:07 <Supercheese> so I'd like to add an OTTD advanced option to name stations according to nearby industry, irrespective of grf-provided stuff
21:55:19 <frosch123> like no station names at all basically
21:55:22 <frosch123> only industry names
21:55:30 <Supercheese> for when stations are beside industries at least
21:55:39 <frosch123> well, yeah, then string/strings.cpp might be the right place for you
21:55:48 <Supercheese> things get a bit hairy for in-town industries like grocery stores
21:55:57 <Supercheese> but even then I like it, others might not though
21:57:07 <Supercheese> yikes, this is a very, very long chain of case XX:, case YY:, case:ZZ...
21:57:57 <Supercheese> strings.cpp seems to be where the magic happens
23:05:21 <Supercheese> ugh, patchwriting is frustrating
23:09:51 <Supercheese> I could have edited FIRS 10 times by now
23:10:06 <Supercheese> rather than try for the "elegant" solution
23:35:05 <Supercheese> Man, where does the town name get put into the nearby station name?
23:44:22 <Supercheese> Meh, I give up for now
continue to next day ⏵
