IRC logs for #openttd on OFTC at 2011-05-31
⏴ go to previous day
01:17:20 *** fjb is now known as Guest2883
01:27:32 *** douknoukem has joined #openttd
01:50:58 *** Eddi|zuHause2 has joined #openttd
02:25:35 *** rhaeder1 has joined #openttd
03:51:39 *** DoubleYou has joined #openttd
04:56:23 *** Eddi|zuHause2 has joined #openttd
05:11:41 *** rhaeder has joined #openttd
05:39:33 *** DayDreamer has joined #openttd
05:43:02 *** HerzogDeXtEr has joined #openttd
05:54:15 *** Cybertinus has joined #openttd
06:02:38 *** Adambean has joined #openttd
06:06:43 *** Kurimus has joined #openttd
06:28:28 *** Br33z4hSlut5 has joined #openttd
06:32:20 *** ChoHag_ is now known as ChoHag
06:39:58 *** tokai|mdlx has joined #openttd
06:48:18 *** Progman has joined #openttd
07:23:05 *** JVassie has joined #openttd
07:42:46 *** ashledombos has joined #openttd
07:44:06 <planetmaker> nice. I thought and hoped you'd add something on the passenger network, Terkhen :-)
07:51:05 <Terkhen> aircrafts are boring though
07:51:34 <Terkhen> I was thinking on trying the vactrain set in a passenger only game but the tracks are insanely expensive
07:52:25 <Terkhen> and it is not easy to make crazy amounts of money in a yacd game
07:53:45 <Ammler> and ships aren't boring? :-)
07:54:26 <Terkhen> you still have to think a bit
07:54:33 <Ammler> you used almost as many ships as trains
07:55:52 <Terkhen> ships are also great for local passengers on big towns
07:56:15 <Terkhen> I prefer metro trains but we didn't have them in that game
07:57:26 <Ammler> specially the fact that ships don't queue or crash is quite boring
07:58:03 <Ammler> on your screen, there are aound 5 ships in one :-)
07:59:41 *** Progman has joined #openttd
08:00:19 <Terkhen> not crashing is boring? random crashing is really annoying and serves no purpose IMO
08:02:54 <Ammler> well, I meant that ships just drive through each other
08:03:26 <Terkhen> yes, it would be nice to have better behaviour for them
08:08:51 <planetmaker> <Ammler> you used almost as many ships as trains <-- yes. But they substitute basically trains which would carry the same
08:09:34 <Ammler> planetmaker: I meant according to the fact that ships are as boring as aircraft... :-)
08:11:41 <Terkhen> that's not a fact for me :P
08:14:00 <dihedral> and planetmaker, Ammler and the lot ;-)
08:15:06 <Ammler> Terkhen: yes, ships are more boring, aircrafts at least have some fun on the airport :-P
08:15:56 <Terkhen> let's end this circular discussion as a tie :)
08:20:04 <__ln__> plural of aircraft is aircraft
08:21:13 <Terkhen> english is so strange
08:25:30 <Ammler> should have used planes
08:28:11 *** Eddi|zuHause2 is now known as Eddi|zuHause
08:36:29 <__ln__> the rain in spain stays mainly in the plane
08:51:41 <Noldo_> in Hartford, Hereford and Hampshire hurricaned hardly ever happen
08:57:14 <peter1139> what's special about that lot?
09:02:40 *** Brianetta has joined #openttd
09:29:23 *** roboboy has joined #openttd
10:02:44 *** Vikthor has joined #openttd
10:11:57 <planetmaker> <Ammler> should have used planes <-- arcraft = [planes|helicopters|plimps|...]
10:13:11 <Ammler> as said, aircraft are exiting :-)
10:13:27 <Ammler> not that boring as ships
10:16:44 <planetmaker> They are exciting as they allow to use different airports at least
10:16:53 <planetmaker> But ships are exciting if FISH is used
10:17:04 <planetmaker> exciting as in look great :-)
10:17:42 <planetmaker> and clearly with our yacd game efficiency was not the only goal. Rather "having fun" and "working somewhat smoothly" and "looking nice" ;-) All at once
10:27:03 <planetmaker> For me the biggest difference from YACD to normal game actually is the money maker phase. It's much more extended
10:29:41 *** Intexon has joined #openttd
10:33:09 <planetmaker> Though it still has some issues. Among them the random desync
10:33:49 <fjb> Random desyncs are no problem in single player. :-)
10:34:39 <fjb> FIRS is incompatible with NARS2 and UKRS2 at the moment?
10:39:30 <Ammler> isn't the desync already fixed in yacd dev?
10:39:41 *** KenjiE20 has joined #openttd
10:40:53 <Ammler> do you earn money if you transport passenger from station A to station A?
10:41:21 <Yexo> I doubt it, since the payment is based on the distance
10:42:01 <Ammler> ah indeed, there was once a discussion to use distance of the "industries", not the stations
10:42:21 <planetmaker> fjb, UKRS2 is fine afaik. Not sure about NARS2: re-gearing is a problem
10:42:29 <planetmaker> if it uses the re-gearing cargo
11:05:47 *** sla_ro|master has joined #openttd
11:13:52 <fjb> There was a discussion about FIRS and regearing.
11:14:23 <fjb> Is UKRS complete enough to be useful?
11:20:22 *** Chris_Booth_ has joined #openttd
11:25:13 <planetmaker> dunno, I think so. It misses IIRC modern engines
11:25:24 <planetmaker> UKRS2 that is. UKRS, of course
11:26:28 *** Progman has joined #openttd
11:27:11 <fjb> Yes, I somewhere lost the 2. :-)
11:28:44 *** Chris_Booth_ is now known as Chris_Booth
11:29:52 <fjb> I'm still playing my r19639 with cargo distribution based game.
11:30:27 <Chris_Booth> how do I reset the engine pool? since I cheated the date forward and back in a game
11:30:35 <Chris_Booth> but don't want all locos
11:33:07 <planetmaker> uhm... did you try "reset_engines"?
11:33:37 <Chris_Booth> thanks planetmaker
11:34:06 <planetmaker> no idea whether it makes engines un-available
11:35:15 <Chris_Booth> no I cheated in the 2 locos I needed
11:41:32 <dihedral> \o/ quake3 at work :-D
12:26:33 <fjb> Didn't play for months. Now I remember what the problem in that games was: airports.
12:30:04 <Terkhen> and it is hard to replace them
12:32:12 <Eddi|zuHause> the size of the airports isn't the problem, it's there absolute inability to manage any kind of traffic
12:32:15 <fjb> But I'm in 2001 now. So they got replaced where ever possible.
12:32:56 <fjb> Eddi|zuHause: True. But you have to build more of them to work around that. And then size becomes an issue.
12:33:45 <fjb> And Pikka's planes are as much fun as Andy's ships.
12:37:10 <Eddi|zuHause> whenever i try, i never have fun with planes...
12:37:45 <Eddi|zuHause> they end up either circling around the airport or waiting in a hanger, depending on which airport is used
12:37:55 <planetmaker> airports are too croweded by the year 2000. Then they need a train station attached to handle the traffic ;-)
12:38:02 <Terkhen> I tried a game with multiple terminals (YACD makes that really simple) but I was bored soon
12:38:24 <planetmaker> and we still can't introduce alternative state machines.
12:38:32 <Eddi|zuHause> the interesting airports come way too late
12:38:48 <planetmaker> Eddi|zuHause, then alter their introduction date... by newgrf ;-)
12:39:20 <Eddi|zuHause> and like i said, you have no means to influence them once they are built, not even timetables help significantly
12:42:51 <fjb> They hangars get overcrowded.
12:43:26 <fjb> And I can not really add a railway station to the most problematic airport.
12:43:56 <fjb> What happend to the new airports branch?
12:46:26 <Eddi|zuHause> some people can't get their head around the fundamental design problems
12:47:52 <fjb> That sounds complicated? What is the problem? The state for the movements at ground?
12:50:32 <Eddi|zuHause> everything except the state machines is in trunk already
12:53:33 <planetmaker> the other rotations and other airports are not done mostly due to missing graphics
13:01:13 <fjb> planetmaker: Nice, but the last small airport got upgraded some years ago. :-)
13:08:38 <Eddi|zuHause> rotating airports isn't all that realistic, since runways are usually built in the (prevailing) wind direction
13:12:03 <planetmaker> then tell me where the wind direction is ;-)
13:12:33 <Terkhen> please, not the wind discussion again
13:12:36 <planetmaker> consider airport wind socks, vehicle smoke and power plant smoke ;-)
13:13:35 <Eddi|zuHause> this isn't about where the wind comes from. just pick a direction for runways and stick with it for all "rotations" ;)
13:14:58 <planetmaker> but as airports usually have at least two perpendicular landing strips ;-)
13:15:14 <Eddi|zuHause> only very few big ones have that, actually
13:15:41 <Eddi|zuHause> and only one of those is used at any given time
13:16:01 <Eddi|zuHause> so unless you introduce weather ;)
13:19:16 <Ammler> [15:12] <Terkhen> please, not the wind discussion again <-- :-D
13:20:21 <Ammler> Eddi|zuHause: usually the landing direction is ruled by the powerful neighbours
13:20:31 <fjb> Hm, now I have two crossing tunnels. Or would have if it would be possible. :-(
13:20:42 <planetmaker> fjb, cheat menu...
13:21:14 <planetmaker> how would it be less boring if it weren't in the cheats?
13:23:52 <fjb> That would also be boring.
13:24:31 <fjb> I found a creative solution. I move one tunnel.
13:37:27 <fjb> Now I only have to reroute some planes to the new airport.
13:43:40 *** Belugas2 has joined #openttd
13:45:55 *** Belugas2 is now known as Belugas
14:29:28 *** SirSquidness has joined #openttd
14:32:50 *** frosch123 has joined #openttd
14:42:31 *** Zeknurn has joined #openttd
14:45:57 *** SirSquidness has joined #openttd
14:46:30 *** lasershock` has joined #openttd
14:47:43 *** supermop has joined #openttd
14:49:05 *** DayDreamer has joined #openttd
14:49:18 *** DayDreamer has left #openttd
14:54:00 *** Chris_Booth has joined #openttd
15:06:24 *** XeryusTC2 is now known as XeryusTC
15:42:40 *** HerzogDeXtEr1 has joined #openttd
15:53:24 *** lucas92 has joined #openttd
15:54:07 <lucas92> I have some problem compiling liblzma
15:55:15 <lucas92> libtool: link: cannot find the library '/mingw/lib/libiconv.la'
15:55:39 <lucas92> but I check if the library is in the folder and it's there... :(
15:55:49 <lucas92> I can compile OpenTTD fine without it
15:56:06 <lucas92> but I would like to compile it fully
15:56:35 <lucas92> I'm compiling with mingw and I followed the steps in the wiki
15:57:02 <lucas92> is there anyone that have an idea?
15:57:21 <lucas92> or should I post in the forums instead?
15:58:07 <Terkhen> lucas92: you probably forgot the etc\fstab step of the "Setting up MSYS" section
15:59:41 <lucas92> I also modified the .bat since I'm on windows 7
16:00:11 <Terkhen> hmm... then I'm not sure of what could be the issue
16:00:15 <lucas92> but I have an idea of the problem
16:00:44 <lucas92> doesn't point to minGW
16:00:56 <lucas92> but an another folder in C: called Cadence
16:01:14 <lucas92> how do I correct this?
16:02:19 <lucas92> it should point to the MinGW directory, right^
16:02:24 <Eddi|zuHause> the ~ directory shouldn't have any influence whatsoever on building openttd
16:02:59 <lucas92> yeah I thought so because I've built it correctly without lzma
16:03:24 <Eddi|zuHause> and the ~ directory would usually be your profile directory on a default mingw installation
16:04:41 <Eddi|zuHause> i.e. "C:\Documents and Settings\Username\" or something... name varies between windows versions
16:06:43 <lucas92> but it's not right now...
16:06:51 <lucas92> is it important at all?
16:07:49 <lucas92> in fact, I would just need the libraries compiled
16:08:04 <Terkhen> the placement of the home directory is not important... but the fact that it is not placed in the default place (\mingw\msys\1.0\home\user) makes me wonder if you followed the tutorial exactly
16:08:08 *** Archimedes has joined #openttd
16:08:40 <lucas92> followed it exactly, I have no idea why it points to this directory
16:09:37 <lucas92> and I don't see any home/user folder in the msys directory
16:09:52 <lucas92> do I have to create these?
16:10:55 <Terkhen> it should have been created automatically
16:11:12 <Terkhen> unless something you installed defined HOME as a global variable
16:11:20 <Terkhen> it happened to me once when mixing different mingw installations
16:12:02 <Terkhen> control panel -> system -> advanced options -> environment variables
16:12:12 <Terkhen> (I'm not sure if those are the right names in english, sorry)
16:12:19 <Terkhen> then again, it might be something in the register too
16:13:09 <lucas92> yes I see the HOME affected to the directory
16:13:17 <lucas92> so where should I put it?
16:14:36 <lucas92> C:\MinGW\msys\1.0\home\user ?
16:15:12 <Terkhen> if you didn't create the HOME environment variable changing it might break the program that created it in the first place
16:15:13 <lucas92> then I copy the contents of the old HOME to this new directory?
16:15:21 <Terkhen> besides, it is probably not the issue
16:15:32 <Terkhen> unless it is doing the same for other directories
16:16:00 <Terkhen> hmm... libiconv should be present by default
16:16:28 <Terkhen> can you try to set in msys "mingw-get update" or maybe it was "mingw-get upgrade", I'm not sure
16:16:33 <Terkhen> just in case some package is broken
16:17:56 <lucas92> libiconv.la is in C:\MinGW\lib
16:18:05 <Terkhen> that's the right place, yes
16:18:24 <Terkhen> if you really edited /etc/fstab it should find libiconv.la in /mingw/lib/
16:18:39 <Terkhen> can you paste your /etc/fstab to pastebin?
16:19:58 <lucas92> but to be sure, /mingw/lib is in which directory?
16:20:05 <Archimedes> Hi! Would anyone happen to know if there is some good documentation avaible about the server list and how it works?
16:21:32 <lucas92> ok, finished updating
16:21:42 <Archimedes> Ah well i might've just found the information I was looking for.
16:22:25 <lucas92> I'm still not sure about where /mingw/lib should be
16:23:00 <lucas92> why is this library so difficult to compile
16:24:43 <lucas92> the library did begin to compile also, I think the error is at the linking part
16:25:45 <lucas92> if anyone have xz compiled for Windows and MinGW say so
16:25:59 <Terkhen> if you have broken paths openttd will fail anyways
16:26:22 <lucas92> OpenTTD works correctly without it
16:26:31 <lucas92> it just cannot save/load maps
16:26:46 <Terkhen> IIRC /mingw/lib is c:\mingw\lib
16:28:17 <Terkhen> try to compile openttd without it (./configure --without-lzma)
16:28:24 <Terkhen> but there is some issue in your mingw
16:28:25 <lucas92> I've already built it
16:29:05 <lucas92> all I need is the compiled library for MinGW
16:29:22 <Terkhen> I have no access to mingw right now
16:32:03 <Terkhen> I have always used self compiled xz
16:32:30 <Terkhen> it mentions that xz has official windows binaries, though
16:32:32 <lucas92> so where do you put the include file and the bin files
16:32:42 <Terkhen> I'll look into that for the next time I update the tutorial
16:32:53 <Terkhen> lucas92: I don't know, follow the tutorial of whoever uploaded the binaries
16:33:01 <planetmaker> why do you want to compile the library and not just download it?
16:33:14 <Terkhen> planetmaker: because the tutorial says so :)
16:33:51 <planetmaker> I'd not compile libraries but try to download them... if I weren't interested in building universal ones ;-)
16:34:23 <Terkhen> that's what you usually do, but it is quite rare to find libraries compiled for mingw if they are not included with mingw itself
16:35:10 <Terkhen> it seems that they only have been posting compiled versions for mingw since 5.0.0, when I updated the tutorial xz was in 4.999 or something like that IIRC
16:35:55 <Terkhen> this will make the tutorial considerably simpler; xz needs a lot of stupid stuff to compile :)
16:36:29 <planetmaker> hm... " Note that the MinGW distribution includes liblzma"
16:37:54 <Terkhen> lucas92: try "mingw-get install mingw32-liblzma" or "mingw-get install mingw32-xz"
16:39:21 <planetmaker> Terkhen: no idea of "since when"... :-) Just google-hopping ;-)
16:40:11 <Terkhen> lucas92: did it work? did it install something?
16:40:25 <lucas92> yes, it installed a .tar.bz2
16:40:35 <Archimedes> Looks like there are some people here with (OTTD) coding experience. Would it be rearlly hard to implement a new new package for the server browser which just exports the game settings (difficultiy, advanced settings)?
16:41:01 <planetmaker> Archimedes: openttd.cfg is exactly that...
16:41:02 *** Alberth has joined #openttd
16:41:02 *** ChanServ sets mode: +o Alberth
16:42:10 <Terkhen> lucas92: I was talking about the mingw-get install ...
16:42:27 <Archimedes> planetmaker but (as far as i know) it's not possible to get the cfg without actually downloading the map (it seems). I'd like to be able to filter out servers that ... let's say have inflation on. I kind of miss such an advanced filtering option.
16:42:59 *** Kurimus has joined #openttd
16:51:19 <Terkhen> mingw-get should have installed the library in the right place
16:51:24 <Terkhen> just try to compile openttd
16:53:09 <lucas92> mingw-get installed me a tar.bz2
16:53:32 <lucas92> the ./configure says lzma is not detected
16:57:43 *** Turbulentor has joined #openttd
16:57:48 <Terkhen> use the compiled version found at the xz homepage and unzip its content in c:\mingw\ then (lib over c:\mingw\lib, include over c:\mingw\include and so on)
16:57:49 *** Turbulentor is now known as TWerkhoven
17:00:19 *** ashledombos has joined #openttd
17:02:34 <Terkhen> hmm... I don't know what else we could try :P
17:03:39 <lucas92> anything that I could try to test if lzma is here?
17:03:52 <lucas92> well, it's in the directory
17:04:03 <lucas92> but to check if they work
17:04:16 <lucas92> or maybe it's their name
17:04:20 <Terkhen> configure does that, that's why it says it can find it
17:04:54 <lucas92> there's no liblzma.la
17:05:37 <lucas92> in the compiled version
17:05:55 <lucas92> while I think there was in the home version
17:07:43 <Terkhen> what is the "home version"?
17:08:59 <lucas92> nvm, I think I found the problem, when I run lzmaininfo.exe, it says that my version of Windows is not supported
17:09:19 <Terkhen> errr... what version of windows do you have?
17:10:12 <Terkhen> me too, so it is supported :P
17:10:43 <lucas92> you tried with the compiled one?
17:11:24 <Terkhen> no, but I have windows 7 and xz works fine
17:11:48 <lucas92> yeah, it might be the compiled on that have the problem
17:12:06 *** Adambean has joined #openttd
17:12:29 <Terkhen> besides uninstalling mingw, deleting the c:\mingw folder completely and starting again I don't know what else you could try, sorry
17:13:01 *** Archimedes has joined #openttd
17:23:04 <lucas92> oh well, I guess I won't compile OpenTTD with this library then
17:25:55 <lucas92> nice, I have a little bit of experience with C++, played with Box2D, boost, sfml, SDL, tinyXML
17:26:11 <lucas92> I'd be willing to help in dev
17:26:52 <lucas92> only used foreach, multi_array and array in boost though
17:27:03 <Alberth> there are plenty of problems and feature requests in the tracker, as well as in the problems and suggestions forums to try
17:27:38 <lucas92> I think it simplifies codes readablity
17:28:08 <lucas92> for example, let's say you want to iterate over every object of a std::vector
17:28:59 <lucas92> for(int i = 0; i < vector.size(); ++i)value = vector[i].value
17:29:10 <lucas92> with foreach you would do that
17:29:31 <Alberth> lucas92: nope, you'd use the STL approach
17:29:32 <lucas92> BOOST_FOREACH(T object, vector)
17:30:06 <lucas92> with iterators you mean?
17:30:33 <lucas92> I don't remember too well how you do it
17:30:35 <Alberth> I don't see the big advantage over an explicit for() imho
17:31:04 <Alberth> for(iter = vector.begin(); iter != vector.end(); iter++) value = iter->value
17:32:09 <lucas92> well, foreach can itterate over other types of list I think
17:32:47 <lucas92> maybe it wants to mimic the C# syntax
17:33:05 <Alberth> boost and STL iterators pre-date C#
17:34:03 <Terkhen> so does foreach, IIRC it is present in some languages older than C
17:34:08 <lucas92> there are other things that boost offers such as signals and multi-arrays
17:34:16 <Terkhen> signals? like qt signals?
17:34:37 <lucas92> you call a functor on the call
17:35:39 <lucas92> anyway, have to go eat I'm back in 20 minutes
17:35:48 <Terkhen> signals are not that great when you have thousand of instances of different classes interacting in many different ways :P
17:45:36 <CIA-1> OpenTTD: translators * r22526 /trunk/src/lang/serbian.txt:
17:45:36 <CIA-1> OpenTTD: -Update from WebTranslator v3.0:
17:45:36 <CIA-1> OpenTTD: serbian - 4 changes by etran
18:16:23 <lucas92> so what is to be done with the code?
18:16:36 <lucas92> is it the to-do list in doxygen
18:16:40 <Terkhen> check the tracker and the forums
18:16:47 <Terkhen> there is no clear todo list anywhere
18:17:19 <Terkhen> the most important thing are always open bug :)
18:19:12 <lucas92> I can't find the "tracker"
18:21:59 *** andythenorth has joined #openttd
18:22:43 <lucas92> what is the color code of the bugs?
18:23:04 *** andythenorth has left #openttd
18:24:01 *** andythenorth has joined #openttd
18:25:26 <Alberth> color is priority, look at the category instead
18:26:00 <Alberth> euhm, sorry, make that 'type' :)
18:26:15 * andythenorth had to register with oftc again
18:26:29 <andythenorth> register / login /s
18:26:33 <SmatZ> why did you have to reregister?
18:26:42 <andythenorth> probably my client is deficient
18:27:11 <SmatZ> more likely than you think :)
18:27:26 <andythenorth> I saw the drive-by-download attack on sunday
18:27:48 <andythenorth> interesting how far it gets + how well crafted it is
18:28:42 <andythenorth> someone just filled the dev forum with a nice kind of spam
18:37:26 *** Intexon has joined #openttd
18:42:59 <Yexo> hmm, the end of grfcrawler is near
18:44:51 <andythenorth> I was going to make some updates to bananas to improve-ify bananas
18:45:08 <andythenorth> but I lack a django dev environment
18:48:06 *** Phoenix_the_II has joined #openttd
18:48:45 <frosch123> is there anything to save from that page?
18:49:21 <frosch123> also i thought it runs on the forums and is controlled by orudge
18:49:32 <Yexo> that's what I thought too
18:50:39 <orudge> and personally, I have no desire to see it closed
18:50:50 <orudge> so, as I said in #tycoon, we'll see what happens
18:51:30 <Alberth> does a full backup exist?
18:51:38 <orudge> GRFCrawler is significantly superior to Bananas in terms of its organisation of GRFs, etc
18:51:52 <orudge> ie, actually categorising things
18:52:02 <orudge> and including little screenshots, etc
18:52:07 <Yexo> only by now bananas is superior in updated content
18:52:18 <Alberth> it needs a bit cleaning up imho
18:52:44 <Alberth> lots of 'old, please use that and that grf
19:00:28 <lucas92> finally I think I'm going to use Visual C++ lol
19:14:23 *** Phoenix_the_II has quit IRC
19:25:05 <CIA-1> OpenTTD: frosch * r22527 /branches/1.1/ (6 files in 4 dirs): [1.1] -Prepare for 1.1.1
19:26:30 <andythenorth> my plan was to add a page to bananas for each grf
19:26:40 <andythenorth> with description, links etc
19:27:47 <andythenorth> but plans are not horses
19:28:12 <Terkhen> that would be great, right now it just links to an outside web
19:31:03 <andythenorth> it would be the best compromise if links are never going to be clickable in-game
19:31:29 <andythenorth> there was also an argument about a url shortener, which I seemed to lose :P
19:31:38 <andythenorth> although I was right
19:31:55 <DorpsGek> Chris_Booth: michi_cc was last seen in #openttd 2 days, 9 hours, 16 minutes, and 40 seconds ago: <michi_cc> frosch123: They don't. Not that it would be hard to hadd.
19:32:15 <Chris_Booth> not here for a long time
19:32:26 <Chris_Booth> any other YACD devs here?
19:32:51 <andythenorth> there is only one
19:32:55 <andythenorth> 2 days is not long
19:33:04 <andythenorth> and he said he would be busy for several weeks
19:33:50 <Terkhen> I guess he will see the highlight :)
19:35:34 <CIA-1> OpenTTD: frosch * r22528 /tags/1.1.1/ (10 files in 4 dirs): -Release: 1.1.1
19:36:16 *** amkoroew has joined #openttd
19:39:53 <Chris_Booth> yes I am sure he will
19:39:58 <Chris_Booth> or read my forum post
19:40:14 <Chris_Booth> I just find it quick to tell in here sometimes
19:49:07 <lucas92> generating OpenTTD is long on Visual C++
19:50:01 *** Phoenix_the_II has joined #openttd
19:55:34 * andythenorth can't see how to fix FIRS industry probabilities at map-gen
19:56:19 <andythenorth> depending on start date, player gets too many or too few of each industry
19:57:13 <andythenorth> any suggestions?
19:59:25 <Hirundo> Too few/many industries in total, or of certain types?
19:59:53 <andythenorth> adjusting the probabilities for date wxyz just causes problems for date abcd
19:59:55 * Alberth considers changing probability handling as well
20:00:25 <andythenorth> separately I think there's an issue with random construction during gameplay, but I can't prove it
20:01:55 <andythenorth> in every test game I've played, if I start prior to 1923, I get no aluminium plants during gameplay
20:02:08 <andythenorth> if I start after 1923, I get aluminium plants at map gen
20:02:10 <Alberth> ie if there is no room for an industry at the map, the game should generate less of that type, whatever its prio says
20:02:22 <andythenorth> (availability is obv. 1923 for alu. plant)
20:02:54 <andythenorth> ^ the issue I describe above is based on limited sample size
20:03:13 <andythenorth> so if it's really random, the behaviour could be correct, and I'm just seeing a particular random effect
20:03:23 <Alberth> priority of the al plant is similar to other industries?
20:03:38 <andythenorth> does map gen try any harder to build an industry?
20:03:43 <andythenorth> alu plant is large and hard to build
20:03:52 <Alberth> it uses different prios
20:03:55 <andythenorth> I suspect I know the issue
20:04:03 <andythenorth> there are limited sites available
20:04:24 <andythenorth> those sites are likely to be more contended by other industry during gameplay than at map-gen
20:05:03 <Alberth> I don't know the order in which industries are generated at the map
20:05:13 <andythenorth> if there are n possible sites, during map gen p from n may be available
20:05:23 <Alberth> besides first trying to generate 1 for each necessary industry
20:05:24 <andythenorth> whereas during gameplay, all n sites may be occupied
20:05:43 <Alberth> yeah, but depending on when the al plant is tried
20:06:09 <andythenorth> is the order deterministic? I think asked a while back and the answer was 'no'
20:06:19 <andythenorth> testing is boring
20:06:29 <Alberth> better read the code :)
20:06:39 <andythenorth> we need a test server for these things, like a nightly buildbot that runs reports on the major industry sets :P
20:06:49 <andythenorth> accessible via hudson on the devzone
20:07:38 <andythenorth> incidentally I've seen the oil rig issue in a FIRS game
20:07:45 <andythenorth> I thought I'd screwed up FIRS code, but perhaps not
20:09:03 <Alberth> I'd like to dig into these problems, but not now, too little time left today
20:09:22 <Alberth> we could make a wiki or have some PMs exchanged or so?
20:10:58 <andythenorth> or a forum thread
20:11:06 <Alberth> about the generation issue, afaik the industry can only tell the priority of getting build. 0 means do not build, but I don't have information when that changes. That makes planning for room for a new industry type complicated
20:11:19 <andythenorth> I think it's a thorny problem
20:11:29 <andythenorth> especially when you mix in availability
20:11:36 <andythenorth> I wonder about pre-seeding the entire map
20:11:42 <andythenorth> but there might be no storage for that
20:12:02 <andythenorth> determine in advance spots where industry x might appear
20:12:13 <andythenorth> it would need to be robust against numerous factors though
20:12:19 <andythenorth> terraforming, town growth, routes
20:12:20 <Alberth> oh, I have that idea too :)
20:12:36 <andythenorth> I'm 99% certain railroad tycoon did it that way
20:12:40 <Alberth> but you'll see many oil-rig issues then :)
20:12:56 <Alberth> I'd make great scenarios imho
20:13:02 <andythenorth> I'm not certain it's feasible by any means, but it's worth exploring
20:13:09 <andythenorth> it might throw up another idea
20:13:29 <Alberth> euhm 32 cargo-types, right?
20:13:46 <andythenorth> and no more required :)
20:14:06 <SmatZ> one bit per cargo type?
20:14:40 <Alberth> ie have 64x64 zones at the map, one bit per zone per cargo for yes/no
20:15:27 <SmatZ> so on a 128x128 map, there would be 4 the same 64x64 zones?
20:16:20 <Alberth> ie at 2048 I see no gain in being very precise
20:16:57 <Alberth> a zone of 32x32 is still nice
20:18:20 <Alberth> not sure that you actually want this for every cargo type though
20:18:50 <SmatZ> I thought it would be done for industry types, not cargo types
20:19:00 <SmatZ> but I didn't follow the discussion :x
20:19:31 <SmatZ> do you want the newgrf to somehow generate a 32x32 bitmap for each cargo type?
20:19:51 <SmatZ> that sounds very complicated
20:19:53 <Alberth> industry types could be used too, I was more thinking about having primary cargo available 'in the ground' at some places
20:20:08 <Alberth> no, I think it is part of a scenario
20:20:23 <Alberth> ie an additional restriction where to build some industry
20:20:51 <Alberth> ie a coal mine can only be build in a zone with coal :)
20:21:28 <SmatZ> on a 64x64 map, the rules would be too strict
20:21:40 <Terkhen> that sounds similar to that old suggestion of having different "zones" in the map, each one with different industries, houses and so on
20:21:59 <SmatZ> wasn't that actually implemented? at least for towns
20:22:06 <Alberth> SmatZ: why? nothing stops you from making all zones have coal?
20:22:33 <Alberth> it is just very precise
20:22:49 <andythenorth> this could be done by newgrf
20:23:01 <andythenorth> it would be arcane currently as no grf-global storage :P
20:23:23 <andythenorth> but using x and y it could be done
20:23:33 <SmatZ> Alberth: yeah, well :) I just have problems getting rid of splitting the map to "super-tiles" :)
20:23:47 <SmatZ> which would have the same size no matter how big the map is
20:24:03 <SmatZ> what you say the "super-tile" would be 1/32x1/32 of the map size
20:24:35 <Yexo> SmatZ: yes, I implemented that for towns/houses
20:24:35 <SmatZ> but I see the advantage
20:24:50 <SmatZ> Yexo: actually I didn't want to highlight you :)
20:24:55 <Alberth> you don't want to split?
20:25:21 <SmatZ> Alberth: thinking about that, your idea is better, since it scales fine with the map size
20:25:23 <Alberth> Yexo has a 6th sense :)
20:25:57 <Yexo> now that would be cool :)
20:37:17 <SmatZ> Alberth: maybe that would either increased or decreased the chance of creating industry that produces/accepts multiple cargos
20:37:44 <SmatZ> or you would have to compensate for that somehow
20:38:17 <SmatZ> like - both bits 1->it can be placed here
20:38:36 <SmatZ> one bit 1-> industry can be placed here if (Random() & 1)
20:38:42 <Alberth> I was thinking either bit
20:39:16 <SmatZ> then "livestock == 1" would imply "grain == 1", and vice versa
20:39:28 <Alberth> and to keep the coal mine example, it would need restrictions on produced cargo
20:40:50 <Alberth> I haven't given it much thought yet, perhaps & is better, or perhaps we need more bits to express other ideas
20:41:19 <SmatZ> with &, "livestock == 0" would imply "grain == 0", and vice versa
20:41:28 <SmatZ> but yeah, that idea sounds good :)
20:42:27 <lucas92> awesome, got it compiled fully with VC++
20:43:28 <Alberth> your idea of doing it for industry types may be better, at least it is more direct
20:44:00 <Alberth> lucas92: you spend an hour waiting for a compiler?
20:44:19 <lucas92> no, actually I didn't installed VC correctly
20:44:30 <lucas92> and installed the directx
20:44:51 *** DorpsGek changes topic to "1.1.1 | Website: *.openttd.org (translator: translator, server list: servers, wiki: wiki, patches & bug-reports: bugs, revision log: vcs, release info: finger) | Don't ask to ask, just ask | 'Latest' is not a valid version | English only"
20:46:16 <lucas92> so what are you working on?
20:46:56 <Rubidium> the most important thing to do right now ;)
20:47:21 <andythenorth> or like me, updating cashflow documents?
20:47:29 * andythenorth suspects Rubidium is probably doing cashflow
20:48:02 <Terkhen> after the freight game passenger only games are boring :/
20:48:35 <Terkhen> lucas92: regarding code, not much, it's been months since I coded anything "big" for OpenTTD
20:48:58 <Terkhen> there are a lot of people here with suggestions for coding if that's what you want :P
20:49:04 * Alberth messes with code segement 2AE1 of Opendune
20:49:19 <Ammler> add static grf support to the gui
20:49:33 <lucas92> ugh how am I supposed to know how lol
20:50:20 <Alberth> while true: read code, try changes, ask questions
20:50:36 <Terkhen> if you don't know what to code, play a game, after a while you will notice something that annoys you :)
20:50:39 <Terkhen> that's what I usually do
20:51:14 <Terkhen> after you get to know the code, you will also find stuff in the code that bothers you :P
20:51:21 <CIA-1> OpenTTD: frosch * r22529 /trunk/docs/ (32bpp.txt openttd.6): -Doc (r22419): Reflect removal of '-i' option in documentation.
20:51:57 <frosch123> someone wants to write a news post? :p
20:52:27 * andythenorth tries not to read source too much, there be dragons
20:53:02 * Alberth wonders how andy manages to change nfo code
20:53:29 <frosch123> Alberth: if the dragon is big enough, you cannot see it
20:53:34 <Terkhen> if openttd code has dragons, nfo has lovecraftian stuff
20:54:22 <Alberth> that's what I like about irc, you learn new words all the time :p
20:54:39 <Yexo> it gets scary when you realize you can read varaction2 code while only looking up what the variables are
20:54:42 <lucas92> the code is so huge! I get lost
20:55:07 <Terkhen> lucas92: I started by checking existing patches, that way you get to know a "single" part of the code and how to modify it
20:55:27 <lucas92> how do you check these?
20:55:29 <Terkhen> you can also check the commit log for commits related to what you want to do
20:55:30 <Yexo> same here, I started with a patch for waypoints
20:55:44 <Terkhen> there are many patches in the development forum and in the bug tracker
20:55:48 <Ammler> Zuu: [22:55] <MrD2DG> Openttd Auto update just failed :/
20:55:59 <Alberth> lucas92: that's normal, I haven't seen all code yet and I dig around in that code since 2007 or so
20:56:12 <Terkhen> yes, there are parts of the code that really have dragons :P
20:56:33 <Ammler> Zuu: [22:55] <MrD2DG> Just got a message "failed to get updated release list from internet"
20:57:26 <Alberth> lucas92: I started by reading code, and adding doxygen comments to undocumented parts
20:57:59 <Terkhen> meh, passenger only yacd is now boring, I'll start a new game tomorrow
20:58:11 <Alberth> then I turned that documented code upside down :p
20:58:21 <Ammler> Zuu: guess it is only for him, the status page works here
20:59:19 *** Devroush has joined #openttd
20:59:41 <lucas92> @param is for documentation?
21:00:01 <Yexo> you'll see it on docs.openttd.org
21:01:34 <lucas92> I see plenty of code improvements, for example, many functions have int types, while they could be const int types
21:02:08 *** Phoenix_the_II has quit IRC
21:02:15 <Yexo> in general the code doesn't use const for integer types for function arguments
21:02:33 <Yexo> it's just not done anywhere in the code
21:02:58 <Yexo> I doubt there would be a measurable performance benefit though
21:03:16 <Yexo> if there was, it would be a very good improvement
21:03:31 <Yexo> if there isn't, there would be no reason to change it
21:04:06 <lucas92> for example, the videoDriver class?
21:04:33 <lucas92> void MakeDirty(int left,int top,int width,int height)
21:04:53 * andythenorth was going to say nfo is trivial compared to real code
21:04:59 <andythenorth> there are no types or anything to deal with
21:04:59 <lucas92> could be void MakeDirty(const int left,const int top,const int width,const int height)
21:05:25 <Yexo> lucas92: if you can show (after testing) that adding const there would make the code faster, it'd be worth it
21:05:32 <Terkhen> andythenorth: that might be true, but it is still more confusing :)
21:05:55 <andythenorth> I am probably a bit special in my approach to coding :P
21:06:22 <andythenorth> also...quite a few times I read src, I find things that are not quite what they claim to be :P
21:06:29 <Terkhen> maybe I am too used to c++
21:06:53 <andythenorth> I may not have tried properly
21:07:15 <lucas92> you mean faster for compilation time or running time?
21:07:21 <fjb> cat /dev/random > code.nfo
21:07:40 <Terkhen> besides VB before starting university, and a small amount of java and php code in some classes, I have only coded c++ :)
21:07:48 <Yexo> for running time, that's the only thing that matters
21:08:08 * andythenorth has only coded actionscript, javascript, python, and nfo
21:08:09 <lucas92> yeah, const only optimize compilation time
21:08:21 <Terkhen> and testing performance on windows is... quite difficult to say the least :)
21:08:34 <Yexo> if it makes a real good improvement on compilation time that's welcome of course
21:08:35 * andythenorth forgot BBC Basic and RISC OS Basic :P
21:08:51 * andythenorth is monologuing and will go away
21:08:52 <Yexo> but it'd have to be significant (let's say 10% or so) for it to be worthwhile
21:08:59 <Terkhen> I started with BASIC too, on an old zx spectrum :)
21:09:13 <Terkhen> I barely understood the stuff, though
21:09:17 <lucas92> ok, I will take a look at it
21:09:20 <andythenorth> I coded animations, and some crappy games
21:09:25 <andythenorth> mostly I coded this though:
21:09:39 <andythenorth> which amused me when I was 12
21:09:45 <Terkhen> I copied examples from the book and messed them up until they did what I wanted
21:09:59 <Terkhen> which now that I think of it is surprisingly similar to what I do now :P
21:10:05 <andythenorth> there was an enhancement in my code above
21:10:14 <andythenorth> color = rnd (16)
21:10:24 <andythenorth> meh, forgot the line number
21:10:31 <andythenorth> what happened to line numbers? They were useful
21:10:42 <andythenorth> it's not at all an insane way to code
21:11:03 <Terkhen> you can use labels on c++ to use goto, but it is not "good code"
21:11:16 <Terkhen> code with gotos is quite difficult to review and understand
21:11:29 <frosch123> lucas92: "readability" is also a goal. so i am not a big fan of making every third word "const" :)
21:11:47 <Terkhen> unless for performance, if that can be proven :P
21:12:33 <__ln__> Terkhen: there's one sensible use case for goto
21:12:54 <Terkhen> which one? openttd has a few gotos anyways
21:13:48 <__ln__> avoiding having a dozen places in a function where the function may return;
21:14:31 <Terkhen> I thought that compilers optimized that at least in part
21:15:24 <__ln__> having a dozen points of exit is not very readable
21:15:42 <frosch123> __ln__: there is a second use case: continueing or breaking nested loops
21:22:29 <andythenorth> it used to be I could remember how to put the swearwords into an array and choose them at random :P
21:23:43 <lucas92> oh well, I think I'll drop the idea with the const keywords
21:24:23 * andythenorth has now written enough useful code today and should go to bed
21:24:32 <andythenorth> 3 lines of highly useful code :P
21:24:40 <Terkhen> good night andythenorth
21:24:51 <andythenorth> maybe I should GPL that program :P
21:25:41 <Terkhen> if you gpl it someone will try to patch the foul language away :(
21:25:44 <andythenorth> I should keep it just for myself?
21:25:53 <andythenorth> it's pretty clever :P
21:26:13 * andythenorth wonders if today's generation of script kiddies have done stuff like that
21:26:26 <Terkhen> "free for any use, as long as the swearing is kept without changes"
21:36:54 <lucas92> Some other thing I see in the code that could be improved for the code readablility is the position
21:37:07 <lucas92> for example, void DrawVehicleEngine(int left, int right, int preferred_x, int y, EngineID engine, PaletteID pal)
21:38:11 <lucas92> could become void DrawVehicleEngine(Vector2 position, Vector2 prefered_position, EngineID engine, PaletteID pal)
21:38:39 <lucas92> Vector2 would be a utility class that defines a 2D vector
21:38:50 <lucas92> with a x and y coordinate
21:39:11 <Terkhen> it looks more or less the same to me, besides, int left and int right are not a position
21:39:59 <Zuu> Ammler: It dose work from here. Either a temporary problem at his side or with users.tt-forums.net.
21:40:10 <Terkhen> IIRC they are the space you have for drawing
21:40:24 <Ammler> Zuu: yes, loading the status page took also long for him
21:40:36 <Terkhen> anyways, code readability changes are not a big priority, unless they are in *really* confusing parts of the code
21:40:38 <Ammler> maybe you have a kind of timeout?
21:40:48 <Zuu> I think so. I don't remember what it is.
21:40:50 <Terkhen> I'm too sleepy to remember an example, though :P
21:41:37 <Zuu> But there is a timeout so that it shouldn't take too long to start it when you are offline.
21:42:55 <Zuu> After it times out with users.tt-forums.net it also tries the secondary server which apparently he also had problems to reach.
21:43:23 <Zuu> The secondary server is at my web host here in Sweden.
21:47:22 <lucas92> what with the FORCEINLINE macro? What is its use?
21:49:41 <Yexo> it expands to __forceinline for msvc
22:10:45 *** KritiK_ has joined #openttd
22:12:11 *** KritiK_ is now known as KritiK
22:12:48 <Eddi|zuHause> hmm... 1.1.1 before midnight?
22:15:47 <Terkhen> that way we can go earlier to bed :)
22:46:46 *** Prof_Frink has joined #openttd
22:50:54 *** Brianetta has joined #openttd
22:51:20 <fjb> Hm, configure doesn't find liblzma.
23:00:56 <fjb> Ah, liblzma has no metadata file for pkg-config. And configure relies on pkg-config.
23:17:33 *** Eddi|zuHause has joined #openttd
23:22:56 <rplnt> Hello, I'm interested in the format of the save file (especially the land/map part) and how could I read it. Could someone point me to the right part of the documentation and/or part of the source tree (I found some load/save file but wasn't quite dure if it is it). Thanks
23:28:45 <Eddi|zuHause> it's a RIFF format that is compressed.
23:29:12 <Eddi|zuHause> the structure of each chunk is described in the files in src/saveload/
23:29:25 <Eddi|zuHause> the map structure is described in docs/landscape.html
23:30:05 <Eddi|zuHause> for easier testing, you can make uncompressed savegames with an option in openttd.cfg
23:30:26 <Eddi|zuHause> "savegame_format = none" or similar
23:37:32 <rplnt> Eddi|zuHause: thank you
23:56:05 *** yorick is now known as Guest2996
continue to next day ⏵
