IRC logs for #openttd on OFTC at 2009-09-12
⏴ go to previous day
00:20:16 *** Dreamxtreme has joined #openttd
01:26:05 *** Progman_ has joined #openttd
01:34:00 *** Progman_ is now known as Progman
01:42:08 *** De_Ghosty has joined #openttd
02:02:45 *** Biolunar has joined #openttd
03:04:19 *** stuffcorpse has joined #openttd
04:41:04 *** ChanServ sets mode: +v tokai
06:01:49 *** [com]buster has joined #openttd
06:19:30 *** ChanServ sets mode: +v tokai
06:31:26 *** Alberth has joined #openttd
07:20:37 *** Cybertinus has joined #openttd
07:37:03 *** Polygon has joined #openttd
07:43:36 *** andythenorth has joined #openttd
08:06:58 *** bubastis has joined #openttd
08:10:07 *** Aankhen`` has joined #openttd
08:23:59 *** tux_mark_5 has joined #openttd
08:26:31 *** Progman has joined #openttd
08:34:27 *** andythenorth has joined #openttd
08:46:36 *** Timitry_ has joined #openttd
08:48:24 *** Timitry_ has joined #openttd
09:01:31 *** nicfer1 has joined #openttd
09:08:29 *** Chris_Booth has joined #openttd
09:09:54 *** fonsinchen has joined #openttd
09:24:02 *** Terkhen has joined #openttd
09:29:45 *** oskari89 has joined #openttd
09:31:21 *** Vikthor has joined #openttd
09:33:09 *** [com]buster is now known as Combuster
09:54:06 *** |Jeroen| has joined #openttd
09:55:17 *** Doorslammer has joined #openttd
10:02:11 *** worldemar has joined #openttd
10:07:47 *** Belugas has joined #openttd
10:07:48 *** ChanServ sets mode: +o Belugas
10:22:13 <fonsinchen> I want to save/load nested containers, for example a std::map<StationID, std::list<CargoPacket *> >.
10:22:31 <Zuu> Up untill about a month ago I hadn't been drinking coffe more than 2-3 times in more than one year. I had copletely stoped drinking it. But now I started again and I drink more a day than I did before. I'm already on my third cup today. :-D
10:22:35 <fonsinchen> Am I the first one with that problem, or has anyone else ever thought about that?
10:25:35 <CIA-4> OpenTTD: alberth * r17507 /trunk/src/order_gui.cpp: -Codechange: Make some methods of the OrdersWindow non-static.
10:37:09 <fonsinchen> In fact I have a solution, which is procedurally loading/saving the stuff in some deterministic order. But that's not so well readable. It would be nicer to have some way of representing that in the SaveLoad struct, like SLE_LST ... not quite trivial, though.
10:38:10 *** andythenorth has joined #openttd
10:52:20 *** worldemar has joined #openttd
11:20:03 * Zuu wished he had a pen with ink that drys more quickly
11:23:35 <Zuu> Doesn't show up as good in a scanner. But hopefully I will be done early enough so that I can hand it in on monday before the door to the post boxes locks. :-)
11:24:33 *** BaronChaos has joined #openttd
11:27:00 *** Azrael- has joined #openttd
11:33:56 *** frosch123 has joined #openttd
11:37:32 *** KenjiE20 has joined #openttd
11:51:18 *** B_Chaos has joined #openttd
12:18:12 *** Dred_furst has joined #openttd
12:22:17 *** bubastis has joined #openttd
12:33:18 *** Aankhen`` has joined #openttd
12:36:43 <frosch123> better call the police directly
12:37:02 <Shhup> A friend of mine can't connect to my server on Internet
12:37:12 <Shhup> It's says "1 Client in front of us"
12:37:44 <Rubidium> so someone else is trying to join your server at the moment
12:37:54 <glx> with a slow download speed
12:37:56 <frosch123> you were fast enough :)
12:38:38 <petern> okay, what's the intel core i5 stuff?
12:38:47 <petern> and why does the core i7 come in two different sockets?
12:39:47 * petern ponders phenom instead
12:41:22 *** Dred_furst has joined #openttd
12:41:29 <Rubidium> ooh... Intel restarted their processor numbering again?
12:41:37 <Rubidium> time to start selling those i386 procs :)
12:42:01 <petern> i guess that i5 is like the celeron version
12:42:11 <petern> then they introduce another new socket tyep
12:42:17 <KenjiE20> i5 is our current C2x series afaik
12:42:20 <petern> AND make a cut down i7 that goes in that socket
12:43:19 <frosch123> i5 sounds like 90mhz pentium
12:43:49 <KenjiE20> sorry yes, i3 is current C2
12:47:33 <frosch123> so it's basically dual-channel vs tri-channel memory
12:47:50 <glx> <@petern> and why does the core i7 come in two different sockets? <-- worse 3 sockets
12:48:28 <petern> glx, i can safely ignore mobile variants
12:48:33 *** Dred_furst has joined #openttd
12:53:05 <petern> "If anything, the QPI on Bloomfield is a performance hindrance for a single socket platform and is totally unnecessary. It's just one more interconnect between the CPU and the video cards that isn't present on the Lynnfield cpu with it's PCIe integrated on die."
12:54:46 <petern> maybe a C2D E8400 is still good enough
12:55:56 <Aali> the E8400 is good enough if you overclock it :)
12:57:01 <CIA-4> OpenTTD: rubidium * r17508 /trunk/src/ (base_station_base.h station.cpp station_base.h): -Fix [FS#3195] (r16859): join station window didn't get updated when stations were removed from the pool
13:06:57 *** Chris_Booth has joined #openttd
13:07:40 *** HerzogDeXtEr has joined #openttd
13:27:00 *** Chris_Booth is now known as Guest2219
13:27:07 *** Chris_Booth has joined #openttd
13:33:41 *** worldemar has joined #openttd
13:49:03 *** andythenorth has joined #openttd
14:31:27 *** bubastis has joined #openttd
14:37:41 <CIA-4> OpenTTD: rubidium * r17509 /trunk/src/vehicle_cmd.cpp:
14:37:41 <CIA-4> OpenTTD: -Fix [FS#3196]: if building a part fails during cloning, sell what was already cloned instead of leaving it 'for free'
14:37:41 <CIA-4> OpenTTD: -Fix: make cloning multiheaded trains possible with with 'max - 1' vehicles existing
14:40:38 <CIA-4> OpenTTD: frosch * r17510 /trunk/src/settings.cpp: -Fix [FS#3197]: When loading GRFConfigs from ini file, validate them wrt. duplicate GRF IDs.
14:41:54 *** De_Ghosty has joined #openttd
14:56:54 *** Bubastis_weg has joined #openttd
14:58:46 *** `Fuco`` has joined #openttd
14:59:52 <Zuu> An interesting thing of traffic theory is to calculate the speed of zero cars. :-)
15:00:31 <Zuu> Which may happen to be eg. 70 km/h
15:01:55 <frosch123> it will be difficult to prove you wrong
15:02:58 * Rubidium proposes it to be `c'
15:04:17 <Zuu> The speed of zero cars is usually described as the free flow speed and is basically the speed you drive if you are alone on the road.
15:04:54 <Rubidium> oh, then it's like 20 km/h :)
15:13:03 <frosch123> wouldn't that be the speed of one car?
15:15:56 <Rubidium> that's how fast I'd drive when there're no cars on the road
15:16:07 <Zuu> Well no, because to calculate the free flow speed you set the density to zero. (Density has the unit [vehicles/length unit])
15:16:31 <frosch123> Rubidium: but hovercrafts are faster
15:16:54 <Zuu> So to be really correct you would have to say zero cars per kilometer for example.
15:25:52 *** Chris_Booth is now known as Guest2227
15:25:58 *** Chris_Booth has joined #openttd
15:26:41 *** DR_Jekyll has joined #openttd
15:39:58 *** TheMask96 has joined #openttd
16:07:47 *** ChanServ sets mode: +v tokai
16:28:47 *** ChanServ sets mode: +v tokai
16:34:06 *** HerzogDeXtEr1 has joined #openttd
16:56:48 *** fonsinchen has joined #openttd
16:57:07 *** fonsinchen has joined #openttd
17:17:20 <CIA-4> OpenTTD: frosch * r17511 /trunk/src/newgrf_gui.cpp: -Codechange: Simplify ShowNewGRFInfo() by passing 'right' instead of 'width'.
17:23:40 *** worldemar has joined #openttd
17:29:30 *** Coco-Banana-Man has joined #openttd
17:32:36 <CIA-4> OpenTTD: rubidium * r17512 /trunk/src/crashlog.cpp: -Add: some basic information about the loaded AIs in the crash log
17:35:14 *** Illegal_Alien has joined #openttd
17:45:32 <CIA-4> OpenTTD: translators * r17513 /trunk/src/lang/ (6 files in 2 dirs): (log message trimmed)
17:45:32 <CIA-4> OpenTTD: -Update from WebTranslator v3.0:
17:45:32 <CIA-4> OpenTTD: catalan - 1 changes by arnau
17:45:32 <CIA-4> OpenTTD: simplified_chinese - 39 changes by EraserKing
17:45:32 <CIA-4> OpenTTD: dutch - 18 changes by habell
17:45:34 <CIA-4> OpenTTD: greek - 55 changes by fumantsu
17:45:34 <CIA-4> OpenTTD: indonesian - 60 changes by prof
17:58:34 *** Doorslammer has joined #openttd
18:00:44 *** kamikaz333 has joined #openttd
18:02:25 <kamikaz333> Is it possible to run openttd on windows 7 ?
18:06:15 <Rubidium> not within such short time :)
18:14:17 <Alberth> maybe he realized it is more efficient to simply try it :)
18:24:32 <Jonis> I have a question, does anyone have a good 64x64 map?
18:24:35 *** zachanim1 has joined #openttd
18:25:45 <Jonis> The terrain should be hard and not to many cities or industries.
18:26:07 <Doorslammer> You aren't going to have much of either anyhow
18:26:32 <Rubidium> 1 town, 0 industries?
18:27:13 <frosch123> Jonis: so open the new game window, turn of the shore on all sides, set the terrain to mountainious, set the number of towns to 2, and similiar for the industries
18:27:14 <Zuu> I played a nice 64x124 multiplayer game once.
18:27:22 <Zuu> We were 4-5 players on that map.
18:27:44 <Jonis> How do i turn of the shore?
18:28:19 <Zuu> Jonis: Which version do you run? We asume you run at least 0.7.x
18:28:21 <frosch123> at the bottom are four buttons
18:29:09 <Jonis> I run 0.6.3 but I will just install the latest version.
18:30:22 <Zuu> http://finger.openttd.org will keep you informed when a new release appears. Or if you don't want to visit that page (which is not really human-friendly) use a program that does that for you and tell you when it is time to upgrade.
18:30:57 <Jonis> Nice, now I know how to turn off the shores
18:30:57 <frosch123> Zuu: there is also a RSS feed
18:31:24 <Zuu> frosch123: Didn't knew there was an RSS for just releases.
18:31:39 <frosch123> i guess it is for all news on the front page
18:31:48 <frosch123> but i do not use it :)
18:31:52 <Jonis> I'm going to add the RSS feed to my mobilephone :D
18:31:57 <Zuu> Still not to much spam there. (on the news page)
18:32:53 <frosch123> well, usually 4 in two months
18:33:14 <frosch123> except in winter :p
18:33:38 <Zuu> Yea, for stable releases which doesn't happen that often, a update tool is a bit of an overkill if you don't already have it for nightlies and all other stuff.
18:40:18 <CIA-4> OpenTTD: frosch * r17514 /trunk/src/newgrf_gui.cpp: -Codechange: Nestify Add-NewGRF dialog.
18:46:35 <jonty-comp> the multiplayer cheating page on the wiki encourages cheating
18:48:34 <jonty-comp> "See if there is an old, deserted company around, and use the share dealing exploit (above) to generate mass funds. Use these to perform land reconstruction."
18:48:43 <jonty-comp> although, I suppose the other person is already cheating
18:52:22 <Jonis> Even when I have but in the number of cities I still only get on large city with like 2000 in population
18:53:54 <Rubidium> jonty-comp: luckily the sharing trick doesn't work on like 99% of the servers
18:54:07 <jonty-comp> Rubidium: not anymore, at least
18:55:05 <Rubidium> but the document illustrates *why* an active admin is needed
18:56:06 *** [com]buster has joined #openttd
18:58:40 <Jonis> How close to the edge can I build a city?
19:00:12 <Zuu> Jonis: Perhaps try it out in the scenario editor?
19:02:58 <Jonis> Discovered a funny thing, I couldnt start build the city to close to the edge but once I had placed it I could expand it to the last tile on the map.
19:25:32 <Zuu> That is probably because the location of a town is the tile with the town name sign on it.
19:26:13 <Zuu> That is always concidered the center of the town, regardless of where the houses and roads are.
19:35:57 *** De_Ghosty has joined #openttd
19:41:43 <fonsinchen> In all known implementations of the standard C++ library multimaps have a deterministic ordering among entries with equal keys. Unfortunately this is not required by the standard. However, it will be required by C++0x. I think I'll just continue relying on that behaviour for cargodist's optimization of cargo loading. The question remains how to load/save (multi)maps in general. My current implementation is an ugly hack ...
19:48:46 <Alberth> a desync in the making...
19:51:21 <fonsinchen> I don't think they'll create any more standard C++ libraries before C++0x
19:51:47 <R0b0t1> Anything new on that yet?
19:51:53 <fonsinchen> and all current multimaps are based on red-black trees which have that properties.
19:52:11 <Noldo> R0b0t1: well it will be C++1x
19:52:12 <Rubidium> well, when will all compilers that we allow OpenTTD to be compiled with follow C++0x?
19:52:24 <fonsinchen> that's not the problem
19:52:37 <Rubidium> Noldo: nah, it'll probably becomes C++0F or so
19:52:53 <fonsinchen> in fact all known implementations behave just the way I need them to. It's just not in the standard.
19:53:23 *** De_Ghosty has joined #openttd
19:53:46 <Rubidium> still "known" is quite fuzzy
19:54:21 <fonsinchen> some people proposing that for C++0x examined that.
19:56:17 *** DaleStan has joined #openttd
20:01:27 <Rubidium> fonsinchen: what in that says that all known implementations behave like the proposal?
20:04:44 <Jonis> What are you discussing?
20:05:18 <fonsinchen> Stability of equal ranges in multimaps across different implementations of the C++ standard library
20:05:26 <Rubidium> "A brief survey of current library implementations reveals differing behavior" (n1780.html)
20:06:20 <Jonis> I don't understand too much of it. :P
20:09:33 <Rubidium> Jonis: don't worry, you're in the majority 'group'
20:10:37 <fonsinchen> That's only with the hint though. The later survey of insertion without hint shows they all behave the same.
20:11:22 <fonsinchen> And I don't have to use hints.
20:14:49 <Jonis> Good, then I can go to sleep tonight knowing that I'm not totally stupid
20:16:17 * Rubidium wonders what STL implementation MSVC uses
20:16:47 <Rubidium> it at least isn't STLport or gcc
20:22:10 <fonsinchen> I'll compile and test this example with MSVC now. Place your bets, please.
20:22:16 *** worldemar has joined #openttd
20:22:49 <Rubidium> well, it's more about the: "all known implementations" when MSVC's implementation has been left out
20:23:05 *** `Fuco`` is now known as Fuco
20:23:55 <Rubidium> whether it works in MSVC or not, "all known implementations" has been proven to be not the truth
20:24:27 <fonsinchen> Maybe I should roll my own multimap.
20:25:11 <Rubidium> well, I don't know what's best
20:25:33 <fonsinchen> I could make that a map<StationID, list<CargoPacket *> >
20:26:00 <fonsinchen> but that would immensely complicate the code. I don't think it's worth it.
20:28:20 <Rubidium> maps can't be directly stored either, so for savegame purposes it doesn't matter
20:31:54 <CIA-4> OpenTTD: rubidium * r17515 /trunk/src/3rdparty/squirrel/squirrel/sqvm.cpp: -Fix [FS#3189]: [Squirrel] In some cases the call stack would not be cleaned up properly during crash handling. Occasionally this causes asserts to be triggered or crashes.
20:33:17 <Jonis> If I have one station near the sea and one up on the mountain but it's just a straight line between them. When should I start the slope? As early as possible or as late as possible?
20:35:05 <frosch123> try to make it smooth, don't terraform much, build tunnels and bridges instead :)
20:35:53 <frosch123> better build less straight than taking 3 slopes at once
20:37:59 <Jonis> But its a big height different, need slopes. But don't know if I'm going to build it late so the train can accalerate before it or take it at the start.
20:41:22 <Zuu> Jonis: Make a test build two tracks side by side where you have the slop early on one and the other have the slope later. That is the wonder of having a simulation environment.
20:44:24 <CIA-4> OpenTTD: rubidium * r17516 /branches/0.7/ (6 files in 4 dirs):
20:44:24 <CIA-4> OpenTTD: [0.7] -Backport from trunk:
20:44:24 <CIA-4> OpenTTD: - Fix: Road vehicles could get lost when the prelimiary destination (for the pathfinder heuristics) is unreachable [FS#3188] (r17491)
20:44:24 <CIA-4> OpenTTD: - Fix: Silence compile warning that is only triggered when you are using a broken iconv (r17479)
20:44:26 <CIA-4> OpenTTD: - Fix: In some cases compilation failed because the search path for ICU was not added to the ldflags (r17476, r17474)
20:44:26 <CIA-4> OpenTTD: - Document: Compiler warning when compiled without asserts in GCC 4.4+ is not our fault but a GCC bug. Includes a reference to the appropriate GCC bug report (r17461)
20:55:41 *** Chris_Booth has joined #openttd
20:57:39 *** Terkhen has joined #openttd
20:59:47 <CIA-4> OpenTTD: rubidium * r17517 /branches/0.7/src/ (6 files in 3 dirs):
20:59:47 <CIA-4> OpenTTD: [0.7] -Backport from trunk:
20:59:47 <CIA-4> OpenTTD: - Fix: [Squirrel] In some cases the call stack would not be cleaned up properly during crash handling. Occasionally this causes asserts to be triggered or crashes [FS#3189] (r17515)
20:59:47 <CIA-4> OpenTTD: - Fix: When loading GRFConfigs from ini file, validate them wrt. duplicate GRF IDs [FS#3197] (r17510)
20:59:48 <CIA-4> OpenTTD: - Fix: When building a part fails during cloning, sell what was already cloned instead of leaving it "for free". Also make cloning multiheaded trains possible with with "max - 1" vehicles existing [FS#3196] (r17509)
20:59:48 <CIA-4> OpenTTD: - Fix: The wrong value was restored to SetAllowDoCommand possible resulting in an AI that was not allowed to do any actions (r17500)
21:14:35 *** tux_mark_5 has joined #openttd
21:55:11 *** Brianetta has joined #openttd
21:57:56 *** Dreamxtreme has joined #openttd
21:59:25 *** DaleStan_ has joined #openttd
21:59:25 *** DaleStan is now known as Guest2261
21:59:25 *** DaleStan_ is now known as DaleStan
22:00:20 *** Nickman_87 has joined #openttd
22:22:34 *** tux_mark_5 has joined #openttd
22:29:46 *** tux_mark_5 has joined #openttd
22:33:45 *** Nite_Owl has joined #openttd
22:38:27 *** Nite_Owl has joined #openttd
22:47:51 *** tux_mark_5 has joined #openttd
22:50:47 *** tux_mark_5 has joined #openttd
22:53:52 <Nite_Owl> it appears that Skynet has become self aware of tux_mark_5
23:04:37 *** tux_mark_5 has joined #openttd
23:06:29 *** th1ngwath has joined #openttd
23:17:33 *** Progman has joined #openttd
23:33:14 *** Eddi|zuHause has joined #openttd
23:52:13 *** ChanServ sets mode: +v tokai
23:52:26 *** KenjiE20 has joined #openttd
continue to next day ⏵