IRC logs for #openttd on OFTC at 2009-01-12
⏴ go to previous day
00:06:34 <CIA-1> OpenTTD: rubidium * r14999 /trunk/src/settings_gui.cpp: -Change [FS#2523]: move service at helipad to the servicing options
00:33:09 *** Eddi|zuHause has joined #openttd
01:02:00 <Zuu> nooooo, nightly forum backup, just as I am about to round off my day by eating an apple and read the dev section of tt-forums :(
01:05:18 <Aali> yeah, that thing always comes at the worst possible time
01:09:00 <Zuu> Though it is nice that backup is taken, as tt-forums is half my life :D
01:09:49 <snappy> When will a steel mill produce steel? I'm transporting iron ore to it.
01:12:07 <Zuu> snappy: Do you use PBI/ECS?
01:12:46 <Zuu> In those you need coal and iron ore to produce steel.
01:13:16 <snappy> i have no idea what a PBI or ECS means? but the steel mill says it "requires iron ore"
01:13:40 <Zuu> snappy: Have you loaded any GRF files?
01:13:49 <snappy> yep, from the original ttdlx
01:14:01 <petern> probably the usual suspect: transfer orders
01:14:22 <Zuu> I even had a friend asking me about it on MSN...
01:15:22 <Aali> snappy: not all tiles need coal, but the steel mill as a whole does
01:15:31 <Aali> build your station closer
01:16:13 <Zuu> If you click on your station, does it says that it accept iron ore?
01:17:18 <snappy> it says it only accepts iron ore
01:17:37 <snappy> so i need to cover the whole steel mill?
01:17:55 <snappy> probably explains why i couldn't do this last time i played ttdlx (in like '97)
01:17:58 <Zuu> snappy: As you have not loaded any NewGRFs (which you probably don't know what it is) your steel mill only need iron ore, no coal.
01:18:34 <Zuu> snappy: You need to cover the right tile(s) of the industries.
01:20:16 <Zuu> snappy: But if your station sasy it accepts iron ore then that part should be okay.
01:20:52 <Zuu> What order have you given your train at the station next to the steel mill?
01:21:14 <snappy> ive been carrying iron ore by trucks
01:21:28 <Zuu> ok, by trucks, but what order did you gave them?
01:21:46 <Zuu> did you used the transfer order?
01:21:48 <snappy> basically full load for iron ore, then goto steel mill
01:22:17 <Zuu> You shouldn't be using the transfer order, so that is okay.
01:24:10 <Zuu> snappy: do you use English OpenTTD or a translation?
01:25:07 <Zuu> Ok, then we can be sure you don't use transfers. (a friend used the Swedish translation and there transfer has been translated, despite that there is no Swedish word for transfer)
01:25:42 <Zuu> snappy: Maybe you can post a savegame?
01:26:44 <snappy> yah sure. But I just don't understand the semantics or the gameplay in this respect. Take for example the factories. Once you feed them livestock, grain, and steel do they start producing goods?
01:26:46 <Zuu> Oh, and you used truck depots (not bus stops)?
01:27:42 <Zuu> Yep in plain TTD (and plain OpenTTD) you only need one of the livestock, grain or steel.
01:27:58 <Zuu> Then goods will start being produced.
01:28:44 <Zuu> Amount of produced goods is proportional to the amount of supplied livestock, grain or steel.
01:29:33 <SmatZ> in TTO, you didn't need to supply anything ;)
01:30:59 <Zuu> snappy: I need to get myself some sleep, but post it to tt-forums OpenTTD problems and I'll take a look on it tomorrow if nobody has done that yet by then.
01:32:13 <Zuu> you're "snappy" also on the forums, right?
01:33:42 *** roboboy has joined #openttd
02:00:21 *** DephNet[Paul] has joined #openttd
02:23:42 *** Roujin_ has joined #openttd
02:26:03 <Eddi|zuHause> AttributeError: 'str' object has no attribute '__cmp__' <- how the hell did i break that one?
02:29:24 *** Roujin_ is now known as Roujin
02:33:18 *** Eddi|zuHause has joined #openttd
02:49:46 *** energetic has left #openttd
03:05:47 *** TinoDidriksen has joined #openttd
03:36:10 *** TinoDidriksen has joined #openttd
03:40:44 *** energetic has joined #openttd
04:01:35 *** welshdragon has joined #openttd
05:38:32 <keiya> OpenTTD supports side-scroll!
05:39:00 <keiya> That means, in 'scrollwheel pans', I can do doubletouch scrolling easily! AWESOME!
05:42:11 <keiya> My little iBook has its advantages ^_^
05:44:06 <Forked> it's nice to look at? =p
05:45:04 <goodger> I liked the ones that looked like purple clams
05:47:14 <Forked> just ignore me.. I'm still thinking that macs are good for things I don't do .. photoshop-related stuff and video editing etc.. also komplett.no (one of, if not the biggest online computer stores here in Norway) had an ad saying something like "making the house looking better for xmas? buy a mac!"
05:47:35 <Forked> so when the stores that sell them can't come up with anything better than that they look good .. =p
05:48:23 <Rubidium> and they're vastly more expensive than your non-mac counterpart
05:49:27 <Forked> What I can imagine is pretty damn good about a mac.. is that there is no home-made setups of the hardware, so the OS can be finetuned to the small selection of hardware that is available. Therefor more stable than an operating system that has to work with every combination possible
05:50:06 <goodger> Forked: the number of drivers are reduced, yes
05:50:25 <goodger> flexibility goes out the window
05:50:36 <goodger> you want a laptop? you can have your choice of six. in two colours
05:50:37 <Rubidium> as long as they deliver an OS that breaks applications due to mysterious reasons I'm not buying it
05:51:56 <keiya> Rubidium: So you avoid Windows like the plague?
05:52:00 <Rubidium> it's one of the few OSes where an application that works fine in version X.Y fails in X.Y+1
05:52:36 <goodger> Rubidium: apple and microsoft have different philosophies
05:52:46 <keiya> That's true of /all/ systems, Rubidium.
05:53:34 <Rubidium> keiya: just install a new videocard driver for your nvidia thingy and see how OpenTTD becomes massively slower
05:53:38 <goodger> microsoft know that their DEVELOPERS DEVELOPERS DEVELOPERS DEVELOPERS have paid a lot of money to make software for windows and will not bother to update stuff or might even move to a less horrible platform if things break
05:53:56 <Rubidium> or how a universal binary can't be build for OSX 10.3 and OSX 10.5
05:54:02 <keiya> Eh? nvidia thingy? There's no nvidia thingy in this one :P
05:54:16 <Rubidium> whereas for Windows a binary made for Windows 2000 works fine for Windows Vista
05:54:28 <goodger> Rubidium: windows 1.0 binaries work on windows vista...
05:54:55 <goodger> apple, on the other hand, don't give two shits, because they know that a certain type of person will unceasingly update software in order to be part of the apple club
05:54:59 <keiya> And those are running under two or three layers of emulation, anyway :P
05:55:18 <Rubidium> keiya: and OSX is free of that... don't make me laugh
05:55:21 <keiya> Anywho, this laptop was /free/, so I can't really complain.
05:55:34 <goodger> Rubidium: actually they dropped classic in 10.4
05:55:39 <keiya> Rubidium: Eh, it's got less of it than in OS 9
05:55:53 <keiya> They dropped classic in 10.4 /Intel/.
05:56:00 <keiya> It's still here in 10.4 /PPC/.
05:56:09 * Forked ponders about throwing a "what linuxdistro is the best?" out there :)
05:56:24 <goodger> they aren't making PPC machines any more, therefore they dropped it at 10.4
05:56:36 <keiya> They didn't want to run an m64k emulator on a ppc emulator, basically :P
05:56:52 <keiya> So I'm /imagining/ chatting to you?
05:56:57 <Rubidium> keiya: in 10.4 your bash is always the PPC which on x86 means it runs on Rosetta, i.e. a PPC emulator
05:57:09 <Forked> Rubidium: interesting. I shall read up on that project.
05:57:23 <keiya> Rubidium: Yes... it would be /if I were on intel/.
05:57:31 <goodger> keiya: no, you're using quite old hardware
05:58:03 <goodger> if you were using any hardware from the era of 10.4 onward, you'd be on an intel
05:58:39 <goodger> apple dropped classic in 2006-or-whenever by moving to intel, and not porting classic to intel
05:58:54 <keiya> You claimed there was none.
05:59:04 <keiya> I am using one, thus you were incorrect.
05:59:20 <goodger> "there was none" --- what?
05:59:24 <keiya> You did /not/ say they no longer manufacture ppc hardware (which would be mostly correct)
05:59:59 <goodger> there was no classic in any new macintosh starting with the release of os x 10.4
06:00:53 <keiya> But OS 9 and OS X are /totally/ different, anyway.
06:01:14 <goodger> when was that a part of the conversation?
06:01:28 <goodger> you can't just whip that commonly-known statement out and say "I win!"
06:01:31 <Forked> well as long as you all know vista is better than any osx :-)
06:01:32 <keiya> You claimed "software doesn't work from one version to the next"...
06:01:48 <keiya> Forked: Heh. OS X actually /runs/ :P
06:02:08 <goodger> keiya: actually rubidium claimed that
06:02:22 <Forked> keiya: what do you mean "sometimes"?
06:02:41 <keiya> Well, anyway, a more accurate comparison would be DOS programs on NT, without the layers and layers of compatability hacks.
06:03:02 <goodger> keiya: a more accurate comparison to what?
06:03:22 <keiya> (The NT kernel doesn't impliment /any/ of the same interfaces as pre-95 kernels. In 98, they introduced an abstract driver system which is still supported.)
06:03:46 <goodger> you're thinking of implement
06:03:58 <Rubidium> keiya: OSX actually runs? It failed to install when I tried
06:04:06 <keiya> Rubidium: On what hardware?
06:04:19 <goodger> keiya: OS 9 to OS X is much more different than win9x to winNT
06:04:31 <Rubidium> well... not a piece of expensive mac hardware
06:04:47 <keiya> goodger: Yes, but not from the Win9x kernel to the NT kernel.
06:05:00 <Tefad> the APIs were the same.. "win32"
06:05:04 <keiya> NT didn't impliment the win32 api, originally... it's hacks on top of it.
06:05:12 <Tefad> a security model was added.
06:05:17 <keiya> Tefad: Actually, that's all implimented above the kernel, now.
06:05:18 <Tefad> win9x had none. only profiles
06:05:47 <Tefad> there are many many implementations of the win32 API
06:05:48 <keiya> The core NT and 9x kernels have /literally/ nothing in common :P
06:05:49 <goodger> what are we arguing about here?
06:06:01 <goodger> well, no, that's bollocks
06:06:02 <keiya> I think we've gotten lost :P
06:06:17 <goodger> there are plenty of parts of the modern windows kernel that haven't been changed since the 80s
06:06:19 <Tefad> win32 even existed for windows 3.11
06:06:35 <keiya> goodger: Oh-so-very false.
06:06:47 <keiya> The layer applications see hasn't changed that much
06:07:01 <Tefad> goodger: i don't think microsoft has retained much of its 16bit asm optimizations...
06:07:32 <keiya> The NT kernel and DOS are totally separate. You might as well say Microsoft Office and StarOffice have code in common :P
06:08:02 <Rubidium> keiya: you know we are actually considering removing support for OSX < 10.5 from OpenTTD?
06:08:24 <keiya> Yeah, and when you do I'll just keep running the last version that handles it. I'm used to that :P
06:08:36 <Tefad> support for beos but not os x?
06:08:41 <goodger> if the kernel was created from scratch for NT, why does it have such appalling security, process management, user management, memory management, filesystem, etc?
06:08:57 <keiya> goodger: Because they had to hack in support for the win32 api.
06:08:58 <Rubidium> OSX gets incompatible APIs, BeOS doesn't
06:09:02 <Tefad> goodger: because microsoft disables security to be compatible with win9x
06:09:13 <Tefad> they're gradually turning it on every windows release since 2k
06:09:16 <keiya> Rubidium: Mostly because BeOS hasn't been updated since the dark ages...
06:09:33 <goodger> apple has illustrated very thoroughly that it is possible to start from scratch and still support legacy software
06:09:44 <Rubidium> even Windows and most Unices have stable APIs
06:09:49 <Tefad> microsoft has problems with that.
06:09:56 <Tefad> they reinvent the wheel too often
06:10:08 <keiya> goodger: Mostly by running emulation...
06:10:13 <goodger> Rubidium: OSS unixes tend to have quite unstable APIs because their software is open-source
06:10:17 <Tefad> recoding the win32 API i mean
06:10:20 <goodger> keiya: not processor-level emulation
06:10:29 <Rubidium> stable meaning that documented API functions are kept
06:10:30 <keiya> goodger: Actually yeah :P
06:10:48 <keiya> A lot of the OS 9 core still uses m64k features :P
06:11:04 <Tefad> Rubidium: i think it's a ploy to make mac users spend more money to upgrade their software from year to year
06:11:06 <Eddi|zuHause> hm... i can't shut down a windows computer from a remote session...
06:11:12 <Tefad> does anything made for 10.3 work in 10.5??
06:11:24 <goodger> Eddi|zuHause: you have to run shutdown from cmd
06:11:34 <Rubidium> Tefad: yes, as long as it doesn't the APIs that got removed in 10.5
06:11:46 *** Moodles has joined #openttd
06:11:55 <Tefad> i don't think microsoft removes many APIs
06:12:06 <Tefad> just deprecates them and ignores them.
06:12:08 <keiya> Tefad: Hence their horrible security, and such...
06:12:11 <goodger> I have plenty of dmgs that specifies individual point releases of os x
06:12:37 <Tefad> keiya: that's when you have to log in as administrator to run legacy apps
06:12:45 <goodger> "requires OS X 10.4.2-10.4.6. incompatible with Panther"
06:12:55 <keiya> MS /can't/ secure WIndows, because they have to keep support for all the horrible flaws..
06:13:06 <Tefad> keiya: they're working on it
06:13:08 <goodger> I forget the name for 10.5, it was really obscure
06:13:16 <Tefad> i still don't use windows or OS X at home
06:13:27 <Tefad> the wife even likes using ubuntu
06:13:30 <keiya> But, uh, have you tried running software for 2000 on ME? :P
06:13:43 <Tefad> easy for her to screw around with additional apps with the package manager
06:13:51 <goodger> keiya: that's completely different
06:14:03 <keiya> At least when Apple breaks backwards compatability, they actually break it instead of just refusing to run arbitrarily :P
06:14:16 <keiya> (Case in point: .net dev kit.)
06:14:43 <keiya> 99% of it runs just fine on 98...
06:15:04 <keiya> (The one exception being a rarely used tool... oh, and the installer, which does a test and refuses to run even though it'd work fine.)
06:15:06 <goodger> keiya: I reckon MS will replace the windows kernel with a bastard UNIX and port the shell, then run a choice of totally enclosed compatibility layers, just as apple did ten years ago, and microsoft could have done twenty years ago
06:15:27 <keiya> goodger: MS did that already...
06:15:37 <keiya> Only not Unix, just a new kernel.
06:15:43 <goodger> keiya: obviously they did not
06:15:52 <Rubidium> and the function that breaks for 10.5 is a function that converts a OSX type string into a C type string
06:16:14 <keiya> Rubidium: That's actually another flaw..
06:16:17 <Tefad> the windows kernel is quite solid. i don't see why they'd run from it
06:16:27 <keiya> If you're not useing objc, you're a second-class citizen
06:16:31 <Tefad> the GUI code that /is/ in the kernel has been run through paces for years
06:16:49 <Tefad> the NT kernel is a hybrid kernel
06:17:01 <Tefad> NT was a joint venture between IBM and MS
06:17:08 <keiya> They then hacked on bits of the old stuff, mind.
06:17:18 <keiya> Tefad: Not really... originally, yes.
06:17:30 <keiya> But, the current releases are so far removed from that... :J
06:17:33 <Tefad> microsoft got the rights to NT through a long court process
06:17:54 <Tefad> there were even some beta floppies of OS/2 that were microsoft branded ; )
06:18:13 <Tefad> anyway, the kernel has roots from there
06:18:26 <goodger> Tefad: the kernel is theoretically just about adequate. everything else of the system is a shittily executed attempt to turn DOS into UNIX; if they replace the entire thing with a real UNIX and put in a totally enclosed compatibility system [think vmware images]
06:18:35 <goodger> then it will all work much more nicely
06:18:36 <keiya> And the 9x kernel is... DOS!
06:18:36 <Tefad> microsoft DOES have a unix kernel-thing sitting next to its win32-thing
06:19:06 <Tefad> but that only comes with the top tier windows versions.. business
06:19:08 <keiya> What's next, are you going to claim that BSD and Linux are the same thing? >_>
06:19:26 <Tefad> interix is heavily influced by bsd
06:19:40 <goodger> keiya: GNU and BSD are both UNIXes, that's good enough for me
06:19:42 <Tefad> i used interix on XP at work for years
06:20:07 <Tefad> interix is a unix subsystem for NT
06:20:17 <Tefad> just like win32 is a subsystem
06:20:23 <goodger> (linux is pretty irrelevant in this context, it's one of quite a few decent UNIX kernels)
06:20:48 <keiya> goodger: They implement the same API, but they're not the same kernel.
06:20:58 <Tefad> one of my freebsd friends has some respect for the linux kernel, but hardly any for the GNU 'experience'
06:21:03 <Rubidium> keiya: so you say that OpenTTD for OSX is a bug?
06:21:35 <keiya> You could /pretend/ to make sense.
06:21:42 <DaleStan> Except that, technically, Linux is not a UNIX. They haven't got the certification as such from The Open Group.
06:22:06 <Rubidium> keiya: you're effectively saying that if it isn't written in objc it is not right
06:22:14 <goodger> keiya: I didn't say that linux and the BSD kernel are the same, I said they might as well be ignored because there's no major differences --- same with GNU and BSD. we're looking at a "windows and unix" world here, wherein unix is practically everything except windows
06:22:25 <keiya> Rubidium: No, I'm saying that's a flaw /with OS X/...
06:22:43 <goodger> DaleStan: we're not talking about certification, we're talking about reality
06:22:51 <keiya> Anyway, I'm tired of this
06:23:18 <keiya> How did we spin off to here from "Ooh, I can set it up to let me scroll sideways with the touchpad!", anyway?
06:23:19 <goodger> but it's only half-ten!
06:23:55 <goodger> keiya: I said I preferred the old ones, then someone said macintoshes were overpriced, then you said the operating system is the only one that works properly
06:24:01 <keiya> I wanted to give Dare a hug before she left :(
06:24:07 <keiya> goodger: No I didn't...
06:24:24 <keiya> What I meant to say, which you all misinterpreted, was that others suck just as much.
06:24:58 <goodger> perhaps we "misinterpreted" you because you didn't say that
06:26:00 <goodger> (I say half-ten with confidence because it's obvious you're californian)
06:26:32 <goodger> do you have a burlap bag and a hybrid car?
06:27:32 <keiya> And, no, I'm not a Californian.
06:27:35 * Rubidium thinks most people in here should first get some real knowledge about developing applications for the different platforms before they say what is more/less compatible (and thus stable w.r.t. application development support)
06:28:09 <Tefad> Rubidium: just use java?
06:28:18 <keiya> Rubidium: Eh, you have to code for the lowest common denominator no matter /what/ platform you're using. That doesn't change.
06:28:35 <keiya> It's just, how big that set is changes :P
06:28:44 <keiya> Tefad: And still slow as hell!
06:28:59 <keiya> (Well, sometimes... it depends on the exact setup.)
06:29:00 <Tefad> so no matter wtf happens, someone can get your shit working if you pay them enough (eg 25years down the road)
06:29:20 <Tefad> java /can/ be slow sure.
06:29:41 <Tefad> that's why you can create platform specific accelerations if you need them
06:29:43 <keiya> Any time you do graphics, here :p
06:30:00 <Rubidium> Java's GUI init is especially slow, but technically it can be faster than normally compiled C code
06:30:18 <Tefad> another key word is 'normally' compiled C code
06:30:24 <keiya> Rubidium: Except that Apple's version sucks horribly :P
06:30:38 <Tefad> properly optimized C code gives Java's JIT a run for its money : )
06:30:42 <Rubidium> without heavy profiling etc. and it requires Java to support JIT on your platform
06:30:44 <Tefad> but it's close competition
06:30:48 <goodger> <dooglus> asg: the biggest package is linux-image-2.6.16-1-686. I don't do any image processing, so I think it's probably safe to remove that one
06:32:04 <keiya> I still want to build an x86 from train logic gates, and run openttd on openttd.
06:32:05 <Tefad> poo. a GCC update. i hate these.
06:32:21 <keiya> (I think you'd need a freaking-huge-map patch, though.)
06:32:51 <keiya> I think maybe sparc, too... has anyone gotten OpenTTD working on an m64k? :P
06:33:40 <Rubidium> I see no reason why it wouldn't work
06:34:10 <Rubidium> assuming it's a 32+ bits processor
06:34:47 <goodger> surely one could port OTTD to 16-bit fairly easily
06:36:23 <Rubidium> is it really m64k or did you mean m68k?
07:05:11 *** Brianetta has joined #openttd
07:23:22 *** Yeggstry has joined #openttd
07:38:07 *** Tim is now known as Timitry
07:39:50 *** Timitry is now known as Tim
07:41:56 *** Tim is now known as Timitry
07:43:11 *** Timitry is now known as Tim-itry
08:09:18 *** sigmund has joined #openttd
08:17:56 <Rubidium> dihedral: then go fix something
08:21:44 *** mortal` has joined #openttd
08:23:56 *** mortal`` has joined #openttd
08:24:01 *** Mortal is now known as Guest69
08:24:01 *** mortal`` is now known as mortal
08:32:07 *** Vikthor has joined #openttd
08:34:45 *** Terkhen has joined #openttd
08:41:03 *** Rubidium_ has joined #openttd
08:41:17 *** prakti_ has joined #openttd
08:41:28 *** canidae_ has joined #openttd
08:41:38 *** wolfryu has joined #openttd
08:41:45 *** mucht_work has joined #openttd
08:44:01 *** canidae_ is now known as canidae
08:51:11 *** WhiteRhino has joined #openttd
08:52:02 *** XeryusTC has joined #openttd
08:52:25 *** ttdopen has joined #openttd
08:52:30 *** planetmaker has joined #openttd
08:52:58 *** planetmaker is now known as Guest82
08:53:13 *** Vikthor has joined #openttd
08:53:20 *** goodger has joined #openttd
08:55:58 *** dih is now known as Guest83
08:56:44 * Tim-itry is busy installing all that stuff needed to compile OpenTTD on his laptop...
08:57:06 *** Rubidium_ is now known as Rubidium
08:58:16 *** [com]buster has joined #openttd
08:58:33 <Tim-itry> If i want to get the latest source of OpenTTD, and write my own patches, and update them to trunk from time to time, is TortoiseHG a good choice and enough? (WinXP)
09:04:31 *** Guest82 is now known as planetmaker
09:11:52 <Tim-itry> Okay, just trying to set it up (TortoiseHG)
09:12:09 <Tim-itry> Oh, Rubidium, which present will we get with Revision 15000? :)
09:13:07 <petern> what more could you want?
09:14:34 <Tim-itry> Hm, are you asking for a wishlist? :D
09:15:27 <Tim-itry> Uh, seems like cloning the trunk to my laptop actually works Oo
09:21:16 <Tim-itry> So, after having a clone of the trunk on my laptop, what next? Do i have to modify the files i want to change over TortoiseHG, or can i just modify them with the editor of my choice and TortoiseHG keeps track of the changes?
09:26:19 <petern> yeah modify them as normal
09:36:18 <Tim-itry> Hm, modified one file and succesfully created a new local revision... However, i can't add files so it will visually show me in the explorer which files have been changed...
09:43:16 *** Guest83 is now known as dihedral
09:43:48 *** dihedral is now known as Guest89
09:44:22 <petern> hmm, msn won't connect :/
09:44:36 *** Guest89 is now known as dihedral
09:46:46 <petern> and fake explorer views piss me off
09:47:01 <petern> you get a nice file listing in what looks like an explorer view
09:47:19 <petern> but you can't universally drag & drop them
09:47:37 <petern> needed to copy them into a temp directory before dragging onto winscp
09:50:42 <petern> something tells me the camera took a while to process it...
09:54:07 *** ConditionalZenith has joined #openttd
10:02:10 <Rubidium> so it isn't bending of time that you've photographed?
10:05:38 *** freiksenet has joined #openttd
10:06:13 <freiksenet> oops, wrong channel
10:35:03 *** const86 has joined #openttd
10:39:18 *** [alt]buster has joined #openttd
10:39:20 *** [alt]buster is now known as [com]buster
10:40:06 *** TrueBrain has joined #openttd
10:48:21 <CIA-1> OpenTTD: truebrain * r15000 /branches/noai/src/ (ai/ai.hpp saveload/ai_sl.cpp): [NoAI] -Fix (r14984): forgot to rename @file too
11:15:31 <petern> we were at 2xxx when i started
11:15:49 <petern> my software mixer patch is coming along.
11:16:07 <petern> only problem is it requires modifying all the sound drivers, so testing is... fun
11:18:40 *** Dred_furst has joined #openttd
11:20:02 <Rubidium> petern: only requires two sound drivers to test (sdl and win), right?
11:20:36 <Rubidium> who cares about cocoa?
11:21:06 *** [com]buster has joined #openttd
11:22:28 <petern> once that's done then there is no fixed reliance on the software mixer
11:22:39 <petern> so an sdlmixer driver can be done
11:22:53 <petern> although that's only a software mixer anyway, heh
11:23:17 <petern> but sdlmixer would provide easy midi support for linux, with no sound device conflicts
11:24:02 <petern> and i think dominic wanted it for whatever port he was doing
11:55:53 *** Progman has joined #openttd
12:16:29 *** Tim-itry has joined #openttd
12:18:34 *** ChanServ sets mode: +v tokai
12:18:44 <Tim-itry> Hm, if i try to get the source code by svn checkout, it tells me "access denied"
12:19:06 <TrueBrain> don't try to commit
12:22:39 <Tim-itry> Well, i'm not trying, i just want to get the code with TortoiseSVN and try "SVN Checkout" or "Import", both gives me access denied - error
12:22:55 *** [com]buster has joined #openttd
12:23:04 <TrueBrain> Import is wrong for sure
12:23:05 <Rubidium> Tim-itry: what url are you trying to download from?
12:23:12 <TrueBrain> checkout should work, if you use svn://svn.openttd.org/trunk
12:24:13 <Tim-itry> That's the one i use...
12:24:43 <Rubidium> and then you said something like use username X?
12:25:21 <Rubidium> or tortoise is thinking it should login instead of anonymously get the data
12:27:07 <keiya> I hate you all for not removing antennas.
12:28:58 <Rubidium> ignor(anc)e is bliss
12:30:20 *** SHRIKEE has joined #openttd
12:35:16 <petern> glx's first patch is before mine, heh
12:35:41 <petern> first commit much later though
12:36:06 <Rubidium> that revision sounds like around my first OTTD releated patch
12:37:16 *** frosch123 has joined #openttd
12:37:43 <petern> hmm, 10,000 commits in trunk
12:38:22 <petern> 3511 is first commit with your name on it
12:42:27 <TrueBrain> petern: what are that?
12:42:39 <petern> commits to trunk per person
12:43:08 <TrueBrain> petern: I count 717 for me ;)
12:44:28 <petern> grep ludde log.txt -c gives 53
12:44:31 <TrueBrain> 52 .. lol :) (ludde had some patches which were committed not under his name! :p)
12:44:34 <petern> but that's not 53 commits
12:45:22 <frosch123> oh, we are already at 15000...
12:45:23 <TrueBrain> r2532 | hackykid | 2005-07-08 21:02:26 +0200 (Fri, 08 Jul 2005) | 2 lines
12:45:23 <TrueBrain> - Fix: Don't waste space using an int where a byte would suffice. (ludde)
12:45:29 <TrueBrain> I think that should be counted to ludde :p
12:45:38 * frosch123 was mentally still at r14953 :/
12:46:11 <petern> TrueBrain, too much hassle to do that for every possible commit
12:46:18 <petern> loads have attribution
12:46:31 <TrueBrain> petern: fair enough :)
12:47:25 <TrueBrain> petern: but I don't find those numbers fair! I did a lot of commits in tgp branch, and that was commited as a single trunk commit :(
12:47:37 <TrueBrain> sorry, I will shut up now :)
12:47:42 <Tim-itry> hm, maybe the firewall here at my workplace is blocking the access to the svn repository... Since it does also not work via MSys
12:48:23 <TrueBrain> Tim-itry: 'access denied' means there is some kind of connection
12:48:28 <TrueBrain> so what are you trying EXACTLY in msys
12:50:02 <TrueBrain> you should add truelight and truebrain together :p (hehehehehe :))
12:50:28 <petern> i guess that's your secret hidden commits
12:51:35 <TrueBrain> well, not really anymore
12:51:39 <TrueBrain> mapgen is no longer part of SVN
12:51:44 <TrueBrain> it is gone completely
12:51:50 <Tim-itry> MSys: svn checkout svn://svn.openttd.org/trunk
12:51:58 <TrueBrain> a few more revisions went byebye
12:52:24 <TrueBrain> Tim-itry: should work, works where :)
12:53:18 <Tim-itry> Message (translated): svn: Can not connect to Host svn.openttd.org: Could not connect because the hostcomputer denied the connection
12:53:37 <TrueBrain> now THAT is your firewall
12:53:42 <TrueBrain> but no message about 'access denied' ...
12:53:47 <TrueBrain> (which suggests a password login)
12:53:52 <TrueBrain> bad Tim-itry, giving us the wrong impression
12:54:03 <Tim-itry> denied the connection... ;)
12:55:55 <TrueBrain> it does bypass your firewall ;)
12:56:23 <TrueBrain> (first redirects to the second)
12:56:41 <Tim-itry> Hehe, fast enough for me, needed only like 10 seconds to download
13:11:18 *** Sacro_ is now known as Sacro
13:12:00 *** roboboy has joined #openttd
13:17:14 *** Vikthor has joined #openttd
13:28:35 *** ConditionalZenith has quit IRC
13:32:51 *** TinoDidriksen has joined #openttd
13:45:45 *** [alt]buster has joined #openttd
13:45:48 *** [alt]buster is now known as [com]buster
13:51:30 <CIA-1> OpenTTD: truebrain * r15001 /branches/noai/src/ai/ai_core.cpp: [NoAI] -Fix: make NoAI network safe again
13:55:50 <TrueBrain> petern: go for it :p
13:59:07 <petern> hmm, railtypes is 76KB !
13:59:44 *** worldemar has joined #openttd
14:06:50 <frosch123> ask the prime minister, he should know
14:06:51 <Mortal> it's divisible by 1 and 15551 and no other integer
14:07:26 <Rubidium> but dividing by two gives me 7775
14:07:50 <Mortal> Rubidium, retry with infinite precision
14:08:15 <Rubidium> int main(int argc, char *argv) { printf("%i\n", 15551 / 2); } <- Q.E.D.
14:08:17 <frosch123> there is nothing as precise as integer arithmetics
14:08:33 <Mortal> yes, fractional arithmetics
14:08:52 <SpComb> Rubidium: and multiplying 7775 by 2 gives you what?
14:09:55 <Aali> Sacro: what did you answer?
14:10:40 <frosch123> 1) looks weird, so I say D)
14:10:57 <Mortal> sacro, 2 of the answers are correct
14:11:06 <Rubidium> Sacro: D is right, thus C isn't and A isn't either, but as A or B or C is right, B must be right too
14:11:07 <Mortal> now, go and finish your homework
14:12:27 <CIA-1> OpenTTD: truebrain * r15002 /branches/noai/ (28 files in 5 dirs):
14:12:27 <CIA-1> OpenTTD: [NoAI] -Codechange: move 'squirrel' to 'script/'
14:12:27 <CIA-1> OpenTTD: [NoAI] -Codechange: renamed 'AISquirrel' to 'AIScanner', as that is what it is doing
14:12:33 <Rubidium> but you not being able to figure out that D is right should fail you the test (you can't use a compiler after all)
14:13:39 <CIA-1> OpenTTD: truebrain * r15003 /branches/noai/projects/ (openttd_vs80.vcproj openttd_vs90.vcproj): [NoAI] -Fix r15002: of course I forgot the MSVC project files
14:13:53 <frosch123> hehe, I knew you would run out of fruits
14:14:33 <Mortal> r15004 [NoAI] -Codechange: renamed 'AIScanner' to 'AISquirrel', as that is way the cutest name~<3
14:14:41 <TrueBrain> blathijs: you think the first thing compiles?
14:14:44 <frosch123> there shouild be "E) implementation specific"
14:15:09 <Mortal> yeah... after all, what is a Vector *really*
14:15:14 <TrueBrain> Mortal: I am glad you don't have commit rights ;)
14:15:39 <TrueBrain> frosch123: not in a fruity mood ;)
14:15:44 *** lewymati has joined #openttd
14:15:54 <TrueBrain> waiting on this lovely girl to reply to my question .. bleh :p
14:16:16 <frosch123> "wanna marry me?" ?
14:16:44 <blathijs> TrueBrain: Oh wait, there is a "result" in between there
14:17:31 <CIA-1> OpenTTD: truebrain * r15004 /branches/noai/ (4 files in 2 dirs):
14:17:31 <CIA-1> OpenTTD: [NoAI] -Change [API CHANGE]: renamed 'SetSetting' to 'AddSetting' in info.nut, as this makes much more sense
14:17:31 <CIA-1> OpenTTD: WARNING: all info.nut using SetSetting will now fail to compile. Change ASAP!
14:17:37 <TrueBrain> blathijs: yeah ... silly, not?
14:17:56 <blathijs> TrueBrain: That does ensure this question has a non-ambiguous answer
14:18:26 <blathijs> TrueBrain: Otherwise, it would be dependent on compiler implementation and Vector implementation
14:18:48 <blathijs> TrueBrain: (ie, if Vector has a copy constructor with observable behaviour)
14:18:53 <TrueBrain> blathijs: then I would have agree'd with you :)
14:18:59 <TrueBrain> and the question would have been really stupid in that case
14:19:05 <Zuu> but still, both B and D can be said to be true. And it says "Which one is true"
14:20:02 <Mortal> nah zuu, you can't really compare the efficiency of one program to the efficiency of uncompiled code
14:20:03 <TrueBrain> Zuu: B requires both to compile, in any sanes persons mind ;)
14:20:09 <TrueBrain> so saying B can be true, is just nitpicking
14:20:31 <Mortal> then B could be true and code section 1 could be line noise
14:22:24 <Mortal> scratch that, we're not comparing efficiency of compiled programs... ugh
14:22:55 <Mortal> I can't wait until I learn this shizzle properly at uni
14:23:39 <CIA-1> OpenTTD: truebrain * r15005 /branches/noai/ (5 files in 2 dirs):
14:23:39 <CIA-1> OpenTTD: [NoAI] -Change [API CHANGE]: remove Event filters, as nobody was using them
14:23:39 <CIA-1> OpenTTD: [NoAI] -Change [API CHANGE]: removed AIEvent::Test(), as it was silly
14:23:40 <blathijs> I would say that code section 2 is most efficient... to fill up a file that has to be 100KiB in size :-)
14:23:53 <blathijs> SpComb: You thought to contribute some line noise? :-)
14:24:47 <Mortal> spcomb is lowering the entropy a bit
14:24:50 <Steve-N> i'm running into this issue (in openttd, not some exam;) and i was wondering if this would be considered a bug or a feature:
14:25:11 *** ChanServ sets mode: +o orudge
14:25:30 <Steve-N> when a vehicle is at a station, without full-load order, it refuses to leave as long as new cargo is arriving at the station and this vehicle is the only one pickup up this specific cargo at this station. this can take a really long time, until the vehicle is completely filled up..
14:26:02 <dihedral> no - it refuses to leave as long as cargo is arriving at the station in the time the train is still loading
14:26:09 <dihedral> and still has space for more
14:26:17 <Steve-N> yes that's what i mean
14:26:26 <frosch123> it has always been like that
14:26:39 <Steve-N> that shouldn't mean it's good ;)
14:26:48 <Rubidium> we have very lenient conductors; they don't close the door when someone is entering
14:27:06 <Steve-N> when i go on the subway, it just leaves if some passengers were getting on while i am walking into the station
14:27:22 <Zuu> and closing doors 30 seconds before departure dosen't exist in OpenTTD.
14:27:28 <frosch123> but not when they are blocking the doors of the train
14:27:33 *** [com]buster has joined #openttd
14:28:21 <frosch123> you should try that in rl, get 50 friends and keep on entering the wagon through one door and leave through the other...
14:28:32 <Zuu> There are these trains where you can open the doors from the outside even if the train has started to roll out from the station.
14:28:43 <Steve-N> then the doors will just open briefly and slam close again, or the subway driver will come out of his cabin
14:28:58 <Steve-N> (believe me, i tried, a long time ago;)
14:29:13 <dihedral> well - they are trains, not subways
14:30:19 <Steve-N> this goes for any vehicle, bus, train, monorail (+metro track)
14:32:01 <CIA-1> OpenTTD: truebrain * r15006 /trunk/src/ (4 files): -Codechange: throw a real instance of a class, instead of '0' (which can also be a throw from within a thread for what ever reason)
14:32:02 <SpComb> goto and longjmp are great for error handling
14:32:09 <SpComb> dismissing them as evil is silly
14:32:23 <TrueBrain> SpComb: also very thread-safe and stuff :p
14:32:46 <SpComb> I'd hate to have to write code somewhere where they decide that "goto is evil and shall not be used, ever, without exception"
14:33:47 <blathijs> SpComb: They are pretty complex, and usually there better tools for the job (goto -> loops, longjmp -> exceptions)
14:33:59 <TrueBrain> longjmp for error handling in a C application, acceptable
14:34:21 <SpComb> I mean `if ((ptr = calloc(1, sizeof(*ptr))) == NULL) goto error;`
14:34:46 <TrueBrain> SpComb: in the C++ world we do instead of 'goto error': 'throw OutOfMemory'
14:34:55 <SpComb> I'm not talking about C++
14:35:02 <TrueBrain> (as can you have one layer somewhere to handle every memory allocation failure)
14:35:05 <SpComb> in C++, you don't even have the if
14:35:21 <TrueBrain> one good reason to start using C++ ;)
14:35:29 <Rubidium> don't forget to mention that break and continue are gotos (without label)
14:35:35 <worldemar> SpComb: how do you replace if in C++
14:35:36 <SpComb> imo, in C, goto is the only sane way of handling errors in functions
14:35:52 <SpComb> and that's not a bad thing at all, it works very well
14:35:53 <ccfreak2k> If goto is evil, why is it in the C spec?
14:36:14 <TrueBrain> so because it is in some spec ... it is not evil?
14:36:21 <SpComb> of course, goto can be abused for stupid crap as related to weird loop structures, but so can anything else
14:36:22 <worldemar> C is a spartan language (c) linux kernel coding style
14:37:14 <ccfreak2k> It was good enough for Kernighan and Ritchie.
14:37:28 <SpComb> libpng uses longjmp for error handling, and using it is alright, although I think I'd just preferr using explicit error return codes for the library API
14:38:01 <TrueBrain> libpng sucks for its longjmp usage
14:38:21 <SpComb> you can set a new jmp_buf in each function where you call libpng
14:38:39 <SpComb> and I don't use threads at that point
14:38:43 <TrueBrain> SpComb: you mostly HAVE to set a new one ...
14:38:50 <TrueBrain> (to gain any sane error information)
14:39:00 <TrueBrain> a simple 'return error' would have worked as well :(
14:39:29 *** einKarl has joined #openttd
14:39:29 <SpComb> the longjmp thing is less code for functions that call libpng many times, and it also works for sub-functions
14:40:01 * Rubidium likes to think that goto should be avoided if easily possible, but certainly not at all cost; deliberately (with thought) using it is fine, but it shouldn't be a habit of using it everywhere
14:40:14 <CIA-1> OpenTTD: truebrain * r15007 /branches/noai/src/ai/api/ai_event.hpp.sq: [NoAI] -Fix (r15005): squirrel_export.sh eats a tab in one file every single time for no good reason ... BAD SCRIPT! (tnx SmatZ for spotting)
14:44:20 <SpComb> but way too often you see some function where each function call with an error return code has its own if block with the same cleanup code copy-pasted into it
14:45:33 <CIA-1> OpenTTD: glx * r15008 /trunk/ (7 files in 3 dirs): -Revert (r12706): remove fibers
14:45:44 <TrueBrain> (I just don't use throws enough :p)
14:46:37 <SpComb> so you were talking about copy constructors earlier...
14:46:52 <SpComb> and how the existance of one impacts the performance of some other method implementation
14:48:17 *** const86 has joined #openttd
14:49:04 <SpComb> it's interesting when you have things like default constructors or copy constructors that do syscalls, like an IPAddress class that calls getaddrinfo
14:49:35 <SpComb> then you strace your program and start wondering why you have seven billion completely useless syscalls for netlink sockets to get network interface addresses
14:50:36 <SpComb> turns out your code or the library code is causing default IPAddresses to be constructed or copied a couple times every time you receive a packet
14:50:58 <SpComb> the compiler can't optimize out calls to getaddrinfo
14:55:36 <SpComb> that's one of example of what I had to deal with when working on my first C++ application
14:56:43 <CIA-1> OpenTTD: frosch * r15009 /trunk/src/train_cmd.cpp: -Fix [FS#2528]: No need to extent the reserved path when the vehicle is still loading.
15:00:03 *** shumway-dk has joined #openttd
15:00:35 <shumway-dk> anyone able to help with a langhuage problem with openttd
15:02:12 <shumway-dk> my problem is it has changed the language in my game from english to danish and that is just plain annoying so i need some way to change it back to english
15:02:15 <TrueBrain> Belugas!!! :) :) :)
15:02:53 <TrueBrain> Game Options -> Language -> English
15:04:03 <shumway-dk> thanks that helped now ill play
15:05:20 <Belugas> that was... VERY silly
15:05:54 <TrueBrain> how is Belugas doing this day? :)
15:06:00 <TrueBrain> it is cold outside here :)
15:06:15 <SHRIKEE> so it's a worthless cold then
15:06:51 <frosch123> no, I already saw someone leaving his bicycle quite fast towards the floor
15:07:06 <petern> gah, what can i use to view a patch file on windows...
15:07:12 <petern> (with colours 'n stuff)
15:08:08 <petern> tortoisesvn and winmerge will show diffs, but not from files, only actual changed files :/
15:08:29 <Belugas> i'm doing fine, thnaks. just that work at work is so fucking demanding :(
15:08:34 <CIA-1> OpenTTD: glx * r15010 /branches/noai/ (67 files in 8 dirs): [NoAI] -Sync: with trunk r14994:15009
15:09:29 *** [alt]buster has joined #openttd
15:09:46 *** [alt]buster is now known as [com]buster
15:13:25 <Belugas> petern, notepad2 is what i use. works like a charm
15:13:32 <Belugas> color highligh and all
15:13:43 <petern> will apply bits of patch tonight, i think
15:14:22 <Belugas> ? Canadian Bankers Association ?
15:14:34 <Belugas> Club de badminton d'Alma ?
15:15:16 *** Tim is now known as Tim-itry
15:19:32 <TrueBrain> petern: Nucleair or Nuclear
15:20:35 <Aali> Nucleair? doesn't make sense :P
15:22:11 <CIA-1> OpenTTD: truebrain * r15011 /trunk/src/newgrf_cargo.h: -Documentation: Nuclear, not Nucleair
15:23:06 <petern> absolutely no need for those extended classes ;)
15:23:16 <petern> they're supposed to be user defined, heh
15:23:49 <petern> 0-7 is standard, 8-15 are custom
15:25:16 <Zuu> hmm, looking for (and fixing) artifacts from a dirty camera lens is a bit tricky on a dirty monitor.
15:25:21 <CIA-1> OpenTTD: truebrain * r15012 /branches/noai/src/ai/ai_core.cpp: [NoAI] -Fix: don't close AIs on a client
15:25:29 *** FauxFaux has joined #openttd
15:25:32 <CIA-1> OpenTTD: truebrain * r15013 /branches/noai/src/ (8 files in 2 dirs): [NoAI] -Fix: make the diff between trunk and NoAI a tiny bit smaller
15:26:39 <scia> long time since i've been here ;)
15:27:23 <TrueBrain> so you missed us? :)
15:27:57 <CIA-1> OpenTTD: frosch * r15014 /trunk/src/ (command.cpp command_func.h): -Codechange: Add a helper function to get the needed DC_xxx flags from the result of GetCommandFlags().
15:29:52 <CIA-1> OpenTTD: frosch * r15015 /trunk/src/station_gui.cpp: -Fix (r14919): Distant-join always failed for docks. Based on patch by PhilSophus.
15:30:24 <CIA-1> OpenTTD: truebrain * r15016 /branches/noai/src/settings.cpp: [NoAI] -Fix: make the current default for suspend-blabla the same as the setting-to-be
15:31:26 <planetmaker> woah... I sense a comit-ing spree here... :)
15:32:35 <scia> hehe, I missed everything :p
15:32:50 <frosch123> planetmaker: would be easier, if you had done the splitting :p
15:33:27 <planetmaker> I'm out of context... what do you refer to, frosch123?
15:33:41 <frosch123> puttint 2) and 3) in one patch
15:34:23 <frosch123> ah, ok, they affect different files
15:34:50 <frosch123> ah, sorry, planetmaker and philsophus both start with 'p' :)))
15:35:21 <planetmaker> Being mixed up with him is no slight ;)
15:35:25 *** stillunknown has joined #openttd
15:39:09 <CIA-1> OpenTTD: truebrain * r15017 /branches/noai/src/ai/api/squirrel_export.sh: [NoAI] -Fix: running squirrel_export.sh no longer gives a diff
15:43:51 <CIA-1> OpenTTD: truebrain * r15018 /branches/noai/src/helpers.cpp: [NoAI] -Fix (sync fix): helpers.cpp should have been removed (spotted by Rubidium)
15:44:49 <CIA-1> OpenTTD: frosch * r15019 /trunk/src/settings.cpp: -Fix (r14919): Close station selection window, when disabling distant-join setting. (PhilSophus)
15:49:05 <George> frosch123: Tell me please, what is the chance to have FS#2521 done this month?
15:53:08 <Eddi|zuHause> slope information, i assume
15:54:02 * worldemar just read the topic >_<
16:01:15 <CIA-1> OpenTTD: truebrain * r15020 /branches/noai/ (3 files in 2 dirs): [NoAI] -Change: put 'script' in their own subfolder for MSVC
16:02:14 *** energetic has joined #openttd
16:03:00 <Belugas> George, it's not out of bad will
16:03:21 <Belugas> let say it's bad to rush on something withouth thinking about it.
16:04:03 <CIA-1> OpenTTD: truebrain * r15021 /branches/noai/src/ai/ (ai_core.cpp ai_instance.cpp): [NoAI] -Codechange: minor changes and movements (nothing groundbreaking or anything)
16:04:34 <frosch123> haha, "minor changes and movements (nothing groundbreaking or anything)" could also qualify for a merge :p
16:05:14 <Zuu> worldemar: entry number 2521 on flyspray (bugs.openttd.org)
16:05:26 <TrueBrain> frosch123: I already said in an other channel: I wonder how many people suspect something else ;)
16:05:36 * Belugas remembers a big commit by tron who was saying something like "nothing" important and was responsible of a lot of fixes ;)
16:07:04 <Belugas> i see petern remebers that one too ^_^
16:07:18 <Belugas> worldemar, why do you always end up your posts with ")" ? does it have a certain meaning for you?
16:07:53 <worldemar> Belugas: two posts makes you think it is always?
16:08:48 <TrueBrain> and out of the 4 (5 by now), that is a pretty big number
16:09:48 <Belugas> well... yes, since those are the only ones i've seen from you, worldemar ;)
16:09:54 <CIA-1> OpenTTD: rubidium * r15022 /branches/noai/src/ai/ (ai_gui.cpp ai_gui.hpp ai_info_dummy.cpp ai_storage.hpp): [NoAI] -Cleanup: coding stylish
16:11:14 <worldemar> Belugas: both "(" and ")" has meaning for me. maybe it's just good evening...
16:11:35 *** [com]buster has joined #openttd
16:12:37 <Belugas> so... ( is good and ) is evening... cool :)
16:14:08 <worldemar> something like a sed script
16:14:09 <TrueBrain> ( ) to you too petern :)
16:15:45 <OwenS> As long as we don't start talking in BF :-P
16:16:04 <worldemar> Belugas: "good evening" was about lots of ")"
16:16:34 <worldemar> OwenS: because it will be really BF
16:17:03 <CIA-1> OpenTTD: glx * r15023 /branches/noai/ (3 files in 2 dirs): [NoAI] -Change: reorder some files in source.list to please MSVC
16:17:52 <OwenS> ++++++++++[>+++++++>++++++++++>+++>+<<<<-]>++.>+.+++++++..+++.>++.<<+++++++++++++++.>.+++.------.--------.>+.>. to you too! :p
16:22:54 *** mortal` has joined #openttd
16:27:54 *** Mortal is now known as Guest118
16:27:54 *** mortal` is now known as mortal
16:28:13 <CIA-1> OpenTTD: rubidium * r15024 /branches/noai/src/ (6 files in 3 dirs): [NoAI] -Cleanup: remove some hacks
16:28:58 *** HerzogDeXtEr has joined #openttd
16:31:20 *** mortal` has joined #openttd
16:31:35 *** mortal is now known as Guest119
16:31:35 *** mortal` is now known as mortal
16:37:55 <Belugas> bank processing, i do totally hate you from time to time :S
16:38:07 <petern> banlist loading doesn't support to work properly
16:38:34 <petern> or saving, one of the two
16:41:02 <Eddi|zuHause> do i have to ask what 72 101 108 108 111 32 87 111 114 108 100 33 10 means? :p
16:41:59 <blathijs> Could be ASCII I guess
16:42:03 <Eddi|zuHause> at least that's what my internal brainfuck interpreter says...
16:42:09 <Eddi|zuHause> of course it's ascii ;)
16:42:22 <Eddi|zuHause> but i'm too lazy to open the table ;)
16:42:26 <blathijs> ending newline (10) most numbers above 97 (lowercase), but starting with an uppercase letter
16:42:43 <worldemar> it is "good evening"
16:42:48 <Eddi|zuHause> what's an A? 65?
16:43:09 <worldemar> Good, to be precize
16:43:17 <blathijs> worldemar: Huh? 101 is not an "o" AFAICS
16:43:42 <blathijs> worldemar: hey wait, the letters don't even match :-)
16:43:55 <Eddi|zuHause> 96? shouldn't a be odd?
16:43:58 <petern> space in the wrong place
16:44:10 <worldemar> seems like my brain isn't brainfuck interpreter
16:44:10 <worldemar> blathijs: yeah, i noticed)
16:44:19 <Eddi|zuHause> should be a difference of 32
16:44:26 <Eddi|zuHause> i.e. one flipped bit
16:44:32 <CIA-1> OpenTTD: glx * r15025 /branches/noai/ (47 files in 7 dirs): [NoAI] -Sync: with trunk r15009:15019
16:44:40 <blathijs> Eddi|zuHause: Yeah, that makes sense. I was thinking 0-based, but it is 1-based
16:44:41 <Eddi|zuHause> ascii is cool, you know ;)
16:45:30 <OwenS> All I can ever remember of ASCII is that 0x3n are the numbers :p
16:45:31 <blathijs> ie, 0110000 + 1-based letter == lowercase and 01000000 + 1-based letter == uppercase
16:45:57 <blathijs> OwenS: Oeh, I didn't know that one. *remembers*
16:46:56 <Eddi|zuHause> we had to implement an upper/lower case converter in hardware design
16:46:58 <OwenS> A lot of BCD math libs actually keep the numbers as ASCII strings actually :p
16:47:11 <petern> BCD is hardly necessary these days
16:47:14 <Eddi|zuHause> with don't cares for non-letters
16:47:26 <Eddi|zuHause> but some special escape letters
16:47:37 <OwenS> BCD is only useful for banking & calculators
16:47:40 <petern> BCD was designed so that conversion from in-memory to output was simple
16:47:43 <OwenS> I.E. where accuracy matters
16:48:04 <Eddi|zuHause> BCD is easy to read in hexdumps ;)
16:48:16 <petern> you don't need BCD for accurate numbers
16:48:39 <SpComb> I tend to store my data as bits
16:48:40 <OwenS> It tends to be used in arbitrary precision applications
16:48:48 <Eddi|zuHause> i vaguely remember that we should design a BCD adder in hardware, too
16:49:13 <petern> one way of doing that is use a regular wide integer value and another value stating where the decimal point is
16:49:19 <petern> although maths might get tricky, heh
16:49:24 <Eddi|zuHause> which is useful for displaying on a 7-segment-LED
16:49:37 <OwenS> I personally like designing processors in Verilog :p
16:49:53 * worldemar remembers his matrix multiplication script on bash
16:50:05 <OwenS> My brain and VHDL never agreeed
16:50:05 <Eddi|zuHause> but it might just have been a BCD counter, not adder
16:50:09 <SmatZ> we use VHDL primarily... but I think Verilog is better :)
16:50:11 <OwenS> Verilog just sticks better for me...
16:50:47 <OwenS> I've been designing a 32-bit pipelined processor
16:51:05 <OwenS> Lots of logic for stage interlocks :p
16:51:36 <Eddi|zuHause> i haven't done much hardware stuff since the 4th semester
16:51:41 <blathijs> 17:49:13 <@petern> one way of doing that is use a regular wide integer value and another value stating where the decimal point is <-- Isn't that exactly what floating point is?
16:51:58 <OwenS> blathijs: No. Floating point says where the BINARY point is
16:52:39 * worldemar keeps all hos floating points in aquarium
16:52:48 <OwenS> It's been fun deciding how to implement strcpy in hardware :p
16:53:00 <Eddi|zuHause> binary is known to have problems with rather common values like 0.2
16:53:11 <Eddi|zuHause> which are "round" in decimal, but periodic in binary
16:53:21 *** [com]buster has joined #openttd
16:53:44 <CIA-1> OpenTTD: smatz * r15026 /trunk/src/ (core/enum_type.hpp spritecache.cpp): -Codechange: use SpriteTypeByte instead of SpriteType in the SpriteCache struct in order to make it smaller
16:54:36 * OwenS doubts it's possible to implement strcpy in software faster than n/8+1 cycles best case
16:56:34 <Eddi|zuHause> abusing MMX opcodes or something?
16:56:48 <OwenS> Even so it would take longer
16:57:06 <Eddi|zuHause> yeah, i think they handle only 4 bytes
16:57:09 *** Vikthor has joined #openttd
16:57:24 <OwenS> And the hardware for strcpy isn't overtly complex either (perhaps my pipeline is bizzare :P )
16:58:13 <Eddi|zuHause> now find software for which strcopy is the bottleneck :p
16:58:29 <OwenS> memcpy is that fast too (actually faster)
16:58:47 <SmatZ> OwenS: what if you are forbidden to read past the terminating 0?
16:59:04 <Eddi|zuHause> memcopy is strcopy without checking for 0?
16:59:04 <SmatZ> you have to read byte-by-byte then
16:59:06 <OwenS> SmatZ: And when does this happen if the strings are 16-byte aligned?
16:59:28 <SpComb> Eddi|zuHause: you give memcpy a length
16:59:32 <OwenS> On x86, you can only fault on 4kb boundaries
16:59:50 <SmatZ> OwenS: when you are using for example debug registers to catch reads out of array borders
16:59:53 <SpComb> strings are 16-byte aligned? :/
17:00:10 <SmatZ> OwenS: segment limits???
17:00:12 <planetmaker> not only strings.
17:00:14 <OwenS> SpComb: Do the first n unaligned bytes as bytes, then do the rest at 16 byte alignment
17:00:19 <OwenS> SmatZ: Name me one OS which uses segments!
17:01:09 <SpComb> Microsoft SQL server uses segmented memory on 32-bit platforms!
17:01:26 <SmatZ> I don't know about linux
17:01:45 <OwenS> DOS uses Real Mode segments, which are vastly different; Windows uses FS and GS for special purposes, but you never store strings in them (pointers at most)
17:02:22 <Eddi|zuHause> this discussion is getting way too technical
17:02:57 <OwenS> Linux will let you use GS for thread local storage; but again you will never have a pointer which uses GS since the C standard hasn't heard of segments ;-)
17:03:50 <Belugas> butterscotch and peanuts popcorn
17:04:08 <Belugas> could hold a whole bag
17:05:37 <blathijs> They usually go hand in hand
17:05:44 <planetmaker> Belugas: I would rather recommend to eat what's IN the bag - not the bag itself ;)
17:05:47 <SmatZ> OwenS: things you are talking about are implementation-specific
17:06:05 *** energetic has left #openttd
17:06:15 <SmatZ> like malloc returning data aligned at 16B
17:06:43 <SmatZ> but you can't suppose the string will begin at 16B boundary
17:06:46 <OwenS> Malloc is guaranteed to return memory aligned to the platform's biggest data structure
17:07:09 <OwenS> But, in any case, have strcpy handle any unaligned leader bytes byte-by-byte
17:08:01 <OwenS> man malloc: "For calloc() and malloc(), return a pointer to the allocated memory, which is suitably aligned for any kind of variable"
17:08:23 <SpComb> is a one-byte malloc 16-byte aligned?
17:08:24 <OwenS> Since x86 has 16-byte aligned types, the memory must be 16-byte aligned
17:08:51 <SmatZ> OwenS: you are still talking about implementation specific problems
17:09:11 <OwenS> If your using SSE, your malloc returns 16 byte aligned variables else SSE is not usable!
17:09:12 <SmatZ> malloc may return "unaligned" pointers, if they can be used to point to any data structure
17:09:21 <SmatZ> implementation-specific
17:10:10 <OwenS> In any case, how did we end up discussing SSE when I was discussing implementing strcpy/memcpy as a CPU instruction? ...
17:10:36 <Eddi|zuHause> hm... "Assi-gnNode" looks like an odd hyphenation
17:10:42 <petern> on x86, mallocs are not all 16-byte aligned
17:11:05 <OwenS> petern: name one OS on which they are not yet SSE is enabled
17:11:18 <petern> maybe they are, and i'm thinking of other stuff
17:11:25 <petern> it's not gauranteed, though
17:11:31 <OwenS> If SSE is disabled, hypothetically they could be 8 byte
17:12:00 <CIA-1> OpenTTD: truebrain * r15027 /trunk/ (311 files in 30 dirs): (log message trimmed)
17:12:00 <CIA-1> OpenTTD: -Merge: tomatos and bananas left to be, here is NoAI for all to see.
17:12:00 <CIA-1> OpenTTD: NoAI is an API (a framework) to build your own AIs in. See:
17:12:01 <CIA-1> OpenTTD: With many thanks to:
17:12:01 <CIA-1> OpenTTD: - glx and Rubidium for their syncing, feedback and hard work
17:12:03 <CIA-1> OpenTTD: - Yexo for his feedback, patches, and AIs which tested the system very deep
17:12:12 <petern> (not gauranteed to be 16, that is)
17:12:27 <OwenS> POSIX: "The pointer returned if the allocation succeeds shall be suitably aligned so that it may be assigned to a pointer to any type of object and then used to access such an object in the space allocated (until the space is explicitly freed or reallocated)."
17:12:45 <OwenS> "it may be assigned to a pointer to any type of object and then used to access such an object in the space allocated"
17:13:49 <OwenS> On x86, it has to be 16 byte aligned because SSE has separate aligned and unaligned load instructions
17:14:09 <CIA-1> OpenTTD: truebrain * r15028 /branches/noai/: [NoAI] -Remove: NoAI branch, as it is now merged with trunk
17:14:17 *** Swallow has joined #openttd
17:14:29 <SmatZ> OwenS: you have no 128bit data type specified in the C standard
17:14:39 <SmatZ> so C standard can't say anything about this
17:15:06 <SmatZ> that is... you have char, short, int, long (and long long in C99)
17:15:13 <SmatZ> all with implementation-specific sizes
17:15:28 * worldemar dreams about long long long long
17:16:06 <OwenS> If your compiler supplies an 128-bit type (Which GCC and MSVC do), then according to both POSIX and the C Standard malloc must return suitable memory
17:16:28 <Swallow> Congratulations on getting NoAI in trunk!
17:16:52 <OwenS> Things are already getting pretty implementation specific if your using SSE anyway ;-)
17:17:42 <blathijs> OwenS: That sounds rather impossible, that libc malloc should behave different depending on what compiler is used
17:18:12 <OwenS> blathijs: libc and the compiler are considered as one unit with regards to the C standard's references to them
17:18:15 <Eddi|zuHause> next merge is cargodest :p
17:18:55 <petern> a0x9d4b008 (1) b0x9d4b018 (2) c0x9d4b028 (4) d0x9d4b038 (8)
17:19:03 <petern> 16 byte alignment, ending on 8? :p
17:19:46 <OwenS> I'm probably too used to the x86_64 ABI
17:20:12 <SmatZ> OwenS: all I can find in C99 specs is that pointers ... are aligned to be able to access data ... but nothing about the alignment as such
17:20:26 <SmatZ> like, on x86, you can access data aligned at 1B boundary (except x86)
17:20:29 <CIA-1> OpenTTD: rubidium * r15029 /trunk/src/lang/english.txt: -Cleanup (r15027): remove strings that aren't needed anymore
17:20:35 <SmatZ> err except some SSE instructions
17:21:14 *** Eddi|zuHause has joined #openttd
17:21:34 <OwenS> (1) 224d010 (2) 224d030 (4) 224d050 (8) 224d070
17:21:46 <OwenS> I'll admit I truncated off the top 4 bytes there :p
17:22:28 <petern> they're all 32 byte aligned
17:22:50 <SpComb> there's the malloc overhead
17:22:50 <petern> 16 byte aligned, then a 16 byte gap
17:22:56 <petern> the hidden stuff you never see
17:23:05 <OwenS> Depends upon the implementation
17:23:14 <OwenS> Two pointers, two size_ts, pointer and size_t,
17:23:24 <SmatZ> ...10 ...30 ...50 ...70 isn't 32B aligned :-P
17:23:55 <OwenS> 16 bytes is quite an important size though since it's the size of a cache line...
17:24:04 <petern> i meant 32 byte offset
17:24:40 <OwenS> O_o theyve been pushing it up
17:24:50 <OwenS> petern: cat /proc/cpuinfo
17:24:51 <petern> if that's a similar thing
17:24:54 <SmatZ> cat /proc/cpuinfo | grep cache :)
17:25:12 <petern> yeah, that only returns cache size
17:25:12 <worldemar> cache size : 4096 KB
17:25:12 <worldemar> cache_alignment : 64
17:25:17 <OwenS> address sizes : 40 bits physical, 48 bits virtual
17:25:31 <petern> i think you only see some things on amd64
17:25:43 <OwenS> cache size : 512 KB <- What happened to my L3 cache?
17:25:51 <OwenS> Intels are 36 physical IIRC
17:25:53 <CIA-1> OpenTTD: rubidium * r15030 /trunk/src/lang/ (43 files in 2 dirs): -Update (r15029): and from the other languages too (yay for WT2)
17:29:15 <TrueBrain> SmatZ: now check what the compile-time does in your graphs ;)
17:29:35 <petern> core 2 anyway, i dunno about xeons
17:29:46 <SmatZ> TrueBrain: +30k lines ;)
17:29:58 <petern> so what does noai allow now?
17:30:03 <petern> and will tron revert it? :p
17:30:23 *** HerzogDeXtEr has joined #openttd
17:30:28 <TrueBrain> I think I will book a ticket and kill him or something
17:31:28 <SpComb> someone on #mysql has 128GB of memory
17:31:57 <CIA-1> OpenTTD: smatz * r15031 /trunk/src/ (core/enum_type.hpp spritecache.cpp): -Fix (r15027): merge reverted r15026, so do it once again
17:32:12 <worldemar> SpComb: not at home, i think...
17:32:38 <yorick> how do I change graphics when landing aircraft?
17:32:52 *** energetic has joined #openttd
17:33:19 <planetmaker> congrats to all devs for NoAI trunk :)
17:33:23 <CIA-1> OpenTTD: truebrain * r15032 /trunk/src/lang/english.txt: -Fix (r15030): Rubidium was too script-happy ;)
17:33:59 <Zuu> Ideed, a good congras to all people behind NoAI :)
17:34:21 <yorick> mh, oh, congrats TB, glx, rubidium, yexo, morloth, TJIP, otherpeople
17:34:23 <planetmaker> I have a peculiar idea what the next game on our servers will be like :)
17:34:43 <SpComb> you could write an openttdcoop AI
17:35:00 <TrueBrain> planetmaker: any problems, let us know ;)
17:35:06 <OwenS> SpComb: What, one which builds rail networks of ridiculous complexity?
17:35:11 <planetmaker> SpComb: yeah... but there are some already :)
17:35:12 <TrueBrain> for all developers: 'make regression' does a big AI test for sanity ;)
17:35:19 <planetmaker> TrueBrain: be sure of that! :)
17:35:37 <OwenS> TrueBrain: It doesn't make a regression? =(
17:36:28 <TrueBrain> it does a regression check
17:36:31 <TrueBrain> two different things ;)
17:37:29 <TrueBrain> just your interpertation of it :)
17:37:54 <goodger> OwenS: suggest you symlink "check" to "make" in PATH :P
17:39:34 <frosch123> \o/ svn up does not conflict
17:39:41 <TrueBrain> concratz frosch123 ;)
17:40:26 <CIA-1> OpenTTD: truebrain * r15033 /trunk/src/ai/api/ai_object.cpp: -Fix (r15027): silent a warning when compiling without network
17:40:46 <CIA-1> OpenTTD: truebrain * r15034 /trunk/source.list: -Fix (r15027): NoAI framework no longer needs threads, so don't make source.list believe it does
17:41:15 <TrueBrain> anyway ... sorry all about the extra compile-time ;)
17:41:21 <TrueBrain> SmatZ: let me know when you have time-stats for that ;)
17:41:38 <Belugas> tron, i invoke you!!!
17:41:39 <SmatZ> TrueBrain: takes about 30 minutes...
17:41:49 <Belugas> Let See Your Mighty Wrath!!
17:47:01 *** tom0004 has joined #openttd
17:48:18 <planetmaker> [SRC] DEP ai/api/ai_object.cpp
17:48:20 <planetmaker> make[1]: *** No rule to make target `/Users/ingo/ottd/trunk.hg/src/3rdparty/squirrel/squirrel/sqvm.cpp', needed by `3rdparty/squirrel/squirrel/sqvm.d'. Stop.
17:48:50 <frosch123> your external fetch failed?
17:49:04 <planetmaker> So... I need squirrel?
17:49:05 <planetmaker> I got no complaint from hg
17:49:37 <TrueBrain> svn has 'svn externals'
17:49:39 <planetmaker> but there's no squirrel.h
17:49:43 <TrueBrain> so you need to do that yourself
17:49:56 <TrueBrain> in src/3rdparty, get 'squirrel' from svn://svn.openttd.org/3rdparty/squirrel
17:49:59 <planetmaker> ok... what do I need? Or where do I find what I need?
17:49:59 <TrueBrain> (of the hg variant :p)
17:50:04 <TrueBrain> we should make configure check for that I guess :)
17:50:16 <planetmaker> ok, will have a look :) - and yes, would be a good idea :)
17:50:48 <OwenS> TrueBrain: Is Squirrel staying? Or is it eventually gonna be replaced?
17:51:17 <TrueBrain> it will be replaced
17:51:22 *** Yeggstry has joined #openttd
17:51:26 <OwenS> Then again, my project will need it's own language anyway since I need to ban loops :p
17:51:48 <orudge> NoAI merged into trunk :o
17:52:17 <planetmaker> hm... where do I checkout that 3rd party thing to?
17:52:32 <yorick> does that still need to be done manually?
17:52:43 <TrueBrain> planetmaker: as I said: src/3rdparty
17:53:28 <yorick> so you mean it can't be done as automagically as it could?
17:53:30 <planetmaker> you might want to include that, so that a simple trunk co actually compiles :)
17:53:49 <glx> planetmaker: it works for svn
17:54:04 <TrueBrain> for svn, it is all automated
17:54:05 <planetmaker> hm... not for hg anymore :)
17:54:12 * OwenS wonders if people would complain that a simple RPN language was too complex for signals :p
17:54:17 <yorick> then remove hg completely
17:54:22 <blathijs> OwenS: You need to ban loops?
17:54:27 <Zuu> hmm hg qpush says "abort: local changes found, refresh first" but 'hg commit -m " "' says "nothing changed" and 'hg qpush' aborts with same abort message again. (I have qpop:ed all patches and made 'svn up')
17:54:30 <yorick> people will go complain
17:54:31 <TrueBrain> @kban 60 yorick hg removed
17:54:31 <DorpsGek> TrueBrain: Error: 60 is not in #openttd.
17:54:32 <OwenS> blathijs: Programmable signals
17:54:38 <TrueBrain> @kban yorick 60 hg removed
17:54:38 *** DorpsGek sets mode: +b *!~Yorick@s55924da0.adsl.wanadoo.nl
17:54:39 *** yorick was kicked by DorpsGek (hg removed)
17:54:52 <Zuu> previously I have fixed this with hg qpush --force, but that seams to give interesting file changes in my patches.
17:55:12 <Aali> hg is really stupid sometimes, try reverting
17:55:14 <TrueBrain> I hate people who can only wine ..
17:55:24 <Zuu> Aali: Reverting hg or svn?
17:55:37 <Zuu> I've made 'svn revert -R src'
17:55:38 *** DorpsGek sets mode: -b *!~Yorick@s55924da0.adsl.wanadoo.nl
17:55:46 <Aali> you have a hg queue in your svn repo or what?
17:56:11 <Swallow> Zuu: qrefresh might be what you need
17:56:18 <Zuu> Aali: Yes, so I update trunk using svn, but manages my queues with hg.
17:56:59 <planetmaker> I pull trunk using hg and have a few clones which pull from my trunk copy
17:57:02 <Zuu> Aali: Because I already know how to checkout with svn. :)
17:58:30 <Zuu> It would be good if 'hg diff' could tell me what files that block patch push. :/
17:58:57 <Aali> Zuu: and it does that in a hg repo :P
17:59:04 <yorick> then I can't use hg addremove anymore :/
18:00:07 <Aali> yorick: hg doesn't traverse sub-repos
18:00:09 <dihedral> TrueBrain: did you not miss two zeros on that kban command?
18:00:23 <Aali> it'll be like that repo isn't even there
18:00:28 <Zuu> Aali: But my idea was that i hg qpop to bottom, make a svn up, and then do hg commit -m "r12345" and then hg qpush me up again.
18:00:48 <Aali> Zuu: why would you do that?
18:02:07 <Zuu> Aali: Because i would not need to learn hg other than the queuing part. But my issues probably show that my shortcut was not really a shortcut. :)
18:02:18 <yorick> it doesn't even error when missing squirrel, only [make] error 2
18:02:29 <Aali> learning hg is not that hard :P
18:02:45 <Aali> and you'll never do it if you don't try
18:03:58 <planetmaker> seems to work now though with two warnings: /Users/ingo/ottd/trunk.hg/src/ai/api/ai_tile.cpp: In static member function 'static bool AITile::PlantTree(TileIndex)':
18:03:59 <planetmaker> /Users/ingo/ottd/trunk.hg/src/ai/api/ai_tile.cpp:216: warning: passing negative value '-0x00000000000000001' for argument 2 to 'static bool AIObject::DoCommand(TileIndex, uint32, uint32, uint, const char*, void (*)(AIInstance*))'
18:04:01 <planetmaker> /Users/ingo/ottd/trunk.hg/src/ai/api/ai_tile.cpp: In static member function 'static bool AITile::PlantTreeRectangle(TileIndex, uint, uint)':
18:04:03 <planetmaker> /Users/ingo/ottd/trunk.hg/src/ai/api/ai_tile.cpp:226: warning: passing negative value '-0x00000000000000001' for argument 2 to 'static bool AIObject::DoCommand(TileIndex, uint32, uint32, uint, const char*, void (*)(AIInstance*))'
18:04:14 <yorick> it doesn't compile here
18:04:22 <yorick> I'll test on another copy
18:04:29 <planetmaker> I did exactly what you posted, yorick ;)
18:04:48 <OwenS> TrueBrain, how about revising that 3600 upwards to 86400? :p
18:05:25 <Zuu> Anyhow, hg qpush -force, and then a diff of my patch before and after svn up shows no problems this time. So I can stick my head into the sand a little while more. :D
18:05:36 <CIA-1> OpenTTD: truebrain * r15035 /trunk/config.lib:
18:05:36 <CIA-1> OpenTTD: -Fix (r15027): if case people don't use SVN (which we advise), they do not
18:05:36 <CIA-1> OpenTTD: receive src/3rdparty/squirrel automaticly. Warn when not found, and advise what
18:05:36 <CIA-1> OpenTTD: to do (this to avoid stupid users like yorick get all confused and agressive)
18:05:43 <TrueBrain> OwenS: I get my fun out of other things ;)
18:06:21 <Forked> ah yes, calling the users stupid.. always fun.. but to their face? :p
18:06:50 <TrueBrain> planetmaker: nice error; I wonder why no other compiler noticed it
18:07:04 <TrueBrain> Forked: not 'their' .. a single person :) Which makes all the difference )
18:07:21 <planetmaker> Aeolus:~/ottd/trunk.hg ingo$ Your script made an error: the index 'SetSetting' does not exist <-- when starting the binary
18:07:46 <planetmaker> TrueBrain: because of that maybe: gcc --version
18:07:48 <planetmaker> i686-apple-darwin8-gcc-4.0.1 (GCC) 4.0.1 (Apple Computer, Inc. build 5363)
18:07:48 <SmatZ> planetmaker: update your AI
18:07:58 <planetmaker> SmatZ: I have no AI ;)
18:08:01 <TrueBrain> planetmaker: update your AIs ;)
18:08:32 <dihedral> TrueBrain: nice commit message :-)
18:08:39 <planetmaker> ok... never did that. Let me guess: RTFM ;)
18:09:12 <TrueBrain> planetmaker: no, if you really don't have an extra AI
18:09:12 <planetmaker> but where's TFM :)
18:09:14 <TrueBrain> it should not error
18:09:27 <planetmaker> TrueBrain: clean trunk. Nothing there at all.
18:09:40 <TrueBrain> but there was an API change today
18:09:47 <TrueBrain> any AI you download from the forum is most likely broken at this moment
18:09:52 <planetmaker> only thing is that hg clone ... which yorick posted to get squirrel there
18:09:52 <SmatZ> planetmaker: and in your ~/.openttd forlder?
18:09:55 <OwenS> TrueBrain: May I suggest that libraries have a minor version number also, which must be greater than or equal to the requested?
18:09:55 <TrueBrain> ls bin/ai gives you what?
18:10:09 <TrueBrain> ah the ~/.openttd/ai folder yes ;)
18:10:21 <planetmaker> library regression wrightai
18:10:28 <Zuu> TrueBrain: It's not wrightAI or does it use the AddSetting?
18:10:31 <planetmaker> ah... there.... let me see
18:10:32 <TrueBrain> OwenS: and what good does that do?
18:10:40 <TrueBrain> Zuu: WrightAI is changed already :)
18:10:48 <OwenS> TrueBrain, Allows you to add functions to a library without breaking backwards compatibility
18:11:01 <TrueBrain> fixed planetmaker's bug, in a way which is consistant through-out the code ;)
18:11:38 <TrueBrain> OwenS: minor versions are completely overrated .. for good reason we removed it from the saveload code :)
18:11:46 <TrueBrain> OwenS: you can do that already; just don't update the version at all ;)
18:11:59 <OwenS> Then some AI crashes when it tries to use a too new function :p
18:12:01 <planetmaker> TrueBrain: in ~/Documents/OpenTTD/ai there're obviously two old admirals ;)
18:12:09 <TrueBrain> OwenS: besides, you are allowed to run multiple library functions next to eachother :)
18:12:33 <OwenS> Minor versions work better with code than files :p
18:12:39 <TrueBrain> planetmaker: see ;)
18:12:56 <TrueBrain> OwenS: well, I can see why you want that, yes, but ... not at this stage :)
18:13:01 <yorick> don't mind the hg sha deprecation warning
18:13:06 <TrueBrain> I will put it on the list to consider :)
18:13:28 <TrueBrain> SmatZ: oh shit .. hmm .. gcc2 .. we did fix that at some point
18:13:33 <TrueBrain> just I guess it got reintroduced ;)
18:13:49 <OwenS> Better question: Why are people still using GCC2 ?
18:13:59 <OwenS> Are they also using Red Hat 9 and Debian Woody?
18:14:01 <planetmaker> ok, with the old AIs gone, there's no warning :) Ty TB
18:14:40 <TrueBrain> SmatZ: if you can check if moving code around a bit can help?
18:15:29 <CIA-1> OpenTTD: truebrain * r15036 /trunk/src/ (ai/api/ai_tile.cpp tree_gui.cpp): -Fix: use the same value for 'random' through-out the code for PlantTree
18:15:32 <TrueBrain> and ty planetmaker for letting me know the warning :)
18:15:37 <OwenS> TrueBrain: Theres GCC 4 for PowerPC also!
18:15:57 <OwenS> Why are they not using a modern compiler?!
18:16:01 <TrueBrain> OwenS: PowerPC != MorphOS :)
18:16:05 <TrueBrain> MorphOS happens to run on a ppc
18:16:21 <OwenS> What I'm saying is... theres no reason for them not to support GCC4
18:16:31 <TrueBrain> because their binary format is not supported by it
18:16:51 <planetmaker> OwenS: it's work...
18:17:03 <TrueBrain> is most likely the reason ;)
18:17:06 <OwenS> planetmaker: And GCC2 is a fossil
18:17:18 <TrueBrain> it is not that easy to port your morphos binary support stuff from gcc2 to gcc4 ;)
18:17:19 <glx> <TrueBrain> planetmaker: nice error; I wonder why no other compiler noticed it <-- win9x build should have it in the logs
18:17:26 <SmatZ> @kban owens you are fossil
18:17:26 <DorpsGek> SmatZ: An error has occurred and has been logged. Please contact this bot's administrator for more information.
18:17:28 <OwenS> TrueBrain: OK, add elf support to MrophOS :p
18:17:56 <TrueBrain> OwenS: see, that is a more sane thing to do ;)
18:18:30 <TrueBrain> SmatZ: it appears to be case sensitive for some weird reason :p
18:18:48 <SmatZ> @kban OwenS you are fossil
18:18:48 <DorpsGek> SmatZ: An error has occurred and has been logged. Please contact this bot's administrator for more information.
18:18:58 <TrueBrain> @kban OwenS you are fossil
18:18:58 *** DorpsGek sets mode: +b *!~OwenS@host86-164-125-149.range86-164.btcentralplus.com
18:18:59 *** OwenS was kicked by DorpsGek (you are fossil)
18:19:08 <TrueBrain> weird, it shouldn't error ..
18:19:09 <planetmaker> it helps to be admin, I was told...
18:19:12 *** DorpsGek sets mode: +o TrueBrain
18:19:32 <SmatZ> TrueBrain: I don't know how to fix that :-/
18:19:36 *** TrueBrain sets mode: -b *!~OwenS@host86-164-125-149.range86-164.btcentralplus.com
18:19:39 *** DorpsGek sets mode: -o TrueBrain
18:19:43 <TrueBrain> SmatZ: I only can ;)
18:19:48 <TrueBrain> you registered yourself to DorpsGek?
18:21:07 <DorpsGek> SmatZ: I don't recognize you.
18:21:25 <TrueBrain> I hate my SSL delay in IRC chat :(
18:21:29 <glx> you identified yourself ?
18:21:40 <TrueBrain> SmatZ: so register yourself to DorpsGek, and I can give you slightly more rights ;)
18:21:44 <TrueBrain> but now for real: FOOD!
18:21:50 <SpComb> what's wrong with chanserv?
18:22:33 <planetmaker> SpComb: nothing, but DorpsGek is more mighty than chanserv ;)
18:22:46 <planetmaker> I assume at least. Or more comfortable
18:22:48 <SpComb> DorpsGek lacks the magical powers of ChanServ
18:28:35 <Zuu> Oh, my compile time only raised form 11 to 15 minutes :)
18:29:35 * planetmaker will time it after foooood!
18:29:36 <Zuu> TrueBrain: But that is useless with only a laptop.
18:29:59 <Zuu> If I recall correctly ccache distributes the job over several PCs?
18:30:33 <glx> <TrueBrain> install ccache :) <-- ask Rubidium about ccache ;)
18:33:04 <Aali> hmm, full release build with some extra patches, 2:20
18:34:59 <Rubidium> yup... ccache is nice, until you notice that it caches too aggressively; like when you notice that with a full recompile without ccache you get lots of compile errors
18:35:48 <glx> or you search the cause of an assert ;)
18:37:27 <yorick> it doesn't compile here with mingw
18:37:40 <glx> then you failed somewhere
18:38:04 <glx> it always compiled for me, same with MSVC
18:39:24 <Belugas> it does not compile in here either
18:39:43 <Belugas> i've got no C++ compiler hehehe
18:39:56 <Wolf01> if you are waiting my hello you fail
18:40:15 <glx> you just did it so you failed
18:40:20 <Wolf01> congratulations devs for that wonderful commit!
18:40:22 <goodger> Wolf01: we've long since given up on expecting you to speak
18:40:43 <Wolf01> goodger, if you didn't notice, I speak when you're not here :P
18:41:03 <TrueBrain> and I can;t blame you Wolf01
18:41:30 <Wolf01> TrueBrain, very nice commit message :)
18:42:43 <Wolf01> so now you should have more spare time to help me developing the sloped stations
18:44:48 <OwenS> Banning & kicking me while I'm away!
18:44:58 <TrueBrain> had a nice dinner OwenS? :)
18:46:49 * OwenS wonders what other instructions he can splice into his string unit
18:57:32 *** [alt]buster has joined #openttd
18:57:35 <Belugas> fossile? somebody older than me??
18:57:37 *** [alt]buster is now known as [com]buster
18:58:54 *** [com]buster is now known as retsub[moc]
18:59:53 <glx> yorick: tried make clean ?
19:00:28 <yorick> I tried a new clone ;)
19:00:34 <petern> hmm, i guess 3rdparty makes hg much harder? :o
19:01:10 <TrueBrain> 'much' ... just a bit :)
19:01:26 <yorick> you need to do 2 clones
19:01:49 <Belugas> and then you'll end up a clown
19:01:57 <petern> i thought squirrel was being ditched anyway :/
19:01:59 <yorick> btw, #include "../../station_base.h"
19:02:22 *** Alberth has joined #openttd
19:02:45 <petern> note: pressing f6 without f-lock on does not achieve compilage
19:03:12 *** sigmund_ has joined #openttd
19:04:07 <petern> right so how do AIs work?
19:05:30 <goodger> petern: when computers become powerful enough and gather enough information, they spontaneously develop consciousness
19:05:59 <goodger> or at least, so is the central premise of a number of star trek episodes and a film :S
19:07:08 <SHRIKEE> so when can we expect a streetview interface for openttd?
19:07:38 <OwenS> SHRIKEE, When we develop an algorithm for extracting perspective from 2D sprites and filling in back faces
19:07:54 <goodger> OwenS: I believe that's been done already :S
19:08:13 <OwenS> And when CSI share with us their image enahncement algorithms
19:08:25 <goodger> someone used it to add extra detail to videos by transplanting still photographs into them
19:08:44 <goodger> SHRIKEE: so you should expect it on the seventeenth of july 2011
19:08:48 <OwenS> goodger: Yeah, but they haven't developed a back face filler :p
19:09:01 <petern> do we only get WrightAI?
19:09:18 <goodger> maybe the friday before...
19:09:28 <TrueBrain> petern: for now; yes. the rest is on the forum :)
19:09:49 <frosch123> there is currently no working AI on the forum :p
19:09:53 <SHRIKEE> i cleaned up my install the other day... stripped out 200mb of i dunno what
19:09:57 * OwenS wonders how hes gonna shift 33k passengers away from this station
19:09:57 <SHRIKEE> but ingame nothing changed
19:10:07 <SHRIKEE> i guess it's all mods and stuff
19:10:23 <goodger> OwenS: quadruple the size and run maglevs
19:10:45 <TrueBrain> Yexo: update your AI already :p
19:10:51 <petern> we need to replace wrightai with something more general
19:11:04 <OwenS> goodger, It's Maglev (well, Shinkansen), and I can't quadruple the size because theres CBD on all 3½ sides
19:11:07 <petern> do the AIs understand the difficulty settings?
19:11:25 <OwenS> WrightAI makes me think of two things: The Wright Flyer, and Will Wright...
19:11:43 <SHRIKEE> hmmm a 64x64 map is a bit smaller than i thought :O
19:11:49 <TrueBrain> petern: understand .. well .. they can read it :)
19:11:50 <petern> OwenS, it's aircraft only. take a guess...
19:11:56 <TrueBrain> OwenS: good thought :)
19:13:29 <TrueBrain> petern: make one ;)
19:15:48 <OwenS> Even worse: Said station with 32k passengers is at capacity already (and has 8 platforms full of 11 tile long trains)
19:17:05 <goodger> OwenS: extend the station
19:17:16 <yorick> there is no such thing as aircrafts :p
19:17:19 <goodger> or add a parallel line of buses
19:17:29 <Yexo> petern: try a more flat map
19:17:34 <yorick> at least now it knows it fails and tries to bankrupt itself asap
19:18:04 <OwenS> goodger: Extend the station where: Into city, into city, into city and canal, or into entrance junction?
19:18:29 <OwenS> It's landlocked on all sides
19:18:36 <Aali> petern: AIs are affected by competitor build speed
19:18:48 <Aali> and start time, of course
19:18:48 *** yorick was kicked by DorpsGek (hello)
19:19:08 <Forked> yorick kick #51256125,2
19:19:11 <petern> what happens to old save games currently?
19:19:33 <Yexo> a random AI is started to replace the original ai
19:19:37 <TrueBrain> a new AI is attached to the AI company, and if it can handle what it has, it will, else .. well .. nothing really :)
19:20:05 <OwenS> TrueBrain: So, if Wright AI was chosen, would it crash, or would it just ignore what was there and build airports? :p
19:20:18 <Yexo> wrightai will ignore the stuff there already is
19:20:24 <petern> i can't load a 0.6.3 save
19:20:29 <TrueBrain> OwenS: WrightAI is an example AI; it ignores EVERYTHING and always starts like it was its first day on the job
19:20:49 <worldemar> and builds only airports?
19:20:51 <OwenS> I just selected WAI because it's all thats included ATM
19:20:52 <TrueBrain> petern: trace it why not :)
19:22:10 <petern> invalid chunk size, so far
19:22:18 <petern> just setting breakpoints
19:24:04 <TrueBrain> so either Rubidium did something bad, or something is funny :)
19:26:34 <CIA-1> OpenTTD: smatz * r15037 /trunk/ (4 files in 3 dirs): -Cleanup: remove unused file src/table/ai_rail.h
19:27:35 <TrueBrain> petern: company_base.h
19:27:36 <TrueBrain> - Company(uint16 name_1 = 0, bool is_ai = false);
19:27:36 <TrueBrain> + Company(uint16 name_1 = 0, bool is_ai = false, bool is_noai = false);
19:27:38 <TrueBrain> does that help at all?
19:29:28 <TrueBrain> petern: company_cmd.cpp
19:29:28 <TrueBrain> - this->is_noai = true;
19:29:28 <TrueBrain> + this->is_noai = false;
19:30:52 <petern> the companies all get renamed to WrightAI
19:31:06 <TrueBrain> part of WrightAI, yes
19:31:12 <TrueBrain> write a better AI ;)
19:31:12 <CIA-1> OpenTTD: truebrain * r15038 /trunk/src/ (company_base.h company_cmd.cpp): -Fix (r15027): loading older savegames failed
19:31:28 <petern> 0.6.3 is not exactly old an uncommon
19:31:50 <TrueBrain> I guess we should retrigger the nightly, as this bug is kind of nasty ;)
19:32:15 <TrueBrain> even a savegame from yesterday would fail to load ;)
19:32:23 <SmatZ> TrueBrain: it's better to assign values in the constructor initializer list (or how is it called ;)
19:32:33 <petern> you should put a warning up
19:32:39 <petern> anyone who plays with AI should not bother
19:32:43 <petern> becuase currently it's not playable
19:33:10 <TrueBrain> petern: it really is easier to just retrigger a nightly in that case ;)
19:33:56 <petern> why? it won't help people who play with AIs
19:34:19 <Zuu> Since there are people out there that reads the commit log, maybe include such a warning in a commit that is somewhat related to the warning :)
19:37:15 <mikegrb> 0.................................
19:37:16 <TrueBrain> then I don't follow you
19:37:27 <TrueBrain> oh, I think I get it
19:37:35 <TrueBrain> just download AdmiralAI, and I think you will have fun
19:37:40 <TrueBrain> @kick mikegrb goodbyte
19:37:40 *** mikegrb was kicked by DorpsGek (goodbyte)
19:37:46 <Yexo> wait a bit with that, current version won't work
19:38:01 <petern> we need a working general purpose AI in trunk ASAP
19:38:29 <Yexo> petern: writing a new one is a *huge* task
19:38:41 *** mikegrb has joined #openttd
19:38:42 <petern> it was easy enough to remove one :p
19:38:54 <SpComb> DorpsGek is pretty trigger-happy today
19:38:55 <mikegrb> elbow on the other keryboad
19:39:12 <TrueBrain> SpComb: you considered those 0's useful?
19:39:33 <SpComb> the kick, rejoin and apology was almost as much noise
19:39:50 <TrueBrain> mikegrb: weird place for your elbow :p
19:40:05 <TrueBrain> (hitting the 0 and enter :p)
19:42:51 <Zuu> Still there is a dot key in-between zero and enter on most keyboards.
19:43:23 <TrueBrain> my point exactly :)
19:47:38 <Steve-N> the noise after the kick rejoin and apology was even more noise
19:48:34 <OwenS> And the commenting about the noise after the kick rejoin and apology was just even more
19:48:35 <yorick> the noise of saying that again is even more noise
19:48:48 <TrueBrain> I am starting to feel trigger happy again
19:49:37 <CIA-1> OpenTTD: smatz * r15039 /trunk/src/ (5 files): -Codechange: fix comments regarding old AI, remove one old AI hack
19:50:29 <Steve-N> so anyone interested in testing my aggressive conductors patch (FS#2530) ?
19:51:58 <SmatZ> Steve-N: it is similiar to not using Full-Load orders, isn't it?
19:52:20 <Steve-N> it only does something when not using full-load
19:52:51 <Steve-N> vehicles will stop loading and loading and loading tiny little pieces of cargo when not using full-load with this option
19:53:20 <OwenS> How hard would it be to recreate the original AS (Artificial Stupidity) with NoAI? :p
19:53:49 <TrueBrain> as said a few days ago
19:53:53 <SmatZ> Steve-N: does it have a significant effect on gameplay?
19:54:37 <SpComb> opponent_vehicle->InduceRandomBreakdown();
19:54:51 <Steve-N> depends on how you play the game :) when using non-full-load orders and multiple trains serving a single platform (or platforms in different directions with cargodest), trains actually will leave the station without waiting for full load
19:57:13 <petern> OwenS: and the savegame data it uses is lost
19:57:16 <frosch123> ohoh, now they will keep downloading old realeases to create some asciiart with the numbers or similiar oO
19:57:28 <Belugas> Steve-N, I do not like the "agressive" part name of your patch. does not really reflect the feature...
19:57:34 *** Brianetta has joined #openttd
19:57:43 *** yorick is now known as Guest150
19:58:02 <SmatZ> TrueBrain: weeeheee stats! :)
19:58:19 * OwenS wonders to himself why towns eat so much CPU
19:59:21 <Steve-N> hm yeah i came up with the name after someone made a remark about the conductors being very friendly (though another word was used, can't remeber it and my irc client didn't log this afternoon;)
19:59:59 <frosch123> Steve-N: it would also be better if it were a loading option in the orders gui, instead of a global setting for all companies
20:00:35 <SmatZ> and then set is as default...
20:00:38 <Steve-N> hmm, yes that might be nice
20:00:42 <SmatZ> it can be stored in the Player struct though
20:00:48 <OwenS> I was about to go "We have enough memory for that?", before realising "this ins't the map array..."
20:00:58 <Steve-N> as no-one seemed to agree with me that it's a bug, i made it configurable and disabled by default
20:00:59 <SmatZ> so it is set per-Company
20:01:38 <frosch123> well, it has been long since the last "load xx%" patch showed up
20:02:09 <frosch123> though with cargodest, that would be silly I guess
20:02:41 <Steve-N> especially with cargodest, default loading really screws up time-tabling
20:03:25 *** welshdragon has joined #openttd
20:03:27 <Aali> a better solution to that may be a timetabling option "always leave on time"
20:03:51 <Aali> which ignores full load or whatever and just leaves the station when the time is up
20:04:38 <Steve-N> i think that would be a nice additional option (or order). i don't want my trains to leave before all waiting passengers have boarded
20:05:02 <Steve-N> (meaning all passengers that were already waiting before the train arrived at the station)
20:05:15 <mrfrenzy> a train *can't* leave until all waiting passengers have entered
20:05:28 <mrfrenzy> as it will be impossible to close the door until they have stepped clear
20:05:30 <Steve-N> it could when forced to keep it's timetable
20:05:46 <mrfrenzy> (going for realism here)
20:05:53 <Steve-N> go to tokyo, take the subway, and be amazed
20:06:13 <mrfrenzy> well I've only used public transport in .se, .de and .no
20:06:19 <Steve-N> (trying more realism here also)
20:06:23 <Swallow> I agree with Aali here that it might be better to make it a timetabling option.
20:06:32 <planetmaker> Steve-N: but there you have a "shoving person" at each door ;)
20:06:44 <OwenS> I disagree with Aali as I don't use timetables :p
20:07:00 <Steve-N> well here (.nl) the conductor usually does give people a chance to get on the train. but when you're just running off the escalator when he blows his whistle, the train won't wait
20:07:23 <Aali> the great thing about such a solution is that you can set up a timetable that only specifies loading times
20:07:31 <Rubidium> Steve-N: in Tokyo they wait till the train/metro is full
20:07:43 <Rubidium> or when there are no people coming anymore
20:07:47 <Aali> so even if you dont play with timetables normally, you can still use it
20:08:24 <Steve-N> so when the train isn't full, it will just keep waiting for all slow people coming in the station?
20:09:05 <Rubidium> Steve-N: effectively yes
20:09:16 <Belugas> [15:07] <Steve-N> (trying more realism here also) <-- freaks
20:09:27 <Steve-N> even when the next train is waiting behind it already?
20:09:33 <mrfrenzy> Steve-N: ofcourse, but if there are people currently going into the train, less than 1m from the doors, they will not close if the train is not full
20:09:36 <Steve-N> yeah i'm a realism freak, sorry
20:09:44 <Rubidium> though people run to the train when it comes and when they can't make it they are far from the train when it leaves
20:09:54 <Belugas> play sims then... not openttd
20:10:06 <Steve-N> oh, wow. here (.nl) trains will just leave, you just should have run faster
20:10:42 <TrueBrain> depends on how nice you are
20:11:05 <petern> now vs won't run it :(
20:11:07 <Steve-N> actually i used to play (a hugely modified verison of) simcity a lot, until my linux server crashed and i trashed my windows workstation to build a new linux server. and then started playing openttd ;)
20:11:16 <TrueBrain> petern: you broke it!
20:11:25 <petern> what've i done wrong? i just get a black console window pop up and then disappear...
20:11:28 <Belugas> well... a little rule, it 's not aimed at realism ;)
20:11:49 <Steve-N> that's why i made it an option, so you can choose!;)
20:12:10 <Steve-N> but even when you don't want realism, it generates more income for your company!:)
20:12:23 <Zuu> Aali: Exactly how I use timetables today, setting a minimum wait time at stations.
20:12:35 <Belugas> strange... i'd be more inclined to have a feature that cost more...
20:12:39 <dyzdyz> i've got a little question
20:12:52 <Aali> yes, but you cant set a maximum wait time.. which sucks sometimes
20:13:00 <TrueBrain> dyzdyz: make sure it is little ;)
20:13:01 <Belugas> i've got a little black book with my poems in...
20:13:04 <Steve-N> (actually i really don't care about company income, as it's a lot more than i can spend after say a month)
20:13:06 <dyzdyz> i've got a scenario with 2048^2 map of europe
20:13:28 <dyzdyz> but i don't know how to use generictrams with it
20:13:36 <dyzdyz> i can use some other grf's
20:13:53 <dyzdyz> what can be a problem here?
20:13:59 <petern> where's the entry point for win32?
20:14:34 <Belugas> dyzdyz, on which version do you play your scenario? P.S.: latest is not am option
20:15:09 <petern> start up project was set at strgen :p
20:15:10 <dyzdyz> i use h3b244a8f Cargodest
20:15:25 <dyzdyz> and Generictrams works fine with generated map
20:16:13 <Zuu> Aali: Why I don't like to use hg more than necessary is that the versions are not sequential.
20:16:17 *** retsub[moc] has joined #openttd
20:16:38 <Zuu> Not in a readable form what i can tell.
20:16:42 <dyzdyz> ok, just found solution for myself
20:16:47 <Steve-N> dyzdyz: would that be CarstsEuropeScenario?
20:16:53 <dyzdyz> sorry for interrupting your conversation
20:16:55 <Belugas> when yuo've started your map, was there already a generic tram set specified in your grf seting? If not, i think you're screwed to re-strt tye scenario
20:16:57 <Steve-N> (no problem with those trams and that map here anyway)
20:17:11 <Aali> Zuu: the little number before the hash in hg log is also a valid identifier ;)
20:17:33 <dyzdyz> Belugas: i seems to conflict with longvehicles
20:22:04 * Belugas is not an expert at mixing grfs
20:24:36 <Steve-N> dyzdyz: it does? i don't get any warnings or errors when adding long vehicles to a game where generic trams (0.4) are already used
20:27:42 <Steve-N> what generic trams version are you using, dyzdyz? (as i'm using (a modified version) of h3b244a8f too, with generic trams 0.4, without any problems. even tried it on windows..)
20:28:36 <dyzdyz> Steve-N: generictrams 0.4
20:29:05 <dyzdyz> Steve-N: when i disable longvehicles then generictrams works fine
20:29:18 <Steve-N> hm, strange, same here.. and where did you get those long vehicles?
20:29:29 <dyzdyz> Steve-N: and i don't get any warnings
20:29:48 <dyzdyz> Steve-N: GRFCrawler i suppose
20:30:12 *** Moodles has joined #openttd
20:30:15 <Steve-N> oh, ok, than we seem to have the same result (but i'm not running any of the vehicles from those long vehicles)
20:30:53 <dyzdyz> Steve-N: what modification have you done to h3b244a8f?
20:31:47 *** sigmund_ is now known as sigmund
20:32:44 <Steve-N> but that shouldn't make any difference for grf's
20:36:08 *** tom0004 has joined #openttd
20:44:40 <michi_cc> TrueBrain: are the changes to projects/openttd_vs80.sln in r15027 wanted?
20:45:19 <petern> they may well be fixed later
20:45:32 <TrueBrain> michi_cc: why wouldn't they?
20:45:59 <TrueBrain> michi_cc: ah, I see what you mean
20:46:02 <TrueBrain> michi_cc: ask glx ;)
20:47:20 <petern> it gets changed to that automatically for me
20:51:29 *** [com]buster has joined #openttd
20:54:29 <michi_cc> glx: it disables x64-builds for vs2005, I'd say that wasn't meant that way
20:56:19 *** [com]buster is now known as retsub[moc]
20:56:25 <SmatZ> someone please fix that sln file, so we have r15040, and I can run another benchmark ;)
20:56:27 *** retsub[moc] is now known as [com]buster
20:58:32 <DorpsGek> Belugas: someone was last seen in #openttd 50 weeks, 3 days, 2 hours, 6 minutes, and 9 seconds ago: * Someone here is gay
20:59:27 <SpComb> mm, 50 weeks old, it's a finely aged joke
21:00:14 <Belugas> like old wine and men, getting better while aging :)
21:01:13 <nicfer> What happened to the custom bridgeheads patch?
21:02:02 <Belugas> waht are the possible answers?
21:02:23 <frosch123> everyone abandoned it
21:02:29 <glx> michi_cc: it's TrueBrain's fault I think
21:02:49 <glx> I can't find any similar changes in noai nor trunk before the merge
21:04:20 <DorpsGek> Zuu: Invalid arguments for _commit.
21:04:40 <DorpsGek> glx: Commit by glx :: r12826 /branches/noai (80 files in 5 dirs) (2008-04-21 21:15:50 UTC)
21:04:41 <DorpsGek> glx: [NoAI] -Sync: with trunk r12780:12824
21:04:47 <DorpsGek> Belugas: seen [<channel>] <nick>
21:05:21 <Steve-N> so belugas, what about these reality freaks
21:05:35 <Steve-N> 'For once, TTD almost looked real,' is what google comes up with ;)
21:06:17 <Belugas> look at the proprotions of the vehicles/houses/else
21:06:20 <nicfer> Do the devs like that patch or is it like tracks over tunnel entrances?
21:06:25 <glx> michi_cc: feel free to revert this change (you're the one with an x64 ;) )
21:07:06 *** lobstar has joined #openttd
21:07:26 <Steve-N> oh proportions, i've been looking at those... why are the catchment areas this unrealistically small?
21:08:06 <frosch123> Steve-N: use distant-join to make them unrealistically huge
21:08:19 <Belugas> because it's a gaaaaaameeeeuh
21:08:52 <Steve-N> distant-join so you can build several stations which are one for the game?
21:09:07 <Steve-N> game doesn't mean it can't be at least trying to be a little bit realistic
21:09:39 <Belugas> and how would you feel if we'd make it so that the bridges would take REALISTIC time to be constructed? like... 2-3 years?
21:09:50 <Steve-N> so in like 1980 when ttd was build, maps couldn't be large so large catchment areas wouldn't make sense. but now we can have huge maps, right?
21:09:52 <nicfer> If you want a realistic game buy Crysis ;)
21:10:21 <SpComb> not very realistic what with the keyboard and mouse
21:10:29 <SpComb> and you don't even die if you get shot in-game
21:10:38 <Zuu> Steve-N: Would you really want a game with all scales 1:1?
21:11:38 <nicfer> Heh, not even Second Life is 100% realistic
21:11:57 <Zuu> Steve-N: Then lets apply for an education in transportation and become a transportation planer :)
21:12:09 <CIA-1> OpenTTD: michi_cc * r15040 /trunk/projects/openttd_vs80.sln: -Fix (r15027): Don't disable support for Win64 in the VS2005 project file.
21:12:11 <TrueBrain> glx: how could you even consider blaming me :) I never ever touched such files :)
21:12:20 <Steve-N> 100% realistic wouldn't be possible without it actually being real, but a tiny little bit closer than we are now would be nice...
21:12:23 <Belugas> Steve-N, well... 1:1 scale means TTD will loose it's curent touch
21:12:29 <Belugas> and it will become BOOOOORINY
21:12:33 <Zuu> Steve-N: That is what I am doing ATM actually
21:13:30 <Steve-N> ATM? automated teller machine?
21:13:45 <Belugas> Steve-N, no, making it closer to reality will just make it loose its essence. There is no reason nor need to make it more real. It's lready popular. it will not gain more popularity with more realism. Look at Locomotion. It's closer to reality. It did not gain as much followers as Opent
21:14:07 <Belugas> nor patch nor original TTD
21:15:30 <Steve-N> Belugas: it is? so i should get a windows machine and try locomotion? damn...
21:16:04 <Belugas> yeah, and leave us alone with the R word
21:16:26 <nicfer> We should focus on making city growth harder in temperate
21:16:44 <Belugas> i dont mind adding stuff that enhance game play, but adding stuff just to make it closer to R is ... <VOMIT NOISE>
21:17:10 <Steve-N> making it closer to realism might even enhance gameplay
21:17:20 <Zuu> nicfer: I have never understood why people so much favor temperate over the other climates.
21:17:38 <Steve-N> Zuu: maybe because most of them are used to living in there
21:18:32 <Steve-N> so about this realism thing... i guess cargodest won't have any chance of ever getting into trunk then?
21:18:55 <Aali> cargodest is not a realism thing
21:19:21 <Zuu> Steve-N: There is a difference between argumenting for something only because it adds realism, or if realism is a bi-product of a gameplay enhancement.
21:19:21 <Steve-N> to me it's a lot more realistic when passengers actually want to go somewhere, instead of paying me for dropping them off anywhere
21:19:54 <Alberth> Steve-N: the keyword is that it must be FUN to play
21:20:04 <Zuu> Steve-N: Still in cargodest where they want to go is highly dependent on what options you provide.
21:20:10 <Aali> it IS more realistic, but it would never have been created if there wasn't any gameplay value
21:21:14 <Belugas> by-product of a gameplay enhancement
21:21:30 <Belugas> newindustry/newcargo : same thing
21:21:37 <Steve-N> sure. i never said realism is the only goal. but in a lot of cases, realism will add to more gameplay value
21:21:50 <Belugas> it's the other way around
21:22:16 <Aali> and yet "everyone" on the forums is crying for more realism in cargodest :/
21:22:39 <Steve-N> :) because realism is fun gameplay!
21:23:04 <Steve-N> before cargodest, i didn't really care about openttd. was way too easy. now with cargodest, it's becoming interesting
21:23:14 <Alberth> Aali: because the current settings is too difficult, and users have no control over it
21:23:50 <Alberth> Steve-N: 'interesting', not 'realistic'
21:24:02 <Zuu> Steve-N: The reason you care for OpenTTD with cargodest is that it has become more difficult and have added gameplay value. This is not strictly connected to realism.
21:24:37 <Aali> cargodest makes OTTD more difficult?
21:24:52 *** tom0004 has joined #openttd
21:24:56 <glx> <TrueBrain> glx: how could you even consider blaming me :) I never ever touched such files :) <-- but you are the one in the log and in svn blame ;)
21:25:08 <Aali> that depends on your playstyle
21:27:29 <mrfrenzy> will this not decrease profit?
21:27:51 <Steve-N> if you're expecting your vehicles to wait until full, without full-load order, it might
21:28:49 <Steve-N> in my case, it hugely increases profit, as there won't be any trains queueing up and cargo (passengers) actually are being moved
21:32:35 *** [com]buster has joined #openttd
21:36:33 <worldemar> if i transfer some cargo from A to B, and then from B to C, where cargo is accepted, then total income from moving A -> C is equal to direct transfer from A to C? (assume transfer times are equal)
21:37:47 <nicfer> Maybe if temperate towns require food or goods it will be more challenging
21:37:56 <worldemar> okay, and when cargo moves from A to B... transport gives income or not?
21:38:14 <Yexo> you don't get any money at that point
21:38:48 <worldemar> so, if B is very close to C, then little line from B to C will give me all the money, including A-B transfer?
21:38:57 <mrfrenzy> you can however use a transfer order for a-b
21:39:07 <mrfrenzy> and the a-b train will get "virtual money"
21:39:22 <mrfrenzy> which will then be deducted from the huge profit b-c train otherwise would have made
21:39:57 <CIA-1> OpenTTD: truebrain * r15041 /trunk/src/company_cmd.cpp: -Fix (r15038): use the ctor auto thing how it is meant to be used (tnx to SmatZ for pointing that out)
21:40:11 <CIA-1> OpenTTD: truebrain * r15042 /trunk/src/ai/ai_info.cpp: -Fix (r15027): don't leak memory in AIInfo (Yexo)
21:40:49 <worldemar> if transfer order 1) is set: i can take money even if cargo dosen't reach destination 2) is not set: all money will income only on arrive to destination
21:41:18 <worldemar> maybe there is wiki page for it? sorry for flooding here)
21:41:46 <Zuu> worldemar: Watch your economy window to see when you earn the money.
21:42:19 <worldemar> i am making huge network of train lines. there are some routes i can do only by ransfer through several stations...
21:42:20 <Zuu> Yellow money (transfer) is not earned for the company until the passenger/goods reach their final destination.
21:49:10 <worldemar> Zuu: so, "transfer" is to share profit between two trains without making one of them "negative-profit"
21:50:12 <Yexo> there is another difference: without using transfer, if the 'middle' station accepts the cargo it'll vanish, but with the transfer option the cargo will always stay at the station
21:53:28 <worldemar> "feeder service" wiki-page was really helpful, thanks!
21:56:02 *** ProfFrink has joined #openttd
21:58:42 *** frederyk has joined #openttd
22:01:32 * Belugas runs home, hoping his repo over there is not going to be too messy
22:03:09 *** ProfFrink is now known as Prof_Frink
22:13:46 *** sigmund_ has joined #openttd
22:16:32 * worldemar remembers about work tomorrow and saves his transport company
22:34:38 *** [com]buster has joined #openttd
22:56:20 *** Ridayah has joined #openttd
23:00:40 *** Terkhen has joined #openttd
23:44:36 *** goodger has joined #openttd
23:56:20 *** Dred_furst has joined #openttd
continue to next day ⏵
