IRC logs for #openttd on OFTC at 2012-03-10
⏴ go to previous day
00:25:19 <morph`> Possible to compile and install JUST dedicated server of OpenTTD?
00:36:16 <frosch123> yes, if you are on *nix
00:36:27 <frosch123> configure has a --dedicated option or similar
00:37:41 <frosch123> actually "./configure --enable-dedicated"
00:53:49 *** Rhamphoryncus has joined #openttd
00:54:01 <morph`> Well, compiling doesn't work for some reason on my Debian Squeeze - I get
00:54:20 <morph`> I can't "make" after Configure ends
04:09:13 <Nat_aS> is there a max height for trains before they start hitting bridges?
04:14:03 <Rhamphoryncus> nothing in the game mechanics. Just aesthetic
04:14:32 <Rhamphoryncus> If you really look close you'll realize that all trains are several times taller than tunnels/bridges :)
04:17:31 <Rubidium> no try that in a rendered 3D world
04:22:48 <Eddi|zuHause> Nat_aS: i think your side view is backwards
04:23:04 <Eddi|zuHause> it looks in <- direction, but should be -> direction on that spot
04:23:48 <Nat_aS> for the lead engine or the tail car?
04:24:16 <Nat_aS> the one facing > is the lead engine, the one facing < is the tail car
04:24:22 <Nat_aS> so they should be the other way around?
04:25:05 <Eddi|zuHause> that. or your preview is backwards
04:25:31 <Nat_aS> the preview is just copy paste
04:25:38 <Eddi|zuHause> in depot, the train is shown in <- direction
04:25:56 <Eddi|zuHause> so i just assumed your full train is like that
04:26:16 <Nat_aS> I just made it facing the direction the picture it is based on is.
04:28:14 *** ricky26 has joined #openttd
05:10:14 *** Maarten has joined #openttd
05:34:13 *** snorre_ has joined #openttd
05:56:23 *** Eddi|zuHause has joined #openttd
07:02:04 *** andythenorth has joined #openttd
07:10:25 *** sla_ro|master has joined #openttd
07:16:26 *** tokai|mdlx has joined #openttd
07:19:03 <Nat_aS> making rounded edges in isometry is hard.
07:27:14 *** tokai|noir has joined #openttd
07:27:14 *** ChanServ sets mode: +v tokai|noir
07:35:21 *** TWerkhoven has joined #openttd
07:39:14 *** tokai|mdlx has joined #openttd
07:39:56 <Rhamphoryncus> If only the Nazis had put their effort towards engineering and not world domination ;)
07:48:52 <andythenorth> why does the trees tool always build random trees, rather than the selected type?
07:49:19 <planetmaker> andythenorth: the first tree on a tile should be of the selected type
07:49:31 <planetmaker> the subsequent ones then follow a certain tile layout pattern
07:50:15 <andythenorth> that is the case
07:50:46 <planetmaker> there's like two dozen different tree tiles. That's it. Times growth stages
07:51:05 <Nat_aS> well in OTTD you can live out the alternate history where Germany was treated fairly after WWI and nobody got buthurt, and all the crazy scientists got nice jobs building crazy shit to rebuild postwar europe.
07:51:23 <Nat_aS> also hitler got accepted into art school.
07:51:53 <Nat_aS> and won a charlie chaplin lookalike contest.
07:52:22 <andythenorth> it's interesting
07:52:29 * andythenorth is feeling flamey today
08:21:31 * Rubidium wonders whether he'd be be welcomed in andy's club now ;)
08:23:29 <andythenorth> I was just surprised I got there first :P
08:27:39 <andythenorth> ...not to mention lots of new newgrf features, continued development of nml, and lots of newgrf developers trying to complete thing to release with next stable
08:33:21 *** valhallasw has joined #openttd
08:40:04 <Rubidium> if you got custom bridge heads and building stuff on top of a tunnel entrance
08:40:15 *** Booth is now known as Chris_Booth
08:40:23 <Rubidium> wouldn't then the best next step be: a bridge head that's also a tunnel entrance?
08:41:12 <Rubidium> that'd be really useful
08:41:54 <Rubidium> and then you're actually (almost) a sort of subway implemented
08:42:27 <Rubidium> just build a row of 0 long tunnels with signals in the entrance
08:42:42 <Rubidium> and build 0 length bridges over that with signals on the bridgehead
08:42:57 <Rubidium> or am I going insane now?
08:44:17 *** zxbiohazardzx has joined #openttd
08:51:36 *** Devroush has joined #openttd
08:53:38 *** kkb110_ has joined #openttd
08:54:40 <andythenorth> Rubidium: you'd need to be able to depth -1
08:54:53 <andythenorth> please don't accidentally do subways
08:54:59 <andythenorth> I like mocking it too much :(
08:58:35 *** Alberth has joined #openttd
08:58:35 *** ChanServ sets mode: +o Alberth
09:01:30 <zxbiohazardzx> moggle, you dutchies?
09:01:39 *** Progman has joined #openttd
09:05:51 <Rubidium> what makes you think that? The company I work for is 50% Dutch and over the last years I've learned quite a few Dutch words ;)
09:06:08 <zxbiohazardzx> :) dutchies rule the world :)
09:06:36 <zxbiohazardzx> and i thought so because moggel is from the KPN commercial, goeiemoggel :P
09:06:46 <zxbiohazardzx> usually its goedenmorgen or morgen :P
09:13:03 <zxbiohazardzx> Rubidium regarding your reply in the multithreading
09:13:24 <zxbiohazardzx> arent the ships heavy on CPU if you have alot of them (mainly due to pathfinding for ships?
09:18:34 <Rubidium> zxbiohazardzx: only under certain circumstances. With proper buoy usages it's actually not that big
09:19:00 <Rubidium> even then, most games I see forbid ships, don't use ships or use ships for very short distances
09:19:18 <zxbiohazardzx> fair enough, it just comes down to the old "too much work for too little benefit"
09:19:41 <zxbiohazardzx> can i ask someone to try to apply a diff and see if it compiles fine?
09:19:52 <Rubidium> well, you'd still need to add loads of synchronisation
09:20:18 <Rubidium> e.g. arrival at stations needs to happen in the same order everytime
09:20:28 <Alberth> andythenorth: number 1200 !
09:20:34 <Rubidium> and you'd need synchronisation for updating the vehicle location caches
09:20:59 <Alberth> andythenorth: it looks like a nice definition to me :p
09:21:07 <Rubidium> all have a negative impact on all other vehicles
09:21:41 <andythenorth> Alberth: I might try coding it later
09:21:54 <andythenorth> so my main.py might become a dispatcher script
09:22:06 <zxbiohazardzx> im not that much of a coder Rubidium
09:22:07 <andythenorth> and the gestalts should probably learn how to run standalone
09:22:13 <zxbiohazardzx> i just updated the house patch
09:22:20 <zxbiohazardzx> till then i hardly did coding on ottd
09:22:33 <zxbiohazardzx> i did do a minor software engineering and got around java alot
09:22:38 <zxbiohazardzx> and i do understand the cpp part
09:22:52 <Alberth> andythenorth: stand-alone? you want to run a gestalt as a child process of main.py?
09:22:53 <zxbiohazardzx> just writing it from scratch is not my thing :P
09:23:10 <andythenorth> Alberth: I'm not sure
09:23:23 <Alberth> I'd use make for that :p
09:23:26 <andythenorth> I want them to be able to be invoked as separate, parallel processes by make
09:23:57 <andythenorth> if we decode filenames, we could, in theory, run one gestalt process per filename
09:24:13 <Alberth> you could make main decode what to do, and then import the right gestalt, and let it do the work
09:24:38 <andythenorth> actually, I'm intrigued by how effective it might be
09:24:43 <zxbiohazardzx> and i liked the idea of being able to place houses manually in the scenario editor
09:24:49 <andythenorth> just the imports (no rendering) appears to take 2s at the moment
09:25:29 <andythenorth> logical error by me :P
09:25:38 <zxbiohazardzx> i think planetmaker even suggested to make it work similar to the object placer, where you can select a house and build it. that for me would be great, but i cant code that well, so i decided that id first update the patch that existed to the 24018 revision (yesterdays head)
09:25:41 <Alberth> ie the first time you have a disk-cache to fill, and possibly have some python compilation going on
09:25:42 <andythenorth> imports are 0.04s
09:26:48 <andythenorth> can imports be dynamic?
09:27:06 <andythenorth> e.g. for item in [list]: import item
09:27:10 <andythenorth> is that even wise?
09:27:30 <Alberth> what if you use "all" as value for stating you want to generate all png of that value? ie tank-trailer-all-1_8-all.png
09:28:09 <andythenorth> I'll see how the code smells when I write it
09:28:25 <Alberth> I wold not do dynamic imports
09:28:29 <andythenorth> BANDIT has to know what 'all' consists of anyway
09:28:36 <andythenorth> to make use of the graphics in spritesets
09:29:15 <Alberth> and it saves you just a few lines of code
09:34:30 <andythenorth> dynamic imports smell :)
09:34:44 <andythenorth> the gestalts change rarely
09:37:49 *** JVassie has joined #openttd
09:53:11 *** mahmoud has joined #openttd
09:53:54 <andythenorth> (1) how to pass things to gestalts: as params in python, or make them use shell args for communication?
09:53:57 *** TGYoshi has joined #openttd
09:54:35 <andythenorth> (2) gestalts need input file (floorplan), should it be passed, or is that an internal detail? It (mostly) relates to length, but might also relate to subtype issue
10:00:46 <Alberth> ad 1: the former should be possible, so you can build the latter if needed :)
10:01:28 <andythenorth> params in python is simpler at this stage
10:01:39 <Alberth> ad 2: what if I use 'all' as subtype? I'd say the gestalt itself should figure it out
10:02:18 <andythenorth> I forgot about cargo variants in my spec
10:02:26 <andythenorth> e.g. STEL can be coils, bars, pipes etc
10:02:40 <andythenorth> not sure yet if that's a gestalt subtype, or a cargo subtype
10:02:44 <andythenorth> think it's cargo
10:06:19 * andythenorth will make code in between chores and babies
10:15:46 *** Zeknurn has joined #openttd
10:17:45 <andythenorth> needs to pass filename obv.
10:18:09 <andythenorth> target_files is for test only, that would be passed by make or such
10:23:42 <Alberth> x.split('_')[0] --> x.split('_', 1), but looking good, imho
10:23:54 *** FLHerne has joined #openttd
10:29:20 <andythenorth> the split('_',1) seems to be failing ;)
10:31:26 <Alberth> oh, it's still a list, you need [0] after it
10:31:36 <andythenorth> my split() only allows two params, docs say three
10:31:56 <Alberth> 'filename' is the first one
10:32:27 <andythenorth> I'm going to rename this as dispatcher.py
10:32:33 <andythenorth> and have main.py act like a fake make
10:33:11 <Alberth> x = C(); x.f(y) becomes x = C(); C.f(x, y) <-- where 'x' is 'self'
10:40:31 *** peteris has joined #openttd
10:42:06 <Alberth> that would be useful indeed
10:42:27 <Ammler> andythenorth: there is a make replacement with python afaik
10:42:54 <Ammler> we once checked that as we were investigating build environments for newgrfs
10:43:26 <Ammler> can't remember the name, something with s
10:44:22 <Ammler> Alberth: but it needs integration to your hook, as we need it for push :-)
10:45:26 <andythenorth> Ammler: I previously found at least one python based make replacement, but the project appeared to be not very supported iirc
10:45:33 <andythenorth> there are probably more...
10:45:51 *** peteris has joined #openttd
10:51:19 *** peteris has joined #openttd
10:54:58 *** FLHerne has joined #openttd
10:58:57 <Alberth> I found the repo with the script, and I have a test setup, but no docs how it worked :p
11:00:48 *** peteris has joined #openttd
11:00:55 <Ammler> pretxnchangegroup.check = python:/home/hg/misc/mercurial/hooks/check_commit.py:check_changegroup
11:03:53 *** peteris has joined #openttd
11:04:57 *** TWerkhoven[l] has joined #openttd
11:09:06 <xiong> I'm not sure what I'm seeing. Has anyone noticed reluctance of industry to accept cargo generated at the same station? I have a train set to 'Unload and take cargo (near end)' then 'Unload and leave empty (far end). The cargo is unloaded but not accepted.
11:11:05 <Ammler> xiong: so it is working?
11:11:24 *** peteris has joined #openttd
11:11:46 <xiong> Ammler, Um, not from my viewpoint.
11:13:00 <andythenorth> Ammler: I did find scons yes
11:14:17 <xiong> I'm not sure the particular use case is relevant; I seem to have similar trouble in another case. There, I definitely want to consume all production without leaving the station. So I had a truck repeatedly loading and unloading. The cargo was not consumed. I eventually created an extra truck stop and sent the truck back and forth, which worked.
11:15:13 <xiong> Um, for even more irrelevant detail, the latter case was a FIRS Coal Mine and Lime Kiln at the same station.
11:15:17 <Alberth> oh, aap, from bram , author of vim :)
11:15:41 *** FLHerne has joined #openttd
11:16:21 <Alberth> afaik you do have to actually transport the cargo, even if it is just one tile
11:16:52 <xiong> The current case is a bit more complex. I've got an excess of FS at several stations so I'm running a big train around to all of them; loading, then conditionally transferring or unloading to consume the excess.
11:17:14 *** peteris has joined #openttd
11:18:26 <xiong> That so, Alberth? I've been successful in so many cases with the small loop trucks ("piglets") rationing out ES/FS and input cargoes. Of course, these come originally from somewhere else.
11:20:03 <xiong> I also use loop trucks to keep output cargo "fresh" by constant transfer-and-load. It seems to me that, back when I started this practice, if I did *not* use a transfer order, my cargo might be consumed at the producing station. Do I just remember that poorly or has the game changed?
11:20:37 <Alberth> oh, I could very well be wrong
11:21:09 <Alberth> I don't setup complicated cargo transfer schemes normally
11:21:11 <xiong> Mm, dunno, Alberth. Your statement is rational, realistic, and consistent with current observations.
11:21:23 *** peteris has joined #openttd
11:21:49 <xiong> You might still be wrong but I'll adopt your view as a working hypothesis.
11:22:16 <Alberth> the only corner case I can think of, is if you first transfer cargo to some station, thn load it, and unload it again there. I have no idea at all what happens then
11:23:17 <xiong> It's not a great burden to add these extra truck stops and alter orders to go back and forth. It's actually a simpler order list -- since to order a vehicle to the same station only, requires a total of 4 orders to avoid the "duplicate order" warning.
11:23:35 <andythenorth> Ammler: scons looks to be the most recently updated
11:24:35 <Alberth> andythenorth: we have too many of these build systems :p
11:25:02 <andythenorth> how many do we have?
11:25:36 <xiong> That's precisely my dominant use case, Alberth. I transfer cargo in bulk -- say, ES -- then pick it up and drop it with a truck, slowly. I actually use a 1 Horse Carriage, which takes nearly a month to station stop and depot. .
11:26:16 *** peteris has joined #openttd
11:26:58 <xiong> I'm going to go along with the "must transport some distance" statement, which forbids consumption at exactly the producing station. It's rational. Thanks Alberth++
11:28:23 *** peteris has joined #openttd
11:28:50 *** frosch123 has joined #openttd
11:30:54 <Alberth> I never got to the point of FS and ES distribution :)
11:41:51 <__mj> hi, I'm debugging something and I've found a train with subtype 1001100(b) but the documentation reads that there are only 6 bits denoting a vehicle subtype, is this information still accurate ?
11:42:28 * andythenorth might get distracted by python multiprocessing module
11:42:46 <__mj> I didn't mean 6 bits in the docu, I meant the enum has only 6 values :) sry
11:43:35 <__mj> still looks strange to have a subtype of 72(d), I need to know if this is a problem in my code that created this train
11:44:01 *** Biolunar has joined #openttd
11:47:26 <Rubidium> are you sure the vehicle is a train?
11:47:35 *** FLHerne has joined #openttd
11:48:05 <Rubidium> though I don't know any reason why the two high bits would be set
11:50:21 *** TWerkhoven[l] has joined #openttd
11:51:23 <__mj> Its a slightly modified version of Train, I use it as intermediate object
11:52:47 <__mj> But I didn't change anything about its type
11:54:31 <Rubidium> are you doing multiple inheritance?
11:57:06 *** Doorslammer has joined #openttd
11:58:10 <__mj> this issue doesn't give any problems, but I should keep a note that it might create problems later on
12:20:09 *** Guest5471 is now known as AD
12:21:29 *** tokai|noir has joined #openttd
12:21:29 *** ChanServ sets mode: +v tokai|noir
12:26:21 <andythenorth> python multiprocessing seems straightforward :o
12:27:41 <__mj> I've been doing a bit of python lately, the only bad thing I can rely say about it, is that it has those ugly indentation semantics ;)
12:30:25 <Alberth> like you don't indent your code otherwise :)
12:33:54 <_maddy> how do I check why my AI is not loaded when I start the game? I made a simple barebones AI according to wiki examples with info.nut and main.nut
12:34:59 <__mj> Alberth: I use chaotic random indentation normally :)
12:35:47 <Yexo> _maddy: open the console, type "debug_level ai=5"<enter> and after that "rescan_ais"<enter>
12:36:09 <valhallasw> __mj: for you, not for anyone else maintaining your code
12:36:25 <valhallasw> the python syntax was not created to be easy to write, but easy to read
12:37:08 <_maddy> Yexo: rescan_ais command not found
12:37:41 <Yexo> try "rescan_ai" or "rescan_scripts"
12:38:17 <_maddy> rescan_ai works, no error messages, I guess for some reason it is not finding it
12:38:27 <Yexo> where did you put info.nut/main.nut?
12:38:30 <Alberth> the 4 letter name is crucial
12:38:31 *** Artanicus has joined #openttd
12:39:50 <Artanicus> howdy. Just installed the lucid 1.1.5 64bit deb and the package seems to be missing the actual binary. Am I doing something wrong or is it just missing? :D
12:39:56 <Yexo> ai debug level has been renamed to "script"
12:40:05 <Yexo> so "debug_level script=5" first, than try "rescan_ai" again
12:40:42 <_maddy> Yexo: still no error message, and no ai loaded
12:41:14 <andythenorth> Alberth: with python multiprocessing, the improvement is ~ linear so far :)
12:41:24 <andythenorth> I know make can handle that part, but I don't have a makefile yet :P
12:41:49 <__mj> valhallasw: did you believe me that I indent my code randomly ? how would anyone get anything done that way ?
12:42:08 <valhallasw> __mj: I've seen enough people *actually* do that
12:42:33 <Yexo> _maddy: can you post your info.nut/main.nut at paste.openttdcoop.org?
12:42:42 <__mj> there is probably a crazy person for everything in the world ;)
12:42:45 <Alberth> andythenorth: nice! :)
12:42:47 *** lofejndif has joined #openttd
12:43:00 <valhallasw> __mj: especially in MATLAB/IDL code... code that was not written by programmers but by random scientists
12:43:24 <Yexo> Artanicus: /usr/games/openttd <- that's the main binary I think
12:43:57 <Artanicus> Yexo, gah, just realized myself. damn I feel stupid
12:44:20 <Hirundo> Hmm... for my test case (ogfxairports), disabling tile compression actually adversely impacts performance
12:44:26 <Artanicus> somehow my eyes thought it was a directory. nvm and thanks for setting me straight :-)
12:44:37 <__mj> valhallasw: and they random-indented their code on purpose ?
12:45:02 <Ammler> openttd installs the bianary in a very silly location per default
12:45:16 <Yexo> Hirundo: lz77 is the slow part, not tile compression
12:45:26 <valhallasw> __mj: it's like the javascript 'written' by web site designers: copy paste stuff from everywhere until it sort of works
12:45:29 <Yexo> and disabling tile compression means more data for lz77
12:47:17 <morph`> Any awesome NoGo scripts for Multiplayer yet?
12:51:18 <Yexo> _maddy: GetShortName is required to return a string of length 4
12:51:31 <Yexo> the debug info should've told you that
12:52:54 <_maddy> yes now the AI loaded, the debug info didn't display any error message though
12:53:25 <andythenorth> for a mixed bag of gestalts, MP pixa is ~5s, non MP is ~11s
12:53:35 <andythenorth> the MP version blocks ~100% of my CPU, as intended
12:53:49 <andythenorth> the non MP version uses ~25%
12:53:54 <__mj> valhallasw: that's the reason why sometimes those people are called 'designers' rather than 'programmers' :)
12:54:02 <andythenorth> must be some overhead somewhere
12:54:38 <__mj> what is a gestalt ? some new feature ?
12:55:02 <andythenorth> Alberth: do you think it's essential to use make for generation?
12:58:24 <_maddy> Yexo: thanks for the help
12:58:31 *** Firartix has joined #openttd
12:58:52 *** FLHerne has joined #openttd
12:58:56 <Alberth> andythenorth: make is used for constructing build sequences, and skipping stuff that does not need a rebuild. If you don't need that, you don't need make imho
12:59:20 <andythenorth> maybe I try to rub along without it
12:59:26 <andythenorth> makefile syntax is a step too far right now
12:59:32 <andythenorth> ducktape coding :P
13:00:18 <Eddi|zuHause> hm... after the "bud spencer" tunnel, and the "i'm lacking a screw" road, now an internet vote (in slovakia this time) resulted in the "chuck norris" bridge
13:01:14 <zxbiohazardzx> thats nonsense, chuck norris doesnt need a bridge
13:03:03 <zxbiohazardzx> being the total naab that i am
13:03:36 <zxbiohazardzx> how do i generate a sln for vc? as ofc the MS VC 2008 guide does not have info on that shit
13:04:36 <Eddi|zuHause> zxbiohazardzx: you execute projects/generate
13:04:54 <zxbiohazardzx> im wintendo on that part
13:05:02 <zxbiohazardzx> so i usually just load a pregenerated SLN
13:05:23 <Eddi|zuHause> zxbiohazardzx: you need only generate the sln if you added files
13:05:31 <zxbiohazardzx> so i recon id have to create a project, import everything and then generate it?
13:05:36 <zxbiohazardzx> i modified files yes
13:05:51 <Eddi|zuHause> and my grammar very german today is
13:06:21 <zxbiohazardzx> hence i want to check if the source compiles without errors, and what results i have ingame
13:06:47 <Eddi|zuHause> zxbiohazardzx: "modifying" files is fine, "adding new files" needs generating the sln
13:07:36 <zxbiohazardzx> lets put it this way Eddi
13:07:42 <zxbiohazardzx> i editted the files in notepad++
13:07:56 <zxbiohazardzx> and now i have to ofc compile & run in order to check how it works out
13:09:14 <zxbiohazardzx> i think i got it, but now its errorspree so i did something wrong for sure
13:10:07 <zxbiohazardzx> it ofc lacks the dependancies, as linking them changed in VS10 compared to older one
13:26:32 <zxbiohazardzx> blegh i keep hitting no sutch file or directory
13:26:40 <zxbiohazardzx> somewhere includes arent going correct
13:31:29 <zxbiohazardzx> D:\Sourcecodes\OpenTTD_SVN\trunk\projects\Debug\ottd_test.exe : fatal error LNK1120: 1 unresolved externals
13:31:54 <zxbiohazardzx> stupid externals, i added them to the MSVC folder in the libs specified for project
13:44:06 *** DanMacK has joined #openttd
13:47:09 <TinoDidriksen> "unresolved external" cannot be solved with #include
13:47:26 <TinoDidriksen> There's some library or file that you're not linking correctly.
13:48:45 <zxbiohazardzx> squirrel.h was also missing
13:48:50 <zxbiohazardzx> and i didnt touch that
13:48:52 <zxbiohazardzx> ill check though
13:49:46 *** TWerkhoven has joined #openttd
13:51:39 <zxbiohazardzx> im not that certain on coding anyway, i just updated the manual house placing patch, and then i failed on setting up the envrionmetnt i recon
14:06:44 *** zxbiohazardzx has left #openttd
14:13:36 <morph`> Is there currently any way of adding company value goals to OpenTTD dedicated server?
14:14:42 *** Chris_Booth has joined #openttd
14:38:48 <_maddy> what type is the return value from AICargoList or AIIndustryList functions?
14:40:36 <Yexo> AICargoList is not a function, it's a class
14:40:39 <Yexo> so the type is "AICargoList"
14:40:51 <Yexo> AICargoList is a subclass of AIList
14:41:49 <_maddy> I have the documentation open, but that class has no functions, so I am assuming I must iterate over it using a loop
14:42:06 <Yexo> as you can see at the top it's a subclass of AIList, which has functions
14:42:20 <andythenorth> Alberth: most of the code in generate() for a gestalt relates to creating all the variations
14:42:25 <Yexo> foreach (item, value in list) { ... } works just fine however
14:42:42 <andythenorth> if, instead, that renders just one spritesheet based on parameters, it might be made common
14:43:21 <andythenorth> then, optionally, passing 'all' as a filename arg, might cause the gestalt to simply construct a list of expanded filenames, then call generate() on them
15:04:52 *** brambles has joined #openttd
15:06:56 <_maddy> how do I get the length of an array in squirrel?
15:13:14 <_maddy> is there a shortcut to get the companyId of the AI's current company?
15:17:28 <Yexo> aicompany.resolvecompanyid(aicompany.company_current) or something like that
15:29:12 <morph`> Weird, UKRS 3.04 doesn't seem to work on my server
15:29:39 <morph`> Works fine when I create single-player game, doesn't appear in list of loaded grfs when I make a multiplayer dedicated server
15:38:35 <andythenorth> it would be nice, for 1.3, to be able to plant farm fields (as industry-owned objects)
15:38:39 *** tokai|mdlx has joined #openttd
15:38:48 <andythenorth> obviously now is not the time for adding newgrf features
15:45:33 <planetmaker> andythenorth: now those things could already be added (again). But... they need writing
15:45:40 *** ChanServ sets mode: +v tokai
15:45:47 <andythenorth> we're all busy :)
15:54:15 *** theholyduck has joined #openttd
16:03:02 <andythenorth> Alberth: I need a new separator for filename parts, underscores aren't cutting it :P
16:03:31 <andythenorth> could use dash for parts, underscore within parts :P
16:05:20 <andythenorth> doesn't that have some exciting OS specific issues? :)
16:05:43 <Rubidium> as if the others don't on Windows
16:06:01 <planetmaker> you could also add a *
16:06:50 <planetmaker> or maybe spaces and \
16:07:16 <planetmaker> breaking FS 101 :-P
16:07:57 <Rubidium> and by that I mean the caret is also bad for FAT
16:08:12 <andythenorth> I don't like it when I run code that I know is broken, but it doesn't fail :P
16:08:14 <Alberth> - and _ are most sane imho
16:08:20 <Rubidium> andythenorth: but where/what do you need to split on?
16:08:37 <andythenorth> example: 'tipping_trailer_4px-fifth_wheel-cc1-7_8-GRAI.png'
16:08:58 <andythenorth> gestalt, colourset, length, cargo
16:09:10 <planetmaker> use camel case naming ;-)
16:09:21 <planetmaker> then you need less separators
16:09:49 <planetmaker> I always have a glass and also use partial separators
16:09:54 <andythenorth> I think python multiprocessing doesn't raise errors in child scripts by default
16:10:26 <Rubidium> andythenorth: but that's 4 things that must always be there, so a single seperator should suffice
16:10:40 <Rubidium> and if you don't have a field just leave it 'empty'
16:10:49 <andythenorth> I need to then further split tipping_trailer_4px etc
16:11:27 <andythenorth> Alberth: I had horrible thought that using 'if 'foo' in filename' fails when an unwanted substring match is found
16:11:45 <andythenorth> e.g foo produces food gestalt instead of foo
16:13:08 <Alberth> that's why I used string.startswith()
16:19:02 <andythenorth> that's interesting
16:20:29 <andythenorth> with MP, this is ~50% faster now
16:26:32 <andythenorth> Alberth: done anything with python multiprocessing ever?
16:26:49 <andythenorth> it would be nice to handle process completion better
16:27:08 <andythenorth> and maybe also optionally limit number of processes
16:27:28 <andythenorth> current code is checked into bandit repo
16:36:49 <Alberth> it beats messing with hg repo code, I am VERY sure
16:41:53 <DorpsGek> andythenorth: 2.16666666667
16:50:30 <Pikka> you really needed the bot to tell you what 100 * 1.3 is, andy? :P
16:52:41 <valhallasw> andythenorth: I've played around with multiprocessing a bit
16:53:06 <valhallasw> most important thing to remember is NOT TO FORGET THE if __name__=="__main__" where you set up the process pool
16:53:13 <valhallasw> or you end up forkbombing yourself
16:53:16 <valhallasw> been there, done that
16:53:18 <andythenorth> Pikka: I like to check the bot is working
16:53:28 *** supermop has joined #openttd
16:53:32 <Rubidium> andythenorth: it's not
16:53:36 <andythenorth> valhallasw: I have ignored that bit of the docs; even the bit where it says "You MUST do this"
16:53:56 <Rubidium> not the expected 0120.6
16:54:19 <valhallasw> andythenorth: I think the docs just say 'this won't work in an interactive interpreter', not 'run this and you have to restart your computer unless you have ulimit configured right'
16:55:45 <valhallasw> I'm always surprised by how quick and effective forkbombs are
16:55:53 <CIA-1> OpenTTD: frosch * r24020 /trunk/src/lang/english.txt: -Fix [FS#4898]: Unify the spacing in 'AI/Game Script' and never just say 'Game' when 'Game Script' is meant.
16:57:09 *** FLHerne has joined #openttd
17:00:12 *** ProfFrink has joined #openttd
17:02:00 *** ProfFrink is now known as Prof_Frink
17:05:55 *** Chris_Booth has joined #openttd
17:30:33 *** welshdragon has joined #openttd
17:35:38 *** HerzogDeXtEr has joined #openttd
18:09:25 <andythenorth> Alberth: handling the 'all' case in filenames would lose the MP speedup, or at least require the gestalt to also spawn processes
18:10:05 <andythenorth> I'm not against it, just not sure if it's worth it right now
18:12:39 <Alberth> oh, you don't expand the 'all' cases in the dispatcher?
18:25:39 <andythenorth> I'd need to figure out a canonical source of data for what 'all' is
18:26:07 <andythenorth> currently that's the gestalt, which is logical but not desirable
18:28:11 <Alberth> add a way to query it from the gestalt?
18:29:25 <Alberth> vars = gestalt.get_variables(); values = gestalt.get_values(vars[2]) with the 3rd var being 'all'
18:31:42 *** urpiotr has joined #openttd
18:33:08 <andythenorth> so that's like saying 'module provides [stuff]'
18:33:41 <_maddy> how do I determine the size (tiles) of an industry in AI code?
18:35:45 <planetmaker> you query each tile for its tile class, I guess
18:35:48 *** Chris_Booth has joined #openttd
18:35:52 <planetmaker> but why do you need to know that?
18:36:56 <_maddy> I need to determine where I can build a station, next to an industry.. I thought it logical AIIndustry class would have something like GetSize, along with GetLocation
18:39:26 <Alberth> andythenorth: you can also use a class instead, but that may be more complicated to use and/or explain
18:39:40 <Alberth> as gestalts are fairly big
18:43:07 <urpiotr> hi, I'm trying to compile with VS2005 but I get this:
18:43:07 <urpiotr> 5>d:\src\openttd\src\network\core\address.cpp(134) : error C2065: 'AI_ADDRCONFIG' : undeclared identifier
18:43:07 <urpiotr> 5>d:\src\openttd\src\network\core\address.cpp(358) : error C2065: 'IPV6_V6ONLY' : undeclared identifier
18:43:07 <urpiotr> Can anyone give me a hint, what could be wrong?
18:43:53 <TinoDidriksen> VS2005? Ancient
18:44:06 <TinoDidriksen> VC++ 2010 Express is free...
18:44:38 <urpiotr> should these be defined in some system, library or SDK headers?
18:45:01 <Alberth> looks like ipv6 support
18:45:17 <Alberth> which may have been non-existing in vs2005
18:45:33 <CIA-1> OpenTTD: translators * r24021 /trunk/src/lang/ (english_US.txt finnish.txt french.txt german.txt):
18:45:33 <CIA-1> OpenTTD: -Update from WebTranslator v3.0:
18:45:33 <CIA-1> OpenTTD: english_US - 13 changes by Rubidium
18:45:33 <CIA-1> OpenTTD: finnish - 9 changes by jpx_
18:45:33 <CIA-1> OpenTTD: french - 9 changes by OliTTD, glx
18:45:35 <CIA-1> OpenTTD: german - 1 changes by planetmaker
18:46:22 <glx> urpiotr: update platform sdk
18:49:27 <Pikka> banks on opposite sides of a 256* map producing around 50 bags of valuables a month
18:49:35 <Pikka> = legitimate use for the Beech 1900 :D
18:51:25 <Pikka> oh look, I never put a sprite image on that page either :)
18:51:42 <Nat_aS> oh I though you said there was a plane that came out in 1900
18:55:55 <guru3> In six days I'll have a 2048 day reporting period for #openttd.
18:56:06 <guru3> Just as a completely random fun fact.
18:57:52 <andythenorth> most of the code was to handle composing all the variations ;)
18:58:01 <andythenorth> now they just dumbly render what they're given
19:04:50 <Alberth> your use of indenting to layout data like code still confuses me completely :)
19:04:50 <CIA-1> OpenTTD: rubidium * r24022 /trunk/src/ (strings.cpp table/control_codes.h table/strgen_tables.h): -Add: CARGO_LIST control code for strings
19:10:22 *** supermop has joined #openttd
19:15:29 <urpiotr> glx: thanks for the hint :) Now it compiles successfully.
19:18:06 <CIA-1> OpenTTD: rubidium * r24023 /trunk/src/ (lang/english.txt station_gui.cpp vehicle_gui.cpp): -Fix [FS#5090]: pass cases down into the list of cargos
19:19:59 <andythenorth> Alberth: you mentioned that the other day, but we didn't discuss further
19:20:08 <andythenorth> my indenting violates PEP-8 I think :o
19:20:10 <CIA-1> OpenTTD: rubidium * r24024 /trunk/src/lang/ (55 files in 2 dirs): -Update: translation w.r.t. the changes in r24023
19:20:43 <andythenorth> can you show me a better way? I don't mind changing it
19:22:15 <andythenorth> Pikka: this bidwell thing should run on batteries :P
19:22:23 <andythenorth> or even stored flywheel power :D
19:22:29 <andythenorth> find a railtype for *that*
19:22:50 <andythenorth> what stopped you?
19:22:53 <Pikka> but there's no "nice" way of doing it that
19:23:10 <andythenorth> wot, you're saying that's no local vars on a vehicle :O
19:23:27 <Pikka> that is, there's no nice way of limiting the time it can spend on normal track
19:23:56 * andythenorth thinks the gameplay result might just be annoying anyway
19:24:21 <Pikka> so, you know, people just have to electrify the routes they want to use it on :P
19:25:27 <Pikka> also, even though sticking 4 of them together is still considerably cheaper to run than a class 66, you won't be replacing all your diesel freights because they're 15mph slower
19:26:43 <andythenorth> you should do one of those green goat things
19:26:49 <andythenorth> variable hp with gensets
19:26:59 <andythenorth> make it adjust to the number of trailing vehicles :P
19:29:05 <Rubidium> 1 engine refittable to everything. Done!
19:35:03 <Pikka> all good all good... just got to do that darn a-train...
19:52:06 <andythenorth> ho 1.5s to generate 42 pngs
19:52:19 <andythenorth> so...who has the fastest box? :P
20:01:16 <andythenorth> seems I upset a Level Crossing
20:03:52 <andythenorth> is the state of...you know...reality calculated single-threaded, or multi-threaded?
20:09:42 <andythenorth> we might need to agree on meta-physics first :P
20:10:24 <andythenorth> stuff like the double slit experiment shows that there is a game-state being calculated for reality
20:10:31 <andythenorth> so is it MP-aware? :P
20:12:45 <frosch123> i think most interesting about rl is whether there is respawning
20:31:14 <Eddi|zuHause> depends on which religion you ask
20:36:30 <andythenorth> also, to what extent could evaluating varaction 2 chains be parallel?
20:37:02 *** lmergen has joined #openttd
20:37:09 <andythenorth> for obvious reason I have overlooked, which is...?
20:37:38 <Eddi|zuHause> andythenorth: the whole underlying design is unfit for parallelization...
20:37:47 <planetmaker> you can query other tiles, towns...
20:38:01 <planetmaker> which can be modified again by other varaction2
20:39:03 <andythenorth> I wondered about the tile case
20:39:06 <andythenorth> what about vehicles?
20:39:19 <andythenorth> there's always some global thing in newgrf :P
20:39:51 <andythenorth> fwiw, I'm not complaining about lack of MP use
20:42:25 <Eddi|zuHause> andythenorth: alone the fact that you can query the "related object" will screw your plans...
20:43:37 <andythenorth> is it theoretically impossible to parallelise vehicle varaction 2 chains?
20:44:39 <Eddi|zuHause> i don't think it can be done without sacrificing some (corner-case?) functionality
20:45:28 <andythenorth> for vehicles, it's probably not critical functionality
20:45:47 <andythenorth> a consist seems to be a pretty self-contained entity to me
20:46:52 <Eddi|zuHause> but it's exactly _inside_ the consists that you will cause trouble
20:47:04 <planetmaker> except for rails or roads
20:47:40 <andythenorth> Eddi|zuHause: I'm assuming you can't parellelise the vehicles in the consist
20:47:53 <andythenorth> you have to choose somewhere to start evaluating, presumably the start :P
20:48:00 * andythenorth is guessing, but could read the code
20:48:13 * andythenorth is actually shirking re-writing cargo sprite generation
20:53:38 <planetmaker> andythenorth, vehicle varaction2 also needs to be sequential or you have to toss all random action2
20:54:08 * andythenorth stops shirking cargo sprite generation rewriting thereof
20:54:10 <planetmaker> or at least the re-randomisation of the vehicle bits needs to be sequential
20:54:58 * andythenorth suggests an optimisation equivalent to MP: reduce vehicle limit, decrease map size :)
20:55:02 <planetmaker> at least vehicle behaviour should not depend on the non-sequential random actions :-). But in principle it can currently
20:55:14 <planetmaker> Like I could return vehicle_speed = 38 * random(10)
20:55:21 <planetmaker> vehicle_max_speed
20:55:46 <Eddi|zuHause> a propos... anyone tested my randomized livery change delay?
20:56:40 <andythenorth> I tested your crossings patch
20:56:45 <andythenorth> what happened to that?
20:56:50 <andythenorth> also your fence patch
20:57:09 <Eddi|zuHause> nothing happened... still waiting on somebody to code the drawing part
20:57:29 <Nat_aS> can anybody give me advice on the windows?
20:57:41 <andythenorth> Eddi|zuHause: the drawing part of crossings?
20:57:51 * andythenorth coded crossings patches once
20:57:52 <Nat_aS> Diagonals are what the player sees most of the time, but they kinda look ugly
20:58:14 <Eddi|zuHause> andythenorth: drawing diagonal tracks on road crossings
20:58:25 <Nat_aS> I can't really get rounded edges in diagonals looking good, there is no easy rule like the 1:2 for straight lines
20:58:46 <andythenorth> first I'd have to draw you that stupid composited road-rail piece so that you could force it to appear under tram tracks
20:59:02 <Eddi|zuHause> Nat_aS: my solution to diagonals looking ugly is making them longer :p
20:59:28 <Nat_aS> isn't that the max lenght for diagonals though?
20:59:29 <Eddi|zuHause> andythenorth: what do you mean?
20:59:40 <Eddi|zuHause> Nat_aS: yeah... i cheat :p
21:00:05 <andythenorth> Nat_aS: your diagonals don't look so bad, you use an odd window colour though
21:00:20 <andythenorth> Eddi|zuHause: tram tracks insist on putting a road piece under rail crossings
21:00:31 <andythenorth> I spent a weekend trying to stop that, but it's pretty tricky
21:00:38 <Eddi|zuHause> andythenorth: yes. draw road, draw rail, draw tram
21:00:49 <Nat_aS> what would be a better color
21:00:56 <Nat_aS> I was not really sure what to use.
21:00:57 <Eddi|zuHause> andythenorth: but that has nothing to do with diagonal crossings :)
21:01:07 <andythenorth> Nat_aS the dark purple used everywhere else ;)
21:01:17 <Nat_aS> which is it on the pallet?
21:01:21 <andythenorth> Alberth: good night
21:01:31 <andythenorth> Nat_aS: start the game, enable newgrf debug tools, and look at some of the default vehicles
21:02:23 <Eddi|zuHause> Nat_aS: it's actually 3 vehicles that just look like one
21:02:47 <Nat_aS> >Trains bending with the track
21:03:43 <andythenorth> *not* 170-177 which some people have used :o
21:04:00 <Eddi|zuHause> Nat_aS: a normal vehicle is 8lu (length-units). to create the illusion of a 10lu vehicle i make three vehicles of length 3-4-3, and make the two 3's invisible, only draw the 4
21:04:45 <Eddi|zuHause> the bending is some additional magic to avoid the vehicle sticking out of the rails too far
21:05:42 <Nat_aS> it's nice to see this game has consistant style
21:08:55 <Nat_aS> so whole tile engines are possible?
21:11:59 <Nat_aS> so these are hand drawn? or traced 3d renders?
21:12:10 <Eddi|zuHause> this is hand drawn
21:12:19 <Nat_aS> anyways, you guys are blowing my mind with newgrif fu
21:12:31 <Nat_aS> I take back what I said about simutrans having better coding
21:12:33 <Eddi|zuHause> i know, i'm kind of an extreme case :p
21:12:41 <Nat_aS> Newgrifs blow them out of the water
21:12:53 <Eddi|zuHause> and i didn't even talk about autogeneration yet :p
21:12:59 <Nat_aS> you are so limited by the engine in simutrans when it comes to making fancy new trains.
21:13:17 <Nat_aS> like they don't even support randomized cargo sprites for doublestack cars
21:13:21 <Eddi|zuHause> most of the magic needed for this to work is only in trunk for 2 months
21:13:45 <Nat_aS> both of which I wish I had when I made doublestack cars for them
21:13:53 <Nat_aS> I think they might still be in there latest build
21:14:20 <Nat_aS> there is no half loaded sprite, and the containers are CC only
21:14:37 <Nat_aS> no multicolor shiping containers :c
21:15:14 <Eddi|zuHause> i'm massively bulding upon other people's foundations here
21:15:33 <andythenorth> gah...I hadn't considered the CETS case when setting up cargo sprite generation :|
21:15:53 <Eddi|zuHause> many of the ideas were there previously, but i pieced them together
21:16:31 *** Scuddles has joined #openttd
21:16:31 <Nat_aS> man, I'm not sure I could sprite cars with that many rotations
21:16:35 <andythenorth> Eddi|zuHause: I'm planning to release a full set of cargo load sprites, GPL, on nml templates, length 2/8 to 8/8
21:16:44 <andythenorth> but only for 8 angles :P
21:16:59 <Nat_aS> yeah, a lot of those angles are hard to draw
21:17:22 <Nat_aS> it's the reason the game is so ugly
21:17:25 <andythenorth> Eddi|zuHause however the sprite generator doesn't know too much about angles, it just uses floorplans
21:17:29 *** supermop has joined #openttd
21:17:47 <Eddi|zuHause> andythenorth: as long as you can use the template as a floor plan...
21:18:05 <andythenorth> but something like a tarpaulin load - that would need drawing for extra angles
21:18:11 <andythenorth> same for machinery, vehicles, etc
21:18:18 <andythenorth> steel coils - I've got covered :P
21:18:28 <andythenorth> make all loads round, you're fine :P
21:18:47 <Eddi|zuHause> well, we can think about that when we get there...
21:18:59 *** Firartix has joined #openttd
21:19:06 <Nat_aS> will there be a geniric shiping container sprite that can be copy pasted for consistancy?
21:19:13 <Eddi|zuHause> i'm more interested in getting some generic shapes that not all vehicles are green blobs, but something that vaguely resembles an engine/wagon
21:19:16 <Nat_aS> and be the same on boats, stations, trucks, and trains?
21:19:50 <andythenorth> Eddi|zuHause: the generator is close to usable....I'll be looking for users :)
21:20:02 <Nat_aS> maybe the game could even track which sprite of container is droped off where so it would look like they are actualy being moved around
21:20:32 <Eddi|zuHause> Nat_aS: that's a little too much... unless you tie the container colour to a cargo
21:21:02 <andythenorth> Nat_aS: there'll be a full set of shipping container sprites, 20' and 40'
21:21:08 <Eddi|zuHause> Nat_aS: or you implement containers as a vehicle, and loading/unloading containers as a special case of shunting
21:22:05 <Nat_aS> well the idea would be to have stations remember which color container they receive, and "store" them in there own sprite, then when they give cargo to a veichile, they will set the loaded veichile to have that sprite
21:22:13 <Nat_aS> it would be an illusion
21:22:30 <Nat_aS> just remember the color it gave you, keep that color, then give that color to the next train
21:22:44 <Nat_aS> probably a lot of work for such a subtle effect
21:23:08 <Nat_aS> man I love shiping containers, will this newgrif have new stations?
21:23:13 <Nat_aS> like a container port station?
21:23:28 <andythenorth> Nat_aS: you've found ISR, right?
21:23:31 <Nat_aS> I like the industrial stations renewal set, but it could be better.
21:23:43 <Nat_aS> it's the best station newgrif
21:23:49 <Eddi|zuHause> it's a little dated meanwhile, could use an update
21:23:52 <Nat_aS> but it lacks stations for some industries,
21:24:07 <andythenorth> there's also DWE stations
21:24:07 <Nat_aS> and also there is no equivlent for PAX stations
21:24:13 <Eddi|zuHause> not as bad as MBs passenger stations, though
21:24:28 <Nat_aS> also the "Classification Yard" set in it seems halfassed
21:25:25 <Nat_aS> that allows placing seaports like stations
21:25:45 <Nat_aS> built over water, as long as one side is touching land
21:25:58 <Nat_aS> and customizible the way industrial stations are
21:26:27 <Nat_aS> and that link is really cool
21:28:03 <Eddi|zuHause> just had to laugh at the lego-smilies... don't think i've seen those before :p
21:53:41 <xiong> Yes, I play with DWE as well. I have quite a few station sets.
21:55:05 <xiong> Just because it's a train station tile doesn't mean you can't build other stations with it -- seaports, bus and truck stops. Just join the tile onto the basic dock, stop, or even airport.
21:55:46 <planetmaker> Eddi|zuHause, what would you add / change with ISR?
21:56:04 <planetmaker> I'd change the forced track type for some
21:56:08 <frosch123> change? the grfid :p
21:56:42 <Eddi|zuHause> don't remember, haven't played in a while so i don't know what annoyed me the last time
21:57:37 <Eddi|zuHause> the most important was that the stockpiles should depend on some average throughput, not the currently waiting cargo (needs some statistics exposed as variables)
21:58:29 <Eddi|zuHause> also: not all parts of the "full (drag&drop) stations" are available as individual pieces
21:58:51 <planetmaker> that I consider not that tragic. Actually I'd rather add some more drag&drop stations
21:59:28 <Eddi|zuHause> i'm not usually using the full stations. mostly i'm making empty stations and put some sprinkles in
22:00:15 <Eddi|zuHause> and the shunting yard needs a matching eyecandy railtype with the same ground tile
22:00:48 <Eddi|zuHause> and more shunting tiles
22:01:04 <Eddi|zuHause> switches in all kinds of directions
22:01:48 <Eddi|zuHause> and eye-candy wagons on the shunting rails
22:03:04 <planetmaker> yeah, the latter makes sense. Though that IMHO is difficult. Not sure I'd do that, if I were the author of a station set
22:03:19 <planetmaker> it's only fit my personal choice of NewGRFs which is... not very versatile necessarily
22:03:33 <Eddi|zuHause> indeed, that's a problem
22:04:11 <planetmaker> thus I think the better solution is invisible engines to place wagons on real tracks
22:04:47 *** supermop has joined #openttd
22:04:50 <Eddi|zuHause> could use some evil AI callbacks like "give me a wagon that carries cargo <Y>, and put that as effect vehicle on location <X>"
22:04:51 <planetmaker> (or maybe not invisible, but looking like a wagon)
22:05:14 <planetmaker> lol, Eddi|zuHause :-)
22:05:35 <planetmaker> or rather a few NewGRFs, iirc
22:06:13 <Eddi|zuHause> planetmaker: well. that's the most extreme way i can immediately think of that will be independent of the vehicle and industry set used
22:07:12 <Eddi|zuHause> planetmaker: the "wagons" could then effectively work like the stockpiles, i.e. get more or less depending on throughput
22:08:18 <planetmaker> :-) It's an interesting concept. Would maybe need a new CB for stations... but might be worth it when someone [TM] cares to implement both
22:10:46 <Eddi|zuHause> the whole shunting yard concept should maybe be split off ISR
22:27:03 <Nat_aS> wait, OTTD works on android?
22:30:20 <Eddi|zuHause> there may be unofficial ports
22:30:40 <Eddi|zuHause> the UI usually needs tweaking
22:31:28 <Eddi|zuHause> like a way to simulate shift and ctrl
22:33:04 <FLHerne> Is there any way to simulate right-clicking?
22:33:34 <FLHerne> Such as [modifier]-click or something
22:33:42 <Nat_aS> apparently that port can't handle click dranging
22:33:48 <Nat_aS> which makes tracks hard to do
22:34:10 <FLHerne> That would be annoying
22:34:35 <FLHerne> Even more so than not being able to move the map
22:35:32 <Nat_aS> almost as anoying as SImutrans autorail feature
22:35:44 <Nat_aS> less because it wont do things you don't want it to do
22:36:19 <FLHerne> What does that do? Never played Simutrans...
22:36:51 <Nat_aS> in simutrans you can lay tracks by clicking in two points, and letting the pathfinding engine pick the best route between them
22:36:59 <Eddi|zuHause> i'd be more worried about assembling trains
22:37:02 <Nat_aS> this sounds like a good idea unless you want to lay parallel tracks
22:37:26 <Nat_aS> because it tries to use existing tracks whenever posible
22:37:32 <Nat_aS> and will connect instead of following
22:38:55 <Eddi|zuHause> i remember how annoying it was in TTT to get a second track next to an existing one
22:40:04 *** lofejndif has joined #openttd
22:45:15 <planetmaker> <FLHerne> Is there any way to simulate right-clicking? <-- there is. At least on OSX
22:55:46 *** supermop has joined #openttd
23:07:40 <DorpsGek> andythenorth: 0.333333333333
23:07:50 <andythenorth> that's quite a bit faster :O
23:10:33 <andythenorth> good night Terkhen
23:24:37 <Nat_aS> Fffffffffffff I hate shitty Bioses
23:24:49 <Nat_aS> can't seem to boot off of anything
23:25:37 <Nat_aS> how am I going to install ubuntu onto this damn thing if I can't get it to boot off of USB, and installing off of a CD crashes halfway through
23:29:52 <TinoDidriksen> Nat_aS, just use VirtualBox?
23:32:20 <TinoDidriksen> http://virtualbox.org/ - it's a virtual machine, but if you have a CPU with virtualization extensions then it's almost as fast as native.
23:33:42 <Nat_aS> well a friend of mine picked up a used computer at goodwill, and I want to reformat it to run ubuntu because it's previous owners left it full of spyware
23:34:10 <Nat_aS> can't get it to boot of of USB, and booting off of CD is really slow and has lots of disk read errors
23:34:35 <TinoDidriksen> Oh...well then, you can put the HDD in another machine and install, then move it back.
23:35:02 <Nat_aS> if only I had another machine, all my computers are laptops
23:35:08 <Nat_aS> but that is a good idea.
23:35:40 <planetmaker> you can also attach a HDD to a laptop, you know...
23:36:19 <Nat_aS> actualy my vaio does have an Esata port
23:36:56 <Nat_aS> so I plug that in, then plug my USB key into another port, then boot off of that, and install Ubuntu into the pluged in drive?
23:37:22 <Nat_aS> careful not to accidently select my laptop's own HD and reformat the Windows install there
23:37:30 <TinoDidriksen> I would remove the laptop drive while doing that, just to be sure it doesn't install Grub there.
23:37:36 *** Rhamphoryncus has joined #openttd
23:37:42 <glx> the other way is network install using a boot cd
23:38:24 <TinoDidriksen> Does Ubuntu have a netinst edition any longer?
23:39:50 <TinoDidriksen> Fewer read errors that way.
23:40:00 <Nat_aS> I'll need to get another disk though
23:40:14 <Nat_aS> Is there a floppy install option :P
23:40:20 <Nat_aS> this computer has a floppy drive
23:40:53 <glx> RW discs exist for a reason :)
23:41:15 <Nat_aS> aren't RW's less reliable though?
23:41:43 <TinoDidriksen> RW is fine these days.
23:42:21 <Nat_aS> i hate optical media SO MUCH
23:42:28 <Nat_aS> I want it to die faster
23:42:50 <planetmaker> you have a window?
23:42:57 <planetmaker> throw it out. It then dies quite quickly
23:44:16 <Nat_aS> I want it to be all digital distribution
23:44:30 <Nat_aS> I want to be able to install oSes from the internet as well.
23:44:32 <Eddi|zuHause> we used to use DVD+RW, but the disks got quite unreliable after several dozen overwrites
23:45:11 <Eddi|zuHause> and suddenly the other guy's burned disks were not readable in my computer anymore, but when i burned over them they worked fine
23:45:22 <Eddi|zuHause> which made it kinda awkward
23:45:32 <Eddi|zuHause> then we invested in a USB stick :p
23:46:05 <Eddi|zuHause> have never seen a BIOS that couldn't boot off USB though
23:46:14 <TinoDidriksen> You can get 500gb USB3 harddrives that only need USB power for €100 these days...wonderful times.
23:46:18 <Eddi|zuHause> at least if it's newer than 10 years
23:46:56 <Eddi|zuHause> 500GB used to be 50€ a year ago
23:47:20 <TinoDidriksen> Only for the versions that need external power.
23:47:37 <TinoDidriksen> These are thin things that fit in a pocket.
23:48:21 <Eddi|zuHause> and if he "got a used computer" it might not have USB3
23:48:51 <TinoDidriksen> Nono, I was more talking instead of USB sticks and DVD-RW in general.
23:49:38 <Eddi|zuHause> i think i paid something like 40€ back then for my sloooooow 8GB stick
23:49:45 <Eddi|zuHause> it's really small though
23:50:09 <Eddi|zuHause> have it on my key ring
23:50:45 *** lofejndif has joined #openttd
continue to next day ⏵
