Log for #openttd on 8th August 2012:
Times are UTC Toggle Colours
00:09:03  *** nicfer [~Administr@] has joined #openttd
00:09:07  <nicfer> hello
00:10:16  *** roadt [~roadt@] has joined #openttd
00:11:45  *** kkimlabs [~kkimlabs@NYUFGA-WLESSAUTHCLIENTS-01.NATPOOL.NYU.EDU] has quit [Remote host closed the connection]
00:12:27  <nicfer> I'm trying to setup mingw for compiling openttd but when ./configure-ing, I get a pack of 'HOSTNAME.EXE - entry point not found' errors
00:13:32  <nicfer> below also says 'Unable to find entry point s_perror in the dynamic link library MSWSOCK.dll'
00:23:24  *** KritiK [~Maxim@] has quit [Quit: Leaving]
00:29:34  *** drac_boy [] has joined #openttd
00:29:44  <drac_boy> hi
00:32:01  <glx> nicfer: followed the tutorial ?
00:34:38  <drac_boy> someone trying to create a small grf?
00:43:33  <drac_boy> guess so
01:15:21  *** Devroush [] has quit []
01:24:58  *** Biolunar [] has quit [Quit: All your IRC are belong to us]
01:44:23  <nicfer> glx, yes I'm following
01:49:13  <glx> the errors are strange
01:54:11  <nicfer> the configuring continues however
01:58:52  *** ivan` [~ivan`] has quit [Quit: ERC Version 5.3 (IRC client for Emacs)]
02:02:41  *** ivan` [~ivan`] has joined #openttd
02:02:49  *** Frankr [] has quit [Quit: ChatZilla [Firefox 14.0.1/20120713134347]]
02:10:56  *** Mazur [] has quit [Remote host closed the connection]
02:14:39  *** LordPixaII [] has joined #openttd
02:14:39  *** Pixa [] has quit [Read error: Connection reset by peer]
02:16:16  *** neofutur [] has joined #openttd
02:17:38  <neofutur> hi all, since luukland will stop operating i ve setup my open citybuilder server on a dedicated server using the city builder patch
02:17:57  <neofutur> the server is working, but i also need a patched client to use it ?
02:18:27  <neofutur> i understand the server have to be patched, but ewhy a default 1.2.1 client couldnt connect and use it ?
02:19:47  <neofutur> ( I used )
02:26:22  <neofutur> no way to use a city builder server with the default binary ?
02:26:31  <neofutur> only people able to patch and compile can use it ?
02:50:57  *** pugi_ [] has joined #openttd
02:56:54  *** pugi [] has quit [Ping timeout: 480 seconds]
02:56:56  *** pugi_ is now known as pugi
03:00:44  *** drac_boy [] has left #openttd [I'm done being in this room!]
03:09:53  *** Mazur [] has joined #openttd
03:11:53  *** pugi [] has quit []
03:12:03  <neofutur> I could build the server on gento but on ubuntu i get a linking error building the client  :
03:12:06  <neofutur>
03:12:49  <neofutur> ( on ubuntu LTS )
03:22:16  *** roadt [~roadt@] has quit [Ping timeout: 480 seconds]
03:26:55  *** glx [glx@2a01:e35:2f59:c7c0:55a0:d5c9:bf87:d4d0] has quit [Quit: bye]
03:28:45  *** KouDy [] has joined #openttd
03:29:21  *** valhallasw [] has quit [Read error: Connection reset by peer]
03:30:45  *** roadt [~roadt@] has joined #openttd
03:52:34  <neofutur> ok i could fix it, the makefile was not lookingfor the good libicu* in /usr/local , using the old ones in /usr/lib
04:34:01  * neofutur playing his first city builder game on his own server
04:45:26  *** Chrill [] has quit []
04:52:12  *** HerzogDeXtEr1 [] has joined #openttd
04:57:09  *** DOUK [] has joined #openttd
04:57:50  *** HerzogDeXtEr [] has quit [Ping timeout: 480 seconds]
05:03:57  *** mahmoud [] has quit [Ping timeout: 480 seconds]
05:32:11  *** Eddi|zuHause [] has quit []
05:32:27  *** Eddi|zuHause [] has joined #openttd
05:42:23  <Rubidium> neofutur: the game state in OpenTTD is too big to be send every tick, or to be updated every few ticks. Imagine downloading a savegame every tick. So OpenTTD makes the server and all clients perform the same tasks by only sending the user initiated commands. If a patch changes anything that modifies the logical program flow of the game state, then both sides need to be the same
05:53:18  *** Prof_Frink [~proffrink@] has quit [Ping timeout: 480 seconds]
06:07:37  *** KouDy1 [] has joined #openttd
06:13:18  *** KouDy [] has quit [Ping timeout: 480 seconds]
06:32:33  *** sla_ro|master [slaco@] has joined #openttd
06:35:13  *** ludde [] has joined #openttd
06:42:59  *** Supercheese [~Password4@] has joined #openttd
06:46:20  *** Rhamphoryncus [] has joined #openttd
06:47:52  *** nicfer [~Administr@] has left #openttd []
07:04:57  <planetmaker> moin
07:05:02  <Supercheese> Salve
07:05:30  <planetmaker> neofutur: if you want citybuilder with default client, then I suggest to implement that via game script
07:05:59  <planetmaker> It's feasible and there are even some game scripts to a similar end around which could possibly be something to get started from
07:08:54  <planetmaker> neofutur: there's already 3 dealing with town growth to some extent
07:12:15  *** Warod [warod@2001:1bc8:1004::1] has quit [Ping timeout: 480 seconds]
07:12:52  *** Defaultti [] has quit [Ping timeout: 480 seconds]
07:15:57  *** ludde [] has quit [Ping timeout: 480 seconds]
07:21:17  *** ludde [] has joined #openttd
07:22:07  *** Pixa [] has joined #openttd
07:26:24  *** KingPixaIII [~pixa@] has joined #openttd
07:27:13  *** LordPixaII [] has quit [Ping timeout: 480 seconds]
07:31:08  *** Pixa [] has quit [Ping timeout: 480 seconds]
07:43:48  *** telanus [] has joined #openttd
07:57:20  <NGC3982> neofutur: city builder game?
07:57:49  * NGC3982 googles.
07:57:55  <NGC3982> C*, G*.
08:25:37  *** Defaultti [] has joined #openttd
08:26:11  *** Warod [] has joined #openttd
08:30:00  *** cyph3r [] has joined #openttd
08:35:50  *** roadt [~roadt@] has quit [Ping timeout: 480 seconds]
08:38:28  *** valhallasw [] has joined #openttd
08:42:00  *** Supercheese [~Password4@] has quit [Quit: ChatZilla [Firefox 14.0.1/20120713134347]]
09:21:42  *** FLHerne [] has joined #openttd
09:25:34  *** pugi [] has joined #openttd
09:49:16  *** TGYoshi [~TGYoshi@] has joined #openttd
10:38:15  *** roadt [~roadt@] has joined #openttd
10:42:29  *** Elukka [] has joined #openttd
11:00:45  *** drac_boy [] has joined #openttd
11:00:59  <drac_boy> hi
11:05:31  *** tycoondemon [] has quit [Ping timeout: 480 seconds]
11:07:44  <FLHerne> drac_boy: Hi :-)
11:08:09  <drac_boy> how're you FLHerne? any new tiles yet? :p
11:08:13  <drac_boy> heh heh
11:23:40  *** perk11 [~perk11@] has joined #openttd
11:26:08  <drac_boy> hi perk11
11:26:19  <perk11> drac_boy: hi
11:27:37  *** Timberdragon [] has joined #openttd
11:28:53  <drac_boy> how're you?
11:33:54  <Timberdragon> Hi guys, I'm keen to work on a 32bit base set for OTTD however ive spent the last week reading the forums and wiki and i seem to be chasing my tail through a mixture of outdated and inccorect information... is there a single absolute source for information regarding building base sets and graphic formats for the current 1.2.* version of the game?
11:34:40  <planetmaker> Timberdragon: the single important information is: 32bpp is not treated different from 8bpp
11:35:03  <planetmaker> you need 8bpp and you can supply in the same package also 32bpp for the sprites
11:35:11  <planetmaker> by the same means virtually
11:35:57  <Timberdragon> yes, i intend to just down sample from the full 32bit versions, but i find all the info confusing
11:37:00  <Timberdragon> im not a fly by nighter i contributed to the opensource project BOSwars massively and im really keen to make a nice set for OTTD
11:37:15  <planetmaker> if you want to create anything 32bpp: just make a normal (8bpp) newgrf. And supply the 32bpp sprites within that newgrf as alternatives to the 8bpp sprites
11:37:49  <planetmaker> I'd like to refer you to ogfx+trains as newgrf example
11:38:00  <Timberdragon> Is the primary tool still the grfcodec?
11:38:06  <planetmaker> you can use that.
11:38:15  <planetmaker> My recommendation is to write in NML, though instead of NFO
11:38:37  <Timberdragon> do you know of any GUI's for either of those by anychance?
11:38:45  <planetmaker> ogfx-trains and zbase are both written in NML
11:39:03  <planetmaker> newgrfs are programmed. It's not just imaged attached to existing objects
11:39:21  <planetmaker>
11:40:51  <V453000> NML is very easy programming :) even I managed to learn that :p
11:40:58  <V453000> somewhat anyway
11:41:09  <drac_boy> :p
11:41:41  <drac_boy> I'm sticking to nfo for a reason but then thats to our own ways tho :)
11:42:00  *** tycoondemon [] has joined #openttd
11:42:29  <Timberdragon> :) im not a programmer either but i learned to hackishly use LUA for the other project, can i start with the standard base set and edit the grphics/details without building the files from scratch?
11:43:03  <planetmaker> you can start with the standard base set and just add 32bpp sprites to it. Yes
11:43:22  <Ammler> doesn't zbase exactly this?
11:44:05  <planetmaker> Which basically is what zbase, too:
11:47:59  <Timberdragon> Is zbase the project to combine all the exisiting 32bit works into 1 file?
11:55:04  <FLHerne> Timberdragon: Not really, it's a project to make a complete 32bpp baseset, ignoring all existing graphics :P
11:56:13  <FLHerne> Apparently licensing issues keep getting in the way when trying to sort out the existing ones...
11:56:28  <Timberdragon> yes i understand
11:58:33  <Timberdragon> i recall now, zbase is controlled pretty much by a single person. i recall reading the forum post, it was the reason i thought of doing a seperate base set
11:58:58  <Timberdragon> i wasnt blown away by what i seen there
11:59:13  <Timberdragon> unlike some of the original 32bit graphics
11:59:33  <FLHerne> At least it'll get done :P
11:59:48  <Timberdragon> perhaps :)
12:00:15  <FLHerne> Zephyris did a lot of the current 8bpp graphics, IIRC
12:00:39  <Timberdragon> for the default opengfx set?
12:01:16  <Timberdragon> if so that kind of makes sence as the 32bit still look a bit 8bittish to me that is all
12:02:01  <FLHerne> Indeed. I think he's trying to do it in the same style
12:02:10  <V453000> that is just because he doesnt use many colours in there. Uses 32bpp colours which arent in 8bpp but the variety is not there
12:02:12  <planetmaker> Timberdragon: Zephyris provided 70% of the 8bpp graphics. He'll get done the 32bpp, too
12:02:24  <V453000> I believe he will change some things in the future :)
12:02:43  <planetmaker> goal for now is: get it done, working. Beautification is for then
12:02:57  <V453000> ^ :)
12:03:43  <Timberdragon> im not having a go at the guy, i think the 8bit set is fantastic and true to the original game. I also understand what he says about "getting it done" and sorting out the details later
12:04:08  <planetmaker> the problem with the existing 32bpp graphics is: *someone* has to sort through. All I always got was "it's basically done", but... nothing I could work with other than I'd have to look at everything myself, too
12:04:31  <planetmaker> as such nothing was really done there
12:04:47  <planetmaker> and each and everything needs checking for the licenses and available models, too
12:05:29  <planetmaker> anyway, that's all beside the point :-)
12:05:36  <FLHerne> Presumably any other 32bpp baseset could use exactly the same NML as zbase, given they'd have the same sprites etc?
12:05:58  <planetmaker> yes, of course. You'll have to adjust all your offsets though, of course, too
12:06:11  <Timberdragon> i understand totally. I'm very good at getting things done however my filing systems suck hard and im not very good at saving sources, hence the reason i would just prefer to work on a stand alone set X)
12:06:23  <planetmaker> except if you do it like Zephyris who alignes the sprites mostly such that offsets are all 0
12:07:21  <planetmaker> well, please do, Timberdragon :-)
12:07:46  <planetmaker> if you like, I'd also be happy to provide you the same technical facilities as to Zephyris' base set
12:07:49  <V453000> you can always provide those sprites/models to anything else in the end
12:08:41  <Timberdragon> plantmaker: you may live to regret the technical assistance offer X)
12:08:54  <Timberdragon> ..and my spelling sucks X)
12:09:42  <Timberdragon> please if you guys have a minute check out boswars dot org for some of my other work
12:10:01  <planetmaker> Timberdragon: technical assistence doesn't include the programming ;-) I'm busy enough, I'm afraid, with 1.5 base sets at my hand
12:10:32  <Timberdragon> yeah i understand
12:10:43  <planetmaker> but it includes the setup of a repository so that the DevZone could - if you use it - build your your base sets on a somewhat daily basis, too
12:11:29  <Timberdragon> ill see if i can get going before i trouble you for that service
12:11:53  <planetmaker> easiest approach probably is to
12:11:57  <planetmaker> - clone opengfx
12:12:04  <planetmaker> - start adding your new sprites to it
12:12:09  <Timberdragon> that is what i am thinking
12:12:36  <planetmaker> the alternative is to make NewGRFs
12:13:00  <planetmaker> which intially might give "better" (in the sense of complete and easier usable) results
12:13:50  <planetmaker> like a newgrf only replacing landscape (tiles, foundations, water...). another all infrastructure (road, rail, bridges, stations), a third rail vehicles, fourth road vehicles, ...
12:13:57  <planetmaker> would not be really more complicated
12:14:26  <planetmaker> and give easier the rewarding "success!" feeling when one part is finished. And when users then will jump to use it - with what ever other graphics they use
12:14:43  <planetmaker> and newgrfs are easier added to a game than a base set switched. I believe
12:16:00  <V453000> the problem is he would have to code that newGRF I guess
12:16:13  <Timberdragon> i understand the principal, though i really hate programming and txt editing, id much prefer to work my way through a base set and replace the sprites
12:16:42  <planetmaker> yes... but programming a simple replacement newgrf is not more difficult than a base grf really. All you need is the grf header
12:17:00  <planetmaker> other than that it's mostly replace(xxx, ...) instead of base_sprite(xxx, ...)
12:17:16  <planetmaker> just telling :-)
12:17:31  <V453000> oh, that way :) well that could be simple
12:17:57  <V453000> how many sprites is opengfx?
12:18:02  <planetmaker> 10k approx
12:18:07  <V453000> yeah, thats a lot
12:18:26  <V453000> making smaller pieces of separate newGRF is a bit more motivating :)
12:18:31  <Timberdragon> what i really need is to have a look insinde a newgrf file, how can i bust one open?
12:18:36  *** roadt [~roadt@] has quit [Ping timeout: 480 seconds]
12:18:53  <drac_boy> grfcodec decoding Timberdragon
12:18:54  <planetmaker> you don#t want to look into a NewGRF file. You want to look into the source code of one
12:19:11  <planetmaker> decoding by grfcodec for that purpose, honestly, is a very bad idea
12:19:21  <NGC3982> I tried that.
12:19:23  <NGC3982> Didn't work.
12:19:24  <Timberdragon> to study file and folder structure?
12:19:27  <planetmaker> it's as unreadable and un-understandable as you can get, close to binary / hex code
12:19:29  <drac_boy> why? it gives you the exact nfo file used :)
12:19:39  <planetmaker> which is... stupid. Really stupid
12:19:41  <V453000> hoho, havent even tried that luckily :)
12:19:56  <NGC3982> To learn NewGRF coding demands looking into NML files, not GRF files, right?
12:20:01  <NGC3982> At least that's how i feel.
12:20:03  <planetmaker> no comments. no structure, no flow, no context
12:20:05  <drac_boy> NGC3982 its not grf :)
12:20:09  <V453000> nml and sprites
12:20:15  <planetmaker> NGC3982: yes, that#s what I'd suggest
12:20:24  <NGC3982> drac_boy: Oh? I kind of fell into the conversation by context here. :p
12:20:36  <drac_boy> planetmaker did you somehow miss that nfo does have comments in a lot of files? :)
12:20:44  <drac_boy> NGC3982 heh ok
12:20:44  <NGC3982> planetmaker: I see.
12:21:03  <planetmaker> drac_boy: action 0C is by far not sufficient to count as "properly commented"
12:21:16  <drac_boy> planetmaker its not action
12:21:28  <planetmaker> you're advising people to learn assembler. While they can learn C++. So to speak
12:21:36  * drac_boy points to the slashes
12:21:54  <planetmaker> drac_boy: THAT will NOT be available if you *decode* like *you* suggested
12:22:06  <planetmaker> try again your argument chain please
12:22:07  <drac_boy> planetmaker really? then how come I have a bunch of them from grf folders?
12:22:25  <drac_boy> hrm
12:22:39  <V453000> why would you decode a newgrf anyway? :D
12:22:45  <V453000> you either have source of it or not
12:23:11  <Timberdragon> :| ekk, sorry , i just meant i want to look at what goes into a newgrf file..... perhaps decoding is not the way to go.. but how else does a visual learner figure out how to do this?
12:23:14  <drac_boy> V453000 when you downloaded it and wanted to figure out snips of it
12:23:16  <planetmaker> V453000: in order to understand what they do / modify / get the "source" of those newgrf which you don't have it from
12:23:19  <NGC3982> I don't understand. A decoded GRF is not even relevant if you can obtain the *correct* source?
12:23:24  <V453000> mhm :)
12:24:06  <drac_boy> V453000 mind you thats how I figured out how the whole vehicle id thing worked (including refit affecting it) among a few other things before
12:24:41  <planetmaker> Anyway Timberdragon: don't mind the fuss. Look at how base set or NewGRFs work in NML. Get the source code of NewGRFs to learn
12:24:43  <V453000> idk the tt-wiki is pretty well written for figuring out things :)
12:24:56  <planetmaker> There are a bunch of people who're happy to work with open source
12:25:19  <planetmaker> so that all people can profit from their knowledge and thought. So that people like you have an easier start
12:25:24  <drac_boy> V453000 :)
12:25:46  <drac_boy> V453000 only thing with the wiki yet is lack of more full examples. but thats probably more of a personal thing
12:26:01  <V453000> I annoy people and ask in that case :p
12:26:03  <NGC3982> V453000: The wiki is neat, but some things really need a look-over.
12:26:10  <drac_boy> V453000 heh
12:26:23  <NGC3982> Like it in so many places refers to "Palnetmaker", for instance.
12:26:42  <V453000> what is bad about that
12:26:58  <NGC3982> Simple spelling errors?
12:27:08  <V453000> oh that you mean
12:27:14  <planetmaker> lol
12:27:18  <Timberdragon> from a noob stand point the wiki's are all over the place and its difficult to figure out what is useful and what is outdated info
12:27:36  <NGC3982> "Pal-netmaker" should be a dating site for bromances, by the way.
12:27:37  <NGC3982> :D
12:27:45  <V453000> idk, in the functions and code-things there arent errors, or at least didnt notice any ... and if there is a typo in the description ... :d so wat
12:27:52  *** roadt [~roadt@] has joined #openttd
12:28:04  <NGC3982> Timberdragon: Im with you there.
12:28:41  <Timberdragon> ok so where can i find zbase source to study how he is doing it?
12:29:01  <V453000> [13:07] <@planetmaker> Which basically is what zbase, too:
12:29:11  <Timberdragon> lol oh yeah
12:29:16  <Timberdragon> its open and all X)
12:29:37  <NGC3982> I bet PM owns a DeLorean
12:29:41  <planetmaker> There's two wikis relevant: with tutorials. and with the language descriptions
12:30:40  <planetmaker> the latter only contains the read-built base set, I think. But the first is the project page
12:31:57  <V453000> I really like how is updated often ... all the new changes in NML are there usually
12:32:28  <V453000> true that if you see it once in a few months it might be a bit confusing as stuff is different from last time but ... :)
12:35:19  <planetmaker> V453000: that wiki is the authorative manual. What's not there, that you should not rely on to continue working ;-)
12:35:40  <planetmaker> same as with the newgrf specs wiki. What is not specified is not guaranteed to work ;-)
12:36:06  <V453000> yes :) but when things I had in the newgrf some time ago and was "right" before and now it disappears :D surprise
12:36:33  <planetmaker> err what?
12:36:41  <V453000> like the length
12:36:54  <V453000> or "return" in switches
12:36:55  <planetmaker> that's described in the changes between 0.2 and 0.3
12:37:15  <V453000> haha punished for not reading changelog
12:37:39  <planetmaker> the return might not. Not sure
12:37:50  <planetmaker> but length is :-)
12:38:13  <V453000> well all is good and happy in the end :) it works and I think I am done with coding any new features so it is just eventual copypaste now :)
12:38:23  <planetmaker> :-) good
12:38:33  <planetmaker> until the next new openttd feature :-P
12:38:43  <V453000> just reached 20k lines of code and 20k sprites yesterday ._.
12:38:50  <planetmaker> hehe
12:38:54  <drac_boy> heh
12:39:08  <V453000> the counts are funny though
12:39:28  <V453000> 9 trains from the set have 11000 sprites as they recolour in 17 different colours in cargo subtypes
12:39:49  <V453000> so effectively it is 11k/17 :) just cloned and recoloured
12:40:04  <V453000> funny when in compare to that another 9 trains without such feature have ... 72 sprites
12:44:16  <Timberdragon> decoding .grf files from the zbase project is not the same as decoding newgrf's is that right?
12:48:34  <planetmaker> Timberdragon: it works the same way. But ... I don't see what information you like to retrieve
12:49:00  <planetmaker> but meh, suit yourself
12:49:24  <Timberdragon> i just want to look at the sprites and the folder structures to see how it's all set out
12:49:55  <Timberdragon> i can reproduce much faster then i can learn from reading
12:50:15  <Timberdragon> basicly im a dumb ass ;)
12:50:37  <drac_boy> heh :)
12:50:52  <planetmaker> Timberdragon: a de-coded grf won't tell you anything about "folder structure". Nor about how it was written
12:51:12  <planetmaker> It will give you all sprites in one big file. Sequentially. Not in the form the artist drew them
12:51:28  <planetmaker> thus you'll learn basically not much. Especially not about how the grf was made
12:51:39  <V453000> oh it even gives separate sprites? :d that could be a huge mess
12:51:57  <planetmaker> V453000: it gives you one file with all sprites next to each other separated by some white space
12:52:14  <planetmaker> with 800px width or so by default. However many fit in one row
12:52:20  <V453000> ahh
12:52:29  <planetmaker> thus: quite ugly
12:52:32  <V453000> interesting
12:52:32  <Timberdragon> ok so i need to see something before its encoded, is that possible?
12:52:48  <V453000> look at the source I think :)
12:53:18  <planetmaker> like this:
12:53:30  <planetmaker> that's a de-coded spriteset
12:53:31  <drac_boy> isn't one file seperated by white pretty much the way to draw anyway? even the raw sprites on forum are laid out the same way
12:53:42  <planetmaker> drac_boy: that's pretty much not the way I draw
12:53:51  <planetmaker> and pretty much not the way most people draw tbh
12:54:05  <planetmaker> they make one file maybe per vehicle, separated those 8 sprites. But in a nice patter
12:54:17  <planetmaker> the decoded stuff has all of that meta-info thrown away
12:55:06  *** Devroush [] has joined #openttd
12:55:22  <FLHerne> planetmaker: That reminds me for some reason: the offset for one of the hair types in OGFX seems to be a pixel or two off...
12:55:24  <planetmaker> heck, people even draw the stuff in proper graphics programmes which support layers. And then export the layers, combining different layers to different sprites
12:55:32  <planetmaker> that even works in a scripted fashion
12:55:35  <Timberdragon> i decoded 32bit trains 2 nights ago and noticed the nfo was useles so i trashed the extracted folder
12:56:14  <planetmaker> imho similar reasoning applies to the spritesets from a de-coded grf
12:56:33  <planetmaker> both are not useless. But are the last thing I want to deal with
12:56:53  <Timberdragon> yes i understand
12:57:01  <planetmaker> as convenient like sanding paper used as toilet paper
12:57:07  <Timberdragon> :)
12:57:45  <drac_boy> hmm I took a quick look at forum and I can't find any seperated raw sprites except for the occassional single-vehicle grfs :/
12:58:01  <Timberdragon> ok, im trying to find the before hand source for zbase but i can only find the file which only has the grf files in them... in the past i use SVN to access project source files.. is there something similar here?
12:58:10  <FLHerne> planetmaker: European female hair 1 seems to be 1 pixel too far right?
12:58:19  <planetmaker> Timberdragon: look at the first link I gave... there's a repo link
12:59:14  <planetmaker> FLHerne: dunno by heart. Maybe :-) Can you make a note in the bug tracker or related forum thread please? Or I'll forget
12:59:24  <drac_boy> heh
13:00:18  <planetmaker> ah, Timberdragon, maybe the link was not that good which I gave... it uses sub repositories
13:00:40  <planetmaker> has the 32bpp sprites
13:01:01  <planetmaker> the 8bpp sprites
13:01:08  <Timberdragon> Thats cool, it just means im not going insane after all X)
13:01:23  <planetmaker> the code for the 8bbp+32bpp combined
13:01:39  <planetmaker> all that is solved, if you simply checkout that project via mercurial ;-)
13:02:27  <Timberdragon> haha, now we are talking
13:03:08  * drac_boy only has ftp and svn but wouldn't comment :p
13:03:55  <Timberdragon> i already have tortoiseSVN, can i just use that?
13:04:05  <planetmaker> you need tortoiseHG
13:04:25  <planetmaker> it's not using svn but mercurial
13:04:32  <planetmaker> (mercurial = hg)
13:05:25  <Timberdragon> i see
13:06:31  <planetmaker> thanks for the reply in the forums FLHerne ;-) I restrained myself to answer. Would have been like "please read again" :-P
13:06:44  <planetmaker> which would not have been very friendly ;-)
13:06:49  <drac_boy> heh
13:07:36  <Timberdragon> restarting, brb
13:07:40  *** Timberdragon [] has quit []
13:08:40  <FLHerne> Sorry, it's Hair 2 actually :P
13:09:31  <drac_boy> just wondering FLHerne how're the tiles going?
13:10:41  <FLHerne> I have a fair number of angles done for the asphalt :-) . I'm not exactly working full-out on it though, just doing odd ones when I get bored :P
13:12:42  <drac_boy> heh...ok :p
13:13:07  * drac_boy is thinking about getting back to my one large grf project when I've finished this model layout yet anyhow
13:13:13  *** Timberdragon [] has joined #openttd
13:15:54  *** Timberdragon [] has quit [Remote host closed the connection]
13:17:22  *** Timberdragon [] has joined #openttd
13:19:10  <planetmaker> what kind of grf you're working for, FLHerne?
13:19:40  <Timberdragon> :S Is there a channel log somewhere?
13:20:03  <planetmaker> you didn't miss anything really
13:20:04  <planetmaker> @logs
13:20:05  <DorpsGek> planetmaker:
13:20:42  <Timberdragon> i didnt save the last few address's
13:25:55  <Timberdragon> SUCCESS!!! well, in 3 days after its finished downloading X)
13:36:04  <__ln__>
13:37:49  <V453000> :>
13:38:18  <FLHerne> planetmaker: CHIPS-style newobjects. I'm trying to redraw the groundtiles for all the slope angles :P
13:38:39  <planetmaker> oh :-) Sounds tasty
13:40:45  <FLHerne> More tedious :-(
13:40:54  <Timberdragon> hmm i keep getting dc'd while downloading, does "command returned code 255" mean anything?
13:40:57  <FLHerne> Fills in time gaps though LP
13:41:04  <FLHerne> s/L/:/
13:42:02  <planetmaker> Timberdragon: downloading what?
13:42:25  <Timberdragon>
13:42:43  <planetmaker> hm, ho. There's an issue with the http checkout... things are too large...
13:42:53  <planetmaker> forgot about that
13:43:17  <Timberdragon> transaction aborted... manually this time at least X)
13:44:00  *** Eddi|zuHause [] has quit [Remote host closed the connection]
13:44:15  <planetmaker> hmpf... the only way which currently works... is via ssh
13:44:15  *** Eddi|zuHause [] has joined #openttd
13:44:25  <Timberdragon> ok
13:44:41  <planetmaker> which... requires you to give me the public key of yours
13:47:00  <Timberdragon> ok, where do i find it?
13:48:55  <planetmaker> you probably need to create it...
14:01:23  *** drac_boy [] has left #openttd [I'm done being in this room!]
14:02:53  <Eddi|zuHause> i think when adjusting the purchase list width, we must switch from centering to left aligning the sprites
14:04:15  <Ammler> just to note, we have a new test http hg server running:
14:04:32  <Ammler> this one should allow clone/pull and push without limit
14:05:10  <Ammler> I need to get celery and rabbitmq running and then we will provide hg.o.o with that
14:05:39  <planetmaker> thus try to clone Timberdragon
14:05:46  <planetmaker> same thing, different interface ;-)
14:05:53  <Eddi|zuHause> STR_TINY_BLACK_COMA <-- is that a typo?
14:06:44  <Timberdragon> thank god for that, i was upto Puttygen for key and was about to grab a gun to put myself out of misery X)
14:07:57  <TrueBrain> Eddi|zuHause: it is code for being drunk out of your mind
14:44:09  *** frosch123 [] has joined #openttd
15:14:10  *** nicfer [~Administr@] has joined #openttd
15:18:07  <nicfer> jello
15:19:12  <nicfer> how do I checkout a specific revision with svn?
15:19:40  <planetmaker> svn up -rXXX
15:20:24  <planetmaker> and svn help co is your friend
15:20:33  <nicfer> I just did that
15:20:59  <planetmaker> and?
15:21:52  <nicfer> I did svn -h and didn't saw I could do svn help co
15:22:47  <FLHerne> Downloading a zBase build from fails :-(
15:23:04  <planetmaker> FLHerne: what's the error?
15:24:47  <FLHerne> After a few seconds, the download stalls. Pausing and restarting seems to work, but then it stalls again...and so on
15:25:10  <nicfer> at which directory shall I apply a patch?
15:25:15  <FLHerne> It's finally downloaded :-)
15:25:16  <TrueBrain> do you have that more often FLHerne, with other sites? :)
15:25:33  <nicfer> at /trunk or /trunk/src?
15:25:39  <FLHerne> TrueBrain: Never :P
15:26:03  <planetmaker> does it, by chance, stop at 23MB, FLHerne?
15:26:18  <planetmaker> Ammler: any idea? ^
15:26:21  <TrueBrain> I remember the day I had a damaged cat-5 cable, where the 4th wire was damaged ... transmissions went fine till a certain kb/s, then it just ... stopped .... took for ever to trace and fix :P :D
15:26:35  <Eddi|zuHause> <nicfer> I did svn -h and didn't saw I could do svn help co <- it says that in like the 3rd line
15:27:44  <FLHerne> I didn't check, will try again
15:27:57  <planetmaker> nicfer: that depends on the patch. And whether svn can deal with the patch at all
15:27:58  <Eddi|zuHause> 1st line: usage, 2nd line: version, 3rd line: use "svn help <command>"
15:28:31  <planetmaker> (or rather your patch programme. It's not svn itself which does the patching)
15:28:55  <Ammler> planetmaker: no clue
15:29:11  <FLHerne> planetmaker: No, it doesn't. Just tried twice, 4.4MB and then 7.4
15:29:17  <Eddi|zuHause> nicfer: most patches are made to be applied at trunk/
15:29:29  <FLHerne> Had to pause and restart about 15 times when downloading...seems to md5sum ok though :-)
15:29:47  * planetmaker tries again...
15:30:00  <planetmaker> you mean, like the 15th time it downloaded all 150MB?
15:30:20  <Ammler> apache crap
15:31:03  <FLHerne> No, I just paused the download when it stalled and then had it resume from where it stopped :P
15:31:25  <planetmaker> lol lol. I get like 74 bytes/s :-P
15:31:33  <Eddi|zuHause> Ammler: when i tried to check out FISH recently, hg clone didn't work, but hg clone --pull did work
15:32:42  <planetmaker> now, that's *really* annoying, FLHerne. Thanks for the notification
15:33:15  <TrueBrain> QoS going haywire? :)
15:33:25  <planetmaker> seems like :'-(
15:33:35  <TrueBrain> some mirrors do that to an extreme. .. you start your download at 100 mbit/s, you go: WOW!
15:33:39  <planetmaker> and it's not my local pipe here...
15:33:42  <TrueBrain> then after 10 seconds, it cuts down to 10 KiB/s
15:34:12  <Eddi|zuHause> TrueBrain: bad rate limiting software that takes too long to kick in?
15:34:27  <TrueBrain> in fact, many (free) mirrors still do that
15:35:08  <FLHerne> Here, it literally drops to 0bits/s and stays there :P
15:35:37  <TrueBrain> talking about mirrors, I am trying to piece together mirror v2 for OpenTTD, but sjees, it is hard :(
15:36:07  <TrueBrain> those 100+ MiB GRFs are a real pain :P
15:36:54  <planetmaker> TrueBrain: and... it's only the beginning ;-)
15:37:17  <planetmaker> zbase currently is at 150MB. And it's like 40...50% done
15:37:18  <TrueBrain> I am trying to piece together a solution, where when a mirror hasn't received a file yet, other can already serve a file if they did
15:37:29  <TrueBrain> but how do you know? How do you store it?
15:37:56  <TrueBrain> does the balancer (which redirects you to a mirror) sends HEAD to each file to see if the mirror already has it?
15:38:05  <FLHerne> Give OTTD a BitTorrent client? :P
15:38:14  <TrueBrain> do you make a DB table to keep track of this kind of information? (terribly slow to query a DB every time)
15:38:23  <TrueBrain> and I can't find any software that does anything remotely like this
15:38:36  <planetmaker> hm
15:38:41  <TrueBrain> FLHerne: played Diablo 3 or World of Warcraft?
15:38:46  <TrueBrain> 2 good examples why you should never do that :P
15:38:55  <planetmaker> FLHerne: your start-stop "solution" works... but... meh
15:38:58  <FLHerne> TrueBrain: Nope. What breaks?
15:39:11  <TrueBrain> they allow you to download the client by torrent or http
15:39:15  <TrueBrain> http via the launcher is rate limited
15:39:31  <TrueBrain> when I had a 100 mbit/s pipe, torrent took 20+ hours, http via launcher 10+ hours, http via browser 1 hour
15:39:48  <planetmaker> uh?
15:39:56  <TrueBrain> and those 2 games are not small (in people downloading etc)
15:39:58  <planetmaker> sounds like torrent is limited. Not http
15:40:19  <TrueBrain> planetmaker: its not; just too many people download from those very small upstreams of people ....
15:40:31  <planetmaker> :-)
15:40:32  <TrueBrain> most connections are 10:1 (downstream:upstream)
15:41:12  <planetmaker> yes, round about
15:41:14  <TrueBrain> its always a nice example for me that those solutiosn work well for many low-quality connections ... but suck if you have a decent connection :P
15:41:16  * FLHerne just had to killall OTTD :-(
15:41:28  <planetmaker> ui?
15:41:32  <FLHerne> 32bpp doesn't seem to like my computer, or vice versa :P
15:41:45  <TrueBrain> you are doing it wrong :D
15:41:57  <planetmaker> surely 32bpp doesn't like your computer. Not vice versa :-P
15:42:03  <FLHerne> Perhaps I should find a stable/trunk build and try that :P
15:42:09  <TrueBrain> anyway, if anyone knows a good mirroring solution, either in software or idea, please, do let me know; kinda running thin in solutions here :D
15:42:12  <planetmaker> you need to paint your box first with some colours. Then they'll feel more at home
15:42:40  <TrueBrain> hmm, I might as well post on the forums, might be easier
15:43:14  * planetmaker goes do some needed shopping before continuing the stupidity path with comments here ;-)
15:43:49  <planetmaker> (I only mean mine)
15:46:33  <FLHerne> Mmm. If I set blitter to 32bpp-anim, start 1.2.1, select zBase as baseset and start a new game, should I be seeing 32bpp graphics? :P
15:47:50  <FLHerne> Oops, my fault :-(
15:48:13  <FLHerne> I should probably remove OGFX+ Landscape first :P
15:50:36  <Ammler> I can't download zbase either
15:52:56  <frosch123> FLHerne: you do not even have to set the blitter
15:53:06  <frosch123> unless you force a 8bpp blitter, it will select a 32bpp one itself
15:53:19  <TrueBrain> btw, Ammler, I guess this is one for you: the last 2 days I am receiving timeouts, refused connectiosn and DNS errors when accesing (CompileFarm reports them)
15:53:32  <FLHerne> Does that override old .cfgs, though? :P
15:55:11  <Ammler> I want back my old server, where I didn't need to care about readiness :-(
15:58:14  *** Prof_Frink [~proffrink@] has joined #openttd
15:59:34  <Sacro> \o/
16:02:22  <TrueBrain> there, posted about my issue :P
16:02:38  <TrueBrain> all (wanna-be) sysops: read and reply! :P
16:03:55  <Sacro> TrueBrain: I am a sysop
16:03:59  <Sacro> what can I read?
16:04:03  <TrueBrain> tt-forums
16:04:05  <TrueBrain> my last post
16:04:08  <TrueBrain> can't miss it, I hope :P
16:04:36  <Ammler> download around 8mb, then stop :-(
16:06:44  *** Doorslammer [] has joined #openttd
16:09:30  *** Doorslammer [] has quit []
16:18:30  *** Biolunar [] has joined #openttd
16:22:36  <Ammler> TrueBrain: this would not have happen, if you wouldn't have such a stupid TOS :-P
16:24:32  <Ammler> and instead to make it more common, you made it more silly with the checkbox
16:27:07  *** perk111 [~perk11@] has joined #openttd
16:28:56  <Ammler> a guy, which took the effort to reclaim newships as his grf will also just click that box
16:29:32  <Rubidium> but they can't say they 'forgot' that fact
16:29:38  <Rubidium> they agreed to it twice
16:30:58  <TrueBrain> Ammler: I don't see how it is relevant. This is an issue how to selectively push stuff to a mirror, because the mirrors want that
16:31:41  <Ammler> TrueBrain: maybe I saw the wrong thread from you then :-)
16:32:39  <Eddi|zuHause> Ammler: that post is months old..
16:32:46  *** perk11 [~perk11@] has quit [Ping timeout: 480 seconds]
16:33:12  <Ammler> Eddi|zuHause: still on top
16:33:12  <Eddi|zuHause> TrueBrain: "We talk about 45k files here." <-- maybe make the decision per-directory, and not per-file?
16:33:33  <Eddi|zuHause> Ammler: that's because it's an announcement
16:33:52  <Eddi|zuHause> Ammler: announcements, stickies, normal topics...
16:37:02  <TrueBrain> Ammler: I said: my last post
16:37:05  <TrueBrain> which is not "top post"
16:37:07  <TrueBrain> which is silly
16:37:13  <TrueBrain> :P
16:37:31  <TrueBrain> Eddi|zuHause: given there are roughly 50 files in 1 dir, that still leaves a huge list :P
16:39:19  <Eddi|zuHause> TrueBrain: and i think you forgot a point, there are 3 cases, the file is not there, the file is there, and an old version of the file is there
16:40:22  <Eddi|zuHause> and that no incomplete files are there must be made sure during the rsync
16:42:59  *** andythenorth [] has joined #openttd
16:47:34  <Ammler> Eddi|zuHause: is clone with --pull like pulling every changeset individually?
16:48:06  <Ammler> which url did you use to clone?
16:48:38  <Eddi|zuHause> my exact command line was "hg clone --pull fish"
16:49:26  <Eddi|zuHause> and i don't know how pull exactly works. i just found this entry in the help and though i should try it. "    --pull              Nutzt das 'Pull'-Protokoll um Metadaten zu kopieren"
16:50:01  <Rubidium> TrueBrain: what about splitting the removal and addition of files for mirroring. Then first add the files, then update the cache (maybe once every so many minutes/seconds) and after that run/trigger the removal
16:51:23  *** FLHerne [] has quit [Remote host closed the connection]
16:51:44  *** FLHerne [] has joined #openttd
16:53:53  <Rubidium> maybe have two classes of 'mirrors': full and partial, where partial is just a small subset. Upon query find the closest mirror. If it's full, return that, if it's not figure out whether the file will be on the mirror (can't be that expensive I'd guess if you store paths; then it's just a hash lookup: O(1)). If it is on the subset mirror, return, otherwise get the nearest mirror in the full mirror list
16:55:15  <Rubidium> with store paths is: have a list with the paths that are pushed to the 'small' mirrors (that's maybe 50 entries?)
16:55:55  <Ammler> or use something like mirrorbrain
16:58:32  <TrueBrain> Rubidium: please post that to the thread, so I can read it later when I am not cooking dinner etc ;)
17:00:17  <TrueBrain> (you say it is not expensive, but I think you forget how that hash gets filled/stored/saved/loaded .. it is not as trivial as you want it to appear here ;))
17:00:55  <Rubidium> write the list to a file, inotify
17:01:15  *** roadt [~roadt@] has quit [Ping timeout: 480 seconds]
17:02:02  <Rubidium> (IN_CLOSE_WRITE)
17:02:26  <Eddi|zuHause> that should have been pynotify :)
17:02:46  <Rubidium> Eddi|zuHause: no, pienotify
17:03:08  <Eddi|zuHause> pyinotify does sound like stuttering
17:03:32  <Rubidium> IN_OVEN_OFF ;)
17:04:54  <Rubidium> filling the file with data, that's the trickier part. However, that's the thing that determines whether something is going to be mirrored or not
17:05:25  <Rubidium> and thus it could be some filter over find $binaries
17:06:10  <Eddi|zuHause> if we fix it on a 2-class system, i'd probably make a positive-list of what files (or directories) are in the "small" mirrors.
17:07:37  *** valhallasw [] has quit [Ping timeout: 480 seconds]
17:07:39  *** M1zera [] has joined #openttd
17:07:47  <Rubidium> and with rsync I'd expect the file cache to be hot-ish. time find $binaries | grep log | wc takes a whopping 0.118 seconds
17:07:58  <Eddi|zuHause> so upon incoming request, you hash the directory, ask the file whether the hash is in there, and if yes, return a "small" mirror, if no, return a "large" mirror.
17:13:09  *** roadt [~roadt@] has joined #openttd
17:22:16  *** glx [glx@2a01:e35:2f59:c7c0:ade3:8c94:d691:67a1] has joined #openttd
17:22:19  *** mode/#openttd [+v glx] by ChanServ
17:27:18  <TrueBrain> Rubidium: so you mean make files like:, to indicate the NL mirror has that directory?
17:27:24  <TrueBrain> (and dn't sync that file, ofc)
17:28:16  <Rubidium> no, just make a file with all the files a particular mirror has
17:28:52  <TrueBrain> hmm I wonder if rsync can tell you that
17:29:09  <Rubidium> you tell which files rsync should sync, don't you?
17:29:39  <TrueBrain> hmm, I guess that is another approach
17:29:47  <TrueBrain> I was thinking about simply running rsync over all dirs you want to rsync
17:29:52  <TrueBrain> but I guess you can also suplpy it with a list he should
17:30:08  <TrueBrain> I wonder if that can be done per directory
17:30:49  <TrueBrain> (also, the idea is that per mirror can be selected which files are there, but that would be allowed in this approach I guess)
17:31:03  <TrueBrain> some mirrors explicitly only want the latest stable (non-nightly)
17:31:13  <frosch123> syncing a list of files?
17:31:18  <frosch123> might be tricky for removing files
17:31:30  <TrueBrain> it should remove any file not on the list anymore
17:31:33  <TrueBrain> rsync has an option for that
17:31:58  <Rubidium> TrueBrain: aren't those mirrors that only want latest stable more or less useless; bananas consumes much more I think
17:31:59  <frosch123> oh, i expected it would only remove files on the list which are not at the source location
17:32:09  <TrueBrain> Rubidium: during releases they are not useless
17:32:20  <TrueBrain> (the first few days, basically)
17:32:35  <frosch123> Rubidium: wouldn't those mirrors also want the basesets?
17:33:13  <Rubidium> frosch123: but that's only for first use, after that you update via bananas
17:33:31  <Ammler> rsync has built in .rsync-filter file support
17:33:47  <frosch123> well, but if it handles zbase that would also be some
17:34:06  <TrueBrain> how still will grow exactly, time will tell .. I just want the ability to change
17:34:09  <TrueBrain> now it is either all or none
17:34:13  <TrueBrain> which are too much of an extreme ;)
17:34:20  <frosch123> alternatively we could modify ottd to only download files when they are going to be used :p
17:34:29  <TrueBrain> Ammler: you said exactly nothing; can you give a bit more context to your comment?
17:34:55  <TrueBrain> cool, --files-from accepts directories just fine, that is nice
17:35:24  <TrueBrain> frosch123: as far as I understand, with --files-from rsync sents that list to the mirror, it compares it with its local list, and removes any file not on it (when you say it should delete, that is)
17:35:34  <TrueBrain> then syncs any file not matchines the checksum
17:35:41  <TrueBrain> (which is mtime I believe?)
17:37:46  <Ammler> TrueBrain: well, mirrors can configure filter on directory level themself (or you edit those), point is, rsync supports that natively (option -F, iirc)
17:38:05  <TrueBrain> -F is write-batch
17:38:42  <TrueBrain> please dont 'iirc', but please give concrete useful information .. tnx :)
17:38:44  <Ammler> -F                          same as --filter='dir-merge /.rsync-filter'
17:38:52  <Rubidium> frosch123: but aren't all files used during scanning?
17:38:58  <Ammler> maybe depends on version
17:39:06  <TrueBrain> you still haven't really told me wtf it does
17:39:14  <TrueBrain> absolutely unclear to me, sorry :)
17:39:19  <TrueBrain> (I might be slow, I dunno)
17:40:04  <Ammler> just how to provide/manage mirror dependend syncing
17:40:14  <TrueBrain> that is the subject, yes :)
17:42:25  <TrueBrain> still no clue what you try to point out ..
17:42:59  <frosch123> TrueBrain: --delete-excluded seems to be the option
17:43:04  <Ammler> imo easiest way to allow mirrors to configure, which files they want on their disk
17:43:15  <TrueBrain> frosch123: yup :)
17:43:38  <TrueBrain> frosch123: it is a USE WITH CARE flag
17:43:49  <TrueBrain> Ammler: and how would we do such thing?
17:44:02  <Ammler> but no clue, how that is done with rsync < 3
17:44:07  <frosch123> Rubidium: we could also just download the 8bpp normal zoom part
17:44:08  <TrueBrain> ...
17:44:22  <Ammler> maybe it has --filter
17:44:24  <frosch123> and update the rest only via password or so :p
17:45:01  <frosch123> TrueBrain: rsync can also read the filelist from the destination
17:45:12  <TrueBrain> frosch123: I expected as much; that is great :)
17:45:22  <TrueBrain> happen to know how?
17:45:38  <frosch123> though i don't know how the balacner would get that info
17:45:54  <TrueBrain> file-cache
17:46:00  <TrueBrain> rsync script starts
17:46:03  <TrueBrain> when finishes, it runs that
17:46:13  <TrueBrain> balancer detects file is changed
17:46:15  <TrueBrain> reloads
17:46:35  <Ammler> replace balancer with mirrorbrain
17:46:38  <TrueBrain> would mean not having to care about rsync giving errors
17:47:13  <frosch123> TrueBrain: you mean --files-from=:/foo/bar with ":" in front of the path?
17:47:24  <TrueBrain> frosch123: I have no clue?
17:47:30  <TrueBrain> what would it do?
17:47:31  <frosch123> i just read the man page
17:48:19  <frosch123> you set a path for a file on the remote machine, which can then contain stuff like "releases/" or "/releases/LATEST" or "bananas"
17:48:19  <TrueBrain> ah, remote host location
17:48:22  <TrueBrain> no, I don't see how that is useful
17:48:33  <TrueBrain> the mirror hosters won't want to maintain such files
17:48:40  <TrueBrain> we can just do that for them, that is perfectly fine
17:48:54  <frosch123> it would not be a list of files, but a list of top-level dirs
17:49:01  <frosch123> but again, the balancer would know about that
17:49:07  <TrueBrain> like: */releases/* ;)
17:49:14  <TrueBrain> how I think it would work, if I get this all right
17:49:20  <TrueBrain> we have a file, say,
17:49:24  <TrueBrain> it contains stuff like */releases/*
17:49:27  <TrueBrain> */bananas/*
17:49:28  <TrueBrain> I dunno
17:49:29  <frosch123> yeah, so not on remote
17:49:32  <TrueBrain> gives directories we want
17:49:38  <TrueBrain> we run rsync
17:49:41  <TrueBrain> it syncs it etc
17:49:51  <TrueBrain> then we run a command (??), which tells us which files are now on the mirror
17:49:53  <TrueBrain> store that in
17:50:02  <frosch123> you can also run that file through some preprocessor and replace $LATEST with 1.2.0 :p
17:50:02  <TrueBrain> the balancer reads that file, and per file now knows who has it
17:50:16  <TrueBrain> yup, would work indeed
17:50:27  <Eddi|zuHause> TrueBrain: so the balancer picks a mirror, looks up whether it has that file, then picks next mirror?
17:50:33  <TrueBrain> Eddi|zuHause: no
17:50:48  <TrueBrain> I wouldn't know why he would do that ...
17:50:54  <TrueBrain> "and per file now knows who has it"
17:50:59  <TrueBrain> it knows, PER FILE, who has it
17:51:11  <TrueBrain> so .... when you ask a file at the balancer, he KNOWS who has it
17:51:15  <TrueBrain> so it picks a mirror from that pool
17:51:37  <Eddi|zuHause> ah, so cache has a mapping from filename:mirror-list
17:51:45  <Eddi|zuHause> for each file
17:52:00  <TrueBrain> let me try this again :)
17:52:05  <TrueBrain> has a filelist, wha tis on the NL mirror
17:52:15  <TrueBrain> the balancer reads this file in an array files[file] = [ "nl", "de", ]
17:52:27  <TrueBrain> more clear?
17:52:32  <frosch123> can ottd handle downloading stuff from different mirrors? i.e. does it notice which mirror server it, or does that not matter at all?
17:52:32  <Eddi|zuHause> yes.
17:52:55  <frosch123> (i mean a scenario where the user selects two files via ingame gui, which are on different mirrors only)
17:52:58  <TrueBrain> frosch123: don't know if I understand your question completely, but you do bring up a good question ..
17:53:04  <TrueBrain> I cannot remember how we told the client about a mirror
17:53:10  <TrueBrain> if we did it once in a package, or for every entry
17:53:21  <TrueBrain> Rubidium: do you remember?
17:53:21  <Eddi|zuHause> TrueBrain: can't use dns-based balancing in that case, only redirects
17:53:25  <frosch123> it was some standard http stuff iirc :p
17:53:30  <frosch123> though i have no idea about http
17:53:33  <TrueBrain> Eddi|zuHause: we don't use dns based balancing
17:53:35  <frosch123> it just sounded standard :p
17:53:49  <Eddi|zuHause> TrueBrain: then i don't see why it would not work
17:54:17  <Eddi|zuHause> assuming the game does not do any magic to combine requests
17:54:25  <TrueBrain> ottd_content can reply in 2 ways: either a http, or an internal entry
17:54:27  <Eddi|zuHause> and starts separate requests for each file
17:54:28  <TrueBrain> the client then downloads it
17:54:34  <TrueBrain> but what I don't remember, is if it caches the redirect
17:55:15  <Eddi|zuHause> TrueBrain: i think that would be bad architecture. the game would contain logic about the structure of mirrors, which it really shouldn't
17:55:31  <TrueBrain> Eddi|zuHause: with full mirrors, that is no problem at all
17:55:36  <TrueBrain> and has nothing to do with bad architecture
17:55:38  <TrueBrain> but with optimization
17:55:47  <TrueBrain> but I think we didn't do it
17:55:50  <TrueBrain> for statistics reasons
17:56:02  <TrueBrain> my memory is so terrible at times :(
17:56:36  <TrueBrain> url = mirrors[m]['url'] + url
17:56:39  <TrueBrain> well, that answers that question
17:56:46  <TrueBrain> so clients get the mirror URL
17:56:52  <TrueBrain> so that would be no issue at all
17:57:22  <Rubidium> that's about what I wanted to say
17:57:36  <TrueBrain> so yeah, then this approach should work
17:57:42  <TrueBrain> only remaining question: how do you get a remote file list? :D
17:58:37  <Rubidium> ugh... already people are mentioning Sintnicolaas :(
17:59:00  <Eddi|zuHause> you mean after rsync runs, get the filelist to doublecheck?
17:59:16  <TrueBrain> Eddi|zuHause: please do read what I wrote :P
17:59:20  <TrueBrain> nothing 'double check'
17:59:20  * Rubidium should eat something
17:59:25  <TrueBrain> Rubidium: enjoy :)
17:59:31  <Eddi|zuHause> TrueBrain: i phrased that badly
17:59:44  <Eddi|zuHause> TrueBrain: i meant: "do not trust that rsync ran correctly, read the actual result"
17:59:51  <TrueBrain> yes
17:59:52  <TrueBrain> it is needed
17:59:59  <TrueBrain> we "often" have mirrors that run out of diskspace
18:00:03  <TrueBrain> how ever annoying, but it happens
18:00:22  <Eddi|zuHause> and rsync itself cannot handle that?
18:00:26  <TrueBrain> it might
18:00:29  <TrueBrain> but that is a lot of work tbh
18:00:37  <TrueBrain> rsync is not really .... userfriendly in its return values
18:00:45  <TrueBrain> what is easier than: rsync && rsync-list ;)
18:00:54  <Eddi|zuHause> so you ssh into it, run ls, and read the output?
18:00:54  <TrueBrain> maybe even rsync; rsync-list
18:00:56  <TrueBrain> ;)
18:00:59  <TrueBrain> we often cannot ssh
18:01:01  <TrueBrain> only rsync-ssh
18:01:07  <TrueBrain> (key-restriction-based-access)
18:01:38  <Ammler> also that would not allow to provide a file as soon as it is uploaded
18:02:05  *** nicfer [~Administr@] has quit [Read error: Connection reset by peer]
18:02:25  <TrueBrain> as soon as rsync finishes, yes
18:02:40  <TrueBrain> which has a hidden benefit: a nightly has to be completely uploaded before a mirror starts serving it
18:02:57  <TrueBrain> might avoid a few cases of: why isn't there XXX, but YYY is there already!
18:03:02  <TrueBrain> its a small benefit :P
18:04:31  <TrueBrain> list-only ofc would be sufficient to list remote
18:04:40  <TrueBrain> as you can just put the remote server as 'source'
18:04:41  <TrueBrain> lolz
18:04:49  <TrueBrain> I was searching for a too complex solution :D
18:05:01  <TrueBrain> btw, Ammler, MirrorBrain has a few 404 pages on their homepage, which is annoying :P
18:05:11  <TrueBrain> and we have a very special balancer
18:05:16  <frosch123> something like running rsync a second time with --dry-run --itemize-changes might give a list of files which failed to transfer
18:05:17  <TrueBrain> as it can also spew out ottd:// urls ;)
18:05:33  *** valhallasw [] has joined #openttd
18:05:33  <TrueBrain> frosch123: that might also be nice, yes
18:05:43  <TrueBrain> good idea tbh
18:05:49  <TrueBrain> as I can trigger emails on that
18:06:30  <TrueBrain> frosch123: wait, no .. what if the filesystem changed in that time?
18:06:34  <TrueBrain> a new upload for example?
18:06:35  <Ammler> TrueBrain: I don't know mirrorbrain, I just know SUSE or KDE uses it
18:06:44  <TrueBrain> another reason btw for you Eddi|zuHause, to list the files remotely ;)
18:06:51  <TrueBrain> the local system can ben changed already :D
18:07:01  <frosch123> TrueBrain: well, that means that you cannot use that mirror for that file
18:07:03  <frosch123> which is correct
18:07:12  <TrueBrain> frosch123: but I cannot trigger emails on it :P
18:07:12  <frosch123> you just need to reconsider your email
18:07:17  <TrueBrain> ;)
18:07:21  <Ammler> there is one url for every suse package, no need to search for mirrors like you do with other distros
18:07:21  <frosch123> except you add some threshold
18:07:22  <TrueBrain> so I am better off listing the remote files, not?
18:07:26  <TrueBrain> as then I am sure I have the right list :)
18:07:27  <frosch123> if it fails on 10 files or more
18:07:59  <frosch123> just listing the files does not tell you whether they are complete, does it?
18:08:15  <frosch123> rsync --dry-run would list you the files which need transfering resp. need no transfering
18:08:23  <TrueBrain> frosch123: rsync promises that
18:08:31  <TrueBrain> if a transfer fails, it removes the file
18:08:47  <TrueBrain> I will have to test what it does if a file already existed, but .. that we can test :)
18:08:59  <TrueBrain> problem about your --dry-run approach, is that it would be a negative list
18:09:03  <TrueBrain> which are annoying to handle tbh :P
18:09:10  <TrueBrain> as ... I have to list all files on the local system
18:09:13  <TrueBrain> then run --dry-run
18:09:16  <TrueBrain> and remove the files it returns
18:09:39  <TrueBrain> but I do agree, we need to be sure files transfered correctly
18:09:50  <TrueBrain> which brings me to the question: how does rsync detect?
18:09:53  <TrueBrain> mtime, md5, ... ?
18:10:12  <frosch123> TrueBrain: it can also list the files which need no transfering
18:10:12  <TrueBrain> Ammler: pretty much like OpenTTD does? :P
18:10:21  <frosch123> TrueBrain: rsync uses checksums
18:10:26  *** NGC3982 [] has quit [Remote host closed the connection]
18:10:28  <frosch123> it can even transfer binary differences
18:10:35  <TrueBrain> that is when it detected a change, yes
18:10:41  <TrueBrain> one does not imply the other
18:10:44  <TrueBrain> or exclude/include even ;)
18:10:47  *** masch_ [] has joined #openttd
18:10:55  <TrueBrain> checksums sounds very expensive
18:10:59  <Ammler> TrueBrain: yes, but that it does because ervery mirror has everything, you need something to support selective, don't you?
18:11:01  <TrueBrain> as it would trigger a lot of traffic on the remote
18:11:12  <TrueBrain> Ammler:<url> always works
18:11:14  <TrueBrain> 1 URL
18:11:17  <TrueBrain> which directs you to a valid mirror
18:11:18  *** DorpsGek is now known as Guest2285
18:11:24  *** DorpsGek [] has joined #openttd
18:11:27  *** mode/#openttd [+o DorpsGek] by ChanServ
18:11:43  <TrueBrain> frosch123: are you sure it uses checksums?
18:12:19  *** Mazur [] has quit [Read error: Operation timed out]
18:12:26  <frosch123> TrueBrain: --checksum
18:12:32  *** Stimrol [] has quit [Remote host closed the connection]
18:12:40  *** Mazur [] has joined #openttd
18:12:44  <TrueBrain> so normally it does mtime + size
18:12:47  <frosch123> so yeah, it might not de by default
18:12:54  <TrueBrain> you have to explicitely say to do checksums :)
18:12:58  <TrueBrain> k, that is what I would expect
18:13:06  <Ammler> TrueBrain: something like
18:13:06  <TrueBrain> TCP transfers are guarenteed
18:13:09  *** Stimrol [] has joined #openttd
18:13:27  <TrueBrain> frosch123: I will have to test what happens exactly if a transfer gets aborted
18:13:31  <TrueBrain> Ammler: yes
18:13:37  <TrueBrain> minus the binaries I beliee
18:13:39  <TrueBrain> not sure
18:13:40  <TrueBrain> :P
18:13:41  *** masch [] has quit [Ping timeout: 480 seconds]
18:13:55  <Ammler> does not matter, both don't :-)
18:13:56  *** NataS [] has quit [Read error: Connection reset by peer]
18:14:08  <TrueBrain> hmm, connection is SLOW
18:14:10  <TrueBrain> holy crap
18:14:27  <TrueBrain> @whoami
18:14:27  <DorpsGek> TrueBrain: TrueBrain
18:14:34  <TrueBrain> so it is me ...
18:14:50  *** Guest2285 [] has quit [Ping timeout: 480 seconds]
18:15:03  * andythenorth ponders forums
18:15:03  <TrueBrain> 120ms latency ... 20% packet loss ...
18:15:05  <TrueBrain> lovely
18:15:37  <TrueBrain> Ammler:
18:15:39  <TrueBrain> as example URL
18:15:54  *** cyph3r [] has quit [Read error: Connection reset by peer]
18:15:55  *** NGC3982 [] has joined #openttd
18:16:19  <Ammler> according to the mirrorbrain docs, that should work
18:16:25  <TrueBrain> okay, with all the ideas here, I think that might work ...
18:16:29  <Ammler> and how suse mirrors are handled
18:16:39  <TrueBrain> Ammler: again, we cannot use that, as we also spit out ottd:// URLs
18:16:52  <TrueBrain> it is just in reply to what you said: Suse does it like that; all I said was: so do we
18:17:03  <TrueBrain> we don't list the mirrors next, no, we just redirect you there
18:17:11  <Ammler> oh, I thought, you need to have all mirrors provide everything
18:17:11  *** Nat_aS [] has joined #openttd
18:17:12  <TrueBrain> I hate those redirect pages :P
18:17:34  <TrueBrain> huh? I think you missed like 90% of what we talked about :P
18:17:43  <TrueBrain> the current software needs full mirrors for listing see thread
18:17:46  <frosch123> TrueBrain: --ignore-errors might be needed as well
18:17:49  <TrueBrain> we just spend 30 minutes talking how to resolve that :)
18:17:56  <Ammler> indeed
18:17:58  <TrueBrain> frosch123: can you make a list of all those commands for me and post it?
18:18:07  <TrueBrain> I will forget, and this channel is a bitch to get a backlog on :D
18:18:47  <TrueBrain> tnx already :)
18:18:55  <TrueBrain> Rubidium / Eddi|zuHause: also tnx a lot for the input; this might just work :D
18:19:29  <TrueBrain> Ammler: for BaNaNaS files protected under ToS from public publishing, we reply ottd:// URLs, which only works for the OpenTTD client
18:19:42  <TrueBrain> I don't see how I can add that to MirrorBrain .. and honestly, the balancer is shit easy/small/fast/etc :P
18:19:57  <TrueBrain> Python and its SimpleHTTPServer ... :D:D :)
18:20:19  <Ammler> those files, you don't mirror anyway
18:20:30  <TrueBrain> but the mirror replies to the request for them :)
18:20:38  <TrueBrain> its a special page with special conditions :)
18:20:43  <TrueBrain> s/mirror/balancer/
18:21:19  <TrueBrain> anyway, I got to go for the evening. Tnx all, happy to finally see some way to resolve this clean and nicely :D
18:21:39  <CIA-4> OpenTTD: translators * r24461 /trunk/src/lang/ (belarusian.txt luxembourgish.txt portuguese.txt turkish.txt):
18:21:39  <CIA-4> OpenTTD: -Update from WebTranslator v3.0:
18:21:39  <CIA-4> OpenTTD: belarusian - 10 changes by Wowanxm
18:21:39  <CIA-4> OpenTTD: luxembourgish - 1 changes by Phreeze
18:21:39  <CIA-4> OpenTTD: portuguese - 51 changes by ricardoespsanto
18:21:41  <CIA-4> OpenTTD: turkish - 151 changes by otrkmen
18:22:06  <Ammler> I just hope, you get it fast working, our devzone seems not to be able to handle it
18:31:23  *** Chris_Booth [] has joined #openttd
18:34:46  *** confound_ [] has quit [Remote host closed the connection]
18:34:48  *** confound [] has joined #openttd
18:36:28  *** lugo [] has quit [Quit: EvoSurge - Free & Premium IRC Bouncers on Demand -]
18:37:41  <neofutur> still trying to configure my city builder server
18:37:51  <neofutur> where shoudl I put the ewgrf settings like [preset-lcb-temp] ?
18:38:27  <frosch123> into openttd.cfg likely
18:38:30  <neofutur> ( planetmaker and Rubidium thanks for the answers ;)
18:38:32  <frosch123> just append it at the end
18:38:45  <Eddi|zuHause> presets are probably fairly useless on a dedicated server
18:38:54  <neofutur> NGC3982: yup trying to setup a city builder server before luukland shutdowns )
18:39:06  <frosch123> Eddi|zuHause: good point
18:39:13  *** Wolf01 [~wolf01@] has joined #openttd
18:39:19  <Wolf01> evenink
18:39:19  <neofutur> patched client and servers are working, but cities are still needeing nothing
18:39:28  <NGC3982> neofutur: I guess i could host my own? :)
18:39:40  <neofutur> I have them in openttd.cfg but they dont seem to be applied, cities needing nothing
18:40:02  <Eddi|zuHause> neofutur: make sure it uses the right .cfg?
18:40:10  <neofutur> NGC3982: eh for sure if you have a dedicated server
18:40:38  <neofutur> Eddi|zuHause: yes i launch it with -c /usr/local/share/games/openttd/openttd.cfg
18:40:56  <NGC3982> neofutur: ;)
18:41:35  <Eddi|zuHause> neofutur: what are the citybuilder-related settings, and what does the console think their values are?
18:41:42  *** GBerten2936 [] has joined #openttd
18:42:23  <Eddi|zuHause> frosch123: "--dry-run --itemize-changes --itemize-changes | grep '^.' " that doesn't look right
18:42:23  *** perk111 [~perk11@] has quit [Quit: Miranda IM! Smaller, Faster, Easier.]
18:43:11  <frosch123> if you mean the duplicated option, that is correct
18:43:29  <Eddi|zuHause> may need better documentation then :)
18:44:52  <frosch123> better?
18:45:13  <Eddi|zuHause> yes :
18:45:14  <Eddi|zuHause> :)
18:46:14  <andythenorth> TrueBrain: I have only scan read the mirrors thing and I am sick and my head is fuzzy
18:46:32  <Eddi|zuHause> hehe :)
18:46:35  <andythenorth> but sounds like a problem usually solved by CDN (the large file type, not the small file type)
18:46:49  <frosch123> duplicate options are sometimes used as enhancement
18:46:59  <frosch123> -v is verbose, -v -v is very verbose and so on :)
18:48:10  <neofutur> Eddi|zuHause: i added the settings found in the pdf
18:48:13  <neofutur>
18:48:21  <neofutur> at the end of openttd.cfg
18:48:40  <neofutur> theres no graphical display on the server, how can I use the console from the client ?
18:48:50  <Eddi|zuHause> neofutur: ah, so you need to copy one of the [preset] entries to [newgrf]
18:49:03  <neofutur> ok
18:49:08  <neofutur> i was suspecting something like that
18:49:15  <neofutur> the pdf is not very detailed about this
18:49:49  <neofutur> ok I just put
18:49:50  <neofutur> cbreqfull.grf = 0 250 500 100 2 40 1000 100 5 250 1500 5 9 111 0 5 10 59 2500 2 11 125 0 10 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
18:50:00  <neofutur> in the [newgrf] section
18:50:15  <frosch123> you can drop the trailing 0 probably
18:50:43  <andythenorth> so we don't pay mirrors, right?
18:50:50  <Eddi|zuHause> well that depends on the grf. there's a difference between "unset" and "zero"
18:50:56  <frosch123> and if you use it like that, you have to make sure that the file is actually in the root "newgrf" dir
18:51:31  <frosch123> andythenorth: we might offer them a cake at the 25k or 30k party
18:51:48  <frosch123> s/cake/piece of cake/
18:52:54  <Eddi|zuHause> unless the cake is a lie.
18:53:36  <NGC3982> Unless it's also a quantum cake?
18:53:50  <NGC3982> So it can be true and false at the same time
18:53:53  <NGC3982> While not being a cake
18:53:59  <NGC3982> ..
18:54:44  <frosch123> if a cake is too continental, i can also make muffins, they are more archipelagish
18:54:53  <NGC3982> :)
18:55:15  <neofutur> I can help providing mirrors if needed
18:55:27  <neofutur> ( also for a free CDN appengine is great )
18:55:29  <frosch123> mirroring cakes?
18:56:52  <frosch123> neofutur: have a look at
18:57:13  <frosch123> since we currently do not support partial mirrors (see discussion above), the requirements are quite big
18:58:57  <frosch123> TrueBrain: is there a more up-to-date list? e.g. one which mention the proper host of our main server :)
19:01:57  <neofutur> ok the cb setting is there now
19:02:40  <neofutur> and the client is required to download a newgrf "city builder cargo requirements"
19:03:08  <neofutur> that cant be found, i have the same problem with the other city builder server on the metaserver list
19:03:49  *** valhallasw [] has quit [Ping timeout: 480 seconds]
19:03:53  <andythenorth> TrueBrain: probably of no help at all:
19:04:58  <neofutur> frosch123: thanks for the link, i ll contact
19:05:12  <neofutur> I have dedicated servers in france with unlimited transfer
19:08:58  <neofutur> 100 GB is possible, recent servers have 2 TB raid
19:12:22  <neofutur> ah i needed the newgrf from the citybuider patch on the client side
19:12:32  <neofutur> it seems theres nowhere they can be downloaded online
19:13:01  <neofutur> yeah towns now need cargo !
19:13:38  <neofutur> the server is named "testing citybuilder server" if anyone wants to help me test it
19:17:16  * neofutur emailed truebrain for the mirror proposal
19:19:56  *** Wolf03 [~wolf01@] has joined #openttd
19:19:56  *** Wolf01 is now known as Guest2302
19:19:56  *** Wolf03 is now known as Wolf01
19:19:56  *** M1zera [] has quit [Read error: Connection reset by peer]
19:22:19  *** Zeknurn [] has quit [Remote host closed the connection]
19:22:45  <neofutur> TrueBrain: you have mail ;) query me for any question for the mirror proposal
19:23:01  *** Zeknurn [] has joined #openttd
19:23:42  <planetmaker> Eddi|zuHause: the failure to find "NoSound" actually might well be related. Every OpenTTD comes with that soundset...
19:25:44  <andythenorth> I take it that we don't have so many donations we can afford a CDN?
19:25:44  *** Guest2302 [~wolf01@] has quit [Ping timeout: 480 seconds]
19:26:31  <planetmaker> what's a CDN?
19:26:31  <planetmaker> nvm
19:27:02  <andythenorth> for us CDN means 'someone else has to worry about distributing the files around the world'
19:27:03  <andythenorth> maybe
19:27:07  <andythenorth> unless they screw up :P
19:27:21  <andythenorth> but it's €some per terabyte
19:27:44  *** valhallasw [] has joined #openttd
19:27:44  * andythenorth points and waves at the 'buy newgrfs, €1 per download' idea :P
19:28:11  <planetmaker> we only allow signed newgrfs
19:29:21  <neofutur>!
19:29:40  <neofutur> that is a cdn, content is distributed on many different ips in different datacenters
19:29:51  <neofutur>
19:30:09  <neofutur> allows you to use appengine as a great free cdn
19:30:34  <neofutur> hum the town says it need steel to grow, but the stops are not needing it
19:31:12  <planetmaker> <-- now. This is a question I happily answered :-) Haven't seen it ever :-)
19:32:05  *** Chris_Booth [] has quit [Ping timeout: 480 seconds]
19:42:21  *** roadt [~roadt@] has quit [Ping timeout: 480 seconds]
19:45:11  *** KouDy1 [] has quit [Quit: Leaving.]
19:58:28  *** Progman [] has joined #openttd
20:17:29  *** telanus [] has quit [Quit: Konversation terminated!]
20:17:55  <neofutur> yeah, got a growing town on my citybuilder server for the fiereet time
20:18:04  <neofutur> first time
20:19:18  <neofutur> anyone know why luuklnd will shutdown his services
20:19:31  <neofutur> and why he sets such a high price ( 200 Keur ) ?
20:23:35  <Xaroth> because he's a muppet?
20:24:11  <neofutur> well he built something great, working well
20:24:17  <Xaroth> he's got something against openttd, but at the same time, he works with it
20:24:24  <Xaroth> which is a classic case of being a muppet
20:24:28  <neofutur> his servers are by far the most active
20:25:00  <neofutur> first i understood the price was 200 eur, I thought I d try to buy it :p
20:25:20  <Xaroth> heh
20:27:49  <frosch123> you can buy luukland for 200k€ ?
20:28:17  <frosch123> who would do that? they rather should ask for 200M€, that might get some facebook inverstors
20:29:24  <neofutur> frosch123: yup 200000 eur
20:29:38  <andythenorth> so that's about £14 sterling?
20:29:39  <neofutur> if no one pays he will shut it all down
20:32:52  <planetmaker> neofutur: that price tag exists as long as the servers exist
20:33:17  <Rubidium> no, it was half of that price
20:34:14  <planetmaker> it's a bad attempt trying to annoy us openttd devs, to illustrate that they don't share their ideas and code with us devils
20:35:09  <neofutur> its related to a licence problem ?
20:35:20  <Rubidium> yes, those OpenTTD devs are bad because they care more about the security of all OpenTTD users than the fun/hacks of a few
20:36:04  <neofutur> k, so better try to build a team around this free city building patch
20:36:19  <Xaroth> better to try and use NoGO
20:36:21  <planetmaker> neofutur: I would suggest... to use the game script approach
20:36:24  <neofutur> is there any plans to accept this city building patch in the official client ?
20:36:30  <planetmaker> no. never
20:36:31  <Xaroth> it can probably do 90%, if not more, of what luukland's servers do
20:36:39  <neofutur> why ? thats something people love ?
20:36:43  <planetmaker> as you can do the same with game scripts
20:36:49  <Xaroth> neofutur: NoGo can already do it
20:36:55  <Xaroth> why implement something that already can be done?
20:36:57  <planetmaker> which is the interface written for that exact purpose
20:37:26  <planetmaker> and there's at least three gamescripts in the online content which deal to some extent with town growth
20:37:28  <Xaroth> more so, why implement something -that cannot be changed without recompiling-, when there's already something in place that works somewhat dynamicly
20:37:30  <planetmaker> did you look at them?
20:38:28  *** KritiK [] has joined #openttd
20:38:37  <neofutur> no, this patch seem to provide a working server, i found no howto links to how to implement this with this nogo
20:38:50  <neofutur> and no working citybuilding server ewither
20:38:53  <Xaroth> that's because with NoGo you're not limited to what the script does..
20:38:55  <neofutur> either
20:39:05  <planetmaker> ehm...
20:39:11  *** Pixa [] has joined #openttd
20:39:16  <planetmaker> game scripts do run serverside...
20:39:44  <planetmaker> really. Try to understand what game scripts do. How they work. How they can help you there
20:39:54  <neofutur> are there some examples of this kinf of scripts ?
20:39:56  <planetmaker> It really was written for that exact purpose. Just as a general API
20:40:03  <Xaroth> game scripts can already do more than the citybuilding patches do
20:40:04  <planetmaker> ... yes.
20:40:08  <neofutur> how many hours of work would I need to make it a citybuilding server ?
20:40:18  <planetmaker> 0
20:40:20  <neofutur> anyone ever tried ?
20:40:28  <planetmaker> just grab one of the game scripts which does that...
20:40:40  <planetmaker> undefined, if you want them to do exactly what *you* want
20:40:46  <neofutur> k I ll search again more info about this
20:40:47  <Xaroth> neofutur: judging from your lack of initiative, more than you think...
20:40:54  <planetmaker> then it's your programming knowledge in squirrel which defines it
20:41:48  <neofutur> Xaroth: lack of initiative on what ?
20:42:04  <neofutur> for now my initiative is setting up a server, coming here on irc, trying to understand ?
20:42:13  <neofutur> what is youre problem Xaroth ?
20:42:20  <Xaroth> neofutur: well, we only mentioned NoGo 5 times, yet you only ask questions about it that can be googled/wiki'ed in seconds
20:42:50  <Xaroth> so to me, that shows a severe lack of initiative.. you want to make it work, but you don't want to spend any effort in doing so
20:43:09  <neofutur> no, I just discover this nogo thing
20:43:21  <neofutur> for now i have a working citybuilding based on the patch
20:43:34  <Xaroth> yes, but as mentioned several times already
20:43:34  <glx> using a patched server is rarely a good idea
20:43:42  <Xaroth> patches are not useful
20:43:54  <neofutur> patched server is not my problem, but the fact it needs a patched client is for sure a problem
20:44:06  <Xaroth> and that's where NoGo comes in :)
20:44:15  *** KingPixaIII [~pixa@] has quit [Ping timeout: 480 seconds]
20:44:25  <Xaroth> there's a big post on the forum about it
20:44:31  <Xaroth> and no doubt several wiki pages
20:45:23  <neofutur> Xaroth: thanks for the link
20:46:00  <Xaroth>
20:46:17  <Xaroth> and
20:46:18  <Xaroth>
20:46:31  <neofutur> thanks, that will help me stop bothering you by talking in here
20:46:44  <Xaroth> it doesn't bother me at all
20:48:51  <planetmaker> @logs
20:48:51  <DorpsGek> planetmaker:
20:52:35  *** M1zera [] has joined #openttd
20:54:20  <neofutur> so , having a look at
20:54:53  <neofutur> the nearest script to a citybuilding game would be CityDomination ?
20:54:58  <andythenorth> bye
20:54:58  *** andythenorth [] has left #openttd []
20:55:12  <planetmaker> It's one of those three I mentioned at 9am this morning, yes
20:57:09  <planetmaker> the "neighbors are important" is another very interesting one
20:59:23  <planetmaker> the city growth limiter might give interesting ideas as well. And as 4th even the transportgoals might be interesting
20:59:23  <neofutur> k now id need to learn this squirrel language
20:59:23  <planetmaker> neither does all what you want, most probably. But going by those 4 things, it's easy to extent them
21:00:00  <neofutur> any good links for learning this squirrel language ?
21:00:13  <frosch123> noone yet coded though an external highscore to put on some website
21:00:18  *** Chris_Booth [] has joined #openttd
21:00:25  <frosch123> but nogo in principal also supports that
21:00:38  <Yexo> neofutur: and maybe NoAI tutorial on
21:03:02  <neofutur> many thanks
21:08:10  <neofutur> ok its near enough to C++ to seem possible
21:14:56  *** drac_boy [] has joined #openttd
21:15:11  <drac_boy> hi
21:21:47  <FLHerne> drac_boy: Hi (yet again) :-)
21:22:11  <drac_boy> hi tiles-bored english person :p
21:22:12  <drac_boy> heh heh
21:22:53  <frosch123> drac_boy: from the other side of the ocean?
21:23:30  <frosch123> i had a question yesterday which i still do not know the answer: do americans normally speak about "cooking a cake" instead of "baking a cake"?
21:24:53  <drac_boy> isn't someone from britan considered english?
21:24:59  <drac_boy> or I must be mixing up my memory :-s
21:25:25  <frosch123> a scotsman might kill you for that
21:25:51  <__ln__> or a welshman, such as batman
21:27:24  <drac_boy> heh ok fair enough
21:28:55  <Rubidium> drac_boy: just like all French people put "Stop" on their "Stop" signs, except those not being French but speaking French near America; they put "Arrêt" on their signs
21:29:52  <__ln__> "near America"....
21:30:02  <Rubidium> likewise Americans, though speaking English are not using the same words/phrases
21:31:42  <Rubidium> e.g. caboose vs guard/brake van
21:32:02  <Rubidium> or cotton candy vs candy floss
21:32:31  <drac_boy> heh
21:32:36  <frosch123> or "i could care less" vs "i coudn't care less" :p
21:33:13  <Rubidium> or streetcar vs tram
21:33:17  <frosch123>
21:39:49  *** Rhamphoryncus [] has quit [Quit: Rhamphoryncus]
21:41:38  <neofutur> [2012-08-08 23:05:09] dbg: [misc] String too long for destination buffer
21:41:55  <neofutur> what does this mean ? a buggy client  or some kind of attack against the server ?
21:41:56  <__ln__> also some verbs have regular past tenses in the US, whereas the same verb is irregular in britain.  can't remember an example.
21:43:13  <neofutur> Rubidium:
21:43:22  <neofutur> stopper is a french verb
21:43:37  <neofutur> you can say "je stoppe" in french
21:43:53  <neofutur> but yes, should be stoppe and not just stop
21:43:58  <Rubidium> neofutur: I know that
21:44:36  <Rubidium> still the French-tongued Canadians use Arrêt on their stop signs
21:45:25  <frosch123> are you sure that the french-tongued placed those signs?
21:45:29  <neofutur> grumph now I have a working citybuilding server its hard to decide to spend hours developping a nogo script ;(
21:45:44  <frosch123> maybe it were rather the english ones, being worried about french driving skills
21:45:50  <Mazur> No, __ln__, I reckon they have irregular conjugations, only the masses in the US never payed attention in school, so they've forgotten to use them.
21:46:16  <frosch123> neofutur: sell it for 200k
21:46:18  <neofutur> using nogo i ll be able to tell the server to rename a town, or to build a bank in  town ?
21:46:34  <frosch123> then you can pay someone to develop the nogo one
21:47:28  <Rubidium> frosch123: I'm quite sure, the English tongued wanted to get rid of the French tongued
21:47:35  <neofutur> eh luukland is selling more than just a server, and the opensource citybuilding script is not yet as good as the luukland one
21:47:58  <__ln__> Mazur: *payed*?
21:49:09  <frosch123> neofutur: if something is worth 200k, you must be able to gain 20k per year with it; or you really have to find someone stupid
21:49:39  <__ln__> Mazur: anyway, that's the prescriptive approach to the phenomenon, but the descriptive is to acknowledge that native speakers actually use the language like that.
21:51:19  <glx> Rubidium: wiktionary says the word "stop" is used since 1792 and is from english
21:52:03  <planetmaker> neofutur: you miss one big advantage of the script approach: with scripts you can offer much more variety in the gameing experience than you ever can do with a patch. Like this very specific "citybuilder"
21:52:13  <glx> so as always french canadians use the original word
21:52:26  <neofutur> frosch123: I agree, that should be sold 200eur, perhaps its possible to earn 100$ /month with it . . . not even sure :p
21:52:30  <planetmaker> the latter is severely limited. While the scripts allow you to keep your players at an edge. Or implement secretly some easter eggs
21:53:01  <planetmaker> well. you don't even need to do it secretly. Most people won't even notice when you do it in plain sight ;-)
21:54:18  <neofutur> but still this question, having a nogo script allows me to drive the server, changing anything ? like renaming a town or adding a bank in a town ?
21:55:01  <frosch123> i am not sure about naming a town
21:55:10  <frosch123> you can place signs or show addtional text in the town gui
21:55:17  <frosch123> but i don't think you can rename them
21:56:22  <frosch123> anyway, night
21:56:26  *** frosch123 [] has quit [Remote host closed the connection]
21:56:53  <neofutur> renaming the town to the owner / player name seems needed
21:57:11  <neofutur> having a system to "claim" a town too
21:57:41  <glx> there is a SetText() function
21:58:39  *** drac_boy [] has quit [Ping timeout: 480 seconds]
21:59:24  <Xaroth> neofutur: look at it this way, if the codebase changes in areas where your patch works, you'll have to fiddle for hours upon end to get it working again
21:59:29  <Xaroth> that is not likely with NoGo
22:00:17  <glx> and it's always possible to add functions in the API if really needed
22:00:23  <neofutur> Xaroth: agreed, its probably a better approach ;) but much more work too
22:01:02  <neofutur> grep SetText gamescripts_nogo/*/* find nothing ;(
22:01:03  <Xaroth> it's a matter of spending time now, and getting it right, or being forced to spend time later :)
22:01:04  *** sla_ro|master [slaco@] has quit [Quit: DANGER is OFFLINE DANGER]
22:01:05  <glx> (via patches of course but easier to get them accepted than a big hack)
22:01:38  *** drac_boy [] has joined #openttd
22:01:41  <glx>
22:01:43  *** Chris_Booth [] has quit [Quit: ChatZilla [Firefox 15.0/20120731150526]]
22:01:46  <drac_boy> sorry...some silly brownout in the area here :-s
22:02:26  <neofutur> ah I needed this class reference function list ;)
22:02:30  <neofutur> thanks glx
22:03:09  <glx> <-- that's the root :)
22:04:21  <FLHerne> Why does ottdcoop bundles have 'ogfx-trains', 'fork-ogfx-trains' and 'fork-fork-opengfx-trains'? :P
22:04:37  <FLHerne> And more to the point, which one should I care about? :P
22:04:39  <glx> because they like forks ?
22:09:44  *** Supercheese [~Password4@] has joined #openttd
22:14:25  <neofutur> strange there is getname and settext, if its the same property, shouldnt they be names getname and setname, or gettext and settext ?
22:18:05  <NGC3982>
22:18:17  <glx> neofutur: it's not the same property
22:20:53  <neofutur> ok, so theres indeed no SetName
22:25:40  <glx> it should be possible to add it
22:26:01  <neofutur> k
22:26:23  <glx> as renaming a town is allowed on the server
22:26:45  <neofutur> also, theres a function to protect a "owned" city ? so that only the owner can build near this town ?
22:26:55  *** Frankr [] has joined #openttd
22:27:40  <drac_boy> neofutur what do you consider 'near' btw?
22:28:43  <neofutur> the area wouls probably be relative to the town sie
22:28:48  <neofutur> size
22:29:18  <neofutur> like 10 tiles around city for a 500 city, 20 tiles for 2000 size . . .
22:29:48  <drac_boy> oh ok just wanted be sure you was not going to block out players from being able to lay lines through a city-close-to-mountain flatland section .. or even a powerplant few tiles outside a city
22:29:56  <drac_boy> and umm 20 is kinda a bit too wide, but 10 is ok
22:30:30  <neofutur> yup 20 could be too big, depending on the map size
22:30:52  <neofutur> this should probably be a setting of the game script
22:31:00  <Wolf01> 'night all
22:31:04  <drac_boy> bye Wolf01
22:31:06  *** Wolf01 [~wolf01@] has quit [Quit: Once again the world is quick to bury me.]
22:32:02  <drac_boy> well neofutur I don't have any complain but on the other hand I can't help, never really play ottd much (less now that theres almost no green dotted servers in list at times :-s )
22:32:30  <neofutur> why do you have no green dot ? yo use an old version ?
22:32:31  <drac_boy> I only can recall that the luk-something server (you did mention that too didn't you?) had some sort of script that blocked you from placing your hq in an already-taken town
22:32:38  <drac_boy> neofutur nope...chrill
22:33:05  <neofutur> more than just Hq on luukland, theeere is this type of protected area around owned towns
22:34:38  <drac_boy> oh luukland mm
22:37:25  * neofutur wonders what chrill means
22:38:22  <drac_boy> the chrill patch version of ottd
22:40:12  <neofutur> citybuilder patch or something else ?
22:40:49  <drac_boy> something else
22:40:59  <neofutur> k
22:41:12  * neofutur now have a working server for the citybuilder patch
22:41:35  <neofutur> gotta add goods requirement for bigger towns and its done
22:42:23  <drac_boy> :)
22:42:32  *** Progman [] has quit [Remote host closed the connection]
22:42:44  <neofutur> since it will probably take me months to make it a nogo script, i ll still try to have good citybuilding servers waiting for it
22:57:11  *** FLHerne [] has left #openttd []
23:12:25  <neofutur> any citybuilder player around to play with me and test the settings of my server ?
23:13:07  <drac_boy> not me sorry, don't know who else could be not asleep on their keyboard now :p
23:13:09  <drac_boy> heh
23:17:51  *** valhallasw [] has quit [Ping timeout: 480 seconds]
23:18:28  *** cyph3r [] has joined #openttd
23:22:35  *** nicfer [~Administr@] has joined #openttd
23:22:57  *** chester_ [] has joined #openttd
23:23:01  <nicfer> hello again
23:24:22  <drac_boy> hi nicfer
23:26:09  *** TGYoshi [~TGYoshi@] has quit [Quit: Hugs to all]
23:26:52  <nicfer> one question, can I create a map with the minimum town distance patch and load that into a patch pack?
23:27:23  <nicfer> that patch pack doesn't have MTD
23:36:46  *** tokai|mdlx [] has joined #openttd
23:42:43  *** tokai|noir [] has quit [Ping timeout: 480 seconds]
23:53:46  *** nicfer [~Administr@] has quit [Read error: Connection reset by peer]
23:53:48  *** drac_boy [] has left #openttd [I'm done being in this room!]
23:59:16  *** chester_ [] has quit [Quit: Leaving.]

Powered by YARRSTE version: svn-trunk