IRC logs for #openttd on OFTC at 2008-12-20
⏴ go to previous day
00:05:56 *** Yeggstry is now known as Yeggzzz
00:16:16 <CIA-1> OpenTTD: rubidium * r14698 /extra/strgen/ (. COPYING Makefile Makefile.bundle findversion.sh readme.txt): [strgen] -Add: a set of makefiles so the compile farm should be able to build (and release) strgen binaries.
00:16:35 <CIA-1> OpenTTD: frosch * r14699 /trunk/src/train_cmd.cpp: -Fix (r14483): Do not use v->dest_tile for trains heading to a station. Reaching that tile does not mean reaching the station.
00:17:19 <glx> btw adobe applications are not meant to be uninstalled :)
00:18:24 <Belugas> and me, i'm meant to go to sleep
00:18:36 <Belugas> mhh... first, i need to go home :S
01:11:25 *** Eddi|zuHause has joined #openttd
01:17:03 <Eddi|zuHause> you guys don't have an algorithm to convert a regexp (already tokenized and parsed) into a finite automaton?
01:17:41 <FauxFaux> I have a book detailing the algorithm.
01:17:52 <Eddi|zuHause> you have the books name?
01:18:17 <Eddi|zuHause> i mean, i could construct the algorithm myself easily, but i am kinda running out of time
01:18:40 <FauxFaux> I expect both the Appel (tiger, modern compiler implementation in *) and the dragon book, and, like, every other parser book ever will have it.
01:19:08 <FauxFaux> Aho, lam, sethi, ullman. Yeah, like I was going to remember that.
01:19:36 <Eddi|zuHause> are those words?
01:19:50 <FauxFaux> Surnames of authors. :p
01:21:29 <Eddi|zuHause> i probably really am faster in making up the algorithm myself than looking for it now...
01:22:19 <FauxFaux> Section 3.7, 14 pages, "from regular expressions to automata".
01:35:24 <CIA-1> OpenTTD: smatz * r14700 /trunk/src/ (oldloader.cpp openttd.cpp station_cmd.cpp): -Fix (r1): loading of very old savegames was broken (STNS chunk is stored before MAP in old savegame)
01:39:41 <Eddi|zuHause> i really wonder where SmatZ always gets those testcases from...
01:40:36 <benjamin_> like all good testcases.
01:40:51 <benjamin_> the best way to test something is to get one of your friends to try and break it...
01:41:58 <SmatZ> :-) sometimes I test compatibility with old savegames... the real-life usability is probably ~0
01:43:33 <Eddi|zuHause> those do not ever look like real life testcases :p
02:41:25 *** sigmund_ has joined #openttd
03:15:59 <Belugas> mmmh... can't remember what George suggested as the frequency of the population callback for houses
03:16:13 <Belugas> and my logs are not showing it either :(
03:18:30 *** sunk is now known as sunkan
03:25:16 <Eddi|zuHause> Belugas: didn't he say something like "i don't really care, as long as it is regularly?"
03:25:48 <Eddi|zuHause> but maybe that was a different discussion
03:26:31 <Belugas> could be Eddi|zuHause, not sure. that's the problem when it's not ported in flyspray :(
03:29:05 <Eddi|zuHause> [Mi Dez 10 2008] [18:58:29] <Belugas> George_, your population callback... when would you think it should be triggered?
03:29:06 <Eddi|zuHause> [Mi Dez 10 2008] [18:59:11] <George_> 1) when industry is build.
03:29:08 <Eddi|zuHause> [Mi Dez 10 2008] [18:59:46] <George_> 2) periodically (like every 256, 1024, or 65536 ticks)
03:29:09 <Eddi|zuHause> is that the one?
03:33:46 *** benjamin_ is now known as benjamingoodger
03:33:58 <benjamingoodger> just a little surprised
03:34:49 <Belugas> proof you know little 'bout your favorite dev ^_^
03:35:20 <benjamingoodger> you're saying sacro's gay?
03:36:41 <benjamingoodger> you said --- my favourite dev --- no --- er --- meh...
03:36:42 <Belugas> are you? DOes it matter if you'r gay of into cats?
03:36:47 <Eddi|zuHause> hm... this function for reachable states in finite automata that i have programmed once upon a time is total rubbish...
03:37:08 <Belugas> sorry to ear that Eddi|zuHause
03:37:10 <benjamingoodger> you go to sleep, joke-failer ¬.¬
03:37:24 <benjamingoodger> good night :)
03:37:25 <Eddi|zuHause> well, it's not that complicated of an algorithm
03:37:31 <Eddi|zuHause> and i never used that function...
03:37:41 <Eddi|zuHause> i just noticed... :p
03:37:55 <Eddi|zuHause> i didn't even check it into the repository :p
03:38:03 <benjamingoodger> if you are "into" cats, that does pose a problem
03:38:18 <benjamingoodger> as you may commit some sort of animal mistreatment crime
03:38:30 <benjamingoodger> and probably have other issues besides.
03:43:07 <Eddi|zuHause> sleep... not the worst idea right now
04:06:56 *** elmex_ is now known as elmex
04:20:54 *** benjamingoodger has quit IRC
04:54:54 *** sigmund has joined #openttd
04:57:23 *** ChanServ sets mode: +v tokai
07:28:30 *** rangaparmastan has quit IRC
07:46:12 *** Doorslammer has joined #openttd
08:03:09 *** frosch123 has joined #openttd
08:41:15 <George3> Is any anybody here with KVIrc. I'm trying to configure it and I cant find where to specify messages filter. I do not want messages like "YYY has joined/quited the channel" and "YYY is now known as YYY2" to appear. I can't find the switch and help is not useful either :(
08:56:05 <petern> often it's called conference mode
08:56:17 <petern> however, those are normal messages that most people do not ever want to filter
08:56:40 <petern> unless you have a specific YYY in mind, in which case just ignore YYY ;)
08:57:02 <blathijs> in irssie you can say /ignore -JOINS -PARTS -QUITS or something similar
08:58:17 <petern> KVirc is a hideous looking program...
09:05:00 <frosch123> George3: do you have processing industries, that need different amounts of input cargo to produce one piece of output cargo, i.e. two units grains for one unit food
09:06:10 <George3> Yes, a lot. For example, you need paper and dyes to produce goods at printing works (hope I understood you right)
09:06:53 <frosch123> no, i mean: is the production 1:1, e.g. 1 uint paper + 1 unit dyes = 1 unit goods
09:07:06 *** Alberth has joined #openttd
09:07:32 *** FloSoft has joined #openttd
09:07:59 <George3> AFAIR, I do not use 1+1=1, I mostly use things like 7+1=6
09:08:49 <George3> meaning there is % of waste
09:09:20 <frosch123> ah, but I guess you always process multiples of 7 and stockpile remaining cargo
09:09:35 <George3> And AFAIR I never use them in 1:1
09:11:06 <George3> frosch123: That is correct for the old industries (that have 6 leves of production). For more modern code I simply make a calculation (production callback with registers).
09:12:04 <George3> There (in calculations) I use constructions like " amount of cargo 3 processed = min (level * ( max (year - 1920; 5) + 20) * base3=8 / 100; v42) in R2"
09:13:22 <George3> and like this "Calculate amount of Energy in R7, add to IR2. 1 ton of coal = 7.5 MW/h, 1000 litres of oil = 10.3 MW/h"
09:14:08 <George3> frosch123: And what do you want to understand with you question?
09:14:39 <frosch123> and those do not work at all :p
09:15:32 <frosch123> they do not stockpile remaining cargo, so when he needs 8 units of coal for 1 unit goods, and 15 units coal are delivered, it produces 1 unit of goods and throws 7 coal away
09:16:18 <frosch123> however, due to cargo packets and gradual loading the amount of cargo transferred to the industry in one step is usally in the range 1 to 5.
09:16:28 <frosch123> So never above 8, so it never produces anything :(
09:16:57 <George3> No, I do not use multipliers, they are always 0 in my GRFs, and if they are not 0 somewhere, than it is a bug :)
09:16:59 <frosch123> But it seems, that your grfs are not affected, when you stockpile remaining cargo
09:17:17 <frosch123> They are only active when not using production callbacks :)
09:17:45 <frosch123> I guess that is also the reason noone noticed the bug, as noone uses that properties :)
09:24:10 <George3> frosch123: I also set props 12, 13 to 0, because their production is added to production of production callback
09:24:48 <frosch123> I meant 1C, 1D and 1E
09:27:07 <George3> I know, but looks like they also affect production callback result (I mean it was like that long time ago when I was testing it, so I made a desision to set it 0 to prevent problems)
09:34:23 *** Progman has joined #openttd
09:39:57 *** Vikthor has joined #openttd
09:47:33 *** George3 is now known as George
09:50:51 <George> wiki says it has no GUI :(
09:51:15 <Forked> it's what I love about it :)
09:55:40 <George> I do not. KVIrc looks not bad, but configuring it looks like a pain :( It is the second day I try to do it without succes. I can't find how to filter (not display) useless messages and how to apply a spellchecker.
09:57:43 <George> ChatZilla, that had a big problem because it was not running well without firefox running, was rather good in configuring it.
10:03:11 <Forked> there is always something :\
10:30:21 *** Purno__ has joined #openttd
10:34:57 *** david is now known as edeca
10:35:20 <edeca> Is it possible to make consumers use goods faster, i.e. make a plantation use up fertiliser more quickly?
10:35:27 <edeca> I'm producing fertiliser like there is no tomorrow :|
10:45:58 <edeca> I find that if I try to drop production, say by delivering less oil, I get bad ratings at the previous part of the industry chain and it all goes wrong ;)
10:46:09 <edeca> The new industries are a lot harder to manage (but it is much more fun)
10:46:57 <xand> whatever setting I choose for loading indicators to be displayed, they do not show at all for any games on this machine... is there a setting other than the one in configure patches -> interface?
10:47:08 <xand> in openttd.cfg I have loading_indicators = 2
10:48:26 <frosch123> there are also transparency options
10:48:36 <xand> bah, it's a transparency option... I wondered what that train icon in there did :)
10:50:26 *** Terkhen has joined #openttd
10:54:51 *** Dred_furst has joined #openttd
10:55:12 *** Dred_furst` has joined #openttd
11:03:21 <CIA-1> OpenTTD: rubidium * r14701 /trunk/Makefile.bundle.in: -Change: Debian's lintian complains about not using the best compression for compressing the man-page.
11:09:12 *** HerzogDeXtEr has joined #openttd
11:29:51 *** roboboy has joined #openttd
11:52:05 <CIA-1> OpenTTD: frosch * r14702 /trunk/src/vehicle_base.h: -Documentation: Add a note about the meaning and validity of Vehicle::dest_tile.
12:02:12 *** Swallow has joined #openttd
12:25:36 *** HerzogDeXtEr has joined #openttd
12:28:08 *** stillunknown has joined #openttd
13:01:57 *** XeryusTC is now known as Xeryus|bnc
13:03:49 *** Hirundo has joined #openttd
13:04:16 *** Hirundo is now known as Swallow
13:15:21 *** Terkhen has joined #openttd
13:40:34 *** entropy has joined #openttd
14:08:06 *** benjamingoodger has joined #openttd
14:48:03 *** HerzogDeXtEr has joined #openttd
15:08:04 *** Hirundo has joined #openttd
15:08:19 *** Hirundo is now known as Swallow
15:17:58 <George> glx: Hi, I'm trying to configure KVIrc and I can't do it well. I find where to switch off logging useless chanel information, but can't switch it off on the display. I made them white on white, but having lots of empty lines is not the best solution. I also found that I can specify their ratings, but I could not find a way to switch off displaying strings with 0 rating. Could you help me?
15:22:11 <George> glx: Are you here? Looks like you are the only one with KVIrc here :(
15:29:20 <xand> why does fast forward turn off if you switch away from openttd and back again?
15:29:58 <Rubidium> because someone thought that'd be useful?
15:30:17 <xand> or I should say, it *sometimes* does that
15:30:31 <CIA-1> OpenTTD: frosch * r14703 /trunk/src/yapf/yapf_road.cpp: -Fix (r8735): [YAPF] YapfRoadVehDistanceToTile() shall not abort searching at v->dest_tile, when it in fact tries to reach another tile.
15:33:46 <Rubidium> and what version of OTTD and the os?
15:34:35 <xand> oh, alt-tab turns it off, but using hte mouse doesn't... linux/fluxbox/openttd 0.6.3
15:36:01 <frosch123> does it also happen when you press alt before tab, and release alt after tab
15:37:08 <xand> I see tab key seems to be momentary fast forward
15:37:58 <petern> people actually press alt and tab together?
15:38:25 <xand> I hold alt and press tab, then let both go
15:38:27 <benjamingoodger> er, yes, it's a "switch to most recently focused window" shortcut on most WMs
15:38:53 <Rubidium> apparantly something tells that tab is pressed and alt isn't
15:39:11 <petern> benjamingoodger, but i meant as opposed to pressing (and holding) alt and then pressing tab
15:39:25 <benjamingoodger> what, tab-alt?
15:39:45 <xand> that's what I do... sometimes I must release alt more quickly than others :P
15:44:17 *** Eddi|zuHause2 has joined #openttd
15:57:50 <George> If I specify for industry tile that it has no animation, build the industry, save the game, and then change the GRF (add the animation to the tile), would a tile have anomation next time I load a game with updated GRF?
15:58:27 <frosch123> you have to start the animation somewhen
16:00:32 <frosch123> any of those animation callbacks which are triggered on cargo delivery or 256-ticks stuff.
16:00:46 <frosch123> just the one on industry-creation is not sufficient :)
16:02:50 <George> would it take much CPU?
16:03:33 *** Eddi|zuHause2 is now known as Eddi|zuHause
16:03:57 <George> Also, if specify prop 0F (animation) to FFFF (no animation) and then return FE 80 with callback 25, should animation run?
16:04:49 <George> frosch123: I mean would running callback 25 take much CPU (do you remember the table you send me?)
16:05:37 <Eddi|zuHause> hm... this more height level thing is a nice idea, but it does not seem ready yet...
16:05:48 <George> Callbacks during one month
16:05:54 <George> 0x real sprite 157 213,00
16:05:56 <George> 0x random trigger 2 108 791,00
16:06:00 <frosch123> prop 0F is 'fund cost' for me :)
16:06:02 *** |Jeroen| has joined #openttd
16:06:41 <George> frosch123: we are speaking about industry tiles. wiki sais "Animation information (0F)"
16:06:44 <Eddi|zuHause> 2 million seems better than the what? 66million? it was before
16:07:09 <frosch123> oh, hehe, and I was wondering why industries do not have animation :p
16:08:46 <George> frosch123: I'm tring to reduce usage of CBs 26 and 27.
16:08:53 <frosch123> just setting property 0F will not start animation in a running game. the callback is of course expensive compared to only using prop 0F
16:09:54 <frosch123> but it is not nearly as expensive as callback 27
16:12:04 <George> so, summery: I can set prop 0F to FFFF, and than later I can change it to FF 01. All, I need to run animation is CB 25.
16:12:20 <George> that means prop 0F is not stored in the save
16:12:42 <frosch123> at least I would be surprised :)
16:12:46 <George> Only animation state (running / stoped) is
16:13:41 <George> newly build industries would not need CB 25
16:16:13 <frosch123> hmm, I guess when you set property 0F to FFFF, you cannot start animation
16:18:17 <George> If it stored in save - yes. But if not - a new GRF would provide a new value FF 01 for prop 0F of the tile and new value for prop 11(CB 25 triggers), so animation should (as I understand it) start
16:19:19 <George> The Idea is - to remove all unused things from GRFs now and add them later when they are used
16:19:28 <frosch123> property 0F is not stored in the savegame. changing it by changing newgrf will not start animation for existing industries. setting it to FFFF will more or less stop the animation
16:19:38 <George> this would reduce CPU usage now
16:19:39 <frosch123> maybe the last part is incorrect behaviour
16:20:54 <frosch123> ohoh, yes, the last part looks very incorrect, as it would continue with the animation of original industries (which will break certain stuff) :p
16:21:09 <George> frosch123: I understood that it would not, but as you said CB 25 can save it, all I need is too redefine prop 11 (hope it is not stored too)
16:21:34 <frosch123> no property is stored
16:22:06 <George> frosch123: And what do you suggest? Currently, there is animation specified, but in fact no animation is provided.
16:22:31 <frosch123> what do you mean by 'animation specified'
16:24:34 <George> prop 0F is set to FF 01 and 0E to 55
16:25:05 <frosch123> had those property already the value, when the industries were constructed?
16:25:35 <frosch123> then animation is running
16:25:52 <frosch123> btw. that is also a performance impact, when you start animation, which is not used
16:26:08 <George> Yes, that is what I mean 'animation specified'
16:26:22 <George> I understand, that is why I want to remove it
16:27:05 <frosch123> so, actually you do not want to start animation, but you want to stop animation when switching the grf
16:28:04 <George> In the early days I used animation as data storage instead of industry registers. Than industry registers were provided, I started to use them, but animation remained.
16:28:37 <George> I want to stop them now and restart it in the future, when new animation is provided
16:29:51 *** Hirundo has joined #openttd
16:29:51 <George> so, 1) whould changing prop 0F to FF FF stop animation ( I suppose it should) in the current state
16:30:29 <George> 2) would cnaging prop 0F to FF 01 in the future would allow to start it with CB 25
16:30:37 <frosch123> no, you can only stop it using cb 25
16:31:02 <frosch123> and currently setting prop 0F to FFFF in a running game, seems to even break it :s
16:31:47 *** Hirundo is now known as Swallow
16:32:12 <George> Do you mean that setting 0F to FF 01 and providing CB 25 is the best solution?
16:32:43 <George> So, now I should use CB 25 to stop it and later to start it
16:33:25 <frosch123> at least that should work
16:35:21 <George> And that would save some CPU usage, right, because animation is stoped
16:37:21 <George> Thank you, I'll try to do it. Could you create a ottd executable for me (for win32) that provides CB usage table like you send me?
16:37:44 <frosch123> for win, no I can't
16:39:59 <George> A pity :( Well, when I shall finish the work may I send you GRFs and you provide me a new table with testing results at the same conditions?
16:40:29 <George> BTW, for prop 11 is 04 the least CPU using value?
16:44:11 <frosch123> 04 means all industries every 3.5 days
16:44:25 <frosch123> don't know how often industries are delivered in the average game
16:45:38 <George> it may happen that never, and than animation would stay, that is not good, because we need to stop it
17:09:30 *** Singaporekid has joined #openttd
17:09:57 <CIA-1> OpenTTD: frosch * r14704 /trunk/src/core/smallvec_type.hpp: -Codechange: Add set capabilities to SmallVector.
17:18:03 <Eddi|zuHause> i could not see it
17:19:14 <George> smile what sybol do you see before smile?
17:19:30 <mrfrenzy> ] on white background
17:19:57 <George> Do you see anything in the bracets ( )
17:20:22 <mrfrenzy> same thing and a whitespace
17:20:46 <mrfrenzy> I suggest you get yourself another client and join a testchannel
17:21:37 <CIA-1> OpenTTD: frosch * r14705 /trunk/src/economy.cpp:
17:21:37 <CIA-1> OpenTTD: -Fix [FS#2460]: First transfer the whole load of a vehicle chain to industries before triggering any processing.
17:21:37 <CIA-1> OpenTTD: This reduces callback usage and resolves critical rounding errors when using input-cargo-multipliers instead of production callbacks.
17:21:37 <George> I alrady have. And I see something like there
17:22:23 <Eddi|zuHause> then you need to enable utf-8 support
17:22:37 <George> frosch123: FS2460 - and what will happen if industry stops accepting cargo during this process?
17:22:39 <mrfrenzy> or you should just stop sending utf-chars with this client
17:22:42 <Eddi|zuHause> my font does not have that symbol
17:22:52 <mrfrenzy> since 30% of all clients can't see them anyways
17:23:03 <valhallasw> well, those clients pretty much suck :P
17:23:17 <mrfrenzy> still, 100% of clients can see this: ]
17:23:24 <George> glx: I know you see it right, you have the same client.
17:23:37 <mrfrenzy> so just configure your client to send a variant of that
17:23:40 <glx> George: emotes are client specific
17:23:47 <frosch123> George: it behaved like that before :)
17:23:55 <Eddi|zuHause> mrfrenzy: the topic once said that this channel prefers UTF-8
17:24:00 <George> mrfrenzy: I do not ;) , use :)
17:24:10 <valhallasw> UFT-8 over iso-8859
17:24:14 <mrfrenzy> George: I sent you a bracket, not a smiley
17:24:28 <mrfrenzy> Eddi|zuHause: maybe it changed it's mind, cause I can't see it ;)
17:24:42 <mrfrenzy> since when does channels have preferences btw
17:24:45 <valhallasw> more importantly, the ] is shared in all iso-8859-* charsets /and/ utf-8
17:25:04 <George> frosch123: what do you mean?
17:25:06 <valhallasw> mrfrenzy: well, look closer
17:25:13 <mrfrenzy> yes, in order to have preferences it must have some kind of consciousness
17:25:32 <mrfrenzy> valhallasw: well that is way too far to the end to fit in my terminal ;)
17:25:39 <mrfrenzy> I see it now typing /topic
17:26:00 <Eddi|zuHause> mrfrenzy: "one person is intelligent, large groups of persons are stupid"?
17:26:35 <George> frosch123: with gradual loading/unloading it should leave the station partly unloaded if undustry do not accept cargo any more, isn't it?
17:26:44 <glx> mrfrenzy: we want to be able to write things like ばか
17:27:02 <Eddi|zuHause> i need glasses for those...
17:27:22 <mrfrenzy> Irssi: Added oftc/openttd/UTF-8 to conversion database
17:27:26 <George> mrfrenzy: And like this " !", but how do you see them?
17:27:29 <mrfrenzy> there ya go, from now on I type utf-8 in here
17:27:40 <frosch123> George: that did not work properly before, does not work now, and does - to my knowledge - not work in TTDP either
17:27:48 <mrfrenzy> but my font does not have the chars glx typed
17:28:16 <mrfrenzy> I can write stuff like '!"#"#%#/&%)/~=()=@$.{[]}\|< though
17:28:21 <mrfrenzy> and you should probably see it
17:28:44 <Sacro> hmm, I don't think I had utf8 in screen
17:28:47 <Eddi|zuHause> George: we do not see those signs correctly unless you switch to UTF-8
17:29:13 <mrfrenzy> I can see them correctly, since my irssi can read both iso and utf
17:29:27 <George> I supposed I switched it on :S
17:29:33 <Eddi|zuHause> mrfrenzy: there are a lot of different "iso" charsets
17:29:42 <mrfrenzy> I am too backwards to change my terminals to utf though, so I will only see characters that exist in iso8859-15 that irssi can convert for me
17:29:45 <Eddi|zuHause> and only ever one can be automatically detected
17:29:48 <Sacro> Eddi|zuHause: are any useful?
17:30:20 <Eddi|zuHause> Sacro: they are useful, as long as you do not need to switch between locales
17:30:57 <Eddi|zuHause> a russian (eastern european) iso charset is not useful in a german (western european) locale
17:31:20 <valhallasw> the iso-* charsets have the same lower 128 characters and locale-specific upper 128 characters
17:32:09 <Eddi|zuHause> glx's russian i can read, George's not
17:32:10 <mrfrenzy> I can still read you fine George
17:32:42 <George> glx: I can't read your text
17:32:59 <George> it is "я не говорю по русски"
17:33:19 <George> Looks like UTF in Win1251 for me
17:33:20 <valhallasw> George: get your client to interpret the text as UTF-8...
17:34:05 <glx> George: on top of the channel there's a button with a "japanese" char
17:34:13 <mrfrenzy> you also need to get your terminal to use utf8
17:34:15 <glx> click on it and select utf8
17:34:17 <mrfrenzy> to see his strange chars
17:34:56 <Eddi|zuHause> the most common utf-8 characters in german are äöüß
17:35:07 <George> glx: I have it in our private cat but not here
17:35:29 <glx> press the little arrow above user list
17:35:36 <Eddi|zuHause> with their upper case equivalents (ÄÖÜ), only ß has no upper case equivalent
17:35:48 <George> Thank you. Please one more test
17:36:12 <Eddi|zuHause> € <- if that is an euro sign for you, it was successful
17:36:32 *** DephNet[Paul] has joined #openttd
17:36:51 <George> Thank you, it works now. BTW, you do, because you made a russian text :lol:
17:36:58 <valhallasw> Eddi|zuHause: heh. well, there is some discussion about /that/
17:37:15 <valhallasw> although SS /is/ considered the upper-case equivalent for most
17:38:23 <tokai> using UTF on IRC is evil :)
17:38:30 <Eddi|zuHause> valhallasw: no, it's an alternate spelling
17:39:10 <Eddi|zuHause> kind of an "emergency fallback", when no äöüß is available, you may replace those by ae, oe, ue, ss
17:39:50 <valhallasw> but without capital-ß it's not really a /fallback/, is it? ;)
17:40:12 <Eddi|zuHause> those are generally understood and percieved correct, but possibly bad style
17:40:12 <valhallasw> but it is included in the unicode character set
17:40:38 <Eddi|zuHause> there were attempts to introduce an upper case ß, but they did not spread
17:41:26 <Eddi|zuHause> ß generally does not appear at the beginning of a word, so there is usually not a reason to have it uppercase, except when you write in all caps
17:41:58 <petern> it's only a problem for those who don't want to change
17:42:07 <petern> and those whose OSes probably don't support it :p
17:42:10 <valhallasw> there is no reason /at all/ *not* to use UTF-8
17:42:21 <Eddi|zuHause> tokai: it's definitely better than having to deal with different iso locales
17:42:26 <valhallasw> it's not like utf-8 is new or something like that...
17:42:27 <SpComb> valhallasw: apart from the fact that other people don't use it
17:42:48 <SpComb> it's very annoying on e.g. IRCNet where people have channels with non-ascii chars in the name
17:42:50 <tokai> even better is to avoid any non-ascii chars:)
17:43:31 <SpComb> so you join #böö, but it's a different #böö from where all the ISO 8859-1 -using people are
17:43:33 <valhallasw> SpComb: yeah, well, bad luck for them...
17:44:02 <Eddi|zuHause> those are easily avoidable, if the servers would enforce "identifier" rules for channel names
17:44:23 <SpComb> the RFC doesn't mandate channel names to be ascii-only :)
17:44:24 <petern> tokai, that's almost like benjamingoodger's statement that translating the game into non-english is a waste of time...
17:44:26 <SpComb> it does for nicknames, though
17:44:33 <Eddi|zuHause> i.e. only contain a-z,A-Z,_,0-9
17:44:39 <Eddi|zuHause> and may not start with a number
17:44:53 <valhallasw> please join channel 0 ;)
17:44:55 <George> tokai: that page sais UTF-8 is the best :)
17:45:24 <Eddi|zuHause> valhallasw: i tried that once, but my client actually joined channel
17:45:40 <tokai> petern: translating a game is a waste of time and usually degrades the quality (the source language can be anything different than english too)
17:45:41 * Rubidium votes for banning tokai whenever he uses a character that is not supported in the same way by *all* of the encodings mentioned in the IRC#Character_encoding webpage
17:46:02 <tokai> I don't mind getting banned :)
17:46:08 <valhallasw> "The Internet Engineering Task Force (IETF) requires all Internet protocols to identify the encoding used for character data, and the supported character encodings must include UTF-8.[2]"
17:46:28 <Eddi|zuHause> i don't remember tokai even talking during the past year :p
17:46:45 <tokai> I always talk when you're not here usually. :)
17:47:50 <George> back on topic means FS2460?
17:48:05 <tokai> petern: f.ex. there are quite some japanese games I wished to be able to get with original language rather than with degraded english dub :) As example. Original language always should be a choice.. like OpenTTD does it, f.ex. that's fine:)
17:48:39 <tokai> there are some style guide problems in the english strings though :)
17:50:13 <George> I think it would be nice that vehicle would stop unloading if industry does not accept cargo any more. How a fix for FS2460 works in that case (industry stops accepting cargo during unloading process)
17:50:58 <Eddi|zuHause> translations do not necessarily need to be a degration from the original language
17:51:38 <Eddi|zuHause> especially if the translated language has a rich set of vocabulary, and the translator is a little more creative than just tansferring words
17:52:12 <frosch123> George: the acceptance callback is called just like before, just the production callback is not called in between
17:52:14 <glx> the translator should know the game too
17:52:53 <Eddi|zuHause> translation usually is a problem when the game/movie/whatever depends on style and other language features
17:53:16 <George> frosch123: What? If I have a very long train and very low unloading speed (like 1 unit) it will stop production callback?
17:53:50 <Eddi|zuHause> what is easily lost are plays on words that sound similar, because those usually do not sound similar in the target language
17:54:26 <Eddi|zuHause> cultural references are occasionally lost, if the cultures are different, too
17:54:42 <Eddi|zuHause> but with increasing globalisation, those are of increasingly smaller issues
17:54:44 <Rubidium> i.e. basically all jokes
17:54:54 <frosch123> No, when gradual loading is enabled it transfers small amounts of cargo to the industry, which accepts it. After that is done for all wagons, the production callback is called
17:55:26 <frosch123> Eddi|zuHause: see the "fahre/fliege/gehe" discussion on forums
17:56:06 <frosch123> problems arise when certain languages need more context to decide cases than other languages do
17:56:19 <Eddi|zuHause> frosch123: i know, i took part in it... but that is not directly a translation problem, it is a problem with the grammar engine not being complex enough
17:56:37 <George> frosch123: and if a smaller train would upload cargo first production would happen for it first, and than for the first very long train?
17:57:03 *** Brianetta has joined #openttd
17:57:14 <Eddi|zuHause> frosch123: i once wanted to investigate in how far the case-system can help there, but i do not have that kind of time
17:57:43 <Rubidium> but you have the time to talk about it for hours...
17:58:15 <frosch123> George: I do not see the relation, but I also do not understand the question
17:58:30 <Eddi|zuHause> making 3 posts in the forum and 2 sentences here means "hours" for you?
17:59:34 <Rubidium> s/it/it and other things/?
17:59:50 <frosch123> fs#1460 causes the on-delivery-production-callback to be called once per train, instead of for every gradual loading step of every cargopacket of every wagon
17:59:59 <frosch123> the acceptance callback is called like before
18:00:15 <Eddi|zuHause> Rubidium: it's called "taking a break" ;)
18:00:40 <frosch123> so as long as the acceptance callback does not modify persistent storage registers, and expects the production callback to be called immediatelly, nothing changes
18:01:04 <frosch123> but I guess, when you would have done that, it would not have worked before either. plus you would encounter certain ttdp/ottd differences
18:01:53 <George> frosch123: Imagine. 2 trains - very long and very short arives for unloading. A very long comes first. A short train comes second, but unloads first. The question is - would the industry process cargo from the short train before long train?
18:02:19 <frosch123> the first train is processed first
18:02:59 <Rubidium> the time a train arrives is the time it starts (un)loading first
18:03:11 <Rubidium> so it is impossible for a train to come second but unload first
18:03:14 <frosch123> and the whole cargo is transferred as much as accepted before it is completely unloaded
18:03:38 <George> so, the cargo from the second train would wait untiil cargo from the first train is processed?
18:03:52 <frosch123> i.e. when you interrupt unloading cargo (e.g. by sending a train to a depot) the industry already got the cargo, that is still on the vehicles
18:04:05 <George> Rubidium: Do you man 2 trains can't unload at the same time?
18:04:18 <frosch123> luckily the cargo knows that it is already delivered, so you cannot deliver it twice :)
18:04:40 <Rubidium> George: that heavily depends on how you define unloading
18:04:42 <Eddi|zuHause> George: trains can unload in the same tick, but still one would be treated first
18:05:25 <Rubidium> as unloading in my book is a while (cargo in vehicle) { real_unload "load amount"; wait a while }
18:05:27 <George> unloading is a PROCESS (happens during a period of time) when cargo is transfered between the vehicle and station
18:05:54 <frosch123> George: just do a testgame. you can see, the the whole cargo amount is stockpiled, when the train arrives. after that the load of the cargo gradually decreases, but the industry already got everything
18:06:20 <George> frosch123: Is it intended that way?
18:06:42 <frosch123> at least TTDP and OTTD agree on that :)
18:07:02 <frosch123> but it does not make a real difference
18:08:27 <frosch123> George: you can compare it with the income animation, you get the whole income when the train arrives, and not multiple small amounts of income
18:09:40 <George> I see, I'm thinking about how bad it is :S
18:10:00 <Eddi|zuHause> it's basically the same as without gradual loadding
18:10:29 <Eddi|zuHause> so gradual loading only affects how long the train is waiting and how that is displayed
18:10:56 <George> The qusetion is - if the industry has limited stockpile, it would be nice to unload only the part of the cargo that is accepts and leave the rest in the vehicle
18:11:06 <frosch123> but remember, the acceptance callback is called in multiple steps. but they are all called at the same time on arrival, and not when you visually see the vehicle load decrease
18:11:49 <Eddi|zuHause> afair the "overdelivered" cargo is left at the station, and may be reloaded by the vehicle
18:12:15 <frosch123> when you do not add any unloading flags it remains in the vehicle
18:13:13 <frosch123> the "station acceptance" is delayed by up to 3.5 days
18:13:37 <frosch123> so when the station still accepts the cargo, it is delivered to valhalla
18:13:37 <George> frosch123: As I understand you, if a station accepts cargo when vehicle starts to unload the first unit of cargo, it will accept all the cargo in the train, right?
18:14:13 <Eddi|zuHause> so what's valhalla doing with all the coal?
18:15:07 <frosch123> at least you are paied for the whole
18:15:31 <frosch123> under which conditions it remains in the train is a bit a mystery also to myself :)
18:18:31 <George> frosch123: :) then what should happen when the second train arrives to the station, which does not accept cargo any more - leaves it with cargo, while the first one is unloading
18:19:28 <frosch123> well the first train will stop unloading, when the station stops accepting
18:20:19 <frosch123> I guess it also stops unloading, though its cargo was delivered :p
18:20:21 <George> But you said it already unloaded all the cargo to the industry and got paid?
18:20:58 <Eddi|zuHause> maybe there is some additional logic required ;)
18:21:04 <George> so, what should than happen with that cargo, that is already unloaded, but is still in train?
18:21:24 <frosch123> yup, when the train arrives, all cargo is converted to imaginary cargo :)
18:21:48 <frosch123> you can then move, load and unload it everywhere as you like
18:22:01 <Eddi|zuHause> "you already signed for recieving, i can now drive off with what i still have in the wagon" :p
18:22:34 <frosch123> hmm, I wonder what happens when you delete the stations :s
18:23:00 <George> so I can deliver it to somewhere else and get paid twice? Looks like a bug :)
18:23:47 <frosch123> no, the imaginary cargo knows that it is imaginary
18:23:52 <Eddi|zuHause> maybe, during gradual unloading, you should assign target industries to each cargo packets
18:24:08 <Eddi|zuHause> and then you unload only those which have a target industry set
18:24:25 <svippy> Uh, I forgot to ask how to fix this lib issue.
18:24:29 <George> but does it take place in the train and has weight?
18:24:32 <svippy> ./openttd: error while loading shared libraries: libicui18n.so.38: cannot open shared object file: No such file or directory
18:24:50 <Eddi|zuHause> svippy: well, install libicu
18:24:59 <glx> or compile openttd yourself
18:25:11 <svippy> I am going for glx's solution.
18:25:26 <Eddi|zuHause> which os is that, anyway?
18:25:42 <svippy> Slackware, specifically.
18:26:16 <Eddi|zuHause> if you don't speak persian or hebrew, you probably won't miss it :p
18:26:32 <svippy> :p Well, I can't - unfortunately.
18:28:18 <frosch123> ah, I see, you can transfer the cargo everywhere you want and it will be accepted at stations as usual, but you won't be paid twice, and it won't be transferred to industries twice
18:31:00 <George> frosch123: Then it looks like a bug that there is something in my train, that uses train space, reduces train speed, and can't make me any profit. It should be throwed away ASAP :)
18:31:32 <frosch123> you can disable that bug by disabling gradual loading :p
18:32:53 <George> :) may be we should find a better way for it?
18:33:04 <Rubidium> what's a bigger bug is that you can unload in a very short time and then let the train leave and it magically loses it's contents
18:33:45 <frosch123> yeah, the user can at least decice which bug he prefers :)
18:34:06 <frosch123> sounds like democracy
18:38:12 <CIA-1> OpenTTD: translators * r14706 /trunk/src/lang/ (finnish.txt indonesian.txt latvian.txt):
18:38:12 <CIA-1> OpenTTD: -Update: WebTranslator2 update to 2008-12-20 18:37:49
18:38:12 <CIA-1> OpenTTD: finnish - 17 changed by UltimateSephiroth (17)
18:38:12 <CIA-1> OpenTTD: indonesian - 81 changed by fanioz (81)
18:38:12 <CIA-1> OpenTTD: latvian - 137 fixed, 123 changed by peerer (260)
18:41:47 *** HerzogDeXtEr has joined #openttd
18:44:56 <Eddi|zuHause> hm... for once a translation message that is not truncated ;)
18:48:19 <frosch123> learn some languages and you can change that
18:50:16 <zircu> finally got to irc, i have to thank the people providing this network at first
18:52:04 <zircu> an then a thanks to everyone making this all happen
18:52:26 <zircu> i'm addictated to the game :)
19:03:50 <yorick> frosch123: I'd like to translate openttd to dutch :p
19:05:54 <frosch123> yorick: you can do frisian if you like
19:06:09 <Eddi|zuHause> i could try dutch, but you would not be fond of the result :p
19:06:40 <yorick> it's quite horrible now...
19:06:47 <Eddi|zuHause> at best it would sound like a parody of rudi carell :p
19:09:06 <yorick> and too much spaces on places they shouldn't be, capitalizing all kinds of stuff that shouldn't, plural errors with "player"
19:10:47 <benjamingoodger> petern: no member of this channel is ever, _ever_ going to let that go, are they?
19:16:52 *** yorick_ has joined #openttd
19:19:57 *** yorick__ has joined #openttd
19:20:10 *** yorick is now known as Guest456
19:20:10 *** yorick__ is now known as yorick
19:24:30 <Eddi|zuHause> well, that's your own fault :p
19:25:15 <Eddi|zuHause> yorick: well, nobody prevents you from fixing it
19:32:37 <jow> I wonder whether there's a possibility to run openttd with 5160x1024 resolution in xinerama. When I use a manual resolution by dragging the window borders, the image is capped on the right display and about one third remains black within the window borders. Are there some maximum constants in the source which I could tweak?
19:34:53 <Progman> you really dont want to scroll on 1024^2 maps, want you? *g*
19:37:52 <jow> petern: allright, thanks. Will compile from trunk now
19:39:11 <yorick> Eddi: it wouldn't go into trunk
19:41:32 <frosch123> you can also read the news page if you like
19:43:32 <Eddi|zuHause> jow: use a nightly, big resoltuions are possible there
19:44:02 <mrfrenzy> Is it possibly to zoom a lot? I don't care if it gets blocky
19:46:08 <jow> Eddi|zuHause, petern: thanks guys, works perfect :) The long openttd night can start, need to fill that 2048^2 map :P
19:52:55 *** Eddi|zuHause has joined #openttd
19:56:52 <yorick> frosch123: I prefr just to submit patches
19:57:28 <yorick> and they "didn't need another translator"
19:58:06 <frosch123> Dutch: fully translated
19:58:38 <yorick> yeah, but the translation is ugly
19:58:49 <yorick> things like "Players List"
20:08:15 <Eddi|zuHause> luckily, "Spieler" is the same singular and plural ;)
20:15:12 *** sigmund_ has joined #openttd
20:27:03 <CIA-1> OpenTTD: frosch * r14707 /trunk/src/ (ai/trolly/pathfinder.cpp aystar.h npf.cpp): -Cleanup: Replace an 'int' by 'Trackdir'.
20:28:29 <petern> you replace with, not by
20:30:39 <frosch123> hmm, yes. as usual two words in english, one in my language. at least I did not use 'at'
20:37:04 *** HerzogDeXtEr1 has joined #openttd
20:44:59 <CIA-1> OpenTTD: rubidium * r14708 /trunk/ (5 files in 3 dirs): -Change: update a few urls and fix some small errors/missing things in a few documents.
21:02:55 *** Brianetta has joined #openttd
21:46:50 *** mortal` has joined #openttd
21:47:10 *** Mortal is now known as Guest470
21:47:10 *** mortal` is now known as mortal
21:48:08 *** Eddi|zuHause has joined #openttd
21:48:44 *** Nite_Owl has joined #openttd
21:51:41 <George> frosch123: I small question. If callback 25 fails (real sprite return), what should happen? Same as FE 80?
22:10:30 <zircu> how active is this channel?
22:12:11 <zircu> and how should i present my issues here
22:14:21 <Eddi|zuHause> mental transmission
22:19:10 <Eddi|zuHause> bääh... argh... ihh... i got caught by a .jpg screenshot again... they are so horrible...
22:20:23 <Nite_Owl> zircu: Try explaining your problem - if anyone can help they will answer
22:24:25 <zircu> Nite_Owl: yeah i dont have a big problem, just would like to hang out and look at what i am doing is wrong
22:24:55 <zircu> i am always looking for the best solution at hand with mhy tracks
22:25:52 <Nite_Owl> Have you gone through the screenshot thread at the TT Forums? I learned more from that then from anyplace else.
22:27:05 <zircu> i'm workign on r14657 on the mac version and may have a suggestion but need to ensure i'm doing things correctly
22:27:44 <zircu> yeah i've been through that
22:28:20 <fjb> There is not one correct way. Find your style and have fun.
22:28:26 <zircu> i'm not exactly who is really ivolved
22:29:01 <zircu> fjb yeah but i think we can help things out
22:29:28 <Eddi|zuHause> zircu: how about you tell us what you _REALLY_ want to ask
22:29:31 <zircu> i guess it would help if i could provid a screen shot
22:31:08 <zircu> well i have a 4 way interesetion (over bridges), the outer briges are slower than than the inner brigde
22:32:08 <Eddi|zuHause> it's really nice of you to have that.
22:33:00 <zircu> what i was thining I have X and Y bridges, if a train only go X spead it will tak that one, if Y spead it will take that one
22:35:40 <zircu> the logic on deciding on what trak to do.. i have my passanger go right though my my fieght cargo go on the alternate path
22:37:43 <Eddi|zuHause> the pathfinder must give path penalties for all trains equally, otherwise the values cannot be cached
22:38:03 *** stillunknown has joined #openttd
22:38:23 <zircu> would it help if i gave you a sceen shot?
22:38:27 <Eddi|zuHause> there were attempts for "routing restictions" like speed limits and stuff
22:39:05 <Eddi|zuHause> you should probably just use waypoints
22:39:51 <zircu> i started to use that but not all tains use the way point that i add (goods)
22:40:39 <Nite_Owl> You have to add the waypoint to a trains orders
22:40:39 <zircu> but somtimes the passger trains use that same path'
22:41:15 <michi_cc> Eddi|zuHause: yapf doesn't cache the path for the first segments so it is possible to have limited train based penalties
22:41:49 <yorick> he talked! || it doesn't?!
22:41:51 <Eddi|zuHause> michi_cc: yes, but i wasn't going to explain that in great detail :p
22:43:33 <michi_cc> simply introducing a configurable bridge speed penalty might be enough, right now the penalty is fixed and not very high
22:44:05 <Eddi|zuHause> well, you need a penalty for >= and <= max speed
22:44:47 <Rubidium> that'd suck when bridge speed == max speed of train
22:45:34 <Eddi|zuHause> you just add both :p
23:13:44 <petern> i hope everyone's cut & paste works :D
continue to next day ⏵