Log for #openttdcoop.devzone on 27th July 2011:
Times are UTC Toggle Colours
00:13:59  *** Zuu has quit IRC
00:42:06  *** Lakie has quit IRC
06:21:26  *** Zuu has joined #openttdcoop.devzone
06:33:33  *** ODM has joined #openttdcoop.devzone
07:11:33  *** Zuu has quit IRC
08:59:08  <Brot6> NewGRF Meta Language - Revision 1543:a98ba3ea3375: Codechange: Do not specify the amount and type... (Hirundo) @
09:31:45  *** FooBar has joined #openttdcoop.devzone
10:46:43  *** ODM has quit IRC
11:02:32  <Brot6> TransRapid Track Set - Feature #2895: option to have custom tunnels with railtypes (foobar) @
11:33:28  *** SmatZ has quit IRC
11:34:40  *** SmatZ has joined #openttdcoop.devzone
11:35:29  <Brot6> Unable to connect to execution expired
12:11:27  <Brot6> TransRapid Track Set - Feature #2895 (Closed): option to have custom tunnels with railtypes (foobar) @
12:11:27  <Brot6> TransRapid Track Set - Revision 26:28a207df5d39: Feature: custom tunnels with railtypes (closes #... (foobar) @
12:11:27  <Brot6> TransRapid Track Set - Revision 27:67ea6b942eaa: Update: documentation for release (foobar) @
12:11:28  <Brot6> TransRapid Track Set - Revision 28:92fd76eb2aa1: Release: 1.0.0 (foobar) @
12:11:31  <Brot6> TransRapid Track Set - Feature #2895 (Closed): option to have custom tunnels with railtypes (foobar) @
12:13:46  <Brot6> transrapidtrackset: update from 0.9.0-beta to 1.0.0 done -
12:19:31  *** ODM has joined #openttdcoop.devzone
13:26:13  <Brot6> Total Town Replacement Set - Bug #1979 (Closed): Spanish translation uses incorrect language-id (foobar) @
13:33:07  *** frosch123 has joined #openttdcoop.devzone
13:34:27  <planetmaker> hm... ttrs could use a transition into NML, too ;-)
13:42:53  <FooBar> well, I just had a look at the code and came to the conclusion that I ain't doing that :P
13:43:58  <FooBar> But yes, that would make it much easier to add a passenger production parameter, as I have no clue where to start now.
13:44:33  <Hirundo> NML could use proper house support first....
13:46:12  <planetmaker> I fear you have a point
13:49:46  <planetmaker> bah, this aluke is such a fool. And that's an euphemism.
13:50:18  <Hirundo> aluke who?
13:50:30  <planetmaker> forums, Transrapid trackset thread
13:50:44  <planetmaker> he's like that everywhere
13:51:16  <Hirundo> Suggestions on a sane way to do multitile houses are welcome
13:51:28  <FooBar> heh, I barely could resist an RTFM post :P
13:52:08  <planetmaker> they need to have subsequent IDs, right, Hirundo ?
13:52:59  <planetmaker> hm... Hirundo what about somewhat copying the layout feature for industries and objects?
13:53:17  <planetmaker> even if they don't really have on grf level
13:54:18  <planetmaker> it would give the potential to expand it sometime to irregular houses :-P
13:54:23  <Hirundo> layouts are intended for irregular stuff, houses only have (1|2)*(1|2)
13:54:33  <Hirundo> even objects don't have layouts
13:55:14  <planetmaker> hm, ok. How do objects handle it?
13:55:23  <planetmaker> They can be rectangles, just like houses
13:55:48  <planetmaker> only that the size definition in nfo is quite different
13:58:03  <planetmaker> FooBar, but seeing your transrapid "bridge": it indeed looks MUCH better than default bridges.
13:58:30  <Hirundo> Multitile houses must consist of multiple items, defined in order
13:58:32  <planetmaker> though people will have a VERY hard time to deboard the train in case of an emergency
13:58:53  <planetmaker> Hirundo, but that can be abstracted away, can't it?
13:59:24  <Hirundo> yes sortof, but it needs abstraction :P
13:59:27  <FooBar> From a user point-of-view, I'd go for something similar to objects. Have NML check that only values 1 and 2 are allowed for x and y.
13:59:41  <planetmaker> yeah, it does. Not exactly 100% trivial that kind of abstraction
13:59:43  <planetmaker> I know
13:59:52  <planetmaker> FooBar, yes, my idea
14:00:24  <Hirundo> Then there's stuff that can be set per tile, stuff that cannot be set per tile, stuff that partially can be set per tile, ...
14:00:26  <planetmaker> and distribute the population, acceptance equally (or assign to north(?) tile) - whatever is asked for by the specs.
14:01:26  <planetmaker> And maybe with a pseudo-property allow to define acceptance of tiles individually, maybe like in an array [[COAL, 0], [BEER, 0], [FOOD, 0], [GOLD, 0]] for my villa ,for example :-P
14:01:48  <Hirundo> :P
14:02:03  <FooBar> planetmaker: yes, transrapid tracks without the transrapid bridge are very ugly. As for emergencies: they have the same problems at the real test facility. Unfortunately.
14:02:10  <Hirundo> I'd suggest to increase the acceptance, you'll be very hungry otherwise ;)
14:02:18  <planetmaker> :-)
14:02:32  <planetmaker> I just ate. So it's time-dependent
14:02:45  <planetmaker> I still had nice quiche from yesterday :-)
14:03:25  <Hirundo> There are callbacks for that sort of stuff :P
14:03:56  <Hirundo> still to be implemented though, for now you have to choose between accepting always or never
14:04:39  <planetmaker> hm, ok. With that choice I'll always accept 8/8 of BEER, GOLD and FOOD. COAL is 4/8 ;-)
14:04:51  <planetmaker> people shall not say I'm greedy :-P
14:12:07  <frosch123> Hirundo: houses have a single item in action3 (no cargos/purchase list etc). so maybe use the nml-action3 to split between house tiles
14:12:23  <frosch123> though i do not know how to do action 0
14:12:43  <frosch123> some properties only need setting once / same for all tiles, some can be set individually
14:14:17  <frosch123> maybe just allow different identifiers in the "graphics" thingie, and accept multiple value for some properties
14:16:14  <frosch123> hmm, that collides with the new callback style
14:17:05  <frosch123> how does the new callback style work with purchase-list/not-purchase-list ?
14:19:24  <planetmaker> frosch123, you can just define 'purchase' as cargo and then do there what you want to be done there
14:19:54  <planetmaker> you'll need to do the usual stuff then subsequently - minus what the purchase list needs not consider
14:20:01  <planetmaker> but you could still do it the other way round
14:20:27  <frosch123> <- i thought i have to handle callbacks at the top now
14:21:08  <planetmaker> no, it's just another 'cargo' which you assign a switch sequence to
14:21:17  <planetmaker> And you can do the same for other callbacks
14:22:44  <planetmaker> but you can still just define the callback in the normal graphics switch sequence as well.
14:22:52  <planetmaker> Whatever is more useful for your application
14:23:01  <planetmaker> And I hope one cannot break it that way by mixing ;-9
14:26:46  <Ammler> hmm, btw. what does l in nml mean?
14:27:44  <FooBar> language?
14:27:49  <FooBar> what did I win? :P
14:27:50  <Ammler> NewGRF Meta Language - language ?
14:28:02  <Ammler> sounds silly :-)
14:28:16  <planetmaker> Ammler, it's NML, not NML language
14:28:16  <frosch123> but common to a lot of abbreviations
14:28:34  <Ammler> planetmaker: nml-language.html :-)
14:29:10  <FooBar> I think you must see that as a book with the name NML, and a chapter therein with the name "language"
14:31:39  <Ammler> ah, I see
14:32:35  <planetmaker> ah :-) And yes, I follow FooBar's reasonong ;-)
14:34:14  <Ammler> well, just that everything is in that chapter :-P
14:48:31  <Brot6> DictatorAI - Revision 116:f72474ef64ef: - Working on slopes & terraforming (krinn) @
14:50:57  <planetmaker> hehe. The AI test seems to have sparked quite a bit activity in that domain
14:51:14  * planetmaker is pleased :-)
15:22:56  <Terkhen> :P
15:26:04  <planetmaker> well, I can with all honesty say: that was always my main intention to run AI tests ;-)
15:54:07  * FooBar ponders someone should devise a turing test for AIs...
15:59:39  <planetmaker> well. I'm just pondering (again) scores.
15:59:59  <planetmaker> I just came to the conclusion that there should not be a penalty for an AI, if it runs a loss with a vehicle type
16:00:10  <planetmaker> Running a loss IMHO must not be worse than not using it. Or?
16:00:37  <planetmaker> after all one might consider a transfer service where the last leg is a bus service or so... that'd never be profitable but still be clever
16:01:26  <planetmaker> my current categories are vehicle Income/#vehicles; company value / years; cargo delivered; load; town rating
16:01:34  <planetmaker> where applicable per vehicle type separately
16:01:48  <planetmaker> and I've not yet a concept for 'reuses infrastructure'
16:02:01  <planetmaker> Maybe indeed via orders... but that has bear traps in it
16:05:45  <FooBar> Depends a bit what the goal is. If it is "beat the human in the high score table" then a loss on certain vehicles is a bad thing. But then you might as well just take the company rating as benchmark...
16:06:48  <planetmaker> Well, yes. But scoring 0 is then bad enough
16:07:06  <planetmaker> as 0 is the worst you can get there anyway, whether you utilize vehicles or not
16:07:13  <FooBar> true
16:07:41  <planetmaker> yes... maybe I could add the company score somewhere to this AI score... :-)
16:07:49  <FooBar> apart from bankrupt, was there another thing that could go negative?
16:08:00  <planetmaker> yes. income per vehicle type
16:08:14  <planetmaker> and NoCAB did score a whopping -360 there for trains ;-)
16:08:21  <planetmaker> and still managed to come out with +70 or so
16:08:27  <FooBar> also apart from that I meant :P
16:08:34  <planetmaker> ah. loan
16:09:51  <FooBar> that seems correct to be negative
16:10:29  <planetmaker> yep, sure
16:10:59  <FooBar> could also be turned around, but doesn't really matter.
16:11:21  <planetmaker> hm... instead of 0 ... 100 for the vehicle incomes I could scale it to -25 ... 75 or so :-)
16:14:02  <FooBar> Also, I think I devised the turing test myself: start with a specific scenario and have say 10 human players play the scenario for say 10 game years. AIs do the same. No competition. After that screenshots are collected of each game (preferrably a small map so everything can fit in one screenshot). Then a quiz-type thing is set up where humans have to decide whether each screenshot is from a...
16:14:04  <FooBar> ...human or AI played game.
16:15:45  <FooBar> AIs that get a lot of "made by human" votes are obviously good at what they do. Quiz contestants will see their score at the end of the quiz (percentage correct) to make it interesting to partake in the quiz.
16:17:36  <FooBar> Could very well end up as a forum game. The data gathered could then be used to award each AI with additional points on top of the tournament score
16:19:24  <FooBar> Would need some scripting to create a website to host the quiz though...
16:21:55  <planetmaker> :-)
16:29:21  *** Lakie has joined #openttdcoop.devzone
16:48:04  *** Lakie has quit IRC
17:08:33  *** andythenorth has joined #openttdcoop.devzone
17:10:04  <Brot6> nml: update from r1532 to r1543 done -
17:18:36  <Brot6> transrapidtrackset: update from r16 to r28 done -
17:20:33  <Brot6> fish: update from r668 to r669 done (4 warnings) -
17:21:00  <Brot6> Following repos didn't need a nightlies update: ogfx-trains (r245), narvs (r37), bros (r52), ogfx-industries (r121), firs (r2183), sub-landscape (r72), opengfx (r681), ailib-tile (r16), 2cctrainset (r750), cets (r90), ailib-list (r32), opensfx (r97), ttdviewer (r34), worldairlinersset (r672), heqs (r605), openmsx (r97), basecosts (r25), nutracks (r202), nml (r1543), 32bpp-extra (r40), manindu (r7), newgrf_makefile (r305),
17:21:00  <Brot6> ailib-direction (r17), ailib-common (r21), snowlinemod (r49), dutchtramset (r84), ai-admiralai (r75), swisstowns (r22), metrotrackset (r56), spanishtowns (r10), frenchtowns (r6), grfpack (r279), ogfx-rv (r107), ogfx-landscape (r71), ttrs (r36), source-test (r2), ogfx-trees (r51), swedishrails (r203), grfcodec (r832), ai-aroai (r39), german-townnames (r34), smts (r19), chips (r143), belarusiantowns (r8), indonesiantowns (r41),
17:21:04  <Brot6> ailib-string (r29), airportsplus (r107), comic-houses (r71)
17:22:25  <Brot6> narvs: compile of r37 still failed (#2789) -
17:23:03  <Brot6> ogfx-industries: compile of r121 still failed (#2792) -
17:25:46  <Brot6> sub-landscape: compile of r72 still failed (#2892) -
17:37:11  <Brot6> ogfx-rv: rebuild of r107 done (Diffsize: 4614) (DiffDiffsize: 536) -
17:42:56  <Brot6> airportsplus: rebuild of r107 done (Diffsize: 16180) (DiffDiffsize: 283) -
17:42:57  <Brot6> Following repos rebuilds successful without any difference to earlier nightlies builds: ogfx-trains (Diffsize: 1606), firs (Diffsize: 95969), cets (482 warnings) (Diffsize: 3517), manindu (Diffsize: 1), newgrf_makefile, dutchtramset (Diffsize: 2584), swisstowns, spanishtowns (Diffsize: 1), frenchtowns, ogfx-landscape (Diffsize: 86), source-test (6 warnings), swedishrails (Diffsize: 67), german-townnames, belarusiantowns (Diffsize: 30),
17:42:57  <Brot6> indonesiantowns (1 warnings) (Diffsize: 1)
18:11:28  <Brot6> Central European Train Set - Revision 91:503ced4ba84b: Added 10/8 and 6/8 legth templates (oberhuemer) @
18:20:10  <Brot6> Central European Train Set - Revision 92:b491de104dbf: fixed too short diagonal on newest templates (oberhuemer) @
18:23:29  <Brot6> Central European Train Set - Support #2784: graphics template (oberhuemer) @
18:27:49  <Ammler> planetmaker: make fails because of the gimp errors, I didn't notice that
18:28:00  <Ammler> is it possible to force make to return 0?
18:58:53  *** Webster has joined #openttdcoop.devzone
18:59:14  *** Brot6 has joined #openttdcoop.devzone
18:59:22  *** Ammler- has joined #openttdcoop.devzone
18:59:23  *** SmatZ has joined #openttdcoop.devzone
18:59:53  *** Terkhen has joined #openttdcoop.devzone
19:00:24  <Brot6> Following repos didn't need a testing update: clientpatches (r22689), openttd-vehiclevars (r22689), serverpatches (r22689), 32bpp-ez-patches (ERROR r22689)
19:00:52  *** Hirundo has joined #openttdcoop.devzone
19:00:53  *** tneo has joined #openttdcoop.devzone
19:01:23  *** V453000 has joined #openttdcoop.devzone
19:01:43  *** XeryusTC has joined #openttdcoop.devzone
19:02:23  *** avdg has joined #openttdcoop.devzone
19:02:54  *** DJNekkid has joined #openttdcoop.devzone
19:03:54  *** planetmaker has joined #openttdcoop.devzone
19:05:10  <planetmaker> 20:36 planetmaker: Hirundo: could you take a look at OpenGFX+ Industries: industry_definitions:499 ?
19:05:10  <planetmaker> 20:36 planetmaker: nmlc: "src/industry_definition.pnml", line 499: layouts must be an array of layout names
19:05:10  <planetmaker> 20:37 planetmaker: And I'm blind. I don't see what's wrong
19:05:11  <planetmaker> 20:37 planetmaker: or maybe Terkhen knows :-)
19:05:13  <planetmaker> 20:38 planetmaker: <-- this diff is additionally required to reach that error message. It fixes a few other syntax changes
19:13:21  <Hirundo> I don't see it either (yet), working on it
19:13:51  <planetmaker> I looked at it already three times. Compared it to other (working) code pieces...
19:13:57  <planetmaker> and it probably is something stupid ;-)
19:15:16  <Hirundo> It is
19:15:23  <Hirundo> there is also a spritelayout with the same name
19:16:13  <Hirundo> Which, in this cases, clashe in a somewhat obscure manner resulting in an error
19:16:33  * Hirundo moves an 's' in that sentence
19:17:15  <Hirundo> Fixing the whole identifier mess is on my todo list. ..  somehwere...
19:18:16  <planetmaker> you're my hero, Hirundo :-) Thanks
19:20:00  * Hirundo meanwhile battles 'feature' r1545
19:21:36  <Brot6> OpenGFX+ Industries - Revision 122:72dd82802613: Fix #2792: NML syntax changes and duplicate iden... (planetmaker) @
19:21:36  <Brot6> OpenGFX+ Industries - Bug #2792 (Closed): DevZone compile failed (planetmaker) @
19:22:35  <Hirundo> Would anything break, if I removed the 'refer to spritesets before their definition'-(mis)feature?
19:25:34  <planetmaker> I *think* not. But who knows?
19:25:42  <Hirundo> Only one way to find out :)
19:25:58  <planetmaker> I always tried to have it defined the usualy way, thus sprites first, then reference them
19:26:00  <planetmaker> But...
19:26:42  <Hirundo> It's not really necessary to remove it, but it'd make coding easier
19:27:31  <planetmaker> I don't mind either way much really
19:28:51  <planetmaker> if it makes things easier... go for it
19:34:16  <Hirundo> I'll think about it... while watching big bang theory :)
19:34:50  <planetmaker> hm, are there new episodes?
19:35:15  <planetmaker> <-- no :-(
19:35:21  <Webster> Title: Watch The Big Bang Theory Online Season 4 (at
19:44:28  *** Ammler- is now known as Ammler
20:09:38  <Brot6> Central European Train Set - Support #2784: graphics template (Eddi) @
20:10:42  <Terkhen> planetmaker: IIRC it was related to changes in nml, but I did not look further into it after being told about that compile error
20:15:35  <planetmaker> yeah, it's already solved :-)
20:17:18  <Brot6> FIRS Industry Replacement Set - Revision 2184:f39df170a1f3: Codechange: Move oil rig to separate ... (planetmaker) @
20:17:42  *** FooBar has quit IRC
20:35:07  <Brot6> Central European Train Set - Revision 93:28ab00988d09: wrong sprite anchor point, fixed in 8lu an... (Eddi) @
20:40:11  <Brot6> FIRS Industry Replacement Set - Revision 2185:dcde560267ae: Fix: The refinery is actually an ore ... (planetmaker) @
20:47:25  <Terkhen> ok :)
20:48:46  <Brot6> FIRS Industry Replacement Set - Revision 2186:6b534da52af1: Codechange: Move oil refinery to sepa... (planetmaker) @
20:50:17  <Brot6> FIRS Industry Replacement Set - Revision 2187:7f2c1ce79b82: Codechange: Move tiles of oil rig int... (planetmaker) @
20:50:58  *** bodis has joined #openttdcoop.devzone
20:56:44  *** ODM has quit IRC
20:59:54  <Brot6> FIRS Industry Replacement Set - Revision 2188:9cafd7eca26b: Codechange: Move sawmill to separate ... (planetmaker) @
21:00:36  <Brot6> FIRS Industry Replacement Set - Revision 2189:39b1b5cbb735: Fix (r2188): Add file (planetmaker) @
21:01:47  <andythenorth> nearly done?
21:01:54  <planetmaker> nearly
21:03:53  <andythenorth> :)
21:05:04  <planetmaker> hm, how is 'steelmill' called? Same or different name?
21:06:37  *** andythenorth has quit IRC
21:11:35  <Brot6> FIRS Industry Replacement Set - Revision 2190:d474bcc55bcf: Codechange: Move steelmill to separat... (planetmaker) @
21:18:27  <Brot6> FIRS Industry Replacement Set - Revision 2191:51fe55a68fcb: Codechange: The rest of 'god_object' ... (planetmaker) @
21:20:04  <planetmaker> ^ andy can be happy now ;-)
21:20:38  <Hirundo> no more god :(
21:21:24  <planetmaker> :-) Nietsche won
21:21:50  <planetmaker> but honestly: would you accept oil wells as your god? ;-)
21:22:09  <Hirundo> Many companies and even countries do :)
21:22:23  <planetmaker> that for sure. But I asked you ;-)
21:24:16  <Brot6> DictatorAI - Revision 117:9068f0a6afef: - more work on slopes (krinn) @
21:24:17  <Hirundo> No, I would not
21:24:30  <planetmaker> well - I call this the personal newgrf deed of the day for me - a good point to say 'good night' ;-)
21:52:45  *** Zuu has joined #openttdcoop.devzone
22:00:21  *** bodis has quit IRC
22:11:42  *** frosch123 has quit IRC
23:21:14  *** Zuu has quit IRC
23:53:25  *** Sylf has joined #openttdcoop.devzone

Powered by YARRSTE version: svn-trunk