Log for #openttdcoop.devzone on 25th July 2011:
Times are UTC Toggle Colours
07:57:56  *** ODM has joined #openttdcoop.devzone
08:12:55  <planetmaker> hm, AI tournament still running ;-)
09:34:15  *** Zuu has joined #openttdcoop.devzone
09:34:28  *** AlexWorX has joined #openttdcoop.devzone
09:34:45  <AlexWorX> himnew
09:35:54  <AlexWorX> want to code some functionality like building houses or fog on map and visibility area and adding house info
10:14:16  <Terkhen> hi AlexWorX
10:14:40  <Terkhen> some of that stuff is not possible with NewGRFs, but for houses you should be able to use NML
10:37:56  <planetmaker> hm... 12 hours AI tournament lasting and still running :-)
10:39:02  <planetmaker> maybe running each game for 50 game years was... lengthy ;-)
10:39:27  <planetmaker> but Zuu wanted to see his refit / replace rules taken into account :-P
10:41:32  <Zuu> hehe, so now I have caused you some inconvience :-)
10:42:12  <Zuu> Do you have any figures on how far it has progressed or is it impossible to see while it is working?
10:43:55  <planetmaker> Zuu, not at all. It's the DevZone's CPU doing the work
10:44:03  <planetmaker> and no, no progress report
10:44:59  <planetmaker> it just shows that extrapolation from 1 year runtime to 50 year runtime does not work linearly
10:45:00  <Zuu> so we can just wait and see :-)
10:45:02  <planetmaker> yup
10:45:24  <planetmaker> 1 year test length, other things the same took about 5 to 6 minutes
10:45:40  <planetmaker> @calc 5*50 / 60
10:45:40  <Webster> planetmaker: 4.16666666667
10:45:46  <Zuu> You didn't include more AIs?
10:46:01  <planetmaker> I included 17 AIs
10:46:03  <AlexWorX> what have to be done for implement building houses?
10:46:11  <planetmaker> but also in the 1 year test. I didn't publish that
10:46:32  <planetmaker> it was just done to get a guestimate on runtime. A bad one it seems ;-)
10:46:36  <Zuu> Does all AIs start with 1 day apart?
10:47:15  <planetmaker> All are started at day one via the at-game-start script every openttd knows in scripts/...
10:47:32  <Zuu> But still, many AIs get into "management hell" at the end when they got so many vehicles to manage, lots of money and should keep building on their empire :-)
10:47:41  <planetmaker> yep :-)
10:47:54  <planetmaker> Well. Each single game has no more than 5 AIs
10:48:11  <planetmaker> But it's running many many games to find the ranking
10:48:51  <planetmaker> I don't quite recall the amount of games. But something with o(2^n)
10:49:04  <planetmaker> or 2(^ln n)
10:49:08  <planetmaker> dunno ;-)
10:49:48  <planetmaker> but what showed yesterday clearly is that 1 year is waaaay too short
10:50:12  <planetmaker> after the 12 years (with the 5 AIs which I then later posted) it showed the competitiveness which I usually saw in my manual tests, too
10:51:05  <planetmaker> But... testing this extrem length also tells me, where to set the time boundaries for a potential somewhat regular test of this kind
10:52:05  <planetmaker> AlexWorX, well... depends, what do you want?
10:52:21  <planetmaker> You "just" have new house graphics and provide alternative houses?
10:52:25  <AlexWorX> coding
10:52:42  <planetmaker> nah, that's a tool I mean: what's your goal?
10:52:56  <planetmaker> "building houses" is ambigeous
10:53:08  <planetmaker> As player in OpenTTD you can never build a specific house.
10:53:21  <planetmaker> It's always the game which decides which house is being built by a town (and if at all)
10:53:38  <planetmaker> Changing that behaviour (e.g. for the SE) would require you to start hacking at OpenTTD source
10:54:10  <planetmaker> if it's "just" new / different / additional houses... then a newgrf approach is right
10:55:04  <AlexWorX> yes thats what i mean. wanna enable for players buying ground if major allows and building house if major allows and to make it possible to be ellected as major..
10:56:17  <AlexWorX> to make it more realistic also in the way building streets. first the player should try buying ground or should pay the city to build a street..
10:56:34  <planetmaker> uhm... you can build roads since the dawn of TTD
10:57:22  <planetmaker> And in the town panel you can always - as player - select "fund new buildings". Which means you give money and the town expands in return
10:57:46  <planetmaker> allowing to pick a specific house... yes, will need coding
10:57:54  <Ammler> maybe you should first learn openttd, before you think about hacking it ;-)
10:57:58  <planetmaker> I'd like to see that function in the scenario editor in particular
10:58:07  <planetmaker> I'm not sure about ingame
10:58:25  <planetmaker> well... ingame. If newgrf developer tools are enabled
10:58:51  <planetmaker> but ok, that's me. It's you who wants to write a patch :-)
10:59:00  <Terkhen> AlexWorX:
10:59:01  <Webster> Title: Transport Tycoon Forums View topic - My patch(es) (place house manually) (at
10:59:13  <planetmaker> ho, that exists?
10:59:21  <Terkhen> I never checked it myself
10:59:30  <Terkhen> you should pay more attention to the dev forum :P
10:59:56  <planetmaker> :-P
11:00:27  <planetmaker> Yes, I briefly saw it. But never tested it myself either. And forgot
11:00:42  <planetmaker> When I saw it my thought was "hopefully he finishes it" ;-)
11:05:02  <Terkhen> looks more like the "release and forget" approach
11:05:28  <Terkhen> oh, or maybe he is talking about that multi climate patch
11:22:44  <Brot6> TransRapid Track Set - Feature #2890 (New): option to have elevated station also with railtypes (foobar) @
14:00:24  <planetmaker> @calc 985/60
14:00:24  <Webster> planetmaker: 16.4166666667
14:00:39  <planetmaker> hm... 16 hours to run the AI tournament...
14:00:54  <Ammler> still running?
14:01:01  <planetmaker> nope. Finished.
14:01:03  <planetmaker> I had it timed
14:01:24  <planetmaker> interesting result...
14:25:57  <planetmaker> @calc log 4
14:26:03  <planetmaker> @calc log(4)
14:26:03  <Webster> planetmaker: 1.38629436112
14:26:06  <planetmaker> @calc log(10)
14:26:06  <Webster> planetmaker: 2.30258509299
14:26:25  <planetmaker> @calc log(5)
14:26:25  <Webster> planetmaker: 1.60943791243
14:26:42  <planetmaker> @calc 2**1.6
14:26:42  <Webster> planetmaker: 3.03143313302
15:12:59  <Hirundo> time to start looking at stupid endian bugs again...
15:20:05  <Rubidium> endian bugs? Where? ;)
15:20:17  <Rubidium> thought I booted them out of grfcodec ;)
15:20:18  <planetmaker> nml. grfID
15:25:15  <Rubidium> what's wrong with grfid?
15:25:46  <planetmaker> inconsistently used little and big endian in different things
15:26:28  <planetmaker> but only in NML. IIRC resulting from the string <-> number conversion or so.
15:26:29  <Rubidium> grfid returns it the logical way, doesn't it?
15:27:32  <Rubidium> i.e. the same way as it's printed in
15:27:55  <Hirundo> IIRC it's loaded as LE in openttd, and BSWAP32'ed for the GUI
15:29:09  <Rubidium> probably yes
15:29:47  *** AlexWorX has quit IRC
15:30:00  <Hirundo> NML just does the wrong thing, which is made worse because grfids can be strings, numbers, or both, depending on where you use it
15:30:44  <Rubidium> oh, it's only NML ;)
15:30:46  <Rubidium> pfew ;)
15:38:39  *** andythenorth has joined #openttdcoop.devzone
15:38:42  <andythenorth> hola
15:38:44  <andythenorth> anything?
15:43:29  <Hirundo> planetmaker: I wonder, how we should handle comparisons with vehicle var 0x25 (grfid)
15:44:49  <Hirundo> If we silently b-swap all numeric grfids, it's bound to go haywire there
15:51:01  <planetmaker> yes...
15:51:16  <planetmaker> what about a compare_grfid(arg1, arg2)?
15:51:26  <planetmaker> which accepts strings and numbers but does the right thing?
15:52:48  <planetmaker> or rather grfids_are_equal(id1, id2 [, mask])
15:56:14  *** andythenorth has quit IRC
15:56:24  <Brot6> NewGRF Meta Language - Bug #2877: engine_override encodes GRFIDs backwards (Hirundo) @
15:57:10  <Brot6> NewGRF Meta Language - Bug #2877: engine_override encodes GRFIDs backwards (Hirundo) @
15:58:31  <Brot6> NewGRF Meta Language - Bug #2877: engine_override encodes GRFIDs backwards (Hirundo) @
16:12:55  *** frosch123 has joined #openttdcoop.devzone
16:18:02  <Hirundo> hmm.... I don't know, endian bugs confuse the mind
16:21:48  *** AlexWorX has joined #openttdcoop.devzone
16:23:36  <planetmaker> hm... can it be enforced that grfIDs are only exposed as 4 byte literal strings?
16:25:01  <Hirundo> perhaps that'd be best, yes
16:25:33  *** andythenorth has joined #openttdcoop.devzone
16:25:39  <planetmaker> I use it that way anyway ;-)
16:25:45  <Hirundo> It implies removing the option to 'mask' grfids in certain places, but since versioning is no longer done via the grfid, I doubt that it's a problem
16:25:52  <planetmaker> like ""
16:26:25  <planetmaker> hm... it would still make sense to mask out, say all base set grfs. Or all ECS grfs
16:26:29  <planetmaker> hm...
16:26:59  <Hirundo> ECS is about the only use case, yes
16:27:14  * Hirundo looks up ECS grfids
16:28:37  <Hirundo> It's 4D656F9x, masking that would be useful
16:30:45  <Hirundo> those masks are currently strings-only
16:31:33  <Hirundo> since no-one requested changing that, I guess it's fine as-is
17:01:56  *** andythenorth has quit IRC
17:10:43  *** AlexWorX has quit IRC
17:18:54  *** AlexWorX has joined #openttdcoop.devzone
17:20:53  <Brot6> cets: update from r88 to r90 done (482 warnings) -
17:22:11  <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), transrapidtrackset (r15), 2cctrainset (r750), ailib-list (r32), opensfx (r97), ttdviewer (r34), worldairlinersset (r672), heqs (r605), openmsx (r97), basecosts (r25), nutracks (r202), nml (r1527), 32bpp-extra (r40), manindu (r7), newgrf_makefile (r305),
17:22:11  <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), fish (r668), ogfx-landscape (r70), 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
17:22:13  <Brot6> (r41), ailib-string (r29), airportsplus (r107), comic-houses (r71)
17:22:51  <Brot6> narvs: compile of r37 still failed (#2789) -
17:23:28  <Brot6> ogfx-industries: compile of r121 still failed (#2792) -
17:44:52  *** andythenorth has joined #openttdcoop.devzone
18:03:32  *** bodis has joined #openttdcoop.devzone
18:34:12  *** andythenorth has quit IRC
18:45:27  <Brot6> NewGRF Meta Language - Revision 1528:0b919fad902a: Change #2877: Make engine_override(), deactiav... (Hirundo) @
18:45:27  <Brot6> NewGRF Meta Language - Revision 1529:7b7278d05b15: Doc #2877: Usage of str2number for certain var... (Hirundo) @
18:45:27  <Brot6> NewGRF Meta Language - Revision 1530:70a6708442ee: Change: Rename object variables to be consiste... (Hirundo) @
18:45:29  <Brot6> NewGRF Meta Language - Revision 1531:3b0043245009: Change #2877: Do the string->number conversion... (Hirundo) @
18:45:33  <Brot6> NewGRF Meta Language - Revision 1532:b47a704c9da7: Cleanup: Remove obsolete comment. (Hirundo) @
18:46:21  <Brot6> NewGRF Meta Language - Bug #2877 (Closed): engine_override encodes GRFIDs backwards (Hirundo) @
19:05:12  *** andythenorth has joined #openttdcoop.devzone
19:05:31  <Brot6> clientpatches: update from r22687 to r22688 done (6 warnings) -
19:10:29  <Brot6> openttd-vehiclevars: update from r22687 to r22688 done -
19:15:20  <Brot6> serverpatches: update from r22687 to r22688 done (10 warnings) -
19:17:32  <Brot6> 32bpp-ez-patches: compile of r22688 still failed (#2446) -
19:21:55  <Brot6> openttd-vehiclevars-demo: update from r22687 to r22688 done -
19:23:04  * frosch123 still cannot believe ttdp 2.5 does not support temporary grf registers
19:56:36  <Brot6> NewGRF Meta Language - Bug #2877: engine_override encodes GRFIDs backwards (foobar) @
20:05:58  <andythenorth> Terkhen: does your nml work?
20:50:38  *** andythenorth has quit IRC
21:14:26  *** AlexWorX has quit IRC
21:23:18  *** frosch123 has quit IRC
21:25:30  *** bodis has quit IRC
21:55:04  *** ODM has quit IRC
23:02:58  *** avdg has quit IRC
23:03:01  *** DJNekkid has quit IRC
23:05:51  <Brot6> Unable to connect to execution expired
23:05:55  *** avdg has joined #openttdcoop.devzone
23:06:26  *** DJNekkid has joined #openttdcoop.devzone

Powered by YARRSTE version: svn-trunk