Log for #openttdcoop.devzone on 28th December 2009:
Times are UTC Toggle Colours
00:05:56  *** Zuu has quit IRC
01:22:48  *** KenjiE20|LT has joined #openttdcoop.devzone
01:27:41  *** welshdragon is now known as Guest734
01:27:43  *** welshdragon has joined #openttdcoop.devzone
01:32:46  *** KenjiE20 has quit IRC
01:32:58  *** Guest734 has quit IRC
03:46:10  *** KenjiE20|LT has quit IRC
03:51:47  *** welshdragon has quit IRC
05:05:52  *** PeterT has quit IRC
06:47:54  *** ODM has joined #openttdcoop.devzone
08:14:35  *** Zuu has joined #openttdcoop.devzone
09:06:23  *** andythenorth has joined #openttdcoop.devzone
09:08:24  <andythenorth> morning
09:16:22  <Ammler> Zuu: you are damn fast in thread reading :-P
09:16:51  <Ammler> heya andythenorth, did you try to push again?
09:22:47  <andythenorth> yes
09:22:54  <andythenorth> seemed to work
09:24:18  <Zuu> Ammler: Thats my expertise. :-p
09:25:33  <Ammler> andythenorth: it won't annouce here yet, need to install rbot first again
09:50:20  <Hirundo> <Ammler>	[23:35:35] is there any part in is, which could be used alone? <- of the 14 patches, only 4 or 5 have to be applied at once
09:51:11  <Hirundo> The first 8 patches are basically codechanges, not features, and the last part that adds the fees is optional too
10:01:00  <Ammler> ok, that sounds like branches is a good idea :-)
10:01:14  <Ammler> !s/branches/branches or patch queue/
10:04:23  <Ammler> OpenSFX is indeed the wrong name for the sound replacement :-)
10:04:36  <Ammler> it should be renamed to FreeSFX ;-)
10:11:11  *** andythenorth has quit IRC
10:12:29  *** andythenorth has joined #openttdcoop.devzone
10:19:58  <Ammler> Hirundo: what about using both?
10:20:04  <Ammler> the base IS as branch and the rest as patches?
10:20:48  <Ammler> hmm
10:20:50  <Hirundo> That'd only complicated matters as far as I can tell
10:21:08  <Ammler> shouldn't it be possible to automatically make the branches from your patches?
10:21:36  <Ammler> or opposite :-)
10:22:09  <Hirundo> I was thinking about the opposite too :)
10:22:38  <Ammler> did you already try to merge trunk with the patch queue?
10:23:09  <Hirundo> I already merged with the patch queue, not yet with the main IS repo
10:35:18  <planetmaker> Ammler: I've been re-thinking your proposal to actually (additionally) rent a root server
10:35:41  <planetmaker> What about doing that and renting (parts) of this server then maybe out to interested people running an OpenTTD server?
10:36:06  <planetmaker> I don't feel like financing two servers, but doing this approach we might again end up with one...
10:36:24  <planetmaker> Given our use of avignon / ap+ we could kinda supply ready configured servers.
10:46:25  <andythenorth> planetmaker, Ammler: I was going to ask you about how you are funding the server...
10:46:30  <andythenorth> I am happy to contribute
10:47:01  <planetmaker> we currently pay it privately for our joy at our expense, 50% each of us
10:49:04  <planetmaker> never thought about donations so far. Maybe we should :-)
10:49:09  <andythenorth> how much is it in eur?
10:50:01  <planetmaker> the current machine is 30€ a month
10:50:56  <andythenorth> do you want to pay €10 each?
10:53:18  <planetmaker> Well... I wouldn't mind :-) But as I asked Ammler: we have currently issues with this VPS we use. It frequently runs out of memory which makes it hang and crash. That's very bad IMO
10:53:42  <planetmaker> So... dunno whether we shouldn't rather rent a root server... Where is Ammler when we need him :-)
10:54:46  <planetmaker> So we should discuss that, too. From my POV such a contribution would make that decision easier.
10:55:19  <andythenorth> I'm happy to contribute €10-15 / month
10:57:00  <Rubidium> just run the processes that seem to leak within supervise (of daemontools) and ulimit the processes
10:58:50  <planetmaker> thanks Rubidium - I'll look at that; didn't know of it so far :-)
11:01:33  <planetmaker> andythenorth: thanks, very much appreciated. Ammler and I'll discuss that :-)
11:08:19  *** welshdragon has joined #openttdcoop.devzone
11:09:20  <Ammler> well, we should also check, how it works, if we move the web to oasi's
11:11:12  *** andythenorth has quit IRC
11:11:12  *** planetmaker has quit IRC
11:11:12  *** XeryusTC has quit IRC
11:11:13  *** SmatZ has quit IRC
11:11:13  *** tneo has quit IRC
11:12:19  *** tneo has joined #openttdcoop.devzone
11:12:19  *** SmatZ has joined #openttdcoop.devzone
11:12:19  *** XeryusTC has joined #openttdcoop.devzone
11:12:19  *** planetmaker has joined #openttdcoop.devzone
11:12:19  *** andythenorth has joined #openttdcoop.devzone
11:14:44  <planetmaker> btw, andythenorth you had a question yesterday, but you left :-)
11:15:04  <andythenorth> about cargo payment rates?
11:15:10  <planetmaker> what was your question? Something with templates... parameters. I still did not ... yeah, something
11:15:27  <andythenorth> probably not easy to understand by irc
11:15:32  <planetmaker> so... 9 different payment schemes, depending upon grf parameter
11:15:33  <andythenorth> can you open excel files?
11:15:44  <planetmaker> yes. OpenOffice or Numbers should do it.
11:17:04  <andythenorth> I'll pm you through the forum
11:17:10  <planetmaker> k
11:18:09  <andythenorth> sent
11:20:53  <Ammler> Rubidium:
11:21:48  <Ammler> <-- ps axuff
11:22:23  <Ammler> <-- free
11:23:12  <Ammler> so it isn't the real memory, more kmemsize which is somehow related to socket buffers but...
11:26:11  <Ammler> he, backup to my local server is still running :-)
11:31:12  <Ammler> We could also update Redmine to the stable 0.9 release....
11:32:43  <planetmaker> dunno. Do you expect it to help? (and wouldn't it be a downgrade?)
11:33:18  <planetmaker> andythenorth: I have the file open now... so high, medium, low... those are the revnue generated by that cargo per tile and unit time?
11:34:28  <Rubidium> so something is leaking sockets?
11:34:48  <planetmaker> and 2nd question: do you already have somewhere code for setting cargo payment rates? where?
11:34:54  <planetmaker> Rubidium: yes... iirc
11:35:07  <andythenorth> planetmaker: it would be prop 12 for cargo (action 0)  "cargopricefactor"
11:35:27  <Ammler> planetmaker: it wouldn't help
11:35:36  <planetmaker> hm... :S
11:35:39  <Ammler> memory we have enough :-)
11:35:40  <andythenorth>
11:35:50  <planetmaker> andythenorth: thx
11:36:16  <andythenorth> I might also want to adjust time factor...prop 11
11:36:23  <Ammler> Rubidium: that is just what the very skilled support answered to my help question :-)
11:36:50  <andythenorth> planetmaker: does my table make sense?
11:37:04  <Ammler> apache seems to use much of that, I disabled some modules, it is a bit better now.
11:37:23  <planetmaker> andythenorth: I think so, yes
11:38:49  <Ammler> hmm, running apache with ulimit?=
11:41:50  <Ammler> I need somehow to drop VSZ and RSS
11:42:00  <Ammler> (from the ps axuff output)
11:47:45  *** KenjiE20 has joined #openttdcoop.devzone
11:57:36  <Ammler> planetmaker: the issues somehow reminds me of the problems we had as the web was on brians server
11:57:59  <Ammler> maybe it is something which changed in newer services
11:58:11  <planetmaker> hm... Might actually be the very same thing. Indeed
11:58:51  <Ammler> and brians server was a root
11:59:06  <planetmaker> yes, true. Very good point.
11:59:32  <Ammler> hmm, I could shutdown blog/wiki for some time
11:59:40  <Ammler> we see quite fast, if it is better then
12:01:33  <planetmaker> is the blog / wiki already moved?
12:01:46  <planetmaker> if not... have Osai move that fast :-)
12:01:56  <planetmaker> but iirc it is... ?
12:02:51  <Ammler> Osai is away until 2nd
12:03:28  <planetmaker> oh, I see
12:09:52  <Ammler> hmm, server is again over limit, can't run any shell command anymore
12:12:34  *** welshdragon is now known as Guest800
12:12:37  *** welshdragon has joined #openttdcoop.devzone
12:18:06  *** Guest800 has quit IRC
12:20:36  <Ammler> initated reboot, cu later :-)
12:24:34  *** Hirundo has quit IRC
12:24:34  *** Ammler has quit IRC
12:30:42  *** Ammler has joined #openttdcoop.devzone
12:31:00  <Ammler> back :-)
12:31:01  *** Hirundo has joined #openttdcoop.devzone
12:32:12  <planetmaker> wb :-)
12:33:29  <Ammler> disabled
12:33:38  <planetmaker> hm?
12:41:31  <andythenorth> planetmaker: I think I follow the template code, but not sure
12:41:35  <andythenorth> can you explain it a bit?
12:42:44  <planetmaker> yes, sure. What do I need to explain? :-D
12:43:35  <planetmaker> My idea was to give the 9(?) different cargo schemes their own names via a define globally (first lines)
12:43:46  <andythenorth> makes sense
12:44:11  <planetmaker> #
12:44:12  <andythenorth> is RATE_VAR_LOW a count?
12:44:13  <planetmaker> #define RATE_VAR_NUMBER 01 <-- defines the parameter number to use for the player
12:45:08  <planetmaker> #define RATE_VAR_LOW    \b01 <-- the value to set, if you want version one. Probably 'low' is wrong, but rather 'RATE_VAR_SCHEME1' would be better
12:45:30  <planetmaker> that probably made it quite unclear, I guess :-)
12:45:34  <andythenorth> ah
12:45:37  <andythenorth> that makes sense
12:45:59  <andythenorth> so RATE_VAR_SCHEME1 could be RATE_VAR_MINING ?
12:46:02  <andythenorth> etc
12:46:10  <planetmaker> yes.
12:47:25  <planetmaker> and all other 'high', 'med' and 'low' should then also be replaced by 'mining', 'farming' etc.
12:48:06  <andythenorth> should line 8 '// in industry template'  read '// in cargo template' ??
12:48:50  <planetmaker> uh, yes, I think. Industries don't have a thing there, we only deal with cargos :-)
12:49:47  <planetmaker> hm... I should template FIRS way more :-)
12:50:01  <planetmaker> This cargo scheme cries for a table simple table.
12:50:07  <andythenorth> I think so :D
12:50:22  <andythenorth> I am hoping this cargo rate thing is quite simple once we have a design
12:50:52  <andythenorth> If we can make it work, my plan is to release new rates gradually over time.  It would maintain interest in FIRS.
12:50:55  <andythenorth> They could e
12:50:57  <andythenorth> oops
12:51:04  <andythenorth> they could be matched with scenarios :)
12:51:47  <planetmaker> you mean v1 has only one rate. And v2... has three different rates etc? Nice. Hm... an idea for add-on newgrf. They just need to know the GRFID of firs.
12:52:10  <andythenorth> yes
12:52:24  <andythenorth> so v1 might have 'Vanilla' (TTD style) and 'Mining' or whatever
12:52:40  <andythenorth> I'd try and match it to available vehicle and house sets as well
12:52:58  <andythenorth> I figure I play more when there are new sets / challenges
12:53:06  <planetmaker> indeed. Good idea. Could even depend upon other grfIDs present :-)
12:53:20  <andythenorth> think that could be fragile?
12:53:44  <planetmaker> well, yes.
12:53:48  <andythenorth> depends on who the other grf authors are :P
12:53:49  <planetmaker> probably a lot of work
12:53:54  <andythenorth> meanwhile
12:54:00  <andythenorth> I am looking at cb 28
12:54:02  <planetmaker> yes. I didn't say to support _all_ other newgrfs, but some
12:54:27  <planetmaker> hm... which properties should depend on your scheme?
12:54:42  <andythenorth> for the cargos?  just prop 12 initially
12:54:45  <andythenorth> maybe prop 11
12:54:45  <planetmaker> ok
12:54:52  <planetmaker> well. Decide now :-)
12:55:10  <planetmaker> I feel like changing it to real code. And 'maybe' is a pain :-P
12:55:24  <andythenorth> ok prop 11 then
12:55:29  <andythenorth> also have a look at prop 17/18
12:55:29  <planetmaker> both. ok
12:55:33  <andythenorth>
12:55:43  <andythenorth> that could be useful?  But it would be for industries not cargos
12:55:59  <andythenorth> I think that could be templated differently
12:56:08  <planetmaker> ah, yes. Another construction site. Make it an issue so it's not forgotten.
12:56:18  <planetmaker> the templating would work the very same, I think
12:57:07  <andythenorth> I'll put both cargos and industries as issues (x2)
12:58:34  <planetmaker> btw, did you try to enter / as the first character in xcode?
12:59:48  <andythenorth> no
12:59:54  <andythenorth> I don't use xcode
13:00:55  <planetmaker> ok
13:01:46  <andythenorth> planetmaker:
13:07:43  <planetmaker> hm, can I set the same action0, property several times? Will only the last take effect?
13:09:18  <andythenorth> yep
13:09:37  <andythenorth> I wrote some HEQS code like that yesterday
13:10:07  <andythenorth> it's also how one newgrf modifies another (in the case of ship capacity multiplier for example)
13:12:06  <planetmaker> ok, good
13:12:16  <planetmaker> then I don't need to care for the defaults
13:12:27  <Ammler> diabling didn't help
13:12:36  <Ammler> we still have memory issues
13:12:48  <planetmaker> :-S
13:13:00  <planetmaker> redmine wasn't on brian's server, was it?
13:13:07  <planetmaker> it was the wiki...
13:13:24  <KenjiE20> wiki and blog
13:14:21  *** welshdragon has quit IRC
13:15:11  <KenjiE20> @invite Phazorx
13:15:13  <Webster> KenjiE20: The operation succeeded.
13:16:08  *** welshdragon has joined #openttdcoop.devzone
13:16:39  *** Phazorx has joined #openttdcoop.devzone
13:16:41  <Phazorx> weee
13:16:43  <Phazorx> thanks
13:16:55  <KenjiE20> lazy bones
13:16:57  <KenjiE20> :P
13:16:59  <Phazorx> well yeah
13:17:04  <Phazorx> and inept client
13:17:32  <Ammler> I have no idea with that fastcgi crap
13:17:32  <Phazorx> so is there short term workaround and long term solution?
13:17:39  <Phazorx> Ammler: i can help with that
13:17:44  <Phazorx> did that few times :)
13:18:03  <Phazorx> and
13:18:05  <Ammler> and we need that, as we run php
13:18:07  <Phazorx> that is no eve necessary
13:18:16  <Ammler> not?
13:18:24  <Phazorx> well for starters you can split  static content from dynamic
13:18:33  <Phazorx> have lighty do staticand apachy php/perl
13:18:49  <Phazorx> but fastcgi is really a good way to deal with php too
13:18:50  <Ammler> that wouldn't help..
13:18:52  <Phazorx> it is more powerfull
13:19:08  <Phazorx> what distro we use?
13:19:08  <KenjiE20> doesn't help when everything is PHP :P
13:19:23  <Ammler> apache needs same amount of memory, dosn't matter whats running
13:19:40  <Ammler> Phazorx: suse
13:19:46  <Ammler> but that shouldn't matter
13:20:23  <Rubidium> tip: don't use svn behind lighty (i.e. svn co http://... when lighty is the http server)
13:20:30  <Ammler> I also tried once nginx, but than I gave up because of the cgi.
13:20:47  <Ammler> then*
13:22:49  <Ammler> you really think, apache is a bad choise, can't it be just wrong config?
13:22:50  <Phazorx> Rubidium: why?
13:23:08  <Phazorx> Ammler: i moved all productionenvironments i ever delt with from apapche
13:23:15  <Phazorx> it is good for development
13:23:19  <Phazorx> when you dont know what yu need
13:23:25  <Phazorx> and it is very compatible
13:23:32  <Phazorx> but for performance it sucks pretty bad
13:23:48  <Phazorx> depending what you do tux/nginx/lighty will work far better
13:24:14  <Ammler> well, we also have rails running, but that could work with fastcgi too
13:24:18  <KenjiE20> Ammler: try setting the EnableMMAP & RLimitMEM ?
13:24:28  <Rubidium> Phazorx: because then I can OOM lighty
13:24:39  <Phazorx> Rubidium: err... how?
13:25:13  <Phazorx> rails are fine with FCGI
13:25:45  <KenjiE20> <-- the Apache section might help
13:25:47  <Webster> Title: Low memory MySQL / Apache configurations - VPSLink Wiki (at
13:26:05  <Phazorx> is mysql on same box?
13:26:19  <KenjiE20> I said "the Apache section"
13:26:45  <Phazorx> that is notthe answer
13:26:56  <Phazorx> we use some RDBMS i just want to know where it hosted
13:29:22  <Rubidium> Phazorx:
13:29:23  <Webster> Title: Lighttpd - Bug #1283: Memory usage increases when proxy+ssl+large file - lighty labs (at
13:31:40  <Phazorx> interestng
13:31:46  <Phazorx> Rubidium: could there be some other issues?
13:31:51  <Phazorx> i used similar setup
13:31:55  <Phazorx> minus SSL
13:32:02  <Phazorx> since it was intranet setup
13:32:06  <Rubidium> Phazorx: you don't need SSL, just a proxy is enough
13:32:35  <Ammler> the ugly thing is, we had the same server running on a pc with half the memory and limits and it worked quite well compared to the new server.
13:32:36  <Phazorx> well your setup mentions it
13:32:40  <Rubidium> and with the attitude of them w.r.t. to OOM killing their servers I don't have much faith in the quality of the rest of the thing
13:33:45  <Phazorx> Rubidium: i personally had Jan helping me with some problems
13:33:56  <Phazorx> so i cant complain about level of support for eve not paying customers
13:34:37  <Rubidium> Phazorx: but seriously... it's a problem that killed our server every few minutes
13:34:59  <Rubidium> and especially if stupid people were doing checkouts of the whole repository over a slow internet connection
13:35:46  <Rubidium> I could OOM the server with svn co and then once it started downloading pressing CTRL-Z so it slept client side
13:35:55  <Phazorx> Rubidium: sad..
13:36:00  <Rubidium> then the server did die of OOM and I didn't use much bandwidth
13:36:05  <Phazorx> have you tried playing with tcp stack and ulimits?
13:36:21  <Rubidium> Phazorx: yes, ulimits 'helps'
13:36:24  <Phazorx> as well as keep alive
13:36:36  <Phazorx> which i tihnk is a good idea to disable on busy boxes
13:36:47  <Phazorx> as well as lowering sack timeouts
13:36:51  <Phazorx> and TTL
13:36:53  <Rubidium> as ulimit kills lighty before it kills the other more important processes
13:36:58  <Phazorx> heh
13:37:02  <Phazorx> not quite qhat meant :)
13:37:22  <Phazorx> Rubidium: i take it yo dont have testing environment available
13:37:26  <Phazorx> to try few things
13:38:19  <Rubidium> Phazorx: the problem is that with proxy it downloads whatever is proxies as fast as it can and then serves that to the client. If the client downloads with 20 kB/s and apache+dav_svn serves at 10 MB/s, lightly would use ~10 MB extra memory per second
13:38:43  <Phazorx> dont really see a reason why it can do that
13:38:49  <Phazorx> but if it is file size dependent
13:38:53  <Phazorx> it implies buffering
13:39:08  <andythenorth> planetmaker: I have a defines question
13:39:15  <Phazorx> in which case not keeping connections alive might help situation
13:39:22  <planetmaker> define the question, andythenorth :-D
13:39:23  <andythenorth> is it possible to have a default value, then over-ride it?
13:39:26  <Phazorx> not sure how proxy <> hostbost talk to each other there tho
13:40:22  <Rubidium> Phazorx: the problem is that mod-proxy downloads as fast as it can, instead of as fast as it uploads. This means that it needs to buffer whatever it can't upload straight away
13:40:30  <Phazorx> but  i dont het Jans attitude on that either
13:40:41  <Phazorx> perhaps it is already fixed/notexistent in 2.0?
13:40:49  <Rubidium> which no keep alive or tcp tweaking can help with
13:41:08  <Phazorx> Rubidium: i can only suggest fining different means when there is no asycn transfer like that
13:41:15  <Phazorx> as in stream file from where it is
13:41:22  <Phazorx> not via another webserver
13:41:40  <Phazorx> Rubidium: it might help for sure
13:41:44  <Rubidium> Phazorx: that would be either having support for dav_svn in lighty, which it didn't or not using lighty
13:41:50  <planetmaker> he... andythenorth that was my question a few minutes ago :-)
13:42:05  <Rubidium> well, we chose for the latter and now lighty doesn't leak anymore (nor run)
13:42:09  <Phazorx> Rubidium: does proxy buffers whole tihng or only part ?
13:42:10  <andythenorth> I'll paste something
13:42:39  <planetmaker> I know what you ask. I don't know the anser, but suspect that it is 'yes'. As you answered yourself :-)
13:42:40  <Rubidium> Phazorx: the whole thing, if it fits in memory
13:43:12  <Phazorx> Rubidium: and issue is on host box corect?
13:43:20  <planetmaker> andythenorth: don't waste your time on the cargo schemes. I'm right now templating it.
13:43:25  <Phazorx> not the proxy
13:43:37  <Rubidium> Phazorx: it's on the lighty doing the mod-proxy
13:43:48  <andythenorth> planetmaker: this is cb 28 industry location permissibility
13:43:49  <andythenorth>
13:44:17  <andythenorth> planetmaker: mistake in that code :)
13:44:48  <andythenorth> fixed
13:44:58  <planetmaker> you confuse me, andythenorth
13:45:00  <Phazorx> so your setup is world > lighty (mod_proxy) <> apache(https) ?
13:45:17  <andythenorth> planetmaker: too many things at once?
13:45:27  <planetmaker> you lost me completely, yes
13:45:31  <Rubidium> world <> lighty (mod-proxy) <> apache + dav_svn (NO SSL/HTTPS)
13:45:39  <andythenorth> oops  :O
13:45:42  <andythenorth> sorry
13:45:48  <Phazorx> okay, so all goes over http
13:45:54  <Rubidium> Phazorx: yes
13:46:15  <Phazorx> and lighty as frontend proxy fetches whole thing but never releases ram used in process if client dies
13:46:16  <andythenorth> planetmaker: I'll see if I can figure out this one on my own for a bit ;)
13:46:30  *** PeterT has joined #openttdcoop.devzone
13:46:34  <planetmaker> well... I didn't get your question concerning that code... is there one?
13:46:44  <Rubidium> Phazorx: that and more related issues...
13:47:03  <Phazorx> and is it never or not soon enough?
13:47:54  <Rubidium> Phazorx: imagine world < 1 KiB/s > lighty (mod-proxy) < 100 MiB/s > apache (dav_svn)
13:48:59  <Rubidium> now the world requests something, so lighty requests it at apache. Apache starts streaming the data at 100 MiB/s to lighty, lighty forwards at 1 KiB/s
13:49:58  <Rubidium> lighty does NOT tell apache to wait for a while, but keeps accepting at 100 MiB/s, as such lighty needs to buffer that, so memory grows with (100 MiB - 1 KiB)/s
13:50:30  <Rubidium> which means that if you got 1 GB of memory you're in 10 seconds out of memory and it'll get OOM-killed
13:51:32  <andythenorth> ow templating hurts my brain
13:51:52  <Phazorx> yeah that is a bad case of usage i'd say
13:51:56  <Rubidium> Phazorx: now replace lighty with nginx and nginx tells apache to slow down (via TCP's behaviour), so nginx does not need to buffer the data, but just forward it to the client. nginx doesn't use extraordinary amounts of memory, doesn't OOM
13:52:25  <Phazorx> yeah nginx is better for that case i totaly agree
13:52:35  <Phazorx> lighty is design for different type of usage
13:52:54  <Phazorx> like it has internal buffering done in a funcky async way
13:53:11  <Rubidium> then... either don't provide proxying or provide it in a way that doesn't kill the server
13:53:17  <Phazorx> which would be taken advanatge of if there would be some other way to access the file
13:53:20  <Phazorx> not via HTTP
13:53:32  <Phazorx> like NFS or soemthing
13:53:51  <Rubidium> dav_svn is no file system, just on-the-fly generated data
13:53:52  <Phazorx> the proxying in apache works marvelously in other circumstances :)
13:54:08  <Phazorx> Rubidium: isnt it timed to some extent?
13:54:13  <Phazorx> and then dumped somewhere
13:54:33  <Rubidium> Phazorx: no, it's generated on-the-fly
13:55:55  <Phazorx> hmm.. i guess i had a bit diff svn setup then
13:56:04  <Phazorx> yeah well throtling can be your saver
13:56:13  <Phazorx> but nginx is better solution for proxy for sure
13:56:23  <Phazorx> it is esentialy a proxy grown into something bigger :)
14:21:35  <planetmaker> andythenorth: //!!Warning (197): Offset 1: <add-in-11> should not be negative.
14:21:35  <planetmaker> //!!Warning (197): Offset 2: <add-in-1> should not be negative.
14:21:36  <planetmaker>  1790 * 15	 02 0A 00 //-1 * 0 0A <set-id>
14:21:55  <planetmaker>  //!!Warning (209): Offset 4: Found byte 1 of a 4-byte escape while reading byte 1 of a 1-byte field.
14:22:36  <planetmaker> ^^ I strongly recommend to fix those errors and warnings before doing too many other things. They are abundant and hide real errors
14:23:01  <andythenorth> I'll have a look
14:23:49  <andythenorth> planetmaker your sprite 1790 != my sprite 1790 :|
14:24:27  <planetmaker> whatever. Just open your local nfo file
14:24:51  <planetmaker> and the output of renum tells you the sprite numbers it warns about
14:24:58  <andythenorth> ah found them
14:28:33  <andythenorth> grr
14:30:22  <andythenorth> not sure how to fix these
14:32:35  <andythenorth> planetmaker: I'm not sure that warning 197 matters
14:34:02  <planetmaker> I'm not sure either, I haven't quite looked at it.
14:34:17  <planetmaker> I just noticed that renum output is a pain to look at.
14:34:34  <andythenorth> I am trying to find the 209 problem
14:34:37  <andythenorth> I don't see it
14:35:04  <Rubidium> newer nforenum?
14:35:14  <andythenorth> I need one, yes
14:35:25  <planetmaker> I use r2281
14:35:31  <andythenorth> I have a horrible feeling mine is  very old
14:35:35  <andythenorth> I thought I changed it
14:35:50  <planetmaker> hm, download the new one and enjoy?
14:36:01  <planetmaker> just typing renum will tell you
14:36:04  <planetmaker> in the xterm
14:36:09  <andythenorth> I remember, I couldn't get the path to it to work
14:36:20  <andythenorth> I have r2251
14:36:25  <andythenorth> I'll get a new one
14:36:30  <planetmaker> that's moderately modern
14:37:04  <andythenorth> I think my makefile.local is still using an old one via crossover
14:37:29  <planetmaker> just get rid of makefile.local :-)
14:39:09  <andythenorth> :(
14:39:23  <planetmaker> well... what do you need there?
14:39:30  <andythenorth>
14:40:19  <andythenorth> permissions?
14:40:27  <planetmaker> one would think so
14:42:34  <planetmaker> but not 100% sure...
14:43:45  <andythenorth> oh dear :|
14:43:53  <andythenorth> the renum in crossover works
14:43:57  <andythenorth> let me try the latest renum
14:45:26  <planetmaker> hm... You don't have by chance an installation of boost on your machine?
14:46:10  <andythenorth> don't think so
14:46:21  <andythenorth> oh maybe
14:46:27  <andythenorth> I might have it via macports
14:54:20  <andythenorth> planetmaker: I have renum 2284 windows version working
14:54:27  <andythenorth> I am defeated by the mac version ;)
14:54:43  <andythenorth> I have a *lot* more warnings now!
14:54:55  <planetmaker> :-P
14:55:03  <planetmaker> yes, probably all I have, too
14:55:29  <planetmaker> andythenorth: if you have boost via macports, you might actually try to compile nforenum yourself.
14:55:35  <planetmaker> svn co svn://
14:55:40  <planetmaker> and just make it
14:57:14  <andythenorth> fails on Boost, probably I need to setup a path somewhere.  another day.  I now have an insane amount of errors to look at :(
14:59:12  <andythenorth> I think error 209 is a failure by nfo renum
14:59:42  <andythenorth> and 197 is renum unhelpfully being helpful
15:02:19  <planetmaker> hm, right. 2cctrainset disables 209
15:03:04  <andythenorth> is that in the 2cc makefile? I can copy it if so...
15:03:19  <planetmaker> no. In the header pnfo file
15:03:44  <planetmaker> warnings are better disabled in the nfo than in the makefile
15:03:50  <andythenorth> actually I can't copy it, redmine won't let me at the 2cc repo
15:03:58  <planetmaker> of course it will
15:04:21  <andythenorth> re
15:04:24  <andythenorth> internal error
15:04:32  <planetmaker> anyone can view it and checkout it. As you have commit rights you can even commit. But then I'll kick your but, if you don't ask DJN before doing so ;-)
15:04:58  <andythenorth> it's back
15:04:59  <planetmaker> oh, once again :-(
15:06:53  <andythenorth> can't checkout via hg either - same problem
15:07:01  <andythenorth> a day of problems  :O
15:07:14  <planetmaker> but... you could just look at nfo_header.pnfo of FIRS :-P
15:08:24  <andythenorth> yup, that might work
15:09:37  <andythenorth> right that takes care of most of the errors - the easy way :)
15:12:17  <planetmaker> well. I don't recommend to add any error there unless unambigously established that it's not a coding error of us
15:14:05  <andythenorth> I agree...
15:14:17  <andythenorth> I don't know what causes 209, but the wiki suggests it's wrong
15:14:31  <andythenorth> I see what 197 is trying to do, but I believe it's wrong
15:18:32  <andythenorth> planetmaker: I've pushed those header changes
15:18:50  <planetmaker> why is 197 wrong?
15:20:19  <andythenorth>
15:20:29  <planetmaker> even if the callback limits to 0 and doesn't allow negative numbers ingame. Why call it with negative numbers in the first place?
15:20:44  <andythenorth> FF FF is not negative if the value is a word?
15:20:57  <andythenorth> the world of negative in hex is admittedly strange
15:21:14  <andythenorth> ah, it's signed
15:21:17  <andythenorth> I could be wrong then
15:21:42  <andythenorth> which makes renum correct  :O
15:22:09  <planetmaker> FF FF of course is negative for signed vars
15:22:19  <planetmaker> EF FF is the positive number
15:22:42  <planetmaker> or FF EF. But rather use \d<number> then maybe?
15:22:52  <planetmaker> or \w<number> Dunno which
15:23:27  <planetmaker> andythenorth: that's why I said: don't add it without checking thoroughly ;-)
15:24:05  <planetmaker> preferrably we shouldn't need any. But then nforenum would need to learn a few things more... alas
15:28:04  <andythenorth> planetmaker: I've corrected the errors that were causing 197
15:28:05  <andythenorth> pushed
15:28:12  <planetmaker> nice :-)
15:28:17  <andythenorth> I've also removed the header suppression of 197
15:28:42  <andythenorth> is 209 a bug with renum?
15:30:43  <Ammler> meh, I give up on the music thread
15:30:59  <planetmaker> andythenorth: I think so...
15:31:00  <Ammler> they tend to use CC crap license
15:31:26  <andythenorth> CC is just a waste of everyone's time
15:31:30  <andythenorth> choice paralysis
15:31:43  <andythenorth> (back to the defines) mumble mumble
15:31:49  <Ammler> :-D
15:33:23  <andythenorth> planetmaker: if I try to undef something that is not defined, do bad things happen?
15:33:33  <planetmaker> no, that's fine
15:39:54  <andythenorth> I've solved my location code problem.  It works.  I think it's a bad design pattern
15:40:33  <planetmaker> which design?
15:41:30  <planetmaker> btw, thanks for fixing the warnings. Or the largest amount thereof. One error, one warning remain :-) But now they're visible
15:41:50  <andythenorth> one of the errors is a tile problem.  FooBar did the tiles...
15:41:58  <andythenorth> ...but I have to learn how they work, so I'll get to it some time
15:42:05  <planetmaker>  //!!Error (188): Offset 20: ID B0 is defined with feature 0A. <-- that one?
15:42:10  <andythenorth> yes
15:42:19  <andythenorth> it's probably trivially solved
15:42:37  <planetmaker>  //!!Warning (94): Redefining ID 10 not used since previous definition at sprite 410. <-- hm, that one has a solution, though
15:43:35  <andythenorth> that one might be gone now I've changed how the location CB handling works
15:43:40  <andythenorth> pull ?
15:44:02  <planetmaker> me?
15:44:08  <andythenorth> yep
15:44:28  <planetmaker> gone, yes
15:44:31  <planetmaker> :-)
15:44:55  <planetmaker> now back to my cargo_scheme.list :-)
15:44:57  <andythenorth> can you try and review my location code?
15:45:06  <andythenorth> it's a bit fragile
15:45:40  <planetmaker> he... I so far never looked at those parts of NFO... so I have not much knowledge how it works there, but yes, I can try
15:45:45  <andythenorth> so not every industry is going to need construction location checks
15:45:55  <andythenorth> but it has to be called from the templated action 2 code
15:46:12  <andythenorth> e.g. template_primary_action23.pnfo
15:46:28  <andythenorth> where it is commented // handle callbacks
15:46:59  <andythenorth> if 04 00 (04 80) is returned to the cb, it just proceeds to build
15:47:20  <andythenorth> so I've set a define to provide that value for all industries using that template
15:47:26  <andythenorth> I figure that's the fragile bit
15:47:38  <andythenorth> if the define is missing from an industry, it breaks the nfo
15:47:39  <planetmaker> is it commited yet or not?
15:47:41  <andythenorth> yes
15:47:43  <andythenorth> committed
15:47:59  <planetmaker> ok. Then I can open the files :-)
15:48:16  <andythenorth> have a look in i_coalmine and the template I mentioned above
15:50:21  <planetmaker> ok, I'm there
15:51:36  <planetmaker> one comment: the subtract amounts... you would want them as defines, I think. Easier to change. What do you think?
15:53:25  <andythenorth> hmm...interesting....but I don't think it's necessary, as they are templated anyway.   They don't change per-industry.  Maybe an abstraction too far?
15:54:08  <planetmaker> Yes, they don't change per industry... But but it makes tweaking the newgrf easier. But well... not necessary.
15:54:27  <andythenorth> ;)
15:54:30  <planetmaker> have you, btw, heart back from foobar at least *somewhen*?
15:54:48  <andythenorth> it's been quite some time...
15:55:08  <andythenorth> October
15:55:22  <planetmaker> hm... :S
15:55:50  <andythenorth> FIRS is kind of just me at the moment, so all your help very much appreciated!
15:56:44  <planetmaker> I'm still on the template... making the sets into a simple csv-readable table :-)
15:56:54  <planetmaker> the cargo schemes template :-)
15:57:01  <andythenorth> ok cool
15:57:24  <planetmaker> the same way I did the whole thing for comic_houses. yet another abstraction, yet another tool with awk ;-)
15:57:29  <planetmaker> but you have it all, so no worries
15:58:00  <planetmaker> awk takes the list and writes pnfo files which then get parsed - as usual - by gcc ;-)
15:58:11  <andythenorth> ok, that seems nice
15:58:23  <andythenorth> so just csv?
15:58:36  <planetmaker> concerning the action23 ... looks nice to me for now.
15:58:50  <planetmaker> csv would do, yes. I call it .list, though
15:59:26  <planetmaker> as the scripts then require it to be in the order they expect it to be.
15:59:55  <planetmaker> well, I'll give you the diff when I have no errors ;-)
16:00:29  <planetmaker> anything in particular with your template which you think is problematic?
16:01:33  <andythenorth> having thought about it I'm not so bothered about the CB 28 being fragile....
16:01:34  <andythenorth> the industries already have to have n defines anyway
16:01:38  <andythenorth> or they break the nfo
16:02:14  <andythenorth> I think the reason I don't like it is that *most* industries won't use CB 28....but I'm adding code to all of them for the sake of just a few
16:02:19  <andythenorth> I don't want to fork the template though
16:02:22  <andythenorth> that would be a pain
16:04:08  <planetmaker> what you could do is: #define THIS_USECB38
16:04:19  <planetmaker> #ifdef THIS_USECB38
16:04:27  <planetmaker> #include "cb38.template"
16:04:29  <planetmaker> #endif
16:04:48  <planetmaker> #undef THIS_USECB38
16:04:57  <andythenorth> that might work
16:05:53  <andythenorth> presumably I can ignore the #include and just write code between the ifdef?
16:05:56  <andythenorth> is there a notifdef?
16:06:03  <andythenorth> or ifnotdef :)
16:06:17  <planetmaker> yes, you could do that, too
16:06:22  <planetmaker> #ifndef
16:06:23  <andythenorth> might try that
16:07:31  <andythenorth> thanks
16:12:03  <Ammler> Rubidium: will you include nosound to the openttd release or shall I rpm that too?
16:12:16  <Ammler> (for dedicated servers)
16:12:24  <Rubidium> not sure yet
16:13:20  <Ammler> same reason, as you kept, I guess :-)
16:13:54  <Rubidium> huh? kept
16:14:06  <Ammler> the error message with dummy file
16:15:38  <Rubidium> that's because the metadata for the original files is part of the packages
16:15:38  <Ammler> hmm, is there somewhere the source for nosound?
16:15:56  <Rubidium>
16:17:12  <Ammler> Rubidium: I meant the discussion we once had about removing the requirement for it, foobar once made a fs post about.
16:17:59  <Ammler> the decision of you devs were, that you like to inform the people, if they don't have sound.
16:21:47  <Rubidium> no, we want to inform them when the files they have are corrupted
16:22:23  *** andythenorth has quit IRC
16:26:09  <Ammler> Rubidium: nvm, that was on time before opensfx....
16:26:33  <Ammler> you could have removed the requirement for
17:29:57  *** welshdragon has quit IRC
17:30:26  *** welshdragon has joined #openttdcoop.devzone
17:55:35  <Rubidium> Ammler: could have, but then there would've been no incentive to make such a set
17:55:49  <Ammler> :-)
17:56:36  <Ammler> I thought and belugas told it that way, you kept it because you feared bugreports like "Why is there no sound..."
17:57:41  <Ammler> which made sense to me and might be the issue, when you include nosound to openttd distribution.
18:25:13  *** andythenorth has joined #openttdcoop.devzone
18:25:21  <andythenorth> ta da
18:25:24  <andythenorth> I'm back :)
18:28:15  <Ammler> WP
18:28:22  <Ammler> WB*
18:34:44  *** Madis has joined #openttdcoop.devzone
18:49:01  <andythenorth> planetmaker: I've pulled the cargo schemes :)
18:49:03  <andythenorth> pdq2s-macbook-3:firs_build andy$ make install
18:49:03  <andythenorth> cat: sprites/nfo/cargo_schemes.pnfo: No such file or directory
18:49:04  <andythenorth> [Generationg Cargo schemes]
18:49:26  <planetmaker> yes. That's unfortunate. But ok
18:49:50  <planetmaker> it's an error in the makefile without consequences
18:50:16  <planetmaker> it doesn't exist when checking for pnfo, it's generated later. I hope I come to changing that :-)
18:50:28  <planetmaker> but there's no quick fix
18:50:35  <andythenorth> so how complete is it? :D  Does it understand parameters yet?
18:51:15  <planetmaker> in principle yes. In practical the values for everything but default are not defined an nothing is changed.
18:51:21  <planetmaker> look at cargo_scheme.list
18:51:39  <andythenorth> I have had a look....I am thinking of testing another scheme
18:51:43  <andythenorth> maybe Mining
18:52:03  <planetmaker> copy it, change the DEFAULT to MINING (which is already defined in id.pnfo) and add sensible values, paste it in the bottom of cargo_schemes.list
18:52:08  <planetmaker> then it should work already
18:52:56  <planetmaker> new parameter values need a name / definition in ids.pnfo and their meaning they get in cargo-schemes.list
18:53:02  <planetmaker> everything else is taken care of
18:53:36  <planetmaker> I just template-ificate the cargos action0s as a whole, converting it to a list.
18:54:05  <andythenorth> so copy inside the file, or copy the file itself?
18:54:14  <planetmaker> like that:
18:54:16  <planetmaker>  //Properties:08 + div.	0E	0F	13+14	15	16				18	19	1A
18:54:17  <planetmaker> // CargoID		Sprite#	Weight	Colour	Freight	Cargo				Subst	Mult	Callback
18:54:19  <planetmaker> //						status	classes				Towngrowth	Flags
18:54:20  <planetmaker> PASSENGERS,		FF FF,	 1,	152,	0,	pax,				00	256	00
18:55:03  <planetmaker> Well... I'm not sure what's best: adding the different schemes for one cargo below eachother or keeping together the schemes as a whole.
18:55:13  <andythenorth> schemes as a whole
18:55:19  <andythenorth> cargos aren't likely to change much
18:55:28  <andythenorth> the other way is a massive copy and paste shuffle ;)
18:55:57  <planetmaker> well... a bit, yes. But might be better for comparison
18:56:14  <andythenorth> think that's just output
18:56:22  <planetmaker> hu?
18:56:33  <planetmaker> what is output of what?
18:56:46  <andythenorth> so to set up a scheme nicely, it's helpful to see all the cargos together
18:57:00  <andythenorth> ignore the output comment ;)
18:57:15  <planetmaker> ok. point.
18:57:47  <andythenorth> is this (from above) just an irc calculator doing something?   div.0E0F13+14151618191A
18:58:22  <planetmaker> that's ignoring formating / spaces
18:58:56  <planetmaker> <-- better
18:58:59  <andythenorth> I'll dig around a bit more in the source...there will probably be...questions :)
18:59:32  <planetmaker> the source is basically in sprites/nfo/*.template and in scripts/*.awk.
18:59:39  <planetmaker> The list files is just the definitions
18:59:44  <planetmaker> or values so to speak
19:01:02  *** Madis has quit IRC
19:02:54  <planetmaker> and if you copy, I meant copy within the file cargo_schemes.list (just to make that clear)
19:03:06  <andythenorth> planetmaker: so the scheme parameter is already on an ID?
19:03:13  <planetmaker> yes. 01
19:03:19  <planetmaker> but you can define that in ids.pnfo
19:03:36  <planetmaker> I just made it a define so that I don't have to worry about where it will be ;-)
19:03:41  <andythenorth> do params start at 0 or 1?
19:03:47  <planetmaker> dunno :-P
19:03:51  <andythenorth> I've never coded them before
19:03:56  <andythenorth> I assume they start at 0
19:04:00  <andythenorth> I'll test
19:04:00  <planetmaker> I think so, too
19:06:21  <andythenorth> planetmaker: will awk freak out if I space the scheme list out on nice tabs?
19:07:08  <planetmaker> it won't. The "," are important
19:07:22  <planetmaker> but they're nice for me...
19:07:34  <planetmaker> so... dunno if it makes sense to change :-)
19:07:39  <planetmaker> what tab spacing do you use?
19:08:05  <planetmaker> I use tab width of 8 spaces
19:08:10  <planetmaker> andythenorth: ^
19:08:36  <andythenorth> 4 spaces I think.  company my text editors work that way
19:09:00  <andythenorth> my tabs will probably convert to spaces anyway (kind of standard for python dev)
19:09:07  <planetmaker> ah, don't do that
19:09:13  <andythenorth> so my 'nicely lined up' might now work for you?
19:09:14  <planetmaker> that will make it ugly.
19:09:36  <andythenorth> can awk work with a different field order?
19:09:46  <planetmaker> it can. If you modify the script
19:10:02  <andythenorth> I figure the values might stay in line better
19:10:12  <andythenorth> I don't mind reading 'backwards' with the cargo names on the right
19:10:26  <planetmaker> honestly: that's a really, really bad idea
19:10:47  <andythenorth> ah ok
19:11:00  <andythenorth> hmm.
19:11:04  <andythenorth> does formatting matter so much?
19:11:09  <planetmaker> you want the lines to start with what is being defined... that's awkward, if otherwise. I'm not rtl reader ;-)
19:11:25  <andythenorth> maybe we just live with the current formatting
19:11:44  <andythenorth> python makes it a habit to care about whitespace :D
19:11:55  <planetmaker> which is generally fine.
19:12:01  <planetmaker> it just needs to be consistent.
19:12:17  <planetmaker> But I certainly don't change tab width for every project I work on here :-)
19:12:36  <planetmaker> If you like space... go for it. But then creation of such lists is... painful IMO
19:12:56  <andythenorth> I might just live with it as it is
19:12:59  <ODM> go use tabs:P
19:13:06  <planetmaker> ODM: I do :-P
19:13:11  <planetmaker> that's the whole issue
19:13:12  <ODM> good
19:13:13  <andythenorth> ah
19:13:22  <andythenorth> list looks awesome in TextEdit
19:13:27  <andythenorth> problem solved
19:13:46  <andythenorth> just one line with a tab out of place
19:13:57  <planetmaker> btw, other topic: is it intended that the cargos all have a different weight per unit
19:14:00  <planetmaker> ?
19:14:04  <andythenorth> dunno
19:14:08  <andythenorth> FooBar did that
19:15:46  <planetmaker> ok
19:16:00  <andythenorth> interesting, I have never set base prices before
19:16:02  <planetmaker> it looks like intended (too much deviation from 16), but not like it being finished
19:26:32  <andythenorth> planetmaker: any reason some commas are missing from the scheme file?  I am about to correct them...
19:27:34  <planetmaker> can you specify where?
19:28:13  <andythenorth> 1 missing from the line for GOODS, 2 for GRAIN
19:28:25  <planetmaker> right. I'll fix it. Thanks.
19:28:48  <planetmaker> ah, wait. You edit it anyway, do you?
19:28:50  <planetmaker> then go for it
19:28:55  <andythenorth> done ;)
19:29:26  <planetmaker> missing , will produce errors
19:32:32  <planetmaker> unlike cargo classes in the cargo list I work on here - they need to be separated by space or tab ;-) - as they're one entry in the whole list only
19:40:58  <Ammler> planetmaker: according to the new license talk
19:41:20  <andythenorth> planetmaker: ho hum
19:41:26  <Ammler> we should push Zeph to push the source of his opengfx stuff ;-)
19:41:29  <planetmaker> dum di dei
19:41:39  <planetmaker> Ammler: yes
19:41:51  <andythenorth> the compiled firs nfo looks right, but the payment rates are not changing
19:41:58  <andythenorth> I'm looking for documentation on parameters
19:42:18  <Ammler> oh sorry, you are still busy on FIRS dev....
19:42:26  <planetmaker> andythenorth: did you try params 0 and 1 ?
19:42:31  <andythenorth> yes
19:42:34  <andythenorth> but I'll try again
19:42:37  <planetmaker> hm
19:44:11  <planetmaker> I mean, like 0 1 and just 1? hm... well
19:44:22  <andythenorth> I've pushed
19:44:29  <andythenorth> in case anyone else wants to try
19:44:40  <andythenorth> doesn't appear to have broken anything so far ;)
19:44:59  <planetmaker> well, yes. I haven't removed anything old so far ;-)
19:45:10  <planetmaker> we just overwrite. Maybe, just maybe, that's the error...
19:45:32  <planetmaker> but then... it should be possible
19:45:36  <planetmaker> or?
19:51:20  <andythenorth> let me try something
19:54:13  <andythenorth> oh
19:54:20  <andythenorth> well I've crashed OpenTTD
19:54:22  <andythenorth> so that's progress
19:59:03  <planetmaker> andythenorth: one thing you shouldn't do is empty lines to the list files :-)
19:59:29  <planetmaker> I know, it's a feature request. But I haven't gotten around to implementing it.
20:00:11  <andythenorth> does commenting solve that problem?
20:00:24  <planetmaker> yes. Don't worry
20:00:30  <planetmaker> I did it ;-)
20:06:11  <andythenorth> now here's a weird thing
20:06:20  <andythenorth> prop 12 makes no difference to cargo payment rates
20:06:24  <andythenorth> even in a straight action 0
20:06:52  <planetmaker> strange.
20:10:55  <andythenorth> actually it does make a difference
20:11:06  <andythenorth> but the behaviour is strange
20:14:01  <andythenorth> might be specific to 0.7.4
20:14:14  <andythenorth> ahem
20:14:27  <andythenorth> front page of openttd: 0.7.5 This fixes the NewGRF cargo payment was that broken in 0.7.4
20:14:30  <andythenorth> :)
20:14:52  <andythenorth> dunno if it will fix my problem
20:15:56  <planetmaker> loooool!
20:16:05  <planetmaker> I just started for a test.
20:16:16  <planetmaker> The only wagon transporting waste is... the armoured wagon!
20:16:44  <andythenorth> ha known issue
20:18:33  <andythenorth> ok so changing prop 12 in main action 0 for a cargo does work
20:18:38  <andythenorth> but not sure if parameters do
20:18:47  <andythenorth> cargo payment rates appear to be somewhat cached...
20:18:57  <andythenorth> the cargo payment graph takes some time to update
20:19:58  <planetmaker> uh? Takes time to update?
20:20:14  <planetmaker> you mean... you change it within the running game?!
20:20:23  <planetmaker> the newgrf parameters?
20:21:50  <andythenorth> yes
20:21:52  <andythenorth> no
20:21:55  <andythenorth> not the parameters
20:22:12  <andythenorth> this is changing the 'main' action 0 and then re-applying the newgrf
20:22:36  <planetmaker> well... that's not the best way to test. Rather start a new game
20:22:36  <andythenorth> which does work
20:23:24  <andythenorth> well I now know that prop 12 does work (with Openttd 0.7.5)
20:23:32  <andythenorth> but to test the parameters...
20:25:23  <planetmaker> better always start a new game
20:26:03  <planetmaker> hm, I guess I push the cargo list. Or do you want to review it before?
20:26:24  <andythenorth> might as well push
20:29:01  <planetmaker> hm... grain has undefined strings
20:33:18  <andythenorth> where?
20:33:30  <andythenorth> also, I have a bit more progress on payment rates :)
20:35:36  <andythenorth> ok
20:36:06  <andythenorth> planetmaker: the DEFAULT scheme was on value 0 for the parameter
20:36:19  <andythenorth> moving it to 02 makes the cargo respect the 'main' action 0
20:36:30  <andythenorth> is not an answer, but it helps me understand what's going on
20:36:40  <andythenorth> so somewhere I need to deal with parameters
20:37:20  <planetmaker> hu? DEFAULT is the same as what was there.
20:37:37  <planetmaker> so setting the param to 0 (=default) doesn't change a thing
20:37:45  <planetmaker> one way or another
20:37:59  <andythenorth> in order of events:
20:38:09  <andythenorth> nvm
20:38:25  <andythenorth> it just means that the game isn't broken :D
20:38:31  <planetmaker> :-)
20:38:37  <andythenorth> and that our code isn't terribly broken either
20:39:00  <andythenorth> just somewhere parameters are not being defined correctly
20:39:15  <planetmaker> I made default such that it doesn't change anything. It does duplicate stuff in the version you have. But that's something I plan to remove in the next commits
20:39:21  <planetmaker> When  I found the missing grain string
20:39:27  <andythenorth> searching the Patch wiki for 'parameter' is unhelpful :|
20:43:55  <andythenorth> planetmaker: does 2CC trainset have parameters?
20:44:43  <planetmaker> yes. For the region availability
20:44:46  <Ammler> andythenorth: my lumber mill grf has good example how to handle defaults
20:44:57  <andythenorth> Ammler: can you help :D
20:45:26  <andythenorth> planetmaker: a parameter is 4 bytes, slicing just one might be the problem...?
20:45:32  <andythenorth> given endianness and such?
20:45:38  <planetmaker> No
20:45:47  <Ammler>
20:45:53  <planetmaker> Endianness of the OS is something which you don't concern yourself with
20:46:05  <andythenorth> no I just mean endianness of nfo ;)
20:46:12  <planetmaker> And I think it's always only one byte
20:46:22  <andythenorth> hmm
20:46:24  <planetmaker> but if you ask for more, two params are added to one big one
20:46:30  <planetmaker> or three or four
20:46:44  <andythenorth> wish I could print  params to the screen
20:46:46  <Ammler> a parameter is a dword
20:46:59  <Ammler> (4 bytes)
20:47:03  *** Frankr has joined #openttdcoop.devzone
20:49:55  <planetmaker> grr.... a pain. Grain string gets lost :-(
20:50:51  <andythenorth> where / how ?
20:58:57  <planetmaker> well. when I autogenerate the whole cargo action0s.
21:00:29  *** welshdragon is now known as Guest873
21:00:30  *** welshdragon has joined #openttdcoop.devzone
21:00:48  <andythenorth> planetmaker: oh
21:01:01  <andythenorth> is it something trivial like missing commas?
21:01:16  <andythenorth> probably not :(
21:02:37  <planetmaker> <--hm... I guess I need to add climate availability
21:02:53  <planetmaker> just see it now. That's the difference.
21:03:09  <planetmaker> but... it shouldn't matter :S
21:03:24  <andythenorth> :o
21:04:11  <planetmaker> and it only fails for grain
21:06:12  <andythenorth> well meanwhile, I have proven that most of the cargo rate scheme works, the problem is isolated to the part of the action 7 that is checking parameters
21:06:13  *** Guest873 has quit IRC
21:06:21  <andythenorth> it works if I use an action 7 checking climate for example
21:06:46  <andythenorth> could be time for a forum post...
21:06:57  <andythenorth> this *will* be worth the effort :)
21:07:55  <planetmaker> :-)
21:08:10  <Frankr> any1 any idea if you can change pax per tick with airport size
21:20:24  <andythenorth> yay
21:20:55  <andythenorth> planetmaker: I have it working :) ...provided I use an action D to define the parameter first :|
21:21:10  <planetmaker> hm
21:32:27  <andythenorth> grr
22:03:36  <andythenorth> right time for bed
22:03:39  <andythenorth> goodnight
22:04:59  *** andythenorth has left #openttdcoop.devzone
23:13:27  *** ODM has quit IRC
23:30:22  *** Phazorx has left #openttdcoop.devzone
23:56:03  *** Frankr has quit IRC

Powered by YARRSTE version: svn-trunk