IRC logs for #openttd on OFTC at 2012-02-29
⏴ go to previous day
00:31:28 *** DanMacK has joined #openttd
02:00:57 *** DanMacK has joined #openttd
02:11:10 <mox> anyone who cand help me instal a mod in chris sawer locomotion?
02:36:45 *** supermop has joined #openttd
04:03:11 *** BootAdmin has joined #openttd
04:31:56 *** cmircea has joined #openttd
05:01:13 <xiong> Mazur, would you be interested in a new FIRS cargo chart?
05:14:48 *** tokai|noir has joined #openttd
05:14:49 *** ChanServ sets mode: +v tokai|noir
05:29:43 *** tokai|mdlx has joined #openttd
05:56:23 *** Eddi|zuHause has joined #openttd
06:46:55 *** sla_ro|master has joined #openttd
07:04:47 *** Twofish has joined #openttd
07:28:01 *** andythenorth has joined #openttd
07:34:27 *** cmircea has joined #openttd
07:44:04 *** JVassie has joined #openttd
07:46:08 * andythenorth is matching animals to where they live
07:55:47 *** kkb110_ has joined #openttd
08:09:51 *** roboboy has joined #openttd
08:23:21 * andythenorth wonders about writing a PIL routine to scale sprites
08:23:28 *** tokai|noir has joined #openttd
08:23:28 *** ChanServ sets mode: +v tokai|noir
08:24:53 <andythenorth> it would need to detect edges
08:31:01 <andythenorth> the toddler has discovered train-rv crashes in the game
08:31:20 <andythenorth> I need openttd for ipad, or I never get my laptop back :P
08:33:15 *** Progman has joined #openttd
08:34:30 <andythenorth> why do I never use 64 tile trains in my games?
08:49:41 *** Progman has joined #openttd
09:07:23 *** roboboy has joined #openttd
09:23:01 *** andythenorth_ has joined #openttd
09:36:04 *** Firartix has joined #openttd
09:38:29 *** tokai|mdlx has joined #openttd
09:49:32 <Pikka> andythenorth stop stirring at once
09:50:09 <andythenorth> what did I do :)
09:50:25 <Pikka> "You can always fork ;)"
09:50:39 <andythenorth> forking seems to be the order of the day for that project
09:50:46 <andythenorth> day / five years :P
09:51:08 <Pikka> going on for ten, isn't it?
09:51:24 <Pikka> the BR set predates newgrf... D;
09:52:15 <andythenorth> we all know forks are bad ;)
09:52:30 <andythenorth> except if you're a git user, which now seems to include world+dog
09:52:51 <Pikka> you'll never eat your spaghetti with a spoon though
09:53:01 <Pikka> you should make a grf called SPAGHETTI
09:53:26 <andythenorth> someone reverse-acronym it, I'll make it :P
09:53:57 * andythenorth is going to take the toddler on his first train ride now
09:54:09 <Pikka> somewhere exciting I hope
09:54:33 <andythenorth> exciting...hmmm: Avon Gorge, then some factories :o
09:55:32 *** roboboy has joined #openttd
10:14:12 *** AcidWeb has joined #openttd
10:15:11 <AcidWeb> One fast question. There is any option that disable purchase of bankrupt companies in multiplayer?
10:15:41 <AcidWeb> I found only one that disabled shares.
10:19:40 <appe> hm, do i normally pay for the amount of stations i have on a map?
10:24:07 <planetmaker> iirc there's no such option, AcidWeb
11:12:30 <appe> bah, game time moves to fast
11:13:47 <appe> thats hard words from someone like that.
11:19:44 <Ammler> he is right, though you can take him serious anyway
11:26:35 <Ammler> also not sure, if it is a security feature to keep away root pw from a local user
11:31:25 <dihedral> i do that on every computer (root or administrator)
11:35:57 <dihedral> i do not hand out the root / administrator password
11:36:39 <Ammler> yes, but not for security, just for support
11:37:36 <Ammler> and in that case, it would be good, he couldn't connect to wlan
11:38:05 *** JVassie_ has joined #openttd
11:38:08 <Ammler> he, just changed my mind, linus is indeed wrong
11:38:24 <dihedral> no, for security - i do not want them bitches fucking around with a running system if they do not know what they are doing
11:38:48 <dihedral> installing / uninstalling software
11:38:56 <dihedral> i do not care for installing printers, wlan, etc.
11:39:17 <Ammler> and simply using a livedisk and resetting root pw isn't possible?
11:40:02 <dihedral> sure it is possible, but that then is intended harm
11:40:15 <dihedral> and then i do not care :-P
11:40:32 <Ammler> yes, so it is for security, it is that your guys don't fuck up your system and waste your time
11:40:58 <dihedral> and it's security on servers where people do not have physical access to the system
11:41:16 <Ammler> yes, I meant physical local users
11:48:09 <planetmaker> depends on what you need an want
11:50:55 <Ammler> it is advertising if linus rants about something :-)
11:52:25 <Ammler> also last time I configured wlan, I was able to checkbox the connection, if it is user or system connection
11:53:00 <Ammler> but maybe it depends on the security policy template
11:56:42 <Ammler> dihedral: the best commentar is "I think that Apple makes a decent Unix distro that runs on the Macbook Air."
11:57:43 <dihedral> well, Unix != Linux ;-)
11:58:20 <Ammler> but it is quite nice, that suse seems to be the best distro to run there
11:58:53 <MNIM> hmmmmh, I thought buntu ran on mac, too?
11:59:19 <planetmaker> but BSD != Unix :-P
12:02:39 <planetmaker> one could argue it, though
12:03:18 <appe> i admire the linguistic skills.
12:04:24 <peter1138> debian's the best distro. always. :p
12:05:12 <peter1138> only reason german-speaking people think suse is any good is cos it was made in germany ;p
12:05:25 *** Progman has joined #openttd
12:13:47 <Eddi|zuHause> peter1138: so why is that not a valid reason then? :)
12:14:34 <Eddi|zuHause> peter1138: if it was made in britain, it would be like a fire extinguisher that catches fire :p
12:14:52 <Ammler> hmm, there is for sure also blog around why linus doesn't like debian :-)
12:15:36 <Eddi|zuHause> anyway, what this whole security crap needs is a decent GUI to set it up
12:20:05 *** andythenorth has joined #openttd
12:22:43 *** TheMask96 has joined #openttd
12:30:04 <andythenorth> so if I've understood correctly, when setting min versions for a grf, I need to pay attention also to which nml version the grf is compiled with
12:30:38 <andythenorth> which means a grf compiled on bundle server might fail for ottd xyz, even though it's fine when compiled locally
12:31:05 <andythenorth> I mean nml version, not grf version
12:31:36 <Ammler> you might use head, devzone uses nightlies, which sometimes is outdated :-)
12:31:50 <andythenorth> more likely other way round in this case I think
12:31:51 <planetmaker> well, NML default branch currently aims to 1.2.0 - compatibility. Which recently means that requirement changed from 1.2.0-beta1 to 1.2.0-RC1
12:32:00 <andythenorth> I compiled and tested with an outdated nml
12:32:07 <andythenorth> then uploaded to bananas the grf from bundle server
12:32:09 <planetmaker> as the newgrf container 2 was newly introduced
12:32:22 <andythenorth> yes this would make 100% sense for the issue I had
12:32:30 <Ammler> planetmaker: not default, you speak about 0.2, don't you?
12:32:34 <andythenorth> means we need FIRS 0.7.2 :|
12:32:44 <planetmaker> why does it mean that, andythenorth?
12:32:53 <planetmaker> Just change the version requirement of the bananas entry
12:33:04 <planetmaker> changing the grf is not needed
12:33:11 <planetmaker> and won't change anything
12:33:40 <andythenorth> so only bananas min version needs to change, not action 14 props?
12:33:46 <planetmaker> basically testing for an earlier openttd version is kinda pointless: those simply don't read the grf
12:34:27 <andythenorth> do we even have a14 for ottd version?
12:34:32 <planetmaker> and those earlier versions not reading the grf is basically what that guy experiences. Though I don#t understand his hesitation to update from a random beta to the RC1...
12:34:40 <planetmaker> there's no such a14
12:34:45 <andythenorth> we have a check instead
12:34:46 <planetmaker> it would be an action7
12:35:05 <planetmaker> but you can't check older versions when using a new grfcontainer
12:35:11 <planetmaker> reading fails before that action is read
12:35:20 <andythenorth> I need to understand a little better what the right thing to do is
12:35:27 <andythenorth> so we have min version in the grf
12:35:30 <andythenorth> and min version on bananas
12:35:33 <planetmaker> the right thing is to change the bananas entry openttd version
12:35:38 <Ammler> but 1.1.5 should handle it, I thought
12:35:48 <planetmaker> Ammler: no: it will also just ignore the grf
12:35:59 <planetmaker> like also the early 1.2.0 betas
12:36:04 <Ammler> I meant, with a proper error :-)
12:36:04 <andythenorth> and this is purely dependent on which nml it was compiled with? nothing intrinsic to the grf?
12:36:29 <planetmaker> as the container was only recently added. To both openttd and nml
12:36:36 <andythenorth> so the min version check in the grf itself - if no change is needed, is it redundant?
12:36:42 <Ammler> if firs does build with nml 0.2, use that
12:36:55 <andythenorth> can I tell bundles server to do that?
12:37:10 <planetmaker> andythenorth: no: that's the min version of FIRS itself it is compatible to. The min openttd version check meanwhile is kinda redundant, yes
12:37:36 <planetmaker> changing to nml 0.2.x for FIRS is not worth it
12:37:39 <andythenorth> ok. so 1) establish min ottd version and change bananas
12:37:46 <andythenorth> 2) remove redundant version check?
12:37:47 <planetmaker> it's compatible to 1.2.0 anyway only and not to 1.1.x
12:38:11 <planetmaker> and I'd keep the version check. To be added later a fitting check ;-)
12:38:19 <planetmaker> will be needed again in the future. I'm sure
12:38:34 <andythenorth> does it need a note adding in comments?
12:38:47 <Ammler> does bananas work in the meantime and is able to download older firs on older clients?
12:39:13 <planetmaker> might be worth, andythenorth, though... cba for me ;-)
12:39:30 <planetmaker> I keep the version check in OpenGFX for the same reason
12:39:48 <andythenorth> seems it should at least be same version as bananas version?
12:39:51 <planetmaker> Ammler: not in the way you mean. nothing changed. no one worked on bananas
12:40:02 <andythenorth> although I see why that makes no difference if the check is never executed :P
12:40:08 <Ammler> andythenorth: you also need to change the md5sum check, afaik, but that might be worth to wait for makefile update
12:40:25 <andythenorth> I leave changes in the grf for now, means I don't have to re-release
12:40:32 <planetmaker> Ammler: a grf has no md5sum check...
12:40:46 <Ammler> however you call taht now
12:41:13 <planetmaker> the only place the md5sum (or the grfid) matters is within openttd.
12:41:26 <planetmaker> but if you upload the devzone bundle... all will be fine
12:41:43 <planetmaker> for the purpose of checking whether you and the devzone build the same: md5sum is still sufficient
12:41:48 <Ammler> openttd calls it still MD5sum
12:42:05 <Ammler> hmm, or does that matter, which container version?
12:42:22 *** Doorslammer has joined #openttd
12:43:59 <Ammler> so you have basically no version confirmation anymore
12:44:01 <planetmaker> as openttd probably shows you the grfid
12:44:46 <planetmaker> though the point (for me) of that file is to check whether *I* build the same thing
12:44:53 <planetmaker> where it doesn't matter
12:45:03 <Ammler> for me it was to check, if I use the same file
12:45:08 * andythenorth should have checked the md5s yesterday
12:45:11 <andythenorth> they probably vary
12:45:16 <andythenorth> usually I do actually check
12:45:18 <Ammler> or to search a unknown grf
12:45:48 <Ammler> andythenorth: as long as you use the same tool on both sides, it should still work
12:45:53 <andythenorth> yeah the md5s vary
12:46:01 <andythenorth> poop, and stronger words
12:46:01 <Ammler> you just can't use openttd as long as we use md5sum
12:46:11 <andythenorth> *usually* I check. yesterday I was rushing :(
12:46:32 <andythenorth> 50 times I check, once I don't, the one time it varies :P
12:46:39 <planetmaker> I started to simply upload the devzone build ;-)
12:46:45 <Ammler> andythenorth: I would bet, you have a M
12:46:52 <Ammler> or you use the version where you tagged
12:46:55 <andythenorth> I have a different nml
12:47:06 <Ammler> that is often the issue someone does locally :-)
12:47:14 <andythenorth> I hadn't accounted that upstream changes in the toolchain could do this ;)
12:47:15 <Ammler> hg tag <tag>; hg up <tag>
12:47:31 <Ammler> hg up <tag> is often forgotten
12:47:50 <planetmaker> changing NML often changes md5sum of the grf
12:47:58 <andythenorth> I was behind the devzone
12:48:07 <andythenorth> so I need to have exact same nml as devzone
12:48:13 <andythenorth> otherwise this will occur again and again
12:48:24 <planetmaker> Snail_: did you btw, check what libpng-config --version give you wrt your grfcodec compilation issues?
12:48:38 <andythenorth> can make bundle_zip curl the devzone version and compare with local?
12:49:02 <Snail_> planetmaker: let me check now
12:49:26 <planetmaker> but rubi is right: it works w/o issues for both andy and me on the same OS revision
12:49:39 <Ammler> andythenorth: IRC does btw announce, if md5sum changes with new nml
12:49:45 <planetmaker> as such you have something special going on
12:50:31 <andythenorth> Ammler: does it raise a ticket?
12:50:33 <Snail_> planetmaker: running libpng-config gives me a list of options
12:50:37 <andythenorth> I'm in irc a lot but not always....
12:50:39 <Ammler> andythenorth: no, why should it
12:50:48 <planetmaker> libpng-config --version
12:51:01 <Ammler> the rebuild does not replace the older release, it does just check for differences
12:51:03 <planetmaker> but if it gives you a number of options that answers my question already
12:51:05 <andythenorth> Ammler: because if the md5 sum changes, the grf has to be treated as failing integrity
12:51:15 <andythenorth> compiled version is effectively untested
12:51:16 <planetmaker> I had it failing on updating libpng
12:51:21 <Ammler> the grf is still ok if build with that version on that time
12:51:34 <Snail_> should I update it now?
12:51:36 <andythenorth> if the md5 sum has changed, any testing of the grf is now invalid
12:51:37 <Ammler> if building fails, it will rise a ticket
12:51:50 <andythenorth> but manual testing is invalid...
12:51:59 <andythenorth> we have no idea if the grf performs as expected
12:51:59 <Ammler> just changing md5sum does usually not hurt
12:52:08 <andythenorth> it means the grf has changed
12:52:13 <andythenorth> therefore the code is untested ;)
12:52:15 <Ammler> as long as you distribute the origin
12:52:34 <Ammler> yes, but the changed grf is not distributed somewhere
12:52:52 <andythenorth> should I stop telling people to use bundles server for downloads?
12:53:15 <andythenorth> hmm - maybe my assumptions about rebuild are wrong
12:53:15 <Ammler> bundles does _not_ publish the rebuild
12:53:27 <andythenorth> that was my faulty assumption
12:53:32 <Ammler> and releases don't get rebuilds
12:53:37 <andythenorth> ok, all is well :)
12:53:46 <Ammler> we use that just for nightlies
12:55:06 <Ammler> but that is quite hidden
12:55:22 <andythenorth> why does MB struggle with ground sprites?
12:55:25 <andythenorth> I miss the issue
12:55:40 <planetmaker> Snail_: do you have any Makefile.local floating around?
12:55:52 <andythenorth> just specify an empty building sprite, or abuse the building sprite for one of the ground sprite layers
12:56:02 * andythenorth sees no gain to the GS / BS distinction
12:56:17 <andythenorth> sprites or childsprites, in a stack, is all is needed
12:56:41 <andythenorth> Snail_: what does 'gcc' in a terminal tell you?
12:56:58 <planetmaker> andythenorth: yes, I could have told that. But given the way he asks without giving any way he wants to go, I chose to answer as verbose as the question was
12:57:02 <Snail_> nope... I just searched for "makefile.local" files on my HD and I found none
12:57:23 <planetmaker> would need to be in the grfcodec dir where you build it
12:57:28 <planetmaker> but with capital M
12:57:47 <planetmaker> hm... do you have a FS where case matters, Snail_?
12:57:57 <planetmaker> andythenorth: do you have a file system where case matters?
12:58:19 <Snail_> I'm on OS X, I don't think it matters
12:58:23 <andythenorth> Disk Utility will know
12:58:29 <andythenorth> Snail_: it's a format option
12:58:37 <planetmaker> Snail_: that depends. It matters on mine and it's an install-time option to choose that or not
12:58:42 <Ammler> Snail_: you tested with a clean fresh checkout too?
12:58:58 <andythenorth> planetmaker: mine is not case sensitive
12:59:01 <andythenorth> Mac OS Extended (Journaled)
12:59:05 <Snail_> in that grfcodec dir, I found Makefile, Makefile.bundle and Makefile.local.sample
12:59:31 <Snail_> andythenorth: it tells me "i686-apple-darwin10-gcc-4.2.1: no input files"
12:59:39 <andythenorth> identical to mine
13:00:09 <planetmaker> so we use the same compiler, all three of us
13:01:09 <Snail_> also, in the dir for grfcodec r900 I also found three makefile's called with the same name
13:01:12 <Snail_> and that one compiles fine...
13:01:15 <andythenorth> something in the repo?
13:01:32 <andythenorth> try rm src/data?
13:01:32 <planetmaker> eh, you got separate dirs for them, Snail_?
13:01:53 <planetmaker> of course? I have one where I checkout, build and done
13:02:05 <Snail_> in each case, I go to the correct directory and run "make all"
13:02:08 <planetmaker> you don't get the source via mercurial?
13:02:13 <andythenorth> there's some folder that I have had to delete a couple of times to get grfcodec to build
13:02:21 <Snail_> I get it from the webpage
13:02:28 <planetmaker> can you give me URL?
13:02:36 <planetmaker> so you build from a tar ball?
13:03:28 <Ammler> that will be forwarded to grfcodec
13:03:30 <Snail_> that page is called nforenum, but it has the sources for grfcodec
13:03:49 <Ammler> planetmaker: it is like you call testing but there are none
13:03:58 <planetmaker> indeed it's re-directed to download-grfcodec
13:04:48 <planetmaker> ok, that trunk tar ball fails here
13:05:46 <planetmaker> hm, no, it doesn't fail really
13:06:03 <planetmaker> though it gives the error you quoted with /bin/sh: line 0: [: =: unary operator expected
13:06:18 <andythenorth> but in that dir, ./grfcodec works
13:06:43 <andythenorth> Snail_: is your priority to have a working compile, or a working grfcodec?
13:07:24 <planetmaker> but... that's the same source
13:07:27 <Snail_> andythenorth: yes it is, because the current nforenum doesn't support property 0A for railtypes
13:07:59 <Snail_> planetmaker: really? where is that downloadable from?
13:08:19 <planetmaker> andy posted the mirror link to where the link you gaves directs people from GB (and near there)
13:08:57 <planetmaker> I took the source from what you gave, snail, and it works, too
13:09:03 <andythenorth> Snail_: is it sorted or not? I can send you a bundle if you need it
13:09:17 <planetmaker> but it is identical. As our main server updates those mirrors...
13:09:19 <Snail_> now I tried to go back to the old version and it compi les...
13:09:41 <Snail_> the funny thing is that I didn't change anything there
13:09:50 <Snail_> anyway... now it looks like it works
13:09:57 <Snail_> sorry for the trouble guys :)
13:10:09 <planetmaker> you changed something: you ran make at least once before
13:10:09 <Ammler> the bin/sh error is here too
13:10:27 <Ammler> on tarball, not with hg checkout
13:10:30 <planetmaker> that might change *something*
13:10:48 <Snail_> well, I put the file I got from Andy's link in another dir and compiled it there
13:11:07 <Snail_> then I came back to the old r920 dir and tried to compile it there, too, where it didn't work yesterday
13:12:45 <planetmaker> would be nice to be able to *not* get it working again (maybe unzip the tar anew and try there or so)
13:12:57 <planetmaker> as it then might point to where the error lies
13:14:49 <Snail_> this is puzzling. I just re-downloaded it from the nforenum page, put it in a different directory, tried to compile it, ad it's compiling correctly
13:17:06 <Ammler> Snail_: aren't you able to close the ticket?
13:17:28 <Ammler> something with the devzone workflow is broken :-/
13:19:46 <Snail_> well, I wrote a message saying it's solved
13:20:57 <Ammler> yes, it's an issue with devzone workflow, author should be able to close/reject a ticket
13:21:26 * andythenorth will soon be offering sprite generator to retail consumers
13:21:34 <andythenorth> Super Discount on first purchase
13:23:24 <planetmaker> Ammler: only members of a project currently can close an issue
13:23:34 <planetmaker> non-members can't close
13:24:00 <planetmaker> they can set the status to resolved. That's fine. IMHO. Exactly as it is
13:25:02 <Ammler> planetmaker: there is additional section: "Additional transitions allowed when the user is the author"
13:25:45 <planetmaker> yep. And 'resolved' is good enough, imho
13:25:55 <planetmaker> and I was talking about that section specifically
13:26:22 <Ammler> ah, but that is not marked, closed is marked
13:28:00 <andythenorth> it's a hard balance between flexible classes for reuse by others, and lots of parameters and scaffolding :(
13:30:46 <Ammler> yes, and you think, resolved is marked?
13:31:35 <planetmaker> closed and rejected are the only inaccessible options for non-members for issues they opened themselves
13:32:46 <planetmaker> actually... I probably just interpret the wrong way
13:40:45 <andythenorth> I never figured out what ottd rev I should set in Bnnananas for FIRS
13:41:11 <planetmaker> the newgrf version
13:41:41 <planetmaker> thus the decimal equivalent of 0x0102008xxxxx
13:41:51 <planetmaker> where xxxxx is the hex equivalent of the svn rev
13:43:15 <planetmaker> actually.... make that 8 a 0
13:43:26 <andythenorth> and the svn rev is...? :)
13:43:29 <planetmaker> the 8 indicates release
13:45:17 <DorpsGek> andythenorth: 1.1.5, 1.2.0-RC1 | Website: *.openttd.org (translator: translator, server list: servers, wiki: wiki, patches & bug-reports: bugs, revision log: vcs, release info: finger) | Don't ask to ask, just ask | 'Latest' is not a valid version, ever | English only
13:45:26 <Ammler> or whatever openttd you have tested it
13:45:49 <Ammler> at least don't use a earlier
13:46:25 <planetmaker> I'd use the rev of openttd 1.2.0-RC1
13:46:48 <planetmaker> using latest nightly is too strong, Ammler
13:47:13 <Ammler> planetmaker: depends which openttd andy is using righ tnow
13:47:32 <planetmaker> r23971, andythenorth
13:47:41 <planetmaker> @ base 10 16 23971
13:48:01 <planetmaker> @base 16 10 010200005DA3
13:48:01 <DorpsGek> planetmaker: 1108101586339
13:48:12 <andythenorth> bananas doesn't use the svn rev?
13:48:20 <Ammler> planetmaker: that is like define /bin/sh as shell but always test with bash ;-)
13:48:28 <andythenorth> I've been doing it wrong in that case :P
13:48:34 *** Rhamphoryncus has joined #openttd
13:48:50 <Ammler> so instead test with other shells, define bash as shell ;-)
13:49:07 <planetmaker> andythenorth: yes ;-) As the svn rev does not give you the branch correctly. This way allows to distinguish
13:49:32 <planetmaker> Like 1.1.5 is much older than an August nightly. But the newgrf version of the August nightly is much higher than the one of 1.1.5
13:49:48 <planetmaker> as it's 010105... vs 010200...
13:50:13 <Ammler> and what is the release flag for?
13:50:14 <andythenorth> so 1108101586339
13:51:35 <planetmaker> Ammler: so that a release is higher than a non-release... and that you can identify a release
13:51:47 <planetmaker> but well. no other purpose than to tell 'release'
13:52:21 <Ammler> ah, as soon as 1.2 is released, you change to 1.3 for trunk
13:52:54 <planetmaker> yup. Already done
13:52:57 * andythenorth tests bananas now with r23969
13:53:03 <planetmaker> as soon as branching, Ammler
13:54:05 <Ammler> but then, he should use the release flag
13:54:40 <Ammler> oh well, doesn't matter
13:54:44 <planetmaker> just any nightly will do from that svn onward
13:55:27 *** Biolunar has joined #openttd
13:55:48 <Ammler> the release flag is basically a workaround, if someone would still use 1.2 for trunk after branching
13:56:07 <andythenorth> ok FIRS 0.7.1 doesn't appear with 32969
13:58:06 <planetmaker> how inconvenient. Trunk crashes on start :-(
13:58:40 <andythenorth> FIRS 0.7.1 doesn't appear now for ottd tip :P
13:59:08 <andythenorth> can anyone else confirm that it's not just me? :)
13:59:25 <andythenorth> 0.7.0 doesn't appear for me either though
14:01:21 <Ammler> andythenorth: no firs with rc1
14:01:33 <andythenorth> no 0.7.0 or 0.7.1
14:01:44 <andythenorth> but I didn't touch the version requirement for 0.7.0
14:01:50 <andythenorth> and people have been downloading it :P
14:02:40 <Ammler> andythenorth: you do not need this version thing on the bananas
14:02:50 <Rhamphoryncus> Hrm. I'm using 0.7.1 atm but the "check online content" only says 0.6.4. Using r23974
14:03:00 <Ammler> on bananas you use the menu
14:03:26 <andythenorth> Ammler: so what value do I set in the menu?
14:04:11 <andythenorth> andythenorth: bananas doesn't use the svn rev? <- planetmaker: andythenorth: yes As the svn rev does not give you the branch correctly. This way allows to distinguish
14:04:37 <Ammler> sorry then, but it seems to be an issue there then
14:04:46 <Ammler> maybe simply set release to 1.2
14:05:00 <andythenorth> I paste 1108101586339 into bananas rev field
14:05:11 <andythenorth> but it re-renders it as 1845493759
14:05:21 <andythenorth> this is very odd
14:05:53 <planetmaker> hm... maybe I added too many digits
14:06:32 <planetmaker> @base 16 10 12005da3
14:06:32 <DorpsGek> planetmaker: 302013859
14:06:37 <planetmaker> ^^ sorry andythenorth
14:06:47 <andythenorth> this seems quite painful bureaucracy :D
14:06:52 <planetmaker> not one byte but one nibble per version number
14:06:57 <andythenorth> one day...a bananas rebuild
14:08:23 <andythenorth> looks horrible :(
14:08:46 <andythenorth> ottd tip still won't see FIRS
14:09:24 <andythenorth> use the 'custom' field not nightly
14:09:36 <andythenorth> 23971 is not a nightly
14:09:36 <Ammler> I think, it would also work, if you would use svn revs only
14:09:43 <Ammler> as long as you do not mix
14:10:04 <Ammler> but as you used this extended version on 0.6.4 already
14:10:14 <Ammler> so that is why 0.7.0 is hidden
14:10:27 *** JVassie has joined #openttd
14:10:36 <andythenorth> it's not easy is it :)
14:10:53 *** George|2 has joined #openttd
14:10:53 *** George is now known as Guest4528
14:10:53 *** George|2 is now known as George
14:11:02 <Ammler> the need to add release version to the version check is very rare
14:11:24 <Ammler> planetmaker: can you tell a scenario, when this could matter?
14:12:47 <andythenorth> FIRS 0.7.1 is also available to r23969
14:13:06 <Ammler> yes, because trunk is 1.3
14:13:08 <planetmaker> sorry, I can't help you now, andythenorth. No working OpenTTD
14:13:51 <andythenorth> FIRS also works in r23969
14:13:55 <Ammler> but as I said, on trunk, only latest working nightly matters :-)
14:14:06 <andythenorth> this is thoroughly confusing tbh
14:14:50 <andythenorth> I might just set no min version
14:15:02 <andythenorth> ottd should handle this better
14:15:09 <Ammler> why do you care about trunk?
14:15:12 <andythenorth> downloading and reporting success for a grf it can't use is dumb
14:15:23 <andythenorth> Ammler: I don't know of any alternative
14:15:27 <andythenorth> what are the options?
14:15:40 <Ammler> not care about outdated trunk
14:16:17 <Ammler> only expeimantal people use those and those you need not to handle, they do that self
14:16:49 <andythenorth> ok, I have to set the min version to something so I've set it to 0
14:17:05 <Ammler> why not keep the version you had from pm?
14:17:23 <Ammler> it is fine for releases
14:17:48 <Ammler> you should care, that 1.1.5 does not download 0.7.1
14:18:15 <planetmaker> right... seems I messed with OpenGFX enough so that it crashes each OpenTTD I tried
14:18:21 <andythenorth> Ammler I don't understand how to do that
14:18:27 <andythenorth> is there a spec for this anywhere?
14:18:37 <Ammler> like pm said, you did alright
14:18:44 <andythenorth> newgrf spec has nothing on bananas
14:18:50 <Ammler> your only fault was to test with a very strange trunk revision
14:19:21 <andythenorth> ottd wiki says "Minimal Version: The minimal game version required to use your work. When selecting Nightly or Custom, fill in the nightly revision number or the name and version of the required patchpack in the adjacent field."
14:19:21 <Ammler> test with openttd 1.1.5 and 1.2.rc1
14:19:59 <Ammler> andythenorth: again, you setup that fine, you just tested wrongly :-P
14:20:13 <andythenorth> but there's no spec
14:20:26 <andythenorth> can't be wrong if there's no spec :P
14:20:41 <Ammler> you think, where does pm has it from?
14:21:54 <planetmaker> should be found there... but also that's what rb told me (long) ago
14:22:06 <planetmaker> unless I have a faulty memory
14:22:17 <planetmaker> not unheard of, though
14:22:30 <Ammler> planetmaker: there is no way to exclude older trunk but include current stable?
14:23:11 <Ammler> [15:14] <Ammler> 12005da3 < 13005da1
14:23:23 <planetmaker> you can set a min and a max version...
14:23:33 <andythenorth> I've set it back to 302013859
14:23:40 <andythenorth> I'm just going to trust it works
14:23:56 <andythenorth> I've got a demanding toddler and a cross wife, and I skipped lunch so far trying to fix this
14:24:03 <Ammler> planetmaker: how to setup a version which is backported from trunk?
14:24:11 <andythenorth> I don't have time to download stables
14:24:20 <andythenorth> if anyone else does, the effort would be appreciated
14:25:09 <Ammler> they just need to check, what version bananas offers for firs on 1.1.5
14:26:05 <Rhamphoryncus> I forgot how to workaround svn's daftness so I have all these tag dirs full of stables..
14:26:38 <planetmaker> lol, you did an svn co of the full svn repo?
14:26:48 <Ammler> Rhamphoryncus: we offer a hg repo with tags on hg.openttdcoop.org
14:27:02 <planetmaker> that's a long download
14:27:05 <Rhamphoryncus> Wasn't until after that I remembered how to do it properly
14:27:17 <Ammler> there is also a git pendant somewhere around
14:27:17 <Rhamphoryncus> even better: my network was acting up and I kept having to resume it
14:27:30 *** kkb110_ has joined #openttd
14:27:37 <Rhamphoryncus> My work is done via hg but for openttdcoop I need to track a svn revision
14:28:00 <Ammler> Rhamphoryncus: not with the repo from hg.openttdcoop.org
14:29:42 <Rhamphoryncus> Yes, the logs have version numbers but I don't want to search the logs for the right hg number
14:31:42 <andythenorth> that's probably success
14:32:04 <andythenorth> if 1.2.0-RC1 can see 0.7.1 that's definitely success imo
14:32:29 <Rhamphoryncus> and.. trunk has 0.7.1 again
14:32:38 <Ammler> Rhamphoryncus: check the repo again, openttdcoop.org != openttd.org
14:33:16 <Rhamphoryncus> Ammler: it looks like it's the same thing
14:33:31 <Ammler> it uses another method to convert svn repo
14:33:52 <Rhamphoryncus> So what is the tag for r23974?
14:34:06 <Ammler> why do you care about that?
14:34:14 <Rhamphoryncus> That's what I need to play openttdcoop
14:34:50 <Ammler> but that works with hg.openttd.org too
14:35:15 <Rhamphoryncus> I may try that later, but with my luck it'll set my monitor on fire *g*
14:35:30 <Ammler> hmm, might need hgsubversion installed
14:35:49 <Ammler> but then you can also simply grep for svn rev, check findversion.sh
14:36:03 <Rhamphoryncus> I'll likely try hg.openttdcoop.org later anyway, as the openttd.org one seems to not do merges.
14:36:36 <Rhamphoryncus> Not merging is bad for working on openttd :)
14:36:50 <Ammler> for working on it, you need trunk only :-)
14:37:17 <Rhamphoryncus> If both I and upstream touched a file it'd do nothing more than throw gvimdiff at me and laugh
14:37:27 <Rhamphoryncus> Which.. is worse than CVS, heh.
14:37:41 <Ammler> but for playing, you can have one repo and simply make BUNDLE_DIR="bundles/ps" etc.
14:37:57 <Rhamphoryncus> what's that do?
14:38:32 <Ammler> make bundle installs the bunlde to ./bundle, you can can change that via bundle_dir
14:38:45 <Rhamphoryncus> what's a bundle? :P
14:39:03 <Ammler> a copy from ./bin basically
14:40:05 <Rhamphoryncus> Well I'm pretty sure I won't be eliminating my other copies, heh
14:40:25 *** TGYoshi has joined #openttd
14:48:01 <andythenorth> Ammler: hg up "svn(23974)" <- :0 I need that
14:48:12 <andythenorth> I'm so bored of opening the log and reading svn revs by hand
14:48:39 <Ammler> check hg for revmap and filemap etc.
14:48:45 <Ammler> there are lots of such things
14:49:00 <Ammler> but the above might need hgsubversion installed
14:49:28 <andythenorth> I've been considering that anyway
14:49:36 <Rhamphoryncus> Oh, that's a given with hg and git: there is an obscure way to do what you want. Probably 6 obscure ways.
14:49:52 * andythenorth has no time to play with hgsubversion today though :P
14:50:16 <Ammler> include the dir in your hgrc [extensions]
14:53:04 <Ammler> I patched hgsubversion so it does add svn rev to the comment, but the hgrev is also part of meta data which you can see with hg log --debug
14:54:57 <Ammler> usually hgsubversion is needed only, if you want to push to a svn repo too, else in most cases hg convert suiffices
14:55:19 *** TWerkhoven[l] has joined #openttd
14:57:47 * Rhamphoryncus upgrades his alcohol trains to americans :D
15:02:53 <MNIM> ...but... americans have low alcohol tolerance!
15:03:53 <Rhamphoryncus> Yeah, and I'm canadian and don't drink anyway, so what do I care what it does to them?
15:15:47 <MNIM> well, your American alcohol trains are bound to be bad.
15:17:14 <Rhamphoryncus> They only look like American's. They're secretly carrying Gavan's Brew
15:40:49 *** theholyduck has joined #openttd
15:46:08 <Belugas> alcohol? somebody mentionned alcohol?
15:46:44 <MNIM> shoo shoo, nobody said anything of the kind. Now kindly move on, please.
15:46:52 * MNIM puts Belugas back on a train to poland.
15:48:16 * planetmaker wonders about the meaning of the word 'back'
15:48:31 * Belugas mumbles and grumbles and picks up his coffee mug with a disgusted look in his eyes
15:48:51 <Belugas> mmh... now that you mention it, planetmaker, me too!
15:57:37 * Rhamphoryncus fires Norris, and all his Norris friends!
16:00:35 <Elukka> wheee orbital rendezvous
16:00:38 <Elukka> kerbal space program is great fun
16:02:12 <Sacro> until you're 230Mm out with no chance of return
16:02:56 <Elukka> someone figured out how to merge save games
16:04:08 <Elukka> rendezvous is hard. much harder than flying to the moon or anything else i've done
16:04:22 <Elukka> it's easy enough to get to the same orbit, the hard part is getting there at the right time
16:10:00 <Rhamphoryncus> Aha, oops. I just realized I'm selling some excess vehicles.. during a recession
16:20:42 *** supermop has joined #openttd
16:26:17 <Rhamphoryncus> That looks comparable to what I use
16:27:15 <Rhamphoryncus> ~45 not including town names
16:29:09 <Rhamphoryncus> I count 75 for them
16:29:40 <supermop> planetmaker: why do you bother using my depot but not the sheds?
16:29:48 <supermop> its not that nice of a depot currently
16:30:13 <planetmaker> you think it's my newgrf list, supermop?
16:30:49 <planetmaker> it's the list of NewGRFs used by the person posting the screenshot with the question there
16:31:05 <planetmaker> reason for his trouble: NewGRF interference ;-)
16:31:17 <Rhamphoryncus> I've had a lot of that :/
16:31:30 <planetmaker> it's easy to arrive there
16:31:51 <Rhamphoryncus> newgrfs are complicated black-boxes
16:32:04 <planetmaker> they need be neither. But can be
16:32:45 <supermop> does oberhmer come here?
16:33:08 <supermop> he was working on helping me by making my depot grf a rail type thing
16:33:15 <Rhamphoryncus> Even in the best case.. I know NARS is supposed to have trains, but what and when? Only reason I know is someone put them on a wiki
16:33:26 <supermop> so that with nutracks, you would get all of the different styles
16:33:34 <planetmaker> Rhamphoryncus: that someone was the author himself
16:34:28 <planetmaker> supermop: better way is though to contact him via forum mail or e-mail
16:40:01 *** tokai|noir has joined #openttd
16:40:01 *** ChanServ sets mode: +v tokai|noir
16:49:45 <morph_> Can anyone try to join 95.143.38.18 (1.1.5-Stable) and check out if it's lagging? Because the stupid server lags for me :(
16:49:46 <sponge> Is it possible to order a vehicle to stop at all stations it passes or do I have to add a Go to order for every station?
16:51:08 <Rhamphoryncus> sponge: the default (not non-stop) stops in each station the vehicle happens to pass through
16:51:25 <Rhamphoryncus> But explicitly including each station is more reliable
16:52:33 <sponge> Rhamphoryncus: how do you mean more reliable?
16:52:51 <planetmaker> sponge: if the train must visit the station on the way, it will stop there if you do not use non-stop orders
16:52:54 <sponge> i have probably 100+ stations
16:53:24 <Rhamphoryncus> Any tweak to your network can alter where it goes. Even other traffic can. If it has a choice.
16:54:28 <sponge> Rhamphoryncus: circular line
16:54:47 <Rhamphoryncus> Alright, that's fine then
16:54:50 <sponge> although i noticed it tried switching direction on the one station i did specify, but specifying two of the most distant stations seems to work
16:54:55 <Rhamphoryncus> Are you using shared orders?
16:55:11 <sponge> got about 40 trains just going in a big loop
16:55:35 <sponge> should be ok then if i specify two stations
16:55:45 <Rhamphoryncus> There's also an option in advanced settings to make goto "stick", so you can click one, click another, etc
16:56:24 <morph_> Can a NewGrf lag your multiplayer server?
16:56:48 <Eddi|zuHause> "Max Payne (2001) removed from The Index"
16:58:23 <planetmaker> morph_: not on their own. But NewGRFs can use additional CPU
16:58:42 <planetmaker> but the lag could also be your computer locally. Which is barely able to keep up with the server
16:58:56 <planetmaker> check whether your openttd uses 100% of the single one core it runs on
17:06:21 <andythenorth> projectile baby sick
17:06:52 <Rhamphoryncus> andythenorth: if you'd like any help redesigning or reprogramming farmsup/engsup just let me know. They're killing it for me.
17:07:08 <andythenorth> there are tickets for them
17:07:32 <Rhamphoryncus> I know, we discussed it a bit before
17:07:38 <Rhamphoryncus> Now I'm offering labour
17:08:17 <andythenorth> planetmaker has a partial patch
17:09:26 <morph_> OpenTTD is using like <1% of CPU
17:09:40 <morph_> But there's like lag on my network server
17:09:55 <planetmaker> 1%? That's ... hard to believe tbh
17:09:56 * andythenorth is a bit interested in other projects right now, FIRS less so ;)
17:09:58 <morph_> It feels like low fps - trains move not smooth but like in jumps, if you know what i mean
17:10:24 <Rubidium> sounds like an unreliable network
17:11:13 <morph_> That's CPU usage and the command I'm using to start server
17:11:54 <morph_> I did a traceroute, yet no packet loss was found
17:12:09 <andythenorth> what sane cases are there for spritesheets where the height of each row varies?
17:12:38 <planetmaker> construction stages. growth stages
17:12:55 <planetmaker> loaded / unloaded vehicles
17:12:57 <andythenorth> that was the only case I could think of so far
17:13:14 <andythenorth> is it necessary for vehicles?
17:13:43 <planetmaker> strictly speaking necessary nowhere, if you crop sprites
17:14:25 <andythenorth> trying to eliminate parameters
17:14:37 <andythenorth> flexibility vs complexity :P
17:14:56 * andythenorth has never designed code for use by other people before :o
17:15:16 <Rhamphoryncus> "((random_bits << 2) & 1)" Isn't that shift backwards?
17:16:40 <Rubidium> depends what your expected result is; if you want a consistent result, this is the way to go
17:17:21 <Rubidium> if you want the result to be something else than 0, then it's wrong... unless it's a programming language where << is a ROL
17:18:01 <Rhamphoryncus> Rubidium: btw, have you tracked down the leader yet? I hear rumours that they're skulking around here somewhere.
17:19:55 <andythenorth> what's left to do?
17:20:14 <Rhamphoryncus> Rubidium: referring to my first conversation with you where I commented on not seeing any leadership
17:20:48 <andythenorth> this is an anarchy
17:20:58 <andythenorth> with Rubidium as chief anarchist
17:21:36 *** mahmoud has joined #openttd
17:21:39 <Rhamphoryncus> He didn't tell me that though. Let me ramble on to him as if he was just another peon
17:22:00 <andythenorth> do you need leaders in a meta-game?
17:22:11 <andythenorth> can't you just have players?
17:23:02 * andythenorth is playing a meta-game callec
17:23:25 <andythenorth> "how many features can you get added without writing any c++?"
17:23:39 *** sla_ro|master has joined #openttd
17:23:44 <andythenorth> dunno who's winning though, me or the game :P
17:24:19 <Rhamphoryncus> My first instinct was to say the game, simply by making you play.. but we are talking about C++
17:24:31 *** lmergen has joined #openttd
17:24:49 *** valhallasw has joined #openttd
17:25:16 <Rubidium> the question is, what is C++ and what isn't?
17:25:30 <Rubidium> that could interpreted as C++
17:25:46 <Rubidium> with the semi colon being on another line, or whatever
17:26:17 <Rubidium> but then basically all written characters can be argued to be valid C++
17:26:40 * Rhamphoryncus suddenly gets the feeling that Rubidium is a very clever and elaborate troll ;)
17:26:57 <andythenorth> I'd say if it's committed to the repo and compiles, that counts for n points ;)
17:27:35 <andythenorth> possibly I'm writing things that are valid c++ if put through a transform :P
17:27:59 <Rubidium> Rhamphoryncus: thank you for noticing I'm a supernatural being ;)
17:28:00 <andythenorth> transform is usually another human being
17:28:35 <Rubidium> although I'd love it if you would use the term jötunn
17:29:47 <Rhamphoryncus> Google seems to think that's a wow character
17:30:12 * andythenorth needs to write a sprite loader
17:30:30 <andythenorth> because sequencing bitmaps manually by (x,y,colour) might not be for everyone :P
17:31:01 <andythenorth> typing is better than click-move-click-choosecolour-move-click blah
17:31:42 <planetmaker> Rhamphoryncus: then you didn't look properly or only in the games' section
17:32:17 <Rhamphoryncus> planetmaker: there's only ~4 pages of results. I'm not picking a section
17:32:32 <planetmaker> I picked the first entry which was a wiki link...
17:32:54 <Rhamphoryncus> I suspect it's giving me a different set
17:33:03 <planetmaker> thus arguing with 4 pages doesn't cut it
17:33:12 *** HerzogDeXtEr1 has joined #openttd
17:33:26 <Rhamphoryncus> google .ca gives 7 pages
17:33:31 <Rhamphoryncus> First result is the wow character
17:33:55 * andythenorth will just spoil it for you
17:33:59 <andythenorth> first result for me
17:34:02 <planetmaker> :-( @ andythenorth
17:34:10 <_maddy> anyone want to start a cooperative multiplayer game with me?
17:34:29 <Rhamphoryncus> See, there's the problem. I trusted rubidium's spelling and google didn't correct it. Clever troll!
17:34:50 <planetmaker> where's the spelling incorrect?
17:35:13 <Rhamphoryncus> _maddy: I would but I have to iron my underwear
17:35:30 <Rhamphoryncus> oh damnit, it was just me that misread & misspelled it :'(
17:36:15 <andythenorth> do I need a pathname generator for spritesheets?
17:36:32 <_maddy> can multiplayer games be saved and continued later, even if not played on an actual server?
17:36:36 <andythenorth> or can people just concatenate foo+blah+cargo+'.png' ?
17:38:09 * andythenorth suspects these questions can't be answered (yet)
17:39:06 <Rhamphoryncus> _maddy: yes, you can go on a multiplayer server, save it locally, and replay it locally
17:40:02 <_maddy> Rhamphoryncus: but if I'm not on a actual server, but just host a multiplayer game on my client?
17:40:52 <Rhamphoryncus> I've never tried setting up multiplayer, sorry
17:42:06 <andythenorth> if I have a Pixa module, and some of my python classes are PixaFoo....should class Bar really be PixaBar (for consistency)?
17:42:16 *** TWerkhoven has joined #openttd
17:47:25 * planetmaker has his own MP servers ;-)
18:02:55 <Zuu> I once developed my own autopilot for running 3 hour long games on a small map, but when it was ready, I realized that I don't have time to support it by beeing an active admin. So it was never opened.
18:08:02 <sponge> can two cities actually merge and become one?
18:08:32 <sponge> if i destroy a building of town A, can town B add a new house there?
18:08:38 <sponge> and, in the long run take over an area
18:09:14 <Zuu> You can destroy all houses of town A, and have town B cover its area, but the town entity is still there.
18:09:41 <Eddi|zuHause> there was something about when one town built on a tile, and it is destroyed, no other town will ever claim that tile
18:09:58 *** break19 has joined #openttd
18:11:01 <sponge> guess il find out in a while
18:13:18 * andythenorth seeks synonyms for 'sequence'
18:13:35 <andythenorth> too much of my code is sequence = Sequence(sequence=trailersequence)
18:18:14 <Rhamphoryncus> I'd use seq for a variable name
18:18:23 *** lmergen has joined #openttd
18:19:29 <valhallasw> andythenorth: maybe take a more descriptive name, such as something that describes what it's a sequence of
18:20:02 <valhallasw> or do some more refactoring to reduce the numnber of sequences ;-)
18:20:18 <Rhamphoryncus> Or use more sets :D
18:20:19 <andythenorth> did that already :)
18:20:32 <andythenorth> fewer sequences ftw
18:23:01 <andythenorth> dictionary just tells me my terms are already correct :P
18:23:16 <andythenorth> what I'm doing is spookily close to DNA mapping (at a high level)
18:23:44 <andythenorth> I have the equivalent of chromosomes mapping to gene sequences
18:24:14 <andythenorth> calling my mappings 'chromosomes' is probably silly and pointlessly quirky
18:29:24 *** Graungaard has joined #openttd
18:30:53 <andythenorth> front_stakes_sequence_collection => front_stakes_sc ?
18:33:34 *** Alberth has joined #openttd
18:33:34 *** ChanServ sets mode: +o Alberth
18:38:52 <Alberth> pixa 2 finished! that was quick :)
18:40:11 <andythenorth> went really quickly once I had time to start :)
18:40:30 <andythenorth> it works, it makes sense
18:40:51 <andythenorth> it could use code review / refactoring
18:41:10 <andythenorth> I need to convert my other gestalts to use it, which might show up oversights in the design
18:42:52 <Alberth> can you point me to some Python code please?
18:43:05 <andythenorth> got BANDIT checked out?
18:44:35 <Alberth> I have it somewhere, I'll update and have a look
18:44:56 <andythenorth> I gain nothing by making an object where a dict really does just work?
18:45:09 <andythenorth> class / object /s
18:45:18 * andythenorth knows a dict is an object :P
18:45:29 <CIA-1> OpenTTD: translators * r23999 /trunk/src/lang/ (5 files):
18:45:29 <CIA-1> OpenTTD: -Update from WebTranslator v3.0:
18:45:29 <CIA-1> OpenTTD: belarusian - 1 changes by Wowanxm
18:45:29 <CIA-1> OpenTTD: french - 67 changes by OliTTD
18:45:30 <CIA-1> OpenTTD: greek - 7 changes by vitalblue
18:45:30 <CIA-1> OpenTTD: indonesian - 4 changes by prof
18:45:30 <CIA-1> OpenTTD: italian - 1 changes by lorenzodv
18:46:52 <Alberth> unless you want something 'extra' (or you want 'less'), no, using a standard data structure is simpler then :)
18:47:23 <Rhamphoryncus> andythenorth: dict is more robust too. A class has limitations on key names
18:47:40 <Rhamphoryncus> (which won't always show up, heh)
18:49:52 <andythenorth> coloursets are just dicts then
18:50:15 <andythenorth> it's tempting to create a class for every entity in the pipeline :P
18:51:42 *** morphium has joined #openttd
18:52:07 <morphium> This is driving me crazy. Can anyone, please please please, connect to "95.143.38.18" (stable 1.1.5) and check if the train running in circles IS LAGGING or not?
18:52:20 *** morphium is now known as Guest4552
18:52:24 <Guest4552> I can't find out whether that's my PC's problem, or my dedicated server's problem
18:52:57 <Alberth> look at the CPU load of both machines
18:53:26 <Alberth> if you have 1 core running at 100%, you have hit the ceiling :)
18:53:59 <Alberth> andythenorth: throw in a few interfaces, and a factory, and you can start as a java programmer :)
18:54:19 <Guest4552> I suppose I have to change my dedicated server or something
18:54:19 * andythenorth considered a factory earlier :|
18:54:24 <Guest4552> This just aint working :(
18:55:07 <Rhamphoryncus> andythenorth: if it grows a helper method then make it a class. Easily to have it still look like a dict and thus be interchangeable
18:55:10 <andythenorth> I'm working on the basis we want classes where we need a stable API (but the internals might change) and built-in entities otherwise
18:55:51 <Rhamphoryncus> the "d['name'] = 42" API is perfectly reasonable for custom classes in many cases
19:03:41 <bolli> i've got an unusual question
19:03:56 <bolli> is there any documentation about the scenario file formats?
19:04:05 <bolli> i can't find anything on the wiki or forums...
19:04:39 <michi_cc> Look into src/saveload/*.cpp
19:05:40 <planetmaker> bolli: it changes without notice. There are 167(?) different versions, as it changes for each new variable and so on
19:06:10 <michi_cc> No idea if somebody has written more anywhere, but the scenario format is identical to the savegame format.
19:07:04 <michi_cc> planetmaker: 174 + some unknown number of minor versions.
19:07:29 <planetmaker> I didn't look it up :-) But... I guess the order of magnitude defines effort
19:07:43 <michi_cc> Probably even more I guess, as adding a new chunk doesn't necessarily require a version bump.
19:10:57 <bolli_> bah... my horrid internet...
19:12:20 <bolli_> what i was hoping to do was create a revision control system for scenarios...
19:12:57 *** Chris_Booth[ph] has joined #openttd
19:15:54 <Alberth> bolli_: it is not a very stable format
19:16:27 <bolli_> is there a way to make a patch for multiplayer scenario editing?
19:16:44 <Alberth> a more useful direction would be to make a file format that is more independent of the savegame format
19:16:53 <planetmaker> you basically have to allow running the SE mode in multi player
19:17:11 <planetmaker> and make sure it's not advertised as game but as scenario editor
19:17:18 <planetmaker> acutally I'd consider that a nice thing
19:17:49 <bolli_> I'm not so sure how to go about that...
19:18:06 <bolli_> my strong point is web development not C++
19:18:55 <Alberth> bolli_: you see that wrong, your biggest area of potential growth is C++
19:21:24 <andythenorth> Rhamphoryncus: have you checked out FIRS?
19:21:42 *** frosch123 has joined #openttd
19:21:48 <Rhamphoryncus> And attempted to import it, such that hg qseries now reports "firs"
19:21:57 *** lmergen has joined #openttd
19:22:15 *** kkb110_ has joined #openttd
19:22:19 <Rhamphoryncus> planetmaker: more specifically?
19:22:24 <planetmaker> but in principle those are normal patches. So no special treatment needed
19:22:45 <planetmaker> the order is given by the naming in which they must be applied
19:23:01 <Rhamphoryncus> So.. manually copy the url of every patch file in the directory?
19:23:07 <planetmaker> you may familiarize yourself with how mercurial queues work
19:23:32 *** Chris_Booth[ph]_ has joined #openttd
19:23:32 <planetmaker> there's no standard way to download the files, no there isn't
19:23:40 <planetmaker> they were uploaded manually after all, too
19:23:48 *** Tintinfan has joined #openttd
19:23:52 <planetmaker> the whole patch queue dir
19:24:03 <planetmaker> you could try things like curl or wget
19:24:52 <Rhamphoryncus> I just need the 5 in that dir?
19:25:13 <bolli_> its vietnam, the road system works on biggest=right of way...
19:25:23 *** Chris_Booth[ph] has quit IRC
19:25:29 *** Chris_Booth[ph]_ is now known as Chris_Booth[ph]
19:25:30 <bolli_> he doesn't need to be able to see, people just have to get out of his way... :P
19:27:24 <planetmaker> Rhamphoryncus: you could put ALL those files in .hg/patches
19:27:42 <Rhamphoryncus> I just used hg qimport on each URL
19:27:43 <planetmaker> and enable mq in your hgrc
19:27:50 <Rhamphoryncus> Already enabled it
19:28:09 <Alberth> just edit the 'series' file :p
19:28:12 <planetmaker> then, if you had them in .hg/patches (relative to FIRS root), you could use hg qpush
19:28:37 <Rhamphoryncus> .. how would pushing help? Especially once I already downloaded them?
19:29:20 *** Chris_Booth[ph] has quit IRC
19:29:25 <planetmaker> learn about mq :-)
19:29:34 <planetmaker> if you imported them already, of course not
19:30:25 <Rhamphoryncus> But .hg/patches was useful. I was able to delete the garbage "patch" it created when I previous attempted to import the dir
19:30:33 <Rhamphoryncus> Obviously, but they should be similar in concept
19:30:59 <Rhamphoryncus> oh right, they're not
19:31:10 <Alberth> not by a long shot :)
19:31:43 <Rhamphoryncus> Yeah, I don't exactly feel responsible for that mistake
19:33:04 <cyph3r> SMS co ti přišla odpovědí?
19:49:33 <andythenorth> Alberth: so my questions include...
19:49:47 <andythenorth> - should I rename Spritesheet to PixaSpritesheet?
19:49:57 <andythenorth> - should I create a class for colourset (probably overkill)
19:51:17 <andythenorth> - can you think of an easy way for authors to specify a colourset when a sequence uses numbers, not variables for colour indexes
20:02:46 <Alberth> andythenorth: what is a colourset in the code?
20:03:08 <Alberth> oh, the cc_1/2 perhaps
20:03:22 <andythenorth> type is dict, purpose is to match colours (or vars) <-> indexes
20:03:42 <andythenorth> I considered making it a type of transform, but it deserves a special case
20:04:48 *** supermop has joined #openttd
20:04:56 <Alberth> I don't understand the question yet
20:06:52 <andythenorth> so the pixarender method expects a colourset
20:07:55 <Alberth> 'cc_1' : dict ( deck_colour = 115, company_colour = 202, ), <-- this thing
20:07:59 <andythenorth> which is just a dict, e.g cc_1 in coloursets
20:08:37 <Alberth> I was wondering why they are in a dict :)
20:08:48 <andythenorth> it was a convenient way to get it working ;)
20:09:14 <andythenorth> but I wonder if they should be a Pixa class
20:09:20 <andythenorth> as the renderer kind of depends on them
20:09:36 <andythenorth> currently my code suggest you could pass colourset=None, but you can't :P
20:10:06 <andythenorth> they compose a mapping of vars:colour indexes, with an id which is used for filenames
20:11:21 *** kkb110_ has joined #openttd
20:12:27 <Alberth> yeah, it would seem better to split those concepts, imho
20:13:58 <andythenorth> no inspiration struck me yet for it though :P
20:14:31 <andythenorth> coloursets or such also need to be able to remap colours for a sequence that's loaded from a bitmap
20:20:19 <andythenorth> I'm never sure about whitespace in args etc
20:20:26 <andythenorth> PEP-8 kind of doesn't guide much
20:22:05 <Alberth> this is mostly trailing white-space removal :)
20:22:19 <andythenorth> I can't get the patch to apply with p1 or -0
20:22:52 <Alberth> from the root of bandit
20:23:20 <Alberth> a/misc/pixel_generator/gestalts/flat_trailer.py <- -p1 means "drop "a/"
20:24:17 <Alberth> and if you are in 'pixel_generator', -p3 should work :)
20:26:04 <Alberth> ie it is just the number of directories stripped from the start
20:26:39 <Alberth> I thought you wanted to compute the numbers before going to the renderer
20:28:24 <andythenorth> I found reasons that might not work
20:28:41 <andythenorth> can't remember what
20:28:44 *** appe is now known as NGC3982
20:28:53 <andythenorth> probably could have found a way around it, but this method seems to work fine
20:31:40 <Alberth> self.id = id doesn't seem very much used :) line 111 of flat_trailer.py
20:31:59 <andythenorth> think it's redundant
20:33:41 <andythenorth> ah, I remember why I don't apply the transforms earlier
20:34:13 <andythenorth> if colours are set by vars, the transforms will fail unless the colourset has been applied first
20:34:49 * Zuu_ sends a great thank to the swap file of Vim. (managed to accidently hit the powerswitch of my computer :-p )
20:35:11 <Alberth> Bram does accept donations :)
20:35:28 <planetmaker> vim is great. If you have xterm only
20:35:32 <planetmaker> or for a quick edit
20:35:57 <Alberth> it is so much more useful than any other editor :)
20:36:05 <bolli_> i just find it impossible to use...
20:36:18 <Alberth> bolli_: it is, for the first 2 months :)
20:36:33 <bolli_> well, i don't have 2 months worth of patience :P
20:36:36 <Zuu_> Also good for coding programs up to the size when you can keep their structure in mind.
20:36:50 <Alberth> but after that you are so much more productive
20:37:10 <andythenorth> I could push the colourset into the sequence earlier in the pipeline
20:37:14 <andythenorth> not sure much is gained though
20:38:10 <Alberth> andythenorth: you indeed need to apply the colourset before transforming as far as I can see, but that is the case no matter when/where you apply the transformations
20:38:47 <Alberth> Zuu_: at an appropiate level of abstraction that always works :)
20:39:14 <Alberth> andythenorth: but what was the question now?
20:39:26 <Zuu> So.., now I can pick up my work again with a fleet management GS :-)
20:39:54 <planetmaker> fleet managment GS?
20:40:01 <andythenorth> Alberth: I thought you wanted to compute the numbers before going to the renderer <- you made me doubt my current method ;)
20:40:52 <Zuu> A GS that will add or remove a train/truck for your line to adopt with the production changes when you are busy working on the network.
20:40:53 <Alberth> andythenorth: I just ask questions :p
20:41:21 <andythenorth> I'm going to leave it as is for now - might change it later though
20:41:30 <andythenorth> I'll try the coloursets patch first
20:41:35 <Alberth> Zuu: for all companies? great, finally the game can play itself :)
20:41:39 <Alberth> andythenorth: ok, bye
20:42:02 <Zuu> Alberth: Yes, for all companies. But only for stations which has a special key in their name.
20:42:21 * Alberth ponders making a trivial AI program :p
20:43:21 <andythenorth> Alberth: make an AI generator ;)
20:43:26 <Alberth> pondering is much nicer than actually doing it :)
20:43:46 <andythenorth> sometimes the pondering is the hard part :(
20:43:55 <andythenorth> code is relatively easy, it has rules
20:43:57 <Alberth> andy I am building enough generator at work atm
20:44:06 <andythenorth> generator generator!
20:44:29 <Alberth> it does use meta-programming yes :D
20:44:48 <Alberth> and runs into Java limitations w.r.t. generics :)
20:46:29 *** peteris has joined #openttd
20:47:29 *** George is now known as Guest4561
20:48:03 <Rubidium> you mean not being able to add to Collection<? extends foo> ? ;)
20:49:09 <Alberth> more that new T[n] with T a type parameter not working :)
20:50:04 <Alberth> of course, that is because "Java generics are superior to C++ templates", according to oracle :)
20:51:12 <bolli_> i have a noobish question now...
20:51:32 <bolli_> i'm compiling in MS C++2008 and i'm getting errors...
20:51:38 <andythenorth> Alberth: colourset patch makes sense
20:51:40 <bolli_> the source is straight from SVN
20:52:00 <bolli_> and i've included the bits that need it
20:54:39 <Rubidium> could you be more specific about "errors"
20:55:05 <bolli_> mostly missing files...
20:55:14 <bolli_> c:\users\sam\documents\openttd\source\s\src\language.h(17) : fatal error C1083: Cannot open include file: 'unicode/coll.h': No such file or directory
20:55:22 <bolli_> ..\src\heightmap.cpp(39) : fatal error C1083: Cannot open include file: 'png.h': No such file or directory
20:55:43 <Rubidium> sound like you haven't installed the "useful" package with all build dependencies
20:57:06 <bolli_> i've added the items into options -> VC++directories
20:58:45 <Alberth> andythenorth: I hve not tested it, but you can have an empty dict now
20:59:16 <Rubidium> bolli_: are you sure you set the right folders for the incude files?
20:59:35 <bolli_> its probably me being stupid...
21:00:01 <Alberth> hmm, does "unload and leave empty" refuse to unload cargo at the station it was loaded from?
21:00:03 <Rubidium> I can't really give in depth help as I'm not using Windows
21:00:04 <andythenorth> well empty dict appears to fail correctly
21:00:30 <andythenorth> i.e. vars in my sequence aren't substituted, so that blows up
21:00:59 <bolli_> i prefer using Linux for jobs like this :p
21:01:11 <bolli_> its just not always useful for playing it on windows...
21:01:50 <Alberth> andythenorth: well, you could do that before as well, of course, but you had two forms of 'colourset' which caused some confusion.
21:04:02 <andythenorth> I think it's time to convert the other gestalts
21:04:11 <andythenorth> I'm running out of 'todo' items :)
21:10:23 <Alberth> andythenorth: I still need people / employees / benches / trashbins / trash / shops / stalls / rollercoaster sprites :)
21:11:37 <andythenorth> mostly generatable :P
21:11:49 <andythenorth> open theme park? :D
21:12:19 <andythenorth> "open lotus turbo esprit" would be nice
21:14:49 <andythenorth> Alberth: got example art anywhere?
21:16:05 <Alberth> www.freerct.org, then redirect to the googlecode project trunk/sprites_src for what exists in the project
21:16:21 <Alberth> otherwise any RCT site :)
21:18:25 <andythenorth> Alberth: looks like it's using CGI models?
21:21:54 <Elukka> heh, i've heard a bunch of people in various places say they wish there was an openttd equivalent for RCT
21:22:11 <andythenorth> it could be time to get a 3D printer :P
21:22:40 <Alberth> Elukka: there are also people saying there should be an open source variant of locomotion
21:22:59 <Alberth> Elukka: point is that 'saying' is cheap. 'doing' is what counts.
21:23:02 * andythenorth wonders if lomo sprites are 32bpp?
21:23:20 <Elukka> i was saying there would be an audience for it
21:23:25 <Elukka> people would like to play it
21:23:28 * andythenorth goes back to doing - sprite generator
21:23:30 <Alberth> andythenorth: the engine is based on RCT2, so I'd doubt it
21:23:38 <andythenorth> Elukka: you could be a sprite generator customer!
21:23:44 * andythenorth is out to make a sale
21:24:05 <Alberth> Elukka: I would too, but there was none, so I ended up writing one first :p
21:24:17 * andythenorth needs a bitmap loader class
21:24:31 <andythenorth> which is probably vanilla PIL, apart from the bit to make it a sequence
21:25:00 <Alberth> ideas to mark (0,0) ?
21:25:38 <andythenorth> I'll see if it's actually a problem
21:25:56 <andythenorth> I don't mark 0,0 for a sequence either
21:25:58 <andythenorth> it might just work
21:26:13 <andythenorth> I'm assuming I'll crop tightly first
21:26:44 <andythenorth> RCT graphics are odd, some are really pretty, some are muddy
21:26:56 <andythenorth> whereas lomo is just muddy
21:26:56 <Alberth> andythenorth: in the sequence you actually give the coords yourself :)
21:27:07 <Elukka> lomo has an ugly palette
21:27:15 <andythenorth> I'm assuming 0,0 is the top-left of the cropped image
21:27:17 <Elukka> i like how RCT looks though
21:27:27 <andythenorth> although that conflicts with how the renderer currently operates :|
21:27:34 <andythenorth> ach - I'll try it and see
21:27:42 <andythenorth> pondering ~= procrastination
21:27:44 <Alberth> andythenorth: perhaps it's a mix of manual and generated pixels
21:29:34 <Rhamphoryncus> nmlc: "sprites/nml/industries/../templates/check_availability.pnml", line 32: Unrecognized identifier 'CB_RESULT_IND_PROBABILITY_FROM_PROPERTY' encountered
21:29:44 <Rhamphoryncus> That's when I run make in firs
21:30:05 <Elukka> oh yeah some objects do look muddy
21:30:19 <Rhamphoryncus> I think I grabbed the stable
21:31:32 <Alberth> Rhamphoryncus: nml is moving quite fast
21:31:35 <andythenorth> Rhamphoryncus: FIRS trunk tip builds for me with nml tip
21:32:38 <andythenorth> is it RCT where you can kill the coaster riders?
21:32:42 <andythenorth> or was that theme park?
21:33:09 <TWerkhoven[l]> you mean drop in the water?
21:33:10 <Alberth> you can crash coasters
21:33:32 <Alberth> oh, dropping in the water is also deadly
21:33:36 <andythenorth> go too fast, they fall out, iirc?
21:33:46 <TWerkhoven[l]> think that was theme park
21:34:10 <Alberth> never seen that, but my coasters were always 'nice' to the public :)
21:34:41 <Alberth> and it does not sound like something I want to implement :p
21:35:50 <Rhamphoryncus> That time it built. Thanks andy, Alberth
21:37:10 <andythenorth> the conversion of the 3D models to paletted sprites has not always gone well
21:37:16 <andythenorth> I guess that's why some are muddy
21:37:41 <Rhamphoryncus> Wait, RCT is still sprite-based?
21:37:47 <andythenorth> they're dithered from a limited palette, and are losing contrast
21:37:53 <andythenorth> even lomo is sprite based isn't it?
21:38:00 <Rhamphoryncus> That's.. very interesting.
21:39:00 <Alberth> Rhamphoryncus: 1 and 2 are
21:39:09 <andythenorth> Alberth: does openrct have full 32bpp support?
21:39:23 <andythenorth> seems that nice renders might be the way to go...
21:39:43 <andythenorth> especially wrt capturing the original style
21:40:02 <Alberth> andythenorth: no idea, I program freerct :D but currently just 8bpp, just like the rct's
21:40:12 <andythenorth> open / free /s :P
21:40:34 <andythenorth> anyway /me is procrastinating and should write code
21:40:42 <andythenorth> I found todo items :|
21:40:51 <Rhamphoryncus> rct is 8-bit too? o.O
21:41:00 <andythenorth> are global constants allowed in python?
21:41:11 <Rhamphoryncus> Geeze, I just assumed it was ugly because they chose that style *g*
21:41:31 <Rhamphoryncus> andythenorth: yes, but they're really just module attributes. Not implicitly shared between modules
21:41:32 <Alberth> andythenorth: 32bpp is definitely an option
21:42:12 <andythenorth> so I can't define a global in module a, import module a to module b and have my global just appear?
21:42:19 <andythenorth> I have to use a.myglobal
21:42:31 <Alberth> Rhamphoryncus: CS is stuck to the same engine at every game he made
21:42:47 <Rhamphoryncus> You can do "from a import myglobal", which is really just the same as "import a; myglobal = a.myglobal"
21:43:32 <Alberth> depends on the import, but generally it is recommended to include the module name when referencing imported symbols
21:43:45 <andythenorth> Rhamphoryncus: in which case I don't need to declare is global :P
21:43:55 <andythenorth> I'll reference the module name
21:44:13 <Alberth> Rhamphoryncus: the ttdx / rct / lomo engines are all the same, just with some evolutionary (small) steps
21:44:23 <Rhamphoryncus> andythenorth: if you actually mean a constant then this is fine. If it's something like an int then you're making copies of the variable and modifying one won't update the other
21:44:31 <Rhamphoryncus> Alberth: *nods*
21:44:34 <andythenorth> I do mean a constant
21:44:47 <Alberth> constants do not exist
21:45:00 <Alberth> just variables that you never assign
21:45:01 <andythenorth> I don't mean a constant
21:45:19 <andythenorth> SPRITEROW_HEIGHT = 40 is same everywhere
21:45:33 <andythenorth> defining it in every gestalt file seems obviously silly :)
21:45:33 <Alberth> you can do from a import * which dumps all global names of a into the module doing the import
21:45:48 <Zuu> If you clone a vehicle that belongs to a group, is the clone automatically added to the same group?
21:46:07 <andythenorth> I'm just using it now as common.SPRITEROW_HEIGHT in my code
21:46:11 <andythenorth> and importing common
21:46:19 <andythenorth> then it's obvious where it originates
21:46:25 <Alberth> that is much saner w.r.t. maintenance
21:46:45 <Rhamphoryncus> Yeah, you can use "from a import *". That's generally frowned upon because it makes it hard to track down where a variable comes from
21:46:56 <Rhamphoryncus> But if you had, say, a constants module, it wouldn't be as bad
21:47:10 <Alberth> not to mention overwriting other names without warning etc
21:47:15 <Rhamphoryncus> (also less likely to produce silent name conflicts)
21:47:33 <andythenorth> python mostly tells you the answer once you try something
21:48:28 <Zuu> Great it works. Then I don't have to figure out where the API function to add a vehicle to a group is hidden.
21:48:45 <Rhamphoryncus> Hmm. Learning FIRS/nml by browsing a random file feels like learning how a car works by examining a seat buckle
21:49:16 <andythenorth> Rhamphoryncus: it's an order of magnitude less complex than openttd ;)
21:49:44 <Alberth> but there are more random files to browse :p
21:50:09 *** Chris_Booth has joined #openttd
21:50:25 <andythenorth> Rhamphoryncus: currently there is a small deficit - I don't know how FIRS works either ;)
21:51:29 <andythenorth> I can tell you the structure of the set though
21:51:47 <Rhamphoryncus> The use of CPP spells doom
21:51:52 <andythenorth> and I can reverse engineer the code
21:51:55 <Rhamphoryncus> (doomy doomy doom)
21:51:59 <andythenorth> I just don't *know* how it works
21:52:21 <Rhamphoryncus> You take one piece, figure it out, make the changes, then forget again?
21:52:39 <andythenorth> I make the changes I can make, release, and wait for other people to work on it ;)
21:52:52 <andythenorth> I keep intending to learn how it works
21:53:00 * Rhamphoryncus feels like he took the bait
21:53:05 <andythenorth> but the road to hell is paved with good intentions etc :P
21:54:07 <Yexo> Rhamphoryncus: if you want to learn NML than FIRS is not a good example
21:54:20 <Yexo> it's a very cool project, but not easy to learn from
21:54:22 <Rhamphoryncus> Yexo: I want to improve FIRS
21:54:39 <Yexo> in that case you should still start with a general NML tutorial
21:54:45 <Yexo> get a feel for how it all works, than dive in the FIRS code
21:54:57 <frosch123> oh, just noticed... i am featured on google today :)
21:55:35 <andythenorth> lol at frosch123 :)
21:55:56 <andythenorth> Rhamphoryncus: the nml aspects of FIRS are straightforward
21:56:13 <andythenorth> and you're working with industry production code which is also straightforward
21:56:48 <andythenorth> you know how CPP variadic macros work?
21:57:04 <Rhamphoryncus> Well I have used and written t hem.
21:57:13 <andythenorth> ok, you're all set then
21:57:22 <Rhamphoryncus> Didn't say I knew how they work ;)
21:58:21 <Rhamphoryncus> Wait, I can't copy you, I have to understand it myself!
21:58:29 <andythenorth> as long as you recognise them, it's fine
21:58:58 <andythenorth> you kind of need to know the newgrf industry spec
21:59:03 <andythenorth> but that's available
21:59:12 <andythenorth> specifically you want to read about the production cb
22:00:37 <andythenorth> then you just need check_primary_production_level.pnml
22:00:44 <andythenorth> and produce_primary.pnml
22:00:55 <planetmaker> uhm... for supplies?
22:02:11 <planetmaker> production of a secondary depends surely on its input
22:02:27 <planetmaker> but check_primary_production_level is geared towards primary industries
22:02:34 <planetmaker> as is produce_primary
22:02:50 <andythenorth> where are supplies handled now?
22:03:02 <planetmaker> hm... sorry, I'm confused
22:03:15 <planetmaker> You#re of course right: Primaries DO consume the supplies
22:04:45 <andythenorth> Rhamphoryncus: you'll need to know about the various FIRS parameters for primaries as well
22:05:35 <andythenorth> is there ever a case for using the windows palette?
22:05:47 <Rhamphoryncus> This is nml with CPP macros wrapped around it? So I'll need to ask here about the syntax, not look in the nml tutorial?
22:06:07 <planetmaker> andythenorth: there's no case (anymore) for it
22:06:22 <planetmaker> it's purely hysterical raisins
22:06:24 <Yexo> the cpp preprocessor is run over it first to create firs.nml, which is then parsed by the nml compiler to firs.grf
22:06:38 <andythenorth> the brittrains project appears to be using windows palette
22:06:43 <Yexo> andythenorth: ttdpatch users with a windows base set
22:06:49 <andythenorth> I might give them the good news
22:06:57 <Yexo> but nml 0.3 is not compatible with that
22:07:13 <planetmaker> it's not compatible with that in many ways
22:07:15 <Yexo> or rather: ttdpatch is not compatible with the grfs produced by nml ;P
22:07:37 <andythenorth> DOS palette is now default for ottd?
22:07:55 <planetmaker> internally all windows paletted newgrfs are converted to DOS prior to use
22:08:57 <Yexo> actually that's not 100% true
22:09:09 <Yexo> openttd's internal palette is neither the newgrf dos palette nor the newgrf windows palette
22:09:41 <frosch123> where do we still differ?
22:09:50 <Yexo> so yes, that looks like the dos palette
22:09:56 <Yexo> frosch123: oh, it's now completely the same?
22:10:05 <frosch123> i think we are completely dos
22:10:32 <frosch123> except maybe for the undefined bits
22:10:52 <frosch123> i.e. dos has only 6 bits per colour channel, while we use 8
22:11:13 <frosch123> so the hue is slightly different for some
22:11:17 <frosch123> including the grey scale
22:15:49 <andythenorth> Rhamphoryncus: for the code you're trying to change, you can apply 'iceberg'
22:15:56 <Rhamphoryncus> NFO is not a programming language because it's hex editing? Errr..
22:15:57 <andythenorth> 90% of FIRS you don't need to worry about ;)
22:16:26 <Yexo> Rhamphoryncus: who says that?
22:16:29 <andythenorth> ^^ where did you get that witty line?
22:17:13 <andythenorth> foobar is mostly write, he just slipped up there for once ;)
22:17:56 <Rhamphoryncus> Speaking of hex editing, I should report a bug against NARS: the grasshopper is bugged. It doesn't have the 65535 HP I remember from railroad tycoon ;)
22:18:12 <andythenorth> see what pikka says about that :P
22:18:31 * andythenorth has procrastinated brilliantly this evening
22:18:34 <Rhamphoryncus> I can't be the only one that did that, can I?
22:19:18 <Rubidium> andythenorth: there are many more in this channel that have succeeded better in procrastination
22:19:30 <Rubidium> they even postponed writing something here
22:19:44 <Rubidium> *if* only they would have postponed joining the channel ;)
22:20:15 <andythenorth> how hard can this code business be anyway?
22:29:52 <frosch123> i guess most in here are busy with porting ottd to a web browser
22:31:47 <Rhamphoryncus> frosch123: they have to port it to pascal first, then write a pascal to python compiler, then use pypy to compile that to javascript
22:32:01 <Rhamphoryncus> (I think pypy dropped that feature, so it'll have to be revived)
22:38:25 <andythenorth> is this evil? 'tank_colour#-1' < then split('#') to get a var which evaluates to a number, the second part of the split being then added to the number
22:38:48 <andythenorth> so (-4, 4, 'tank_colour', -2),
22:39:00 <andythenorth> where the last element is the shift
22:39:11 <andythenorth> and it's optional
22:39:23 <Rhamphoryncus> Evil will always triumph because good is dumb
22:39:25 * andythenorth hates lucky dip tuples and lists
22:39:34 <andythenorth> tell me what those numbers mean
22:39:39 <andythenorth> or where I should document it :P
22:39:44 <Rhamphoryncus> I'd use a class
22:39:56 <Rhamphoryncus> Foo('tank_colour#-1') is okay
22:40:40 <Rhamphoryncus> Of course Foo('tank_colour', -1) works just as well
22:40:51 <andythenorth> you took the words out of my mouth :P
22:41:10 <andythenorth> this definitely smells (-4, 4, 'tank_colour', -2)
22:41:13 <andythenorth> and so would a dict
22:41:19 <Rhamphoryncus> And if tank_colour is meant to refer to a variable, not a constant, then have it stash the name 'tank_colour' and look it up later
22:41:33 <andythenorth> that would help a lot actually
22:42:32 <Rhamphoryncus> If it's a colour *reference* then treat it as a reference
22:42:49 <andythenorth> it's a var, evaluated at render time
22:43:01 <andythenorth> hence why it's string, not var :P
22:43:11 <andythenorth> I use it as a dict key
22:43:31 <Rhamphoryncus> the name? With the -2?
22:44:01 <andythenorth> the -2 is applied to whatever the result is from the lookup
22:44:16 <andythenorth> colourset['tank_colour'] - 2
22:44:19 <Rhamphoryncus> I was going to say it's easy to make a class usable as a dict key
22:45:14 <andythenorth> that I had to do this also smelt funny
22:45:15 <andythenorth> colourshift = i[3]
22:45:29 <andythenorth> when evaluating tuples that may not have the shift value
22:45:46 <andythenorth> an object can do it cleanly
22:46:06 <andythenorth> tomorrow's problem
22:46:14 <andythenorth> or today's problem, deferred :p
22:46:14 <valhallasw> andythenorth: the 'pythonic' version would be try: colourshift = i[3] except IndexError: colourshift = 0
22:46:41 <Rhamphoryncus> Eh, exceptions are overrated for that stuff
22:46:52 <valhallasw> yeah, I don't find it much clearer
22:46:55 <andythenorth> by moving this to an object, I can also get rid of:
22:46:56 <andythenorth> point.colour + 1
22:47:02 <Rhamphoryncus> It's easier to ask forgiveness than ask permission.. except when permission is easier.
22:47:08 <valhallasw> if would be nice to have the dict version
22:47:19 <valhallasw> {}.get('a', 'b') = 'b'
22:47:34 <valhallasw> or getdefault, I always forget which one is which
22:47:42 <andythenorth> setdefault is my favourite
22:47:56 <andythenorth> 'the oddly named method'
22:48:08 <valhallasw> ah yea, that one. list = data.setdefault('key', [])
22:48:10 <andythenorth> as it doesn't actually set anything but the value returned
22:48:32 <andythenorth> dict remains unmodified
22:48:43 <valhallasw> if data['key'] exists, it's equivalent to list = data['key']
22:48:59 <valhallasw> if it doesn't exist, it's equivalent to list = data['key'] = <given value>
23:31:20 <DorpsGek> Zuu: Don't ask to ask, just ask
23:31:38 *** KenjiE20 has joined #openttd
23:33:05 <Nat_aS> sorry bout that, anyways, how does the upgrade order in the veichile list work?
23:33:15 <Nat_aS> when does it take effect?
23:33:42 <Nat_aS> I like to feel like I am doing things myself, but doing it by hand is really complicated
23:34:08 <Zuu> If you have added a replacement order, you can then select to send all vehicles to depot if you want to enforce it.
23:34:36 <Nat_aS> is there a minimum ballance for it to work?
23:34:39 <Zuu> However, all vehicles that are in subject for replacement will visit their depot as soon as it gets there.
23:34:42 <Nat_aS> like it wont spend more than half your money?
23:35:08 <Zuu> So usually it is not a good idea to force all vehicles to depot as that will include also vehicles with no replace order.
23:35:16 <Zuu> Yes, there is one in advanced settings.
23:55:00 <Rhamphoryncus> planetmaker: Can I bounce something off you regarding farmsup/engsup?
continue to next day ⏵
