IRC logs for #openttd on OFTC at 2010-05-14
⏴ go to previous day
00:00:57 <SmatZ> and I am almost sure "Unloaded" cargo had rating too
00:03:23 <Eddi|zuHause> can you check around r14417?
00:04:04 <Eddi|zuHause> that's the last merge from trunk in the version of cargodest that i played
00:04:27 <Eddi|zuHause> cargodest version was 2ee9ce8ab5fe
00:04:58 <Eddi|zuHause> or more specifically h2ee9ce8aM
00:05:29 <Eddi|zuHause> where M was mostly the timetable patch
00:32:23 *** roboboy has joined #openttd
00:37:20 *** last_evolution has quit IRC
00:38:59 *** Devedse has joined #openttd
00:54:56 *** HerzogDeXtEr has joined #openttd
01:32:28 *** robotboy has joined #openttd
02:23:43 *** Keyboard_Warrior has quit IRC
02:51:17 *** robotboy has joined #openttd
02:51:18 *** amalloy1 has joined #openttd
03:51:20 *** r0b0tb0y has joined #openttd
04:56:23 *** Eddi|zuHause has joined #openttd
05:31:35 *** roboboy has joined #openttd
07:28:51 *** einKarl has joined #openttd
07:31:59 *** JVassie has joined #openttd
07:38:00 *** Alberth has joined #openttd
07:51:52 *** last_evolution has joined #openttd
08:15:17 *** valhallasw has joined #openttd
08:30:24 *** oskari89 has joined #openttd
08:31:57 *** Progman has joined #openttd
08:45:34 *** Devedse has joined #openttd
08:57:14 *** ChanServ sets mode: +v tokai
08:58:52 *** roboboy has joined #openttd
09:08:10 *** HerzogDeXtEr1 has joined #openttd
09:13:32 *** Brianetta has joined #openttd
09:16:48 <iconiK> What is the URL of the Mercurial repository?
09:18:55 <iconiK> Lovely, thanks yorick!
09:29:09 *** |Jeroen| has joined #openttd
09:33:28 <ccfreak2k> That makes another project I know of that uses mercurial.
09:37:12 <Alberth> hg is only a mirror, it is not the master repo.
09:49:10 *** JVassie has joined #openttd
09:59:04 *** Devroush has joined #openttd
10:05:05 *** TheMask96 has joined #openttd
10:06:32 *** ChanServ sets mode: +v tokai
10:08:33 *** devilsadvocate has quit IRC
10:35:21 *** Phoenix_the_II has joined #openttd
10:39:25 <TrueBrain> we even have git, and who still uses git!
10:39:54 <Eddi|zuHause> KDE is switching to git...
10:40:22 <planetmaker> what do they use now?
10:41:06 *** KenjiE20 has joined #openttd
10:41:45 *** fonsinchen has joined #openttd
10:42:01 <__ln__> their cvs→svn conversion had quite epic proportions some years ago
10:42:06 <Alberth> TrueBrain: same people that use perl?
10:42:23 <TrueBrain> Perl is not a language
10:43:45 <fonsinchen> TrueBrain, does the compile farm still have the problem of labeling 3rd party versions, e.g. cargodist, as "modified"?
10:44:38 <TrueBrain> it never had that problem
10:44:58 <TrueBrain> (mostly as 'problem' indicates something unwanted)
10:45:45 <Eddi|zuHause> TrueBrain: yes, it is unwanted that a manual compile from git repository gets called "unmodified" but a compile farm build from that exact same repository gets called "modified"
10:46:17 <fonsinchen> I don't understand. If it compiles my cargodist repository it generates a version "<git-id>-cargodistM"
10:46:27 <TrueBrain> Eddi|zuHause: that is not the CF which does that :) The modified bit is added by find-version.sh
10:46:40 <fonsinchen> A version "<git-id>-cargodist" would be enough to make it clear that this is not trunk
10:46:42 <TrueBrain> the compile-farm has no capabilities anymore to apply a patch of any kind
10:46:59 <TrueBrain> and all compiles I did were named correctly, like headtohead branch and such
10:47:08 <TrueBrain> (which are mercurial repos, but the idea holds)
10:47:15 <fonsinchen> If I compile locally I don't get the "M".
10:47:17 <TrueBrain> any M addition is because of find-version failing, not in my control :)
10:48:23 <Eddi|zuHause> TrueBrain: if that is what happened, why does it not happen on local checkouts, but only on the CF?
10:48:39 <TrueBrain> I wonder why it only happens on CFs I didn't start :p
10:48:55 <TrueBrain> the CF does for git: git checkout <url> && ./configure && make
10:48:56 <Eddi|zuHause> TrueBrain: maybe it's git-related
10:49:06 <TrueBrain> so you tell me how the CF can cause a M addition :p
10:49:19 <TrueBrain> (git clone, what ever :p)
10:49:41 <TrueBrain> the only problem of CF is that it can only checkout the master branch
10:50:27 <peter1138> lots of projects use git
10:51:44 <fonsinchen> Maybe I patch findversion to never generate an "M" in a special branch, make that the master branch for the compile farm and when it's done I delete the branch.
10:52:25 <fonsinchen> TrueBrain, could you build another version of cargodist on the compile farm later today?
10:53:06 <TrueBrain> and it really is find-version.sh in fault, as I don't see any reason why it would think the thing is modified
10:53:18 <TrueBrain> if you make a branch-less clone, and compile it, does that not add a M?
10:54:07 <fonsinchen> ah, I should try that.
10:54:21 <TrueBrain> as that is what the CF does (as I told you many times before I believe :p)
10:54:22 <fonsinchen> Can you give me the exact command line the CF uses to do that?
10:54:45 <TrueBrain> pff, takes forever btw ..
10:54:51 <TrueBrain> you should pack from time to time
10:55:14 <fonsinchen> I'm moving to a new server in a few days
10:55:20 <fonsinchen> Thins will be faster there
10:55:38 <TrueBrain> I still want to roll out 3rd party project website for openttd.org :(
10:56:13 <TrueBrain> like <name>.openttd-projects.org
10:56:30 <TrueBrain> then again, I want more CPU cores too ..
10:58:14 <TrueBrain> nope, there really is no reason for the CF to add a M
10:58:30 <TrueBrain> and according to the info, the last compile made on the CF, didn't had the M
10:58:48 <TrueBrain> schedule 0000018, last compile: g5b3732ce
10:58:55 <fonsinchen> We tried, people couldn't connect because of it.
10:59:35 <TrueBrain> btw, the name is: <git-id>-cd
10:59:39 <TrueBrain> not <git-id>-cargodist
11:00:13 <Eddi|zuHause> that's a recent change in the repository, right?
11:00:22 <fonsinchen> I changed the branch name because the version string was longer than 15 characters which caused problems
11:00:45 <fonsinchen> lots of "waring: string too long for destination buffer"
11:02:03 <TrueBrain> weird, findversion.sh gives g5b3732ce-cd back to the CF, but the binary says g5b3732cd-cargodistM
11:02:07 <TrueBrain> that doesn't add up :p
11:03:29 <planetmaker> oh hello, a TrueBrain here :-)
11:04:10 <TrueBrain> I was going to get some lunch ...
11:04:58 <fonsinchen> I have changed the branch name
11:05:24 <fonsinchen> before findversion would tell us g5b3732cd-cargodist
11:09:51 <TrueBrain> so, I wonder why ingame it still added cargodist :p
11:10:52 <fonsinchen> I can reproduce it now.
11:13:44 <fonsinchen> why are we using "git diff-index" instead of simply "git diff --numstat" in findversion.sh?
11:14:20 <fonsinchen> for some reason "git diff-index" generates a lot of output on branchless clones.
11:15:15 <planetmaker> fonsinchen, make a patch for that, if one way works in all cases, and the other not
11:15:45 <fonsinchen> maybe there is a case where it doesn't work. But I'll make a patch.
11:15:47 <planetmaker> @svn commit 16462
11:15:47 <DorpsGek> planetmaker: Invalid arguments for svn.
11:15:54 <planetmaker> @openttd commit 16462
11:15:54 <DorpsGek> planetmaker: Commit by rubidium :: r16462 trunk/findversion.sh (2009-05-29 21:24:51 UTC)
11:15:55 <DorpsGek> planetmaker: -Change [FS#2930]: use a safer way to detect the hash of a mercurial repository (planetmaker)
11:16:17 <planetmaker> well, you should test the patch for corner cases :-)
11:18:11 <fonsinchen> I guess "git status --porcelain" is the safest bet.
11:28:13 <Eddi|zuHause> random suggestion: if servicing is disabled, reliability should influence running costs
11:29:44 <Eddi|zuHause> so basically: divide running costs by the reliability, so a vehicle with 75% reliability will have 33% higher running costs
11:30:57 <Eddi|zuHause> ... or make this independent from the servicing setting
11:50:24 <fonsinchen> and I have also pushed it into my git repository.
11:50:29 <fonsinchen> TrueBrain, can you build it now?
11:50:54 <TrueBrain> I love it when I am not to blame :p
11:51:26 <TrueBrain> will take a while :p
11:52:11 <Eddi|zuHause> isn't that like the second step in the guide of software problem management? "find someone else to blame" :)
11:53:20 <Eddi|zuHause> (the first step is "deny the existence of a problem")
11:58:51 <TrueBrain> error: unknown option `porcelain'
11:58:54 <TrueBrain> fonsinchen: expect failures
11:59:17 <fonsinchen> oh, old git version I guess
11:59:37 <fonsinchen> does it know "--short" instead of "--porcelain"?
11:59:39 <TrueBrain> so your patch is wrong in either case
11:59:52 <fonsinchen> you can stop the build I guess
12:00:32 <TrueBrain> that is something that is not possible :)
12:01:01 <Rubidium> well, you could remove unscheduled jobs from the queue, but that's tricky
12:01:13 <fonsinchen> ok, then we have a very strange version on this one.
12:02:23 <fonsinchen> It doesn't matter much. I'll post a guide in the forum thread on how to reproduce the same version string, once I see it.
12:02:50 <Rubidium> my git says it doesn't know porcelain either
12:03:01 <fonsinchen> does it know "--short"?
12:03:08 <planetmaker> so much for proper testing of the patch... :S
12:03:26 <fonsinchen> porcelain is the same as short, but it is guaranteed to give the same output in later versions.
12:04:00 *** roboboy has joined #openttd
12:04:00 <Rubidium> but then, that's only 1.7.0(.4?)
12:04:17 <Rubidium> Unknown option: --short
12:04:18 <fonsinchen> git status --short
12:04:27 <fonsinchen> and "git status --porcelain"
12:05:22 <Rubidium> but then, I've got a quite new version
12:06:51 <fonsinchen> The build will work anyway. It marks everything as unmodified like this.
12:29:05 <fonsinchen> hrm, "git status" used to be quite dumb until fairly recently
12:37:29 *** ajmiles has joined #openttd
12:38:57 *** Devroush has joined #openttd
12:57:15 <fonsinchen> gah, what happened? I cannot reproduce the behaviour of diff-index anymore ...
13:02:51 *** JostVice has joined #openttd
13:05:23 <fonsinchen> I must have changed something on the server with my last commit ...
13:07:04 *** nighthawk_c_m has joined #openttd
13:07:47 <nighthawk_c_m> Where is the difference between symmetric and asymmetric in the cargodist?
13:09:16 <fonsinchen> symmetric means the same amount of cargo is sent in both directions between two stations
13:09:24 <fonsinchen> asymmetric doesn't do that.
13:17:12 <Eddi|zuHause> rule of thumb: symmetric for passengers, mail, valuables; asymmetric for everything else
13:19:32 <fonsinchen> the build looks good.
13:20:08 <SpComb> the ottdc cargodist builds?
13:37:00 <nighthawk_c_m> WHat falls under the Express cargo class?
13:38:00 <fonsinchen> usually goods, but certain industry grfs might redefine it.
13:38:27 <fonsinchen> In the source code there is a statement that it can be used for passengers
13:39:10 <Eddi|zuHause> mainly food, it looks like...
13:39:34 *** Chruker has joined #openttd
13:52:08 *** devilsadvocate has joined #openttd
13:52:09 <Yexo> <fonsinchen> In the source code there is a statement that it can be used for passengers <- ECS tourists are in both the passengers class and the express class
13:52:27 *** Weeknie has joined #openttd
13:54:05 <fonsinchen> This is why its distribution method can be configured separately.
13:55:45 <Weeknie> I wonder, how does it actually work behind the screens in openTTD development?
13:56:46 <planetmaker> I'd bet similar as in front of the screen.
13:56:51 <Weeknie> Lol, I'll just ignore that:P
13:56:56 <planetmaker> though you don't see what's on the screen when behind it.
13:57:00 <Weeknie> No but I mean, how does anyone know what to do?:P
13:57:13 <Belugas> Eddi|zuHause, yeah, i know... as every friday. We have a meeting at 9:00h
13:57:18 <Weeknie> I'm thinking about helping out a little
13:57:29 <Weeknie> Want to learn C++ and also help with openTTD so:P
13:57:40 <Alberth> everybody does what he likes to do
13:57:55 <Alberth> there is no plan whatsoever
13:58:00 <Belugas> WHEN he likes to do it...
13:58:47 <Yexo> Weeknie: have you programmed before?
13:58:50 <planetmaker> Belugas, Weeknie in a volunteer project there's no leverage to force anyone to do anything, especially not NOW
13:59:00 <planetmaker> oh, hello Belugas :-)
13:59:01 <Alberth> If you run out of ideas, look at the bugs in FlySpray, read requested features in FlySpray, read the suggestions forum :)
13:59:09 <Weeknie> Planetmaker, I know about that;)
13:59:40 <Weeknie> Yexo, yes I have, not in C++, but I have a lot of experience in PHP and quite some in Python
14:00:06 <Weeknie> I learned python through doing an open source project (called pyfa), but that was on a much smaller scale
14:00:09 <Yexo> you might want to read up a bit on pointers then
14:00:09 <Alberth> Weeknie: find a book about C++, read it, find a nice little problem, and start fixing it
14:00:11 *** devilsadvocate has quit IRC
14:00:28 <Weeknie> Oh yeah, Yexo, I'm doing that as we speak:P
14:00:47 <Alberth> Weeknie: and bits are important too :)
14:00:58 <Weeknie> What do you mean exactly?
14:01:34 <Alberth> Belugas: not entirely, atm I am quite fed up with the newgrf gui, yet I push through :)
14:01:38 <Eddi|zuHause> Weeknie: an "easy" task for you to solve: introduce a method to divide the running costs of the vehicle by its reliability
14:02:12 <planetmaker> lol. Outsourcing your latest idea and desire, Eddi|zuHause ? ;-)
14:02:14 <Alberth> Weeknie: a lot of information is packed in collections of bits.
14:02:29 <Eddi|zuHause> planetmaker: well, it is the easiest task from the top of my head ;)
14:02:47 <Alberth> Weeknie: PHP and Python programming don't generally do single bit operations very much
14:03:06 <Weeknie> As it so happens I have been working with it a little though (in php anyway)
14:03:40 *** devilsadvocate has joined #openttd
14:05:05 <Weeknie> Would it be possible to do any work with the git repos? I read you're using subversion as your main repo manager, are they all synced every so many time or is the subversion one just copied over to git and all others?
14:05:28 <Eddi|zuHause> Weeknie: a "small" detail like i suggested lets you browse some parts of the code with a specific goal in mind ("where is the reliability stored", "where is the running cost calculated", etc.) and lets you learn a few programming techniques ("how do you do a division with fractional values when you can only use integers") that are crucial to the workings of openttd
14:05:43 <planetmaker> Weeknie, they're usually synced
14:05:49 <planetmaker> upon every commit to svn
14:05:56 <Eddi|zuHause> Weeknie: you can use git or hg for your personal development
14:06:08 <Alberth> highly recommended even
14:06:12 <Eddi|zuHause> Weeknie: the repositories are synced automatically with each svn commit
14:06:20 <SpComb> Weeknie: the git/hg repos are purely read-only
14:06:45 <Weeknie> Well atleast there's enough people here to help me out when I have a problem obviously:P
14:07:08 <Eddi|zuHause> Weeknie: if you're done developing your patch, you usually send it to the bug tracker or open a thread at the forum [or both]
14:08:11 <Belugas> Alberth, it's normal. I got fed up with some stuff myslef quite oftenly.
14:08:23 <Belugas> In fact, yo're more tenacious then me...
14:14:37 <Weeknie> Does anyone here know what the dev status is on that newGRF Airports state machine?
14:14:48 <Weeknie> Are there people working on it or?:P
14:15:35 <Weeknie> I mean, really no one has any overview?
14:15:37 <Belugas> Weeknie, ask Yexo. peter1138, have i gooffed?
14:16:03 <Weeknie> Yexo, see question above please?
14:16:36 <Yexo> I have not really done anything on newgrf airports the last weeks (months?)
14:16:37 *** SirSquidness has joined #openttd
14:17:07 <Weeknie> Do you know anyone who might know something?
14:17:23 <Alberth> everybody knows something :)
14:17:34 <Weeknie> I was worried you were going to say that...
14:17:38 <Yexo> status: next target is building some newgrf airports
14:17:56 <Yexo> that is not too hard, it can probably be ripped out of the airports.hg repo pretty easily
14:18:14 <Alberth> but not a C++ problem :)
14:18:20 <Yexo> after that the main part has to be done: implementing the statemachine callback
14:18:46 <Yexo> there is a 'working' version in my hg repo, but I don't really like it
14:18:59 <Weeknie> And that sounds like what I'd like to try to do, or atleast try to help with
14:19:05 <Yexo> it's pretty much stalled because I haven't found a better way yet
14:20:23 <Yexo> now that is a good reason for me to work on it again :)
14:20:31 <Yexo> and I'll be very happy with any help I can get
14:21:17 <Weeknie> Well, I'd have to start with another question though
14:21:33 <Weeknie> Could you explain to me what you exactly want to do?
14:23:14 <planetmaker> In laymen's terms it would be "a newgrf interface to airport implementation"
14:23:32 <planetmaker> but that's as far as my knowledge there goes :-)
14:23:48 <Weeknie> Yeah well that helps me:P
14:24:07 <Yexo> make it possible to define airports + their statemachines in newgrf files
14:24:22 <planetmaker> in a sane way :-)
14:24:45 <Weeknie> statemachines being the ones that handle planes lading/taking off?
14:25:01 <Alberth> all movement of planes afaik
14:25:09 <Yexo> the statemachine is the code that handles all plane movement
14:27:48 <Weeknie> Quite funny to see something I never had to deal with in python or php
14:29:20 <Alberth> then you programmed the wrong software in python and php :p
14:30:12 <Weeknie> In php I haven't been programming anything that would've required these kind of restrictions
14:30:15 <Weeknie> They were just websites
14:30:46 <Weeknie> And no biggies at that, so there wasn't any need to pay attention to memory management
14:30:52 <Weeknie> Same goes for the python one actually
14:31:35 <Weeknie> What do you exactly mean with "Every airport has 16 4-byte slots persistant storage (var 7C).", especially the 7C part
14:31:36 <planetmaker> it's not memory only. But also speed of implementation ;-)
14:31:54 <Weeknie> Oh brb, keep talking though:P
14:32:31 <Yexo> Weeknie: that is/will be part of the newgrf spec
14:34:43 <Weeknie> Ah yes, that was my question
14:34:56 <Weeknie> What's actually the difference between TTDPatch and openTTD?
14:35:01 <Weeknie> Or what is TTDPatch exactly
14:35:34 <Yexo> TTDPatch adds new features to TTD by modifying the executable once it's loaded in memory
14:35:45 <Yexo> as such it needs the original executable to run
14:36:44 <planetmaker> basically: ttdpatch is something entirely different
14:36:53 <planetmaker> it's a modified ttd
14:36:57 <Weeknie> And doesn't have anything to do with openTTD
14:37:01 <planetmaker> whereas openttd is ttd re-written
14:37:17 <Weeknie> Then I don't care about TTDPatch from now on:P
14:37:26 <planetmaker> don't start that.
14:37:49 <planetmaker> such statements are prone to cause flame wars. Especially in the forums
14:38:22 <planetmaker> and as you may have noted: the newgrf wiki is hosted on the TTDP website
14:38:33 <planetmaker> while the ttdp nighlies are built by the openttd server
14:38:53 <Weeknie> I won't say it anymore
14:39:05 <nighthawk_c_m> And there are quite a few people that play both versions
14:39:16 <nighthawk_c_m> Each has its advantages and disadvantages
14:39:33 <Weeknie> tbh people shouldn't make a fuss about it
14:39:39 * planetmaker plays OpenTTD only, though ;-)
14:39:57 <Alberth> Weeknie: people should think before they speak
14:40:01 <Weeknie> And neither do I care about whether you only play openTTD or also TTDPatch or do I know what
14:40:16 <nighthawk_c_m> Actually, a question, why is the town rating influenced so dramatically by plowing away trees?
14:40:25 <Weeknie> But if me saying that hurts you than I think I'm not the only one at fault so to speak
14:40:32 <planetmaker> Weeknie, one starts to care about that when people mix up those games in questions ;-)
14:40:56 <nighthawk_c_m> The town rating is not based on People, its based on politicans
14:40:56 <Weeknie> That is important yeah
14:40:58 <planetmaker> nighthawk_c_m, you destroy the recreational areas for them!
14:41:09 <nighthawk_c_m> They got parks, no need to whine
14:41:12 <planetmaker> besides. Try to do that in real life. Quite realistic ;-)
14:41:28 <nighthawk_c_m> Not if the city/government says we need that railroad
14:41:48 <Weeknie> But in this case they don't:P
14:41:50 <planetmaker> that doesn't change that people get antagonized by cutting down trees.
14:41:57 <nighthawk_c_m> then the activists can cry and beg and demonstrate as much as they want - just during early gameplay its very very annoying
14:42:03 <Weeknie> Hmm, makes me wonder about something else lol
14:42:07 <Weeknie> If a town has a subsidy under it
14:42:14 <Weeknie> Will the rating fall just as hard?
14:42:21 <Yexo> Weeknie: I'm pretty sure planetmaker doesn't care personally what you think about ttdpatch, but in general the "I don't care about ttdpatch" attitude is not going to work
14:42:24 <planetmaker> nighthawk_c_m, quite not true. Then a lengthy process through all instances of courts has to be walked
14:42:29 <Weeknie> With the thing you just said, that wouldn't really be realistic would it?
14:42:30 <Yexo> and not just because the newgrf specs are shared
14:42:32 <planetmaker> which will drag such thing through several years
14:42:49 <Weeknie> yeah yeah, I'll just stop about it:P
14:42:58 <Zuu> why wouldn't they like trees?
14:43:05 <Alberth> Weeknie: realism is *not* a design criterium for OpenTTD
14:43:19 <Zuu> Ops, not 100% scrolled down
14:43:26 <nighthawk_c_m> True, I'm just seeing this in a game I have running, just started - and youi can cheat yourself to better terms by ervicing inner town stations
14:43:49 <Alberth> planetmaker: indeed, thanks
14:44:11 <planetmaker> Alberth, common mistake for German / Dutch people :-)
14:46:17 <planetmaker> it's a non-intuitive thing.
14:46:43 <planetmaker> usually the latin words are the same in all languages. Except here English uses the Greek stem instead of the Latin
14:47:25 <Weeknie> Wait, what are you talking about exactly?
14:47:56 <planetmaker> hm... or is it another declination in Latin? I'm talking about 'criterion' / 'criterium'
14:48:48 <Weeknie> But I don't know what's wrong with criterium
14:48:58 <nighthawk_c_m> I'll have to pass on that, even so I had latin in school
14:49:03 <Yexo> planetmaker: I think criterium (criteria for plural) is the latin form
14:49:07 <Zuu> In Swedish it is 'kriterium' eg. it also ends with 'um'.
14:49:18 <planetmaker> Yexo, yes, that's true.
14:49:34 <Weeknie> I'm still in school, had latin 2 years ago
14:49:41 <planetmaker> But the singular is the Greek singular. I don't know how Greek deklination works, though
14:49:49 <Alberth> Weeknie: 'criterium' is not an english word
14:50:18 <planetmaker> sorry Alberth that I make such fuss about it :-) It was not my intention.
14:50:27 <planetmaker> I hope you don't mind ;-)
14:50:57 <Weeknie> What should it be anyway?
14:51:46 <Alberth> Weeknie: 'criterion' (s/um/on/ means 'substitute on for um')
14:52:08 <Zuu> criterion at least got more hits at google than criterium.
14:52:48 <Alberth> Weeknie: not a gvim user nor a sed user, apparently :)
14:53:12 <Zuu> Next, send him to vim school? ;-)
14:53:37 <Alberth> well, if you want a good editor.... ;)
14:54:28 <Weeknie> I wonder how fast I get murdered when I say I'm using netbeans
14:54:35 <Weeknie> And actually on windows?
14:55:24 <planetmaker> δῶρον (st. δωρο-) present <-- seems to be a valid Greek declination with -on and -a for singular and plural respectively
14:55:49 <planetmaker> the omicron declination for neuter :-)
14:56:07 <planetmaker> s/declination/declension/ :-)
14:56:32 <IPG> i think i found a bug... in the latest nightly
14:56:38 <IPG> sorry if anybody said it before
14:56:48 <Alberth> Weeknie: I don't mind you having to fight with Windows or netbeans, or both :)
14:57:01 <Weeknie> Lol, well that's a first from anyone:P
14:57:07 <IPG> if i track the minimap to the NW end of the map, it shows a big black nothin
14:57:44 <IPG> it also does with my NE end
14:58:06 <planetmaker> can you scroll, IPG?
14:58:17 <Yexo> what else do you expect outside the map?
14:59:12 <planetmaker> Weeknie, and now probably Alberth decides that it's an interesting problem and will take charge of that :-P
14:59:18 <planetmaker> Or I'm terribly wrong :-)
14:59:25 <planetmaker> But anyway it's self-organizing here ;-)
14:59:31 <IPG> but if i am starting to arrive to the end, it suddenly goes black
15:00:06 <Alberth> planetmaker: I am only confirming the problem, I'll ket Weeknie do the solving ;)
15:00:16 <Yexo> the minimap is completely black here, but that might be due to a very small map size
15:00:18 <planetmaker> good choice actually
15:00:40 <Weeknie> But would there be any place to see who is working on what, so I, having just missed this conversation and for whatever reason hearing it in the evening, gonna do it too, and end up doing the same work as Alberth? (or myself, apparently)
15:00:47 <Alberth> Yexo: it should never be completely black, the center should always have a point at the mapo
15:00:50 <Weeknie> Sneaky typing while I'm typing bastards that you are:p
15:01:19 <Yexo> Alberth: try a 64x64 map
15:02:49 <Weeknie> Lol, industry is going nuts, whole island is full
15:02:58 <Weeknie> But I see the black map anyway
15:03:35 <Alberth> Weeknie: there is no such place
15:05:01 *** Keyboard_Warrior has joined #openttd
15:06:23 <Alberth> It looks like a uint underflow problem to me
15:07:39 *** Keyboard_Warrior has quit IRC
15:07:39 *** holyduck has joined #openttd
15:08:51 *** orudge_ has joined #openttd
15:09:12 <Alberth> does anybody how to setup make to use -j2 by default?
15:10:39 *** Coco-Banana-Man has joined #openttd
15:13:25 <planetmaker> Alberth, you want it to run with less than all cores available?
15:13:55 <planetmaker> "With no argument, make runs as many jobs simultaneously as possible."
15:14:28 <Alberth> that's make -j but I want to type 'make' and get 'make -j2'
15:14:58 <Alberth> make -j is sort of useless, as starting 50-60 compile processes is not going fast
15:15:17 <Alberth> the machine would just start swapping :p
15:15:27 <Weeknie> What system are you working on?
15:15:37 <planetmaker> ah... I understood that wrongly, Alberth :-) Thanks, you're right
15:15:51 <planetmaker> I always assumed that it would do that automatically... even without -j
15:15:55 <Weeknie> Not sure which file it was
15:16:13 <Weeknie> I believe the ~/.bash-profile (or any that comes close) was the right one
15:16:16 * Alberth guesses Weeknie picks the wrong file
15:16:42 <planetmaker> still the alias thing would probably work
15:16:50 <Alberth> Weeknie: I use z-shell :)
15:16:52 <Weeknie> The alias thing is how you should do it:P
15:16:59 <Alberth> yes, it is a good suggestion, thanks
15:17:03 <Weeknie> Then the bash-profile one won't work:P
15:17:13 <Weeknie> But just make sure the alias gets evaluated at every boot
15:17:43 <Alberth> every new shell even, aliases don't get exported to child processes
15:18:06 <Weeknie> Well, the bash-profile would do both for bash:P
15:18:11 <Weeknie> (when you have the correct file anyway)
15:18:35 <Alberth> afaik ~/.bash_profile is only for login
15:18:50 <Alberth> (I need to edit ~/.zshrc )
15:19:10 <Weeknie> I'm happy I remembered that much lol
15:19:32 <Weeknie> I had a course, failed it miserably (not really my fault, they promised quite the opposite of what I got, but that aside)
15:19:37 <Weeknie> Atleast it's helped me 1 time now:P
15:22:32 *** fonsinchen has joined #openttd
15:25:32 <Weeknie> hmm, now that I'm thinking about it, C++ development is quite different compared to dev of PHP or Python
15:25:42 <Weeknie> Those files you can just save and then hit f5 or restart the program
15:25:54 <Weeknie> In C++ you'll have to compile it continouisly, right?
15:26:19 <planetmaker> depends upon the definition of 'continuously'
15:26:35 <Weeknie> Erm well, every time you make a change and want to test it out
15:26:48 <Weeknie> Next chapter in my tutorial
15:28:39 <Zuu> Weeknie: In Visual studio you will just hit F5 and it will compile and then run your program. :-p
15:29:00 <Weeknie> But first it has to compile
15:29:15 <Weeknie> About how long would that take with openTTD (with an average pc)
15:29:44 <Zuu> If you compile from scratch.
15:30:06 <Zuu> But if you only make changes to a cpp file, then only that file needs to be re-compiled.
15:30:08 <Weeknie> (yes I'm really totally new to this:P)
15:30:42 <Zuu> If you on the other hand change a .h file that gets included by many cpp files, then all those cpp files need to be re-compiled. :-p
15:30:47 <planetmaker> Zuu, that's a long time span...
15:31:06 <planetmaker> even my 3-year old laptop is faster
15:31:15 <Weeknie> Header files, yeah, I guess I'll run into them somewhere along the way
15:31:24 <glx> true for english.txt changes too
15:31:32 *** theholyduck has joined #openttd
15:31:43 <Zuu> planetmaker: Since I don't know exactly the "average pc" it's hard to give a small interval.
15:32:27 <Weeknie> Yeah ok, that's a little hard
15:32:38 <Weeknie> But well, it takes some time, a lot longer for a page to refresh anyway:p
15:32:49 <Weeknie> On the other hand, like you said, depends on what I change
15:33:17 <planetmaker> if you just play with a single cpp file, it re-compiles very quickly
15:33:24 <planetmaker> if you built it once before.
15:33:43 <Weeknie> Yeah, but else you'll have to build the whole project and that takes a while
15:34:32 <planetmaker> it always depends on what changed
15:34:39 <Zuu> If you would be using visual studio, you should be compiling debug builds as the release builds take minutes to link. Not sure if that also holds for netbeans. But if netbeans uses gcc, then it is probably not that bad compared to Visual Studio link times.
15:35:04 <Weeknie> netbeans can use gcc yeah
15:35:09 <Weeknie> It can use any compiler:P
15:35:21 <Weeknie> Set it up with minGW, and that one has the gcc compiler
15:35:41 <Belugas> ho... tenacious is not a word :S "stubborn" then !
15:37:46 <Zuu> A really good thing with compiled code is that the compiler eliminates many mistakes in the code even before you try to run the code.
15:38:07 <Zuu> In php etc. you will only find some coding errors when that code is executed.
15:38:26 <Weeknie> in php you just press f5
15:38:42 <Weeknie> And you can instantly fix it
15:38:55 <Weeknie> If you were doing a complete compile and then only halfway the compiler detects the error
15:39:00 <Weeknie> You can start all over again
15:39:22 <planetmaker> you continue to compile where it failed
15:39:32 <planetmaker> unless you change part of what is compiled, too
15:43:20 <Weeknie> Hmm well, pointers aren't that hard
15:43:39 <planetmaker> until they get you into segmentation faults and null pointer assignments
15:44:10 <planetmaker> hm, not assignment. but dereference
15:44:13 <Terkhen> yeah, they are cleverly disguised to look harmless :P
15:44:33 <Alberth> Weeknie: a crash of the program
15:44:34 <planetmaker> Weeknie, they are those things which drive you nuts ;-)
15:45:26 <Weeknie> Null pointer dereference, would that happen to be a pointer to a memory block that has already been cleared? (aka doesn't really belong to the program anymore)
15:45:33 * planetmaker had yesterday probably two dozen or so of some kind of that
15:48:54 <Alberth> Weeknie: NULL is usually outside the addressable memory of the application, so you get a nice crash
15:49:36 <Alberth> Weeknie: accessing memory that you free()-ed is indeed much nastier. What happens, depends on the contents that happens to be there.
15:49:39 <Weeknie> yeah, and we all love those
15:49:54 <Alberth> Weeknie: if you are unlucky, the program will continue to run
15:50:10 <Weeknie> With what is actually just random data?:P
15:52:28 <Alberth> Equally nice is to have 2 pointers that you think point to different memory, actually use the same memory.
15:53:02 <Weeknie> Can imagine that yeah
15:53:10 <Alberth> or they share only a small part
15:53:35 <Weeknie> I do wonder how you'd do that though:P
15:54:06 <Alberth> I once spend 3 weeks debugging, only to find that the compiler generated the wrong code
15:54:22 <Weeknie> And how did that happen?
15:56:05 <Alberth> it was with a 6809 compiler, where a 16bit integer was stored in 2 8bit registers. The -- in x[--i] got computed wrong, so it actually added 255 or so.
15:56:47 <Alberth> that was a really bad compiler, we found several errors in it, more than in our own code :p
15:57:04 <Weeknie> It was substracting from the first 8bit register instead of the 2nd one like it should've, right?
15:58:19 <Weeknie> that is a bad compiler indeed anyways
16:01:02 <Alberth> Another fun example was "abc\0xyz", which got translated to "abc0xyz"
16:01:32 <Weeknie> Is that a general compiler thing or that bad compiler failing?
16:02:37 <Alberth> I hope that bad compiler failing, afaik \0 means a null byte (ie 0x00), instead of '0' (ie 0x30). I never checked it with another compiler though
16:04:26 <glx> undefined behaviour are nice to chase too
16:04:52 <Alberth> or 32bit vs 64bit behaviour :)
16:05:13 <Weeknie> Sounds like fun too yeah
16:05:36 <Alberth> Weeknie: so many new ways to make a mess, compared to python :)
16:05:58 <glx> (all cited things already encountered in openttd IIRC)
16:06:37 <Alberth> and 'int' versus 'unsigned int' (as with the minimap)
16:08:00 <Weeknie> Did you fix that black bug?
16:08:53 <Alberth> I do have a fix, just checking whether I can reduce it, and whether it then still works always
16:09:12 <Alberth> hmm, that sentence is wrong :p
16:09:25 <Weeknie> erm, i actually meant to say "Did that fix that black bug"
16:09:29 <Weeknie> As in was that the problem?:P
16:10:10 *** Aylomen has joined #openttd
16:11:27 <Alberth> it was introduced in 19816, where several values changed from int to uint, so it is a pretty safe bet that it is the cause of the black map :p
16:15:06 <Weeknie> Well, I wonder how the map worked before the
16:15:33 <Weeknie> int to unint sounds like negative and positive values with 0.0 in the middle to 0.0 top left and counting upwards
16:16:27 *** Luukland has joined #openttd
16:16:28 <Alberth> uint is the unsigned integer from 0 to 2^32-1 int is the integer value of -2^31 to 2^31-1
16:17:25 <Alberth> oh, you said that, I see now
16:17:54 <Alberth> the fun is when you have a negative number and consider it to be a uint. It gets very large then
16:18:07 <Weeknie> Yeah I know how that works:P
16:18:46 <Alberth> well, trunk gets modified very heavily, so every now and then, something breaks
16:19:25 *** Brianetta has joined #openttd
16:21:40 <Alberth> it is quite seldom, if you look at how many changes happen
16:22:09 *** lobstar has joined #openttd
16:24:10 <Aylomen> Am I blind? Or where can I change my password on account.openttd.org?
16:26:06 *** Grelouk has joined #openttd
16:29:26 <Ammler> Aylomen: somewhere on the footer is a Link
16:29:58 <Aylomen> there is just privay policy profile logout and contasct
16:30:23 <Aylomen> and profile says: This page is not yet done, but will be in the very near feature.
16:35:17 *** |Jeroen| has joined #openttd
16:38:34 *** Polygon has joined #openttd
16:40:36 <fonsinchen> The cargodist test game is running at mz.openttdcoop.org:3999
16:45:50 <Ammler> Aylomen: sorry then, it was there sometime
16:46:02 <Ammler> maybe before the merge
16:58:00 <Zuu> yep, indeed you can't change your password.
16:58:21 *** George is now known as Guest90
16:59:05 <Ammler> Zuu: maybe you should change ottdau to cargodist instead dest
17:01:50 <Weeknie> What's that cargodist thing about?
17:03:02 <Weeknie> Hmm, what is the #openttdcoop channel for?
17:10:00 <fonsinchen> cargodist is a system to automatically transfer and distribute cargo in your network
17:10:25 *** TheMask96 has joined #openttd
17:10:34 <fonsinchen> like "cargo destinations", but a little more.
17:11:41 <planetmaker> Weeknie: have a guess by the name
17:11:57 <Weeknie> But what does it mean for openTTD?:P
17:12:58 <planetmaker> Well, I could also have referred to that question, yes ;-)
17:13:08 <fonsinchen> if you have a pax line from a to b and one from b to c with cargodist passengers will be automatically transferred at b so that some of them go from a to c and from c to a.
17:13:09 <planetmaker> but that was already answered and not your last question :-P
17:13:52 *** Jolteon has joined #openttd
17:19:12 <Weeknie> So that you make more money
17:21:00 <planetmaker> Weeknie: not quite
17:23:04 <planetmaker> making money is a no-brainer with OpenTTD
17:27:18 *** welshdragon has joined #openttd
17:28:58 *** theholyduck has joined #openttd
17:31:26 <Alberth> Weeknie: you get a lot of transport over multiple hops, so a heavier load of the network.
17:31:31 <Zuu> Ammler: does cargodist has a finger? Otherwise it will be hard. Cargodest is on finger.openttd.org.
17:32:05 <Weeknie> Erm, what's cargodest for?:P
17:32:10 <Ammler> TrueBrain: could you also write a finger to openttdcoop.org?
17:32:30 <Weeknie> Found it in openttd wiki
17:32:32 * TrueBrain gets a pen and paper
17:32:54 <TrueBrain> if I had any idea what the fuck you were talking about, most likely not
17:32:55 <Ammler> maybe I could do it with a cron
17:33:29 <Ammler> the CF does move the bundles to bundles.openttdcoop.org, could it also write a versions file there?
17:34:17 <TrueBrain> I always wonder what make people think I can read their mind
17:34:27 <TrueBrain> I know I can act godlike, but seriously?
17:34:34 <glx> because you're a true brain ?
17:35:00 <Ammler> the CF of openttd.org does compile bundles for some 3rdparty patches and move those to bundles.openttdcoop.org
17:35:37 <Zuu> TrueBrain: I think Ammler want something similar to finger.openttd.org for openttdcoop.org. :-)
17:36:09 <Zuu> So that he can pass me the link, and then I add a new target to OpenTTD Auto Update for eg. Cargodist.
17:36:10 <Ammler> well, for those bundles compiled at openttd.org
17:36:27 <TrueBrain> Zuu: owh, that; he should fix that himself :)
17:36:48 <Ammler> well, you can push it, I would need to pull it
17:36:59 <Ammler> but that is fine then :-)
17:39:17 <TrueBrain> put a script on the ssh account or something
17:43:56 <TrueBrain> we can also execute something over the ssh connection after compile is done btw
17:44:06 <TrueBrain> that is how our finger updates
17:44:41 <TrueBrain> for example: /var/www/finger.openttd.org/scripts/update-versions.sh "custom/cargodest" "@@revision@@" "@@date@@"
17:44:43 <Ammler> hmm, that would be nice
17:44:53 <TrueBrain> is excuted over the ssh connection
17:45:03 <TrueBrain> (the CF is completely isolated and can only push data)
17:45:27 <Ammler> so I need to create a bash script, which writes the versions.txt
17:45:48 <Ammler> hmm, you like to share your script?
17:46:04 <CIA-6> OpenTTD: translators * r19820 /trunk/src/lang/ (10 files): (log message trimmed)
17:46:04 <CIA-6> OpenTTD: -Update from WebTranslator v3.0:
17:46:04 <CIA-6> OpenTTD: traditional_chinese - 78 changes by josesun
17:46:04 <CIA-6> OpenTTD: german - 17 changes by planetmaker
17:46:04 <CIA-6> OpenTTD: italian - 17 changes by lorenzodv
17:46:05 <CIA-6> OpenTTD: lithuanian - 28 changes by BlinK_
17:46:05 <CIA-6> OpenTTD: polish - 1 changes by silver_777
17:46:53 <TrueBrain> yes; I think that is a good excercise for you
17:47:02 <TrueBrain> (mostly as our script does much much more, like generating the index.htmls and stuff)
17:48:44 *** Jolteon has joined #openttd
17:50:22 <CIA-6> OpenTTD: alberth * r19821 /trunk/src/smallmap_gui.cpp: -Fix (r19816): Minimap displayed a black screen instead of the upper edge of the map (reported by IPG).
17:51:04 <Alberth> thanks for the report
17:55:38 <planetmaker> moin andythenorth
17:58:58 * andythenorth wonders what to work on
17:59:18 <Weeknie> Care for a state machine thing?:p
17:59:23 <Zuu> You're welceome to finish my thesis report.
18:00:30 <andythenorth> Zuu: what's the subject?
18:00:47 <Zuu> improving a roundabout capacity model
18:01:09 <Weeknie> Real life transportation shit?
18:01:10 <andythenorth> civil engineering or comp sci?
18:01:30 * andythenorth did one year of civil engineering degree then switched
18:01:47 * Alberth experimented with a roundabout in OpenTTD some months back :p
18:02:21 <andythenorth> Zuu: if you need anything about politics, philosophy of science, cultural theory, or post-modern/post-structuralist analysis, I can help
18:02:23 <Weeknie> Cargo dest thingy not working
18:02:48 <andythenorth> and also html, css, and some very basic python
18:02:52 <andythenorth> roundabouts...no
18:03:13 <andythenorth> planetmaker: how is our GUI adventure?
18:03:16 <Zuu> Actually the model I'm studying model roundabouts like a series of T-intersections.
18:05:18 <Zuu> The GUI adventure, sounds like a good topic.
18:05:35 <planetmaker> I kind play currently around with moving the heightmap into that window
18:06:00 <planetmaker> as that window shares many of those options, it should get integrated there, too.
18:06:02 <andythenorth> screenie? Or do I need to compile? :)
18:06:15 <planetmaker> Which is nicer than making everything twice for that one
18:06:39 <planetmaker> but without the heightmap
18:06:45 <planetmaker> which is not looking well so far
18:07:36 <fonsinchen> Weeknie, what are you trying to do and how does it fail?
18:07:56 <Weeknie> What gave you that impression?:P
18:08:00 <Weeknie> I'm not working on anything yet
18:08:45 <Weeknie> Still have to finish my C++ tutorial:P
18:08:54 <planetmaker> I haven't yet decided to change the panel colour. But I guess it looks much better, if done, just as you proposed, andythenorth
18:09:11 <fonsinchen> You said "Cargo dest thingy not working"
18:09:28 <Weeknie> NO, by that I mean that I got me a cargodest install through the auto updater
18:09:38 <Weeknie> Missing sample.cab or something like that?
18:10:03 <planetmaker> dIst not dEst, Weeknie
18:10:09 <fonsinchen> you'll still have to put the data files in the right place, then
18:10:21 <planetmaker> same idea, but principally different approach
18:10:30 <Weeknie> It would've compained about that
18:10:38 <andythenorth> planetmaker: I might have a go with the GUI code.
18:10:42 <Weeknie> But now that I think about it
18:10:44 <andythenorth> not on a friday night though
18:11:01 <planetmaker> andythenorth: it's no principal thing. I know exactly where to do it :-)
18:11:10 <Ammler> TrueBrain: Usage: /home/openttd/public_html/update-versions.sh <project> <version> <date> [<branch>]
18:11:21 <planetmaker> but sure, go ahead :-)
18:11:35 <planetmaker> it's just that I wanted to keep steps small
18:11:44 <planetmaker> and changing that colour is a step possible independently
18:11:47 <TrueBrain> Ammler: you have seen the format available by the one I gave you
18:11:52 <planetmaker> and actually at any time within that process
18:12:52 <Ammler> TrueBrain: yes, should match
18:12:53 <TrueBrain> Ammler: and I am not going to adjust our script to comply to your format .. so please give me a line of execution in the format I gave you
18:14:03 <Ammler> "branch" is for the projects on the DevZone
18:14:24 <Ammler> maybe I should rename it to type
18:14:24 <TrueBrain> I care 0.0000 what is for what :)
18:14:29 <TrueBrain> I want an exact line I can copy/paste
18:19:28 <Ammler> ok, I test it from my local over ssh
18:23:31 <Ammler> ssh -t openttd@openttdcoop.org /home/openttd/bin/update-versions.sh "cargodist|is2" "@@revision@@" "@@date@@"
18:24:12 <Ammler> dunno the var for the project :-)
18:24:25 <Ammler> but else it is like your paste, isn't?
18:24:58 <Zuu> Weeknie: Not sure if someone helped you as I was on the phone, but OpenTTD Auto Update do not copy the TTD data files and/or OpenGFX/OpenSFX to the new installations. Instead, it is good practice to place those in your personal OpenTTD dir where all installations can find them.
18:25:32 <Weeknie> I know, that's probs what I forgot
18:25:39 <Weeknie> But I don't care actually
18:25:49 <Weeknie> The homedir thing is a good idea though
18:26:06 <Ammler> Zuu: does the finger file match the format you have from finger.openttd.org?
18:26:54 <Zuu> Just what I'm checking (got to take a phone call, so I haven't came that far yet)
18:27:35 <Zuu> Weeknie: The OpenTTD readme explains where it searches for data files.
18:27:54 <Weeknie> Sorry, was too lazy:P
18:29:05 <Ammler> hmm, I do not have tabs
18:29:34 <Ammler> my vim does convert tabs to spaces..
18:29:41 <TrueBrain> Ammler: I just copy/pasted the line into the settings, it is up to you to make sure it works
18:31:48 <TrueBrain> Ammler: you want that line for cargodist or is2?
18:32:15 <Ammler> oh, if you copy paste, wait
18:32:24 <TrueBrain> I told you like 3 times
18:32:26 <Ammler> cargodst: ssh -t openttd@openttdcoop.org /home/openttd/bin/update-versions.sh "cargodist" "@@revision@@" "@@date@@"
18:32:38 <DorpsGek> andythenorth: pikka was last seen in #openttd 3 weeks, 1 day, 21 hours, 29 minutes, and 2 seconds ago: <Pikka> @calc andythenorth
18:32:43 <Ammler> is2: ssh -t openttd@openttdcoop.org /home/openttd/bin/update-versions.sh "is2" "@@revision@@" "@@date@@"
18:33:00 <TrueBrain> for that I asked, didn't I? So hard to do what someone asks? omg ..
18:33:17 <Ammler> well, i have no idea, how your script works
18:33:34 <Ammler> so you do that for every project again?
18:33:49 <TrueBrain> I told you what I wanted, didn't I? Just comply, all you can do :p
18:33:52 <TrueBrain> either way, installed
18:34:05 <TrueBrain> no, it is a global mask which magically works for all projects ....
18:34:09 <TrueBrain> of course it is for every project
18:34:52 * andythenorth offers TrueBrain something for his frazzled nerves
18:35:01 <Ammler> I need to convert to tabs, but that is on my side
18:35:32 * planetmaker compiles *.gui... and gets an error :S
18:35:37 <TrueBrain> andythenorth: tnx, but no tnx, I am fine ;)
18:35:53 * andythenorth buys a depressing song on itunes
18:38:11 <TrueBrain> andythenorth: I am just always greatly suprised that it seems impossible to follow simple instructions :p
18:39:08 <planetmaker> it depends upon 'simple' and the assumed familiarity with a topic
18:39:35 <TrueBrain> planetmaker: well, make a line identical to mine which I can copy/paste, should be easy enough for any person, not? :)
18:39:59 <TrueBrain> andythenorth: says more about you than your wife I guess ;)
18:40:49 <andythenorth> planetmaker: maybe we made it too nice :P
18:40:57 <andythenorth> lots of the default game isn't that nice :)
18:44:28 <Zuu> Ammler: It will probably work, I'm currently adding support to specify which binary server to use. All previous finger targets use binary.openttd.org, so I need to override that for cargodist+is2.
18:46:14 *** theholyduck has joined #openttd
18:48:28 <fonsinchen> btw, the compile farm still produces the "M", but only on windows binaries.
18:48:36 <fonsinchen> That sucks. Royally.
18:48:53 <Ammler> there it doesn't use findversion, afaik
18:49:03 <TrueBrain> findversion.vbs is your friend
18:49:34 <fonsinchen> I can't test vbs scripts here
18:49:58 <fonsinchen> is it possible to make an isolated recompile of the windows binaries?
18:50:43 <fonsinchen> ah, but the git revisions won't match then
18:51:08 <Ammler> well, we could remake the whole thing, so we know finger update works too :-)
18:51:17 <TrueBrain> not worth my time to trigger 1 target .. maybe Rubidium wants to, but I rather recompile the whole set
18:51:27 <planetmaker> hm... that diff is way too
18:52:05 <fonsinchen> I guess I'll just switch off the modification detection completely now.
18:52:31 <fonsinchen> It doesn't work and I cannot reproduce the problem. It's somewhat nondeterministic
18:53:17 <Ammler> well, is2 made a release changeset and reverted it right after compile with tagging
18:54:11 <Ammler> long time ago since last is2 release :-)
18:56:25 <planetmaker> andythenorth: the repo for this stuff is on the DevZone, project openttd-gui
18:58:33 <Zuu> I haven't tested it using the client program yet as I don't have a developer client on this program, so I can't make it available only for me at the moment. hhm, I could use my IP though.
19:00:01 *** Brianetta has joined #openttd
19:00:24 *** PhoenixII has joined #openttd
19:03:28 *** Phoenix_the_II has quit IRC
19:04:06 <Zuu> A problem with cargodist is that the version string in the binary (.exe) is not the same as the one finger tells.
19:05:04 <Ammler> yeah, I already told fonsinchen
19:05:07 <Zuu> For cargodEst it looks like this: "Development h3b244a8f". For cargodIst it looks like this: "Development g3c5595c9M-cd"
19:05:20 <fonsinchen> I'm working on that.
19:06:01 <Zuu> Are you planing to remove the "-cd" part also?
19:06:18 <Ammler> or adding it to findversion.sh :-)
19:08:10 <fonsinchen> no, the "cd" will stay
19:08:12 <Zuu> The IS2 is working fine though. (added it as well to ottdau.
19:08:21 <fonsinchen> I just switch off the modification detection
19:08:52 <fonsinchen> TrueBrain, how does the compile farm determine the names of the output files?
19:09:13 <fonsinchen> Ie, why does it call them "*cargodist*" instead of "*cd*"?
19:09:29 <TrueBrain> cargodist is in the configure I believe
19:09:45 <TrueBrain> the version comes from findversion.sh, via a bit of magic
19:09:49 <Ammler> =folder on bundles.openttdcoop.org
19:10:09 <Ammler> version is the 3rd field, afaik
19:10:48 <fonsinchen> I don't get it. Which 3rd field? Can I do anything about it?
19:11:20 <Ammler> well, dunno, how proper it is to patch findversion in that respect
19:11:21 <TrueBrain> I also have no idea :p
19:11:47 <fonsinchen> so you get findversion to output a directory somehow?
19:12:15 <TrueBrain> this conversation is too vague for me ;) Let me know if you have any more questions about the CF itself :)
19:12:19 <Zuu> In the binaries there is a embeded version that show in --help in linux, or in the file properties in windows. For branches etc. this should as far as I've understood it be "Development <version>".
19:12:57 <Zuu> If <version> is not the same as finger reports, then OpenTTD Auto Update will think there is a newer version and suggest that it should be updated.
19:13:34 <Zuu> So you can't have "-cd" at one place, but not the others.
19:13:34 <planetmaker> maybe it works for you
19:14:01 <Zuu> Unless I strip of everything after "-" in the version I get from the binaries.
19:15:12 <fonsinchen> planetmaker: that sets the version to a fixed string
19:15:29 <Ammler> fonsinchen: revert after compile
19:15:48 <Eddi|zuHause> Ammler: that isn't possible on the compile farm
19:16:04 <planetmaker> Eddi|zuHause: an not needed either
19:16:35 <Ammler> Eddi|zuHause: yes, is2 made the release commit, asked for compile, after compile tagged and reverted
19:16:44 <planetmaker> the CF just needs to build - with proper filename and version. The rest is repo issues outside the cf
19:17:03 <planetmaker> might be sub-optimal, but works reliably
19:17:22 <Eddi|zuHause> but that's not the point of findversion.sh
19:17:38 <planetmaker> true. But releases are different
19:17:43 <Eddi|zuHause> ... or findversion.vbs
19:18:02 <Ammler> well, the "M" is a bug of git detection
19:18:13 <fonsinchen> I still don't know how the compile farm gets the "cargodist" filename part from findversion.sh
19:18:22 <Eddi|zuHause> yes, and you propose a workaround, instead of working on a fix
19:18:22 <Ammler> fonsinchen: it doesn't
19:18:34 <Ammler> that is set in the setting for the project
19:19:06 <Ammler> with the location of the repo, of the bundles location etc.
19:19:07 <TrueBrain> fonsinchen: that I can answer very simple:
19:19:08 <TrueBrain> # cat 0000018/version
19:19:10 <TrueBrain> cargodist-@@revision@@
19:19:13 <TrueBrain> that is a configuration setting :)
19:19:16 <fonsinchen> so we can just change the setting there and I can commit the removal of modification detection now, I guess.
19:19:30 <TrueBrain> the @@revision@@ part comes via findversion.sh
19:19:56 <Ammler> fonsinchen: patch findversion.sh 3rd field for the release
19:20:17 <TrueBrain> fonsinchen: but we dictate that string
19:21:26 <fonsinchen> we can just change "cargodist-@@revision@@" for "cd-@@revision@@", can't we?
19:23:05 <Ammler> fonsinchen: it would be nice to have cd in the version
19:23:19 <Ammler> so someone can guess the patch from it
19:23:27 <TrueBrain> fonsinchen: cargodist and cargodest are too closed named, so it will remain cargodist as long as the CF produces the binaries
19:24:12 <Zuu> Ammler: but then it wolud be nice if the "cd" appears not just in the binaries, but also in the version string used all over.
19:24:48 <Ammler> Zuu: that would happen, if fonsinchen does patch findversion.sh in that respect
19:26:00 <fonsinchen> In what way shall I patch findversion.sh?
19:26:08 <Zuu> but if not, I guess I'm fine. I can make a hack for the case when the cargodist release target is selected, if I'm just told how to convert a binary reported version into the version reported by the finger url.
19:26:46 <Zuu> eg, removing '-' and everything after that is fairly simple.
19:27:11 <Ammler> Zuu: you are not the only one, who needs to hack
19:27:50 <fonsinchen> I cannot permanently fix the version to a constant string ...
19:27:58 <Ammler> echo "$REV $REV_NR $MODIFIED $CLEAN_REV-cd" for example
19:29:07 <Ammler> fonsinchen: again, that changeset for IS2 releases got reverted right after the CF made the bundles
19:29:47 <fonsinchen> OK, then. I'll do it.
19:30:42 <Ammler> if you don't want to make release commit, you could simply add the suffix -cd
19:30:58 <Ammler> that wouldn't hurt for non releases, would it?
19:31:52 <fonsinchen> you mean in findversion?
19:31:58 <fonsinchen> And that would help?
19:32:13 <Ammler> that is the only difference I see
19:32:24 <Ammler> well except the "M" glitch
19:32:40 <fonsinchen> so I'll do that then
19:32:50 <Ammler> but the "M" is gone now?
19:33:10 <fonsinchen> but will the compile farm still understand it then?
19:33:29 <Ammler> the compile farm does use your findversion.sh
19:33:35 <fonsinchen> I have removed the modification detection in findversion.sh and determineversion.vbs
19:34:09 <fonsinchen> so I better not change the output. Otherwise it will glitch out again.
19:34:15 <Ammler> cargodist-<3rd-findversion.sh-field>
19:36:21 <fonsinchen> if that helps anyone why doesn't that person read the first field of findversion's output instead of the last?
19:36:48 <Ammler> because it wants the clean rev?
19:37:29 <fonsinchen> So now I go and make that "clean" rev "unclean" by adding the branch name.
19:37:38 <fonsinchen> sort of absurd, isn't it?
19:38:03 <Eddi|zuHause> ... i don't understand anymore what you're talking about...
19:38:18 <TrueBrain> Eddi|zuHause: I gave up a few minutes ago too :p
19:38:26 <TrueBrain> if they try to modify the name of the files, they will fail
19:38:38 <TrueBrain> if they try to modify the string displayed in the game, they are doing it in a weird way
19:38:49 <Zuu> I guess the binaries use the "clean" rev in their version info. And if you want "-cd" in there, then you have to properly modify the "clean" rev, or things will break.
19:39:05 <Eddi|zuHause> i have a feeling Ammler, fonsinchen and Zuu are talking about completely different things each...
19:39:21 <TrueBrain> does the left hand ever know what the right hand does?
19:39:47 <fonsinchen> alve@schaf:~/projekte/openttd$ bin/openttd --version
19:39:48 <Zuu> Or at least all three of us have different perspectives.
19:39:48 <fonsinchen> OpenTTD gaf524ac1-smallmap-zoom-in
19:39:50 <Eddi|zuHause> no... but the brain should know what each hand does ;)
19:39:55 <fonsinchen> definitely not the clean version
19:40:35 <Zuu> hmm, yep indeed, I got it backwards.
19:40:39 <fonsinchen> I'll leave it as it is now and push it.
19:41:06 <Zuu> It is in the finger file that it uses the clean version, and that is why matching the finger info with the version from the binary failed.
19:41:18 <Ammler> just run findversion, if 1st field is like 4th, it is fine
19:42:11 <Weeknie> Could someone tell me how pointers to pointers are usefull?
19:42:20 <TrueBrain> Zuu: does your tool work for the head-to-head branch?
19:42:34 <fonsinchen> TrueBrain, would you please build another version now?
19:42:36 <TrueBrain> then I really don't understand the fuzz :p
19:43:02 <Ammler> TrueBrain: head-to-head is no branch, afaik
19:43:08 <Belugas> fuzz : It's a device that generates a rounchy sound, very distorded
19:43:12 <Zuu> head-to-head reports "Development h35be7d1d" as version string while cargodist reports "Development g3c5595c9M-cd"
19:43:30 <TrueBrain> I believe we dismissed those postfixes a while back ..
19:43:33 <Belugas> songs that used the fuzz : Stones : Satisfaction
19:43:47 <Zuu> Yep, but fonsinchen decided to add it manually to his cargodist.
19:43:56 <Alberth> Weeknie: if you want to change a pointer from within a function, you need its address, ie a pointer to the pointer
19:44:02 <Zuu> without making sure that finger also reports the post-fix.
19:44:17 <fonsinchen> the only thing I did now was removing the modification detection
19:44:18 <Ammler> fonsinchen: where is "-cd" from then?
19:44:21 <TrueBrain> Weeknie: open a good C book ;)
19:44:24 <fonsinchen> everything else is stupid
19:44:35 <fonsinchen> that's the branch name
19:44:37 <glx> Weeknie: arrays can be accessed using pointers
19:45:12 <glx> and 2D arrays are arrays of arrays
19:45:13 <Alberth> GRFConfig lists use pointers to pointers
19:46:02 <Weeknie> And I thought pointers were easy:P
19:46:06 <TrueBrain> glx: 2d (or 3d) arrays are accessed by pointers, they have no other way ;)
19:46:22 <Alberth> conceptually, at least
19:46:24 <TrueBrain> Weeknie: pointers are the most difficult concept to understand, but once understood the most easiest to understand :)
19:46:24 <Ammler> fonsinchen: either you remove branch name or you add it to findversion.sh
19:46:38 <glx> autoreplace used to (or still) use pointers to pointers to pointers
19:46:41 <Weeknie> But to remember what yo ucan all do:P
19:46:55 <TrueBrain> char **array_of_strings;
19:47:06 <Ammler> maybe the compilefarm can't handle branches of git
19:47:10 <glx> yup for non fixed size 2D arrays
19:47:21 <Eddi|zuHause> three-star, four-star and five-star programs ;)
19:48:00 <glx> Weeknie: in C you can do almost everything
19:48:02 <TrueBrain> glx: well, only optimizers which are 'clever' enough make 1D arrays from 2D arrays, yes :p
19:48:35 <glx> that's why using pointers when you don't understand them is dangerous :)
19:48:43 <fonsinchen> The compile farm has detected the branch name fine
19:49:06 <TrueBrain> [2010-05-14 19:48:19] Task 0003257 created (git://fickzoo.com/fonsinchen/openttd.git, g9e7e0a6b)
19:49:13 <andythenorth> planetmaker: I've got a compile error with the gui project
19:49:17 <Eddi|zuHause> you can not prove whether an arbitrary C program stops, or two arbitrary C programs produce identical output
19:49:27 <Eddi|zuHause> not even in C ;)
19:49:29 <TrueBrain> somewhere in a dark basement a computer spins up and starts to make this weird noise
19:49:53 <Alberth> Eddi|zuHause: you cannot do that for much simpler languages already
19:49:57 <andythenorth> are pointers difficult?
19:50:07 <fonsinchen> ... just before making this other weird noise and spinning down for the last time.
19:50:25 <Alberth> andythenorth: just like NFO, if you get it, it is trivial
19:50:26 <TrueBrain> fonsinchen: lets hope not :) Would be a bit of a disaster :p
19:50:32 <glx> andythenorth: not really, but incorrect usage can have nice effects
19:50:38 <Eddi|zuHause> Alberth: yes, there are very simple languages that are turing-complete
19:50:46 <andythenorth> I only get it because they're one of Joel's hiring criteria, so I tried to understand them
19:51:01 <Ammler> [21:48] <TrueBrain> [2010-05-14 19:48:19] Task 0003257 created (git://fickzoo.com/fonsinchen/openttd.git, g9e7e0a6b) <-- no -cd
19:51:09 <Eddi|zuHause> brainfuck, for example, has only 8 different instructions
19:51:25 <fonsinchen> Well, I'll go and fetch my hopefully clean laundry. I'll be back in a few minutes.
19:51:26 <TrueBrain> Ammler: not even allowed by the CF
19:51:28 <TrueBrain> we didcate the name
19:52:15 <TrueBrain> but I guess this battle is endless, as it was held from the beginning of automated compiles in OpenTTD .. and we have had many formats
19:52:39 *** devilsadvocate has quit IRC
19:52:40 * andythenorth always wonders with pointers
19:52:56 <andythenorth> if conceptually you understand what something like array.slice does, where's the problem?
19:53:09 <andythenorth> or array.pop etc
19:53:15 <Ammler> well, we needed around 10 IS2 compiles until we found the solution :-)
19:53:16 <Alberth> andythenorth: I always draw pictures to understand the operations
19:53:18 <planetmaker> interesting, andythenorth
19:53:27 <planetmaker> that's nothing I did touch, though
19:53:31 <andythenorth> planetmaker: OS X 10.5.7
19:53:34 <planetmaker> that's the midi interface...
19:53:36 <andythenorth> let me see if trunk broke
19:54:45 <Ammler> it is just bad, that we now need a exception for cargodist only
19:57:14 <Eddi|zuHause> andythenorth: pointers are something that allows heavy optimisations, but can be very dangerous if done incorrectly
19:58:43 <andythenorth> planetmaker: looks like doomsday for my OS X compile :P
19:58:58 <andythenorth> is this the day I stop using openttd?
19:59:21 <Weeknie> I certainly hope not!
19:59:28 <planetmaker> try --without-libtimidity?
19:59:44 <planetmaker> or some other option to disable the music options
20:00:33 <andythenorth> --without-libtimidity didn't solve it
20:00:44 <andythenorth> where do I look for other options?
20:00:56 *** theholyduck has joined #openttd
20:07:37 <planetmaker> I can only assume so far that possibly r19814 broke it
20:07:48 <planetmaker> you could - for a starter - revert that
20:08:00 <planetmaker> that's not a solution but a workaround
20:08:07 *** Kovensky has joined #openttd
20:09:17 <planetmaker> hm, interestingly qtmidi compiles for me
20:09:51 <planetmaker> but still, it worked for you and there's no functional change really
20:10:06 <andythenorth> how do I hg revert when I only have the svn revision number :|
20:11:00 <Ammler> Zuu: I guess, you don't to make a exception for cd only, you could cut -$BRANCH from every version
20:11:42 <planetmaker> hg log -k 'svn rXXXXX'
20:11:59 <planetmaker> ^ that will give you the proper hg revision
20:14:02 <planetmaker> andythenorth: what's your compiler version?
20:14:14 <andythenorth> pdq2s-macbook-3:openttd andy$ gcc
20:14:14 <andythenorth> i686-apple-darwin9-gcc-4.0.1: no input files
20:15:01 <planetmaker> right. It's a compiler thing
20:15:06 <planetmaker> 4.0 doesn't compile it
20:18:14 <andythenorth> what do I have to do?
20:18:47 <planetmaker> src/music/qtmidi.cpp:44
20:18:52 <planetmaker> add a ; to the end of line
20:19:29 <planetmaker> that was forgotten and gcc 4.2 obviously cares shit about a forgotten ; there
20:20:19 <planetmaker> actually not to the end of the line but to the end of the statement
20:20:25 <planetmaker> But I think you know what I mean :-)
20:22:38 <planetmaker> any dev currently here who cares to fix src/music/qtmidi.cpp:44 without me opening a bug report right now?
20:22:58 <glx> remember OSX is not supported?
20:23:29 <glx> feel free to open a bug report, or add it to the OSX bugs list
20:23:33 <planetmaker> statements without terminating ; are not supported by any OS
20:23:40 <planetmaker> it's not an OSX bug. It's a copy&paste bug
20:24:08 <planetmaker> introduced in r19814
20:25:34 <planetmaker> but before we have here again a "not supported rant" ... I'll happily make bug reports for any typo separately
20:25:55 <glx> that's an easy to fix bug
20:26:02 * andythenorth thinks "if it doesn't compile I can stop doing all this work...."
20:26:07 <planetmaker> that's what I outlined above before I asked
20:26:10 <glx> kind of regression fail ;)
20:26:14 <andythenorth> "....and have my weekends back" :P
20:26:49 *** JostVice has joined #openttd
20:26:57 <PeterT> Zuu: isn't this what you need to set up auto-updating for IS2 binaries?
20:28:43 <planetmaker> that's how my local diff looks like, yes
20:29:03 <planetmaker> then gcc4.0 didn't complain anymore. Funnily gcc4.2 didn't bother
20:29:40 <CIA-6> OpenTTD: glx * r19822 /trunk/src/music/qtmidi.cpp: -Fix (r19814): missing semicolon
20:30:07 <glx> that's what happens when we can't compile ;)
20:30:17 <Ammler> [22:26] <PeterT> Zuu: isn't this what you need to set up auto-updating for IS2 binaries? <-- what you think, how long that exists?
20:30:27 <PeterT> planetmaker and andythenorth are your new bug testers
20:30:37 <PeterT> Ammler: I didn't mean it like that
20:30:47 <glx> we'd prefer have maintainers instead
20:30:55 * peter1138 boggles... a project that uses CVS
20:30:58 <PeterT> What I should've said, Is this something you could use for ottdau?
20:31:50 <planetmaker> it was setup for him, I think, right?
20:32:11 <planetmaker> like 2 years ago?
20:32:28 <andythenorth> planetmaker: so how do I merge that change into your branch (or should I just do it manually?)
20:32:49 <planetmaker> andythenorth: do it manually for now
20:32:53 <PeterT> Those are IS2 and cargodist versions
20:33:00 <planetmaker> it will be in the repo with the next trunk merge
20:33:22 <Ammler> we added that tonight, so just be a bit patient please
20:33:30 <PeterT> IS2 and cargodist auto-updating is not in OTTDAU
20:33:59 * andythenorth thought the diff could be applied, but it fails :P
20:34:38 *** fonsinchen was kicked by DorpsGek (Wrong channel. Retry in #openttdcoop.)
20:35:02 <PeterT> anyone know what supybot plugin does that?
20:35:02 *** fonsinchen has joined #openttd
20:35:48 <andythenorth> PeterT: I'm not trying to compile trunk
20:35:54 <Zuu> PeterT: Huh? IS2 was added earlier tonight.
20:36:09 <Zuu> And worked fine back then when I tried.
20:36:20 <PeterT> Sorry, I haven't used it in a while
20:36:31 <PeterT> I'm just remebering what it had since I last used it
20:41:41 <peter1138> buhhh cvs up doesn't appear to ... work
20:42:07 <Zuu> Hmm, is '-' a special char in regular expressions?
20:42:29 <Zuu> Or do I need to enclose it with []?
20:43:21 <fonsinchen> Thanks again for building them.
20:44:36 <TrueBrain> fonsinchen: thank the people who donate the money to allow us to have such a beast of a machine :)
20:44:55 <TrueBrain> (still it is too small)
20:45:44 <peter1138> buhhh cvs failing again
20:45:57 <TrueBrain> cvs? Does that still exist?
20:47:45 <andythenorth> planetmaker: does your compile of the gui branch crash on start?
20:48:49 <planetmaker> what's the repo version you use?
20:49:30 <Zuu> "Last Cargodist" should now work for OpenTTD Auto Update, unless you happen to be on the minor server which will shortly also be updated.
20:51:53 <andythenorth> NOT REACHED triggered at line 139 of main_gui.cpp
20:52:39 <planetmaker> by just starting it? Without clicking anywhere?
20:53:27 <andythenorth> I reverted the repo earlier, let me delete and clone again to be sure
20:53:59 <planetmaker> that's not how my line looks like
20:54:09 <planetmaker> it's a trunk bug which was already fixed which you discovered
20:54:33 <planetmaker> maybe it's needed :-)
20:54:40 <planetmaker> I added two heads to the repo
20:55:05 *** WizzleBLincoln has joined #openttd
20:55:27 <Belugas> TONIGHT TONIGHT TONIGHT... HOOO Hooo...!
21:03:09 *** Chillosophy has joined #openttd
21:16:19 <andythenorth> planetmaker: successful compile, no assert :)
21:19:42 <andythenorth> planetmaker: do you see drawing artefacts on tabs?
21:21:39 <Weeknie> Little question, what are all the things that you can't pass to a function dereferenced?
21:23:49 <Weeknie> Erm, well, like arrays, you can't pass them to a function unless you use a pointer, right?
21:24:09 * andythenorth thinks 'map edges' could be improved
21:24:47 <peter1138> C only has pass by value
21:24:58 <andythenorth> planetmaker: we'll find a nice way to choose between height map / random seed...
21:25:22 <planetmaker> random seed is also needed for height map
21:25:23 <peter1138> and the 'value' of an array is a pointer to its first element
21:25:37 <peter1138> (not strictly correct wording, but... i can't be arsed to look it up ;))
21:25:52 <andythenorth> planetmaker: is it? I learnt some thing new :)
21:26:13 <planetmaker> Pretty sure. Industry placement and alike
21:28:03 <andythenorth> from me? not tonight
21:29:14 * andythenorth wonders. Often the words are better than the pictures. But some of the map gui might be better with pictures....
21:30:18 <Weeknie> Lol, andy, you sound like you're writing a book this way:P
21:30:57 <andythenorth> very flat / flat / hilly / mountainous being one example
21:30:57 <andythenorth> is choosing the tree algorithm still relevant?
21:31:03 * andythenorth doesn't even know the difference.
21:31:18 <andythenorth> I just use 'improved' because it sounds best
21:33:39 <andythenorth> "TREE ALGORITHM This is how the trees are aligned. Keep it where it is, or it wont look good, unless you don't want any trees, which is 'None'."
21:34:18 <andythenorth> planetmaker ^ is the tree algorithm a choice players don't need?
21:35:22 <planetmaker> none might be a viable option. So a keep
21:36:05 <andythenorth> "Trees: Yes / No" ?
21:36:05 * andythenorth does ottd have radio buttons?
21:36:10 * andythenorth doesn't remember any
21:39:06 * andythenorth contemplates a game with the original map generator, it looks kind of nice
21:39:10 <Eddi|zuHause> there is a problem with trees, there are two different issues: the initial tree placement, and the tree growth during the game
21:42:30 <andythenorth> does map generator (original / terragenesis) need to be a choice in map gen screen?
21:43:26 <Weeknie> I don't think any average openTTD player will notice, will he/she?
21:43:40 <Weeknie> notice as in notice the difference between the 2 generators
21:43:48 <ccfreak2k> andythenorth, it seems the closest analog to radio buttons in OpenTTD are left/right buttons.
21:43:59 <andythenorth> planetmaker: "Land generator" could move to advanced options?
21:44:40 <planetmaker> it makes a big difference. And is directly a map option
21:44:47 <planetmaker> I wouldn't move anything away from there
21:44:59 <andythenorth> for each item shall I suggest moving and you answer "no" ? :P
21:45:14 <andythenorth> ok well that saves time then :)
21:45:22 <Weeknie> He said yes, quick, perhaps you can smuggle one in!
21:45:55 <andythenorth> it's quite a lot of....stuff in that dialog
21:46:01 <andythenorth> and it got to be more recently
21:46:17 <andythenorth> variety distribution + 5 buttons for map edges
21:46:34 <Weeknie> That variety distribution, what does it do
21:46:38 <Weeknie> Make sure things are spread out?
21:46:49 <andythenorth> adds variety....
21:47:13 <andythenorth> depending on other settings it will cluster mountains, add noise to sea shores etc.
21:47:38 <andythenorth> it can interact a bit unpredictably with 'Terrain type' and 'Smoothness'
21:48:29 <planetmaker> It gives nice results with "very large" and not too lengthy maps
21:51:00 * andythenorth wants to take some options out the back and shoot them
21:51:13 <andythenorth> however, if that's not permitted, perhaps it's time for bed :)
21:56:46 <Zuu> Bed sounds like a good idea either way. :-)
21:58:09 <Eddi|zuHause> last time i checked, results were "unpleasant" with very non-square maps (2:1 ratio seemed fine, but 4:1 not)
22:04:04 <planetmaker> maybe it should compensate for that somehow...
22:04:23 <planetmaker> the 2:1 stretch can actually look nice as it gives elongated mountain ranges
22:05:58 <planetmaker> anyway... good night
22:08:08 *** Devedse_ has joined #openttd
22:30:50 *** Weeknie has joined #openttd
continue to next day ⏵
