IRC logs for #openttd on OFTC at 2011-08-17
⏴ go to previous day
00:02:41 <KittenKoder> Okay, time to learn breaking the nml files into sections. :p
00:14:42 <Eddi|zuHause> ask planetmaker to help you set up the "standard" makefile?
00:15:39 <KittenKoder> I'm going to have to. >.< I'm getting lost in other people's code and I'm not good at reading other people's code.
00:15:48 * KittenKoder thus why not a good team player normally.
00:58:53 <KittenKoder> I found a copy of it.
01:01:46 *** Eddi|zuHause has joined #openttd
01:08:17 *** mos|work has joined #openttd
01:23:07 *** Eddi|zuHause has joined #openttd
01:55:23 <KittenKoder> planetmaker, are you active now?
02:03:22 *** lessthanthree has joined #openttd
02:14:06 *** lessthanthree has joined #openttd
02:26:03 *** rhaeder1 has joined #openttd
04:48:11 <pjpe> is it just me or does the openttd web server through a 502 every night at around this time
04:56:18 *** Eddi|zuHause2 has joined #openttd
05:01:21 <KittenKoder> Probably system maintenance or something.
05:01:48 <KittenKoder> Though not on the OTTD web server, necessarily.
05:03:31 <pjpe> lessthanthree it's back up
05:37:32 *** sla_ro|master has joined #openttd
05:46:26 *** JVassie has joined #openttd
06:04:19 *** DayDreamer has joined #openttd
06:12:06 *** Progman has joined #openttd
06:22:42 *** Br33z4hSlut5 has joined #openttd
06:37:40 *** Cybertinus has joined #openttd
06:54:36 <KittenKoder> I was needing you a while ago ... but now I think I have a solution that works better for me. :p
06:55:31 <KittenKoder> Now, I just need to figure out the custom tags part.
06:56:57 <planetmaker> "my" makefile framework has the advantage that it it understood by the DevZone's compile farm, should you want to host and (nightly) build your NewGRFs there
06:58:04 <planetmaker> and it automatically takes care of versioning of NewGRFs. Which IMHO is the most important part
06:58:35 <planetmaker> using the repo's version as newgrf's version, but allowing a tag to act as displayed release version
06:59:00 <KittenKoder> Yeah, but I have problems with it, primarily, it doesn't work for me.
06:59:07 <planetmaker> and adding install and bundle targets for easy use
06:59:16 <planetmaker> wow, that's a good error description
06:59:34 <KittenKoder> Well, I bumble around, get one thing to work, then I get another problem.
07:00:03 <KittenKoder> Last one "no rule definition for some.png file"
07:00:36 <KittenKoder> I don't like editing or modifying another person's source, thus why I have never been a great team player.
07:01:18 <planetmaker> either you're mixing case or that shouldn't happen
07:01:40 <KittenKoder> Not mixing cases, checked all that.
07:02:12 <KittenKoder> The file I am using as a test case is really the first NML I wrote, just broken into a few files so I can test it.
07:02:36 <planetmaker> that's what the framework with the makefile is made for
07:02:43 <planetmaker> it expects *.nml and *.pnml files
07:02:55 <planetmaker> and searches their includes for further deps
07:03:17 <KittenKoder> Other than that, almost no changes to it were made. The Makefile.config I modified for it, had to change a lot of other values before I got that far, then the png rule error hit and I just got annoyed.
07:04:29 <KittenKoder> So, I'm just making a simple Makefile. ;) It's not you, just other people's source code has never worked right for me unless I use it unchanged, and I mean completely unchanged.
07:04:59 <planetmaker> sucks to be you then ;-)
07:05:24 <KittenKoder> I can copy-pasta bits, or use it as examples, so it's still good.
07:06:00 <planetmaker> I still would like to see why it failed for you... do you have a zip of the dir?
07:06:48 <KittenKoder> Not the one that was causing the error anymore, sorry, it's just such a common event for me I just started over on reflex.
07:07:59 <KittenKoder> The swedish rails original file works almost perfectly, but it's outdated compared to the new nmlc and has a few errors in it from that, but the Makefile works fine as long as I don't change anything for that.
07:08:31 <KittenKoder> The problem with make that has always bothered me, the very useless error reporting.
07:09:22 <planetmaker> you can change all code in it...
07:09:32 <planetmaker> appe: it's a trackset
07:09:49 <KittenKoder> I see where you write to the tag file, but I don't see where you create it .... but I'm REALLY behind in makefile code, too many years of IDEs.
07:10:24 <planetmaker> replacing the two default rail tracks with "better" tracks, thus that they better work with all kinds of ground
07:10:38 <planetmaker> and roads and so on :-)
07:10:54 * appe is swedish and would like to compare it.
07:11:11 <KittenKoder> It's nice .... not my personal favorite, but the source archive is a great learning tool.
07:11:16 <planetmaker> it's all drawn by a Swede ;-)
07:11:28 <Markk> appe: Are you sure about that? :o
07:11:34 <Markk> appe: You're from Småland you know.
07:11:54 <appe> Markk: smland = sverige.
07:11:54 <planetmaker> KittenKoder: it was the first NML NewGRF ever ;-)
07:12:04 <KittenKoder> That explains it.
07:12:33 <KittenKoder> Gathering it was also a testing ground.
07:12:50 <planetmaker> yes. It needs indeed some code updates
07:13:01 <planetmaker> maybe when I come around to re-touching tunnels
07:13:13 <planetmaker> my graphics artist vanished :-(
07:15:24 <KittenKoder> I'm more of a code explorer than anything.
07:15:29 <KittenKoder> Any chance to learn something new.
07:16:15 <KittenKoder> Ironically, the code I use for work most often now was only because I was "oooh, new languages to play with."
07:17:47 <KittenKoder> Why can't I find where you invoked the rule for tags!?!?!?! >.< LOL
07:20:54 <KittenKoder> Wait, I see it now.
07:23:04 <planetmaker> patches in principle are welcome
07:25:04 <KittenKoder> I've been up too long again.
07:28:09 <KittenKoder> My neighbors I think are having a noise war ... and it's starting to grate my nerves.
07:41:29 *** Belugas has joined #openttd
07:41:29 *** ChanServ sets mode: +o Belugas
07:45:04 <KittenKoder> Trying to think how I will do the maglev for Fantasy Rails.
07:46:27 <KittenKoder> My favorite idea would make a lot of the other maglevs not quite fit in with the set.
07:47:06 <KittenKoder> Using two different tracks for the three different propulsion systems (like in reality) ....
07:47:51 <KittenKoder> But then I'd have to define one as a "default" for the levs not defined in such a manner.
08:13:10 *** Eddi|zuHause2 is now known as Eddi|zuHause
08:24:39 <KittenKoder> Anyone know where I can find the stats for the maglev tracks used in OGFX?
08:27:05 <KittenKoder> Price, speeds, multipliers, etc.
08:28:04 <Pinkbeast> Tracks don't have much in the way of stats as a rule
08:28:16 <planetmaker> I guess openttd source... src/table/...
08:28:43 <KittenKoder> I want to make the original tracks I designed as a secondary type instead of replacing the default ones.
08:29:01 <Pinkbeast> You might try looking at what NuTracks does
08:29:22 <KittenKoder> There's speed limit, curve multiplier, and cost.
08:29:24 <Pinkbeast> ... but the total number of track types is absurdly small (why?) so it would be hard not to be incompatible with them.
08:29:54 <KittenKoder> I think NuTracks is over board on types, personally. >.<
08:30:47 <Pinkbeast> Well, I appreciate them representing the different electrical systems for simulationists.
08:30:48 <pjpe> i don't get what the nutracks subway tracks are supposed to do
08:31:01 <Pinkbeast> pjpe> IIRC, look like houses.
08:31:28 <pjpe> last time i tried it just made parts that look like a city
08:31:28 <KittenKoder> But for my fantasy rail set, I want two types of maglev tracks.
08:31:30 <pjpe> isn't that what a city is for
08:31:48 <Pinkbeast> Kitten> And the "fast-expensive" or "slow-cheap" choice is a meaningful one, but it's too discrete, but can it not be without excessive simulatonism
08:32:16 <Pinkbeast> pjpe> Well, not after you build tracks through it, given OTTD's... lacking... support for underground railways.
08:32:40 <pjpe> oh trains can still go on it?
08:32:45 <KittenKoder> The maglev track idea is that one of the tracks will be specifically for fantasy only trains.
08:32:51 <pjpe> looks like it would just stop a train
08:32:59 <Pinkbeast> pjpe> Yeah, that's the idea. It's a bodge at best.
08:33:43 <Pinkbeast> Kitten> At least I and others talked the NuTracks guy out of having the not-totally-slow track start in 1920.
08:34:09 <KittenKoder> Or I could make a real maglev set that distinguishes between them ... but that may be a lot of extra work. >.<
08:34:53 <Pinkbeast> Really it would be nice if whatever limits the binary to 16 track types was fixed # ahem but I am not fixing it, I am presently trying to comprehend YAPF
08:35:36 <KittenKoder> I have dealt little with pathfinders.
08:35:59 <Pinkbeast> I'm not joking, the thing I most want fixed is in YAPF. However, I'm a Perl programmer and my understanding of C++ is limited. :-(
08:36:23 <KittenKoder> I'm also really bad at understanding other people's code.
08:36:45 <Pinkbeast> Kitten> so am I when it has template programming in, that stuff is deep magic
08:36:49 <KittenKoder> I think YAPF is actually awesome, myself, I have yet to confuse it ... and I have tried.
08:37:07 <Pinkbeast> pjpe> Presently when a train wants automatic servicing it goes to the nearest depot.
08:37:21 <Yexo> Pinkbeast: that has nothing to do with YAPF
08:37:37 <KittenKoder> Isn't that part of the core?
08:37:40 <Yexo> and if you want to prevent that just use at least one "service at depot" order
08:37:48 <Yexo> it's part of the core, yes, but not of yapf
08:37:53 <Yexo> yapf being only the pathfinder part
08:38:08 <KittenKoder> Yeah, OTTD can use other pathfinder programs.
08:38:18 <KittenKoder> Just that there really is no need with YAPF.
08:38:20 <Pinkbeast> It should instead consider both the train->depot and the depot->next destination issue
08:39:33 <Pinkbeast> Yexo> "service at depot" orders are kind of a reflection of the way automatic servicing leads to comedy # and some timetable separation patches are broken with it although yes that is their bug to fix
08:40:08 <Pinkbeast> Kitten> So ideally some kind person would implement in every pathfinder "find depot with net lowest total journey" but I'd settle for doing it in YAPF
08:40:24 <KittenKoder> >.< Pooh, that doesn't have it either.
08:41:02 <Yexo> Pinkbeast: that is not a complete solution, since when a slow train requires servicing "now" it should travel another 1000 tiles just because it'll make the total journey a few tiles shorter
08:41:17 <Pinkbeast> Yexo> Aha, I thought of that. :-)
08:41:19 <Yexo> it needs servicing say within 50 tiles, and than again after another 500, and again after 1000 etc.
08:41:22 <KittenKoder> Sorry, trying to find the defaults for the maglev track still.
08:41:50 <Pinkbeast> You would add a weighting factor to make the pre-depot journey cost more than the post-depot journey...
08:42:17 <planetmaker> KittenKoder: "no need" might be a bit strong. Actually a path finder which is both deterministic, threadable and efficient would be nice ;-)
08:42:35 <Pinkbeast> ... and also appreciate that, yes, things can go wrong with this setup, but presently they go wrong with a train taking a journey up the one-way line to arseendofnowhereville.
08:42:44 <KittenKoder> Okay, no pressure then? :p
08:42:56 <Pinkbeast> In particular in my proposed approach adding depots can never make things worse, whereas now it can.
08:43:22 <KittenKoder> Pinkbeast, I have found a few track techniques that actually help out with that.
08:43:41 <KittenKoder> One is forced depot stops, just after a full load command.
08:43:50 <Pinkbeast> Kitten> Yes, so have I! That is part of why I am so frustrated as to want to learn C++ just to fix it!
08:44:15 <KittenKoder> The other is putting a reversible PBS facing the "wrong" way, which discourages trains not needing that track from going there.
08:44:29 <KittenKoder> Pshaw ... c++ is easy.
08:44:39 <Pinkbeast> Kitten> Indeed. But wouldn't it be nice not to have to do these things?
08:44:42 <KittenKoder> You can complain when you have to learn Java. :p
08:45:01 <dihedral> what's wrong with coding java?
08:45:02 <KittenKoder> *shudder* I loath Python's syntax.
08:45:13 <KittenKoder> Java is the least forgiving language ever.
08:45:19 <Eddi|zuHause> i love python's syntax :p
08:45:19 <SpComb> what I'd like would be lookahead for routing on station orders
08:45:27 <dihedral> and why should a language be 'forgiving'
08:45:29 <KittenKoder> One bad call and the compiler spits it out saying "what's wrong with you?" :p
08:45:33 <dihedral> you are telling a machine what to do
08:45:40 <SpComb> i.e. if you have a station with different platforms that have different routing, pick a platform that lets the train continue on to its next destination
08:45:42 <KittenKoder> I like Java, just saying, it's hard for beginners.
08:45:44 <dihedral> and you expect the machine to forgive faulty syntax?
08:45:57 <Pinkbeast> SpComb> Well, waypoints, but yes.
08:46:00 <KittenKoder> Well, actually I LOVE Java, it's my preferred language.
08:46:00 <pjpe> java ain't hard for beginners
08:46:06 <SpComb> waypoints can be hard to fit in
08:46:11 <dihedral> i dont think java is hard for beginners either
08:46:13 <SpComb> if you have cramped layouts
08:46:20 <Pinkbeast> SpComb> I don't disagree with you!
08:46:36 <pjpe> then again i can't think of anything i'd think is hard for beginners
08:46:42 <Pinkbeast> That there is a workaround doesn't mean I don't wish it wasn't necessary (as with servicing)
08:46:50 <SpComb> of course, you could have asinine counterexamples where you'd need more than one step of lookahead
08:46:55 <KittenKoder> pjpe, when you are an old school coder who used c and c++ for too long ... it can be.
08:47:02 <SpComb> but it would be useful in places!
08:47:10 <Pinkbeast> Particularly if you combine it with conditional orders.
08:47:17 <dihedral> KittenKoder, that's just because the person is thinking in too small a box ;-)
08:47:27 <SpComb> s/asinine/pathological/
08:47:28 <KittenKoder> Can't argue that.
08:47:53 <KittenKoder> Gotta say though, this is nice, no language elitists here.
08:48:03 <Pinkbeast> Well, Perl is best, obviously. # joke
08:48:48 <Pinkbeast> I mean, I love Perl programming, but it's like Frank Cooper's, I don't expect everyone else to like it. :-)
08:49:22 <KittenKoder> I was in one programming channel where if you mentioned anything but ansi-c ... you got attacked.
08:50:01 * Pinkbeast occasionally still reports bugs in NetHack. _ANSI_ C? New-fangled nonsense. :-)
08:50:23 <KittenKoder> You still play NetHack?
08:50:41 <Pinkbeast> Well, these days, Derek Ray's "SporkHack" variant, but yes.
08:51:23 <Eddi|zuHause> "you still play Transport Tycoon?"
08:51:34 <Pinkbeast> Not so much, though. I play a lot of Kornel Kzhardtospell's marvellous DoomRL
08:51:56 <Pinkbeast> And Jeff Lait's POWDER (and any roguelike by Lait is worth a look)
08:52:47 <KittenKoder> >.> Sheesh, get to a hospital, you have a broken bone.
08:53:28 <KittenKoder> Can't find it. >.<
08:58:46 <KittenKoder> NuTrack has it at 26667 ... isn't that a bit high since it's a multiplier?
08:59:33 * Pinkbeast assumes his trademark blank look. :-O # sorry, I don't understand
09:00:01 <KittenKoder> The cost multiplier for the maglev track.
09:00:26 <Pinkbeast> NuTrack does like to make high speed tracks _very_ expensive, dunno if they mean that.
09:01:03 <KittenKoder> Because the example track in the NML says 32 is "steep" for the costs.
09:01:59 <Pinkbeast> Not to be Captain Obvious, but seen if NuTracks does actually charge a million a mile for maglev?
09:03:00 <KittenKoder> That's where I'm lost, the math in my head doesn't coincide.
09:03:34 <Pinkbeast> Well, while I'm being the cardboard programmer, what are the multipliers for other track?
09:03:52 <planetmaker> Pinkbeast: nutracks means to be very expensive
09:04:37 <KittenKoder> Okay, test is 46 per mile of maglev .... see ... math doesn't add up. >.<
09:04:47 <Pinkbeast> planet> yabbut nutracks slowest ain't 47 times the cost in vanilla, which I think is key to Kitten's issue
09:05:31 <KittenKoder> The highest speed electric comes to 9,346.
09:07:49 <KittenKoder> I think another of my GRFs is interfering with NuTracks, because the highest rail seems correct.
09:08:09 <KittenKoder> 9364 with a cost value of 9334.
09:09:51 * KittenKoder is actually more confused now really.
09:11:37 <KittenKoder> Okay ... maybe I'll just ditch this whole line of thought,
09:12:19 <KittenKoder> Some aspects I just can't find enough information on ....
09:12:42 <KittenKoder> Though ... come to think of it, I could just make a NML file and play with values there.
09:12:48 <Pinkbeast> It's struck me as one of the odd things about OTTD, as free games go
09:13:25 <KittenKoder> Not really, Open Source devs have a VERY hard task when it comes to documentation maintenance.
09:13:41 <KittenKoder> It's not easy with a lot of contributions to keep everything in full detail and up to date.
09:14:00 <Pinkbeast> Even so, OTTD is unusually full of magic.
09:14:26 <KittenKoder> The program devs do a great job at what they do.
09:14:49 <Terkhen> svn log | grep -i Magic
09:14:50 <SpComb> you can set the base cost that the multiplier applies to
09:14:53 <KittenKoder> ... and with the NewGRF ... almost impossible to expect that to be complete.
09:15:00 <Pinkbeast> I think you're responding to a criticism I wasn't making.
09:16:05 <KittenKoder> SpComb, I almost figured it out, but it slipped out of my brain for some reason.
09:16:10 <KittenKoder> The math is just not working for me.
09:17:17 <KittenKoder> Is there a conversion between pounds and dollars?
09:17:51 <Terkhen> in OpenTTD? 1£ == 2$ IIRC
09:17:53 <Pinkbeast> Er... yes, all the other currencies have hard-configured conversions to pounds. # perhaps that was not your question
09:18:22 <KittenKoder> Okay .... that's part of my problem.
09:18:30 <KittenKoder> I thought it just changed the display. >.<
09:18:51 <Pinkbeast> Er, I think it does - the internal game arithmetic is always in pounds.
09:19:04 <Yexo> internal game is always in pounds, yes
09:19:31 <KittenKoder> There's a lot of my confusion's source, I didn't think of that until just now.
09:20:59 <KittenKoder> Even then, I'm still thinking of ditching the idea of adding another track type anyway. LOL
09:21:41 <KittenKoder> Unless I make one monorail+maglev compatible ... but that may unbalance things a lot.
09:22:58 <KittenKoder> ... as well as complicate matters .... forgot about acceleration.
09:42:07 *** Biolunar has joined #openttd
09:43:18 *** Sc00by22 has joined #openttd
09:51:50 <KittenKoder> Not going too high on the detail for 3D modeling is hard when you're use to high res.
09:52:29 <Pinkbeast> You should tell that to the P1SIM guy # sorry
09:53:19 <KittenKoder> I am trying to remember that when taking the 3D model down to an 8bbp sprite, detail becomes dots.
09:53:36 <Pinkbeast> Someone should tell that to the P1SIM guy # oh oops again
09:54:26 <KittenKoder> Okay, gotta ask, who?
09:55:52 <Pinkbeast> He wants to implement something - alone - that will be 80000 times better than OTTD by simulating EVERYTHING
09:56:42 <Pinkbeast> Now admittedly that worked for the Dwarf Fortress guy but 1) he's very unusual 2) he worked for ages before releasing anything 3) he did have to live on instant ramen for years after that before the money came in
09:57:48 <blathijs> He does draw pretty pictures
09:58:01 <blathijs> and actually has code already, it seems
09:58:49 <Pinkbeast> blathijs> Some code, yes, but I have code to make an @ walk round the screen and I ain't claiming to have written a roguelike yet
09:59:18 <Pinkbeast> As some OTTD dev put it in the thread, OTTD's interface code alone is a pretty awesome one-man job.
09:59:27 <blathijs> Lots of pretty pictures, even
10:00:04 <KittenKoder> Well, I'm a bit of a realist.
10:00:30 <blathijs> heh, "It would take many years to make a fork of ottd so that it will work like the concept i decribed."
10:00:41 <blathijs> At least he has some realism on his side as well
10:00:54 <Pinkbeast> Kitten> I do, around page 38, try and encourage him to do something one man can do
10:01:15 <KittenKoder> I didn't read that far. :p
10:01:35 <blathijs> And looking at the vast amount of detailed pictures he drew up, he does have plenty of time on his hands :-)
10:01:37 <KittenKoder> Kinda busy trying to make a low res version of the Ghost of Mars train right now.
10:01:37 * peter1138 ponders attempting to migrate stuff from vserver to lxc
10:02:00 <blathijs> peter1138: I'm in the middle of that right now, so feel free to shoot questions
10:02:29 <peter1138> network setup seems vastly more complicated :S
10:02:30 <KittenKoder> Which is better, long or tall engines?
10:02:47 <blathijs> peter1138: Also, realize that lxc is not as foolproof as vserver right now (it does isolation, but there are still parts that are not isolated or secured)
10:03:10 <KittenKoder> I have seen some long engines, but they don't look right taking turns, though they are nice otherwise.
10:03:34 <Pinkbeast> Kitten> er if they are fictional why have too much of either
10:03:47 <blathijs> I like the track layout idea of p1sim, though. I once designed something like it for ottd :-)
10:03:49 <peter1138> well, they're all for service separation rather than dishing out to untrusted customers etc...
10:04:15 <blathijs> peter1138: If you have a functioning vserver system right now, you might want to stick with that for a while (though it's fun playing with lxc, of course)
10:04:28 <KittenKoder> Pinkbeast, I am trying to work out balance .... it's just sometimes it's hard.
10:04:49 <blathijs> peter1138: I'm setting up a new server right now, so I thought to save me the migration trouble later on by using lxc right now
10:05:04 <peter1138> yeah, i'm setting up a new server
10:05:30 <peter1138> basically i installed it as vserver
10:05:40 <peter1138> *then* read that vserver's being deprecated :S
10:05:43 <blathijs> peter1138: And indeed, my goal is mostly separation to make management more easy and slightly reduce the impact of a security breach, not give root to untrusted people
10:05:55 <blathijs> peter1138: Same here
10:06:15 <blathijs> peter1138: Or rather, I knew vserver was to be deprecated, but only found out later that lxc was somewhat usable already
10:06:47 <blathijs> peter1138: Though I think that the vserver kernal patch will be deprecated by Debian, but it's likely that the vserver userspace tools will end up using lxc kernel utils eventually
10:06:49 <Pinkbeast> I like many of the ideas in P1SIM (although not the "simulate everything and see what drops out" idea, that worked once for Dwarf Fortress) but he'd do better hacking on OTTD
10:06:56 <blathijs> peter1138: Not sure about the timeframe, though
10:07:45 <peter1138> lxc seems to be lacking in the shutdown procedure
10:17:58 <blathijs> peter1138: I've wondered about that. shutdown -h now seems to work, but I usually use lxc-stop to stop a container. I suppose that just sigterms all processes or something
10:18:17 *** AlexLivingstone has joined #openttd
10:18:28 <peter1138> yeah, lxc-stop doesn't do a clean shutdown
10:18:42 <blathijs> peter1138: And you can't seem to run a single command within a container if it's already running, so not sure how to script the shutdown -h now
10:18:45 <blathijs> perhaps over SSH ;-)
10:19:21 <blathijs> I wonder if something like "echo ctrl-alt-del | lxc-console" exists
10:19:38 <blathijs> or sending an ACPI power button signal :-)
10:20:12 <peter1138> (that's a year ago mind you)
10:20:46 <blathijs> seems the lxc-stop is no longer neede right now
10:22:40 <peter1138> i wonder how libvirt handles it
10:23:42 <blathijs> "chroot /var/lib/lxc/vm0/rootfs telinit 0" seems to do the trick
10:24:15 <blathijs> (which is a bit of a hack, but takes advantage of the fact that telinit just dumps something in /dev/initctl I think)
10:24:38 <peter1138> and /dev/initctl isn't actually a device node
10:25:03 <blathijs> I was just wondering about that, yes
10:25:05 *** Brianetta has joined #openttd
10:25:08 <blathijs> just a FIFO created by init I guess?
10:25:18 <peter1138> should be in /var/run ;)
10:25:30 <blathijs> hysterical raisins, I guess
10:25:45 <peter1138> now... can you do lxc without the network bridge
10:26:02 <peter1138> i find promiscuous mode disconcerting, heh
10:26:39 <blathijs> though I had a look at that and then eventually just went for the bridge configuration
10:27:09 <blathijs> since that static approach requires configuring a (possibly identical) address for every veth pair on the host side
10:27:29 <peter1138> hmm, i see. painful
10:27:38 <blathijs> peter1138: Just using a veth pair and putting that in a host-side bridge seems to work
10:28:12 <blathijs> peter1138: I also coded up a patch for lxc-start to setup the default gateway inside the guest (autodetected from the bridge address), so the guest doesn't need to do any networking configuration anymore
10:30:35 <KittenKoder> Pink = CC1, green = CC2
10:31:12 <Pinkbeast> I think it would be more helpful to show it in a variety of CC, frankly # or at any rate less alarming ones
10:31:20 <Pinkbeast> But it looks basically good
10:31:52 <KittenKoder> Well, that's the rough render, and the CCs change to the defaults when I do my palette trick.
10:33:27 <Pinkbeast> Kitten> It's an awkward shape. In the future I hope we can expect trains to be either super streamlined (like the new shinkansens) or bluff freight full of brute force.
10:33:54 <Pinkbeast> But I don't think that design is bad # but I'd like to see it in more CC
10:33:58 <KittenKoder> Thus why I'm calling this "Fantasy" instead of sci-fi.
10:34:14 <KittenKoder> Not all sci-fi is realistic, and most is pretty fugly. :p
10:34:52 <Pinkbeast> Well, in fantasy, we could have designs like UKRS1's 7AT steam locomotive # ahem sorry
10:34:53 <KittenKoder> But hmm ... lost a little more detail than I expected ... need to tone down the shading stuffs or use less variation in shading colors.
10:35:33 <Pinkbeast> But more seriously, in fantasy, it should exude power, with grilles over vast engines # inasmuch as we can do in so few pixels
10:36:09 <KittenKoder> Yeah, that's the balance I'm trying to work out, the detail levels, right now.
10:36:17 <Pinkbeast> LOL> I dunno, if I won umpteen million pounds I would build the 5at
10:36:47 <Pinkbeast> Well, again, I dunno about the CC, but that design does seem to have a certain brute something.
10:36:51 <KittenKoder> I need to choose a better base color.
10:37:24 <KittenKoder> It's a long distance passenger transport in the movie.
10:37:53 <KittenKoder> Maglev, but slower than most diesel. >.< LOL
10:38:26 <KittenKoder> I think I need to put more red in the texture so it converts better, it goes all grey.
10:40:20 <KittenKoder> Ooops, figured out why the CC amount was so low.
10:40:30 <KittenKoder> Forgot to change the specular values.
10:44:13 <KittenKoder> Thinking of doing the unit style.
10:45:38 <Pinkbeast> No, that looks much worse.
10:46:50 <Pinkbeast> The grey with 2CC stripes one looked better.
10:48:02 <KittenKoder> Well, the grey would fit other sets better.
10:50:54 <KittenKoder> How about darker grey?
10:52:24 <KittenKoder> Refresh it, I think this time I got it.
11:00:33 <peter1138> just freed up 1.7GB of ram
11:00:35 <peter1138> by closing chrome :S
11:38:59 <KittenKoder> The blue on the train itself is CC1.
12:07:56 <Eddi|zuHause> sprites are difficult to judge on their own
12:10:13 <KittenKoder> The problem is that all the others I have seen everything looks longer from the side angles than the diagonal ones
12:11:00 <Eddi|zuHause> yep, that's a general problem
12:11:00 <KittenKoder> But mine keep winding up the same.
12:11:09 <KittenKoder> Is it because of using a 3D rendering?
12:11:18 <Eddi|zuHause> the - view must be stretched about 40%
12:12:09 <KittenKoder> Wow, that's a lot. >.<
12:14:22 <Eddi|zuHause> total width should be 32px
12:15:15 <Eddi|zuHause> yours looks more like 24px
12:15:40 <DorpsGek> Eddi|zuHause: 1.33333333333
12:15:54 <Eddi|zuHause> 33% stretch needed
12:16:17 <KittenKoder> I'll have to figure it out tomorrow,
12:17:03 <Eddi|zuHause> it's difficult to recognize any kind of shape in your images though
12:17:25 <KittenKoder> I was thinking that to.
12:18:17 <KittenKoder> I still need a lot more practice.
12:19:23 <Eddi|zuHause> this stretch is actually very annoying, but it's kinda deep inside the game mechanics
12:19:54 <KittenKoder> Working with isometric is a pain no matter what game.
12:20:21 <peter1138> would be better if it was proper isometric ;p
12:20:27 <KittenKoder> Hexagonal is the only way to solve it really.
12:20:38 <Eddi|zuHause> hexagonal has other problems
12:21:09 <KittenKoder> With square the angles won't fit, but yeah, hexagonal is VERY hard to code for.
12:21:11 <Eddi|zuHause> like you cannot do proper 90° crossings
12:21:48 <KittenKoder> Something like TT ... it'd be virtually impossible to use hexagonal.
12:22:09 <Eddi|zuHause> actually, the idea is very intriguing, but probably not worth the effort
12:22:25 <KittenKoder> ... and if the isometric as it is was fixed, it'd be a lot of trouble with having to remake everything.
12:23:43 <KittenKoder> That's why I said I need practice. :p
12:26:17 <Pinkbeast> What would Gresley do?
12:48:38 <georg> i just downloaded openttd-1.1.2-linux-generic-amd64 and started a dedicated server, it keeps changing the version number to 1.0.3/10384F2B in the config file and returns me a "version mismatch" in my 1.1.2 win64 client
12:50:56 <glx> are you sure you start the right one ?
12:51:02 <georg> nevermind, my fault ofc, openttd != ./openttd, had an old version installed via apt
13:11:44 *** TheMask96 has joined #openttd
13:15:25 *** Hirundo has joined #openttd
13:16:39 *** planetmaker has joined #openttd
13:16:39 *** ChanServ sets mode: +o planetmaker
13:16:52 *** ^Spike^ has joined #openttd
13:17:06 *** Terkhen has joined #openttd
13:17:06 *** ChanServ sets mode: +o Terkhen
13:17:39 *** XeryusTC has joined #openttd
13:17:39 *** V453000 has joined #openttd
13:18:10 *** ChanServ sets mode: +o Yexo
13:18:39 *** DJNekkid has joined #openttd
13:29:11 *** Adambean has joined #openttd
13:34:41 *** douknoukem has joined #openttd
13:37:27 *** andythenorth has joined #openttd
13:50:51 *** Sacro_ is now known as Sacro
13:55:21 *** TWerkhoven has joined #openttd
14:10:09 <Hirundo> Eddi|zuHause: Is a NML var that captures vehicle var 45 in its entirety still useful to you?
14:10:48 <Eddi|zuHause> haven't changed much in the last days
14:11:37 <Eddi|zuHause> so i still check those same 4 cases... and i still have not checked the behaviour in 90° turns
14:11:50 *** Devroush has joined #openttd
14:12:29 <Eddi|zuHause> and the actual plan was to replace that with 60+ vars
14:12:46 <Eddi|zuHause> so var45 possibly won't be used at all in the end
14:14:02 <Hirundo> At least for now, NML has a var equivalent to var45
14:14:31 <Hirundo> I found an old patch in my stack which I just finished + committed
14:15:05 *** andythenorth has left #openttd
14:15:16 <Eddi|zuHause> maybe i check it out later
14:15:21 <Eddi|zuHause> not in the mood right now
14:45:19 *** douknoukem has joined #openttd
14:52:32 <planetmaker> hm... maybe I should borrow FIRS' ground tile?
14:59:27 *** DayDreamer has joined #openttd
15:37:27 *** frosch123 has joined #openttd
15:37:54 *** AlexLivingstone has quit IRC
15:45:27 *** Prof_Frink has joined #openttd
15:45:57 *** andythenorth has joined #openttd
15:46:29 * andythenorth has been burnt by that :P
15:52:25 *** HerzogDeXtEr has joined #openttd
15:53:53 *** Xrufuian has joined #openttd
15:55:13 *** supermop_ has joined #openttd
16:01:30 *** ar3k is now known as ar3kaw
16:18:16 *** Kurimus has joined #openttd
16:22:37 *** Phoenix_the_II has quit IRC
16:29:19 <planetmaker> andythenorth: can you explain me the need for tile 226 (basetiles.pnml:59ff) vs. tile 225 (basetiles.pnml:47ff)?
16:29:47 <andythenorth> basetiles got crufty
16:29:53 <andythenorth> it was a nice idea, but impossible in practice
16:30:09 <andythenorth> it failed on Proper Separation of Concerns
16:30:29 <andythenorth> basically shared items + necessary variation = bad code smell
16:30:55 <andythenorth> but to answer the question...
16:31:47 <andythenorth> I'd have to read the nfo
16:32:43 <andythenorth> planetmaker: if they're tiles shared by multiple industries, they should probably be unshared
16:32:46 <andythenorth> and moved into the industry
16:32:55 <andythenorth> the shared basetiles gain almost nothing
16:44:04 <Terkhen> I did not touch the shared industry tiles either
16:44:17 <planetmaker> I'm on them right now
16:49:51 <planetmaker> Terkhen: the building_zextend: what units is that?
16:50:25 <planetmaker> thus 31 for a building of height 0?
16:50:48 <planetmaker> or 8 for a one height level building?
16:51:10 <planetmaker> thus (sprite height - 31) ?
16:51:35 <Terkhen> I don't know, I have not been touching them at all
16:51:55 <planetmaker> oh... nvm. There's an zextend given in the layout :-)
16:52:23 <Terkhen> I know it is in pixels and that it marks the size of the bounding box but nothing more
16:52:26 <Terkhen> I did not have to change them
16:53:48 <Terkhen> what are you planning? :P
16:54:27 <planetmaker> it has the tanks for plastics plant
17:02:16 <Terkhen> I have been thinking about a file for defining common spritesets / spritelayouts
17:02:29 <Terkhen> for example, most of the industries use the same ground tiles
17:02:39 <Terkhen> the spriteset defining them is duplicated all over
17:02:52 <planetmaker> that might make sense
17:03:07 <Terkhen> the sprites for those common industry tiles could be converted into one of the shared spriteset
17:03:25 <planetmaker> One spritelayout for all industries ;-)
17:04:07 <Terkhen> there was an animation particularly nasty in which I had to duplicate the ground sprite in the ground sprite set like 7 times
17:04:17 <Terkhen> so unless you want to define all sprites 7 times... nope :P
17:04:57 <planetmaker> you haven't reached the industry with the 14 animation frames then?
17:05:36 <Terkhen> there are others with 48 animation frames
17:05:42 <Terkhen> just not 48 sprites :)
17:05:50 <Terkhen> I don't know if there is one with 14 sprites
17:06:02 <Terkhen> 14 of "our" sprites, not ttdsprites
17:06:30 <planetmaker> I don't recall and probably didn't even look at the layouts referenced ;-)
17:16:00 *** Brianetta has joined #openttd
17:18:41 *** TheMask96 has joined #openttd
17:18:48 <andythenorth> one spriteset? controversially logical...
17:21:13 <planetmaker> oh.... how boring :-(
17:21:26 <planetmaker> And I thought it would be on the same line as reducing cargos to "stuff"
17:23:19 <Terkhen> it would be simpler if we didn't have to deal with the "all spritesets have the same size" spritelayout restriction
17:23:34 *** |Jeroen| has joined #openttd
17:23:38 <Terkhen> but I still don't see the point to unify all spritesets in one
17:24:44 <planetmaker> nmlc: "sprites/nml/basetiles.pnml", line 10: All parts of the ternary operator (?:) must be integers. <-- nice
17:25:17 <planetmaker> sprite: (terrain_type == TILETYPE_SNOW) ? spriteset_plain_concrete_snow : spriteset_plain_concrete; doesn't work for spritesets, I learnt (again) ;-)
17:26:14 <Terkhen> that's why the spritelayout templates have two/three fields for ground sprites
17:26:32 <Terkhen> if they only use ttdsprites, you can use the ternary operator
17:26:44 <Terkhen> if they only use spritesets, we can use an index to the spritesets
17:26:46 <planetmaker> both is custom in this case
17:26:50 <Terkhen> but if they use both...
17:27:29 <Terkhen> you can use spriteset_common(terrain_type == TILETYPE_SNOW) then
17:27:38 <Terkhen> but then... your ground spriteset must have two entries too :)
17:27:49 <planetmaker> that can be arranged
17:27:55 <planetmaker> it's a groundsprite only layout
17:28:06 <planetmaker> concrete or snow. No otherfrills
17:28:44 <planetmaker> thus I'll just make a short custom thingy for it
17:29:01 <planetmaker> as no such template seems to exist nor is it worth templating that, I think
17:31:00 <planetmaker> as it probably will be removed anyway ;-)
17:32:01 <Terkhen> I had to do a lot of custom stuff
17:32:12 <Terkhen> not worth templating :)
17:33:13 <planetmaker> if everything would be templated, the template would be toooooooo complicated
17:34:09 <Terkhen> yes, the current templates are relatively simple and cover most of the cases
17:35:05 <planetmaker> which is how it should be IMHO
17:45:27 <CIA-2> OpenTTD: translators * r22756 /trunk/src/lang/ (danish.txt korean.txt unfinished/persian.txt):
17:45:27 <CIA-2> OpenTTD: -Update from WebTranslator v3.0:
17:45:27 <CIA-2> OpenTTD: danish - 2 changes by zyx
17:45:27 <CIA-2> OpenTTD: korean - 2 changes by telk5093
17:45:27 <CIA-2> OpenTTD: persian - 18 changes by Peymanpn
17:58:11 <Zuu> Cool, my new computer compiles OpenTTD in 37 seconds for debug build and 1:38 for release builds. So now there is no excuse for making awsome patches :-p
17:59:35 <__ln__> Zuu: the carbon footprint?
18:02:08 <Zuu> But I have a green energy contract and Sweden got most electricity from water and nuclear plants.
18:03:53 <Terkhen> IIRC I managed 20 seconds... with ccache :P
18:03:59 <Zuu> But actually, the main issue with my old computer was not compile time, but that word/excel was too slow to bother open documents.
18:04:54 <frosch123> Zuu: with disk cache or without?
18:05:10 <frosch123> i.e. is the second full compile faster?
18:05:34 <Zuu> frosch123: Compile time after a "clean solution".
18:06:07 <Zuu> I should add that I haven't added direct X yet which might impact things.
18:08:18 * planetmaker would not compile w/o ccache if it can be avoided :-)
18:09:27 <__ln__> are the danes bothered by Barsebäck?
18:09:37 <Rubidium> I seldomly seen much compile time improvements from ccache
18:09:53 <Rubidium> except for the make && make clean && make case
18:10:04 <Zuu> Basrsebäck has been shut down, although it will take long time to complete the shuting down operation.
18:10:44 <Rubidium> generally you change something in a header and everything that includes that header changes, so the ccache is pointless
18:11:44 <Zuu> IIRC in VS 2008 they added some detection of changes that doesn't affect other parts or some other reduction of re-compilation.
18:12:37 <Zuu> Not sure if VS 2010 have any such boosts. Have not yet learned how to work with the new way of adding global include dirs to 2010.
18:13:49 <Rubidium> I doubt they'd remove them
18:22:25 <Zuu> It's not all bad. It looks like I can create a OpenTTD property sheet that is not global but that I add to all OpenTTD projects. That way the OpenTTD includes are stored in one place but does not affect other builds.
18:23:25 <Zuu> Currently I don't have problem with having the OpenTTD includes global, but if I would, this would be a useful improvement.
18:23:58 <Terkhen> IIRC someone did a tutorial on how to add the global folders to MSVC 2010 already
18:24:14 <Terkhen> ah, you don't want that :)
18:24:47 <Zuu> I also remember seeing something like a 2010 guide, but it has not been added to the compiling category.
18:25:08 *** andythenorth has joined #openttd
18:28:14 <Terkhen> he probably never got around to finish it
18:32:44 *** andythenorth has joined #openttd
18:38:01 <andythenorth> my spanish is rusty
18:40:10 <andythenorth> you broke my FIRS again :P
18:42:17 <andythenorth> so my opengfx+ food processor didn't go down well :|
18:44:29 <andythenorth> I drew too much shadow on the walls
18:44:38 <andythenorth> forum feedback :P
18:47:28 <Terkhen> it is quite better than the old one IMO
18:47:40 *** kasuga_ has joined #openttd
18:48:07 *** kasuga_ is now known as kasuga
18:53:41 *** Hirundo has joined #openttd
18:59:26 *** dihedral has joined #openttd
19:05:01 <andythenorth> and went in suggestions forum :|
19:05:23 <Rubidium> oh darn... too late to suggest rivers then :(
19:06:38 <andythenorth> I have to lay those sprite sheets out
19:06:41 <andythenorth> or change the nfo
19:07:18 *** ChanServ sets mode: +o SmatZ
19:07:22 <andythenorth> so that tropic and arctic use 100% identical sprite layout to temperate?
19:07:51 <andythenorth> (copy + rename temperate png for tropic green, tropic desert, arctic brown, arctic snow)
19:08:12 *** V453000 has joined #openttd
19:08:36 *** welterde has joined #openttd
19:08:45 <andythenorth> I'd do it but I have to take apart a three wheel truck :P
19:09:57 <andythenorth> I'll happily (grumpily) do the graphics, but the layout is just one step too much right now
19:13:30 <Wolf01> I'll happiny (grumpig) do the refactoring of my website php code
19:25:12 *** ^Spike^ has joined #openttd
19:26:42 *** XeryusTC has joined #openttd
19:33:41 *** ar3k is now known as ar3kaw
19:34:12 *** Terkhen has joined #openttd
19:34:12 *** ChanServ sets mode: +o Terkhen
19:34:48 <Rubidium> andythenorth: what did you mean by copying? Just copy the temperate one to the arctic etc ones?
19:37:36 <andythenorth> even if it made sense once, I think it's decomped grfcodec output
19:37:54 <andythenorth> and yes, just duplicate the temperate png and rename for arctic-snow etc
19:38:04 <andythenorth> then I'll fill / replace / tweak for colours
19:40:12 *** ChanServ sets mode: +o Yexo
19:40:42 *** DJNekkid has joined #openttd
19:55:12 *** DJNekkid has joined #openttd
20:15:33 <Adamos> hi everybody, is this right place to find players tonight?
20:16:29 <Terkhen> you can try if someone wants to :P
20:18:04 <Terkhen> no thanks, playing is not my thing :)
20:20:33 <Adamos> ok, thanks for info... anyway, some recommandations about settings? i.e. Town cargo limitations, prizes simulating real bussiness etc.?
20:21:57 <Adamos> there is lot of possibilites, i would appreciate some pre-set settings - which would simulate reality event with it's high difficulty
20:23:21 <Adamos> btw. i really admire work on openttd by all of developers - it's really great to see this game not just being alive, but also living
20:25:04 <planetmaker> Adamos: everybody has his or her own preference settings and might have different view on what constitute 'difficult' or 'realistic' or whatever
20:25:23 <planetmaker> try a few servers and check which you like, I'd recommend
20:25:43 <planetmaker> save the game where you like the settings and find out which are the interesting ones
20:26:10 <Adamos> thanks for recommendation
20:29:34 <Adamos> this variability is in one way really great - because of such many possibilities, but on the other hand - it lacks any "standard" - maybe some level you want to compete at... you know, something like general conditions you have to fit to if you want to success... when you make your own conditions, you are losing some quite important feeling of the game
20:33:06 <planetmaker> if you want to compete, do that on one map with other players. Then it's the same for all
20:33:21 <planetmaker> and the different maps are just different 'levels'
20:35:43 <Adamos> you are right... and what about scenarios - it is possible to create them with specially adjusted advanced settings?
20:36:02 <Terkhen> settings are saved for each scenario, yes
20:36:51 <Adamos> cool, thanks for informations, gyus!
20:38:04 <Adamos> if i find some oldies i made for TTD, i'll put them to right place to download.. but i am afraid, that they are already lost
21:09:50 <Eddi|zuHause> i think it's usually "too early to tell"
21:10:38 <__ln__> in fact, in a monty python sketch they say "too early to say, too early to tell"
21:15:16 *** DJNekkid has joined #openttd
21:23:34 *** Phoenix_the_II has joined #openttd
21:36:12 *** andythenorth has joined #openttd
22:08:05 *** Nowaker has joined #openttd
22:08:27 *** Nowaker has joined #openttd
23:17:10 *** lessthanthree has joined #openttd
23:45:14 <Sacro> Pikka: your wiki is le broked
23:45:41 <Pikka> it needs upgrading, orudge is possibly maybe going to do it when he has time
23:45:56 <Pikka> he updated le php on the server you see
23:45:57 <Sacro> :( but I need the knowledge that it holds
23:46:19 <Sacro> me and my friends are having a very enjoyable game
23:46:27 <Sacro> but we refer to the wiki for info
23:46:36 <Pikka> anything in particular you want to know?
23:46:59 <Pikka> you can usually just wiki the real vehicle for stats and usage and suchlike
23:48:13 *** Progman has joined #openttd
23:48:44 <Pikka> also, is it a private game? :P I'm always interested to see how people are using my grfs (and often disappointed, for example with UKRS1 where most players just put A4s and deltics on everything)
23:49:27 <Sacro> errm, it's a CHPP game, Coop's server
23:50:02 <Sacro> heh, we are quite realistic
23:50:12 <Pikka> I guess I should download CHPP some time then
23:50:14 <Sacro> can't afford to run a4s and deltics
23:50:29 <Sacro> tbh we've had a daylength of between 3 and 5, this game has gone on for weeks
23:50:41 <Sacro> started in 1920 and we've only just got to HSTs
23:54:59 *** rhaeder has joined #openttd
continue to next day ⏵
