IRC logs for #openttd on OFTC at 2012-06-10
⏴ go to previous day
00:17:20 <pete1> is the sequence in the readme.txt the sequence, in which openttd will search for the data files? In which order are they processed?
01:16:46 *** Biolunar_ has joined #openttd
02:14:48 *** KnogleAFK has joined #openttd
02:26:38 *** kkimlabs_ has joined #openttd
03:06:07 *** Stimrol has joined #openttd
03:09:57 *** Djohaal has joined #openttd
03:46:07 *** Stimrol has joined #openttd
03:59:08 *** Stimrol has joined #openttd
04:41:37 *** Stimrol has joined #openttd
04:56:17 *** Eddi|zuHause has joined #openttd
05:06:37 *** Stimrol has joined #openttd
05:12:22 *** kkimlabs_ has joined #openttd
05:22:51 *** telanus has joined #openttd
05:51:22 *** kkimlabs_ has joined #openttd
06:04:28 *** andythenorth has joined #openttd
06:22:42 *** andythenorth has joined #openttd
06:27:42 <andythenorth> can't see how to make cargo routing work for PAX *without* routing individual packets :|
06:28:49 <andythenorth> it's too intuitive that PAX want to go to specific places
06:53:51 *** LordAro has joined #openttd
06:55:07 *** sla_ro|master has joined #openttd
07:07:43 *** kkimlabs_ has joined #openttd
07:36:42 *** Alberth has joined #openttd
07:36:42 *** ChanServ sets mode: +o Alberth
07:45:42 <andythenorth> Alberth: I was thinking more about supply / demand
07:45:49 <andythenorth> unconvinced by my spot price idea
07:46:03 *** mahmoud has joined #openttd
07:46:14 <andythenorth> might be helpful to figure out some goals
07:47:42 <Alberth> sorry, but the whole mess with mb got to me, and I am reconsidering what to make of it, and/or what the fuck I am doing here.
07:47:54 <andythenorth> oh did that continue? :P
07:48:06 <Alberth> so I am not exactly in the mood to do deep thinking today, I am afraid
07:48:22 <andythenorth> it is an unfortunate side effect of his total lack of proportion :P
08:17:13 *** Progman has joined #openttd
08:20:12 * andythenorth puzzles at cargo routing
08:20:22 <Terkhen> why? it is quite easy
08:20:25 <andythenorth> can't see a way to do it nicely without assigning a destination to each packet
08:20:43 <Alberth> assigning destination is not the problem :)
08:20:54 <andythenorth> I've been trying to invent methods that are quite blind to where cargo 'wants' to go
08:21:01 <Terkhen> ah, you are looking at a new way, I thought that you were puzzled by the current way :P
08:21:21 <andythenorth> I was hoping to effectively make it work like water
08:21:32 <andythenorth> water doesn't know where it's going, it just goes...down
08:22:11 <Alberth> cargo has an owner, and he is very much aware of where the cargo should go :)
08:22:29 <andythenorth> I can think of ways to do 'down', but I also want to improve transfers, and make auto-refit smart
08:22:37 <andythenorth> and that means understanding the cargo's destination
08:24:36 <Alberth> depends whether you want a slightly different openttd where the user still has much control over where cargo goes, or no control (or very little)
08:24:51 <Alberth> in the latter case, you need destinations and routing, I think
08:27:47 <andythenorth> I was hoping with spot price that game would handle majority of routing, but load/transfer orders could be used by player to move cargo against price gradient (for whatever reason)
08:28:05 <andythenorth> I need to figure out the goals a little more probably
08:28:19 *** TGYoshi has joined #openttd
08:29:07 <andythenorth> although imho, 'goals' are not the best way to design something like this
08:30:36 <Alberth> you just need to keep track of a percentage of cargo moved in each down-hill direction, don't you?
08:30:59 <andythenorth> I have to figure out 'move to station'
08:31:11 <andythenorth> and also rules for loading cargo to a vehicle, or not
08:34:25 <andythenorth> what I really want to do is find a system which makes interesting emergent behaviour possible
08:34:46 <andythenorth> then let GS mess with various parameters that produce different behaviours
08:35:10 <Alberth> so the trouble spot is 'find' :)
08:35:28 <andythenorth> I think cargo needs to 'want' to go somewhere
08:35:37 <andythenorth> almost certainly on a per-packet basis
08:35:39 <Terkhen> andythenorth: subsidies? :)
08:36:09 <andythenorth> I think subsidies are a modifier of 'where cargo wants to go'
08:36:22 <Alberth> Terkhen: I mentioned those yesterday too, but perhaps more fine grained, eg station to station
08:36:59 <Alberth> then you can really influence cargo payments
08:37:05 <Alberth> hmm, but not amounts :(
08:37:34 <Terkhen> yes, subsidies are not a great solution :P
08:38:16 <Alberth> subsidies are perhaps just one form of cargo payment
08:40:32 <Alberth> if you control payments, you can trick the user into handling amounts, but that's micro-management, and splitting cargo streams will never happen, I think
08:40:40 <Terkhen> they would be great for making destinations that are far away or have small amounts of cargo more interesting
08:40:59 <Terkhen> but you need destinations for this to work :)
08:42:30 <Alberth> you need to have some incremental way to build the graph
08:47:24 *** Devroush has joined #openttd
08:51:11 <andythenorth> subsidies would modify demand
08:51:17 <andythenorth> demand would affect routing and price
08:51:25 <andythenorth> how, I can't figure out, I'm not good at maths
08:53:19 <andythenorth> I would also really like to enable payment for delivering only part-way to the destination
08:54:12 <andythenorth> so cargo that wants to go A-D is delivered A-B and player gets paid, assuming that B is on the A-D demand gradient
08:54:17 <Terkhen> sounds quite complicated
08:54:25 <andythenorth> *and* cargo is put back on the map at B
08:54:35 <andythenorth> this would then get very interesting, especially in MP
08:55:06 <Terkhen> I agree, it would be awesome :P
08:55:15 <andythenorth> paying for partial routes is not complicated
08:55:21 <andythenorth> loading cargo onto vehicles for partial routes is mind-boggling
08:55:21 <Terkhen> but you also need a mechanism for shared stations then
08:55:35 <andythenorth> or storing cargo on map tiles
08:57:08 <andythenorth> I have a horrible feeling that this could be solved with a lot of data
08:57:24 <andythenorth> each source-destination pair forms a vector
08:59:09 <andythenorth> and every tile on the map has a relation to that vector
09:02:19 <andythenorth> storing a demand on every tile for every source-destination pair is insane :P
09:03:57 *** Biolunar has joined #openttd
09:06:23 <andythenorth> ho, I think I just reinvented cargodist :P
09:06:34 <andythenorth> not useful, done already :P
09:08:02 <andythenorth> what I need is a metric for cargo at A to know that B is 'closer' to D than A
09:08:17 <andythenorth> and cartesian co-ordinates are a bad metric for that
09:08:38 <andythenorth> link efficiency is one possible measure, but afaict, that's what cargodist does
09:15:04 <andythenorth> michi_cc: in YACD, does the pathfinder have to run for every single cargo packet waiting at a station?
09:17:18 <NGC3982> am i understanding PBS correctly when using it like this?
09:19:27 <Alberth> seems ok, what part is troubling you?
09:21:35 <NGC3982> nothing, it works properly. im just not sure if it's the most effective way.
09:21:48 <NGC3982> im thinking about not having that tight turn on the double-track
09:22:00 <NGC3982> since the train passing is longer then the two turns
09:27:05 <Alberth> I never reach that level of optimization :)
09:28:43 *** FLHerne has joined #openttd
09:30:57 * andythenorth needs to be better at pure maths
09:31:09 * andythenorth is reading about topologies and manifolds
09:32:06 <andythenorth> reading, but not understanding :P
09:33:47 <andythenorth> 'cargo always moves downhill' is trivial until you have a vehicle with orders that initially go uphill
09:33:48 <Alberth> manifolds are math? I thought they were for controlling liquid flows
09:33:59 <andythenorth> do you want to load cargo on that vehicle or not?
09:34:07 <NGC3982> Alberth: well, the double-track is going to have lots and lots of trains on it. the single-track will only house one.
09:34:13 <andythenorth> sometimes you will, because you know that the route is ultimately valid
09:34:34 <andythenorth> and sometimes you won't because it looks stupid for cargo to be taking a circuitous route
09:35:03 <Alberth> andythenorth: just don't care, and load it. The user will stop doing when he does not get paid
09:35:20 * NGC3982 introduces reynolds number.
09:35:42 <andythenorth> also, coal is waiting. Train 1 is in the station, and will take 40 days to deliver. Train 2 is about to arrive and will take 10 days via a shorter route. Load which train?
09:35:54 <Alberth> NGC3982: I'd build a bridge then :)
09:36:12 <andythenorth> I *do not* want to be measuring link capacities, speed per ton mile etc :P
09:36:37 <andythenorth> dealing with this behaviour is then the 'game'
09:37:09 <andythenorth> simple rule: cargo will always move 'downhill'
09:37:11 <Alberth> LordAro: good developers see the point and adjust goals :)
09:37:19 <andythenorth> so any vehicle with 'down' in its orders will be loaded
09:37:40 <LordAro> Alberth: indeed so, and kudos to him for doing so :)
09:38:19 <Alberth> andythenorth: indeed, let the player handle the problem of train 1
09:38:38 <Alberth> if he wants to use more resources for travel, let him
09:39:02 <Alberth> maybe he considers it more realistic that way :p
09:39:22 <Terkhen> LordAro: that's a way of saying "there is no point in upgrading performance in a platform currently broken for OpenTTD"
09:39:23 <NGC3982> Alberth: bridges are no fun :(.
09:40:42 <andythenorth> so a topology for each source-destination pair for each cargo: gradient value for each tile on the map
09:41:12 <andythenorth> gradient is 'flat' or 'down'. Not sure if 'up' is useful or not
09:41:20 <andythenorth> gradient calculated by pathfinder every so many ticks
09:41:43 <Alberth> charge the user for taking it in the wrong direction :p
09:41:59 <andythenorth> gradient based initially on some radius
09:42:08 <andythenorth> building routes modifies the gradient
09:42:44 <andythenorth> I'm trying to keep it scale free: so measuring cargo waiting, link capacity etc are all undesirable imo
09:43:42 <Alberth> that's just for balancing cargo evenly over the network
09:44:01 <Alberth> at least 2 steps further from here :)
09:44:03 <andythenorth> I did like the idea that piling up cargo en-route modifies the gradient
09:44:21 <andythenorth> but then a decision has to be made: is 500t too much waiting? 5,000t?
09:44:37 <andythenorth> there's no way to decide that without enforcing a certain play style
09:45:17 <andythenorth> if underlying routing is scale-free, GS could modify the gradient according to rules of GS
09:45:32 <andythenorth> => supports varied play styles / challenges
09:45:43 <Alberth> keep track of average cargo movement speed, multiply by distance
09:47:05 <Alberth> or perhaps even simpler: gradient * amount
09:48:20 <Eddi|zuHause> andythenorth: how does a "gradient" work for "symmetric" cargos? (passengers, mail, valuables)
09:49:50 <andythenorth> it's based on a source-destination pair
09:49:53 <andythenorth> so should work ok
09:50:08 <andythenorth> without giving cargo packets destinations, it works horribly
09:50:14 <andythenorth> not at all in fact :(
09:50:47 <DorpsGek> andythenorth: 4194304
09:51:38 <andythenorth> I'm guessing that for PAX, there could be hundreds of thousands of source-destination pairs
09:52:09 <andythenorth> and each needs to store a full gradient topology - i.e. a gradient value for every tile on the map
09:52:14 <andythenorth> so not looking very viable :P
09:53:46 <andythenorth> also, even if the gradient pathfinder only runs 8-9 times a month, it's probably too much to calculate gradient for hundreds of thousands of source-destination pairs :|
09:55:52 <andythenorth> perhaps this does need to a topology based on stations, not tiles
09:56:02 <Eddi|zuHause> today Alberth "owns" almost the whole forum :)
09:56:54 <Alberth> Eddi|zuHause: fix it, post some new discussions :)
09:59:39 <Terkhen> or post in existing ones :P
10:00:23 <Alberth> or release a new cargo-d*st :D
10:02:39 <andythenorth> someone remind me
10:02:47 <andythenorth> in what way do newgrfs not *have* to be GPL?
10:03:31 <andythenorth> do they run in a shared address space?
10:04:44 <Terkhen> in the same way that you can create an odt file that contains non GPL stuff and load it with libreoffice
10:04:58 <andythenorth> does odt exec code?
10:05:21 <andythenorth> not that it matters, just thinking out loud :P
10:05:41 <andythenorth> even if newgrfs *had* to be GPL to be distributed, the MB case would still apply
10:05:57 <andythenorth> it's a copyright infringement :|
10:06:15 <andythenorth> his primary problem is 'the internet'
10:06:20 <Terkhen> if every GPL program that is able to "read" code would be limited to read only GPL code they would not be used at all
10:06:29 <andythenorth> newgrf is executable no?
10:06:36 <andythenorth> the bytecode runs
10:06:42 <Terkhen> for example, you can use the gcc compiler to compile non GPL code
10:07:11 <andythenorth> ach, even if I'm right, it solves nothing :P
10:07:31 <andythenorth> he wants to distribute but not be re-distributed
10:07:38 <andythenorth> which is like trying to push water uphill frankly
10:08:02 <Terkhen> the problem is not the copyright infringement itself, those issues are taken seriously and they are always solved as fast as possible
10:08:24 <andythenorth> the problem is that MB feels he has no control so goes nuclear
10:08:33 <andythenorth> fear leads to lack of proportion
10:09:04 <andythenorth> lack of proportion leads to headache for everyone else
10:10:48 <Terkhen> I guess that you can call "actively trying to destroy OpenTTD infrastructure" a headache, yes
10:11:20 <andythenorth> are his actions sufficient to warrant a forum ban warning?
10:11:37 <andythenorth> they're legal, but disproportionate
10:12:08 <andythenorth> my guess is that if he'd had OpenTTD taken offline, we could have sued him
10:12:34 <andythenorth> for at least the monetary value of the lost hosting
10:12:53 <andythenorth> we would have been able to demonstrate compliance with DMCA/EDEC
10:12:56 <Xaroth> andythenorth: MB feels the world revolves around his own head
10:13:04 <andythenorth> we all have that
10:13:49 <andythenorth> is our hosting donated by a sponsor, or paid for by player donations?
10:14:30 <Xaroth> Thought you needed that :)
10:15:48 * andythenorth wonders how a class action would have gone - players suing MB for loss of services paid for by donations
10:16:49 <Xaroth> suits like that are a complete circlejerk
10:16:58 <andythenorth> definitely adds to the fun, involving lawyers in FOSS community stuff :P
10:17:25 <Xaroth> only adds more headache for those who already have a headache from this
10:17:45 <Eddi|zuHause> andythenorth: it would have been thrown out due to lack of success-expectation
10:18:09 <andythenorth> we could have had fun for months discussing it though :P
10:18:17 <andythenorth> way better than working on the game
10:18:27 <Xaroth> depends on your definition of 'fun'
10:19:04 <Xaroth> I can imagine a lot of devs already being mightily pissed off, I wouldn't want to see them being faced by this same shitstorm day after day for several weeks, if not months
10:20:22 <planetmaker> Xaroth: your assumption can't be more to the point
10:20:39 <Xaroth> planetmaker: who said anything about it being an assumption?
10:25:30 <andythenorth> if we'd been taken offline, how many kittens would have died?
10:25:37 <andythenorth> [just trying to establish proportions]
10:25:54 <andythenorth> I'm going to reply in that thread later
10:26:24 <TrueBrain> andythenorth: hmm ... it would be such a cool lawsuit tbh
10:26:36 <andythenorth> it would be a remarkably stupid lawsuit
10:26:37 <TrueBrain> as mb clearly defamed us, and we have enough evidence to proof that
10:26:55 <Xaroth> TrueBrain: only problem is proving the amount of damage caused
10:26:56 <andythenorth> 'bunch of semi-aspergers train fans sue and counter-sue each other over free game'
10:27:19 <TrueBrain> Xaroth: that is the fun part: the damage doesn't has to be in terms of money lost due sale
10:27:27 <andythenorth> sponsorship has a valuer
10:27:31 <TrueBrain> we just say: a few developers burned down because of it
10:27:38 <TrueBrain> couldn't do their own real-life work
10:27:48 <TrueBrain> so that is the first demand
10:28:00 <TrueBrain> then you start about the loss in name (OpenTTD)
10:28:03 <TrueBrain> the damage that has caused
10:28:06 <TrueBrain> which is a 'feeling' value
10:28:07 <andythenorth> there's probably some case could be made that it's vandalism of an artistic work
10:28:22 <andythenorth> do we have ad revenue on the openttd site?
10:28:25 <TrueBrain> and let there be no mistake: he would have lost
10:28:53 <andythenorth> ad revenue can be a poisoned chalice anyway; then we're "making money from copyright infringement"
10:29:05 <TrueBrain> andythenorth: no, we are not
10:29:30 <TrueBrain> I posted the same to Hyro a moment ago: we are in no violation up to the point we are informed of said violation (DMCA / EDEC)
10:29:37 <andythenorth> and then it's process
10:29:38 <TrueBrain> then we are giving a "reasonable amount of time" to solve the issue
10:29:48 <andythenorth> as long as we follow process, we're correct
10:29:55 <TrueBrain> so a single upload can never been defined as "making money from copyright infringement"
10:30:11 <TrueBrain> of course, this under the assumption it doesn't happen every single minute of every single hour of every single day
10:30:13 <planetmaker> We might order a cease and desist with court to have him on penalty of a fine never repeat such claim unfounded
10:30:14 <TrueBrain> (like the piratebay)
10:30:25 <TrueBrain> planetmaker: hihi; yes :D
10:30:31 <planetmaker> we'd come through with that. Or *he* would have to show that his claim of repeated infringement is just
10:30:38 <planetmaker> Actually... I really like that thought...
10:30:40 <Xaroth> a cease and desist is easy :P
10:30:47 <andythenorth> if he has no evidence base, then he is at risk of defamation
10:30:58 <andythenorth> did he publish his assertion publicly anywhere?
10:31:15 <planetmaker> he made it to support his claim towards 3rd party
10:31:20 <TrueBrain> but I can publish it, as a takedown notice is part of a legal process, thereby public domain
10:31:58 <andythenorth> but it's written, and defensibly untrue claim?
10:32:06 <andythenorth> did he claim it was a true claim?
10:32:22 <TrueBrain> he made 2 statements which made our 3rd parties go: euh?! WUTH?!
10:32:31 <TrueBrain> one in very clear wording, the other is a clear implication of wording
10:32:38 <andythenorth> so he claimed 'fact' not 'opinion'?
10:33:01 <andythenorth> libel -> actionable defamation
10:33:10 <andythenorth> but you know that already :)
10:33:13 <planetmaker> luckily that makes the difference whether a cease and desist order would work :-) This way it does
10:34:04 <andythenorth> I would just get him forum warned as 'unhelpful'
10:34:34 <Xaroth> only warned? Pff, softie
10:34:55 <andythenorth> I suspect with a warning, he might extract himself from the forums
10:35:01 <Xaroth> I see two threads, that's at least 2 warnings.. how many did it take to get banned again? :P
10:35:10 <andythenorth> he's not actually going to go away anyway
10:35:28 <andythenorth> he'll sit making grfs and occasionally sending takedown notices
10:35:43 <Xaroth> which doesn't actually solve the problem
10:35:46 <TrueBrain> all I wish, is that he would show some sign of realise that what he did was so far out of line, the line is not visible anymore
10:36:20 <Xaroth> He won't do that, as that would mean he'd have to admit the world doesn't revolve around him anymore
10:36:23 <andythenorth> I'm going to post that the measures to remedy are significantly out of proportion to the harm done
10:41:11 <Eddi|zuHause> i'll put this thread in the "don't feed the troll" bin...
10:41:29 *** Chris_Booth has joined #openttd
10:41:44 <Eddi|zuHause> there's only one possible way this thread can end. a flamewar...
10:42:06 <Chris_Booth> and I bring my Fire-Arms?
10:42:07 <Eddi|zuHause> every "sane" thing that could ever be said was already said
10:43:07 <TrueBrain> Eddi|zuHause: except the words: I am sorry
10:43:21 <TrueBrain> but agree'd, that might not be "sane" ;)
10:44:05 <Eddi|zuHause> TrueBrain: there is always four truths...
10:44:20 <Eddi|zuHause> your truth, their truth, "the truth" and what actually happened
10:44:28 <andythenorth> you clearly haven't met my wife
10:44:32 <andythenorth> there is only one truth
10:44:58 <Eddi|zuHause> andythenorth: that's a clear case of "their truth" :)
10:45:18 <andythenorth> can you send her that memo, thanks :)
10:46:05 <Chris_Booth> dear andythenorth's wife there are always four truths; your truth, their truth, "the truth" and what actually happened
10:48:58 *** brambles has joined #openttd
10:49:40 <TrueBrain> Eddi|zuHause: despite the amount of truths, I could at least hope he realised he was overreacting, and that it would deserve at least a "sorry"
10:50:31 <Eddi|zuHause> i don't think you'll ever hear those words from his mouth
10:50:54 <TrueBrain> like said, I guess that it doesnt falls under that "sane" quotation
10:51:11 <TrueBrain> in that case your statement would be right :P
10:54:07 <andythenorth> do / can cargo packets store a final destination, and/or a 'next unload'
10:55:54 <Alberth> sure, you may have to split up cargo packets, as the same cargo with different destinations cannot be combined any more
10:56:40 <Alberth> it's a matter of adding a field
10:57:04 <andythenorth> I'm not sure it's needed yet
10:57:12 <andythenorth> I'm bouncing around different levels of detail :P
10:57:38 <andythenorth> also....can we fork tt-forums? :)
10:57:51 <andythenorth> it's become an ottd monoculture anyway :P
10:59:01 <andythenorth> we could establish it as 'an unsafe place'
10:59:18 <andythenorth> - you will be flamed for discussing multi-threading without knowledge
10:59:25 <andythenorth> - you must use GPL for your newgrfs
10:59:39 <andythenorth> - asking for underground metro is a warning, then ban
10:59:48 <andythenorth> - OS X developers will be punished
11:00:55 <TrueBrain> andythenorth: your second point, we have been talking about only allowing open source licenses on BaNaNaS a few times now
11:01:08 <andythenorth> pikka is your blocker to that :P
11:01:19 <TrueBrain> well, its not a single person tbh
11:01:31 <andythenorth> can you issue DMCA takedown for content not yours?
11:01:34 <TrueBrain> but I personally really favour sending the vibe: we are OSS, so should you
11:01:42 <TrueBrain> andythenorth: DMCA: no. EDEC: yes
11:01:53 <TrueBrain> so in US: no, in EU: yes
11:02:16 <andythenorth> I only joke about forking forums
11:02:30 <andythenorth> but slowly creative, helpful people are being driven away by ottd group-think
11:03:13 *** valhallasw has joined #openttd
11:03:34 <Terkhen> andythenorth: what do you mean?
11:04:19 <Eddi|zuHause> <TrueBrain> andythenorth: your second point, we have been talking about only allowing open source licenses on BaNaNaS a few times now <-- how does that help?
11:04:19 <andythenorth> we drove out SAC, RichK67, now we drive out MB?
11:04:21 <TinoDidriksen> What is EDEC? Google finds nothing related to EU...
11:04:46 <TrueBrain> Eddi|zuHause: it is in no relation to mb, if that is what you mean
11:05:04 <TrueBrain> Eddi|zuHause: it is just about the general vibe, I personally dislike close source stuff being uploaded for an open source game
11:05:07 <TrueBrain> but that is my personally opinion
11:05:35 <TrueBrain> TinoDidriksen: EDEC is the European Directive on Electronic Commerce
11:05:42 <TrueBrain> it contains DMCA basically
11:06:07 <TrueBrain> Article 12 to 14 are what we talk about
11:06:34 <andythenorth> can't see a scale-free way to set demand on a tile
11:06:39 <Terkhen> andythenorth: and what were the causes of each one of them being driven away?
11:06:50 <andythenorth> they don't like the prevailing attitude
11:07:10 <andythenorth> primarily the attitude of ottd devs and contributors
11:07:15 <andythenorth> I didn't say any of this was bad :P
11:07:24 <andythenorth> but we're forking the forum from the inside :P
11:07:39 <Eddi|zuHause> andythenorth: what did SAC have to do with openttd?
11:07:46 <TrueBrain> the people I have seen go away, were issues on their side: their god complex got so big, they no longer received the praise they wanted
11:08:10 <TrueBrain> which is fundemental to human beings
11:08:16 <TrueBrain> and why OSS is so hard to maintain
11:08:44 <TrueBrain> we had patches of people where they demanded their name to be in the source code
11:08:49 <andythenorth> 'how many marxist-leninist factions does it take to change a lightbulb?'
11:08:51 <TrueBrain> never underestimate the human ego :D
11:08:53 <Eddi|zuHause> i'd like to add to the list of people "driven away": eis_os and OzTrans
11:08:58 <andythenorth> 'answer: 1; the others split into other factions and left' :P
11:09:08 <TrueBrain> Eddi|zuHause: eis_os was ever part of OpenTTD? wasn't he TTDp?
11:09:12 <andythenorth> joke needs work :P
11:09:14 <TrueBrain> andythenorth: lolz :D
11:09:34 <Eddi|zuHause> TrueBrain: yes, but...
11:09:47 <TrueBrain> [13:07] <Eddi|zuHause> andythenorth: what did SAC have to do with openttd? <- that was why I asked
11:10:19 <andythenorth> the attitude is now:
11:10:26 <Eddi|zuHause> TrueBrain: if we assume the "wider TT community", to include the SAC problem, then...
11:10:28 <andythenorth> - release something or get off the pot
11:10:33 <andythenorth> - no individual heroes
11:10:44 <andythenorth> - don't violate the GPL
11:10:51 <TrueBrain> Eddi|zuHause: fair :P
11:10:57 <andythenorth> - your dramas are not welcome
11:11:16 <Xaroth> both sac and mb are butthurt because they want to publish copyrighted stuff, but don't want the responsibility behind it
11:11:29 <TrueBrain> andythenorth: that sounds like a good env, not?
11:11:36 <andythenorth> I didn't say it was bad :P
11:12:03 <Eddi|zuHause> MB always had something against BaNaNaS, and i never really understood what...
11:12:18 <Terkhen> andythenorth: I would rephrase the first point as "if you don't plan to work on it, don't talk about how things should be done"
11:12:26 <andythenorth> it's not invented in the german forum in 2003?
11:12:47 <andythenorth> it's not grf-crawler?
11:12:52 <TrueBrain> "no control" I would say
11:12:52 <Xaroth> Eddi|zuHause: probably because, in BaNaNaS, the world does not revolve around him
11:12:58 <Xaroth> he wants to have a site where you can only see his stuff
11:13:04 <TrueBrain> they want to keep very close control
11:13:05 <Xaroth> and force people to only use his site to get his stuff
11:13:11 <TrueBrain> and they are very afraid old versions are kept in rotation
11:13:15 <TrueBrain> on one hand I can understand that
11:13:24 <TrueBrain> it is bad if an old defunctional version is floating around
11:13:35 <TrueBrain> but ... that doesn't kept us from releasing 0.4.7
11:13:50 <TrueBrain> for years we had a 0.6.3 server ... how ever bad it was, it was his choice
11:14:07 <Eddi|zuHause> that was "interesting" development back then :p
11:14:08 <TrueBrain> on one hand I would like I could remove his server, but on the other ... this is a free world
11:14:14 *** telanus1 has joined #openttd
11:14:37 <TrueBrain> but from what I have read over these years, it comes down to: I want to be able to remove my file any time I want
11:14:41 <TrueBrain> I am that important
11:14:48 <TrueBrain> even if it breaks all savegames, I want to do that
11:14:57 <TrueBrain> which, again, I can kindof understand
11:15:20 <Eddi|zuHause> on the german forum we today have a guy who plays a long-running game on 0.6.3, and won't update "because the industries change production too fast"
11:15:25 <TrueBrain> giving away control to a 3rd party is scary .. means you have to trust people etc :P
11:16:04 <Alberth> and yet they are fine with normal users having the old version
11:16:40 <Eddi|zuHause> and another guy a while back refuses to update his game from 0.7.x because it screws up all his timetables because the trains are underpowered wrt the tweaked "realistic" acceleration
11:16:50 * andythenorth wonders if 'demand' can be scaled to the size of the map
11:16:52 <TrueBrain> funny enough, in contrast, we still publish 0.1.0 on our binary servers :P
11:17:23 <TrueBrain> and I Can honestly say I am proud on the iterations we made
11:17:33 <Terkhen> Eddi|zuHause: they don't know about our never ending list of advanced settings, I guess
11:17:54 <Eddi|zuHause> Terkhen: well, those are exactly items which cannot be set by settings
11:18:03 <TrueBrain> then again, our MasterServer protocol for example still supports the first client which had it ... which is old :P
11:18:09 <andythenorth> 'demand' unit is arbitrary, just needs to create a gradient between current tile, and accepting tile
11:18:25 <Terkhen> oh, sorry, I did not read the "tweaked" part :)
11:18:28 <Eddi|zuHause> Terkhen: there is no switch between the "old unbalanced realistic acceleration" and the "new balanced realistic acceleration"
11:18:43 <andythenorth> python import helper
11:18:49 <andythenorth> from helper import mathematician
11:19:00 <Terkhen> well, in that case still having the old versions around is what lets them play the game as they want
11:19:16 <Eddi|zuHause> and likewise there is no switch between the old industry behaviour (monthly change not scaled to map size) and the new one (constantly change scaled with map size)
11:19:35 <TrueBrain> andythenorth: ImportError: class not found
11:19:59 <Eddi|zuHause> TrueBrain: i don't think that is a real error :=)
11:20:13 <TrueBrain> I dont think it is a real import
11:20:49 <Eddi|zuHause> andythenorth: what you maybe want is a "network flow", and then CargoDist does model it this way
11:21:01 <andythenorth> yeah, wondering that
11:21:02 <Eddi|zuHause> andythenorth: but it does that per station, not per tile
11:21:40 <andythenorth> wondering if I'm just reinventing cargodist
11:23:04 <andythenorth> "it's a pirate bay situation then"
11:26:06 <andythenorth> do we know why we *don't* try and maintain control?
11:26:14 <andythenorth> -> makes us liable if we do
11:26:35 <andythenorth> if we start attempting to verify copyright, we're liable for any failures
11:26:47 <andythenorth> whereas if we do nothing for that, it's covered as 'abuse' under ToS
11:29:26 * andythenorth wants to now place money on the prediction by Eddi|zuHause wrt flame war
11:29:30 <andythenorth> who will take my bet?
11:31:03 * andythenorth needs a rubber sheet
11:31:42 <andythenorth> I need to forget cartesian map co-ordinates when trying to route cargo
11:32:10 <andythenorth> for a given pair of source-destination.....let 0 be the state of the map before calculations
11:32:20 <andythenorth> let -100 be the value at the destination tile
11:32:28 <andythenorth> let +100 be the value at the source tile
11:32:35 <andythenorth> now how to deform the topology?
11:33:54 * andythenorth should draw this in PIL
11:41:00 *** frosch123 has joined #openttd
11:43:01 <Alberth> or in inkscape with svg :)
11:43:11 <Xaroth> andythenorth: it's not fair to bet on something you're helping the outcome with :P
11:45:04 <DorpsGek> frosch123: peter1138 was last seen in #openttd 2 weeks, 6 days, 1 hour, 27 minutes, and 50 seconds ago: <peter1138> also it's a cog with opengfx
11:45:10 <DorpsGek> frosch123: zuu was last seen in #openttd 4 days, 0 hours, 28 minutes, and 40 seconds ago: <Zuu> Eddi|zuHause: You need to find a post to comment in the AI/GS section and then you will own the entire forum :-)
11:47:14 <andythenorth> TrueBrain: libel not slander
11:47:21 <andythenorth> slander is spoken, libel is written ;)
11:47:41 <TrueBrain> where did I write that?
11:49:03 <TrueBrain> and ... popcorn time
11:49:12 <TrueBrain> without letting others censor my writing
11:49:15 <TrueBrain> can't be a good thing :P
11:51:17 <andythenorth> but also...wtf cargo distribution :)
11:51:45 <andythenorth> I need a v-shaped profile, extruded along the vector between source and destination
11:52:12 <yno> is it possible to know the cargo amount/month of a station?
11:53:32 <andythenorth> TrueBrain et al - fwiw, when I got into a GPL argument with SAC, I found it quite distressing :P
11:53:38 <andythenorth> and I am quite thick skinned.
11:53:46 <andythenorth> so you have my sympathies here
11:54:15 <TrueBrain> it just sickens me that mb keeps playing the victim: I did nothing wrong. Wtf. You told our 3rd parties they should terminate us on unfound claims ...
11:54:18 <TrueBrain> how is that being a victim?
11:54:49 <andythenorth> I could think of answers, but the only one that sticks is:
11:55:01 <andythenorth> it's just a big lot of 'meh'
11:55:07 <andythenorth> unless you're stuck dealing with the madness
11:55:16 <TrueBrain> it just amases me how he thinks he changed the world
11:55:28 <TrueBrain> like he thinks we only handled his caes because the mirrors contacted us
11:55:38 <TrueBrain> the HU mirror emailed me: I cannot find the file he mentioned
11:55:45 <TrueBrain> mostly, as I already removed it by then :P
11:55:58 <TrueBrain> but if you put everything in a very small timeframe, things appear to overlap I guess
11:56:56 <Xaroth> absolutely win reply, tb
11:56:57 <Alberth> yno: roughly the production of the supplying industries times the percentage
11:58:47 *** Djohaal has joined #openttd
12:02:11 <andythenorth> "Are my methods unsound?"
12:02:16 <andythenorth> "I don't see any method at all, sir."
12:02:24 <TrueBrain> nice reply andythenorth
12:02:48 <andythenorth> ^^ Apocalypse now / heart of darkness
12:04:56 <andythenorth> really, he lacks moral sense
12:05:20 <andythenorth> wasting time for mirrors
12:05:40 <andythenorth> urgent, dramatic takedowns are for when you're accidentally distributing malware that lifts credit cards
12:05:51 <andythenorth> or other such horrible things
12:06:07 <TrueBrain> I made an edit on my post, I forgot a piece :D
12:06:21 <andythenorth> not when you are distributing 729KB of grf with pixel boats in it
12:06:26 * andythenorth deletes newships
12:07:22 <TrueBrain> andythenorth: among ISPs, when there is a website changed to steal bank details, we are asked to take care of it within 24h, but no longer than 48h
12:07:28 <TrueBrain> for all the right reasons
12:07:49 <TrueBrain> after 24h, they contact your AS provider already .. for all the right reasons :P
12:07:56 <TrueBrain> top priority those things have :P
12:08:41 <Xaroth> at which point your AS will contact you again, and give you another 24h :P
12:08:49 <TrueBrain> Xaroth: on bank-fraud?
12:09:01 <TrueBrain> I almost once cut off a person in this room because he did not reply fast enough :P
12:09:07 <Xaroth> yep, the timeframe for those are insanely large
12:09:08 <TrueBrain> I had to text him ;)
12:09:14 <TrueBrain> Xaroth: euh, not in my experience :P
12:09:33 <Xaroth> then our experiences differ :)
12:09:36 <TrueBrain> within 48h those things are removed after first notification ;)
12:09:44 <TrueBrain> by what ever means ;)
12:09:54 <TrueBrain> our AS told us to either cut the IP, remove the site, or be cut off :P
12:10:00 <Alberth> nuke the data center :p
12:10:01 <TrueBrain> for all the right reasons, again :)
12:10:22 <TrueBrain> phishing shit is the worst of the worst
12:10:53 <Xaroth> don't get me wrong, I agree with the urgency, but we've had one case where we were given 72 hours in total.. 48 for what we could do, then our AS provider would get another 24h
12:11:23 <Xaroth> (the content itself was gone after 2 hours though)
12:12:21 <Eddi|zuHause> <Xaroth> s/sense/brains/ <-- insulting people has never solved anything...
12:12:44 <Eddi|zuHause> (this, obviously, is the point of no return for a flamewar)
12:12:45 <Xaroth> Eddi|zuHause: neither has making wild accusations
12:14:12 <Eddi|zuHause> Xaroth: neither has pointing fingers what the other side did wrong
12:15:30 <Xaroth> Eddi|zuHause: also, do you see a flamewar here?
12:15:44 <Xaroth> I just made a statement, and nobody cared enough to openly disagree
12:15:49 <Xaroth> until you brought it up again :)
12:16:18 <Eddi|zuHause> Xaroth: i saw a brooding flamewar two hours ago...
12:16:46 <Eddi|zuHause> [10.06.2012 12:41] <Eddi|zuHause> there's only one possible way this thread can end. a flamewar...
12:20:04 <TrueBrain> I always hope one can have a decent conversation about these things
12:20:08 <TrueBrain> but I always forget that never happens :P
12:21:05 <Eddi|zuHause> and that's why things like ACTA only make it worse...
12:21:22 <Eddi|zuHause> it further cements a downward spiral
12:21:35 <Eddi|zuHause> that can only end in a catastrophe
12:21:53 <TrueBrain> ACTA basically means: stop writing software :P
12:21:54 <Eddi|zuHause> even if the actual step it takes is marginal
12:22:05 <TrueBrain> well: stop using Internet
12:23:34 <yno> Alberth: hm thanks, but what of passengers at airports? :p
12:23:38 <TrueBrain> a free world is a dream, only kept alive by people who are not in politics
12:24:25 <Eddi|zuHause> yno: it's the sum of all houses in the catchment area
12:25:03 <Eddi|zuHause> yno: only that you can't see the amount on a per-house basis. but the town window gives the sum over all houses in the town
12:25:13 <Eddi|zuHause> so you can use that as an estimate
12:25:34 <yno> well, that's the potential, but this doesnt give me the carried passengers
12:25:52 <Eddi|zuHause> again, it is multiplied by the transport rating
12:36:33 <TrueBrain> hmm ... what to do with my sunday
12:36:50 <Xaroth> post on a forum to point out somebody on the internet is wrong?
12:37:02 <Xaroth> either that, or diablo III :P
12:37:37 <TrueBrain> cant' get passed act2
12:38:26 <Xaroth> would happily team up, but got people comming over in .. 20 minutes
12:40:27 <Terkhen> Xaroth: unless new information is provided in the thread, I would advocate for diablo III
12:40:34 <Terkhen> and I don't see that happening
12:40:45 <TrueBrain> there is always someone wrong somewhere on the internet :P
12:41:04 <TrueBrain> when ever I think about it, the xkcd cartoon keeps flashing in my mind :P
12:44:41 <andythenorth> I think I don't need a topology for every source-destination pair
12:44:57 <andythenorth> I need a topology for every cargo-destination pair
12:45:01 <andythenorth> which is simpler
12:45:09 <andythenorth> well, smaller :P
12:45:16 <andythenorth> maybe no simpler :P
12:45:35 <TrueBrain> for the first time in weeks I am reading the forums; it is funny how after years the same topcis are still current :)
12:46:05 <TrueBrain> multithreading, mac, mobile platforms, IS, patchpacks, <random patches not following any coding style>, .. :)
12:46:21 * TrueBrain hugs this community; how ever long you are gone, it is always the same when you come back :P
12:47:04 <TrueBrain> yeah, why is underground not done yet? :D
12:48:28 * andythenorth wonders if it (demand) must be based on station nodes....probably yes
12:49:31 <andythenorth> in which case working out backwards from the destination becomes trivial
12:54:13 * FLHerne demands multicore support and infra-sharing, NOW! :P
12:54:30 <TrueBrain> FLHerne: demand away; I demand your patch for it :D
12:57:09 <andythenorth> destination = 0; first set of nodes = 1, nodes connected to first nodes = 2 etc
12:57:21 <andythenorth> node weighting = shortest number of links from this node to destination
12:57:56 <andythenorth> this method fails for: payment for partial delivery; putting partial delivery back on the map for other players
12:58:18 <andythenorth> this method fails for: modifying payment rates to account for demand
12:58:23 <andythenorth> but it's conceptually simple
13:00:20 <andythenorth> assuming a connection, cargo will join the network at a certain node, and always flow 'downhill'
13:01:02 <andythenorth> moves between equally weighted nodes would be forbidden for practical reasons, even if the overall route would be beneficial
13:01:18 <andythenorth> a 'move' is load/unload, not 'pass through'
13:01:35 <Terkhen> FLHerne: some stuff such as drawing is already done in different threads, and you can share as much infrastructure as you want as long as all players use the same company
13:01:53 <andythenorth> node weighting would be recalculated using pathfinder code at some sane frequency
13:02:37 <andythenorth> this method fails for: giving GS ways to screw around with node weightings
13:04:21 <andythenorth> need to be able to lift one end of the gradient :P
13:04:27 <andythenorth> or drop the other end :P
13:08:15 <andythenorth> or provide some additional 'link quality' rating
13:08:29 <andythenorth> meh, node weightings are just another form of directed links I guess
13:08:43 <andythenorth> maybe they only need to store the direction (up, down, neutral)
13:08:55 <andythenorth> and have something else to cause links to be preferred or not
13:09:57 * andythenorth has horrible ideas
13:10:25 <andythenorth> arbitrary +ve / -ve values for directed links, not just the direction
13:12:46 * andythenorth reads how cargodist works
13:15:31 <andythenorth> my idea would penalise transfers
13:15:49 <andythenorth> it would leave cargo on one slow ship rather than reload it between two fast trains
13:16:01 *** Stimrol has joined #openttd
13:16:56 <andythenorth> assuming both are in the station simultaneously
13:19:42 <TrueBrain> no flaming, constructive replies
13:19:49 <TrueBrain> you didn't see that coming now, did you? :P
13:21:03 <TrueBrain> *and I am trolllliinngggg*
13:24:40 <andythenorth> I am thinking now that demand and routing might be entirely separate
13:25:24 <andythenorth> probably that makes no sense without more spec
13:26:28 <andythenorth> every cargo packet would have a destination
13:26:37 <andythenorth> every destination would have a node graph
13:26:54 <andythenorth> cargo always routes downhill once on the network
13:27:50 <Terkhen> constructive replies, I'm honestly surprised :)
13:27:53 <andythenorth> transfer orders would be likely unnecessary with this
13:29:33 <andythenorth> the node graph is formed from the set of vehicle orders
13:29:48 <andythenorth> or maybe from the set of connected stations, not sure which
13:30:29 <andythenorth> vehicles seems more intuitive; stations might be more interesting
13:30:53 <NGC3982> what does 4-2-0 mean in context of old train engines?
13:31:09 <NGC3982> it sounds like some kind of wheel setup?
13:32:50 <valhallasw> NGC3982: something like 4 front axles, 2 driving axles, no rear axles
13:32:53 <valhallasw> or possibly wheels
13:33:15 <valhallasw> so wheels, not axles
13:36:17 <NGC3982> what on earth am i missing? the trains wont go into depot for replacing :(
13:37:56 <andythenorth> ho ho ho andythenorth has an idea
13:38:14 <andythenorth> instead of calculating links on a company basis, per station
13:38:45 <andythenorth> calculate links for all companies, for any tile in a catchment
13:39:03 <andythenorth> I think there's an edge case where stations could be walked to the destination :P
13:39:20 * andythenorth wants 'unloading' to make the cargo available at that tile for any player
13:39:26 <andythenorth> that would be insanely awesome
13:39:48 <andythenorth> any station covering that tile could pickup the cargo
13:40:29 <andythenorth> awesome potential for all three of collaboration, competition and griefing :)
13:41:45 <Alberth> NGC3982: disabled breakdowns, and enable 'disable servicing when breakdowns are enabled' ?
13:42:40 <NGC3982> breakdowns are disabled
13:42:48 <NGC3982> two trains have replaced since i pasted the link
13:43:06 <NGC3982> its seems like they have choosen the depot to replace in
13:43:19 <Alberth> you were too impatient :)
13:43:47 <NGC3982> but what, it doesnt just route to the closest depot?
13:43:48 <Alberth> they look for depots every now and then
13:43:53 <andythenorth> maybe explicit 'unload and get paid' order
13:44:00 <andythenorth> the idea of partial payment interests me
13:44:15 <andythenorth> feeder payments remain stupid afaict
13:48:47 <andythenorth> meh, what do we think about infrastructure sharing stuff?
13:48:54 <andythenorth> e.g. shared stations
13:49:36 <Alberth> you steal all my platforms!
13:49:39 <andythenorth> not necessarily shared tracks
13:49:52 <andythenorth> but a union of waiting cargo
13:51:28 <andythenorth> is accounting by vehicle silly?
13:52:00 <andythenorth> i.e annual profit / loss
13:52:15 * andythenorth is finding lots of things are unfortunately related :P
13:52:54 <Alberth> first do the core thing
13:53:03 <Terkhen> andythenorth: personally I think "too much work for such a little gain"
13:53:09 <Terkhen> but if someone else does the work...
13:53:32 <andythenorth> Alberth: first got to figure out what the core thing is :)
13:53:32 <Alberth> this 'someone' is such an awesome guy :)
13:53:52 <andythenorth> I don't care if this ever ships, it's a good brain-exercise for me
13:54:08 <andythenorth> if it does get coded, even better
13:54:33 <Terkhen> andythenorth: you should learn multiplatform C++ thread code and do multicore support for us
13:54:40 <Terkhen> that would be an awesome brain exercise for you
13:54:53 <Terkhen> maybe brain melting :P
13:54:53 <andythenorth> awesome in the traditional sense of the word
13:55:05 <Terkhen> no idea of that sense :P
13:55:39 <Terkhen> those are the best exercises
14:07:41 * Alberth would settle for consists
14:19:20 *** HerzogDeXtEr has joined #openttd
14:43:52 <andythenorth> so for this graph, if we stored the direction of the link rather than the weight...that would be interesting
14:44:36 <andythenorth> and only load cargo for -ve value
14:45:00 <andythenorth> but now allow values such as -2 etc
14:45:09 <andythenorth> thus things like GS could modify values
14:45:31 <andythenorth> for cases like 'avoid congested hubs'
14:45:43 <andythenorth> or 'this town doesn't permit nuclear fuel transport' or such
14:46:13 <andythenorth> a floor / ceiling would be needed to prevent such modifications causing loops
14:46:37 <andythenorth> I'm not very convinced by the idea
14:47:04 <Alberth> you need cycle detection :)
14:47:51 <andythenorth> I'd rather need 'avoid complicated rules' :)
15:13:09 * andythenorth is not convinced that 'number of hops' is a good way to route
15:13:23 <andythenorth> edge cases could see cargo travelling from one side of map to the other
15:14:21 <andythenorth> this is an inevitable consequence of any scheme to decouple routing from map cartesian co-ordinates :P
15:15:05 <Alberth> as a first approximation it should be fine imho
15:18:48 <andythenorth> the key thing I keep returning to is to have a simple method for loading cargo onto any given vehicle at any given station
15:19:14 <andythenorth> imho everything else in this area should be decoupled cleanly from that
15:19:31 <yno> does the variety of transported stuff impact town growth?
15:20:59 * andythenorth wonders if real transport fees are per ton mile travelled, or per ton mile as crow flies
15:23:41 <Alberth> wouldn't it be a fixed price contract?
15:24:33 <Alberth> yno: nope, it does not matter which transport you use, except for the properties of the transport itself (eg max speed, and capacity and such)
15:25:14 <andythenorth> Alberth: not sure :) It's of no consequence to routing, except that the routing mechanic could cause more money to be paid :P
15:25:26 <andythenorth> currently we use manhattan distance between stations?
15:26:13 <Alberth> something in that direction
15:26:16 <Terkhen> the code is your friend :P
15:26:36 <Alberth> I don't know whether the real source / destination is also taken into account
15:26:42 <andythenorth> I am scared to read the transfer code that I need to read :P
15:29:28 <frosch123> oh, why all the bananas discussion?
15:29:44 <frosch123> there are currently 256 newgrfs uploaded, that's the limit, isn't it?
15:37:02 <NGC3982> ukrs2+ and FIRS is hard.
15:44:35 <andythenorth> magic 'warehouses' attached to stations, bridging company networks. The graph could then be the union of the bridged networks
15:44:55 <andythenorth> and player 1 would get paid for source - warehouse leg, player 2 for warehouse -> destination leg
15:45:22 <andythenorth> warehouses must be built explicitly, possibly per-cargo
15:46:00 <andythenorth> cargo won't be put onto network unless there graph includes a route from source to destination
15:46:15 <andythenorth> so no partial payment unless the cargo can actually be delivered
15:46:33 *** brambles has joined #openttd
16:22:59 <Xaroth> Rubidium: you should be on vacation, shoo :P
16:29:48 <NGC3982> i have noticed one thing
16:30:06 <NGC3982> if i put a new station next to an industry and never use it, will that promote the industry closing?
16:32:14 <andythenorth> industry closing is quite simple, you can read the code for it if you checkout src
16:32:20 <NGC3982> and if i use the station once (make a pickup and leave)?
16:32:55 <andythenorth> Alberth: I'm pasting another version of that with word-wrap :P
16:34:33 <NGC3982> andythenorth: oh, how does one ..find that? it's in the binarys?
16:35:10 <andythenorth> industry_cmd.cpp
16:35:53 <NGC3982> i think i shall brake the asberger overlay of this channel with:
16:36:02 * NGC3982 farts smells like tacos
16:36:29 <andythenorth> brake or break? :P
16:37:35 <andythenorth> NGC3982: is your industry primary or secondary?
16:37:48 <andythenorth> newgrf or vanilla?
16:43:03 <andythenorth> mostly l2497 - 2555 in the file I linked :P
16:43:15 <andythenorth> actually it's not simple :P
16:48:24 <kjgh> hihihihihihihhihihihihihihihihihihihihihihihihihihihihihihihihihihihihhhhhhhhhhhhhhhhhhhhhhhhh
16:50:43 <andythenorth> does SAC personally approve every file uploaded to her forum?
16:50:51 *** tokai|mdlx has joined #openttd
16:51:44 <andythenorth> must....not....reply
16:51:58 <andythenorth> last time I got into an argument with SAC I nearly quit ottd stuff completely
16:55:02 <andythenorth> stay out of it is my advice :P
17:04:03 <Terkhen> meh, I barely coded anything this weekend
17:04:06 <TrueBrain> andythenorth: so please, stay out of it :P I would hate you quitting OpenTTD over SAC :P
17:04:16 <andythenorth> TrueBrain: too late :P
17:04:25 <andythenorth> can't help myself
17:04:34 <TrueBrain> this is so MPAA vs Google
17:04:41 <TrueBrain> MPAA saying Google has to screen content
17:04:47 <TrueBrain> Google saying it can't (legally) be done
17:04:49 <andythenorth> I saw the head of MPAA talk at a film festival once
17:04:56 <andythenorth> the MPAA are nuts
17:05:07 <TrueBrain> their vision is unrealistic
17:05:25 <andythenorth> they get the fundamentals completely utterly wrong
17:05:35 <TrueBrain> it seems that artists tend to create such vision, where they are the center piece; somewhat I can understand it, but ... wake up, this is 2012
17:05:40 <andythenorth> DVDs ship with a message I can't skip saying "don't steal films"
17:05:53 <andythenorth> rather than "thankyou for buying a legitimate DVD"
17:06:14 <TrueBrain> in this country, we have to pay a fee which allows us to copy music
17:06:25 <TrueBrain> in the old days, with casettes, you had to pay a fee over empty casettes
17:06:31 <TrueBrain> as you wer emost likely going to copy other casettes
17:06:32 <andythenorth> imagine going into a restaurant, and the first thing they do is put a knife through your hand, staking you to the table in case you try to leave without paying
17:06:37 <TrueBrain> this made it legal to make one copy for yourself at home
17:06:47 <TrueBrain> this fee continues on to CDs, DVDs, and now even MP3 players
17:06:54 <TrueBrain> all fine and all, but the fee is calculated like this:
17:07:03 <TrueBrain> a Casette contains 60 minutes of audio
17:07:09 * andythenorth is tempted to join Simuscape
17:07:16 <TrueBrain> a DVD can hold 1 minute for ever 3.5 MiB
17:07:21 * andythenorth probably wouldn't fit the entry criteria for Simuscape
17:07:26 <TrueBrain> 3GB / 3.5 * 60 * N = fee
17:07:42 <TrueBrain> it is THROUGH THE ROOF, and highly unfair ... as how many DVDs do you use to store music these days?
17:08:03 <TrueBrain> it is how old companies react to the modern world
17:08:12 <NGC3982> with UKRS2+, i cant use the replace function for 10T to 20T brake vans
17:08:16 <andythenorth> buggy whip scenario
17:08:18 <Terkhen> andythenorth: who would take care of your babies if you left? :P
17:08:33 <NGC3982> with the error message "Autorenew failed on Train8, This train requires a brake van"
17:08:49 <andythenorth> Terkhen: more to the point, who would invent new cargo routing mechanics?
17:08:57 <andythenorth> I happen to think this idea is rather good btw
17:09:08 <andythenorth> it's conceptually simple
17:09:22 <andythenorth> it might require quite a lot of....cached data....and updating of caches
17:09:34 <andythenorth> but it's only link graphs, we can do those fast right?
17:09:50 <Terkhen> that too, who would start random suggestions just to make us think about them? :P
17:10:04 <andythenorth> Eddi|zuHause of course
17:10:58 <Terkhen> your suggestions are usually far more farfetched :P
17:12:16 <andythenorth> usually 'remove xyz' :P
17:12:42 <andythenorth> I removed two whole sections this week from an app I maintain
17:14:55 <Terkhen> sadly we have to take care of hysterical raisins... forever :P
17:15:24 *** Rhamphoryncus has joined #openttd
17:16:29 <andythenorth> we've just convinced ourselves we do :P
17:17:35 <andythenorth> we coud do stuff like aggressively deprecate support for newgrfs older than (current spec -1)
17:17:51 <Terkhen> what purpose would that server?
17:17:57 <andythenorth> not sure there's any benefit at all
17:18:09 <andythenorth> we could rip out the default gameplay and replace it with newgrf
17:18:16 <andythenorth> simplifying industry code :P
17:19:29 <michi_cc> andythenorth: If you replace "store several graphs" with "walk link graph if needed" that reads suspiciously similar to YACD :p
17:19:43 <andythenorth> I wondered if it would converge :P
17:20:01 <andythenorth> is it expensive walking the link graph as needed?
17:20:58 <jgh> I notice that some "regular" scenarios work on chill's patchpack, but some of them don't and throw a chuck size error. Might it be possible to load and save them with a vanilla openttd to make work?
17:21:30 <Yexo> jgh: most likely the scenarios that give an error are too new
17:21:39 <Yexo> chill's patchpack is based on a really outdated version
17:22:04 <Yexo> it basically means you have to convert the scenario from a current version to an older version, which has never been supported
17:22:10 <Alberth> cpp is compatible upto some point, after that you get errors, but you can thus only load them with those newer versions
17:23:11 <michi_cc> andythenorth: Yes. This is why YACD is slow on large maps/big networks right now, as I spend little time on the cargo routing and a lot more on the destinations "Using means of your choice" :)
17:23:56 <andythenorth> my idea is basically inspired by playing YACD without understanding the implementation much
17:24:07 <michi_cc> Cargo routing needs something different than an A* pathfinder, but that is the only thing readly included in the trunk sources (e.g. lazy :)
17:24:19 <andythenorth> I started from the idea of a spot-price economy, storing cargo price on every tile....and found problems with that idea :P
17:24:43 <andythenorth> routing according to price is insanely hard :P
17:24:48 <frosch123> Zuu: yesterday i added a feature to the newgrf, content and sign list filter to filter on multiple words. does that also make sense for the script breakpoints?
17:29:05 <andythenorth> where is SAC's terms of service / mechanism for reporting abuse?
17:29:13 *** Matulla has joined #openttd
17:30:20 <Matulla> hi all i have seen that 1.2.1 is out ! Im on linux lucid 10.04 do i need to uninstall the old packet or just run the new deb over it ?
17:32:40 <andythenorth> I need to check that simuscape is not distributing my copyrighted work
17:35:21 <Terkhen> Matulla: most package managers ask you when you want to update a package, isn't that the case with ubuntu?
17:36:14 <Matulla> openttd is not updated
17:36:29 <Matulla> maybe it is not in the repros at that bersion
17:36:49 <Terkhen> what I mean is: when you try to update any package, your package manager should usually remove the old one and install the new
17:37:14 <Terkhen> that should also be true if you grab a package from outside the repositories and try to install it manually
17:40:23 <Matulla> let me find the main fetures now to get a good running game
17:40:46 * andythenorth has a simuscape account. Can't see any of my copyright being infringed :)
17:41:02 <TrueBrain> andythenorth: I can fix that if you like?
17:42:45 <Alberth> Matulla: starting of the program is a great feature :p
17:42:55 <andythenorth> TrueBrain: an intriguing offer but let's not eh? :)
17:43:38 <andythenorth> as a copyright holder, and de-facto the guardian of copyright for other contributors to my sets, I needed to check Simuscape was not a source of infringements
17:44:02 <andythenorth> specifically as SAC has posted screenshots at tt-forums showing modified graphics from my sets
17:45:18 <CIA-8> OpenTTD: translators * r24338 /trunk/src/lang/latvian.txt:
17:45:18 <CIA-8> OpenTTD: -Update from WebTranslator v3.0:
17:45:18 <CIA-8> OpenTTD: latvian - 1 changes by Parastais
17:45:45 <andythenorth> as a responsible copyright holder, I feel bound to check simuscape
17:46:46 <andythenorth> as I hold the opinion (no evidence, these are only ungrounded suspicions, not presented as truth or fact) that Simuscape may be a private membership club for the distribution of digital artefacts that infringe copyright, colloquially known as 'warez'
17:47:46 <Terkhen> andythenorth: create an account and check that forum and simutrans everyday
17:48:03 <andythenorth> there is an additional private section of Simuscape that I do not have access to yet
17:48:35 <andythenorth> "Artists Guild". My suspicion is that if there is infringement taking place (I have no evidence there is, nor do I claim factually there is), that is where it might be.
17:49:10 <Pinkbeast> Going back a bit, RRT3 did route-by-spot-price and this produced endless trouble with getting past local max/minima
17:49:29 <andythenorth> I don't fit the rules for admission to Artist's Guild
17:50:03 <Terkhen> andythenorth: what are those rules?
17:50:10 <andythenorth> Terkhen: you can't see them
17:50:16 <andythenorth> you have to register
17:50:23 <andythenorth> and I can't repost them - that would infringe copyright
17:50:42 <andythenorth> I am going to PM SAC requesting access to Artists' Guild, and if I have not been granted access, or told that access is in process within a reasonable timeframe, I'm going to notify the ISP that I suspect copyright infringement
17:50:52 <andythenorth> what is a reasonable time frame? 10 days?
17:51:39 <Terkhen> andythenorth: I think that you should sleep on that decision :)
17:52:02 <andythenorth> I need to protect my rights, and the rights of those on who's behalf I act :)
17:52:10 <TrueBrain> andythenorth: think about your blood preasure
17:52:13 <TrueBrain> and just leave the boy alone
17:52:19 <andythenorth> I should file this under "don't be a wanker"
17:52:58 <andythenorth> shall we discuss cargo routing instead?
17:53:46 <Terkhen> sounds more constructive
17:56:18 <Zuu> frosch123: If you want to match on spaces or two words, how does one do that with your changes?
18:00:05 <TrueBrain> andythenorth: or much more serious: I am out of Coca Cola
18:00:25 <andythenorth> fortunately there are shops
18:00:31 <TrueBrain> the next best thing I have is beer
18:00:39 <TrueBrain> there are shops, but none I can get into without some kind of tool atm :P
18:00:45 <andythenorth> and your currency is stiil not collapsed
18:00:54 <andythenorth> oh shops are shut?
18:01:00 <TrueBrain> I live in a country with a strong market, so that won't happen no :P
18:01:14 <andythenorth> we used to have those in the uk
18:01:18 <andythenorth> now we have retailing instead
18:03:06 <frosch123> Zuu: you can quote with " and ' like in a unix shell
18:04:09 <frosch123> "bla 'boo'"'doo " hoo' matches for bla 'boo'doo " hoo
18:04:32 <andythenorth> so no comments on the 'Warehousing' suggestion I made for orders in my paste above?
18:05:22 <andythenorth> strikes me that if cargo is routed by a specific graph, it's trivial to merge graphs across companies
18:06:18 <Zuu> frosch123: I guess it could make sense to allow OR:ing togeather strings. Which is what you do by threading un-quoted spaces as OR.
18:06:47 <Alberth> s/threading/treating/ I hope :)
18:07:16 <frosch123> Zuu: they are treated as AND
18:07:29 <frosch123> quoted strings are taken as a whole
18:09:09 <frosch123> so you could still do the same as before by quoting the whole input
18:09:12 <Zuu> frosch123: hmm, I don't know if it is inutuative, but I guess it might be useful for someone. Did you see FS#5206 (a recent patch by me "fix" break on log to also work with game scripts)
18:09:42 <Zuu> Your change is however not really in conflict with my patch.
18:10:43 <frosch123> i would only change the two lines containing strstr and strcasestr
18:11:10 <frosch123> i think it would basically allow to place multiple breakpoints
18:11:33 <Zuu> I guess, since you have made this change for the other search/match fields in OpenTTD, it makes sense to have it here for consintency.
18:12:00 <Zuu> As you say it also adds some new ways to use it. (but requires existing usage of it to change a bit (add quotes))
18:14:05 <frosch123> ok, i'll add it, so it can be tested :)
18:36:36 <Alberth> read about the focussed D* algorithm ?
18:37:30 * Alberth is pondering some sort of Python experiment
18:38:25 * andythenorth has been trying to find Simuscape's abuse email :P
18:39:44 <andythenorth> what are we going to do with D* ?
18:40:19 <Alberth> it's A*, but once you have a path, you can adjust it for changes
18:40:36 <Alberth> without recomputing everything
18:41:14 * andythenorth had better read about A* again
18:41:55 <andythenorth> asking SAC how she can help me protect my rights is going to be painful not fun :( Might avoid that for now
18:43:23 <andythenorth> in my plan for weighting nodes by number of links, it would be permissible to abort
18:43:41 <andythenorth> for any given node, if another sequence uses more than than the currently stored weight, it's time to stop
18:43:51 * andythenorth is a graph theory newbie :P
18:44:21 *** andythenorth has left #openttd
18:44:25 *** andythenorth has joined #openttd
18:44:25 *** andythenorth has left #openttd
18:44:51 *** andythenorth has joined #openttd
18:46:16 <andythenorth> my scheme demands walking the entire graph
18:47:11 <Alberth> I am also wondering how to use the ant colony idea here
18:47:40 <Alberth> it seems that once you have arrived you deposit pheronomes along the path you traveled
18:47:57 <andythenorth> on the way back :P
18:48:31 <andythenorth> my scheme is pretty simple - it just stores num nodes on a link. It would be easy for graphs that are not very interconnected
18:48:47 <andythenorth> a nice tidy tree would be easily done
18:48:55 <andythenorth> but if a lot of nodes are multiply connected to other nodes might be...inefficient
18:48:59 <andythenorth> oh, but they would form a set
18:49:05 <andythenorth> and could be optimised out
18:49:20 * andythenorth suspects it's not that easy :)
18:49:51 <andythenorth> it's just a shortest path problem :P
18:53:18 <andythenorth> I don't even need shortest path measuring link distance
18:53:24 <andythenorth> just shortest number of nodes
18:53:48 <al3x> hey folks... i'm just working on a presentation about openttd as an outstanding open source software game project. i'm about to finish my researches and am just wondering: is there a special hidden detail, a funny spoof, eastereggs or crazy features being planned for upcoming versions? any hint would be appreciated :)
18:54:15 <Sacro> it's no fun if we tell you
18:55:05 <andythenorth> definitely planned for 1.3
18:55:24 <andythenorth> definitely not started ;)
18:55:38 <andythenorth> instead of encoding newgrfs in nfo (bytecode), we'd encode them in xml
18:55:47 <Yexo> al3x: what are you looking for? Something a player might notice, something about development, something about user cooperation or newgrf development?
18:56:16 <andythenorth> the format would be <node type="byte">0A</node><node type="word">3F80</node>
18:57:09 <Yexo> s@</node>@ @ s@<node type="byte">@
18:57:09 <al3x> andythenorth: thanks, sounds reasonable, why didnt you (programmers) do it in the first place?
18:57:27 <Yexo> s@</node>@ @ s@<node type="byte">@\bx@ s@<node type="word">@\wx@
18:58:07 <andythenorth> you asked for spoofs :P
18:58:16 <andythenorth> the humour here isn't of the highest standard tbh
18:58:26 <andythenorth> we won't win any awards
18:58:56 <andythenorth> there are no easter eggs I know of
18:59:12 <andythenorth> big new features aren't really known in advance ;)
18:59:15 <Yexo> wasn't there some easter egg grf?
18:59:22 <al3x> Yexo: well, since it's a computer science course, i'm interested both in development and playing-related information :)
18:59:29 <andythenorth> there are various easter egg / silly grfs
18:59:37 <andythenorth> one that changes 'goods' to 'stuff'
18:59:45 <andythenorth> one that makes the ufo available as a plane
18:59:50 <andythenorth> one that adds christmas tree
19:00:37 <al3x> cool, i will ask google for it
19:00:55 <andythenorth> try tt-forums search
19:01:14 <andythenorth> sounds like a nice presentation :)
19:01:57 <al3x> every ottd-presentation is a nice presentation i guess ;)
19:02:04 <frosch123> al3x: the plan is to move more functions to the right mouse button
19:02:12 <frosch123> to spoil users of touchscreens
19:02:34 <Terkhen> al3x: you can check some features that already prepared and implemented for 1.3.x in the changelog
19:02:45 <Terkhen> some are quite nice but I wouldn't call them "crazy" or "easter eggs" :P
19:02:58 <frosch123> Terkhen: he, that's the first answer, which is no trolling :o
19:04:07 * andythenorth could work on some trucks or something :P
19:04:11 * andythenorth is tempted to do actual work
19:04:19 <andythenorth> making actual internet
19:04:26 <al3x> frosch123: hehe :) well to be honest, i like the right mouse-button scrolling ^^
19:06:15 <al3x> Terkhen: thanks, im going to look into the changelog
19:07:57 <al3x> i hope i can find some comparison images.. original ttd vs. new graphics
19:08:31 <Terkhen> for that you might want to check the OpenGFX thread
19:08:53 <Terkhen> and, if you own the original TTD, you can just use the original graphics set and do some screenshots yourself
19:09:27 <al3x> yeah... would be so easy... ;)
19:11:21 <al3x> hmm... well i will consider it. might be a good live comparison
19:11:57 *** kkimlabs has joined #openttd
19:12:52 <Terkhen> ooh, good idea to link those, I forgot about them :P
19:13:15 <al3x> Alberth, thank you very much. thats great, better than enything expected :)
19:13:58 <Alberth> it just has this annoying intro window everywhre :)
19:16:04 <andythenorth> fellow traveller
19:16:09 <Terkhen> I have never tried it so I can't talk about it
19:16:20 <Terkhen> I went into OpenTTD because I loved the original TTD
19:16:25 <Alberth> games are totally different
19:16:39 <Terkhen> since simutrans isn't exactly the same thing, I did not feel compelled to try it
19:16:46 <Terkhen> they are games for different tastes I guess
19:17:19 <NataS> I hate Simutrans, but wish OTTD had all it's cool features.
19:17:50 <Alberth> __ln__: he wants to hate OpenTTD too :p
19:19:36 <al3x> Alberth: thanks for the additional screenshots - now i have to choose between all them... :( hehe
19:19:51 * andythenorth wonders if players would care about cargo not taking shortest / fastest route
19:20:44 <andythenorth> would you though? :)
19:21:03 <Alberth> al3x: everybody plays temperate, almost nobody plays toyland, yet it has very nicely animated industries
19:21:19 <andythenorth> actually it might be a non problem
19:21:37 <NataS> i play tropical almost exclusively :p
19:21:49 <al3x> i didnt look into simutrans, but i think it said -no-multiplayer- so i dismissed it.
19:22:02 <Alberth> I think it's not a big problem, as long as it does not happen too much
19:22:17 <andythenorth> shouldn't happen at all, except in pathological cases
19:22:39 <FLHerne> I would probably create such cases on a regular basis, though :P
19:22:48 <andythenorth> depends on definition of 'shortest' and/or 'fastest' :P
19:22:49 <Alberth> not sure, people also expect cargo to evenly use all routes
19:23:16 <andythenorth> I am only counting number of hops. I read the cargodist docs on balancing; trying to estimate link capacity seems hard / foolhardy
19:23:28 * FLHerne tends to create massive interconnected messes of all transport modes, with about 10 possible routes for everything :D
19:23:39 <al3x> ever heard about locomotion? (by chris sawyer)
19:23:54 <Alberth> andythenorth: just amount of waiting cargo?
19:24:15 <Alberth> al3x: he, you are here with transport fanatics :)
19:24:45 <al3x> hehe.. its all a test, you know ;)
19:25:06 <NataS> even worse than Simutrans
19:25:19 <NataS> Uglier than simutrans, and less features
19:25:48 <Alberth> al3x: pretty pictures, but much less scalability
19:25:53 <andythenorth> in my routing system, both of these routes are valid
19:26:12 <al3x> before i started researching ive never heard about locomotion. looking at some screenshots, im not really sad about it
19:26:15 <andythenorth> despite that one might be much longer (measured in tiles)
19:26:21 <FLHerne> andythenorth: Worrying :-(
19:26:28 <andythenorth> work it through...
19:26:42 <Yexo> al3x: ever heard about transport empire? or p1sim?
19:26:45 <andythenorth> the long route might use faster vehicles, or lower running cost vehicles, or be less congested
19:26:59 <andythenorth> there's no simple deterministic way to measure that, too many variables
19:27:10 <Alberth> andythenorth: seems fine to me to have 2 routes
19:27:17 <FLHerne> Depends how good it is at figuring it out then, I suppose
19:27:22 <andythenorth> therefore the abiding rule in my scheme is 'cargo hates to be transshipped'
19:27:31 <andythenorth> irl, cargo hates to be transshipped
19:27:31 <al3x> Yexo: no, but im about to google it right away :)
19:27:51 <Alberth> al3x: tt-forums.net :)
19:28:04 <Yexo> al3x: don't bother, but you can save yourself some time and just look at tt-forums.net
19:28:07 <FLHerne> My entire network is based on transhipping everything, everywhere :P
19:28:16 <andythenorth> mine tend to be too
19:28:33 <andythenorth> transshipping still works, but cargo hates it
19:28:56 <Alberth> andythenorth: ant optimization may do cargo link capacity all by itself
19:29:02 <FLHerne> Which is bad if some routes have 10 or so transfers :-(
19:29:14 <andythenorth> I think capacity is for the player :)
19:29:22 <andythenorth> work it through, the short route on the left, the vehicles cycle faster
19:29:35 <andythenorth> my scheme makes no attempt to determine optimum total route
19:29:41 <andythenorth> it just favours fewest hops
19:29:52 <andythenorth> the decision is made when cargo is loaded to a vehicle
19:30:11 <andythenorth> so adjusting the number / cycle time of vehicles lets player control where cargo actually goes
19:30:44 <andythenorth> it's slightly pathological to get exactly the same number of hops on such varying route lengths anyway :P
19:30:52 <FLHerne> Hop-counting sounds like a bad idea, would certainly wreck my network at the moment...
19:31:45 *** GBerten2936 has joined #openttd
19:31:48 <andythenorth> there are certain conditions I should mention, e.g. cargo *destination* is not determined by number of hops
19:31:51 <andythenorth> just the routing
19:32:17 <andythenorth> and it's scale free with respect to actual number of hops in use
19:32:22 <FLHerne> If a straight route across a bay is [RV -> Tram -> Train -> Tram -> Ship -> Tram] (all but train very short) versus [Train to far side of map -> Train back again], that will be a problem :-(
19:33:35 <andythenorth> you can use 3, or 300 hops, the routing wouldn't care. It would just pick the smallest number of hops to route by, from the available choices
19:34:01 <andythenorth> if route A has 300 hops and route B has 299, route B is used
19:34:39 <Alberth> hmm, can you attach different orders to a autorefit ?
19:34:46 <Yexo> what happens if you both two routes, route A with 4 hops and route B with 5 hops. Will the cargo start to take route B when route A is overloaded?
19:35:38 <Alberth> Yexo: my guess is 'no'
19:42:07 * FLHerne worries about oversimplistic cargo routeing
19:42:33 <Terkhen> like the one we have right now? :P
19:47:13 <Alberth> no routing at all causes less problems :p
19:52:40 * FLHerne is quite happy with CargoDist
19:53:09 <FLHerne> YACD is annoying, I don't like the game dictating where to run trains
19:53:59 <FLHerne> I don't get the performance issues that people moan about - why can't it just be trunked as an option?
19:54:01 <NataS> but cargo dist is exploitable, and can break in long chains
19:54:13 <NataS> but it's better than nothing
19:54:18 <NataS> I can't imagine playing without it
19:54:22 <andythenorth> Yexo: if route A is overloaded, fix route A :P
19:54:42 <FLHerne> I haven't noticed any bugs or breaking, what's exploitable?
19:55:16 <andythenorth> but also my scheme pays no attention to load factors. If a vehicle travelling route B arrives at your station, it will be loaded with cargo for route B
19:55:24 <NataS> well there's nothing stoping you from only sending goods to the town on the other side of the map
19:55:31 <NataS> but that's possible in normal too
19:56:09 <NataS> i guess the problem is, there's no real advantage to distributing as opposed to just shiping as much as you can as far as you can.
19:56:09 <andythenorth> FLHerne: in your diagram, the route on the left is shorter so would be favoured
19:56:22 <andythenorth> actually, where's the start point? :P
19:56:35 <FLHerne> Top left to top right :P
19:56:50 <andythenorth> ok so cargo would take the very long route
19:57:09 <Alberth> nice and easy to predict :)
19:57:13 <FLHerne> That's the problem, yes...
19:57:23 <andythenorth> very easy to predict
19:57:54 <Alberth> andythenorth: do you count station, or transfer stations?
19:58:06 <Yexo> is a "hop" defined as a place where cargo has to be transfered between vehicles or a place where a vehicle has to stop?
19:58:07 <andythenorth> transfers would be automatic
19:58:07 <Terkhen> FLHerne: IIRC it was relatively easy to add cargodist to YACD
19:58:25 <Yexo> nvm, you answered that :)
19:58:30 <andythenorth> hops are explicit (not implicit) orders I think
19:58:31 <FLHerne> Most of my 'intended' routes have myriad transfers, so I'd probably get all the cargo using accidentally-formed ones instead...
19:58:42 * andythenorth wonders if implicit orders would need to be counted
19:58:50 <Alberth> so if you add say 100 drive-through station to the long route, it would take the short route instead?
19:58:51 <andythenorth> FLHerne: your case is well made
19:58:51 <FLHerne> Terkhen: to produce what effect?
19:58:54 <Yexo> explicit and implicit orders need to be treated exactly the same
19:59:02 <andythenorth> Alberth: possibly :P
19:59:31 <Alberth> cargo hates stations in general :p
19:59:36 <Yexo> FLHerne: so the user can chose between the yacd and cargodist routing options
20:00:06 <FLHerne> Oh, ok. That would be ideal :D
20:00:08 <andythenorth> my plan is that cargo prefers links with lower 'weight'. Default calculation for 'weight' is num hops
20:00:17 <andythenorth> then expose it to GS or plugins or such
20:00:17 * FLHerne wants it yesterday :D
20:00:20 <andythenorth> for complicated crap :P
20:01:01 <Alberth> FLHerne: find a light-weight cargo routing algorithm :)
20:01:31 <andythenorth> find a light-weight, predictable, reliable cargo routing algorithm, that is MP safe and supports all play styles
20:01:41 <Terkhen> FLHerne: a triple switch allowing you to set "none", "cargodist" or "yacd" behavior
20:02:03 <Terkhen> but yeah, performance is the biggest issue here
20:02:31 <FLHerne> I've been using CargoDist for 18 months now, still don't see any problems with it :-)
20:02:54 <FLHerne> I don't have an insanely fast PC, either :P
20:03:39 * andythenorth ponders FLHerne's diagram
20:03:56 <andythenorth> FLHerne: I conclude your route is silly :)
20:05:14 <andythenorth> the route shows the possibility of unintended consequences :|
20:05:34 <FLHerne> Well - I use HEQS lorries in the mountains, and trains to the coast, and trams to get through towns to the port, and ships across the bay, and trams to the transfer yard on the other side, and then more trains, and then trams :D
20:06:17 <FLHerne> *Generic idea, not necessarily specific description of a particular service
20:07:34 <andythenorth> calculating the actual route length would be possible I guess
20:09:46 <frosch123> maybe you should just leave the routing decision to the player :)
20:10:07 <frosch123> calculate prices depending on supply and demand
20:10:18 <frosch123> but let the players decide which destination shall hop on which vehicle
20:11:00 <frosch123> magic automatics do not necessarily improve gameplay
20:14:28 <andythenorth> I wanted to find a way to simplify transfers
20:16:27 <andythenorth> keep the entire current system, change only the economy? :P
20:17:07 <frosch123> you still need to control when to transfer or when to deliver arriving cargo
20:17:41 <frosch123> so, you could configure at a station: accept this percentage of cargo, transfer this percentage to this group of vehicles, the rest goes to the other group
20:17:44 <andythenorth> I didn't have an answer for 'when to unload'
20:18:12 <frosch123> more like the simutrans approach: first the orders, then the vehicles :p
20:18:32 <andythenorth> everything is connected :P
20:18:51 <andythenorth> routes is shared orders, shared orders are (according to some) groups, groups are (according to some) consists
20:18:54 <frosch123> the main advantage: if the player decides what amount of cargo goes on which route, you do not have to bother about conditional orders and such stuff
20:18:55 <andythenorth> the wheel of life turns :P
20:22:40 <frosch123> you also do not have issues with supply cargos. you can just deliver where you want them to and not bother about payment
20:23:03 <frosch123> so you have no conflicts between cargo routing and industry production behaviour
20:23:29 <andythenorth> localised demand curve
20:28:13 <andythenorth> seems hard to untie 'demand' from newgrf industry / houses
20:30:00 <frosch123> if you let newgrfs decide the payment, and the routing decide the payment everything should be fine, shoudn't it?
20:30:18 <andythenorth> payment / demand /s ?
20:30:30 <frosch123> if a industry refuses to accept cargo, while the routing things it shall take a lot of cargo, you get high payment for few cargos
20:30:44 <frosch123> if the newgrf accepts unlimited, the payment limits what it worth delivering
20:31:15 <andythenorth> so how would demand be modified? Newgrf decides? Or based on frequence of delivered cargo or such?
20:31:20 <frosch123> [22:29] <frosch123> if you let newgrfs decide the payment, and the routing decide the payment everything should be fine, shoudn't it? <- i meant: newgrfs decide acceptance, routing decides payment
20:31:49 <andythenorth> I like the clean separation
20:31:55 <michi_cc> andythenorth: Does Simuscape have any abuse contact information, maybe hidden inside the registration wall, or is SAC not doing what she's demanding?
20:32:14 <andythenorth> it has the PHPBB standard
20:32:25 <andythenorth> - the PHPBB standard ToS
20:32:41 <andythenorth> - which disclaims any responsibility
20:32:44 <michi_cc> (Better check twice or my post is going to be silly)
20:32:50 <andythenorth> - it has an administrators page
20:33:25 <michi_cc> Wants me to register, so doesn't count.
20:33:31 <andythenorth> - it has a 'welcome center'
20:33:42 <andythenorth> no there's no information for you without registering afaict
20:34:10 <andythenorth> beware that there are some links that are so low contrast - they don't just fail 508 / DDA - they fail to be visible to me who has grade A eyesight
20:34:16 <andythenorth> so there may be links I'm failing to see
20:35:05 <andythenorth> as a rights owner, I had to register to see if my work is being infringed
20:35:26 <andythenorth> I then have to be approved for registration in a special members area where the graphics work is shared
20:36:01 <andythenorth> I can't see any ToS etc
20:36:44 <andythenorth> it also illegally sets cookies
20:37:05 <andythenorth> and provides no privacy policy, no cookie policy, no ownership information afaict
20:37:37 <andythenorth> it's not clear who I would contact for:
20:37:44 <andythenorth> - a DMCA / EDEC violation
20:37:55 <andythenorth> - a privacy violation or concern about how my data is used
20:38:12 <andythenorth> e.g. it collects more than one piece of sensitive personal data (email, my IP I assume)
20:38:25 <andythenorth> - it sets cookies without my consent
20:40:21 <andythenorth> whois does give a gmail address, which might be a useful first point of contact
20:40:47 *** kkimlabs has joined #openttd
20:41:21 <planetmaker> who has access to 'whois'? (yes, everyone, but... who 'normal' user knows about it?
20:43:32 <andythenorth> I make no judgement
20:43:40 <andythenorth> I'm sticking to facts I can substantiate
20:45:01 <andythenorth> demand-per-cargo on every tile
20:45:23 <andythenorth> ho ho, that would have the side effect of slightly penalising station walking
20:45:59 * andythenorth thinks the Warehouse order is genius and should be explored until it's broken :P
20:46:13 <andythenorth> "Goto X and Warehouse"
20:46:46 <andythenorth> puts the cargo back on the map at that point
20:47:02 <andythenorth> distributing it to any nearby station
20:47:04 <planetmaker> demand per tile... isn't that what yacd basically did? Actually... what houses and industry DO?
20:47:18 <andythenorth> doesn't vary price ;)
20:47:37 <andythenorth> I'm doing without routing etc in this (new / old) idea
20:47:42 <andythenorth> just adjust price
20:47:50 <andythenorth> routing is...hard :P
20:48:25 <andythenorth> frosch123: conditional order: goto A if price at A > price at B, else go to B
20:48:33 <andythenorth> shipping lines do that :P
20:49:12 <andythenorth> ships wait at sea to see which market offers best spot price on commodities
20:58:28 *** Chris_Booth has joined #openttd
21:05:35 <andythenorth> if I figure it out right, a demand (price) function does something to reduce the annoying part of the 'get paid to transport halfway across the map' issue
21:05:47 <andythenorth> the annoying part being when there are other, closer sources of supply
21:06:01 <andythenorth> whereas the payment method (distance, speed) is correct
21:06:57 <andythenorth> the presence of a nearby source of supply could reduce the demand at destination, thereby reducing the price gradient between the destination and far-off sources of supply
21:07:25 <andythenorth> depends if I figure it out right though :P
21:08:22 <NGC3982> i dont even understand how that happend
21:08:49 <Terkhen> powerful lobbying and lots of money
21:09:12 <Terkhen> and... people preferring to believe simpler things :)
22:03:59 *** Zeknurn has joined #openttd
22:17:52 * NataS is slowpoke, but Supply and demand would be nice
22:18:26 <NataS> because it's kind of anoying how the best way to play is to ignore the closest and just send goods to the other side of the map
22:52:40 *** Hazzard has joined #openttd
23:06:14 *** Djohaal has joined #openttd
23:58:13 *** Hazzard has joined #openttd
continue to next day ⏵
