IRC logs for #openttd on OFTC at 2018-09-04
⏴ go to previous day
00:13:11 *** Wacko1976 has joined #openttd
00:28:49 <Samu> and for the airport case:
00:28:51 <Samu> if (AITile.IsStationTile(t) && (AIAirport.IsAirportTile(t) || AITile.GetOwner(t) != AICompany.ResolveCompanyID(AICompany.COMPANY_SELF) && AIController.GetSetting("is_friendly"))) {
00:29:18 <Samu> together, i shall see more airports, I hope
03:18:38 *** Gustavo6046 has joined #openttd
03:36:03 *** WWacko1976-work has joined #openttd
04:37:41 *** rocky113844 has joined #openttd
05:22:50 *** Supercheese has joined #openttd
05:35:44 *** haudrauf has joined #openttd
06:15:49 *** sim-al2 has joined #openttd
06:24:04 *** wodencafe has joined #openttd
06:27:54 *** rocky1138 has joined #openttd
06:29:50 *** greeter has joined #openttd
07:54:08 *** Supercheese has joined #openttd
07:55:53 *** sla_ro|master has joined #openttd
08:01:42 *** snail_UES_ has joined #openttd
08:31:17 *** Progman has joined #openttd
08:35:31 *** andythenorth has joined #openttd
09:32:44 *** Supercheese has joined #openttd
09:50:48 *** andythenorth has joined #openttd
09:52:10 *** snail_UES_ has joined #openttd
11:11:51 <Groope> Hey, does automatic signal placement works in Chris Sawyer's locomotion or just in OPEN TTD ?
11:18:08 <peter1138> Does it it work in Microsoft Train Simulator?
11:20:21 <peter1138> Locomotion is a commercial game from 2004. It's not related to OpenTTD.
11:23:43 <Groope> okey, but there is any way to put signals automatically ? or I have to install them one by one ?
11:30:34 <peter1138> In Locomotion? No idea, haven't played it for over a decade.
11:30:52 <peter1138> IIRC Locomotion's signals are too basic to be able to do much useful with.
12:07:58 <Samu> andythenorth: hey, what is the name of that roadhaul newgrfs that was used to test notroadtypes? i wanna try the equivalent version for 1.8.0, do youknow the name?
12:09:04 <Samu> btw, did you fix the articulated vehicles issue?
12:14:44 <Samu> gonna test the articulated part precisely with the ai
12:42:11 <Samu> you don't have any articulated mail truck that is faster than the non articulated counterpart. The way I have the AI set up, it would always look for the fastest engine
12:43:02 <Samu> which means, no articulated chosen, hmm i gotta make a few changes
12:45:47 <Samu> oh, i see a window of opportunity between 1968 and 1971
12:46:33 <Samu> before goldmire courier truck is available, quickset box truck is the fastest mail truck and is articulated
12:57:48 *** andythenorth has left #openttd
14:02:23 *** chomwitt has joined #openttd
14:10:23 <Samu> it is done, AI is now refit and articulated viable!
14:50:31 *** sla_ro|master has joined #openttd
15:16:22 *** snail_UES_ has joined #openttd
15:20:23 <Samu> darn, lost helicopters are annoying
15:20:45 <Samu> how do they end trying to go to an hangar of an airport that does not exist anymore
15:58:09 <Samu> okay, i found the problem
15:59:18 <Samu> this does not include orders of the kind "nearest hangar"
15:59:39 <Samu> "service at nearest hangar"
16:00:03 <Samu> that a vehicle is heading to the hangar
16:14:09 *** chomwitt has joined #openttd
16:21:19 <Eddi|zuHause> nielsm: so, are you volunteering to host a party for the next commit, then? :p
16:22:25 <Eddi|zuHause> (also, didn't we already have an r25k party, how can we then celebrate 23k now?)
16:23:26 <nielsm> the 23k is linear count from now until beginning of the master branch, while 25k is number of svn revisions across all branches
16:26:37 *** HeyCitizen has joined #openttd
16:30:41 <Samu> what is ResolveOrderPosition doing
16:31:17 <Samu> I am trying to skip orders to the next
16:32:01 <Samu> AIOrder.SkipToOrder(i, AIOrder.ResolveOrderPosition(AIOrder_ORDER_CURRENT) == 0 ? 1 : 0);
16:32:14 <Samu> orderlist only contains 2 orders
16:32:23 <Samu> is this gonna do what I want?
16:34:06 <Samu> AIOrder.SkipToOrder(i, AIOrder.ResolveOrderPosition(AIOrder.ORDER_CURRENT) == 0 ? 1 : 0);
16:34:46 *** Tweeterman has joined #openttd
16:37:18 <Tweeterman> I can't find a Japanese Track set for the Total Bridge Renewal Japan GRF. Any help?
16:38:25 <Samu> not familiar with that newgrf
16:45:58 <Samu> I am the worst coder in the world
16:46:11 <Samu> AIOrder.SkipToOrder(i, AIOrder.ResolveOrderPosition(i, AIOrder.ORDER_CURRENT) == 0 ? 1 : 0);
16:46:24 <Samu> 3rd fix on the same line
17:14:16 *** IgnoredAmbience has joined #openttd
17:27:57 *** Wormnest has joined #openttd
17:30:06 *** WWacko1976-work has quit IRC
17:53:30 *** sla_ro|master has joined #openttd
18:02:56 *** Thedarkb has joined #openttd
18:11:56 *** frosch123 has joined #openttd
18:14:34 *** andythenorth has joined #openttd
18:31:42 *** Thedarkb has joined #openttd
18:34:21 <DorpsGek> andythenorth: alberth was last seen in #openttd 2 days, 2 hours, 46 minutes, and 1 second ago: <Alberth> less confused now :)
18:37:19 *** HerzogDeXtEr has joined #openttd
19:02:43 <flitz> Hi. I just wanted to revive an old openttd project of mine. I switched the trunk URL I used to use to the new github one and wanted to update the +openttd master branch on it. But during the merge my old version of the game's files and the current master have merge conflicts in almost 400 +files. So it seems like a nightmare to merge the 2 versions.
19:05:57 <nielsm> not sure if this is the best way, but assuming your old work is based on the pre-github move tree, it might be best to first rebase your work onto the equivalent revision of your old base in the new repository
19:06:19 <nielsm> and then merge in the changes from then until current head
19:07:00 <nielsm> i.e. start out with what is hopefully a clean rebase and then do the "dirty" merge after that
19:07:18 *** Wormnest has joined #openttd
19:08:46 *** greeter has joined #openttd
19:10:15 <Eddi|zuHause> yeah, don't switch to github "head", but to github "old revision that you had back then"
19:11:39 <Eddi|zuHause> from there, you can also try to make smaller steps to upgrade to non-conflicting revisions
19:12:37 <Eddi|zuHause> as always with jobs that seem overwhelmingly complex and hard, separate them into easier smaller steps
19:17:46 *** IgnoredAmbience has left #openttd
19:24:56 <LordAro> flitz: rebase --onto is useful here
19:28:21 <flitz> I was also already thinking about just dropping my old git history and applying the patch in small chunks onto the new tree. that would give me the opportunity to review it. but its still a lot of work with about ~10k lines of code in the patch
19:33:36 <Eddi|zuHause> really, i would start patching the old revision, as then you can commit it to git and have an easy fallback if you make mistakes updating
19:34:10 <Eddi|zuHause> then maybe try bisecting the update so you identify which revisions conflict
19:34:29 <Eddi|zuHause> if you know the revision that conflicts, you often get an idea from there how to resolve the conflict
19:49:56 *** ToBeFree has joined #openttd
19:52:09 <andythenorth> so should I get Dwarf Fortress?
20:00:42 <frosch123> only if you can survive blinking text
20:02:01 *** HeyCitizen has joined #openttd
20:03:11 *** Progman has joined #openttd
20:05:20 *** Wacko1976 has joined #openttd
20:06:46 <frosch123> hmm, it's a new one
20:06:47 *** ChanServ sets mode: +v tokai
20:06:59 <frosch123> it's the standard message when people change vehicle grfs in game
20:07:03 <frosch123> but the gamelog is fine
20:07:04 <nielsm> let's see where the message exists
20:08:03 <LordAro> road vehicle trying to split itself?
20:08:33 <nielsm> FollowPreviousRoadVehicle returning INVALID_TRACKDIR
20:08:39 <frosch123> i can only imagine it being a timing problem
20:08:42 <LordAro> removing a road piece under an articulated road veh?
20:08:49 <LordAro> i'm speculating wildly here
20:08:54 <nielsm> yeah it smells a bit like the road disappearing
20:08:57 <frosch123> like the ai being able to reverse vehicles in weird spots
20:09:02 <frosch123> or selling/buying them
20:09:17 <frosch123> or removing tracks beneath a vehicle
20:12:12 <LordAro> can't manage to remove a road beneath a road vehicle myself
20:12:18 <LordAro> could need some AI preciseness
20:12:28 <nielsm> far too many ways FollowPreviousRoadVehicle can return INVALID_TRACKDIR
20:12:49 <frosch123> if it is reproducible with the ai, you can print the commands
20:13:00 <frosch123> LordAro: maybe depot, road stop, level crossing, ...
20:13:03 <nielsm> might be reversing while entering/leaving a depot or wormhole
20:13:22 <frosch123> yeah, or tunnel, bridge, ...
20:14:09 <Wolf01> Had a lot of those in NRT when vehicles tried to take the wrong path
20:14:31 <Wolf01> Like, eltrams trying to go on non electric rail
20:14:36 <frosch123> ah, so also removing/adding tram/road to existing road/tram :)
20:16:56 <nielsm> if it's even reproducible (need to let the save run for 11+ years??) it's probably a good idea to augment the code to print some more details on the failed movement
20:17:34 <Wolf01> I had the same discussion with my coworkers today
20:17:35 <frosch123> usually crash.sav is sufficient
20:17:55 <frosch123> crash.sav contains everything
20:19:10 <LordAro> frosch123: not if it's AI triggered
20:19:43 <frosch123> crash.sav, not crash.png
20:20:03 <frosch123> crash.sav contains the broken rv, and you can check the landscape around it and guess what changed
20:20:12 <nielsm> well it happens inside the roadveh movement code, not as a result of another command
20:20:35 <frosch123> it most likely happens directly after a command
20:22:04 <nielsm> oh it's one-and-a-half year, not eleven years, the save needs to run
20:22:27 <andythenorth> I was testing a set of patches for NML
20:22:40 <frosch123> nielsm: yes, but it is not exactly a light-cpu game :p
20:22:43 <andythenorth> it was 3 weeks ago :P
20:23:42 *** [1]sim-al2 has joined #openttd
20:23:42 *** sim-al2 is now known as Guest2407
20:23:42 *** [1]sim-al2 is now known as sim-al2
20:24:12 <andythenorth> and then there's the prop 28 question
20:24:24 <andythenorth> nielsm: ^ not sure how to best integrate those patches
20:26:00 <frosch123> 1999-05-30, this takes ages :p
20:26:41 <andythenorth> also what are those terrible newgrf ships :P
20:26:59 <frosch123> dbg: [sprite] Tried to load character sprite #15 as a recolour sprite. Probable cause: NewGRF interference <- is that their fault?
20:27:31 <andythenorth> more terrible grf choices :P
20:28:22 <andythenorth> so many things I should do :P
20:28:49 <LordAro> still no idea how to fix that properly
20:29:15 *** gelignite has joined #openttd
20:29:30 <LordAro> smacking apple upside the head for using a completely separate version scheme
20:29:48 <andythenorth> class action suit? :P
20:30:02 <andythenorth> it's the only thing that gets Apple's attention :P
20:31:27 <LordAro> could hardcode a "if apple, version number is actually this" set of rules
20:31:42 <LordAro> or could finish TB's cmake build system
20:32:02 <andythenorth> for now, I have a workaround
20:32:20 <andythenorth> it builds, we have a probably cause
20:34:06 <nielsm> okay doesn't reproduce in this build at least
20:34:37 <frosch123> LordAro: is it known that cmake fixed it, or is that only hopeful thinking?
20:37:41 <LordAro> frosch123: hopeful, but expected, iirc
20:42:55 <nielsm> okay doesn't reproduce in my vanilla 1.8.0 install either
20:46:28 <frosch123> nielsm: there are tons of Random calls in the ai
20:47:36 <nielsm> but aren't those supposed to be deterministic too?
20:47:55 <nielsm> or are they Actually Random?
20:48:17 <nielsm> (since AI only runs on the host in multiplayer, and not every client, afaik)
20:48:18 <frosch123> i think ai random seed it not store
20:48:27 <frosch123> iirc ottd only has two random seeds
20:48:35 <frosch123> one for gameplay, which ai must not used
20:48:41 <frosch123> and a second one for client stuff
20:52:07 <nielsm> late, haven't had much food yet, need to run out
20:53:04 *** sim-al2 has joined #openttd
20:59:37 <Eddi|zuHause> i just made some freezer food in the oven
21:00:18 <frosch123> do you also cook food in the fridge?
21:01:44 <andythenorth> I cook the oven in the fridge
21:01:50 <andythenorth> and freezer the fridge in the food
21:02:24 <Eddi|zuHause> or i fridge food in the cooker?
21:02:34 <Samu> doesn't seem to work, I am randomizing PASS and MAIL choice
21:05:18 <Eddi|zuHause> oha, it's no longer "public transport", it's now "Mobility as a Service"
21:06:53 <andythenorth> I need alberth as a service
21:07:02 <andythenorth> or at least someone else who will indulge my GS idea
21:08:23 <andythenorth> I could have called it 'credit mobilier'
21:08:39 <andythenorth> which is has echoes of mobility-as-a-service :P
21:09:01 <Samu> it always picks PASS, nor MAIL, why!
21:12:28 <Eddi|zuHause> that name sounds familiar somehow
21:19:25 <Eddi|zuHause> andythenorth: using fixed regions based on map size might not be that useful of an idea, if that region doesn't actually contain a significant portion of land
21:20:59 <Eddi|zuHause> andythenorth: in theory you could go through the map adding up all the land tiles, and then choosing a point where x% of the landmass is to decide the origin area
21:22:49 <andythenorth> this is oriented to railroad construction
21:23:01 <andythenorth> I hadn't considered water map cases, or non-rail transport types :P
21:24:40 <Eddi|zuHause> "Can a cargo monitor detect origin of cargo on delivery?" <-- in theory the game could do that, but i doubt that GS currently has access to that kind of information
21:28:38 <andythenorth> the stuff at the bottom is older ideas
21:29:03 <andythenorth> I reduced it to 'connecting towns'
21:30:12 <Eddi|zuHause> so essentially an extension of the "subsidies" of the original game
21:30:45 <Eddi|zuHause> "establish a connection of <cargo> between <town> and <town>"
21:31:46 <Eddi|zuHause> the existing code to check for subsidy completion could be used to trigger a response from the GS to fulfill a goal?
21:45:34 <andythenorth> yes I wondered about that
21:45:49 <andythenorth> it's perhaps just a more structured way to generate subsidies?
21:45:55 <andythenorth> or at least that might be worth exploring
21:46:15 <andythenorth> so the subsidies are for a deliberate graph, in sequence
22:13:42 <Eddi|zuHause> yeah, something like that
22:18:13 <andythenorth> is there a name for a graph that has a specific layout on a grid?
22:18:27 <andythenorth> most graphs I've encountered are not anchored to a plane
22:19:33 <andythenorth> I am way out of my depth
22:19:49 <andythenorth> but I need something like, given an array of town x,y pairs
22:20:00 <andythenorth> generate a weighting of nodes that implies an order
22:20:09 <andythenorth> with some nodes equally weighted to allow randomness
22:20:44 <andythenorth> then x% of nodes at each weight must be connected before unlocking the next tier
22:21:39 <Eddi|zuHause> well, you already chose an origin, so you can calculate a distance to that origin
22:22:02 <Eddi|zuHause> and then divide a range of distances into groups
22:22:31 <Eddi|zuHause> you don't even need a graph for that
22:23:31 <andythenorth> so just bands of the map
22:23:54 <Eddi|zuHause> more like x% of the map
22:23:55 <andythenorth> so it's a side note that it's a graph, not the implementation
22:24:26 <Eddi|zuHause> x% of the distance between the chosen origin and final destination
22:24:50 <Eddi|zuHause> where the method of choosing those is a separate issue
23:04:26 <Samu> Wormnest: can you help me ?
23:04:48 <Samu> CONFIG_BOOLEAN in AIInfo doesn't work for me, or I can't manage to make it work
23:07:40 <Samu> Users won't explicitly set the value, it'll get a random value between min_value and max_value.
23:07:51 <Samu> i can't manage to make it work :(
23:10:23 <Wormnest> Check how other ai´s use it
23:11:17 <Samu> do you know any ai using it? :(
23:20:29 <LordAro> well that's not an issue
23:29:06 <Samu> Wormnest: can't find any ai that uses it
23:29:20 <Samu> doing the search alphabetically
23:29:33 *** andythenorth has joined #openttd
23:35:14 <Wormnest> Samu: idk where you are looking but at least WormAI uses it
23:37:11 <Samu> oh crap, I meant to say CONFIG_RANDOM
23:40:12 <Samu> CONFIG_RANDOM is what doen't work for me, sorry about that
23:48:12 <Samu> I looked through them all
23:48:25 <Samu> no AI uses CONFIG_RANDOM flag
23:48:38 <Samu> I wonder if this feature actually works
23:50:49 *** rocky113844 has joined #openttd
continue to next day ⏵
