Back to Home / #openttd / 2013 / 08 / Prev Day | Next Day
#openttd IRC Logs for 2013-08-21

---Logopened Wed Aug 21 00:00:13 2013
00:28<@Rubidium>Rieksts: the main problem with cargodest is the enormous amount of resources that must be spent on routing all the cargo
00:29<@Rubidium>it simply made games unplayably slow much earlier than cargodist would
01:13-!-Prof_Frink [] has quit [Remote host closed the connection]
01:23-!-Pecio [] has joined #openttd
01:31-!-Eddi|zuHause [] has joined #openttd
01:39-!-Devroush [] has joined #openttd
02:01-!-DDR [] has quit [Quit: DDR is not Dance Dance Revolution.]
02:08<Supercheese>I take it if I add a new advanced setting, I have to bump the savegame version?
02:17<Xaroth|Work>depends if that setting needs to be saved with the savegame, no?
02:19-!-V453000 [] has quit [Ping timeout: 480 seconds]
02:19-!-V453000 [] has joined #openttd
02:24<Supercheese>how would I determine that?
02:25<Supercheese>I'm thinking it would, but I'm not sure
02:25<maddy_>does it change the game state or mechanics in a meaningful way, or does it just change something like the UI behavior?
02:26<@planetmaker>Supercheese, savegame relevant is a setting when it influences the game, like map or any game element behaviour
02:27<@planetmaker>non-savegame are per-user settings like language, signal distance, sound, display units...
02:27<Supercheese>Well, you know which patch I'm writing
02:27<Supercheese>station-naming related stuff, stored in savegame or not?
02:27<@planetmaker>that's savegame relevant
02:28<@planetmaker>as it has to be the same for all players
02:28<Supercheese>I have to dive into saveload code then?
02:28<@planetmaker>For a simple setting: not really. Just add it. And increase savegame version by one
02:28<@Rubidium>no, you can also wade into the saveload code; much less messy ;)
02:29<@planetmaker>lol, Rubidium
02:31-!-tokai|mdlx [] has joined #openttd
02:33-!-Midnightmyth [] has joined #openttd
02:37-!-tokai|noir [] has quit [Ping timeout: 480 seconds]
02:49<Supercheese>Broken Savegame - Referencing invalid persistent storage
02:51<maddy_>I just coded saveload yesterday, only I added new chunks, instead of editing existing ones
02:52<Supercheese>I'm completely new to this
02:53<Supercheese>I added one new setting, bumped the savegame version
02:53<Supercheese>evidently I need to do something else
02:53-!-JVassie [] has joined #openttd
02:55<Supercheese>or perhaps this typo was causing issues...
03:00<Supercheese>nope, is apparently something else
03:18<maddy_>paste it to somewhere so others can see the code
03:19<Supercheese>that's the thing, I haven't modified anything else in saveload other than the savegame version
03:22<Supercheese>time for bed; I'll figure this out later
03:22-!-Supercheese [~Superchee@] has quit [Quit: ChatZilla [Firefox 23.0.1/20130814063812]]
03:29-!-JVassie [] has quit [Ping timeout: 480 seconds]
03:34-!-Pensacola [] has joined #openttd
03:46-!-wakou2 [] has joined #openttd
03:51-!-sla_ro|master [] has joined #openttd
04:02<roboboy>grr. Not happy with the telco
04:07<@planetmaker>telstra? :D
04:12-!-valhallasw [] has joined #openttd
04:14<roboboy>well yes
04:15<roboboy>I was with their ex partner and want to switch to Aldi which keeps getting rejected
04:15<roboboy>ie Kogan
04:16<roboboy>it's all a mess
04:22<roboboy>how'd you know it would be telstra planetmaker?
04:34<@planetmaker>default assumption for AU :-)
04:35<@planetmaker>got some experience with telstraclear some years ago when it was still their subsidy...
04:35<@planetmaker>can't say I was particularily impressed
04:36<roboboy>they've just disconnected an entire telco that was using their network and I was with them
04:36<roboboy>Iv'e heard lot's of people not have good customer experience with Telstra but they have the best mobile coverage in the country
04:46-!-Alice3 [] has joined #openttd
05:06-!-Dark-Ace-Z [~BillyMays@] has quit [Ping timeout: 480 seconds]
05:37-!-George|2 [~George@] has joined #openttd
05:37-!-George is now known as Guest4013
05:37-!-George|2 is now known as George
05:42<maddy_>planetmaker: did you have some ideas for how to represent links between signals visually? I was thinking just something simple like drawing lines between the signals when the programmin GUI is open
05:43<maddy_>or, optionally, just highlighting the corresponding tiles, but that is not so good
05:43<Xaroth|Work>any of you lot happen to have a clue how to best do master-master replication on a set of postgres databases?
05:43-!-Guest4013 [~George@] has quit [Ping timeout: 480 seconds]
05:43<dihedral>Xaroth, i was playing around with that at work, with actually no good result
05:44<@peter1138>i've only just done master-slave :S
05:44<dihedral>peter1138, i would love to know more about that, that's something i could do with
05:44<@peter1138>documentation for that is pretty lacking
05:44<Pinkbeast>maddy_: Is there a limit on link length?
05:44<Xaroth|Work>crap :|
05:45<@peter1138>but master-slave is at least built in to postgresq now
05:45<maddy_>Pinkbeast: no
05:45<@peter1138>master-master will need external tools
05:45<dihedral>Xaroth, it will probably be cheaper getting some external guru in to do the job before you waste hours (as i did) just hunting and trying and failing
05:45<Pinkbeast>maddy_: In that case I think you have to consider how it will look in the perverse case where someone is doing programming in them without regard to physical proximity, which they can and will.
05:45<@peter1138>dunno how up to date that is
05:45<dihedral>i know of pgpool2, but then you also must connect to pgpool2 instaed of connecting to the pg database directly
05:45<@peter1138>Postgres-XC (eXtensible Cluster) is a multi-master write-scalable PostgreSQL cluster based on shared-nothing architecture
05:45<Xaroth|Work>aye, was looking at that as well
05:46-!-oskari89 [] has joined #openttd
05:46<maddy_>Pinkbeast: yes, in which case a simple line between the signals would still work
05:46-!-George is now known as Guest4015
05:46-!-George [~George@] has joined #openttd
05:46<@peter1138>PostgreSQL core modifications required: Yes
05:47<@peter1138>well that's a bummer
05:47<Pinkbeast>Yeees, except maybe in the middle of the map where 8000 lines cross, or the cases where you have no chance to see that signal A is connected to A' because 600 other lines go the same way...
05:48<@peter1138>there's pgpool
05:48<maddy_>Pinkbeast: well, yeah, but those cases are hard to represent visually in any case I think...I am thinking what would work for the average use case (which they are intended for): that they are used somewhat sparingly
05:48<Xaroth|Work>well for this it's not -that- bad, as we're setting up a new db system
05:48<Xaroth|Work>however, I can't find if -XC is stable and production-ready or not :|
05:48<@peter1138>it's a pain in the arse though
05:49<@peter1138>security update wise and all that
05:51-!-Guest4015 [~George@] has quit [Read error: Operation timed out]
05:51<Xaroth|Work>I don't think symmetricds allows for true master-master setups though
05:51<@peter1138>yeah, might be crippled, it's yet another one of those open source projects piggy backed off commercial software
05:52<Xaroth|Work>looks like it
05:53<@peter1138>dihedral, the master-slave configuration i've done so far is based off their transaction log shipping
05:54<@peter1138>however it isn't high availability or load balanced
05:54<@peter1138>so if the master goes down it's manual work to switch to the slave
05:54<@peter1138>but at least everything is there read to go
05:56-!-George [~George@] has quit [Ping timeout: 480 seconds]
05:56-!-montalvo [] has quit [Quit: Textual IRC Client:]
05:56<@peter1138>i do hope they tackle HA at some point
06:00-!-Dark-Ace-Z [~BillyMays@] has joined #openttd
06:04<@peter1138> .. hmm
06:05<@planetmaker>maddy_, yeah, maybe that works. Maybe you then even can borrow some code from cargodist's linkgraph
06:06<Xaroth|Work>but that doesn't allow read scaling
06:06<@peter1138>no, that's just HA, not load balancing
06:07<Xaroth|Work>I'd prefer to get both out of it :P
06:07<@peter1138>pgpool-ii seems to have active development
06:09-!-TheMask96 [] has quit [Ping timeout: 480 seconds]
06:10<maddy_>planetmaker: yeah I will try to look at to for example
06:10-!-TheMask96 [] has joined #openttd
06:12-!-flaa [~flaa@] has joined #openttd
06:14<@peter1138>Someone™ just needs to bang it all together
06:23-!-Pensacola [] has quit [Ping timeout: 480 seconds]
06:37-!-HerzogDeXtEr [] has joined #openttd
06:39<dihedral>Xaroth, peter1138 : either of you use something for 'version control' regarding structure and content?
06:41-!-tycoondemon [] has quit []
06:44<@peter1138>of postgresql? no
06:46<@peter1138>simplest thing i can think of is a pgdump of the structure, stored in git
06:46<dihedral>it would come quite handy, as our product is configured via db and I could that way get 'revisions' corresponding to releases of the software
06:46<dihedral>it's not just the structure
06:46<@peter1138>well you can do the same with data too
06:46<@peter1138>though i would keep initial data separate from structure
06:47<dihedral>makes sense
06:48<@peter1138>if you change the structure you'll always need some manual way to update that structure when you upgrade a live install
06:50-!-valhallasw [] has quit [Ping timeout: 480 seconds]
06:51<dihedral>that actually is already in place
06:51<dihedral>changing the structure during updates
06:52<@peter1138>bah, everything on gets 5 star reviews
06:53<@peter1138>sounds like "i bought it so it must be good" syndrome
06:53<@peter1138>i'm sure there's a name for that
06:54-!-wakou2 [] has quit [Quit: Konversation terminated!]
06:55<maddy_>delusion, denial?
06:56<@peter1138>need to replace a couple of old 120GB drives... dunno what to get :S
06:56<Pinkbeast>As in, which hard drive manufacturer?
06:56-!-Guest3949 [] has quit [Remote host closed the connection]
06:57-!-roboboy [] has quit [Ping timeout: 480 seconds]
06:57<Pinkbeast>We-employer use Seagate which I believe is the result of some sort of experience.
06:57<@peter1138>heh, 4TB for £125...
07:01-!-valhallasw [] has joined #openttd
07:07<@peter1138>not sure i trust a 4TB drive... all that data loss if it dies :p
07:07<@peter1138>i'm sure we said the same thing about 4GB drives :D
07:07-!-oskari89 [] has quit []
07:08<fjb>We did.
07:08<fjb>Use backups.
07:09<@peter1138>and what is a suitable backup media for up to 4TB of storage?
07:09<fjb>A 4TB storage?
07:09-!-phatmatt [] has joined #openttd
07:09<phatmatt>here's a code question out of nowhere: what's the easiest way to go from a Trackdir to a Direction?
07:10<phatmatt>i can see Trackdir -> Track, but not Trackdir -> Direction conversion helpers
07:11-!-frodus [~frodus_de@] has joined #openttd
07:11<@planetmaker>TrackdirToExitdir maybe?
07:11<@peter1138>quicker than me :(
07:12<phatmatt>isn't that outputting a DiagDirection?
07:12<fjb>peter1138: You were lost on your storage.
07:14<@peter1138>DiagDirToDir(TrackdirToExitdir(...)) maybe
07:15<phatmatt>well, that loses information for non-diagonal trackdirs
07:15<phatmatt>i guess i have to write my own
07:15<@peter1138>converting trackdir to direction will always do that
07:16<@peter1138>hm, i see what you want
07:17-!-Pereba [] has joined #openttd
07:19<phatmatt>i guess it doesn't exist because i'm meant to be just using diagdirs, and toggling the trackdir's lowest bit to switch non-diagonal track sides
07:19<phatmatt>as i'm moving across the map
07:19<phatmatt>but i do actually need just the dir here, oh well
07:22<maddy_>sounds like you need to write a small utility function yourself for that purpose
07:23<phatmatt>guess so
07:24<phatmatt>i was just suprised there wasn't one already, since every other conversion seems to be covered ;)
07:26<@peter1138>i fancy a giant marblemadness style controller
07:26<@peter1138>and then scroll around the map using that :D
07:26-!-valhalla1w [] has joined #openttd
07:27-!-valhallasw [] has quit [Read error: Operation timed out]
07:34-!-Pensacola [] has joined #openttd
07:37-!-DanMacK [] has joined #openttd
07:37<DanMacK>Hey all
07:48<Malinux->will there be diagnoal briges any time?
07:48<DanMacK>When you code them
07:48<phatmatt>funny you ask.
07:48<phatmatt>guess what i'm coding right now
07:49<Pinkbeast>Non-linear tunnels?
07:49<DanMacK>is that possible w/the current engine?
07:49-!-tycoondemon [] has joined #openttd
07:49<phatmatt>DanMacK: i don't see why not. i'm just experimenting right now
07:49<MINM>Sure is. as long as you spend a bit of time implementing a 'tunnel' layer/dimension
07:50<phatmatt>oh, i was talking about just diag tunnels/bridges
07:50<phatmatt>not non-linear. sorry ;)
07:50<MINM>because you'd need a way to save these tunnel coordinates.
07:50*DanMacK is still waiting for Custom Bridgeheads in trunk
07:50<Pinkbeast>When I looked at it, it didn't seem like it would need a true-3D layer, just a bit of cleverness to know which tunnels fouled which areas of the map.
07:51<phatmatt>it would be easy if the train actually disappeared while in the tunnel, but you can still see its position by following it of course
07:52<phatmatt>i guess you could fake the track layout in the tunnel
07:52<MINM>for which, Pinkbeast? non-linear or diagonal?
07:52<phatmatt>would be hard if the terrain was weird though
07:52<MINM>because the former would seem like it needs a 3d layer to me
07:55<phatmatt>it wouldn't *need* one, but it would be very icky without one.
07:56<phatmatt>and on the flipside, it would be pretty "easy" with one
07:56<Pinkbeast>It certainly doesn't need one. Fundamentally, now, when you build a tunnel, the game checks it doesn't intersect any existing tunnels or poke out of the ground. That would still need to be done.
07:58<Pinkbeast>And yes, when a vehicle's in a tunnel, its position and speed are determined from knowing where it entered the ground and the metadata about the wormhole, it's just at the moment there isn't very much metadata
08:01<Malinux->DanMacK: yeah.. I am one of those who complain and wish things, but dosen't code myself... :p
08:01<Pinkbeast>It's all harder for bridges because you'd have to draw them </captain obvious>
08:03<phatmatt>Pinkbeast: perhaps if i add diag bridges, and force a caged artist to draw some art for them, that part will be easier :P
08:04<phatmatt>first compile after first pass modifying stuff... let's see how it goes
08:06<Malinux->I also wonder about singals on bridges and in tunnels. I think this is very challenging to do as it hasen't been done yet :)
08:07<maddy_>isn't there a patch for it?
08:07<Pinkbeast>Malinux-: It has (but rarely well)
08:07<Pinkbeast>... I didn't even mention forking tunnels, did I? :-)
08:10<phatmatt>stations on brdiges
08:10<phatmatt>and, of course, in tunnels.
08:10<MINM>there is/was a patch for it, yes, but it's hacky
08:11<MINM>signals on bridges/tunnels I mean
08:17-!-frodus [~frodus_de@] has quit [Ping timeout: 480 seconds]
08:18<Malinux->maddy_: I have seen a kind of patch for singals in tunnels
08:21-!-TrueBrain [] has quit [Quit: reboot]
08:29-!-TrueBrain [] has joined #openttd
08:30-!-Midnightmyth [] has quit [Ping timeout: 480 seconds]
08:42<phatmatt>compiles and runs. place bets on how quickly it crashes
08:43*DanMacK loves having a canal system
08:43<phatmatt>MINM: never mind. i was talking about my diagonal bridge/tunnels changes
08:44-!-Elukka [] has joined #openttd
08:57-!-ST2 [] has joined #openttd
09:02-!-xT2 [] has quit [Ping timeout: 480 seconds]
09:11-!-Pecio [] has left #openttd []
09:12<fjb>Moin Sir Belugas
09:15<fjb>Moin __ln__
09:18<__ln__>guten moin fjb
09:19-!-oskari89 [] has joined #openttd
09:24-!-Midnightmyth [] has joined #openttd
09:29<dihedral>peter1138, regarding ssd i am personally a fan of samsung
09:29<dihedral>esp. when it gets to natively encrypted ssd's
09:29<dihedral>enter the wrong password 10 times, ssd gets trimmed :-P
09:29<dihedral>and i used to purchase things from
09:31<@peter1138>i'm not buying an ssd
09:31<@peter1138>i've already got one :p
09:31<@peter1138>but my bios is too old to understand hdd passwords or something
09:50-!-t1488t [~id@] has joined #openttd
09:55<juzza1>do those ssds with native encryption have some kind of bootloader where you input the password?
10:03-!-mindlesstux [] has joined #openttd
10:09-!-ntoskrnl [] has joined #openttd
10:15<@peter1138>no, it's meant to be handled by the BIOS
10:16<@peter1138>but i've never seen a BIOS with the option, but that may just be because it's only visible with the feature available on the drive
10:16<@peter1138>the encryption is done in hardware on the drive, which is handy
10:17<@peter1138>it's basically "free" encryption, though not usuable if you need to do things like key rotation
10:21<@peter1138>i guess it's supported by more modern computers
10:22-!-pugi [] has joined #openttd
10:22<juzza1>modern cpus already support aes through hw, so it doesnt seem that useful to me
10:23<@peter1138>it still requires the cpu to do it
10:24<@peter1138>if you were doing some heavy encryption load, that's aes cycles taken away from whatever you needed to do
11:07-!-perk11 [] has quit [Ping timeout: 480 seconds]
11:24-!-alluke [] has joined #openttd
11:28-!-TheMask96 [] has quit [Ping timeout: 480 seconds]
11:30-!-TheMask96 [] has joined #openttd
11:42-!-LordAro [] has joined #openttd
11:43-!-Prof_Frink [] has joined #openttd
11:46-!-tycoondemon [] has quit []
12:05-!-valhalla1w [] has quit [Ping timeout: 480 seconds]
12:09-!-pugi [] has quit []
12:13-!-namad7 [] has quit []
12:16-!-namad7 [] has joined #openttd
12:20-!-Progman [] has joined #openttd
12:20-!-tycoondemon [] has joined #openttd
12:21<LordAro>hai Terkhen
12:28-!-valhallasw [] has joined #openttd
12:32<DanMacK>oops :P
12:35<Pinkbeast>DanMacK: Not _another_ A1 Class, my brain hurts. :-)
12:36<DanMacK>that's the first one :P
12:37<Pinkbeast>Fair point. I have just been having the "no, Tornado is a Peppercorn A1, that's actually after the Gresley A4s" conversation quite a lot recently :-)
12:38<DanMacK>And yes, it is a Peppercorn A1
12:39<DanMacK>You're British I assume?
12:39<Pinkbeast>Yes, and went to York to see the six surviving A4s so had the conversation with non-enthusiasts quite a lot.
12:41<DanMacK>The new construction there blows my mind
12:41-!-valhalla1w [] has joined #openttd
12:41<DanMacK>I wish they'd do it in North America
12:42<Pinkbeast>DanMacK: Well, it's not new construction, but I read UP have bought and are restoring a Big Boy, and there's those mad Project 130 types too...
12:42<DanMacK>The Big Boy project's been stalled for years AFAIK
12:42<Pinkbeast>If I had a squillion quid I'd be building the Wardale 5AT.
12:43<Pinkbeast>DanMacK: No, quite recently, it was in _Steam Railway_ this month
12:43-!-valhallasw [] has quit [Ping timeout: 480 seconds]
12:43<Pinkbeast>,0,2255582.story - they just bought one. (I think there might be another Big Boy restoration going nowhere?)
12:43<DanMacK>Well shit...
12:44<DanMacK>If anyone can do it, it's UP
12:44<Pinkbeast>DanMacK: Certainly the situation where an ordinary mainline operator restores a locomotive is unknown here (pity)
12:45<DanMacK>but the reconstruction of a new locomotive or reconstruction of a long abandoned railway is pretty unheard of in North America
12:45<Pinkbeast>Project 130 seem a bit confused about what they're doing, but if they can have a decent crack at the speed record that'll be interesting.
12:46<DanMacK>The Big Boy thing is cool
12:47<DanMacK>Love to see an new CPR F2 4-4-4 built here
12:48<Pinkbeast>DanMacK: I have no real idea how many classes the USA is "missing", in the sense the Peppercorn A1s were before Tornado.
12:48<DanMacK>well, there are tons
12:48<DanMacK>because the US had so many different railroads
12:49<Pinkbeast>DanMacK: Well, so did we before the "Big Four", but I think most of the gaps are being closed.
12:49<Pinkbeast>Depends how far back you go, but they're (say) building a GWR "Saint" which is the main hole there.
12:50<DanMacK>Which they're doing
12:50<Pinkbeast>How far back: for example to my knowledge Firefly is the only running broad gauge locomotive (a replica) and you can travel perhaps 50 feet behind it
12:50<DanMacK>well... sort of
12:54<Pinkbeast>DanMacK: I don't know why the US doesn't have more new-build, you're not exactly short of rich eccentrics
12:54<DanMacK>Well, I'm Canadian, but yeah
12:55<Pinkbeast>DanMacK: *shame*
12:55<Pinkbeast>DanMacK: I am terribly sorry.
12:55*Pinkbeast did live with a Canadian for eight years, he really ought to think of that.
13:00<Rieksts>Does cargodist recalculate trough what route to send cargo based on previous cargo delivery times?
13:01-!-frosch123 [] has joined #openttd
13:01<Pinkbeast>Rieksts: Not as I understand it - just percieved link capacity.
13:01<Rieksts>I did some reconstruction on one of mainlines, had to stop many trains and some strange cargo started transfering trough very inefficient routes
13:01<Rieksts>nem the *strange
13:02<Pinkbeast>That would be consistent with a chance in measured capacity.
13:02<Rieksts>ahhh I meant cargodest. Wtf, guess my brain is not working properly atm, time for a pause :D
13:03<Pinkbeast>Rieksts: I don't know what cargodest does, sorry
13:04<Rieksts>Gold just piling up in station far away from its destination.
13:05-!-andythenorth [~Andy@] has joined #openttd
13:05<Rieksts>ok i guess I am on my own, if I wont figure this out I will submit my save to cargodest thread
13:05<andythenorth>DanMacK: I think we can consolidate open and flat
13:05<andythenorth>also van and fridge
13:09<LordAro>cargodist* there is a difference
13:09<Pinkbeast>LordAro: Er, surely that is why Rieksts specifically corrected themselves to say "cargodest".
13:10<LordAro>oh right. :p
13:11-!-glx [] has joined #openttd
13:11-!-mode/#openttd [+v glx] by ChanServ
13:12-!-Progman [] has quit [Remote host closed the connection]
13:30-!-andythenorth [~Andy@] has left #openttd []
13:40-!-JVassie [] has joined #openttd
13:45<DanMacK>Hey James
13:45<@DorpsGek>Commit by translators :: r25739 /trunk/src/lang (3 files) (2013-08-21 17:45:24 UTC)
13:45<@DorpsGek>-Update from WebTranslator v3.0:
13:45<@DorpsGek>afrikaans - 218 changes by mulderpf
13:45<@DorpsGek>simplified_chinese - 1 changes by siu238X
13:45<@DorpsGek>traditional_chinese - 1 changes by siu238X
13:47-!-t1488t [~id@] has quit [Read error: Operation timed out]
13:49<maddy_>my patch has lots of things left to do, but I am going to just release the first version
13:51-!-Alberth [] has joined #openttd
13:51-!-mode/#openttd [+o Alberth] by ChanServ
13:52<@planetmaker>hi Alberth
14:03<maddy_>oh my, my forum post was not approved, any of you a mod at tt-forums who can approve it?
14:04<JVassie>ill take al ook ;)
14:09-!-Pensacola [] has quit [Remote host closed the connection]
14:11-!-Hendrick [~Hendrick@] has joined #openttd
14:20<maddy_>cool it's published, thanks, hope you guys go check it out when you have time
14:23<Eddi|zuHause>i should do something either useful or fun, but i get immediately bored when i start anything... what's wrong with me?
14:24<LordAro>depression? :L
14:24<LordAro>seriously, i've been somewhat similar lately
14:24<LordAro>i think perhaps the key is to stick with it, and keep going
14:24<JVassie>o/ Eddi|zuHause long time no speak :)
14:25-!-Wolf01 [] has joined #openttd
14:26<Wolf01>hi o/
14:31<@Alberth>hi hi
14:33<@Alberth>Eddi|zuHause: in my experience fun comes when you get in close contact with the problems, and try to find your way around them. From a distance it does not look like a lot of fun to solve stuff
14:45-!-HerzogDeXtEr1 [] has joined #openttd
14:48-!-Jomann [] has quit [Ping timeout: 480 seconds]
14:51-!-HerzogDeXtEr [] has quit [Ping timeout: 480 seconds]
14:56-!-Jomann [] has joined #openttd
14:57-!-tokai|noir [] has joined #openttd
14:57-!-mode/#openttd [+v tokai|noir] by ChanServ
15:01-!-tokai|mdlx [] has quit [Ping timeout: 480 seconds]
15:02-!-DanMacK [] has left #openttd []
15:11-!-Rieksts [] has quit [Quit: ChatZilla [Firefox 23.0.1/20130814063812]]
15:19-!-smallfly [] has joined #openttd
15:22-!-pugi [] has joined #openttd
15:30-!-smallfly [] has quit []
15:34-!-mindlesstux [] has quit [Ping timeout: 480 seconds]
15:50-!-alluke [] has quit [Quit: Page closed]
16:03-!-mindlesstux [] has joined #openttd
16:08-!-Alberth [] has left #openttd []
16:11-!-ntoskrnl [] has quit [Ping timeout: 480 seconds]
16:14-!-George [~George@] has joined #openttd
16:17-!-Hendrick [~Hendrick@] has quit [Ping timeout: 480 seconds]
16:22<Eddi|zuHause>i have an idea: if you "force" a train through a path signal, you force to unreserve all paths that it crosses, and tell the trains that this path backtracks to to "wait for free path" instead
16:22<Eddi|zuHause>that way, the forced train will get a valid path
16:23<Eddi|zuHause>and the other trains won't crash it as long as they have enough distance to stop
16:23<@Rubidium>... and the other train will not have a path and continue running into the just reserved path and... takes the 90 degree corner to crash and burn
16:23<Eddi|zuHause>no, "wait for free path" would imply "stop immediately"
16:24<Eddi|zuHause>same deceleration as if you hit the stop button
16:25<@Rubidium>then it can still stop shy of the just newly reserved path and cause the same issue
16:25<Eddi|zuHause>that is then the player's fault
16:25<@Rubidium>and now to make things worse...
16:25<@Rubidium>... it isn't realistic to unreserve a path
16:26<@Rubidium>if that happens in the real world several minutes need to pass before the unreserving actually takes place
16:26<Eddi|zuHause>Rubidium: let's assume you had a near-miss and you stopped all trains in time
16:27<Eddi|zuHause>Rubidium: now you want to cancel all reservations and start the trains one-by-one to clear the crossing
16:27<@Rubidium>turn them around to cancel the reservation?
16:28<@Rubidium>but... in the first place, you shouldn't be working on a line that is in service!
16:28<Eddi|zuHause>Rubidium: but you can't turn around a broken down train
16:28<Eddi|zuHause>Rubidium: and this is totally not obvious to users
16:29-!-George [~George@] has quit [Read error: Connection reset by peer]
16:29<Eddi|zuHause>Rubidium: i'm just trying to make this a little more useful, currently there is no situation whatsoever where you can force a train through a path signal and not get a crash
16:30<Eddi|zuHause>the way i proposed is more like "ok, this train now gets priority"
16:30<Eddi|zuHause>and assuming all other trains are stopped, there should be no bad side effects
16:31<@Rubidium>well... I'm not going to implement or test it
16:32<Eddi|zuHause>(small detail: you should prevent unreserving a trackbit which has a vehicle on it. and if unreserving fails, the train should not start unless you force it a second time)
16:32<@Rubidium>and this whole thing is already complex enough
16:32<@Rubidium>and IIRC a path signal is only passed after forcing twice
16:32<Eddi|zuHause>yes. which is weird at best
16:33<Eddi|zuHause>i have no clue what the mechanics behind that even is
16:33<@Rubidium>me neither
16:33<@Rubidium>anyhow... the easier solution would be:
16:33<@Rubidium>* remove 90 degree turns
16:33<@Rubidium>* and remove forcing to pass signals at danger
16:33<@Rubidium>* and possibly... remove breakdowns
16:34<Eddi|zuHause>but how are you going to resolve deadlocks?
16:34<@Rubidium>* and if they're really whiny... remove crashes
16:35<@Rubidium>that'll definitely have a speed improvement bonus ;)
16:35<frosch123>Eddi|zuHause: "one force" means "unstuck train", "double force" mand "unstuck and ignore signal"
16:35<Eddi|zuHause>Rubidium: but the cause of the crash is that the path is not unreserved in the first place. not that following the path violates the 90° setting
16:35<frosch123>Eddi|zuHause: anyway, if rb argues that it takes several minutes irl for the line to clear, you should ask what daylength factor applies in ottd for that :p
16:39<JVassie>anyone else here using Zernebok hosting?
16:42<Eddi|zuHause>why would anyone in his right mind do that :p
16:43<Eddi|zuHause>(why would anyone here be in his right mind?)
16:44<JVassie>trying to work out if its blocked outgoing ports for SMTP
16:47<@Rubidium>just looked it up... if a signal is revoked the switches of the reserved path will remain locked into position for at least two minutes. This to allow the driver to perform an emergency stop (or to allow the computer in the train to do so). Only after those two minutes the switches can changed so another path can be taken.
16:47<@Rubidium>I propose to use 2 minutes IRL as "safety" margin then ;)
16:47<@Rubidium>(or little more than the time it takes for a crashed train to clear)
16:58-!-Supercheese [~Superchee@] has joined #openttd
17:00<Eddi|zuHause>but that would require us keeping track of "stopping distance"
17:00<Eddi|zuHause>which is not at all what i intended to say
17:01<Eddi|zuHause>what i meant: "if all trains are stopped and one is forced through, that force should take a safe state if possible"
17:08-!-frosch123 [] has quit [Quit: be yourself, except: if you have the opportunity to be a unicorn, then be a unicorn]
17:11-!-oskari89 [] has quit []
17:16-!-Dark-Ace-Z is now known as DarkAceZ
17:17-!-Progman [] has joined #openttd
17:28<LordAro>what's the best way to get the char* of a StringID ?
17:29-!-Progman [] has quit [Remote host closed the connection]
17:31-!-valhalla1w [] has quit [Read error: Operation timed out]
17:44-!-Wolf01 [] has quit [Quit: Once again the world is quick to bury me.]
17:48-!-sla_ro|master [] has quit []
17:52<Supercheese>Well, the assertion of, "X old tyrannical group would have been envious of Y new tyrannical group" will almost always be true, simply due to new technologies
18:07-!-LordAro [] has quit [Ping timeout: 480 seconds]
18:17-!-tokai|mdlx [] has joined #openttd
18:21-!-Alice3 [] has quit []
18:22-!-tokai|noir [] has quit [Ping timeout: 480 seconds]
18:52-!-flaa [~flaa@] has quit [Quit: leaving]
19:17-!-Midnightmyth [] has quit [Ping timeout: 480 seconds]
19:25-!-Devroush [] has quit []
19:27-!-JVassie [] has quit [Ping timeout: 480 seconds]
19:29-!-pugi [] has quit []
19:33<Eddi|zuHause>well, in dutch population surveys they asked for religion, and when the nazis invaded they said "hey, cool, now we know who to deport"
19:34-!-HerzogDeXtEr1 [] has quit [Read error: Connection reset by peer]
19:35-!-HerzogDeXtEr [] has joined #openttd
20:00-!-HerzogDeXtEr [] has quit [Read error: Connection reset by peer]
20:01-!-HerzogDeXtEr [] has joined #openttd
20:24-!-Pereba [] has quit [Quit: better late than never -]
20:58-!-DabuYu [DabuYu@] has joined #openttd
21:05-!-fjb is now known as Guest4114
21:05-!-fjb [] has joined #openttd
21:06-!-valhallasw [] has joined #openttd
21:09-!-Guest4114 [] has quit [Ping timeout: 480 seconds]
21:09-!-phatmatt [] has left #openttd []
21:16-!-DDR [] has joined #openttd
21:21-!-glx [] has quit [Quit: Bye]
21:23-!-montalvo [] has joined #openttd
21:47-!-Virtulis [~virtuall@] has quit [Ping timeout: 480 seconds]
21:56-!-JGR_ [] has joined #openttd
21:56-!-perk11 [] has joined #openttd
21:59-!-JGR [] has quit [Ping timeout: 480 seconds]
22:09-!-roboboy [] has joined #openttd
23:03-!-HerzogDeXtEr [] has quit [Read error: Connection reset by peer]
23:23-!-zeknurn [] has quit [Ping timeout: 480 seconds]
23:35-!-Virtulis [~virtuall@] has joined #openttd
23:53-!-George [~George@] has joined #openttd
23:56-!-George [~George@] has quit [Read error: Connection reset by peer]
---Logclosed Thu Aug 22 00:00:14 2013