Back to Home / #openttd / 2013 / 11 / Prev Day | Next Day
#openttd IRC Logs for 2013-11-24

---Logopened Sun Nov 24 00:00:47 2013
00:45-!-andythenorth [] has joined #openttd
00:45-!-andythenorth [] has quit []
00:56-!-Eddi|zuHause [] has quit []
00:56-!-Eddi|zuHause [] has joined #openttd
01:20-!-retro|cz [] has joined #openttd
01:34-!-Jomann [] has joined #openttd
01:39-!-sla_ro|master [slamaster@] has joined #openttd
01:40-!-andythenorth [] has joined #openttd
01:41-!-abchirk_ [] has quit [Ping timeout: 480 seconds]
01:45-!-Japa_ [~Japa@] has joined #openttd
01:53-!-Japa [~Japa@] has quit [Ping timeout: 480 seconds]
01:54<Supercheese>hey ho
02:06-!-LuHa [~LuHa@] has joined #openttd
02:20-!-andythenorth [] has quit [Quit: andythenorth]
02:28-!-DarkAce-Z [~BillyMays@] has joined #openttd
02:33-!-andythenorth [] has joined #openttd
02:34-!-DarkAceZ [~BillyMays@] has quit [Ping timeout: 480 seconds]
02:53*andythenorth wonders if there's a way to provide the 'game demands routes' aspect of YACD via a hack on subsidies
02:54<andythenorth>or some kind of latent routes
02:55-!-LuHa [~LuHa@] has quit [Quit: Leaving.]
02:55<andythenorth>like a shadow link graph of unconnected nodes, which some portion of cargo is assigned to
02:57-!-Supercheese [~Superchee@] has quit [Quit: ChatZilla [Firefox 25.0.1/20131112160018]]
02:57<Eddi|zuHause>yes. if you can find an efficient algorithm to find "all nodes"
03:02<andythenorth>that's super hard for pax / mail, and simpler for most primary / secondary cargos?
03:09-!-Pensacola [] has joined #openttd
03:20-!-andythenorth [] has quit [Quit: andythenorth]
03:27-!-Alberth [] has joined #openttd
03:27-!-mode/#openttd [+o Alberth] by ChanServ
03:34-!-adf88 [] has joined #openttd
03:46-!-Progman [] has joined #openttd
03:53-!-alluke [] has quit [Quit: Page closed]
04:10-!-andythenorth [] has joined #openttd
04:30-!-adf88 [] has quit [Read error: Connection reset by peer]
04:30<@DorpsGek>Commit by rubidium :: r26080 trunk/src/industry_gui.cpp (2013-11-24 09:30:30 UTC)
04:30<@DorpsGek>-Fix: possible out of bounds array access
04:33-!-adf88 [] has joined #openttd
04:35-!-Alice3 [] has joined #openttd
04:36-!-GriffinOneTwo_ [] has joined #openttd
04:37-!-GriffinOneTwo [~oftc-webi@] has quit [Quit: Page closed]
04:38-!-GriffinOneTwo_ is now known as GriffinOneTwo
04:50-!-LordAro [] has joined #openttd
04:52<@DorpsGek>Commit by rubidium :: r26081 trunk/src/company_base.h (2013-11-24 09:52:35 UTC)
04:52<@DorpsGek>-Fix [FS#5815]: memset with virtual functions doesn't quite work. Upon further review CompanyProperties doesn't need to be virtual in the current code base
04:59-!-Wolf01 [] has joined #openttd
05:04-!-adf88 [] has quit [Read error: Connection reset by peer]
05:04-!-wakou2 [] has joined #openttd
05:05-!-adf88 [] has joined #openttd
05:08-!-adf89 [] has joined #openttd
05:12-!-adf88 [] has quit [Read error: Operation timed out]
05:17-!-zydeco [] has joined #openttd
05:21-!-adf89 [] has quit [Quit: adf89]
05:23-!-adf88 [] has joined #openttd
05:39-!-kais58__7 [] has quit [Ping timeout: 480 seconds]
05:50-!-Midnightmyth [] has joined #openttd
05:51-!-andythenorth [] has quit [Quit: andythenorth]
05:59-!-montalvo [] has joined #openttd
06:06-!-retro|cz [] has quit [Ping timeout: 480 seconds]
06:12-!-Pensacola [] has quit [Remote host closed the connection]
06:13<juzza1> "Don't use the pink colours like the ones marked "WinAPI" in the Windows palette."
06:13<juzza1>why not?
06:16<scshunt|nospoilers>juzza1: my guess would be that they are loaded from Windows and subject to change
06:21<@planetmaker>yeah. from windows 3.1 ;-)
06:23<@planetmaker>juzza1, the API colours are not used by OpenTTD. OpenTTD converts every windows (legacy) palette to its internal default (DOS) palette
06:23<@planetmaker>the API colours will not be converted
06:24<@planetmaker>as for the pink colours in the Default palette: you could use them. But they'll all be pink. And there's no guarantee that their interpretation and way to be displayed will be kept
06:26<juzza1>they were almost a mystery for me, not so much anymore :)
06:26<@planetmaker>yeah, they were the colours MSDOS and windows used internally and which were not available to other stuff :-)
06:34-!-andythenorth [] has joined #openttd
06:35<@peter1138>planetmaker, MS DOS didn't...
06:38-!-skyem123 [] has joined #openttd
06:39<@planetmaker>well... why ever then the reserved colours might exist :-)
06:39<@planetmaker>maybe those were ms windows 3.1 and the so-called windows palette was for windows 95
06:40<@planetmaker>long ago :-)
06:41<@planetmaker>skyem123, what is a 4th rail railtype?
06:41<skyem123>what ae you talking about?
06:41<@peter1138>Maybe they're just not used.
06:42<@planetmaker>skyem123, do you know ?
06:42<@peter1138>You get free reign of the graphics palette under DOS. DOS didn't use graphics modes...
06:42<@planetmaker>talking about
06:43<skyem123>ah ok.
06:43<@peter1138>4th rail is, e.g. London Underground.
06:45<skyem123>I used fourth rail in my newGRF because i was thinking of the London Underground.
06:45<@planetmaker>ah, so that is not covered by foobar's treatise?
06:46<Qantourisc>Is there an easy way to buy up and then sell company assets ?
06:46<skyem123>4th rail with catenary doesn't exist.
06:47<@peter1138>it could exist with mixed traffic
06:47<@peter1138>dunno if it does though
06:48<skyem123>(in the rail scheme)
06:48<@peter1138>metropolitan line has mixed traffic, but it's mixed with diesels...
06:49<skyem123>and steam engines. (London Underground 150th Anniversary)
06:49<@planetmaker>Suggest a new character to indicate 4-rail with overhead wires?
06:49<@planetmaker>and add that to that wiki page, I think
06:49<@planetmaker>why not, yeah
06:49<skyem123>becuase 3rd rail + catenary is Z
06:50<@planetmaker>sounds plausible
06:50-!-Elukka [] has joined #openttd
06:51<skyem123>so i can just add it to the table?
06:52-!-KritiK [] has joined #openttd
06:52<@planetmaker>yes. It's a wiki, even when it's a spec wiki :-)
06:53<@planetmaker>E.g. george and andy also edit the cargo label pages when needed :-)
06:53*skyem123 updates wiki and his newGRF
06:54<skyem123>my newgrf will be the first to support it :-p
06:56<@planetmaker>I'm not exactly sure how many vehicle NewGRFs actually support that extensive railtype scheme yet. But in my eyes it makes sense and keeps it reasonably flexible, thus the choice for players
06:56-!-GriffinOneTwo [] has quit [Remote host closed the connection]
06:57<skyem123>will my newGRF be backwards compatible when i add an new railtype?
06:57<@planetmaker>vehicle newgrfs which want to support railtypes reasonably when they define a railtype translation table
06:57<@planetmaker>yes, if you add fallback railtypes
06:58<skyem123>i meant with itself.
06:58<@planetmaker>with itself... if you change the railtype of vehicles rather not
06:59<skyem123>i have 4th rail
06:59<@planetmaker>honestly, the backward compatibility of newgrfs with itself is overrated. Just mark it incompatible and don't worry
06:59<@planetmaker>it's safe and doesn't hurt in all usual situations
06:59<skyem123>and it will be disabled by defualt.
06:59-!-GriffinOneTwo [] has joined #openttd
06:59<skyem123>i have parameters that disable 4th rail. disabled by default
07:00<@planetmaker>well, backward compatibility of a newgrf with itself is only important for one case:
07:00<@planetmaker>you play a game which uses the old one. delete the old newgrf, download the new. Can you then continue the map with the new version of your newgrf without the map being borked?
07:01<@planetmaker>changed railtypes of vehicles would render vehicles broken
07:01<@planetmaker>changed meaning of parameters, too
07:01<skyem123>not removing. adding.
07:01<@planetmaker>undefining labels used as well.
07:01<@planetmaker>defining something additional: that's safe
07:02<@planetmaker>but doesn't it mean that you undefine the label you used before?
07:02<skyem123>no. i commented out 4th rail with catenary
07:02<@planetmaker>taking a game which enabled your new railtype by means of that parameter? Or is that parameter also new in the new version?
07:02<@planetmaker>then it's safe
07:02-!-GriffinOneTwo [] has quit []
07:02-!-GriffinOneTwo [] has joined #openttd
07:03<@planetmaker>or at least backward compatible
07:04<skyem123>i think i know know why the grf crashed the servers
07:05<skyem123>the grf block was after the graphics block.
07:06-!-yorick [] has joined #openttd
07:11<@Alberth>sounds like a bug in nml
07:12<@planetmaker>yeah, it should catch that and remedy it
07:13<@planetmaker>skyem123, btw, the issue you opened at the nml tracker: it's not a grf which I can build; it's missing all graphics
07:13<zydeco>it seems the changes in r26077 prevent it from opening grfs on osx
07:13<@planetmaker>you surely use some source control and could just bundle the whole thing?
07:14<skyem123>i have a odd compiling environment.
07:15<skyem123>do you use windows or linux?
07:16<skyem123>or i could give you the gfx folder...
07:16<@planetmaker>what is odd about your environment?
07:16<skyem123>i have windows.
07:16<skyem123>gnuwin32 make
07:17<skyem123>and strange things so i can use echo and del in make.
07:17<@planetmaker>I'll figure that out
07:18<skyem123>i can just give you the files after preprocessing...
07:18<@Alberth>hmm, after 5 patches, you find out you forgot to add a new file to the first patch :p
07:18-!-frosch123 [] has joined #openttd
07:18<@planetmaker>hehe, Alberth :-)
07:18<@planetmaker>Alberth, hg amend might be your friend :-) and subsequent hg evolve :-)
07:19<@Alberth>I have qpop and qpush friends :)
07:19<@planetmaker>well, will do, skyem123 :-)
07:20<skyem123>that should work!
07:22<@planetmaker>do you also have cutom_tags.txt? :D
07:23<skyem123>you get the new custom tags.
07:23<skyem123>doesn't make any difference
07:25<frosch123>btw. skyem123: did you really meant to put a mailto address into the bananas "website" entry?
07:25<skyem123>i have no website.
07:26<skyem123>did i do somthing wrong?
07:26<frosch123>no, but you are the first one to do so :)
07:26<frosch123>usually people put there a link to a forum thread or something
07:26<skyem123>i don't have a fourm thread.
07:27<@planetmaker>I believe you should create a public repository for your newgrf ;-)
07:29<@planetmaker>:-) Have you ever worked with any kind of version control software, like mercurial, subversion or git?
07:29<skyem123>i use git
07:30<@planetmaker>so I assume you know github :)
07:30<@planetmaker>alternatively we could also host it on the DevZone
07:32<skyem123>whitch one is better?
07:32<@planetmaker>I'm biased :-P
07:32<skyem123>to what?
07:33<@planetmaker>well DevZone:
07:33<andythenorth>does devzone also have git?
07:33<@planetmaker>andythenorth, yes. and no.
07:33<@planetmaker>repositories no not a problem
07:34<@planetmaker>The build service might. As it currently assumes mercurial repositories
07:34<@planetmaker>but that's not necessarily linked
07:34<skyem123>is the build system on linux?
07:35-!-Japa__ [~Japa@] has joined #openttd
07:35<skyem123>then it cant compile anyway.
07:35<@planetmaker>Makefiles can be adjusted, you know
07:36<skyem123>makefile.devzone makefile.skyem123?
07:36<skyem123>seperate files?
07:36<@planetmaker>simply by detecting system
07:37<skyem123>can it check environment varibles?
07:38<@planetmaker>^ see there
07:38<@planetmaker>yet I'm sure that Makefile will fail on mingw as it hasn't been tested for ages there
07:38<skyem123>that wont work on my build system
07:39<skyem123>i use mcpp
07:39<@planetmaker>don't underestimate how modular that makefile is ;-)
07:40<skyem123>how do i make a project?
07:41<@planetmaker>usual approach is to register with the DevZone and apply for a project. We can skip the last step and clear details here
07:42-!-Japa_ [~Japa@] has quit [Ping timeout: 480 seconds]
07:42<skyem123>i'm signed in to the devzone
07:43-!-retro|cz [] has joined #openttd
07:44<skyem123>so what do i do?
07:44<@planetmaker>I basically need two pieces of info: project name. And the 'project id' which basically would be the unique URL part / repository path
07:44<@planetmaker>and then I need the public part of your ssh key
07:45<skyem123>"skyem123's rails" skyeRail
07:45<@planetmaker>mind if I keep it to small letters, skyerail?
07:48<skyem123>ssh key...
07:49<skyem123>whats that?
07:50<@planetmaker>it's means to authenticate. Basically see on how to create it. Obviously ignore the upload part - but I'll need that public key they tell you there to upload
07:52<skyem123>should tortiose-git work?
07:52<@planetmaker>I assume so
07:52<@planetmaker>I've very little experience myself with both, git and tortoiseXXX
07:56*skyem123 reinstalls PuTTY
07:56<@planetmaker> @ skyem123
07:56<@planetmaker> alternatively for repo view
07:57<Rubidium>zydeco: ... but only when there is no internal path in the tar, right?
07:58<Rubidium>zydeco: that's why I didn't notice it as basically everything I got reachable for OpenTTD is from bananas
07:58<zydeco>it failed for me when loading opensfx from bananas
07:59<Rubidium>hmm... that doesn't use paths? Could be?!?
07:59<zydeco>and opengfx too when I removed it
08:00<zydeco>it would happen on anything that doesn't have a prefix
08:00<@DorpsGek>Commit by rubidium :: r26082 trunk/src/fileio.cpp (2013-11-24 13:00:06 UTC)
08:00<@DorpsGek>-Fix [FS#5816] (r26077): tar files with more than one file in the root directory would not be read properly (zydeco)
08:00<zydeco>unless it's the first header
08:00<Rubidium>zydeco: next time, if it's a bug and you have a patch, please mark it as a bug instead of patch
08:01<zydeco>oh ok
08:01<Rubidium>besides that, thanks ;)
08:01<zydeco>I wasn't sure about that
08:05-!-KopjeKoffie [] has quit [Quit: Leaving]
08:05<skyem123>planetmaker, how do i push to the repository?
08:07<@planetmaker>that needs the public key installed on our server first. Then git remote add origin ssh://; git push -u origin master
08:07<skyem123>how do i add the public hey?
08:08<@planetmaker>give it to me. via irc, via link to a paste service, via link to that file...
08:08<@planetmaker>whatever suits you
08:08<@planetmaker>sorry, that's not automated yet
08:09<skyem123>sent it to you in private chat.
08:09<@planetmaker>k, just saw that. My IRC window was too small to see that :D
08:11<@planetmaker>installed. Please try whether you're successful in pushing
08:12<skyem123>fatal: '/skye_rail' does not appear to be a git repository
08:12<skyem123>fatal: Could not read from remote repository.
08:13<@planetmaker>strange. The push seems to have been successful
08:14-!-kais58 [] has joined #openttd
08:14<@Alberth>~/hg ?
08:15<@planetmaker>that looks quite strange
08:15<@planetmaker>not ~hg/
08:15<@planetmaker>you scared me
08:15-!-Japa_ [~Japa@] has joined #openttd
08:16<@planetmaker>yeah, maybe gits wants its paths differently. It's the first git repo on our server now :-P
08:16<skyem123>i'm doing a lot of firsts!
08:17<@planetmaker>not a bad thing usually
08:17<skyem123>first to use 4th rail with catenary, first to use a mailto: address, first to use a git repo on
08:19<@planetmaker>I can imagine a lot of worse 'firsts' ;-)
08:21<frosch123>first to not put "grf { }" at the top of the nml file :p
08:21<@planetmaker>probably yeah :P
08:22<zydeco>oh so it doesn't have to say grf?
08:22-!-Japa__ [~Japa@] has quit [Ping timeout: 480 seconds]
08:22<skyem123>that still is a bug in NMLC
08:22<@planetmaker>could be a bug in the documentation, too ;-)
08:22<Qantourisc>how do you turn of ufo ?
08:22<@planetmaker>disable disasters, Qantourisc
08:30<skyem123>how does the openttdcoop build service work?
08:31<@planetmaker> @ skyem123
08:32<@planetmaker>basically: it needs a Makefile with the described targets. And setting up the project
08:32<@planetmaker>with the CF
08:33<skyem123>i need to set up my make file environment detection.
08:49-!-montalvo [] has quit [Quit: Textual IRC Client:]
08:54-!-mindlesstux [] has joined #openttd
08:54-!-GriffinOneTwo [] has quit [Quit: Page closed]
09:00-!-HerzogDeXtEr1 [] has quit [Read error: Connection reset by peer]
09:15<andythenorth>I have to do railtype support for Iron Horse :(
09:15<andythenorth>this does not fill me with enthusiasm
09:16<@planetmaker>andythenorth, that's easy... copy the railtype file I linked earlier and slightly adjust it
09:16<andythenorth>I'll have to learn the spec first
09:16<andythenorth>I never understand anything until I understand it :(
09:17<Xaroth|Work>isnt' that always the case?
09:18<@planetmaker>all you need to do is basically: think about a priority list of railtypes for your vehicles, down from the most specialised one (preferred) to RAIL/ELRL (least preferred)
09:19<andythenorth>ok, so it cascades, most specific preference first
09:20<@planetmaker>each preference-cascade gets its own name within your vehicle grf. That name is used as pseudo-railtype
09:20<@planetmaker>for your vehicles
09:20<@planetmaker>that itself is (and must be) meaningless outside your grf
09:22-!-montalvo [] has joined #openttd
09:23-!-TomyLobo [] has joined #openttd
09:24-!-valhallasw [] has joined #openttd
09:26<andythenorth>ho ho what fun
09:26<andythenorth>so many labels
09:27<andythenorth>but we had that conversation last time :)
09:28<andythenorth>ok, so the table contains mappings
09:28<andythenorth>in that case, how hard can this be? :P
09:28<@planetmaker>yup. railtype translation table
09:29<@planetmaker>not hard. tedious
09:29*andythenorth will now reveal poor understanding of logic
09:29<@DorpsGek>Commit by rubidium :: r26083 /trunk/src (station_gui.cpp station_gui.h) (2013-11-24 14:29:32 UTC)
09:29<@DorpsGek>-Codechange: pass the CommandContainer for the station joiner around using a reference instead of by-value
09:29<andythenorth>so take the case of metro trains
09:29<andythenorth>I want them to *only* use Metro track, if available
09:30<andythenorth>and fallback to ELRL if no metro
09:30<@planetmaker>METRO: [MTRO, ELRL]
09:30<andythenorth>(this may be a bad idea for confusing players, but let's assume not)
09:30<@planetmaker>and use railtype 'METRO' for that vehicle
09:30<LordAro>Rubidium: you've been busy this weekend :)
09:30<@planetmaker>and the first quoted line within the RTT
09:30<andythenorth>and if my game happens to contain both MTRO and ELRL, the vehicle will run on MTRO only, or both?
09:31<@planetmaker>it doesn't mean it'll run on both, MTRO and ELRL if available. Only the most preferential
09:31<andythenorth>I guess I'm asking about compatibility vs. exclusion
09:31<@planetmaker>compabibility is determined by the actual NewGRF defining the most-preferred, available railtype
09:31<andythenorth>so I can delegate that
09:31<@planetmaker>NML's RTT bascially resolves to a chain of action6
09:31<@planetmaker>and action7/9
09:32<@planetmaker>so the vehicle has only one railtype - depending on available railtypes. Compatibility is for the railtype to define
09:32<andythenorth>ok thanks
09:33<Rubidium>LordAro: yeah... holiday sprint season ;)
09:33<Rubidium>or rather... just not having done lots of coding for $JOB lately
09:34<@DorpsGek>Commit by rubidium :: r26084 trunk/src/newgrf_gui.cpp (2013-11-24 14:34:33 UTC)
09:34<@DorpsGek>-Fix: don't allow executing the palette toggling code when it's not editable
09:35<andythenorth>MTRO is not used by Metro
09:35<Qantourisc>how to prevent trains from reversing ?
09:36<Rubidium>and being mostly finished with moving house increases the amount of time that can be spent on frivolous things such as openttd
09:36<LordAro>frivolous? :O
09:36<Qantourisc>smaller maps ?
09:36<Qantourisc>(well routes)
09:37<Qantourisc>ok the train is trying to ride off the rails ..
09:38<Rubidium>Qantourisc: depends on the reason why they are reversing; could be a setting, or something broken in the network (signal, electrification)
09:38<Qantourisc>ow rails wasn't electrifiled
09:41<@DorpsGek>Commit by frosch :: r26085 /trunk/src (27 files) (2013-11-24 14:41:19 UTC)
09:41<@DorpsGek>-Codechange: Pass ResolverObjects as reference instead of pointer since they are never NULL.
09:46<@DorpsGek>Commit by rubidium :: r26086 /trunk/src (15 files in 3 dirs) (2013-11-24 14:46:26 UTC)
09:46<@DorpsGek>-Codechange: use AutoDeleteSmallVector instead std::list for dropdowns
09:46-!-robotboy [] has joined #openttd
09:49<LordAro>why the hate for the STL?
09:50<LordAro>or, what's the difference?
09:50<Qantourisc>stl:: ? or a differnt stl ?
09:50<frosch123>stl::vector != boost::ptr_vector
09:51<frosch123>yeah, something looked wrong :p
09:51<frosch123>but could not figure out out
09:51<LordAro>oh, is AutoDeleeteSmallVector an implementation of boost::ptr_vector?
09:51<frosch123>if you ignore the details
09:51<frosch123>then yes :p
09:52<LordAro>you could just use std::list<std::[shared|unique]_ptr<Type>> :p
09:53-!-roboboy [] has quit [Ping timeout: 480 seconds]
09:53<frosch123>yeah, and decrease the max train count to 50
09:53<@planetmaker>:D so much bloat there?
09:53<LordAro>slight difference in outcome ;)
09:54<LordAro>planetmaker: ?
09:55<frosch123>well, unique_ptr could actually work
09:55<LordAro>but no c++11 allowed yet, i guess :L
09:58<LordAro>despite there being a 'hidden' static_assert() in stdafx.h ;)
10:00-!-DarkAce-Z is now known as DarkAceZ
10:02<@DorpsGek>Commit by rubidium :: r26087 trunk/src/fileio.cpp (2013-11-24 15:02:05 UTC)
10:02<@DorpsGek>-Fix: file descriptor leak when finding a corrupt tar
10:02<andythenorth>implementing electro-diesel engine (bi-mode)
10:04<Rubidium>those would be interesting on some routes in the NLs
10:04<Rubidium>and it shouldn't be that hard; it's mostly placing a panto on an already diesel-electric train ;)
10:04<andythenorth>so set the railtype to RAIL
10:05<andythenorth>but use cb36 to check for ELRL
10:05<andythenorth>is current railtype of the depot available in buy menu? :P
10:06<frosch123>no, it isn'T
10:06<andythenorth>power varies according to railtype
10:06<andythenorth>I'll work around it
10:06<frosch123>put it in the description
10:06<frosch123>don't surprise the player
10:07<@peter1138>Yup, specify both.
10:13<Qantourisc>are there "better" maglev textures ?
10:14<Qantourisc>opengfx donwload from ingame ?
10:14<frosch123>there are specific maglev track sets
10:14<frosch123>just search for maglev
10:17<@DorpsGek>Commit by rubidium :: r26088 /trunk/src (rail_gui.cpp widgets/dropdown_type.h) (2013-11-24 15:17:53 UTC)
10:18<@DorpsGek>-Fix (r26086): Windows compilation
10:21<@DorpsGek>Commit by rubidium :: r26089 /trunk/src (fileio.cpp openttd.cpp) (2013-11-24 15:21:38 UTC)
10:21<@DorpsGek>-Fix: a number of minor memory leaks
10:21<Qantourisc>Would this change work inj the "background-map" ?
10:25<@DorpsGek>Commit by rubidium :: r26090 trunk/src/game/game_text.cpp (2013-11-24 15:25:41 UTC)
10:25<@DorpsGek>-Fix: leaking of file descriptor when reading strings for game texts
10:28<@Alberth>Qantourisc: you mean in the game playing on startup? no
10:30<Qantourisc>smolly zbase is ungly :)
10:31<Qantourisc>If I can get 16 other blender peaple, or a lot of time, i'd start on it :)
10:31-!-Superuser [] has joined #openttd
10:31<@planetmaker>Qantourisc, it doesn't need 16 more people. It needs to acually *start*
10:31<@planetmaker>Every journey starts with a first step
10:31<Qantourisc>ow hmmm
10:31<@planetmaker>And not with trying to gather a whole climbing party
10:32<Qantourisc>well ... imo, for 32 bit you either need good 2D drawing artists, or peaple with a 3D tool to render it ...
10:32<@planetmaker>yes. Sure. For instance blender like you said.
10:33<@planetmaker>But if you're one of the blender people you could already make a start
10:33<@planetmaker>Or rather the 2nd step. zBase is a first
10:33<Qantourisc>howmany sprites we talking about btw ? :)
10:33<@planetmaker>a whole base set is around 10k
10:34<@planetmaker>That's a lot. But as said: every start is small
10:34<Qantourisc>10k ?
10:34<@planetmaker>one can start with a house. Or with a certain ground type
10:34<Qantourisc>howmany of those are "duplicate" ?
10:35<Qantourisc>hmmm not a lot it seems :p
10:35<@planetmaker>what's your definition of 'duplicate'? Each looks different. But many are ground tiles, different rails and roads
10:35<Taede>or different views of the same vehicle
10:35<Qantourisc>well if you do it in 3D, you can auto-rotate-and-render so it's 2.5k the :p
10:35<@planetmaker>yes... somewhat
10:37<Qantourisc>sounds like fun :p, but i'll probably end up making like 128 max....
10:37<@planetmaker>if you checkout zBase's sources, you'll see and find that it actually uses that kind of technique
10:40<@planetmaker> and especially its (sub-)repo
10:40-!-retro|cz [] has quit [Ping timeout: 480 seconds]
10:42<Qantourisc>hmm it's blender !
10:42<@planetmaker>yes, it is
10:43*Qantourisc checks it, but fears the contents :)
10:46<LordAro>Qantourisc: also note that zBase was drawn entirely by 1 person, over a period of about 3-6 months ;)
10:47<Qantourisc>nice work on that timespawn
10:47<Qantourisc>but I directly something weird
10:47<LordAro>zephyris is unmatched in drawing speed/skill
10:47<Qantourisc>the max-res-render is not rendered at aliased edges
10:48<Qantourisc>while rescale at first sight would ?
10:49<Qantourisc>ow my bad
10:49<Qantourisc>the default view was just on this mode :)
10:51-!-adf88 [] has quit [Ping timeout: 480 seconds]
10:53*Qantourisc stops looking and releases he has other unfinished projects
10:56*Qantourisc is still confused why the multi-export setting is inside blender :p
10:56<Qantourisc>o well :)
10:57<@planetmaker>probably easier scripted?
10:57-!-treaki_ is now known as treaki
10:58-!-zydeco [] has quit [Quit: Miscellaneous hardware exception error]
10:58<Qantourisc>well it's not the easy part i'm woried about
10:58<Qantourisc>it's the "what if you want to change something ?"
10:58<Qantourisc>like for example use a differnt interpolation scaling method
10:58<Qantourisc>edit each blender file by hand ?
10:59<@planetmaker>dunno whether you can apply a setting template to a file with one click (or a few)
10:59<Qantourisc>well you can import the node system iirc, but feels like just hurting yourself :)
10:59<Qantourisc>then again
10:59<@planetmaker>Qantourisc, if it's possible to specify those things outside the single blender files: that's likely a good thing
10:59<Qantourisc>windows users don't have bash :)
11:00<Qantourisc>so it would make a lot of sence to do it in blender on windows :)
11:00-!-kais58_ [] has joined #openttd
11:01<@planetmaker>there's also bash for windows ;-)
11:02<Superuser>y'all should just use emacs, which reimplements a lot of gnu tools in elisp
11:02<Superuser>(instead of cygwin)
11:02-!-kais58 [] has quit [Ping timeout: 480 seconds]
11:03<LordAro>urgh, emacs
11:03<Qantourisc>i know but notper default or easy :p
11:04<@planetmaker>windows doesn't ship with blender by default either
11:04-!-chester_ [] has joined #openttd
11:04<@planetmaker>windows ships with basically nothing by default ;-)
11:04<V453000>has internet exploder what else do you need
11:05<Qantourisc>problaby want to add imagemagick to the list of tools :)
11:05<@planetmaker>true enough. You get everything else you ever want by simply using that. With or without your consent :-P
11:05<Qantourisc>convert -resize 64 iirc :p 128.png 64.png
11:06<Qantourisc>yea i see a lot of computer passing by with shit from inet
11:06<Qantourisc>i always wonder if those peaple click on stuff or not to get them
11:07<Jomann>is there any trick to increase production of banks?
11:07<Qantourisc>Jomann: fund another ?
11:07<Qantourisc>Jomann: increase popuplation of town ?
11:07<Jomann>increasing population helps?
11:08<Kjetil>Is there any way of disabeling industrytypes from being generated ?
11:08<Qantourisc>Jomann: don't know sorry
11:09<@Alberth>Kjetil: new game -> amount of industries -> funding only ; generate
11:09<Qantourisc>how do you peaple manage to keep motivated to contribute ?
11:09<@Alberth>although technically, it generates industries rather than industry types
11:10<@Alberth>Qantourisc: depends on the person; for me it's finding out how things work
11:10<LordAro>Qantourisc: good question :)
11:10<Qantourisc>personally i'm to stressed often
11:11<Kjetil>Alberth: then I don't have enough space credits to get any industries :P
11:11<@Alberth>Qantourisc: you should make a few changes in your life then
11:11<Qantourisc>or are your lives "ok" enough that you don't have this issue, and have enerefy left ?
11:11<Qantourisc>Alberth: if only depended on me would be done by now, but takes some pushing and time
11:12<@Alberth>Kjetil: set the amount of indsustries ; do not press generate ; go to the scenario editor ; make yourself a bunch of industries ; save ; play scenario
11:12<Kjetil>hm. ok
11:13<@Alberth>Qantourisc: messing with code and stuff is the relax activity :)
11:13<Qantourisc>can also be fustrating
11:13<Qantourisc>depends on the amount of "pressure"
11:13<@Alberth>you are the one making pressure
11:14<@Alberth>which you shouldn't do :)
11:14-!-Ristovski [~rafael@] has joined #openttd
11:14<@Alberth>time is irrelevant here
11:14<@Alberth>unlike the real world :p
11:17-!-LuHa [~LuHa@] has joined #openttd
11:21<@DorpsGek>Commit by rubidium :: r26091 trunk/src/openttd.cpp (2013-11-24 16:21:19 UTC)
11:21<@DorpsGek>-Fix (r26089): double frees... woospie
11:23<Qantourisc>double free ? smells like c
11:24<LordAro>it does, unfortunately :L
11:24<LordAro>OTTD was originally written in C, and still retains quite a few C-like features
11:26<andythenorth>32bpp should be done by rendering from a voxel generator :P
11:26<andythenorth>it would be like drawing pixels, only more dimensions :P
11:26<frosch123>cubicle constructor :p
11:26<andythenorth>and would retain the TTD feel, instead of feeling like cheap, amateur CGI
11:26<andythenorth>there should be a palette of textures for it, limited to say 256 or so :P
11:27<andythenorth>constraints = useful
11:27<andythenorth>voxel games are actually appealing to look at, most mesh stuff is kind of ugly, or sacharine, or uncanny valley
11:28<@planetmaker>zbase might actually be in that exact valley
11:29<Qantourisc>it feels like amateur becaus eit probalby is :) good lookin is hard
11:29<andythenorth>just render everything in minecraft?
11:29<andythenorth>might work
11:30<andythenorth>or in LDraw
11:33<Qantourisc>what you need is good textures + a set of materials
11:33<@planetmaker>he's not an artist by education, yes
11:33<Qantourisc>well good news you don't learn that at school :)
11:33<Qantourisc>well helps :)
11:40-!-LuHa [~LuHa@] has quit [Quit: Leaving.]
11:44-!-mindlesstux [] has quit [Ping timeout: 480 seconds]
11:45<@planetmaker>well, there are art schools. But art is probably not a requirement with what he does ;-)
11:46-!-Devroush [] has joined #openttd
11:46<@planetmaker>he seems to like image processing / generation / stuff like that a lot, though :-)
11:49<Qantourisc>maybe I should try 1 blender ?
11:49<Qantourisc>ow and we need a HRDI texture
11:49<Qantourisc>yes :)
11:50<Qantourisc>for anything reflective
11:51<@planetmaker>I know that as HDR :-)
11:53-!-LeandroL [~leandro@] has quit [Ping timeout: 480 seconds]
11:53<Qantourisc>i is for image :)
11:54<@planetmaker>yeah. HDRi instead or HRDi ;-)
11:55<Qantourisc>oeps my bad
11:55<frosch123>high-definition random image?
11:55<@planetmaker>I'm not exactly sure how hdr helps though
11:55<Qantourisc>would work
11:56<frosch123>highly-random definition image?
11:56<@planetmaker>frosch123, high dynamic range. Thus more than 8 bit per channel
11:56<frosch123>ah, range
11:58<@planetmaker> <-- like composed of three different exposures. And logarithmically added
11:59<@planetmaker>difficult to convey without reduction to 8 bit again :-)
12:00<frosch123>is that something the camera does on its own?
12:00<frosch123>or do you need a tripod and switch the camera settings?
12:00<@planetmaker>at least not mine
12:00<@planetmaker>yes, a tripod helps tremendously :-)
12:00<@planetmaker>tripod + exposure sequence
12:01<Qantourisc>that show hdri in "action" :)
12:02<frosch123>Qantourisc: pm is my reference-photographer :p
12:02<frosch123>he is not my reference-astronomer though
12:02<frosch123>my brother does that :p
12:02<Qantourisc>frosch123: some camares do have a multiple-exposure feature
12:03<@planetmaker>he :-)
12:03<Qantourisc>frosch123: but not all, and it assumes "still" images
12:05<@planetmaker> <-- also HDR :-P
12:05<@planetmaker>3*30 minutes exposure :D
12:05<@planetmaker>but still not nice :-(
12:06-!-LeandroL [~leandro@] has joined #openttd
12:10<Eddi|zuHause>"also HDR" is a weird phrase, because it means something entirely unrelated if you read it in german :p
12:12<@planetmaker>hm... what does mean? :-O
12:13<frosch123>lotr maybe
12:13<frosch123>but it would be HdR
12:14<frosch123>if you start with capticalisation, you need to do it right :p
12:14<@planetmaker>doh. I should have known that one :-)
12:14<Eddi|zuHause>but then it would be LotR
12:14<frosch123>all lower case is valid in irc
12:15<frosch123>all upper case is also valid, but results in kick :p
12:16<frosch123>planetmaker: is that with or without telescope?
12:16<@planetmaker>that's with an 8" telescope f/8
12:17<frosch123>hmm, i might have no adjust my astronomer reference then
12:17<@planetmaker>sadly the electronics are broken and spare parts non-existant even at the manufacturer :-(
12:17<frosch123>iirc my brother did not manage to photograph through his telescope
12:18<@planetmaker>and without those, no star tracking, no photography
12:18<frosch123>what kind of electrioncs?
12:18<frosch123>oh, star tracking
12:18<@planetmaker>basically the main board of the telescope
12:19<frosch123>can't you just stop the earth?
12:19<@planetmaker>so it doesn't allow positioning anymore properly etc. yeah. I could. I'm scared of side-effects, though
12:19<@planetmaker>mainly on wheather
12:20<@planetmaker>the change would not be backward-compatible. Loading old savegames then would fail ;-)
12:20<frosch123>i've heard the earth rotation has been mainly driven by musicians rolling over in their graves in the past years
12:20<@planetmaker>maybe. But how could I stop them doing so?
12:20<@planetmaker>It's not like I can change contemporary music :P
12:20<frosch123>remove the atmosphere
12:21<frosch123>no air, no sound
12:22<@planetmaker>hm... that's bad for photography again. Air kinda makes for the rather smooth intensity variations between shadow and sun light
12:22<@planetmaker>w/o atmosphere it's either sun light or pitch dark
12:23<frosch123>does it need air for dust?
12:23<@planetmaker>no. The moon is my proof
12:23<Eddi|zuHause><planetmaker> so it doesn't allow positioning anymore properly etc. yeah. I could. I'm scared of side-effects, though <-- you just need two servo elements and a PLC?
12:23<@planetmaker>Or the suits of the astronauts who walked there
12:24<Eddi|zuHause>or possibly 3
12:25<@planetmaker>quite a lot of logic to programme :-)
12:26<Eddi|zuHause>it's just 3 "simple" formulas, input is position, direction and time, output is 3 angles
12:27<@planetmaker>not exactly. that's the principle, yes. But this telescope allows to add corrections for a CCD, to allow tracking in sub-pixel accuracy when using an appropriate camera
12:27<@planetmaker>so you need to allow for feedback
12:27<Superuser>programme is a noun. Program is the verb
12:27<Superuser>planetmaker, ^
12:28<@planetmaker>and you still haven't then taught how to adjust to latitude
12:29<@planetmaker>there's one guy who can somewhat put a replacement system into telescopes. But 1000€ is also a lot of money
12:43<Qantourisc>and programmed is the victem :)
12:44<Qantourisc>wait no
12:44<Qantourisc>programme is also the victem
12:46<andythenorth>these 3-part 'vehicles are made of articulated vehicles' make my brain hurt sometimes
12:46-!-George [~George@] has quit [Read error: Connection reset by peer]
12:47<frosch123>don't try to get rid of it
12:49-!-namad7 [] has joined #openttd
12:51-!-Progman [] has quit [Ping timeout: 480 seconds]
12:52-!-namad8 [] has quit [Ping timeout: 480 seconds]
12:55-!-George [~George@] has joined #openttd
13:02<chester_>andythenorth: hi, a year ago we were discussing js and python, remember?
13:04<chester_>i said i'll find a job with js and i have half year exp with it and can say its not as bad as you said
13:06-!-Alice3 [] has quit [Read error: Connection reset by peer]
13:07-!-Alice3 [] has joined #openttd
13:14-!-glx [] has joined #openttd
13:14-!-mode/#openttd [+v glx] by ChanServ
13:36-!-montalvo [] has quit [Quit: My MacBook Pro has gone to sleep. ZZZzzz…]
13:38-!-Alberth [] has left #openttd []
13:47-!-LuHa [~LuHa@] has joined #openttd
13:49-!-Progman [] has joined #openttd
14:17-!-HerzogDeXtEr [] has joined #openttd
14:20-!-oskari89 [] has joined #openttd
14:24<Eddi|zuHause>chester_: i'm sure we have over two dozen $language vs. $language discussions per year. they bear very little meaning
14:27-!-Sonny_Ji1 [~Sonny_Jim@] has quit [Ping timeout: 480 seconds]
14:28<Rubidium>that jsharp still exists
14:30<chester_>Eddi, i took part only in one, we had no conclusions then
14:46<@DorpsGek>Commit by rubidium :: r26092 /trunk/src (command.cpp script/api/script_object.cpp) (2013-11-24 19:46:16 UTC)
14:46<@DorpsGek>-Fix [FS#5818]: prevent scripts from crashing OpenTTD when they send text with command codes to user editable texts such as sign and station names
14:53<@DorpsGek>Commit by rubidium :: r26093 trunk/src/script/api/script_object.cpp (2013-11-24 19:53:35 UTC)
14:53<@DorpsGek>-Fix (r26092): compilation error
14:56-!-DarkAce-Z [~BillyMays@] has joined #openttd
14:57<@DorpsGek>TrueBrain: Commit by rubidium :: r26094 /branches/1.3 (19 files in 5 dirs) (2013-11-24 19:57:23 UTC)
14:57<@DorpsGek>TrueBrain: [1.3] -Backport from trunk:
14:57<@DorpsGek>TrueBrain: - Fix: [Script] Prevent scripts from crashing OpenTTD when they send text with command codes to user editable texts such as sign and station names [FS#5818] (r26093, r26092)
14:57<@DorpsGek>TrueBrain: - Fix: Occasional hanging when client joins [FS#5811] (r26043)
14:57<@DorpsGek>TrueBrain: - Fix: Multi line text was handled incorrectly causing glitches [FS#5809] (r26037, r26036)
14:57<@DorpsGek>TrueBrain: (...)
14:59<@DorpsGek>Commit by rubidium :: r26095 /branches/1.3 (6 files in 4 dirs) (2013-11-24 19:59:51 UTC)
14:59<@DorpsGek>[1.3] -Update: documentation
15:01-!-DarkAceZ [~BillyMays@] has quit [Ping timeout: 480 seconds]
15:03<LordAro>another release?
15:03<LordAro>it's only been a week
15:03-!-Cirno [] has joined #openttd
15:04-!-DarkAce-Z is now known as DarkAceZ
15:04<Eddi|zuHause>where did you read "release"?
15:05<LordAro>well, the documentation has dates
15:06<Eddi|zuHause>nobody ever reads the documentation :p
15:06<LordAro>including you :p
15:06<Xaroth|Work>for once I'd have to agree with Eddi|zuHause
15:07<Xaroth|Work>how's rct going, Aro?
15:07<LordAro>not bad
15:07<LordAro>making progress
15:07<LordAro>we can display coaster tracks now :)
15:08<@DorpsGek>Commit by rubidium :: r26096 /tags/1.3.3-RC2 (3 files in 3 dirs) (2013-11-24 20:08:28 UTC)
15:08<@DorpsGek>-Release: 1.3.3-RC2
15:08<LordAro>Eddi|zuHause: there: ^ :p
15:09<Eddi|zuHause>LordAro: it's still "just" a release candidate, so you can expect another "release" fairly soon anyway
15:10<LordAro>oh i know, but i still think a week between release candidates is quite quick
15:10<LordAro>as per usual, anyway
15:10<Xaroth|Work>rubi has been doing a looot of commits the past few days
15:12<LordAro>wiki updated :)
15:19-!-andythenorth [] has quit [Quit: andythenorth]
15:23-!-tomlane [] has joined #openttd
15:25-!-andythenorth [] has joined #openttd
15:25<andythenorth>chester_: there are plenty of good js jobs, js is just a language I like to disparage
15:26<andythenorth>I used to write it when it was broken
15:26<andythenorth>that was boring
15:26<andythenorth>I got burnt, I've maintained a small level of hatred ever since :)
15:26<andythenorth>otherwise it's just a language ;)
15:27<tomlane>I am having trouble getting the soundtrack to play on OS X, I have tried two different sound packs so I am guessing it is a system issue of some sort. I have freepats and timidity installed via homebrew. Any ideas?
15:31<+michi_cc>OS X isn't using timidity. There used to be a MIDI/Core Audio control panel somewhere (no idea if it is still there in current OS versions), maybe a wrong output device is activated.
15:31-!-oskari89 [] has quit []
15:32<tomlane>michi_cc, I have sound effects but no music
15:32<@planetmaker>then, how did you install the music sets?
15:33<+michi_cc>Check the OTTD jukebox if music is in fact running and if not, hit play.
15:34<@planetmaker>did you select the music set in the game options?
15:35<chester_>ive been full-time js programmer for half a year, its not just a language, its front-end language, 'front' meaning modern, from which other languages borrow
15:35<tomlane>*doh* too used to it autoplaying, thanks for the help ><
15:39<frosch123>chester_: interesting, can you name a thing from js from the past 10 years, that other languages borrowed?
15:39<andythenorth>ask doug crockford
15:39<andythenorth>he would probably know
15:41<andythenorth>some people even *like* javascript, which is odd
15:41<andythenorth>but anyway
15:43<andythenorth>what's a sensible default for a switch checking current_railtype ?
15:43<chester_>i was not very clear maybe, there are things from other languages, like lisp, which were realized in js before java or c++, i mean anonymous functions and those functional things
15:43<andythenorth>is that the same as a lambda?
15:43<frosch123>sounds like
15:43<chester_>cpp has lambdas?
15:44<andythenorth>chester_: I'm not arguing against you :)
15:44*andythenorth is just dicking around
15:45<frosch123>the wiki article rather suggests js borrowing stuff from python :p
15:45<frosch123>but well, all interpreted languages are the same, once they head for professional usage
15:46<chester_>guido ripped reduce afaik, and wanted to slay map and other things also
15:46<andythenorth>hmm, this bi-mode engine works
15:46<andythenorth>5000hp on electric, 1250hp otherwise
15:47<andythenorth>rather easy, once I'd figured out how my own framework works :x
15:47<Eddi|zuHause>i rather like map
15:47<andythenorth>now I probably get stoned with '10 things you wish you had known about railtype compatibilty' :(
15:47<Eddi|zuHause>andythenorth: so what about multi-power tracksets?
15:47<andythenorth>the railtypes spec makes sense, but the implementations baffle me
15:48<andythenorth>Eddi|zuHause: dunno, I think we'll find out empirically
15:48<andythenorth>people report oddness, I might try and fix it
15:48<andythenorth>otherwise, I have fingers in ears, singing la la la la
15:48<Eddi|zuHause>andythenorth: i wrote about action6/7/9/D operators checking railtype compatibility somewhere
15:49<Eddi|zuHause># lalalala
15:49<Eddi|zuHause># lalalalalala
15:49<Eddi|zuHause>that is a weird song
15:49<LordAro>chester_: as of c++11, lambdas are a thing
15:49<andythenorth>now I have it in my head, thanks
15:49<Eddi|zuHause>well YOU brought it up :p
15:50<andythenorth>these steam ships would look better with smoke, eh?
15:50*andythenorth changes topic
15:50<Eddi|zuHause>loads of languages had lambda before js did
15:54<Rubidium>thousands of years before js ;)
15:55<chester_>i've seen '((){}[])() is now valid in c++' pic in our cpp department, yes it was added recently
15:57<chester_>or [](){} or ()[]{}
15:57<chester_>something like that
15:58<LordAro>[](){}, methinks
16:00<Eddi|zuHause>can't be much less weird than for(;;);
16:00<chester_>this is breakthrough for cpp, when its been for thousands of years in other languages
16:00<frosch123>Eddi|zuHause: you cannot beat '*'
16:01<frosch123>it's by far the worst c construct
16:01<Eddi|zuHause>Rubidium: i do believe that is illegal :p
16:02<chester_>showd give 2*i + 1
16:02<Eddi|zuHause>(and even if it werent, it would be undefined behaviour)
16:02<frosch123>planetmaker: not an lvalue
16:02<LordAro>int i = 0; i = i+++++i;
16:02<chester_>(i++) + (++i)
16:02-!-DanMacK [] has joined #openttd
16:02<DanMacK>Hey all
16:02<andythenorth>ohai DanMacK
16:02<frosch123>chester_: no, the lexer is greedy
16:02<Eddi|zuHause>chester_: it could be 0+1 or 1+1 or 0+2
16:03<frosch123>Eddi|zuHause: no, it's a compile error, "++ expected lvalue"
16:03<LordAro>"lvalue required as increment operand"
16:03<LordAro>yup :)
16:03<Eddi|zuHause>frosch123: yes
16:03<@planetmaker>frosch123, but is it a compiler thing or a language thing?
16:04*Rubidium wonders whether the majority of C is actually undefined behaviour ;)
16:04<Eddi|zuHause>chester_: even with the parentheses, it's undefined behaviour, since there are two assingments without a sequence point
16:05<+glx>and + operator argument evaluation order is implementation dependant IIRC
16:05<frosch123>planetmaker: without the parantheses it is a language thing, with them it is a compiler thing
16:06<chester_>would i++ + ++i work? js eats it
16:06<frosch123>hmm, maybe eddi was also in my cert training course
16:06<frosch123>chester_: what are the sequence points of js?
16:06<LordAro>chester_: the fact that js eats it is meaningless - i'm sure php eats it too :p
16:06<frosch123>in any case: it is bad style
16:06<Eddi|zuHause>frosch123: i did attend a language lecture a few years ago
16:07<Eddi|zuHause>glx: operator evaluation order has actually nothing to do with assignment evaluation order
16:08<Eddi|zuHause>i++ just means "increment i at some point after you evaluated this operator, and before the next sequence point"
16:11<Eddi|zuHause>so it could be "evaluate the left operand" (=>0), "assign the right ++" (=>i=1), "evaluate the right operand" (=>1), "assign the left ++" (=>i=1)
16:11<Eddi|zuHause>result of the operation would be 1
16:12<Eddi|zuHause>err, the last one should be (=>i=2)
16:12<Eddi|zuHause>or it could be "evaluate the left operand" (=>0), "assign the left ++" (=>i=1), "assign the right ++" (=>i=2), "evaluate the right operand" (=>2)
16:12<Eddi|zuHause>result of the operation would be 2
16:16-!-skyem123 [] has quit [Read error: Connection reset by peer]
16:20<chester_>never heard of sequence points, Eddi's explanation is clear though, does it mean if its undefined its not valid at all or compiler dependent or what?
16:20<Eddi|zuHause>it means it's compiler dependent
16:20<@planetmaker>the latter
16:21<frosch123>it may also be random, a compiler may compile it differently at various lines
16:21<Eddi|zuHause>yeah, it could be shuffled during optimization
16:21<@planetmaker>that's the special form of compiler-dependent :-)
16:21-!-tomlane [] has quit [Quit: Textual IRC Client:]
16:21<frosch123>for certain platforms (maybe ia64), the may even be really random due to parallel execution
16:22<Eddi|zuHause>ia64 is not really parallel, it's just 4 instructions compressed into one
16:23<frosch123>i thought it was the platform without hardware sheduler, where the compiler optimised for the pipelines
16:24<Eddi|zuHause>you might be thinking about early mips
16:24<Eddi|zuHause>where when executing the second operation the results of the first are not yet in
16:24<Eddi|zuHause>so if you have data dependencies you have to insert a nop
16:25<Eddi|zuHause>but i'm not a hardware expert
16:25-!-HerzogDeXtEr1 [] has joined #openttd
16:26<frosch123> <- what a chart :p
16:27<Eddi|zuHause>frosch123: well sometimes things just come at the wrong time
16:27<Eddi|zuHause>frosch123: if it had come 4 years earlier, that figure might have been accurate
16:28<Eddi|zuHause>but compared to amd64, ia64 was just too incompatible to the older platforms
16:29<Eddi|zuHause>and compatibility just outweighs the possibly cleaner architecture
16:29<+michi_cc>Anything without delay slots after a branch is lame :) Some DSP-like microcontroller architecture have three or even more.
16:31<Eddi|zuHause>michi_cc: you mean things where it discards the results of the operations when it discovers that the branch was taken earlier?
16:31-!-HerzogDeXtEr [] has quit [Ping timeout: 480 seconds]
16:34<+michi_cc>No, not discard. A jump on a FPU condition on a TI TMS320..something will modify the IP only after the next three instructions were already fetched. And somebody thought it a nice idea to simply not discard anything already fetched.
16:38<Eddi|zuHause>ah, so you mean they are always executed. yes, that is the "simple" architecture
16:38<Eddi|zuHause>(simple for the hardware, complicated for the compiler)
16:40<Eddi|zuHause>how many instructions are fetched depends on your pipeline length
16:46-!-Supercheese [~Superchee@] has joined #openttd
16:48<Eddi|zuHause>chester_: a sequence point (in c(++)-specs speak is the end of the statement (';') or the comma operator (',') [but not the comma that separates function call arguments]
16:49<frosch123>|| and && as well
16:49<Eddi|zuHause>yeah. i knew i forgot something
16:49<Eddi|zuHause>openttd's calculation of the station rating makes heavy use of that
16:51<__ln__>have a look at this disney movie instead, everyone:
16:53-!-sla_ro|master [slamaster@] has quit []
16:54<Eddi|zuHause>__ln__: now let them make real faces
16:54<Eddi|zuHause>(without mapping them on a real actor's motions)
16:57<Eddi|zuHause>|| and && are odd anyway, because it's the only place where it specifies evaluation order
17:01-!-DanMacK [] has quit [Quit: Page closed]
17:11-!-frosch123 [] has quit [Quit: be yourself, except: if you have the opportunity to be a unicorn, then be a unicorn]
17:11-!-Tulitomaatti [] has quit [Ping timeout: 480 seconds]
17:16-!-Ristovski [~rafael@] has quit [Quit: Leaving]
17:21-!-valhallasw [] has quit [Quit: leaving]
17:25-!-HerzogDeXtEr1 [] has quit [Read error: Connection reset by peer]
17:32-!-Haube [] has joined #openttd
17:40-!-Djohaal [] has joined #openttd
17:45-!-Devroush [] has quit []
17:54<andythenorth>good night
17:54-!-andythenorth [] has quit [Quit: andythenorth]
18:06-!-Tulitomaatti [] has joined #openttd
18:09-!-Midnightmyth [] has quit [Read error: Operation timed out]
18:17-!-Progman [] has quit [Remote host closed the connection]
18:31-!-Wolf01 [] has quit [Quit: Once again the world is quick to bury me.]
18:42-!-Alice3 [] has quit []
18:45-!-TomyLobo [] has quit [Quit: Standby mode...]
18:46-!-DabuYu [DabuYu@] has joined #openttd
18:47-!-MNIM [] has joined #openttd
19:06-!-wakou2 [] has quit [Quit: Konversation terminated!]
19:09-!-GriffinOneTwo [] has joined #openttd
19:29-!-Elukka [] has quit []
19:37-!-treaki_ [] has joined #openttd
19:44-!-treaki [] has quit [Ping timeout: 480 seconds]
19:49-!-Djohaal [] has quit [Quit: Leaving]
19:52-!-yorick [] has quit [Remote host closed the connection]
19:55-!-DarkAce-Z [~BillyMays@] has joined #openttd
20:00-!-DarkAceZ [~BillyMays@] has quit [Ping timeout: 480 seconds]
20:05-!-KritiK [] has quit [Quit: Leaving]
20:15-!-Superuser [] has quit [Remote host closed the connection]
20:22-!-DabuYu [DabuYu@] has quit []
20:34-!-LordAro [] has quit [Remote host closed the connection]
20:37-!-DabuYu [DabuYu@] has joined #openttd
20:41-!-Jomann [] has quit [Remote host closed the connection]
21:45-!-Haube [] has quit [Read error: Connection reset by peer]
21:46-!-glx [] has quit [Quit: Bye]
21:49-!-George|2 [~George@] has joined #openttd
21:49-!-George is now known as Guest6856
21:49-!-George|2 is now known as George
21:51-!-Guest6856 [~George@] has quit [Ping timeout: 480 seconds]
22:11-!-Dark-Ace-Z [~BillyMays@] has joined #openttd
22:16-!-DarkAce-Z [~BillyMays@] has quit [Ping timeout: 480 seconds]
22:19-!-Dark-Ace-Z is now known as DarkAceZ
22:20-!-Pereba_ [~UserNick@] has joined #openttd
22:26-!-Pereba [~UserNick@] has quit [Ping timeout: 480 seconds]
22:26-!-Pereba_ is now known as Pereba
22:32-!-Pereba_ [~UserNick@] has joined #openttd
22:37-!-Pereba [~UserNick@] has quit [Ping timeout: 480 seconds]
22:37-!-Pereba_ is now known as Pereba
23:16-!-LuHa [~LuHa@] has quit [Quit: Leaving.]
23:41-!-DabuYu [DabuYu@] has quit [Ping timeout: 480 seconds]
23:58-!-Pereba [~UserNick@] has quit [Quit: Little strokes fell great oaks.]
---Logclosed Mon Nov 25 00:00:49 2013