IRC logs for #openttd on OFTC at 2011-08-28
⏴ go to previous day
00:40:59 *** welterde has joined #openttd
01:13:16 <supermop_> any dutch people around?
01:14:55 <supermop_> does that mean yes?
01:15:08 <supermop_> I dont know much dutch
01:15:41 <Markk> I learned some dutch words from a guy I shared a flat with before.
01:16:28 <Markk> I say Goedemoergen alot (on the morning that is).
01:16:48 <Markk> Especially because it's pretty close to the swedish words for it: God morgon.
01:17:35 <supermop_> they just started selling aged bols in the us
01:32:04 <supermop_> ... i guess that is sort of a dutch thing
02:27:00 *** rhaeder1 has joined #openttd
02:55:14 *** supermop_ has joined #openttd
03:48:39 *** goldspirit has joined #openttd
03:48:49 <goldspirit> i need some information
03:49:02 <goldspirit> so if you can chat me it will be very apreciate
03:52:08 <goldspirit> i want to know how to make a multiplayer game for me and my uncle
03:52:29 <goldspirit> when i try to make the server it's say server offline
04:01:03 <goldspirit> can i make a multiplayer game
04:01:12 <goldspirit> for me and my oncle and friend
04:01:24 <goldspirit> cuz when i try it.s making an ofline server
04:03:03 <DDR> Odds are against it, but what's up?
04:04:20 <goldspirit> against multiplayer online with friend?
04:05:04 <DDR> I'd rate it as maybe 40/60 odds, against your favor, that I can help you.
04:05:27 <goldspirit> but what's the problem :O?
04:06:50 <DDR> I'll give it maybe 80% chance you actually tell me the problem, and then maybe half-and-half that I know the answer.
04:07:06 <DDR> But, that's just an uneducated guess. I just got here!
04:07:35 <goldspirit> an online multiplayer game for my uncle and my friend
04:07:41 <goldspirit> but everytime i try it make
04:08:39 <DDR> Hm, I've had problems with that myself. Never figured out the answer, the other guy I was playing with could make. :/
04:09:00 <goldspirit> no problem if you finaly have an answer
04:09:10 <DDR> But, what happens every time? Perhaps someone else knows.
04:09:33 <goldspirit> i try to make the game and it'sa offline
04:10:06 <DDR> Ah. Tried making a private online server?
04:10:18 <DDR> That might work around the problem, but you need internet access. :/
04:10:39 <goldspirit> lol if im chatting here i have it :)
04:11:29 <DDR> Well, I have a friend who has three offline computers which we like to hook into a simple LAN to game. Those computers don't have access to the internet, but my laptop right beside it can hook into his wireless.
04:14:44 <goldspirit> information in the game
04:15:12 <goldspirit> that we can change with
04:26:36 <pjpe> why does doing ./configure --enable-profiling && make run-prof only show me function calls related to os x drawing the game
04:56:23 *** Eddi|zuHause has joined #openttd
05:01:09 *** Kurimus has joined #openttd
05:21:24 *** HerzogDeXtEr1 has joined #openttd
07:01:04 *** andythenorth has joined #openttd
07:09:41 *** TWerkhoven has joined #openttd
07:54:11 *** fjb|mobile has joined #openttd
07:56:05 *** Alberth has joined #openttd
07:56:05 *** ChanServ sets mode: +o Alberth
08:03:12 *** andythenorth has joined #openttd
08:03:41 <pjpe> huh someone made a patch for 10.7 full screen support
08:10:08 <andythenorth> 10.7 is nonsense
08:10:13 * andythenorth ponders manual-dist
08:10:24 <andythenorth> uses the yacd cargo routing
08:11:22 <andythenorth> but instead of yacd calculating destinations, player would go to (e.g. coal mine) and use a routing menu select a valid destination for coal
08:11:30 <andythenorth> with % allocations between n destinations
08:11:51 <planetmaker> you can do that without fancy menu...
08:13:54 <planetmaker> just distribute it to different sources and don't use yacd
08:14:03 <andythenorth> that's what gave me the idea
08:14:06 <planetmaker> that's 95% of what you asked...
08:14:19 <andythenorth> I am playing a non-yacd game
08:14:29 <andythenorth> I miss the smart cargo routing
08:14:39 <andythenorth> I don't miss some of the weirder side effects of routing primary cargo
08:17:08 <pjpe> am i wrong or is "./configure --enable-profiling && make run-prof" supposed to give me more than os x specific functions involving NSWindows and NSApplications and such
08:17:34 <andythenorth> also...if the routes could (optionally) be specified in SE, but not edited in game, that would be interesting for scenarios
08:17:57 <planetmaker> NS* is osx-specific
08:18:35 <pjpe> i figured the profiling would show me game code functions
08:18:40 <__ln__> not necessarily, those could be available also in GNUStep
08:20:17 <andythenorth> planetmaker: did you upgrade to lion?
08:26:23 <Eddi|zuHause> planetmaker: the difference is automatic transfer (even in two directions)
08:28:02 <Eddi|zuHause> andythenorth: i think most of that can be done using the "cargodist" type of routing
08:28:30 <Eddi|zuHause> andythenorth: i.e. the available destinations are chosen from the ones actually connected
08:28:49 <planetmaker> andythenorth: I didn't yet
08:29:27 <planetmaker> after cloning of my hdd seemed to fail on some circular symlinks, I was too pissed at that and too lazy to install everything anew on the new hdd
08:31:12 <andythenorth> I only just got snow leopard
08:31:20 <andythenorth> I won't upgrade to lion soon - if ever
08:31:51 <andythenorth> at some point in the next few years I probably have to find a new OS :|
08:32:29 <andythenorth> it's almost impossible to believe Apple won't make an almighty mis-step
08:32:48 *** fjb|mobile_ has joined #openttd
08:33:06 <pjpe> what makes you think that
08:33:51 <Eddi|zuHause> apple will likely disappear into the marginal company it came from as soon as they miss out on one innovation step
08:34:01 <Eddi|zuHause> just look at nokia...
08:34:11 <andythenorth> they'll miss a bunch in a row
08:34:35 <pjpe> nokia is a marginal company?
08:34:50 <pjpe> their revenue last year was 42 billion
08:34:58 <__ln__> last year was last year
08:35:19 <__ln__> nonetheless, they are becoming an OEM manufacturer
08:35:38 <Eddi|zuHause> missing out on the whole smartphone thing, not getting any of their products launched, then practically bought out by microsoft
08:35:50 <pjpe> first quarter this year it held 30 percent of the market share of cellular devices
08:35:54 <Eddi|zuHause> market share dwindling drastically during that time
08:37:13 <__ln__> nokia is in that position because of religious reasons. (and the religion is symbian.)
08:38:04 <andythenorth> when I got OS X beta there were lots of older guys on forums saying "system 7 was quite good enough for me. I certainly don't want any of this unix nonsense"
08:38:09 <andythenorth> I have become that guy :P
08:39:12 <Eddi|zuHause> __ln__: what about meego?
08:39:52 <planetmaker> anyway, two tasks less in our bug tracker - and that without work
08:40:55 <andythenorth> why change an OS :P
08:41:04 <__ln__> Eddi|zuHause: it's been said most nokians were interested in meego only in the sense "how much will it cut down sales of symbian?"
08:41:17 <planetmaker> andythenorth: two reasons only: security and application support
08:41:26 <andythenorth> used to also be performance
08:41:26 <planetmaker> both don't call yet for lion
08:41:35 <andythenorth> lion == much less application support
08:41:54 <andythenorth> performance won't get much better until I can afford an SSD
08:42:11 <andythenorth> stability is a solved problem
08:42:24 <andythenorth> and new GUI stuff I can't be bothered to learn any more
08:45:56 *** JVassie has joined #openttd
08:46:06 *** Vikthor has joined #openttd
08:47:12 *** Cybertinus has joined #openttd
08:48:07 <planetmaker> hardware support is actually a third criterion for OS upgrades
08:49:49 <Eddi|zuHause> the main problem is the lack of backwards compatibility
08:51:48 <Eddi|zuHause> (ranging from something just not working anymore because it was badly programmed in the first place, to deliberately deprecating fairly new APIs, meaning a rewrite for most apps older than 2 years)
09:05:40 *** Progman has joined #openttd
09:42:12 *** fjb|mobile has joined #openttd
10:00:40 *** frosch123 has joined #openttd
10:03:35 *** |Jeroen| has joined #openttd
11:13:48 *** Adambean has joined #openttd
11:18:36 *** andythenorth has joined #openttd
12:02:09 <DorpsGek> andythenorth: Yexo was last seen in #openttd 13 hours, 27 minutes, and 23 seconds ago: <Yexo> or gui from openttd pasted over graphics from some other game
12:03:37 *** DayDreamer has joined #openttd
12:08:48 *** ProgVal has joined #openttd
12:53:14 <Hirundo> What happens (or rather, should happen) if a railtype grf does not define all the 'required' entries in the action3, but does define some others?
12:54:10 <Hirundo> For example, a grf that defines only depot or catenary gfx, but no overlays/underlays
12:58:04 <frosch123> what currently happens: it draws nonsense
12:58:16 <frosch123> like mousecursors and character glyphs
12:59:25 <frosch123> what should happen? either draw the questionmark, or revert to default graphics
12:59:37 <frosch123> no idea how much work are either of those
13:03:33 <planetmaker> default graphics might be the nicer fallback and allow for more flexible customization
13:07:42 <Hirundo> Basically you need to supply all the stuff with rails on them in one go
13:08:04 <Hirundo> Having different overlays and underlays just makes no sense
13:08:15 <Hirundo> But supplying catenary and nothing else might, though
13:10:31 <Hirundo> IMO it should be 'all or nothing' for the required stuff (at least NML could enforce that), but all the others are optional and could be defined individually
13:16:12 *** supermop_ has joined #openttd
13:51:02 * andythenorth is bored of station ratings
13:51:09 <andythenorth> maybe a grf to keep them at 100% :P
14:39:12 *** Zeknurn has joined #openttd
14:57:42 *** douknoukem has joined #openttd
15:13:25 *** Brianetta has joined #openttd
15:42:54 *** JVassie has joined #openttd
16:26:14 *** Progman has joined #openttd
16:42:52 *** TWerkhoven has joined #openttd
16:46:35 *** andythenorth has joined #openttd
17:08:53 *** TWerkhoven2 has joined #openttd
17:21:44 *** Zeknurn has joined #openttd
17:35:20 <Yexo> <Hirundo> [15:10:31] IMO it should be 'all or nothing' for the required stuff (at least NML could enforce that), but all the others are optional and could be defined individually <- but it's perfectly valid to check if a certain railtype is defined and if so, override only the depots
17:35:28 <Yexo> at least from what I remember that works correctly
17:38:19 *** LadyHawk has joined #openttd
17:40:48 <LadyHawk> hmm so i'm playing ottd again with these huge amounts of trains over a small piece of track
17:41:13 <LadyHawk> and i found something that's a bit strange with path signals
17:42:02 <LadyHawk> basically, if a train entering a path signal block has 3 possible exits
17:42:37 <LadyHawk> it will pick the one that's been free the longest amount of time
17:42:49 <LadyHawk> even if that exit track's signal is green, but blocked off by another train
17:43:03 <LadyHawk> instead of picking a perfectly free path, the train STOPS and waits
17:43:54 <frosch123> do not mix path signals with exit signals
17:45:13 <LadyHawk> would it be possible to adjust the behaviour so it picks the SHORTEST free route instead?
17:45:26 <CIA-2> OpenTTD: translators * r22852 /trunk/src/lang/ (finnish.txt unfinished/persian.txt vietnamese.txt):
17:45:26 <CIA-2> OpenTTD: -Update from WebTranslator v3.0:
17:45:26 <CIA-2> OpenTTD: finnish - 2 changes by USephiroth
17:45:26 <CIA-2> OpenTTD: persian - 3 changes by Peymanpn
17:45:26 <CIA-2> OpenTTD: vietnamese - 2 changes by nglekhoi
17:45:51 <Ammler> LadyHawk: maybe your penalty settings are screwed up, try with a new cfg
17:46:00 <frosch123> LadyHawk: your observation is wrong, the pathfinder always picks the shortest wrong
17:46:04 <LadyHawk> or is that a lot of coding? (i know hassle in the past with pbs)
17:46:16 <frosch123> likely your signal layout is wrong, but you did not show us a screenshot
17:46:35 <LadyHawk> i could show a screenshot, hang on
17:46:40 <LadyHawk> gotta take one firstg
17:46:48 <planetmaker> that is never required when talking about signals and vehicles moving or loading funnily
17:47:21 <LadyHawk> and i'll also have the issue i'm describing on the4 screenshot as well
17:50:11 <LadyHawk> tryiing to get into my ftp now..
18:01:21 <LadyHawk> all tracks go to the exact same spot, so the waiting train should just go straight over, which is the shortest route and also the only free route available
18:01:23 <LadyHawk> but it doesn't want it
18:01:57 <LadyHawk> because it was the longest free track until someone else reserved a path to it
18:02:08 <Eddi|zuHause> you might want to change the amount of signals considered for lookahead
18:02:48 <LadyHawk> oh that sounds like something that's been annoying me for a while
18:03:21 <LadyHawk> where could i find this setting?
18:04:29 <Eddi|zuHause> try "list_settings lookahead" or something
18:04:59 <Eddi|zuHause> default value is 10
18:05:03 <frosch123> at first you should remove at least the first three signals on every line after the junction
18:05:47 <LadyHawk> frosch123 i do that and the trains block up in front of the junction, there's too much traffic it needs to go fast ;)
18:05:56 <LadyHawk> Eddi|zuHause it's not actually showing me a value when i type that in
18:06:15 <Eddi|zuHause> LadyHawk: then the name is something else
18:06:32 <Eddi|zuHause> LadyHawk: try "list_settings signal"
18:06:49 <Eddi|zuHause> LadyHawk: or "list_settings pf.yapf.rail"
18:07:32 <LadyHawk> hmm list_settings signal showed me a bunch of things
18:08:02 <LadyHawk> pf.yapf.rail_look_ahead_max_signals 10
18:08:29 <LadyHawk> i'm guessing this is the one
18:08:59 <LadyHawk> so how would i change this one
18:09:09 <LadyHawk> setting signal pf.yapf.rail_look_ahead_max_signals 5
18:09:32 <Eddi|zuHause> LadyHawk: without the first "signal"
18:10:20 <Eddi|zuHause> so it would have been "look_ahead" earlier
18:10:27 <LadyHawk> done, it didn't give me any confirmation, just a new command line
18:11:55 <Eddi|zuHause> mind you: all changes to pathfinder settings may cause trouble at the other end of your network
18:12:13 <Eddi|zuHause> a guy recently had trouble because this setting was too low
18:13:35 <LadyHawk> and i must say, the trains stop behaving randomly
18:14:03 <LadyHawk> now they're showing a clear pattern as to which exit track they prefer when they have more than 1 choice
18:16:04 <LadyHawk> it's going a lot smoother on busy blocks like that now (116 trains in total on that track in the screenshot)
18:16:15 <LadyHawk> thanks a lot for showing me this :D
18:16:53 <Eddi|zuHause> beware of loops when setting this to low values
18:17:18 <Eddi|zuHause> a train might decide to go in circles instead of waiting for a signal to become green
18:18:06 <LadyHawk> the chosing depots thing looks broken now but at least it makes sense now as to what the train wants to do
18:18:15 <LadyHawk> so i can fix it properly
18:18:30 <LadyHawk> that's another thing the trains just kept doing at random... can't fix something if it seems random
18:18:58 <Eddi|zuHause> nothing is random in this game. not even the random values :p
18:19:29 <LadyHawk> if i can't spot a behaviour pattern i find it random enough
18:20:53 <LadyHawk> thanks again Eddi|zuHause, i'm gonna go now :)
18:21:16 <Eddi|zuHause> what you mean is chaotic :)
18:25:08 <Eddi|zuHause> chaotic means there is a system behind it that we do not (yet) recognize
18:25:49 <Eddi|zuHause> that's basically the opposite of random (which means there is no system behind it at all)
18:30:28 <supermop_> chaos means 'without order'
18:32:08 <Eddi|zuHause> yes, until you bring in the phrase "the genious brings order to the chaos"
18:32:44 <supermop_> that implies that chaos has no order
18:33:10 <Eddi|zuHause> but it has, the non-genious is just unable to see it
18:33:32 <supermop_> if improved perception divines order, then it was not a chaos to begin with
18:33:44 <planetmaker> chaotic in a scientific term means that it depends on the initial conditions more sensitively than you can measure
18:33:51 <planetmaker> (or boundary conditions)
18:34:39 <Eddi|zuHause> chaos is the oldest god (in greek mythology)
18:34:49 <Eddi|zuHause> then gaia was born
18:35:36 <supermop_> yes, chaos is the primordial state before order was created
18:36:40 <planetmaker> hm... can an industry read the road bits of an adjacent tile?
18:36:55 <planetmaker> I only find that I can see "road yes/no" - but maybe I miss something
18:37:17 <Eddi|zuHause> i don't think the road bits can be read
18:37:21 <planetmaker> I want to draw a gate where the road goes towards the industry tile :-)
18:38:00 <Eddi|zuHause> planetmaker: instead of road bits, you need something like "road continuation info" (similar to rail stations)
18:38:49 <Eddi|zuHause> planetmaker: makes it more flexible wrt. bridge heads, road stations, depots, "traffic objects", ...
18:39:03 <planetmaker> yes, that would suit better
18:40:37 <Eddi|zuHause> also, you might want to draw a gate for rails as well ;)
18:41:37 <Eddi|zuHause> for future flexibility, "road continuation info" may be a list of lanes with an offset and a direction
18:42:40 <Eddi|zuHause> by default, offsets might be 6 and 10. in/out depending on road side
18:42:55 <planetmaker> hm... interesting, yes :-)
18:49:22 <Eddi|zuHause> if you limit number of lanes to 4, you could bitstuff that into a dword (4 bit offset (0-15) and 3 bit direction (DIR_N..DIR_NW))
18:49:52 <Eddi|zuHause> leaves 1 bit for used/unused
18:52:41 *** supermop__ has joined #openttd
18:53:59 <Hirundo> Yexo: The guy in the NML topic complained that his depot-only grf didn't work
18:54:31 <Hirundo> As far as I've been able to see, depots are the only case where having the 'required' stuff is a prerequisite
19:02:59 <Hirundo> Which gave the confusing result that his grf only worked if loaded after a grf that re-defined "RAIL" itself (eg swedishrails)
19:07:11 *** TWerkhoven has joined #openttd
19:24:35 <perk11> It used to work in previous versions
19:26:11 <Terkhen> what does it do now and what did it before?
19:26:57 <glx> exit signals without entry is strange
19:26:58 <perk11> the train went to the depot when all the paths are full
19:27:24 <perk11> and now they just choose the nearest path
19:30:01 <perk11> It worked maybe because there was a combo signal in the depot
19:31:00 <michi_cc> My guess: You have pf.yapf.rail_firstred_twoway_eol set to false (the default since 1.1 or maybe 1.0, not sure) while the example assumes a value of true. There are a lot more situations where false makes more sense than those few that only work with true, which is why it was changed.
19:33:38 <Eddi|zuHause> i think there are variants of overflow depots that work without "eol" being set to true
19:33:56 <perk11> michi_cc: you are right
19:34:10 <Eddi|zuHause> basically you need to make sure that firstred_exit_penalty is larger than the depot penalty
19:34:35 <perk11> Eddi|zuHause: maybe someone will add one of those to the wiki?
19:34:52 <Eddi|zuHause> i am pretty sure i have seen it somewhere
19:35:13 <Eddi|zuHause> perk11: the wiki page should rather more clearly say that this value has to be said
19:37:55 *** Cybertinus has joined #openttd
19:43:52 <frosch123> ah, it allows replacing the depot without the tracks, right?
19:44:57 <frosch123> is this the only place? what about the gui?
19:45:14 <frosch123> i.e. the orientation picker
19:45:49 <Hirundo> The GUI sprite selection is not so picky
19:45:56 <Hirundo> either you replace or do not replace all sprites
19:46:44 <frosch123> i mean the orientation picker, not the toolbar
19:48:12 <frosch123> DrawTrainDepotSprite() needs the same change
19:48:51 <Hirundo> can't test right now as my just-compiled OpenTTD crashes on startup
19:50:42 <Hirundo> And now I did *something* that made MSVC think it needs a full recompile *argh*
19:51:55 <Hirundo> string change doesn't rebuild blitters/drivers :S probably touched some config while messing with my repo
19:53:05 <xahodo> Which advanced setting regulates how long trains wait to find a path past a signal?
19:53:48 <xahodo> I've got trains waiting on a busy peace of track while a continued path is available.
20:03:30 <Eddi|zuHause> they calculate which path is the shortest and then wait until that exact path gets free. if an actually free path is longer, they never take that
20:04:29 <xahodo> problem is, they wait for the free route, even though they can already take it for some time.
20:05:29 <xahodo> it takes them about a second to realize they can move on.
20:05:39 * planetmaker senses again a discussion about "this train should go here or there" without savegame or least a screenshot
20:06:12 * xahodo foobared some patch setting, wonders which
20:07:44 <Eddi|zuHause> xahodo: try the path backoff interval
20:09:50 <Hirundo> AFAIK, all other non-required stuff (catenary, gui) works the same
20:11:51 *** sla_ro|master has joined #openttd
20:18:53 *** JVassie has joined #openttd
20:20:49 <xahodo> Thanks for the help Eddi|zuHause, but it turned out to be pf.wait_oneway_signal
20:20:56 <xahodo> I reduced its value to 2
20:22:50 <xahodo> errr... I also reduced pf.path_backoff_interval to 1 at first. but the latter setting did the trick.
20:28:16 <CIA-2> OpenTTD: frosch * r22853 /trunk/src/ai/api/ai_tile.hpp: -Doc: Add a missing bit of AI documentation.
20:33:42 <CIA-2> OpenTTD: frosch * r22854 /trunk/src/rail_cmd.cpp: -Change: [NewGRF] Allow replacing depot sprites without having to provide rail overlays. (Hirundo)
20:46:17 <Eddi|zuHause> two people thanked each other and you missed the point.
20:50:20 *** lessthanthree has joined #openttd
20:56:17 <pjpe> how did you get liblzma on os x
20:57:13 <planetmaker> sudo port install lzma
20:57:28 <pjpe> i tried brew install lama and opened doesn't recognize it
20:57:54 <planetmaker> everytime I looked at their pages I was like "wth?!"
20:58:07 <planetmaker> if people follow their advice they're on the path to sudo rm -rf /
20:58:58 *** andythenorth has joined #openttd
21:02:25 <planetmaker> basically you'll have to teach the system to look for the header files and libaries in the proper paths, though
21:05:02 <planetmaker> and... the proper package name is liblzma of course
21:05:24 <planetmaker> but I guess the wiki describes it
21:05:42 <pjpe> brew only has one called lzma
21:06:02 <planetmaker> you need not the archiver but the development libraries and headers
21:06:13 <planetmaker> macports has both, lzma and liblzma
21:08:36 <pjpe> now why does make run-prof profile right on ubuntu but not on os x
21:09:20 <planetmaker> osx is a bsd, not a linux ;-)
21:14:09 *** JVassie has joined #openttd
21:15:31 <__ln__> is there any unit of length that cannot be converted to metres by multiplying with a constant?
21:29:13 <Eddi|zuHause> __ln__: you mean like a logarithmic scale?
21:32:00 <__ln__> Eddi|zuHause: that, and/or zero has an offset.
21:33:01 <Eddi|zuHause> (off the top of my head) the only units which have an offset are for temperature and pressure
21:37:29 <Eddi|zuHause> (pressure used to be given in "Atmosphärenüberdruck" (atü) with the offset 1 bar = 0 atü)
21:41:05 <planetmaker> energy has an arbitrary offset
21:41:14 <planetmaker> and needs gauging
21:46:22 <Eddi|zuHause> energy, similar to distance, is a unit of differences, the offset is irrelevant there
21:46:44 <Eddi|zuHause> basically all vectorial units have that
21:46:58 <Eddi|zuHause> it's the very nature of a vector
21:47:17 <planetmaker> energy is a scalar, not a vector
21:47:34 <Eddi|zuHause> depends on what you define as a vector ;)
21:48:01 <planetmaker> not really. Energy is never a vector
21:48:43 <planetmaker> you may calculate with the energy-momentum tensor. But that's a matrix rather
21:49:03 <Eddi|zuHause> a vector is a member of a set that has operations vector+vector and scalar*vector defined. clearly energy has that ;)
21:49:25 <planetmaker> no. As there's no vector
21:49:39 <Eddi|zuHause> but in comparison, temperature+temperature or pressure+pressure doesn't make a lot of sense
21:50:14 <planetmaker> with your definition of vector "1" is also a vector
21:50:33 <planetmaker> which is ... not true according to the usual algebra
21:50:57 <Eddi|zuHause> yes, 1 (of set R) is a member of the vector-space R^1
21:51:32 <Eddi|zuHause> obviously, any field is in itself also a vector-space
21:52:51 <Eddi|zuHause> it's perfectly legal for "1" to be both scalar and vector at the same time ;)
21:53:26 <planetmaker> and it's similarily pointless
21:53:49 <Eddi|zuHause> only if you have no clue of maths ;)
21:54:11 *** andythenorth has joined #openttd
21:55:03 <planetmaker> sure. Good that you know then how to treat energy as a "vector". Which is physically nonsense
21:55:17 <Eddi|zuHause> you'll have to do away with the interpretation that "vector" is "something like a direction"
21:57:35 <planetmaker> you're forgetting that you argue physics - where energy is not useful to treat as vector
21:58:08 <planetmaker> thus you have to use actually something more than mathematical deduction ab initio
21:58:38 <Eddi|zuHause> no, i forgot that i was discussing with a physicist :)
21:59:05 <Eddi|zuHause> (and didn't make my nomenclature clear in the beginning)
21:59:50 <Eddi|zuHause> let's just conclude that your idea of "vector" and my idea of "vector" don't quite match
22:00:03 <Eddi|zuHause> and thus i meant something different than you understood
22:00:28 <planetmaker> I know well the mathematical definition of vector. It simply does not apply to the context you used it
22:01:50 <planetmaker> saying a scalar quantity is a vector may be mathematically correct - but is physically nonsense
22:03:38 <Eddi|zuHause> what is the core of a vector (in my sense above) was that there is no defined "origin"
22:07:14 <Eddi|zuHause> what may make this distinction clear is the concept of time. there is a defined origin called "birth of christ" (even if that does not really match the actual event). and there are absolute times (like 29.8.2006 0:07) and relative times (2 minutes 3 seconds)
22:07:34 <Eddi|zuHause> the relative times are "vectors" in the mathematical sense
22:08:03 <Eddi|zuHause> in the sense that you can freely add and substract them, or scale them
22:08:31 <Eddi|zuHause> but you cannot add absolute times, or scale them
22:09:20 <Eddi|zuHause> (you can substract absolute times, and get a relative time, though)
22:10:05 <Eddi|zuHause> (mathematically, the combination of absolute and relative times is an affine space)
22:10:21 * __ln__ 's Theorem: a vector is a more or less straight line drawn on paper, with an arrow in one end
22:10:22 <planetmaker> with that definition everything is a vector
22:11:22 <planetmaker> and... time is actually one of the worst things to speak of when using "absolute" in the same sentence
22:11:58 <planetmaker> your GPS would fail, if there was anywhere an absolute time
22:13:03 <planetmaker> temperature has an absolute zero, though
22:13:08 <Eddi|zuHause> ok, when you add the relativistic element, things get complicated :p
22:15:04 <Eddi|zuHause> planetmaker: the same argument holds with absolute temperatures and temperature-differences as well
22:15:48 <planetmaker> you can't "add" temperature
22:16:08 <Eddi|zuHause> yes, but i can add temperature-differences
22:16:20 <planetmaker> You can add or remove energy
22:16:39 <planetmaker> which may or may not result in a temperature change
22:16:56 <Eddi|zuHause> that is technical blah :)
22:17:11 *** Progman has joined #openttd
22:17:23 <planetmaker> without this distinction time direction would be undefined as it defines entropy
22:17:48 <planetmaker> (around a few corners at least ;-) )
22:18:37 <Eddi|zuHause> in a simplified temperature-model, every material has a caloric constant, which directly scales an energy into a temperature difference
22:18:55 <Eddi|zuHause> this means, temperature-differences and energy are isomorphic vector spaces
22:19:22 <planetmaker> this model is invalid if you want to include the zero-point
22:19:26 <planetmaker> for ALL materials
22:19:26 <Eddi|zuHause> (if you leave out all the crazy freezing stuff)
22:19:41 <planetmaker> even without phase transitions
22:20:56 <Eddi|zuHause> well, a temperature model as simplified as this would have no absolute zero, since that cannot happen in a vector space :)
22:21:41 <Eddi|zuHause> (vector spaces must be defined over a field, which means it must be "open")
22:22:00 <planetmaker> a vector space has a well-defined 0 for the add operation
22:22:32 <Eddi|zuHause> yes. but that is not the same as an "absolute 0"
22:22:45 <Eddi|zuHause> it also has nothing to do with an "origin"
22:23:35 <Eddi|zuHause> planetmaker: the "0" you think of would be a "0 temperature-difference", not a "absolute 0 temperature"
22:24:10 <Eddi|zuHause> confusing the two clearly demonstrates that you still have not understood my point
22:26:33 <planetmaker> so you argue physics, throw away the physical boundaries, simplify to an unphysical, unbound vector space and I have mis-understood your point?
22:27:35 <planetmaker> I'd rather argue you're confusing some things ;-)
22:27:50 <Eddi|zuHause> clearly, as the point was to throw away all the physical nonsense and discuss the underlying mathematical concept
22:28:15 <planetmaker> that was not what we started discussing nor I ever discussed
22:29:03 <planetmaker> the starting point was your statement that energy is a vector. Which remains of dubious quality
22:29:05 <Eddi|zuHause> what i meant to explain was that my concept of "vector" as used in the initial sentence was "anything that can reasonably be described as the difference of two absolute values"
22:30:29 <planetmaker> well. Energy already is not an absolute value as it's subject to the gauge invariance
22:31:49 <planetmaker> (though by all practical terms, the offset can usually be set to 0)
23:01:14 *** supermop__ has left #openttd
continue to next day ⏵
