IRC logs for #openttd on OFTC at 2007-11-03
⏴ go to previous day
00:02:11 *** ThePizzaKing has joined #openttd
00:46:06 *** lolman is now known as John
01:31:31 *** Eddi|zuHause3 has joined #openttd
01:49:51 *** Diabolic1Angel has joined #openttd
01:56:11 *** Diabolic-Angel has quit IRC
02:11:41 *** Maarten has joined #openttd
02:15:54 *** Maarten has joined #openttd
02:48:52 *** Diabolic1Angel is now known as Diabolic-Angel
02:51:11 <BigBB> Markkisen, you are german, right?
02:52:24 <BigBB> I think "Miau" don't understand the people. Miaow for british-english and meow for americ.-english.
03:06:58 *** XeryusTC has joined #openttd
03:26:05 *** Diabolic-Angel has quit IRC
04:10:36 <BigBB> can someone please tell me how is the action for NewGRF to check if TTDP or OTTD?
04:11:32 <BigBB> I think there is one ?!
04:11:41 *** DaleStan has joined #openttd
04:12:41 <BigBB> ah, DaleStan :) can you please tell me how is the action for NewGRF to check if TTDP or OTTD?
04:12:59 *** Gonozal_VIII has joined #openttd
04:51:42 *** dotnine has joined #openttd
05:38:10 <DaleStan> BigBB: Action 7, var 9D.
06:06:12 <Jello> can i buy out other buisnesses even if there not bankrupt?
06:08:02 *** Deathmaker has joined #openttd
06:17:55 <Gonozal_VIII> multiplayer/singleplayer
06:20:48 <CIA-1> OpenTTD: rubidium * r11370 /trunk/src/order_cmd.cpp: -Fix [FS1384]: manually replacing a vehicle with shared orders makes it lose it's order index and service interval. Patch by jthill.
06:33:13 <Jello> where do i go to buy it out?
06:34:14 <Rubidium> company window -> buy shares till you've got all of them
06:34:55 <Rubidium> maybe you need to enable the patch option that makes you able to take over companies (somewhere under economy I guess)
06:35:17 <huma> Rubidium, you're on linux?
07:04:50 *** gfldex_ has joined #openttd
07:12:15 *** ThePizzaKing has joined #openttd
07:28:58 *** Progman has joined #openttd
07:52:37 *** John is now known as lolman
08:31:52 *** |Bastiaan| has joined #openttd
09:23:47 *** dihedral has joined #openttd
09:33:58 *** Alltaken has joined #openttd
10:15:38 *** egladil has joined #openttd
10:17:47 <Markkisen> BigBB, no I'm swedish
10:19:01 <Gonozal_VIII> ? i don't see bigbb here
10:20:44 <Gonozal_VIII> i thought there was a problem with my irc client or something with you answering a question i couldn't see
10:21:23 <Markkisen> He ased it 3:51AM (UTC +1)
10:21:47 <Markkisen> Wait, isnt that CET?
10:22:12 <Gonozal_VIII> [07:00:05] *** BigBB has signed off IRC (Remote host closed the connection).
10:25:46 <Wolf01> stop turning your head in the wrong direction, you make me have pain in my neck
10:26:03 <TrueBrain> yeah, I can't rotate my head 180 degrees
10:26:22 <TrueBrain> (you are truly internet addicted, when you rotate your head 90 degrees to smile)
10:26:40 <TrueBrain> (and then you should turn your head, and smile, to whoe ver you tell it)
10:28:29 <Wolf01> anybody here used to eggdrops (maybe you TrueBrain ;) )
10:29:00 <Wolf01> i can't make mine working, it doesn't accept any ctcp chat or dcc chat session
10:29:12 <TrueBrain> I made it to ignore those 2 :p
10:29:19 <TrueBrain> but dcc mostly is a firewall issue
10:29:33 <TrueBrain> (it might require incoming signal)
10:32:28 <Gonozal_VIII> i think i wrote some script stuff years ago but can't remember because my memory is... what did i want to say? :S
10:51:12 *** Farden123 has joined #openttd
10:57:44 *** Farden123 is now known as Farden
11:04:00 <Markkisen> No, time to go home (:
11:08:34 *** LeviathNL has joined #openttd
11:10:44 *** Farden123 has joined #openttd
11:15:00 <Wolf01> good, now works, now i should only find an talian language for the chatbot
11:15:13 <Wolf01> and maybe teach him some infos about ottd
11:17:39 *** Farden123 is now known as Farden
11:23:49 *** Gonozal_VIII has joined #openttd
11:34:58 *** bluebear____ has joined #openttd
11:35:05 *** bluebear____ is now known as bluebear
11:43:11 *** |Bastiaan| has joined #openttd
11:49:36 *** bluebear____ has joined #openttd
12:02:33 *** dexterl1 has joined #openttd
12:05:52 *** dexterle has joined #openttd
12:27:55 *** Diabolic-Angel has joined #openttd
12:28:34 *** Ammlller has joined #openttd
12:31:04 *** Farden123 has joined #openttd
12:33:56 *** dexterle has joined #openttd
12:34:11 <SpComb> hmm... the number of active servers just went down drastically
12:34:16 <SpComb> I wonder if that was supposed to happen
12:34:16 *** HerzogDeXtEr has joined #openttd
12:34:27 *** HerzogDeXtEr has joined #openttd
12:36:24 *** ThePizzaKing has joined #openttd
12:37:54 *** Farden123 is now known as Farden
12:56:00 *** bluebear_____ has joined #openttd
12:56:05 *** bluebear_____ is now known as bluebear
13:05:45 <SpComb> but the point is, MyOTTD isn't dead, even though I haven't touched it for ages!
13:08:55 <TrueBrain> I assume that with 'ages', you mean game-years?
13:12:59 <SpComb> ages as in three weeks
13:18:19 *** Wolf01 is now known as Guest340
13:25:29 <Ailure> the online hosting service
13:25:57 <Ailure> wouldn't it be tricky to make sure that like
13:26:06 <Ailure> a hosted server won't take all avaible CPU time on the server
13:27:49 <SpComb> Ailure: yes, you'd need to figure out what configuration settings influence the CPU usage
13:33:44 <Ailure> funny how a game that that was orginally made to run on 486 computers
13:33:56 <Ailure> can run so slow on the most modern computers with the wrong configuration :)
13:34:03 <Ailure> well, it can be mostly blamed on the newer pathfinders heh
13:34:25 <Ailure> especially when ou use them on the ships
13:34:44 <SpComb> there's one process that's using a fair amount of CPU
13:37:17 <Ailure> the most processor intesive game I played
13:37:39 <Ailure> was that wwottdgd game
13:38:14 <Ailure> and while my processor hardly got to reach it's max, I noticed that the longer the game went on, the less people were able to play on it
13:40:52 *** Wolf01 is now known as Guest341
13:43:19 *** Wolf01 is now known as Guest342
13:52:28 *** Alltaken has joined #openttd
13:57:45 *** Vikthor has joined #openttd
13:58:41 *** Greyscale has joined #openttd
13:59:35 <Eddi|zuHause3> <Ailure> and while my processor hardly got to reach it's max, I noticed that the longer the game went on, the less people were able to play on it <- the server must be the slowest machine
13:59:56 <Eddi|zuHause3> the slower the server, the bigger the chance that the clients can keep up
14:00:14 <Ailure> that's how some online games works actually
14:00:20 <Ailure> the game slows down to the slowest computer
14:00:25 <glx> that way the game is laggy for clients, but they can still play
14:31:59 *** lolman is now known as John
14:32:09 *** John is now known as lolman
14:55:51 <MarkSlap> Where the f*ck is Bjarni? :D
14:56:16 <MarkSlap> He didn't like that I talked about drugs (Especilly acid) :>
14:56:24 <TrueBrain> I am seriously not interested in his fucking or not
15:00:12 <TrueBrain> yeah, that was what I was thinking
15:00:44 <MarkSlap> I dont understand it, so sure :>
15:03:42 <Eddi|zuHause3> see, that is because of the drugs
15:13:44 *** Dark_Link^ has joined #openttd
15:19:06 *** mikk36|work has joined #openttd
15:21:39 <CIA-1> OpenTTD: rubidium * r11371 /trunk/src/industry_cmd.cpp: -Fix [FS#1396]: the industry protection only kicked in when it should not kick in.
15:32:19 <CIA-1> OpenTTD: glx * r11372 /trunk/src/industry_gui.cpp: -Fix [FS#1394]: don't stop on first invalid cargo type when displaying industry acceptance
15:36:07 *** Brianetta has joined #openttd
15:45:39 <_42_> yorick, dihedral (~dihedral@p54A0DECE.dip.t-dialin.net) was last seen quitting #openttdcoop 6 hours 11 minutes ago (03.11. 09:34) stating "Remote host closed the connection" after spending 10 minutes there.
16:07:07 <TrueBrain> want to see something really cool?
16:08:05 <TrueBrain> just don't get too close to the border, it fucks things up ;)
16:33:00 *** redmonkey has joined #openttd
16:33:16 <Ailure> I'm trying to figure out what that does
16:33:48 <valhallasw> TrueBrain: it's not loading? :(
16:34:08 *** skidd13 has joined #openttd
16:34:09 <Phazorx> it makes a nice tile map
16:34:51 <Ailure> probably just strained
16:57:35 <TrueBrain> it should really be pretty fast
16:57:40 <TrueBrain> just it requires a lot of a browser :p
17:01:14 <Gonozal_VIII> what am i loading there?^^
17:01:24 <SpComb> TrueBrain: it's not working
17:01:47 <Gonozal_VIII> well, the loading has stopped
17:02:51 * Phazorx used opera and did not have issues, it loads pretty fast
17:04:24 <SpComb> I mean, it's not loading, at all
17:05:07 <Gonozal_VIII> is it supposed to be a blank page?
17:05:09 <fjb> Did anything break newindustries after r11362?
17:06:05 <Gonozal_VIII> didn't see any problems with ecs
17:06:40 <TrueBrain> SpComb: it was stuck :p Try again :)
17:07:00 <MarkSlap> AND NOW the acid has been kicked in
17:07:42 <Gonozal_VIII> there's a part on the left side that looks funny
17:07:46 <fjb> Hm, when I create a new game in r11367 or 11370, then the ECS factory, ECS textile factory and the ECS vehicle factory don't accept anything.
17:08:34 <Gonozal_VIII> do you have the new grf?
17:08:51 <Gonozal_VIII> 2 days old or something
17:09:19 <fjb> Yes, the new machinery vector
17:09:21 <Gonozal_VIII> those buildings changed a lot
17:09:24 <SpComb> TrueBrain: now it's managed to load itself, but the frame thing isn't loading
17:09:27 <Eddi|zuHause3> hm, the page is very screwy in konqueror
17:09:28 <Gonozal_VIII> but they work with me
17:09:32 <fjb> And all ECS grfs loaded in id order.
17:09:32 <TrueBrain> SpComb: enable JavaScript ;)
17:09:43 <TrueBrain> and it needs to load a 100 KiB library(prototype)
17:09:45 <TrueBrain> that might take a while...
17:09:46 <SpComb> it just says "Loading..."
17:09:49 <SpComb> for a minute or so now
17:10:07 <SpComb> right, now I managed to get it
17:10:16 <SpComb> yay mouse-draggable map
17:11:01 <SpComb> uh oh, I scrolled to the edge of the map :(
17:11:08 <TrueBrain> I need to fix that :)
17:12:00 <Eddi|zuHause3> hm, i should not scroll too fast
17:12:17 <Gonozal_VIII> i have some white tiles now
17:12:17 <Eddi|zuHause3> where "fast" is not the right description
17:12:18 <SpComb> does it do a request for every tile or something?
17:12:27 <TrueBrain> Gonozal_VIII: don't go near the border ;)
17:12:32 <TrueBrain> SpComb: no, per line
17:12:38 <TrueBrain> so 40 tiles at once
17:12:41 <Gonozal_VIII> they are in the middle
17:12:45 <Eddi|zuHause3> TrueBrain: when i move, the next row of tiles is flat first, and only slowly builds up to the real heights
17:12:46 <TrueBrain> Gonozal_VIII: that happens, an update failed
17:12:59 <TrueBrain> Eddi|zuHause3: see it as fast moving in Google Maps
17:13:03 <TrueBrain> you first get a grey sprite
17:13:04 <Gonozal_VIII> no new request for those tiles?
17:13:05 <TrueBrain> then it loads the real thing
17:13:11 <TrueBrain> Gonozal_VIII: not yet, no
17:13:20 <TrueBrain> the code is there already, just didn't finished it yet :p
17:14:25 <TrueBrain> it is just a first version ;)
17:14:59 <Eddi|zuHause3> TrueBrain: occasionally, i get a white row instead of a row of flat tiles
17:15:15 <Gonozal_VIII> and for stupid people like me... what's different to ingame? is that 3d?
17:15:45 <TrueBrain> Eddi|zuHause3: only looks that way: the flat tiles are on height 0, and you look at a hill, so it appears white, as you look into the hill
17:15:48 <TrueBrain> the flat tiles are there :p
17:16:17 <Eddi|zuHause3> it looks quite disturbing though
17:16:37 <TrueBrain> Eddi|zuHause3: shit happens ;) It will update itself in the end... when ever I finish that part :p
17:18:24 <Gonozal_VIII> i don't really understand what it does but looks nice^^
17:19:07 <fjb> Can anybody reproduce it? Some ECS factories don't accept anything when the game got createt in r11367 or r11370. If the game got created with r11362 they accept the the usual things, even when the save gets loaded into r11367.
17:20:11 <Gonozal_VIII> hmmm i started the last game with 64..
17:20:28 <Wolf01> TrueBrain, how i can set up an eggdrop to reply on some commands (and how to add them) in chan instead of dcc chat only, like !seen !logs etc
17:20:38 <fjb> Gonozal_VIII: Can you create a new game and see what happens?
17:21:41 <fjb> Gonozal_VIII: What happened?
17:21:51 <Gonozal_VIII> the industry says it accepts but the stations don't
17:22:15 <Wolf01> too muck material on industry stockpile
17:22:25 <fjb> Gonozal_VIII: Click on the Industry with the questionmark. No tile accepts anything.
17:23:26 <Gonozal_VIII> seems he forgot to place the tiles that accept stuff
17:24:10 <fjb> Yes, but it works with the same ECS grfs in r11362 .
17:24:25 <TrueBrain> Wolf01: dunno, it was too long ago for me
17:24:32 <TrueBrain> Gonozal_VIII: it is a TT landscape in a webbrowser ;)
17:26:12 <Gonozal_VIII> i didn't have much time to play the last days, i enjoyed the new looks of those industries and that they seemed to have new stockpiling ability but did not actually play with that
17:26:53 <Gonozal_VIII> so i don't know if it was broken before or not
17:27:37 <fjb> The new industries are great. And the are working for me when the game got created in r11362.
17:27:38 *** lolman is now known as John
17:27:51 *** John is now known as lolman
17:29:02 * TrueBrain is happy with his scrollable map :)
17:29:30 <fjb> TrueBrain: Can you give me the URL of your new masterpiece please?
17:31:42 <fjb> YOu enhanced it from a few days ago?
17:31:52 <TrueBrain> rewrite from scratch, to be exact
17:32:06 <TrueBrain> bah, it is stuck again...
17:32:16 * TrueBrain restarts the server... I have to find out what is causing the lockups...
17:32:40 <Gonozal_VIII> that's really nice and such truelight but isn't that feature in the game since, like... 12 years?
17:33:00 <TrueBrain> Gonozal_VIII: there is kind of a big difference between a standalone GUI and a webbrowser
17:33:21 <TrueBrain> fjb: you can now clck on the map, and drag around
17:34:14 <Gonozal_VIII> <-- too n00b to understand that
17:34:49 <TrueBrain> Gonozal_VIII: basicly, this allows you to use any browsers anywhere in the world to login to this virtual world
17:35:09 <TrueBrain> no need to download, and things like 'max clients' is kind of different ;)
17:36:17 <Gonozal_VIII> soooooo you could make something like a 64k^2 map with 1k players at the same time?
17:36:20 <TrueBrain> it is a proof-of-concept of mine, to show that it is possible to do
17:36:25 <TrueBrain> Gonozal_VIII: for example, yes
17:36:37 <TrueBrain> AttributeError: 'HTTPConnection' object has no attribute 'rfile' <- lol, nice crash...
17:36:53 <TrueBrain> clearly the server has some stability issues :)
17:39:23 <Gonozal_VIII> openttd the browsergame :D
17:39:24 <TrueBrain> Gonozal_VIII: if you check the time in which the pages are generated, and know a thing or 2 about dynamic pages, you will see that it is VERY fast; it is a combination of several modern techniques, to show nowedays it can be done to serve over a 1000 clients, while having a real-time game running, via a webbrowser
17:45:19 <fjb> r11364 is already broken...
17:46:40 <Gonozal_VIII> buuut in a real game you often need more than just a small part of the map
17:49:17 <TrueBrain> Gonozal_VIII: only the server does ;)
17:49:19 <Gonozal_VIII> well... i have work to do... java coding... lalala.. n00b anyways
17:49:29 *** XeryusTC has joined #openttd
17:50:22 <fjb> Oh, ECS is not broken, the newly generated industried just need some days till the accept anything. But that is the case only with some industries.
17:51:01 *** michi_cc has joined #openttd
17:51:01 *** ChanServ sets mode: +v michi_cc
17:54:14 *** Leviath has joined #openttd
17:58:27 <fjb> Does the generatet landscape in TTDP look that different from a landscape in OpenTTD?
18:02:47 <TrueBrain> not if you use the old landscape generator ;0
18:13:03 *** Ammlller has joined #openttd
18:22:52 *** Dephenom has joined #openttd
18:26:19 *** skidd13 has joined #openttd
18:28:50 <fjb> How does a landscape look like that got generated by the old generator?
18:30:03 <skidd13> fjb: try yourself and compare :P
18:30:45 <Eddi|zuHause3> in TTO, landscapes were just very few basic hill layouts that were assembled to a map
18:31:24 *** ChanServ sets mode: +v tokai
18:31:51 <Eddi|zuHause3> i don't know how far that changed in TTD
18:33:17 <fjb> Oh, the world looks strage now, but ECS industries get a lot faster placed. :-)
18:36:12 <fjb> The new landscapes are looking much better. Hope we get rivers one day.
18:37:26 <hylje> sealevel stuff will get there
18:38:20 <fjb> That means tunels uder the sea?
18:39:05 <Gonozal_VIII> chrisin has that
18:39:42 <Gonozal_VIII> but it sounds fancier than it is.. long tunnels are not very usefull anyways
18:40:37 <fjb> ChrisIN gets really slow at the end of the game, and my computer is not thar slow (Athlon64 @ 2GHz).
18:41:28 <Gonozal_VIII> well... the last games we played were 256^2 that runs fast...
18:42:26 <Gonozal_VIII> with ecs that is enough for long hours of connecting all the industries
18:43:00 <MarkSlap> What is #openttdcoop?
18:43:22 <Gonozal_VIII> exactly what it sounds like
18:43:46 <fjb> 256 is too small for realistic trains...
18:44:02 <Gonozal_VIII> realistic trains are?
18:44:21 <hylje> more than 256 tiles long
18:45:24 <fjb> And it's not realistic that one station touches the next one.
18:45:46 <Gonozal_VIII> well... where i live i hardly ever see a train with more than 5 waggons
18:45:47 *** Kamikaz333 has joined #openttd
18:46:41 <fjb> You didn't see real coal trains then. Goods trains ususlly have more than 100 axles.
18:48:09 <Gonozal_VIII> they don't send many freight trains here, it's a small line
18:48:32 <MarkSlap> hylje and Gonozal_VIII :o
18:48:43 <fjb> Only small trains on small lines of course, but it is different on the main lines.
18:49:18 <Gonozal_VIII> mostly passenger trains with three wagons and a driver thingy at the back so they don't have to turn around
18:49:35 <hylje> no multiple unit stuff?
18:49:43 <Tefad> landscape in old generator is a bunch of random rectangles
18:50:01 <fjb> On a small line you will hardly ever see that.
18:50:26 <Tefad> landscape in new generator looks like random gauss turned into heightmap.
18:50:50 *** lws1984 has joined #openttd
18:51:04 <Tefad> where i live trains have hundreds of cars
18:51:09 <lws1984> Anyone here having issues with OpenTTD and OS X 10.5?
18:51:10 <Tefad> and three or more engines
18:51:15 <Eddi|zuHause3> <hylje> no multiple unit stuff? <- i have never seen any term closely resembling "multiple unit" being used in german train vocabulary
18:51:46 <Tefad> and they usually have a bunch of tractor trailers and or shipping containers
18:52:03 <hylje> what do you call those complete trains then?
18:52:22 <hylje> as opposed to loco-wagons
18:52:44 <Eddi|zuHause3> well, you have the term "Triebwagen"
18:52:49 <Tefad> i don't see much raw materials on the lines around here, except coal
18:53:15 <Eddi|zuHause3> which means a self driving wagon
18:53:33 <Eddi|zuHause3> occasionally "Schienenbus" (=railcar?)
18:53:46 <Gonozal_VIII> they sometimes use very overpowered engines here... "taurus" pulling 3 passenger cars and such
18:54:05 <hylje> isnt "taurus" a maglev loco in ttd?
18:54:14 <Tefad> they have interesting cars that hold tall objects (i've never seen one loaded) with warning label "do not unload one side all at once"
18:54:18 *** michi_cc has joined #openttd
18:54:18 *** ChanServ sets mode: +v michi_cc
18:54:45 <Gonozal_VIII> it's the strongest electric engine in the dbset
18:56:19 <Eddi|zuHause3> hylje: anyway, a distinction between single self driving units and multiple self driving units is kinda artificial in my eyes
18:57:28 <lws1984> Belugas? (If I remember correctly, you're the resident Mac expert)
18:57:43 <Eddi|zuHause3> 182? i don't remember that one, when does it come out?
18:58:23 *** XeryusTC has joined #openttd
18:59:34 <Gonozal_VIII> i use them on almost every train in the late game
18:59:59 *** Sacro` is now known as Sacro
19:00:19 <Eddi|zuHause3> i think i never really did a late game
19:01:48 <Gonozal_VIII> a game is so short without daylength patch...
19:01:51 <skidd13> lws1984: IIRC it's Bjarni not Belugas :D
19:02:17 <Sacro> yes, Belugas is the residient diving canadian
19:03:12 *** Kamikaz333 has left #openttd
19:09:08 <Gonozal_VIII> i used to play with 14 part trains, just because that station lenght was easy to build, later i just disabled very long trains and made them old max length
19:09:57 <lws1984> and bjarni isn't here. :(
19:10:12 <_42_> lws1984, Bjarni (~Bjarni@0x50c79a03.virnxx14.adsl-dhcp.tele.dk) was last seen quitting #openttd.notice 2 days 17 hours 33 minutes ago (01.11. 01:36) stating "Quit: zzz" after spending 4 hours 16 minutes there.
19:10:23 <Gonozal_VIII> more ticks per day
19:10:36 <Eddi|zuHause3> increase the length of days, so you can play longer in a certain era (like steam)
19:11:07 <Ailure> what I like the idea with something like the daylength patch
19:11:15 <Ailure> is that it would allow a game to go on for a month
19:11:29 <Ailure> so you can like, play once a hour or something heh
19:11:29 <Gonozal_VIII> max factor is 30
19:11:48 <Eddi|zuHause3> well, there are some disputes about the actual effect of the patch
19:12:05 <Gonozal_VIII> i think that's somewhere around a minute per day... 66 seconds or something
19:12:15 <Ailure> there's been severeal versions
19:12:37 <Tefad> does this affect vehicle speeds?
19:12:38 <Eddi|zuHause3> one type wants to simply make the game slower (industrial development, train speeds, payment rates)
19:12:41 <Gonozal_VIII> breakdowns are not affected
19:12:59 <Tefad> breakdowns piss me off, i generally disable them : x
19:12:59 <Eddi|zuHause3> and the other type simply wants the same game with slower advancing dates
19:13:18 <Eddi|zuHause3> the current patch achieves the second version
19:13:34 <Tefad> no matter how many depots i have, they always manage to block in important route for an extended time period.
19:13:39 <Ailure> I would prefer the game slower heh
19:13:54 <Ailure> which would mean you could spend a bit on building a network
19:13:59 <Ailure> then leave for a few hours and come back
19:14:05 <Ailure> without people having catched up too much :p
19:14:16 <Eddi|zuHause3> yeah, i would prefer that also, but i could not really convince any of the different coders that previously attempted a daylength patch
19:14:32 <Tefad> aww i kind of liked pre-r11366 behavior
19:14:48 <Ailure> sounds like something I would code in
19:15:16 *** gono_ping_timeout has joined #openttd
19:15:17 <Ailure> make income the same, but costs multiplied by the daylength multiplier
19:15:41 <Eddi|zuHause3> well, the difficulty in this is finding a system as to which game rules should be tick-based (keep current speed) or day-based (make slower)
19:16:12 <Ailure> or at least tweaked so a train would make as much money in 1 year on 1x setting compared to 1 year on 30x setting
19:16:24 <Ailure> the game is pretty much tick-based too
19:16:57 <Ailure> and some things should stay that way too
19:17:01 <Ailure> such as the movement of vehicles
19:17:16 <gono_ping_timeout> 30 times slower ships^^
19:17:22 <Ailure> otherwise it would look ridicolus at 30x
19:17:22 <Eddi|zuHause3> well, Celestar said he would put something like that into the game balance branch, but then he dropped out
19:17:33 <Ailure> with a maglev train going 20 km/h
19:18:10 <Ailure> It's something I would want, since it might improve multiplayer
19:18:34 <Eddi|zuHause3> well, there it gets difficult, if you don't want to modify speed, this means a train arrives 30 times as often on 30x daylength
19:19:01 <Ailure> that's not a problem in itself
19:19:28 <gono_ping_timeout> it also has 30 times more running cost
19:19:33 <Ailure> and same for industry logic
19:20:09 <Eddi|zuHause3> gono_ping_timeout: which is stupid, because running costs are supposed to be year based
19:20:12 <Ailure> it's running cost should be less in terms of ticks
19:20:16 <Ailure> but the same in terms of ingame days
19:20:28 <gono_ping_timeout> well but they are in chrisin
19:20:36 *** gono_ping_timeout is now known as Gonozal_VIII
19:20:51 <Ailure> I think that's a daylength patch
19:20:57 <Ailure> where the economy isn't really affected
19:20:59 <Eddi|zuHause3> Gonozal_VIII: which is exactly the problem
19:21:20 <Gonozal_VIII> they are year based, the values are 30 times higher
19:21:42 <Eddi|zuHause3> Gonozal_VIII: which is again exactly the problem
19:22:17 <Ailure> is that a train on the same route, with same engine
19:22:26 <Ailure> should earn as much on 1x setting as on 30x setting
19:22:29 <Eddi|zuHause3> the value should stay constant independent on daylength, because it is running cost per year, and the vehicleset designers chose them for a realistic aspect
19:22:37 <Ailure> without the vehicle being slowed down
19:23:24 <Gonozal_VIII> then let the industries produce the same ammount, not 30 times more, then the train can only make one tour :-)
19:23:38 <Eddi|zuHause3> it is totally stupid if a vehicle costs 3 times more to run for a year than to buy
19:25:07 <Tefad> how hard would it be to just lengthen a tick
19:25:20 <Tefad> to slow the whole thing down
19:25:34 <Eddi|zuHause3> Tefad: this is not wanted, because the trains should still (visibly) move at the same speed
19:25:36 <Tefad> i know there's the ">>" button which is like simcity african swallow
19:25:47 <Tefad> do the opposite of that
19:26:16 <Tefad> opposite..converse whatever
19:26:17 <Eddi|zuHause3> Tefad: this is easy, just turn on YAPF for ships ;)
19:26:36 <Eddi|zuHause3> or better, NPF ;)
19:26:52 <Tefad> there's a bunch of off-by-one problems in gui window placement btw
19:26:55 <Tefad> unless it's intentional
19:27:39 <Tefad> for some reason, when there's a 1px gap in gui elements, my cpu cycles go way up
19:28:09 <Tefad> (1px is example, i'm guessing it occurs when there's any gap)
19:28:38 <Tefad> does anyone else have this same observation?
19:29:26 <hylje> probably something with the way its drawn
19:31:12 <Tefad> it's off by two actually
19:31:20 <Tefad> it's the vehicle window placement
19:31:52 <Tefad> hell window placement in general seems pretty hosec
19:32:32 <Eddi|zuHause3> hm, funny, i have 3 recordings of reruns of the same show distributed over a year, each with two commercial breaks, the first commercial break is always at the same spot, and the second break always at a different spot...
19:34:09 <Gonozal_VIII> ... why would somebody analyse that?
19:35:05 <Eddi|zuHause3> well, i want to cut the 3 different airings into one recording that does not have any interruptions
19:35:31 <Gonozal_VIII> you only need one to do that?
19:36:07 <Eddi|zuHause3> not for the "popup" stuff
19:36:28 <Eddi|zuHause3> which is often accompanied by a loud noise
19:37:04 <Gonozal_VIII> i hate that kind of commercial
19:38:48 <Gonozal_VIII> at least sometimes there are good films in austrian tv... free of commercials... but only sometimes
19:40:18 <Eddi|zuHause3> well, it's technically not a commercial, there are 3 kinds of "popups", one is the show name after each commercial break, or the name of the next show, one is a popup like "buy DVDs of this show on [blah]" or "get this melody as ringtone" and the most annoying kind is preview popups like "series XY starts tonight at 20:15"
19:42:47 <Eddi|zuHause3> although i did notice that the 3rd kind reduced over the past few months
19:43:27 <Eddi|zuHause3> there were times where you got two of these in every 1h show
19:43:42 <Eddi|zuHause3> sometimes announcing shows a week in advance
19:44:13 <Eddi|zuHause3> meaning you got to see the same stupid interruption with the same loud noise over the course of one week
19:45:01 <Eddi|zuHause3> now it's only occasionally, when a "big event" is upcoming
19:46:11 <Eddi|zuHause3> on the other hand, the second kind used to be without noise, but now they added a "beep"
19:46:57 <Gonozal_VIII> you really pay a lot of attention to those commercials...
19:47:54 <Gonozal_VIII> i watch lots of german tv stations but i have no idea what you're talking about^^
19:48:01 <Eddi|zuHause3> well, you usually do not pay attention when watching a live show
19:48:29 <Eddi|zuHause3> but when you have recordings, especially different recordings of the same show in parallel, you easily notice that stuff
19:52:31 <Eddi|zuHause3> Gonozal_VIII: clearly someone notices when he wants to watch a show, but the whole picture is rotating because they are doing an announcement that spiderman is playing next sunday
19:57:31 <Eddi|zuHause3> which is even more annoying if that announced spiderman airing was over a year ago
20:01:38 *** Greyscale has joined #openttd
20:05:03 <Ailure> it's even more weird when you have downloaded it over internet
20:05:28 <Ailure> and live in a way diffrent country
20:11:34 <Rubidium> lws1984: I wouldn't expect much "help" from Bjarni as he kinda refuses to install Leopard on his system.
20:11:55 * lws1984 is using it in Parallels anyway, so no matter
20:12:40 <Rubidium> fjb: you say ECS starts behaving wrongly between r11362 and r11364?
20:13:19 <Gonozal_VIII> no it was only some kind of construction state thingy
20:14:14 <Gonozal_VIII> [18:50:21] fjb: Oh, ECS is not broken, the newly generated industried just need some days till the accept anything. But that is the case only with some industries.
20:21:31 *** LordAzamath has joined #openttd
20:22:15 <fjb> Yes, I just didn't notice that behavior before. But george confirmed that that can happen, so no fault of OpenTTD, just a bit confusing.
20:33:38 <Phazorx> is it possible to calculate "proper income" knowing current and starting year with inflation being on ?
20:33:57 <Phazorx> is there a formula somewhere how inflation is calculated?
20:34:40 *** UnderBuilder has joined #openttd
20:35:33 <Eddi|zuHause3> from a theoretical point of view, inflation is a geometric series
20:36:29 <Eddi|zuHause3> assuming constant inflation rate q, the effective price after n years is p*(1+q+q^2+q^3+...+q^n)
20:37:31 <Eddi|zuHause3> you can expand this sum to (1-q^(n+1))/(1-q)
20:37:36 <Gonozal_VIII> that's why i always play with inflation off... too much math
20:38:22 <UnderBuilder> we should introduce a campaign for increasing the amount of nightly game servers
20:38:47 <Eddi|zuHause3> note, if the inflation rate is 3%, q=1.03
20:41:33 *** lolman is now known as John
20:41:40 *** John is now known as lolman
20:53:38 *** lolman is now known as John
20:53:53 *** John is now known as lolman
20:54:47 *** lolman is now known as John
20:54:55 *** John is now known as lolman
20:59:09 <Gonozal_VIII> wasn't there something with 89?
20:59:55 <DorpsGek> Sacro: I have not seen JohnUK89.
21:00:05 <Gonozal_VIII> wow, my brain works
21:03:23 <TrueBrain> we really need a better patch-tracker...
21:04:39 <hylje> think i could take up mine again
21:04:43 <TrueBrain> Sacro: sucks even worse
21:05:20 <hylje> ill see to it, no promises though :(
21:05:44 <Gonozal_VIII> what about a forum section where every patch has a topic and no other topics are allowed?
21:06:35 <TrueBrain> Gonozal_VIII: not useful either
21:06:43 <TrueBrain> what I miss is to see if I replied to a patch
21:06:46 <TrueBrain> and if there is a new version
21:07:03 <TrueBrain> what hylje designed, on my draft (:p), is a system to do that, even allowing you to mark when a comment is done
21:11:29 <TrueBrain> maybe a maillist like ffmpeg
21:11:31 <TrueBrain> that is pretty useful
21:12:07 <TrueBrain> I liked it on ffmpeg, although a lot is coming through
21:12:26 <TrueBrain> but it was simple, as developer you could see really fast what had a reply and what not..
21:12:34 <TrueBrain> downside is that patches tend to get lost when not applied fast
21:15:15 *** blathijs_ is now known as blathijs
21:16:02 *** skidd13 has joined #openttd
21:17:32 <skidd13> Does anyone can tell me something about the performance of _ffb_64[...]?
21:17:47 <TrueBrain> not even what it does :)
21:19:00 <skidd13> TrueBrain: It' used in the KillFirstBit and FindFirstBit functions.
21:19:08 *** DaleStan_ has joined #openttd
21:27:07 <skidd13> TrueBrain: A profiling session with only one compiler is not representative IMO. There should be tests with the common compilers (windows, linux, OS X).
21:27:27 <TrueBrain> skidd13: agree'd. But a profile of one compiler really is a good start ;)
21:27:32 <TrueBrain> what brings you here? :)
21:27:38 <TrueBrain> (in other words: nice to see you1)
21:27:56 <ludde> i just felt like coming by and saying hello
21:28:06 <TrueBrain> well: hello back to you :)
21:28:10 <TrueBrain> how are you these days?
21:28:43 <skidd13> TrueBrain: what profiling tool do you suggest?
21:28:53 <TrueBrain> skidd13: I always use gprof, but any will do I guess
21:29:21 <ludde> i'm good, now that I quit utorrent I work with Spotify instead.
21:29:32 <ludde> it's not released yet, but it's a music player
21:30:23 <TrueBrain> as always, lightweight ;) Not bad!
21:30:42 <Tefad> so bascially itunes replacement?
21:31:09 <ludde> you don't play mp3 from your disk. but from a central server
21:31:21 <ludde> so you can find any song and play it instantly
21:31:28 <TrueBrain> ludde: and how much is it going to cost? :)
21:31:41 <Tefad> so it's ondemand streaming?
21:31:46 <ludde> well, you don't really care about what format it is.. just click on a song and it starts playing faster than it would if you were using itunes locally :)
21:32:13 <ludde> TrueBrain: it's free, but then you get ads every 30 mins or so.
21:32:25 <TrueBrain> so infinite free music, for some ads?
21:32:38 <Tefad> so basically it's like radio..
21:32:40 <TrueBrain> sounds hard to believe ;)
21:32:40 * Smoovious touches the brim of his black hat to ludde
21:32:52 <TrueBrain> but would be really nice :)
21:32:56 <Tefad> audio ads or visual ads?
21:32:58 <Smoovious> how ya doin', ludde?
21:33:11 <Tefad> anything that spawns browsers?
21:33:13 <ludde> Tefad: it works just like radio ads
21:33:20 <ludde> Tefad: only if you click the ad
21:33:39 <ludde> Smoovious: i'm doing well
21:33:51 <ludde> my back still isn't healed thou :(
21:33:53 <Tefad> is it community based or all behind closed doors?
21:34:09 <Smoovious> oh that sucks... from the train incident?
21:34:42 <ludde> Tefad: we do have beta testing though
21:34:44 <Tefad> audio format is arbitrary?
21:34:48 <ludde> Tefad: but only for swedish ppl
21:35:01 <Tefad> or are you guys locked to one codec
21:35:02 <ludde> Tefad: audio format is mp3 right now, but it's not really something the user notices.
21:35:14 <ludde> Tefad: we're looking into switching to some other format.
21:35:15 <TrueBrain> ludde: Windwos only?
21:35:44 <Tefad> if you guys are worried about license fees, i'm sure you know where to look ; )
21:36:01 * Smoovious is a truck driver now.
21:36:11 <Tefad> even if this product was available for linux i don't see many linux users that would be interested
21:36:20 <TrueBrain> hmm, now a songs popup into mind... tnx Smoovious :p
21:36:21 <Tefad> except those on ubuntu ; )
21:36:30 <TrueBrain> Tefad: and why would that be?
21:36:51 <Tefad> because many linux users are linux users for a reason
21:37:00 <TrueBrain> talking about music... that I didn't put on today!
21:37:04 <TrueBrain> Tefad: to not listen music?
21:37:08 <TrueBrain> now that is a rather silly statement
21:37:25 *** DaleStan_ is now known as DaleStan
21:37:30 <Eddi|zuHause3> i'm linux user because windows annoyed me with constant need of rebooting
21:37:38 <Tefad> i'm not trying to start flamewar : \
21:37:45 <Tefad> i listen to plenty of music
21:37:48 <Eddi|zuHause3> and i am listening to music all the time
21:37:58 <Eddi|zuHause3> like, right now...
21:37:58 <Tefad> however, i stay away from binary blobs as much as i can
21:38:31 <Eddi|zuHause3> "blob" means "binary large object", so the "binary" is redundant
21:38:52 <Tefad> blob can also mean goo-like creature
21:38:59 <Tefad> so i am reduntant on purpose.
21:39:06 <Tefad> also, i can't spell redundant.
21:39:47 <Eddi|zuHause3> the german term is "doppelt gemoppelt", but i'm not sure if there is even a close translation
21:40:16 <Gonozal_VIII> pudelhund & spaghettinudel
21:41:29 <skidd13> Eddi|zuHause3: Try latin -> Hendiadyoin
21:42:15 <Eddi|zuHause3> that supposed to be latin?
21:42:53 <skidd13> Its rhetoric so latin should be valid for many languages as technical term
21:43:18 <Eddi|zuHause3> i have absolutely no clue what you are talking about...
21:43:38 <skidd13> Eddi|zuHause3: "doppelt gemoppelt"
21:43:51 <TrueBrain> huma: I have something for you:
21:44:37 <TrueBrain> too bad any higher resolution than 320x200 as TT screen needs.. well.. a beast of a client-machine ;)
21:44:45 <TrueBrain> btw, don't go near edges, it is broken :)
21:44:45 <Gonozal_VIII> still doesn't like the borders ;-)
21:45:03 <skidd13> Eddi|zuHause3: Hendiadyoin is the technical term (in latin) for a thing that is colloquial called "doppelt gemoppelt"
21:45:06 <huma> TrueBrain, is there arctic? :)
21:45:13 <TrueBrain> huma: nope :p Still no arctic :)
21:45:23 <TrueBrain> I was about to add it, but... I found more important things :)
21:46:01 <skidd13> TrueBrain: WTF you added mouse scroll :O
21:46:05 <huma> what can be more important than that?
21:46:37 <TrueBrain> as I said long ago: it is just waiting for the time a whole OS comes to you via a browser :p
21:47:04 <huma> he's serious about it :)
21:47:04 <TrueBrain> if there would be an easy way to add OpenTTD to python via bindings, you could in fact have the game done in several days
21:47:37 <Gonozal_VIII> just put a dos floppy online
21:47:56 <TrueBrain> skidd13: I rest my case ;)
21:48:22 <Eddi|zuHause3> only problem i have is that in conventional TT you scroll with right mouse button...
21:48:27 <huma> eh, i'm client #5, it doesn't greet me by name :(
21:48:49 <TrueBrain> I am more amazed by the performance :)
21:48:53 <TrueBrain> hmm, yet an other crash...
21:48:54 <hylje> python bindings for openttd, ingenious
21:48:58 <TrueBrain> HTTPConnection seems faulty..
21:49:20 <TrueBrain> skidd13: eyeos is very nicely done :)
21:50:00 <TrueBrain> haha, a process list :)
21:50:43 <Gonozal_VIII> no desyncs and instant join on any mapsize :-)
21:50:59 <huma> Tefad, it's a webtt - web based openttd game
21:51:16 <huma> TrueBrain promises to finish it next week
21:51:26 <TrueBrain> only in your dreams
21:51:31 <TrueBrain> which reminds me: I don't want to know what you dream about
21:51:45 <Tefad> is it just me or is scrolling absurdly slow
21:51:52 <huma> my christmas vacation :)
21:51:56 <TrueBrain> Tefad: only if your computer can't handle it :)
21:52:02 <TrueBrain> it moves 30 divs every time you scroll
21:52:07 <Tefad> i guess 2.4GHz isn't enough
21:52:09 <TrueBrain> on some computers this is as fast as lightning
21:52:12 <TrueBrain> on others it is dead slow..
21:53:02 <TrueBrain> so it really is your computer :p
21:53:07 <Phazorx> oh... nice to see ludde here
21:53:13 <skidd13> Tefad: 2.4GHz doesn't mean anything.... Compare a 2.4GHZ sparc with a 2.4GHz intel
21:53:13 <Gonozal_VIII> 40 tiles per request? the white lines that appear sometimes are only 20 tiles
21:53:25 <TrueBrain> Gonozal_VIII: yup, non-lineair requests :)
21:53:29 <Tefad> skidd13: heh, no long it does
21:53:34 <Tefad> it used to.. back in the day
21:53:42 <Tefad> it's hard to gauge performance now : \
21:53:57 <Phazorx> i wonder if anyone noticed a name of global company manager on the wwottdgd game
21:54:35 <huma> damn, why it gives me like 5 alerts about session timeout? :)
21:54:47 <TrueBrain> huma: because that part is pretty unfinihsed :)
21:54:58 <TrueBrain> but it already detects it! That is something ;)
21:55:46 <TrueBrain> tnx :) I wonder if I will ever continue on it, but okay :p
21:55:53 <Tefad> i get much better formance in ff under winde than i do with native ff.
21:55:55 * huma gives TrueBrain a cookie
21:56:13 <TrueBrain> Tefad: weird... maybe you installed FF wrong? :p
21:56:24 <Tefad> i compiled it myself if that's what you mean : x
21:56:36 <Gonozal_VIII> openttd as a massive multiplayer browsergame thousands of players on a giant map with a daylength of 24 minutes or something like that^^
21:57:31 <TrueBrain> I just really wonder what the server can handle... how much concurrent connections and stuff...
21:57:34 <Tefad> i get like 1-2fps in wine-ff
21:58:15 *** cybermage has joined #openttd
21:58:24 <TrueBrain> it takes 0.6ms to handle a single client, which happens on scrolling once in the 100ms.. the game logic needs 5ms every 30ms, so that is 150ms every second; a client needs 6ms every second.. that gives..
21:58:31 <TrueBrain> @calc (1000 - 150) / 6
21:58:31 <DorpsGek> TrueBrain: 141.666666667
21:58:36 <TrueBrain> 141 clients per server
21:59:00 <TrueBrain> so I guess it won't be a 30ms per tick game ;)
21:59:03 <cybermage> hi guys, Is it possible to disable autosave on running server ?
21:59:17 <Tefad> yeesh, native-ff just sucks completely
21:59:29 <TrueBrain> Tefad: install the pre-built binary
22:00:14 <Tefad> argh message bombed in opera
22:00:27 <Tefad> java script errors one after the other.
22:00:48 <TrueBrain> what kind of errors?
22:01:00 <TrueBrain> server had a hick-up, so needed a restart
22:02:53 <TrueBrain> dunno... something hits an infinite loop, so it seems
22:03:03 <TrueBrain> I use stackless python, and some custom http servers.. not really tested ;)
22:03:33 <Sacro> hmm, this spotify looks cool
22:03:34 *** jonty-comp has joined #openttd
22:03:45 <huma> yea, i wrote a neat xsl once that kept crashing websphere :)
22:04:10 <TrueBrain> making things crash is easy... keeping things from crashing on the other hand...
22:04:48 <TrueBrain> but okay, byebye server :) Showtime over ;)
22:05:03 <hylje> stackless silently crashes on deadlocks
22:05:28 <TrueBrain> hylje: I don't make deadlocks ;)
22:05:41 <huma> maybe i should try jwebtt :)
22:06:51 <TrueBrain> let me know if you can get the same performance (Server-wise)
22:07:13 <huma> and there won't be green fields :)
22:08:00 <Tefad> arctic is too bright for me
22:08:24 <Tefad> it'd be fun to see the night-time set cover everything
22:09:08 <huma> there's a patch for that
22:09:22 <Tefad> i'm not fond of convoluted graphics systems and limited palette art *shrug*
22:09:36 <Tefad> nor do i claim to be a great artist
22:09:46 <Gonozal_VIII> there's a monitor setting for that, it's called gamma :-)
22:09:54 <Tefad> i'd be more interested in implimenting a better AI
22:10:03 <TrueBrain> Tefad: so, load up NoAI
22:10:23 <TrueBrain> you are intersted in AIs, and don't know NoAI?
22:12:13 <TrueBrain> very simular to C++
22:12:59 <Tefad> so it's a competitor to LUA
22:13:49 <TrueBrain> it is an addition to it :)
22:13:58 <TrueBrain> (lua doesn't have objects, Squirrel does)
22:14:20 <TrueBrain> it is in Squirrel's case
22:14:32 <TrueBrain> as 'objects != lightweight' is NOT by default true
22:14:41 <Sacro> there is no sunos_sparc build :(
22:14:45 <Tefad> assembly is lightweight ; )
22:14:53 <TrueBrain> Sacro: give me the binaries and I make it for you :)
22:15:02 <TrueBrain> Tefad: depends very much on your point of view
22:15:10 <TrueBrain> Tefad: I see you porting TTDp to linux?
22:15:22 <TrueBrain> means writing a virtual machine... there your 'zero overhead' goes ;)
22:15:31 <TrueBrain> Sacro: I need files from a sunos sparc
22:15:51 <Sacro> but i need a converter cable
22:16:07 <Tefad> however, my sparc has VGA
22:16:13 <Tefad> and also doesn't post : \
22:16:20 <Tefad> my alpha almost posts, then fails it.
22:16:23 <TrueBrain> Sacro: I also still need someone who has a 64bit sunos
22:16:52 <Tefad> isn't sunos rather dated?
22:16:52 <cybermage> TrueBrain: virtualization ?
22:17:08 <TrueBrain> as the compiler for it is there, but I need certain OS specific files to get the compile-farm to work ;)
22:17:26 <Tefad> last i checked virtualization isn't too heavy.. but that wasn't my point
22:17:28 <Sacro> Tefad: well my sparcstation only has 2*40Mhz
22:17:30 <cybermage> TrueBrain: If you have 64b processor, you can run virtualized enviroment :)
22:17:45 <Tefad> modern computers "virtualize" into uops anyway
22:18:05 <Tefad> hmm my sparc is a few hundred MHz
22:18:16 <TrueBrain> Tefad: and my point was, that lightweighted is _very_ subjective :)
22:18:59 <TrueBrain> (I seem to be randomly dropping letters...)
22:19:05 <TrueBrain> might be the packet-loss :p
22:19:25 <Tefad> i like the binary bf derivative
22:19:48 <hylje> let me implement bf in python
22:21:57 <TrueBrain> I once tried to write a program in it
22:22:07 <TrueBrain> (and not many languages I failed to write in :p)
22:22:49 <ludde> TrueBrain: my work mate is co author of shakespearelang
22:22:58 *** ChanServ sets mode: +o Bjarni
22:23:05 <TrueBrain> ludde: then I have great respect for him :)
22:23:11 <TrueBrain> as I couldn't make anything out of it :)
22:23:27 <TrueBrain> (and no, Bjarni, not you)
22:23:43 <TrueBrain> (I knew you wanted to say something like that, so I thought, let's beat him too it)
22:24:16 *** |Bastiaan| has joined #openttd
22:28:51 <skidd13> Bjarni: As you are the man for OS X may I recruit you for some profiling? -> FS1392
22:30:10 *** dihedral has joined #openttd
22:31:40 <Bjarni> skidd13: shouldn't this produce the same executable code?
22:31:56 <Bjarni> but then again I guess you want me to verify that it actually does that :)
22:32:15 *** ThePizzaKing has joined #openttd
22:33:26 <Bjarni> no I will not read ASM to ensure that it produces the very same code
22:33:36 <Bjarni> and odds are that md5 wouldn't do the trick either
22:33:59 <Bjarni> though that could be interesting to try... then it really would be the same code and no need to test
22:34:07 <skidd13> Bjarni: It hopefully should not. My hope is that the compilers will do a better job with the inline stuff instead of the defines.
22:34:14 <Bjarni> but somehow I don't think that will happen
22:34:32 <TrueBrain> Bjarni: don't worr,y I am 100% sure the asm will be different
22:34:39 <skidd13> Bjarni: I just want to know if there is any difference in performance.
22:34:41 <TrueBrain> what we need to know, if there is a noticable speed difference
22:35:37 * Bjarni prepares for some testing
22:35:43 <Bjarni> aka freeing up CPU time
22:36:29 <dihedral> and skidd13 of course too
22:37:29 *** michi_cc_ has joined #openttd
22:37:29 <Bjarni> how to test this in the best way...
22:38:25 <Bjarni> so far I did profiling to learn how much CPU time the different functions use compared to each other and TIC/TOC for actual function speed
22:38:44 <Bjarni> but TIC/TOC has to be used at the correct location in the code :/
22:40:27 <Bjarni> so how should this be tested to get the best results...
22:40:44 <Bjarni> best as in: best info, not fastest ;)
22:42:07 <DorpsGek> skidd13: Smatz was last seen in #openttd 2 weeks, 1 day, 8 hours, 13 minutes, and 40 seconds ago: <SmatZ> hmm yes, is not in unifont.rf
22:42:26 <TrueBrain> if (addr[0] == 0xE9) { // 0xE9 == jmp
22:42:26 <TrueBrain> addr[0] = 0xE8; // 0xE8 == call
22:42:31 <TrueBrain> I love things that work :)
22:43:04 <Tefad> so jmp and call are the same thing in this vm?
22:43:19 <TrueBrain> that is no code from any vm
22:43:24 <TrueBrain> this is code from a dlopen for windows :)
22:43:41 <TrueBrain> it inverts an optimization of a compiler :)
22:43:48 <TrueBrain> dlopen, as in: dynamic library loader
22:43:57 <TrueBrain> but not in dll, as this supports lazy binding
22:43:57 * Tefad confused and shuts up : D
22:44:07 <hylje> dynamic libraries are horrible hacks
22:44:26 <hylje> but they tend to work..
22:44:53 <Bjarni> except when they don't and then it's really a pain to deal with them :(
22:45:21 <Tefad> i've not had to deal with dynamic lib problem in forever
22:45:23 <TrueBrain> Tefad: in *nix world, you can load library without having them to find all the functions right away.. example: I have a lib1 which creates func1(). I have a lib2 which creates func2(). Now lib2 wants to call func1(), and lib1 wants to call func2(). Under linux this is no problem, and works out of the box
22:45:34 <TrueBrain> in windows, via DLLs, you need a bunch of ugly hacks to make it to work, if at all
22:45:46 <TrueBrain> so, I made a library that works like *nix
22:45:53 <Bjarni> Tefad: so you haven't had OpenTTD crashing on you with a "lazy pointer not found"?
22:46:18 <Bjarni> I fixed that by moving away from SDL
22:46:36 <Bjarni> it just tries to do as little as possible
22:47:03 <TrueBrain> Tefad: in my example, when you load lib1, the call to func2() doesn't point to anywhere yet
22:47:10 <Sacro> Tefad: he's a dutch mac user
22:47:11 <TrueBrain> when you call it, it looks up where it should point to
22:47:23 <TrueBrain> this is called a lazy symbol (not a lazy pointer)
22:47:28 <Bjarni> there is a Dutch mac user in here?
22:47:35 <TrueBrain> in *nix land this is _very_ common
22:47:40 <Tefad> only gets lookedup when it's used
22:47:45 <TrueBrain> in Windows land this is advised against (by MS)
22:47:59 <TrueBrain> basicly: *nix: runtime, MS: compiletime
22:48:22 <Tefad> those lead to problems on my system when i upgrade packages, sometimes there's updates to both and they have circular dependency
22:48:25 <TrueBrain> this also means: *nix: no need to have all the files when compiling, MS: you need all files when compiling ;)
22:48:40 <TrueBrain> euh, yeah, it is not that, but he, what ever :)
22:48:43 <TrueBrain> at least I tried ;)
22:48:51 <Tefad> i know what you're talking about
22:49:01 <Tefad> what i'm saying is similar, at least i think it is
22:49:26 <Tefad> lib1 uses func2 in lib2 while lib2 uses func1 from lib1, yes?
22:49:32 <Tefad> lib1 depends on lib2 and viceversa
22:49:46 <Tefad> assuming neither is installed, how would a package manager know what's up
22:49:57 <Tefad> unless it realizes mutual dependency is ok.
22:49:58 <TrueBrain> in a package system it is slightly different
22:50:17 <Tefad> in windows land, i dont think i'd be able to compile that at all
22:50:37 <Tefad> in unix land, i think the config script would be confused, but could probably be bootstrapped, eh?
22:50:50 <Tefad> unless you have binaries anyway.
22:50:56 <Tefad> then there's no config script
22:51:17 <Tefad> this problem occurs at compile time in windows land? unless you have a workaround in place
22:51:48 <TrueBrain> Windows requires all the symbols to be known at compile-time
22:51:58 <TrueBrain> there are workarounds, but they are hard to write
22:52:22 <Tefad> if you can manage create a binary in unix you're fine
22:52:35 <TrueBrain> short: it helps porting Unix -> Windows a lot :)
22:52:49 *** Eddi|zuHause has joined #openttd
22:52:59 <Tefad> totally facetious there yes?
22:53:33 <Tefad> not meant to be taken seriously or literally: a facetious remark.
22:53:46 <Eddi|zuHause> hm, so much for my uptime
22:53:53 <TrueBrain> I don't understand what you try to say, sorry :)
22:54:25 <Tefad> how can being able to find symbols at compile time be in unix but not windows be beneficial to porting unix to windows?
22:54:40 <TrueBrain> Tefad: it isn't. But my library solves tha tproblem ;)
22:54:56 <Tefad> ok, "it" wasn't defined in your sentence : D
22:55:04 <Tefad> i assumed you were referencing the whole concept
22:56:22 <hylje> Tefad: i dont think you got the subtlety there.. :p
23:07:55 <skidd13> TrueBrain: I did some profiling and noticed that some functions increase their speed while others loose :(
23:09:32 <skidd13> GetTileType +37% Random -18%
23:09:47 <TrueBrain> but Random doesn't use any of your functions, does it?
23:10:09 <TrueBrain> so, keep the ones that improve
23:10:13 <TrueBrain> revert the others ;)
23:10:33 <skidd13> Maybe there is some space to rewrite the functionality ;)
23:10:41 <skidd13> to increase it even more
23:13:33 <skidd13> But all at all I need input form other testers. IIRC smatz did some huger profiling tests but he sems to be away :(
23:16:31 <skidd13> It's quite late. good night and happy profiling ;)
23:34:46 <CIA-1> OpenTTD: rubidium * r11373 /trunk/src/industry_cmd.cpp: -Fix [FS#1398]: do not even think about building disabled industries.
23:39:22 <TrueBrain> hylje: I have a request for the patch-review: dev-talk (hidden for all other users)
23:42:47 <Sacro> Rubidium: heh, i can just imagine a motherly figure telling people off
23:43:45 *** mikk36|w has joined #openttd
continue to next day ⏵