IRC logs for #openttd on OFTC at 2007-05-15
⏴ go to previous day
00:00:13 <Bjarni|uni> remote host forces the connection to drop
00:00:36 <Bjarni|uni> the file is a binary file of nearly 500k
00:00:46 <Bjarni|uni> is there some sort of file size limit in svn?
00:01:43 <Bjarni|uni> the "funny" part of this one is that we need it to work for the assignment where we have a deadline, which is tomorrow morning :(
00:01:46 <glx> I already commited big stuff (syncs)
00:02:29 <Bjarni|uni> but now it fails and with that error message... it says very little
00:03:10 <Belugas> maybe binaries are different then patches, diffs...
00:03:24 <Bjarni|uni> I'm in the lab right now
00:03:37 <Bjarni|uni> and there are more robots than people in here right now o_O
00:03:41 <Belugas> at least, you have a good reason why you cannot meet the deadlines ;)
00:04:22 <Bjarni|uni> we already pushed the deadline one day
00:05:43 <glx> can you commit smaller diff?
00:06:13 <Bjarni|uni> normal sized text files and source files works just fine
00:06:57 <Bjarni|uni> it's just the binary files, that fails
00:07:14 <Belugas> or maybe a link to a ftp site where the data will be?
00:07:55 <Bjarni|uni> I think we will try to figure out something like that, but it's kind of silly to do that
00:08:06 <Bjarni|uni> now that we work based on svn checkouts
00:08:24 <Belugas> do you know the admin of the svn server?
00:08:47 <Bjarni|uni> actually I'm trying to solve the problem together with him
00:08:56 <Bjarni|uni> he is as lost regarding this as I am
00:09:01 <Bjarni|uni> the server tells us nothing
00:13:51 <Bjarni|uni> err, now it worked. Moving the files to the computer next to the one that failed to commit appears to fix the problem
00:14:21 <Bjarni|uni> both computers can commit small files
00:14:49 <Belugas> ho... i think i saw that once. It may be related to the speed at which the connection is established.
00:15:07 <Belugas> sometimes, i get a connection failure at the office, when there's too much activity
00:15:26 <Bjarni|uni> it's not like we have too much activity here right now
00:16:10 <Bjarni|uni> in fact I think we are pretty much alone on the network right now
00:17:05 <Bjarni|uni> anyway it works now, so I will go offline again to finish this (and not pay attention to whatever weird links or whatever you post :p )
00:21:11 <Sacro> damn, wanted to say some more offensive things in denmarkian :(
00:21:57 <Belugas> next time, prepare a file :)
00:22:03 <Belugas> copy and paste will be faser ;)
00:25:45 <Sacro> its only Bjarni who has ever had !kick capabilities
00:25:49 <glx> too bad I don't have this right
00:27:45 *** Sacro was kicked by glx (:))
00:28:26 <Belugas> whooo... it worked :)
00:28:46 <glx> and he don't have autoreconnect ;)
00:43:29 <_42_> Belugas, DaleStan (~Dale@74-140-61-198.dhcp.insightbb.com) was last seen quitting #openttd 1 day 3 hours 55 minutes ago (13.05. 20:48) stating "Quit: Leaving" after spending 12 minutes there.
01:31:38 *** Eddi|zuHause3 has joined #openttd
01:38:36 <Sacro> Belugas: do you require him?
01:40:01 <Belugas> i would very much like him to explain me what he's doing exactly on oilpower.grf
01:50:11 <Sacro> Belugas: irc.quakenet.org #tycoon
01:50:41 <Sacro> Belugas: <patchbot>> Sacro: DaleStan last spoke 2d 2h ago. 1d 5h ago DaleStan left with the message: Quit: Week of biking, then Europe, then, ... ? Who knows?
01:50:55 <Sacro> he seems to have fled the scene
01:52:04 <Belugas> good for him :) it sounds like vacations :)
01:52:56 <Sacro> could ask patchman, but he will be asleep currently, or maybe eis_os
01:53:11 <Belugas> it's about oilpower.grf
01:53:25 <Belugas> in fact, i would have hoped born_acorn could answer
01:53:38 <Belugas> the nfo seems a bit strange
01:53:46 <Belugas> in fact, i don't knwo what to do with it
01:54:21 <Belugas> well... with the specs in front of you, it's not too bad
01:54:37 <Belugas> but when they do not match anymore,
01:54:56 <Belugas> then it's head scratching time,
01:55:07 <Sacro> i'm trying to code a train set
01:57:45 <Belugas> i know the feeling :)
01:57:57 <Belugas> although it is in the other side :S
01:57:58 <Sacro> i have some nice sprites though
01:58:07 <Sacro> i just wish i could code them in xml or something
01:59:58 <Sacro> i want to do a UK Signals set
02:00:05 <Sacro> cos i *really* hate the TTD signals
02:00:15 <Sacro> but it requres a 5/2/3 NFO
02:08:29 <Belugas> you mean action 05, actgion 02 and or action 03??
02:13:02 <Belugas> i leave that to you :)
05:27:04 *** ThePizzaKing has joined #openttd
05:48:05 *** Taikaponi has joined #openttd
05:49:53 *** Frostregen_ has joined #openttd
05:50:07 *** boekabart has joined #openttd
05:57:36 *** XeryusTC has joined #openttd
06:20:43 <CIA-1> OpenTTD: rubidium * r9840 /trunk/src/economy.cpp: -Fix (r9838): MSVC is complaining about signedness again (Belugas).
06:35:53 *** lolman_ has joined #openttd
06:40:31 *** lolman_ is now known as lolman
06:56:24 *** Thomas[NL] has joined #openttd
07:22:04 *** Brianetta has joined #openttd
08:10:37 *** Vikthor has joined #openttd
08:28:41 *** raimar2 has joined #openttd
08:42:36 *** dihedral has joined #openttd
08:47:36 <dihedral> sitting at school...
09:03:36 <TinoM> mission flicken erledigt, ab zum ifgi
09:03:45 *** peterbrett has joined #openttd
09:03:47 *** Frostregen has joined #openttd
09:29:52 *** Tino|R152 has joined #openttd
09:43:43 *** Vikthor has joined #openttd
09:55:58 *** Tino|R152 has joined #openttd
10:05:11 *** Brianetta has joined #openttd
10:05:25 *** Tino|R152 has joined #openttd
10:06:36 *** Tino|R152 has joined #openttd
10:07:03 *** ChanServ sets mode: +o Bjarni
10:08:16 <Brianetta> xchat works exactly as I want, and I couldn't get irssi to emulate it. I think nine months is ample time to grow to hate something.
10:10:24 *** Tino|R152 has joined #openttd
10:13:32 <Brianetta> I wish I knew why my dedicated server keeps on exiting
10:14:31 <Brianetta> I'm recompiling with debug
10:14:37 <Brianetta> I specify -d on the command line, but still nothing
10:14:44 <Brianetta> No end of game specified
10:14:54 <Brianetta> It normally rolls on to 2398
10:14:58 <Brianetta> if I forget about it
10:15:50 <Brianetta> Hopefully I'll see an assert or something next time
10:15:56 <Brianetta> or at least some output
10:16:05 <Brianetta> How do you increase verbosity?
10:16:39 <Rubidium> or -d <specific_target>=<num>
10:17:07 <Brianetta> I don't have any clue where to look, so I'll just bang them all up
10:17:24 <Rubidium> that would be -d 9 which isn't nice ;)
10:17:43 <Brianetta> What's likely to be a useful level?
10:17:46 <Rubidium> quite 'floody' if I may say so
10:17:58 <Brianetta> flooding matters not - I have IRC rcon
10:18:01 <Rubidium> I've got not a real idea
10:18:37 *** Thomas[NL] has joined #openttd
10:21:45 <Brianetta> OK, that's running in screen, teeing to a log fil
10:22:13 <Brianetta> /msg sarah_pilot playercount
10:49:02 *** ChanServ sets mode: +o orudge
11:20:18 *** setrodox has joined #openttd
11:28:46 <CIA-1> OpenTTD: rubidium * r9841 /trunk/src/ (7 files in 2 dirs): -Codechange: add a little more type strictness to the vehicle types.
11:32:53 *** Tino|R152 has joined #openttd
11:48:19 *** Tino|R152 has joined #openttd
12:04:04 *** boekabart has joined #openttd
12:32:14 *** Progman has joined #openttd
12:38:42 <Belugas> hey Bjarni. SO, how did your svn fight ended yesterday?
12:51:49 *** peterbrett has joined #openttd
13:02:16 <CIA-1> OpenTTD: rubidium * r9842 /trunk/src/ (functions.h stdafx.h): -Codechange: now NOT_REACHED is also triggered when debugging is disabled.
13:02:37 *** Vikthor has joined #openttd
13:07:57 *** setrodox has joined #openttd
13:14:47 <Bjarni> <Belugas> hey Bjarni. SO, how did your svn fight ended yesterday? <-- I won and we met the deadline.... the sideeffect is that I went to bed around 7:30.... I'm not feeling too well right now :(
13:15:36 <Bjarni> I have to leave in a moment
13:18:13 <Belugas> you will enjoy your bed much more tonight :)
13:18:39 <Belugas> hold on, and grab some coffee (or whatever suits you) until then
13:19:16 <TrueBrain> Belugas did a 'svn update'
13:32:08 *** Osai^2 is now known as Osai
13:38:16 *** boekabart has joined #openttd
13:38:56 <CIA-1> OpenTTD: rubidium * r9843 /trunk/src/newgrf.cpp: -Fix (r9838): tell the newgrfs that we now support proper FIFO loading.
13:39:09 <boekabart> welcome back you mean... my internet has gone down 4 times already today
13:42:21 <Eddi|zuHause3> hylje: i think the german version sounds a little more polite :)
13:42:51 <hylje> the fact of it being german makes it about equal in politeness
14:09:03 <CIA-1> OpenTTD: truelight * r9844 /trunk/ (22 files in 3 dirs):
14:09:03 <CIA-1> OpenTTD: -Codechange: replace zoomlevel with an enum
14:09:03 <CIA-1> OpenTTD: -Codechange: use predefined enums for viewport zoomlevels
14:09:42 <peterbrett> TrueBrain: is that your commit?
14:09:57 <boekabart> TrueLight: One of the good things of the 32bbp patch :) ?>
14:10:38 <TrueBrain> boekabart: just one of those things that should have been in the trunk long ago :p
14:10:56 <peterbrett> TrueBrain: I'm puzzled as to why you would want to use predefined enums for zoomlevels.
14:11:13 <TrueBrain> peterbrett: as now it is a mess, with single numbers nobody knows the meaning of
14:11:22 <TrueBrain> and if you start adding new zoom-levels
14:11:23 <peterbrett> TrueBrain: Theoretically, zoom is a geometric value, i.e. real.
14:11:27 <TrueBrain> it gets even more weird
14:11:33 <TrueBrain> peterbrett: currently, yes. Soon: no
14:12:15 <peterbrett> So it should theoretically be represented by a real number, not by a selection of predefined ones.
14:12:36 <peterbrett> Or am I just going nuts? Or am I looking at it too much like an engineer and not enough like a hacker?
14:12:56 <TrueBrain> this exactly cleans up the 'hackish' part
14:13:06 <TrueBrain> and you assume every zoom level is 2 times the last one
14:13:08 <boekabart> peterbrett: It will be in OpenTTD 3.0 '3D'
14:13:14 <TrueBrain> where there is no real reason for doing so
14:13:37 <TrueBrain> and as we don't use OpenGL
14:13:43 <TrueBrain> don't expect zoom level of 1.2223232`
14:15:02 <peterbrett> So what's wrong with documenting that "the zoom level is an integer z. for a given z, a sprite will be scaled by 2^(-z)
14:15:52 <boekabart> you assume a fixed base size
14:16:03 <peterbrett> boekabart: No, I don't.
14:16:20 <boekabart> which might be an untrue assertion in the future
14:16:50 <TrueBrain> peterbrett: you will see in the near future that this is very useful :)
14:16:56 <peterbrett> True. However, I assume that each sprite has a "natural size" associated with it.
14:17:10 <TrueBrain> but, to give a simple example: we now have zoom normal, 2x out, 4x out
14:17:14 <TrueBrain> say someone makes 3x out
14:17:35 <boekabart> "for any zoomlevel,all sprites should be scaled to a specified width/height" might be better in the future. allows any zoom factor, and any input size for sprites
14:17:36 <TrueBrain> this means that you need to replace all '2' to '3' and add a '2'
14:17:39 <peterbrett> Then the scaling is 2^(<sprite dependent factor> - z)
14:17:44 <TrueBrain> or make '3' be in betwen of value '1' and '2'
14:17:51 <TrueBrain> so you get very weird things ;)
14:17:52 *** Vikthor has joined #openttd
14:18:06 <peterbrett> TrueBrain: Clearly the scale factor should be base e
14:18:14 <peterbrett> But it will always be exponential.
14:18:18 <TrueBrain> peterbrett: and that is not really needed :)
14:18:34 <peterbrett> In order to percieve a continuous rate of zooming out (optical illusion)
14:18:49 <peterbrett> This is why the default is 1, 2, 4, IIRC
14:19:04 <TrueBrain> but okay, this really is just code cleanup
14:19:43 <peterbrett> So actually making it work "properly" is something planned?
14:20:14 <peterbrett> So I can go into the console and type "zoom <random real number>" and it will try to do it
14:20:38 <TrueBrain> no, that won't be possible anytime soon; that is: for X zoom levels it will
14:20:44 <TrueBrain> or rather: N zoom levels
14:20:58 <TrueBrain> as we have like 10 switches() for zoomlevel
14:21:15 <peterbrett> TrueBrain: But I've just demonstrated a system by which you can do zooming consistently and accurately using a single integer value.
14:21:25 <TrueBrain> good for you, but OpenTTD isn't ready for it :)
14:21:31 <peterbrett> TrueBrain: And there's no reason you can't gate it in a certain range, 0-2 say
14:21:33 <TrueBrain> there is more needed then just a theory or an idea :)
14:21:40 <peterbrett> TrueBrain: Which will give you your zoom levels.
14:21:43 <boekabart> TrueBrain: these switches are going to be replaced by lookup tables, i presume?
14:21:47 <TrueBrain> if you clamp it between 0 and 2, you can make the console command right now ;)
14:21:56 <TrueBrain> boekabart: not for now
14:22:52 <peterbrett> TrueBrain: That sounds like a challenge ;)
14:23:06 <boekabart> from what I remember, most of them can be. The ones deciding font size/ hiding things actually should be I think.
14:23:20 <TrueBrain> peterbrett: I dunno if it is something we want... do you really want to zoom out 32 times? :)
14:24:15 <peterbrett> TrueBrain: No, but surely it would be useful to be able to zoom out to view the whole map. Of course, the frame rate would be horrendously slow, but hey
14:24:17 <boekabart> I actually think that one more level 'in' and one more 'out' are enough. But I'm not a 32bpp artist, of course
14:24:33 <peterbrett> That's not a good reason not to do it ;)
14:24:35 <TrueBrain> peterbrett: and that is a problem that will be hard to fix, so I don't think it is useful at this stage :)
14:24:50 <TrueBrain> boekabart: I doubt very much if we need a zoom-in level
14:24:54 <TrueBrain> but okay, code-wise it is now possible
14:25:40 <boekabart> TrueBrain: On a 1280x1024 17" screen, sometimes details are getting small, I use CTRL-D once in a while now...
14:25:52 <TrueBrain> boekabart: then it might be worth considering
14:25:52 * boekabart needs new glasses, maybe.
14:28:48 <CIA-1> OpenTTD: rubidium * r9845 /trunk/src/zoom.hpp: -Fix (r9844): svn:eol-style/svn:keywords were missing...
14:30:50 <Brianetta> boekabart: You asked about my live webcam
14:30:58 <Brianetta> boekabart: It's done with the screenshot command
14:31:17 <peterbrett> Brianetta: Live webcam? Where? ;)
14:31:31 <Brianetta> ppcis.org/standards
14:31:32 <boekabart> peterbrett: doesn't show Brianetta, just the game ;)
14:31:37 <boekabart> Brianetta: yes, someone told me... I didn't know server could make a screenshot
14:31:51 <boekabart> I was suggesting to add < > ^ buttons around it
14:32:02 <boekabart> to control the scrollto
14:32:06 <Brianetta> I would do the buttons, but the server load...
14:32:15 <Brianetta> people would click the buttons
14:32:45 <boekabart> maybe accept only from one person, until that person times out, then accept the next
14:33:02 <Brianetta> Once the devs find a way to make a dedicated server that doesn't render the game into an off-screen framebuffer, they'll lose the screenshot function.
14:33:26 <boekabart> Devs: Why is de dedicated server rendering?
14:33:29 <Brianetta> Not much point includin garrows if they only move about ten tiles
14:33:42 <Brianetta> boekabart: dedicated server is a hacky kluge
14:33:54 <boekabart> no, as long as 1 user keeps pressing, he doesn't time out.
14:34:01 <Brianetta> Basically, it's a regular game but with the SDL bits chopped out
14:34:14 <Brianetta> will press lots and lots
14:34:32 <boekabart> how much does making a shot from the already-existing buffer cost!?
14:34:45 <Brianetta> It's that, and also the mogrification that puts the caption on
14:34:53 <Brianetta> and the script that copies it to the web page
14:35:09 <peter1138> you know you want to...
14:35:27 <boekabart> yeah!! instead of sdl or gdi, use mjpeg!
14:35:43 <boekabart> or mpeg::: the motion vectors are already known....
14:36:05 <Eddi|zuHause3> make giant screenshots in real time, and then cut out the view of each person :)
14:36:15 <boekabart> anyway, so why is it actually rendering? an IF in DrawViewport isn't possible?
14:37:00 <Eddi|zuHause3> boekabart: it has to be drawing into a buffer, because the drawing functions are not clearly separated from the game logic
14:37:40 <boekabart> hm... I didn't notice that so much actually. Did anyone ever spend serious time on that?
14:42:45 <RamboRonny> hi nerds, a good friend of mine once said a nerd is a good thing
14:43:15 <TrueBrain> RamboRonny: the fact that you need to add it directly after you saying so, indicates you know it can be read offensive
14:43:30 <TrueBrain> so it is doubtful you agree with your good friend
14:43:52 <boekabart> and a nerd isn't a thing, it's a person.
14:45:12 <peter1138> it's an attribue ;p
14:54:29 <gregor> What chmod needs the sample.cat ?
14:55:08 <boekabart> r... like everything except openttd.cfg and save/
14:56:13 <blathijs> gregor: It needs to be readable by the user running openttd
14:57:08 <blathijs> so if you want to run openttd for one user, which also owns sample.cat, 400 or 600 is fine
14:57:17 <blathijs> if you want it to be run by all users, 644 is needed
14:58:51 <valhallasw> I changed the symlinks to a apache rewrite with [R=permanent]. that would be the best solution to keep, I suppose?
14:59:10 <Brianetta> apache rewrite is a write-only language to me
14:59:23 <Brianetta> If it works, go with it
14:59:35 <peterbrett> Brianetta: perl is a write-only language to me ;)
14:59:51 <valhallasw> interesting, 'write-only' languages :p
14:59:52 <hylje> that's the common consensus anyway
15:00:12 <Brianetta> well, by write only, I mean that the only way to modify it is to erase it, and try again
15:00:25 <valhallasw> regexp's are write-only in general
15:00:29 <peterbrett> valhallasw: indeed
15:00:31 <Brianetta> no, they can be modified
15:00:44 <valhallasw> apache rewrites just are regexps ;)
15:01:01 <Brianetta> I'm sure they're related, but since I can't read it, I don't know
15:01:04 <hylje> apache rewrites are regexps
15:01:10 <Brianetta> I have written rewrite rules
15:01:25 <Brianetta> but I don't understand how they work, and to be honest I'm not even sure of the syntax.
15:01:46 <Brianetta> You said "[R=permanent]" and I just didn't know what that meant
15:01:50 <Brianetta> or even where that would go
15:01:58 <Brianetta> or even if that was what you meant by a rewrite rule
15:02:05 <valhallasw> permanent redirect
15:02:12 <Brianetta> It does in the line somewhere?
15:03:55 <boekabart> me, i want only ontopic chatter to keep me off my work
15:07:28 <valhallasw> boekabart: /part O+ ;P
15:11:57 <boekabart> is that the same as /leave?
15:12:12 <valhallasw> more like /leave *
15:13:51 <Brianetta> Originally, IRC only had numbered channels
15:14:21 *** GoneWacko has joined #openttd
15:15:38 <boekabart> sounds like IRC used to be only for nerds
15:16:18 <hylje> there were no way to chat on the internet back then
15:16:25 <hylje> more or less just one-to-one
15:16:29 <blathijs> boekabart: I still have a fairly strong channel-to-number connection in my head, for irssi shows activity by channel number :-)
15:16:58 <Noldo> blathijs: window number
15:17:01 <boekabart> blathijs: see, still. just for nerds. and yes I know I'm here.
15:17:08 <hylje> channel names take my precious screen estate
15:17:50 <Brianetta> I just ditched irssi
15:17:54 <boekabart> so that's why the # prefix...
15:18:32 <Brianetta> boekabart: # for named network-wide channels
15:18:37 <Brianetta> @ for local named channels
15:19:25 <boekabart> local == the random server you log on to?
15:19:30 <Brianetta> Some servers don't allow local channels to be created, but that's just because they have miserable admins
15:19:42 <boekabart> what are those useful for
15:19:45 <Noldo> boekabart: not all people connect to random servers
15:19:46 <Brianetta> boekabart: Before round-robin DNS, they were far from random
15:20:03 <Brianetta> You connected to your local efnet server, which was in your company or university
15:20:10 <Noldo> IRCnet has + and ! channels too
15:20:19 <Brianetta> and channels there could be properly local
15:20:46 <boekabart> is it very obvious that I never was in the University Unix club?
15:21:08 <Brianetta> Just obvious that you weren't on the net way back then
15:21:30 <Brianetta> I remember when the first Windows users came onto IRC
15:21:40 <boekabart> actually I was, both... I remember having a @stack.tue.nl email address....
15:21:42 <Brianetta> Mardam-Bey was almost assassinated, I recall
15:21:53 <Noldo> our computer club has a local channel on unis irc server
15:22:09 <Brianetta> The first version of mirc was cool
15:22:32 <Brianetta> All these whiney Windows users came online, yacking on about games and stuff, with no hacker values or appreciation of net culture
15:22:46 <Brianetta> because if you just said a Ctrl-A into the channel, their clients crashed
15:22:49 <boekabart> up to now, I haven't found a really decent IRC client for win
15:22:58 <boekabart> using gaim/pidgin nowadays
15:23:14 <Brianetta> There isn't a really decent *anything* for win
15:23:19 <Brianetta> or lose, as I think of it
15:23:40 <boekabart> no faildows sounds better
15:23:55 <Brianetta> Win is to lose as succeed is to fail
15:24:05 <Brianetta> It isn't Succeeddows
15:24:26 <valhallasw> Brianetta: InDesign works
15:24:29 <boekabart> wait. msvc is decent for win, actually
15:24:29 <peterbrett> Windows Vista is a particularly good joke.
15:24:46 <valhallasw> *insert random game* works
15:25:04 <Brianetta> Yeah, sure. No recent game works on my XP box.
15:25:30 <boekabart> I have a p4 3.2 with a geforce 6600... can't even fly flightsim 10 decently
15:25:40 <valhallasw> Maybe you should install DirectX :p
15:25:55 <boekabart> no, i'll just play around with ottd
15:26:33 <Brianetta> valhallasw: Actually, I think it has more to do with my 1800 CPU
15:26:41 <Brianetta> Whereas on my PSP, all PSP games work.
15:26:50 <Brianetta> The PC isn't a gaming platform any longer.
15:26:51 <peterbrett> Microsoft are getting really desperate, judging by their patent trolling against FLOSS
15:26:55 <Brianetta> Fewer and fewer games are available.
15:27:24 <Brianetta> Wander into Game, and you get two stacks. Latest releases, and budget.
15:27:34 <Brianetta> The shop must have 100 stacks.
15:28:13 <valhallasw> Brianetta: I agree with you console gaming is taking a leap
15:28:59 * boekabart is dreaming about a xbox360 port of openttd.. which you play with keyboard and mouse
15:29:10 <Brianetta> So when there are no more PC games
15:29:16 <Brianetta> there will be no more use for Windows
15:29:29 <Brianetta> and Mac will become the new evil
15:29:35 <Brianetta> and Linux will fight Mac instead
15:29:44 <boekabart> because without evil there can be no good?
15:29:47 <hylje> microsoft is migitating that with xbox30 and easily portable games between the platforms
15:30:05 <Brianetta> Yeah.. and their QA doesn't compare with Nintendo's
15:30:37 <peterbrett> hylje: Did you mean mitigating?
15:31:06 <peterbrett> Nintendo is absolutely kicking ass in the consoles arena
15:31:52 <hylje> peterbrett: wow, never seen that one
15:32:14 <boekabart> brains are weird....
16:09:32 <TrueBrain> -Codechange: introduced ZOOM_LVL_MIN and ZOOM_LVL_MAX for the obvious reason
16:09:32 <TrueBrain> -Codechange: introduced ZOOM_LVL_DETAIL to show/remove details
16:09:32 <TrueBrain> -Codechange: changed << and >> operator with ZoomLevel to a simple wrapper (
16:09:33 <TrueBrain> that in theory also allows zoom-in besides the current zoom-out)
16:09:33 <TrueBrain> -Fix r9845: missed some int -> ZoomLevel
16:09:35 <TrueBrain> That it should have been...
16:10:42 <DorpsGek> TrueBrain: Commit by truelight :: r9846 /trunk/src (13 files) (2007-05-15 16:08:46 UTC)
16:10:44 <DorpsGek> TrueBrain: -Codechange: introduced ZOOM_LVL_MIN and ZOOM_LVL_MAX for the obvious reasons
16:10:45 <DorpsGek> TrueBrain: -Codechange: introduced ZOOM_LVL_DETAIL to show/remove details
16:10:46 <DorpsGek> TrueBrain: -Codechange: changed << and >> operator with ZoomLevel to a simple wrapper (that in theory also allows zoom-in besides the current zoom-out)
16:10:47 <DorpsGek> TrueBrain: -Fix r9845: missed some int -> ZoomLevel
16:14:50 <TrueBrain> peterbrett: and now we can add new zoom levels in a very simple way
16:14:57 <TrueBrain> as the compiler immediatly starts bithcing if you forget a case somewhere :p
16:15:20 <TrueBrain> (or you assert directly :p)
16:16:16 <CIA-1> OpenTTD: truelight * r9846 /trunk/src/ (13 files):
16:16:16 <CIA-1> OpenTTD: -Codechange: introduced ZOOM_LVL_MIN and ZOOM_LVL_MAX for the obvious reasons
16:16:16 <CIA-1> OpenTTD: -Fix r9845: missed some int -> ZoomLevel
16:21:23 <gregor> MD5 of TRGT.GRF is ****INCORRECT**** - File Corrupt.
16:21:25 *** setrodox has joined #openttd
16:25:52 <glx> you may have applied a grd exe on it
16:38:42 *** Osai^2 is now known as Osai
16:43:55 *** Osai^2 is now known as Osai
17:02:21 *** skidd13 has joined #openttd
17:05:55 *** skidd13 has joined #openttd
17:13:52 *** Osai^2 is now known as Osai
17:16:37 *** Osai^2 is now known as Osai
17:19:27 <_42_> MeusH, Sleepie (~Sleepie@p54b37224.dip.t-dialin.net) was last seen quitting #openttd 17 hours 6 minutes ago (15.05. 00:12) stating "Quit: 3...2...1...quit" after spending 1 hour 54 minutes there.
17:19:38 <DorpsGek> TrueBrain: sleepie was last seen in #openttd 20 hours, 41 minutes, and 57 seconds ago: <Sleepie> hmm, I think it is a very special kind of transport so I'm not sure if their is really a need for it in OTTD
17:19:57 <TrueBrain> Difference between being here, and really talking :p
17:20:36 <TrueBrain> fast, commit many tihngs!
17:22:08 <Belugas> stange... <I> think <WE> found a bug...
17:22:18 <maddy> too bad that brianetta isn't here, he could look up in the logs what exactly happend
17:22:51 <peter1138> you shut down his server again?
17:22:52 <Belugas> <I> think, but <WE> do not know that <I> think...
17:22:54 <maddy> Belugas, i think that we (my associate and me) found it ;)
17:23:03 <TrueBrain> Belugas: exactly :)
17:23:08 <Belugas> so, we think we've found a bug!
17:23:12 <maddy> peter1138, just because we tried to demolish a bridge
17:23:15 <TrueBrain> Belugas: only he thinks it
17:23:18 <TrueBrain> the other guy has no clue :p
17:24:18 <Belugas> anyway, maddy, please explain what you've found
17:25:01 * Belugas likes "you", as it could be singular or plural without distinction
17:25:23 * peter1138 likes Belugas cos he's finished newindustries
17:26:27 *** boekabart has joined #openttd
17:26:28 * Belugas should detect british humour by now :S
17:26:42 <hylje> british humour is most silly
17:27:02 <peter1138> that is the point of humour...
17:27:02 <geoffk> hylje, wait til i start mies so bad
17:27:04 <boekabart> Can't get my wife to not-loathe it.
17:27:10 <maddy> Belugas, last night it was a wooden bridge with a train on it and my asso tried to demolish it *crash*, and just 5 minutes ago was it a steel bridge with a train on it *craseaswell*
17:27:12 <Belugas> no, not silly. Just different
17:27:59 <Belugas> maddy : which version of program?
17:28:39 <Belugas> i think it has already been fixed in nightlies, iirc
17:29:04 <maddy> then we try to not demolish a bridge with a trzin on it XD
17:29:31 <boekabart> maddy: that's silly anyway
17:29:35 <Belugas> yup, a trzin as well as a train ;)
17:29:46 <maddy> boekabart, don't tell me...
17:30:04 * boekabart just had to use that word one more time
17:30:47 * Belugas thinks he might as well shuts up, since his rate of type is quite high
17:31:04 <Belugas> furthermore, i have stuff to finish...
17:31:26 *** Wolf01 is now known as Guest76
17:31:46 <Wolf01> i think OFTC kick me when i say idiocies :\
17:32:05 <maddy> Belugas, NewIndustries is finished?
17:32:21 <TrueBrain> see what you did peter1138
17:33:13 <Wolf01> Belugas, NewIndustries is finish..ehm can it allow industries on slopes?
17:33:19 <yeti_> who of you broke zooming in in latest trunk ;) ?
17:33:41 *** skidd13 has joined #openttd
17:34:00 <Belugas> unfortunately, i cannot tell you, Wolf01, since it's not finished yet :D
17:34:17 <Belugas> but surely it will, given the grf authors provided the callbacks for it
17:34:47 <yeti_> according to the svn log, it was truelight ;)
17:34:54 <Wolf01> why cia logs site is dead?
17:35:22 <TrueBrain> yeti_: what is broken about it?
17:35:41 <Wolf01> i hope is not another dns problem... i'll kill somebody tomorrow
17:36:30 <yeti_> TrueBrain: on my linux build, when i zoom in, it doesn't zoom in, and when i try to scroll around the map when "zoomed in", the whole screen turns into a big mess
17:36:55 <yeti_> TrueBrain: also, in the city windows, the city does not appear any more
17:36:56 <TrueBrain> yeti_: latest version?
17:37:16 <hylje> r9829, linux, zooming works as intended
17:37:33 <TrueBrain> hmm... git <-> svn sync failed badly :s
17:37:40 <yeti_> of course it does in 9829, the zooming code changes were done in 984* :)
17:39:11 <peterbrett> TrueBrain: git?!?! Where!!
17:39:31 <boekabart> Q: how do I set the server password to nothing? (dedicated server, console)
17:39:32 <TrueBrain> I use git for OpenTTD developing
17:39:39 <TrueBrain> just pushing things to SVN doesn't always go as planned
17:40:16 *** Sacro|Laptop has joined #openttd
17:40:35 <TrueBrain> the question now is, where did it go wrong...
17:40:54 * glx tries to find the faulty rev
17:41:39 <TrueBrain> too bad recompiling each revision takes for ever :s
17:41:56 <peterbrett> TrueBrain: Can you put your git clone of the svn master on repo.or.cz so I can clone it?
17:42:14 <TrueBrain> lol, why would I? :p Hehe :)
17:42:31 <DorpsGek> MeusH: MiHaMiX was last seen in #openttd 3 weeks, 0 days, 23 hours, 45 minutes, and 14 seconds ago: <MiHaMiX> Rubidium: err... i don't really know, i'll test it
17:42:41 <TrueBrain> they have many many gits :s
17:42:43 <peterbrett> TrueBrain: By attempt to use git-svnimport keeps stalling on r8253
17:42:58 <TrueBrain> peterbrett: here it only failed in the branch -> branches renaming
17:43:02 <TrueBrain> so I just trashed everything older :p
17:43:11 <peterbrett> TrueBrain: I got over that quite easily using the -b switch
17:43:21 <TrueBrain> peterbrett: which doesn't work as you hoped ;)
17:43:35 <TrueBrain> but okay, I might make the git public in a while
17:43:37 <TrueBrain> first let me find this bug
17:43:41 <peterbrett> TrueBrain: Do you use -m?
17:43:59 <peter1138> yeah, that's really funky
17:44:01 <TrueBrain> glx: it is 9846 itself
17:45:02 <TrueBrain> peterbrett: as git is so easy to clone, it is enough for me to publish a single git via the openttd.org domain
17:45:06 <TrueBrain> the rest it up to who ever cares :p
17:45:17 <TrueBrain> I just need to make some scripts to it keeps in sync with svn
17:45:24 <peterbrett> TrueBrain: Yes, but it'll destroy your bandwidth if you don't make a git daemon available
17:45:38 <TrueBrain> which is all I need to launch ;)
17:45:46 <TrueBrain> personally I use ssh, but... that isn't available for non-devs :p
17:45:55 <TrueBrain> but okay, git has some downsides with OpenTTD
17:46:04 <TrueBrain> lack of patches which state SVN revision
17:46:14 <peter1138> it's crap, basically
17:46:24 <TrueBrain> peter1138: not if we would all switch to git :p
17:46:35 <peterbrett> I use stg to manage patches for another project, and then push them into CVS
17:47:05 <peterbrett> TrueBrain: yes, I know. I'm trying to persuade them to switch to git or mercurial
17:47:17 <peterbrett> Actually, Mercurial might be better for openttd.
17:47:18 <TrueBrain> mercurial.. that is basicly git, not? :)
17:47:28 <peterbrett> Similar ideas, different implementation
17:47:36 <TrueBrain> the web.cgi looked VERY simular :p
17:47:39 <peterbrett> I don't think it's as good, but I'm sure other people would beg to differ.
17:48:08 <TrueBrain> anyway, ssst, I have 12 minutes to find this bug
17:49:29 <boekabart> Q again: how do I set the server password to nothing in the dedicated server console?
17:52:36 <TrueBrain> peter1138: because nobody had a better suggestion back then :)
17:52:46 <TrueBrain> ("" wasn't supported yet)
17:53:30 <TrueBrain> peterbrett: why do you tihnk mercurial might be better for openttd? (I haven't looked it up yet or anything)
17:54:09 <peterbrett> Better Windows support
17:54:17 <peterbrett> And you do have a lot of Windows users
17:54:29 <peterbrett> Basically, git is very Unixy, and uses lots of pipes and forks
17:54:40 <peterbrett> Whereas Mercurial is written in Python
17:54:57 <peterbrett> It's a bit slower, and has fewer really powerful tools written for it
17:55:00 <TrueBrain> yeti_: bug fixed, tnx for reporting :)
17:55:07 <peterbrett> AFAIK there's no StGit equivalent, for instance.
17:55:16 <boekabart> better windows support than TortoiseSVN??
17:55:21 <DorpsGek> TrueBrain: Commit by truelight :: r9848 trunk/src/zoom.hpp (2007-05-15 17:54:46 UTC)
17:55:22 <DorpsGek> TrueBrain: -Fix r9846: how every many times you check something, of course you always switch them at the last moment... UnScaleByZoom was ScaleByZoom and visa versa... (tnx to yeti_ for noticing)
17:55:48 <TrueBrain> better windows support isn't that much harder ;)
17:55:52 <glx> just in time for the nightly
17:56:01 <yeti_> TrueBrain: thx for fixing, i'll try it :)
17:56:10 <boekabart> hm, i'm pretty happy with tortoisesvn.. works like a charm
17:56:27 <TrueBrain> SVN is nice and stuff, but not for development on the skill I want to :)
17:56:28 <Wolf01> TrueBrain, i hope now it will be easier to make new zoom levels
17:56:34 <peterbrett> TrueBrain: I had a nice conversation with Linus about that. He said that it wasn't his fault Windows was broken.
17:56:53 <TrueBrain> peterbrett: that is not the first time I hear that argument, and how much it might or might not be true, we have to deal with Windows
17:57:19 <peterbrett> TrueBrain: Quite. Which is why I suggest OpenTTD might prefer hg over git
17:57:40 <peterbrett> Mercurial. Hg = standard symbol for Mercury
17:57:48 <peterbrett> Also the name of the command
17:58:22 <TrueBrain> hg does have revision numbering :p
17:59:38 <TrueBrain> it looks really simular to git :s
18:00:20 <peterbrett> TrueBrain: "A revision number is a simple decimal number that corresponds with the ordering of commits in the local repository. It is important to understand that this ordering can change from machine to machine due to Mercurial's distributed, decentralized architecture."
18:00:39 <peterbrett> i.e. Mercurial revision numbers mean, well, just about nothing
18:00:44 <TrueBrain> peterbrett: which is fine, as OpenTTD will have a single centralized system that serves all things
18:00:48 <TrueBrain> so it means enough :)
18:01:15 <peterbrett> TrueBrain: But each developer will have their own repo which may or may not have numbers matching hg.openttd.org
18:01:37 <TrueBrain> peterbrett: but that is not important, as I assume that on every synchronize, the one from hg.openttd.org is used
18:01:47 <TrueBrain> if that is not true, it is useless :p
18:01:49 <Rubidium> why is everybody so into other source code managament tools? Don't think SVN will handle a 10 000 commits?
18:02:04 *** ChanServ sets mode: +o Bjarni
18:02:06 <peterbrett> Rubidium: SVN isn't as nice as git for things like maintaining patchsets
18:02:06 <TrueBrain> Rubidium: I personally think SVN sucks, I have like 100 checkouts, taking up a lot of space
18:02:26 <TrueBrain> the tons of patches get lost in .patch files, which are impossible to sync up in a simple way
18:02:33 <Bjarni> don't switch stuff when it works :p
18:02:38 *** Bjarni was kicked by DorpsGek (TrueBrain)
18:02:38 *** ChanServ sets mode: +o Bjarni
18:02:42 <Rubidium> TrueBrain: you need stsvn or something like that :)
18:02:43 <boekabart> Well the biggest disadvantage of SVN is managing your own and other peoples patches. Like TrueBrain says, you need a checkout for every one of them... or it's impossible to do version control on your patches
18:03:15 <peterbrett> stgit makes it SO EASY to manage patches it's actually mind boggling
18:03:22 <TrueBrain> anyway, I am using git now, what, 2 weeks, and I am really happy with it
18:03:26 <Rubidium> don't know whether it exists for svn
18:03:31 <TrueBrain> my own personal branches
18:03:35 <TrueBrain> many commits to make up 1 SVN commit
18:04:04 <TrueBrain> no worries when I want to try something, I can always revert
18:04:16 <TrueBrain> (trying things inside a patch)
18:04:38 <TrueBrain> one more time and you both go
18:05:17 <TrueBrain> Rubidium: I personally rather make a local branch then use stgit :)
18:06:27 <peterbrett> TrueBrain: Yeah, that works too. But when you have a set of changes which depend on each other, and you want to commit them into the main VCS in a sane order, being able to go backwards & forwards through the patches changing them is pretty damn handy
18:06:54 <TrueBrain> and what I like about git too, is that you can make a local branch, share it with someone else and work together on something
18:06:58 <TrueBrain> without doing that directly in SVN
18:07:17 <TrueBrain> (as SVN is nice and stuff to do so in a branch, but somehow it does suck when you are just trying something)
18:07:22 <peterbrett> Doesn't pasky work on OpenTTD? IIRC he maintains cogito.
18:07:26 <TrueBrain> so I guess I miss most the localbranch in SVN :)
18:07:43 <TrueBrain> he does something with git yes, we noticed last time :p
18:07:48 <TrueBrain> and he did newgrf work I bleieve
18:09:25 <TrueBrain> anyway, I have to look into how to get git done on OpenTTD in a nice way
18:09:34 <TrueBrain> currently it kind of sucks, mostly because of the SVN link :)
18:10:01 <TrueBrain> (when there is a SVN commit, I need to update the svn git, my local git on openttd.org, rebase my local branches on openttd.org, then sync it with my working station :p)
18:10:04 <peterbrett> Why not use git as the backend, and put a svn frontend on it?
18:10:17 <peterbrett> There's a (IIRC pretty good) git-backed SVN server available
18:11:30 <TrueBrain> which results in the above complications :)
18:12:06 <TrueBrain> an other annoying thing: when I have a localbranch, and I commit that to master, to svn-git, to SVN, and update back again
18:12:13 <TrueBrain> I have a conflict in the localbranch from which the change came
18:12:19 <TrueBrain> as the commit has an other uid
18:12:29 <TrueBrain> so I need to reset the branch the change came from :p
18:12:39 <TrueBrain> (which I of course mostly forget to do)
18:19:16 *** Mucht_ is now known as Mucht
18:26:19 *** Sleepie has joined #openttd
18:31:10 <skidd13> Can anyone of the dev's comment my order movement patch?
18:41:31 <TrueBrain> peterbrett: I am redoing all git dirs and stuff as we speak, it should be ready in some hours for public usage
18:45:39 <peterbrett> TrueBrain: That is full of win
18:46:03 <TrueBrain> sadly enough the git-daemon doesn't allow pushing
18:46:22 <peterbrett> TrueBrain: I don't think that's a big problem
18:46:44 <peterbrett> I'm sure contributors can use the normal patch-submission process for small changes
18:46:46 <TrueBrain> it would have been nice if we could open up a git where everyone could push their branch to
18:47:12 <TrueBrain> as we had a talk about that some time ago
18:47:22 <TrueBrain> but it really needs either http via webdav or ssh...
18:47:28 <peterbrett> If you push the openttd repo to repo.or.cz
18:47:53 <peterbrett> Then it would save you bandwidth ;)
18:48:01 <TrueBrain> bandwidth is not really an issue
18:49:16 * Belugas is reading skidd13'slast opus
18:50:04 <TrueBrain> I even doubt git usage will show in the graphics..
18:50:37 <TrueBrain> (bandwidth graphics that is)
18:51:14 <peterbrett> What sort of bandwidth do you guys get through? Who bankrolls you?
18:51:28 <TrueBrain> Google Ads bankrolls it for 50% or something
18:51:35 <TrueBrain> the rest is on every page :p
18:52:57 <TrueBrain> Donations do currently not go into hosting costs
18:53:53 <TrueBrain> he handles tt-forums, which is partly by donations and google ads I believe
18:54:46 <TrueBrain> (but okay, tt-forums isn't directly openttd related :))
18:55:13 <TrueBrain> reminds me I should update the hosted by thingy..
18:55:34 *** boekabart has joined #openttd
18:56:47 <TrueBrain> pff, svn import to git takes for ever :p
18:56:59 <TrueBrain> orudge: nothing important :)
18:57:22 <hylje> i for one know how you liek highlights :p
18:59:01 <TrueBrain> lol, there are 200k visits on www.openttd.org a month :)
18:59:16 <TrueBrain> not bad for a simple game :p
18:59:42 <peterbrett> TrueBrain: I'm surprised you haven't received legal threats yet
19:00:55 <peterbrett> From whichever shell company currently "owns" the TTD IP
19:01:03 *** |Jeroen| has joined #openttd
19:01:35 <peterbrett> Transport Tycoom Deluxe "intellectual property"
19:01:44 <TrueBrain> Ah, I was wondering what TTD and IP had in common ;)
19:01:54 <TrueBrain> I believe Belugas knows more about that :p
19:02:17 <glx> Atari is the current owner
19:03:06 <peterbrett> Do you ever worry that they'll turn around and C&D you one day?
19:07:20 * Belugas is amazed by the quality of skidd13's work
19:07:28 <Belugas> and the quantity too...
19:07:32 <Noldo> peterbrett: I supose devs don't worry about it as the code rewriten quite a bit and the graphics a users' problem
19:07:56 <Belugas> Belugas knows about what?
19:08:23 <peter1138> qualtiy not quantity? ;p
19:08:46 <Belugas> both, actually, from my point of view
19:09:00 <Belugas> granted, some code style have to be done
19:09:08 <Belugas> like spaces, tabs etc...
19:09:46 <peterbrett> Belugas: I take it you guys don't use indent (1)
19:09:52 <Belugas> personnaly, apart from that, i don't see anything bad...
19:10:14 <Belugas> peterbrett, we do indent, but using tabs when required
19:10:18 <Belugas> and spaces when required
19:10:25 <peterbrett> Belugas: No, indent(1), the command
19:10:41 <peterbrett> `man indent` on most systems
19:10:41 * Belugas does not know such command
19:11:00 <hylje> no manual entry for indent
19:11:08 <glx> peterbrett: we do it by hand :)
19:11:14 * Belugas is working on winXP, msvc. TAB is a key
19:11:26 <Belugas> nice enough and simple enough to use
19:12:28 <Belugas> [15:02] <peterbrett> Do you ever worry that they'll turn around and C&D you one day? <--- there is already an official query been addressed to Atari
19:12:34 <Belugas> it's not like they are not aware
19:12:41 <peterbrett> Belugas: Glad to hear it.
19:12:44 <Belugas> we are waiting for an answer
19:13:06 <Belugas> as ling lang long as it will please them
19:13:34 <hylje> looks like `indent` could help code submissions
19:14:01 <peter1138> only if you find the magic incantation to make it like we want it
19:14:14 <peterbrett> peter1138: Dammit, you beat me to it ;)
19:14:28 <Belugas> no better tool than your own will
19:14:28 <TrueBrain> which you can't in indent :p
19:14:32 <peterbrett> peter1138: I was just typing out the phrase "magic incantation"
19:14:37 <TrueBrain> there is an other tool that does it better, but still not like we want it ;)
19:14:47 <peterbrett> TrueBrain: `indent` is true wizardry
19:15:02 <TrueBrain> astyle yes :) I even wrote the authors if they could add a switch
19:15:20 <peter1138> and it doesn't do things like complain about 'if (p && *p) ...'
19:15:41 <TrueBrain> so we have to make our own :)
19:15:55 <peter1138> i suppose that would be 'if(p&&*p)' if it was dalestan, heh
19:16:16 <TrueBrain> and if(*p) if you like to write broken code
19:17:06 <peterbrett> I tend to write out the tests explicitly
19:17:19 <peterbrett> It's more verbose, but it's oh so much easier to read.
19:18:06 <peter1138> we like 'if (p != NULL && *p != '\0') ...'
19:18:14 <peter1138> assuming p is a char *
19:18:37 <TrueBrain> peter1138: if (char *p != NULL && *p != '\0')
19:19:00 <peter1138> i don't think that's valid :p
19:19:18 <Belugas> prissi did something that i liked very much, by the way... it was somehting like "if (char *p != NULL && p != '\0')"
19:19:32 <Belugas> two spaces separating conditions
19:20:29 <TrueBrain> mkdir: unrecognized option `--bare'
19:20:51 <peter1138> Belugas: i hate that ;p
19:21:31 <peter1138> less annoying than 'if ( ( p != NULL ) && ( p != '\0' ) )' though
19:21:56 <peter1138> hmm, my big toe still hurts :/
19:22:06 <peterbrett> peter1138: So that's why you're so grouchy ;P
19:23:03 <Rubidium> because he had to work when he was sick?
19:23:50 <peterbrett> Rubidium: Sounds like a good reason to me
19:25:39 <Belugas> peter1138, it is a matter of style. i agree it looks strnage, but it has an advantage over clarity
19:25:51 <Belugas> it's easier for the eye to catch it
19:26:00 <Belugas> although the brain can be as fast
19:26:41 <Belugas> i think he hopes some more critics
19:30:44 *** Rambo^^ has joined #openttd
19:31:18 <TrueBrain> Belugas: like tons of other patch-creators...
19:32:54 <Belugas> skiddl13 is good. he has some good ideas, knows a lot the code, and is very easy to work with. Like Wolf01
19:33:02 <TrueBrain> doesn't take away my comment ;)
19:33:06 <peter1138> Wolf01's a bit pushy ;p
19:33:13 <peter1138> (no offence Wolf01)
19:33:21 <peterbrett> Are there any plans for multi-threaded OpenTTD?
19:33:29 <TrueBrain> OpenTTD is multi threaded!
19:33:33 <TrueBrain> the NoAI branch even more!
19:33:37 <Wolf01> don't worry, i'm stupid too ;)
19:33:41 <peterbrett> E.g. having one thread for pathfinding, one thread for AI ?
19:33:50 <TrueBrain> each AI has his own thread in NoAI
19:34:00 <TrueBrain> okay okay, it doesn't mean anything
19:34:06 <TrueBrain> else only one thread is active at a given time
19:34:11 <Bjarni> but will that be desync free?
19:34:22 <TrueBrain> Bjarni: no, therefor we made it
19:34:41 <TrueBrain> peterbrett: the basics of OpenTTD doesn't allow two threads to be active at the same time with map read/write access
19:34:48 <TrueBrain> there simply isn't any locking on that level
19:34:53 <TrueBrain> nor is it designed with that in mind
19:35:11 <peterbrett> (if you can work out how and whether they work)
19:35:34 <Bjarni> I asked in the wrong way...
19:35:54 <Bjarni> TrueBrain: how do you make sure that using multiple threads will not cause any desyncs?
19:36:03 <TrueBrain> Bjarni: by doing it the right way
19:36:24 <Rubidium> Bjarni: by not running more than one at a time...
19:36:26 <peter1138> Caught signal 11. Server aborting
19:36:30 <Bjarni> and what is the right way (apart from doing it without desyncs)
19:36:44 <TrueBrain> Rubidium: I already told him that, doesn't seem to mean anything to him :p
19:37:02 <Bjarni> what is the point in having multiple threads if only one of them will be active at any given time?
19:37:14 <TrueBrain> who said there was need for any point?
19:37:20 <Bjarni> that will not really make good use of multi core systems
19:37:29 <TrueBrain> absolutely no use at all
19:37:31 <Rubidium> Bjarni: go read like 2 months of IRC backlog
19:37:51 <TrueBrain> I think I will blog it tomorrow or something :)
19:38:00 <Bjarni> damn. Have I been that busy with uni for the past two months?
19:38:33 <TrueBrain> peterbrett: anyway, don't expect real threads in any near future :)
19:39:30 <peterbrett> TrueBrain: Is it on the "not-near-future to-do list"?
19:39:39 <TrueBrain> it is in no todo list
19:39:49 <TrueBrain> as I tried to tell you, the basics of OpenTTD simply do not allow it
19:39:53 <TrueBrain> (which in map-access)
19:42:27 <Eddi|zuHause3> <Bjarni> what is the point in having multiple threads if only one of them will be active at any given time? <- the point is easy switching of contexts
19:42:37 <Belugas> peterbrett, it will require a total rewrite
19:42:45 <TrueBrain> I am so glad one OpenTTD users understands it :)
19:44:08 *** Vikthor has joined #openttd
19:44:13 <TrueBrain> in fact, as we have most map-writes in a nice wrapper, we could add locking ;)
19:44:37 <peter1138> have fun keeping sync for network games
19:44:38 <TrueBrain> we only would need to split up the map in multiple locks, so we can lock only a small part of the map :p
19:44:45 <TrueBrain> no real need, as any client would do the same
19:44:49 <Eddi|zuHause3> parallel threads will be a network nightmare :)
19:45:32 <Eddi|zuHause3> with free interleaving of threads, how can you be sure that all clients do things in the same order?
19:45:41 <TrueBrain> you just need good moments to make sure all threads catch up :)
19:46:01 <TrueBrain> Eddi|zuHause3: we also are sure that clients execute commands in the same order :) All about administration :)
19:46:15 <TrueBrain> but okay, I don't see how threads will be useful in OpenTTD in general
19:46:17 <TrueBrain> saveload was useful
19:46:31 <TrueBrain> and I am sure there are other useful parts
19:46:52 <TrueBrain> (okay, TGP isn't a real thread)
19:46:53 <Eddi|zuHause3> but that "administration" thing would create a pretty big overhead
19:47:09 <TrueBrain> Eddi|zuHause3: depends how you do it, or more: what you expect the threads to do
19:47:17 <Bjarni> anyway I think the point of the original question was if there are plans to make threads to run in parallel and we agree that the answer to that is no, right? (apart from background saving)
19:47:45 <TrueBrain> take for example pathfinding
19:47:53 <TrueBrain> how would the game benifit from that running in a (parallel) thread?
19:48:04 <TrueBrain> (say, SP only for now)
19:48:25 <Zr40> one possible option would be for the GUI to run in a separate thread
19:48:31 <TrueBrain> the only thing that might worth giving time, is splitting screen-updates from gamelogic
19:48:33 <Rubidium> it wouldn't, because the vehicle still has to wait for the parallel thread
19:48:40 <TrueBrain> Rubidium: exactly :)
19:48:43 <Rubidium> Zr40: GUI uses like 0.001% of the CPU
19:48:53 <TrueBrain> Rubidium: more the other way around: if you have 1000 trains
19:48:55 <peter1138> not on my machine it doesn't
19:48:56 *** boekabart has joined #openttd
19:48:57 <Zr40> Rubidium: then why does scrolling when zoomed out spike the CPU usage to 40%?
19:48:57 <TrueBrain> it locks normal window scrolling
19:49:08 <TrueBrain> so if you split it in threads, you can keep the window scrolling realtime
19:49:12 <TrueBrain> while the rest is slower
19:49:28 <Zr40> TrueBrain: not on dual core machines
19:49:44 <TrueBrain> Zr40: I mean: while the rest hogs the CPU
19:49:44 <peter1138> you still have to wait
19:50:01 <TrueBrain> peter1138: not if you make locking around map-access :)
19:50:08 <Bjarni> I wondered about that. Having a thread for each vehicle type and then run all of them in parallel would speed up the game, but I realised that it would only work if the CPU load is evenly spread over multiple vehicle types (sucks for trains only) and it would produce problems like a train and a bus loads at the same station, but not in the same order on all clients in MP, so I discarded that idea
19:50:12 <peter1138> because you can't draw when things are in a semi-set state
19:50:22 <TrueBrain> peter1138: not if you make locking around map-access :)
19:50:30 <peter1138> that'll just be slow ;p
19:50:40 <Bjarni> solving issues like that would take forever and cause us to add a not so nice overhead
19:50:44 <hylje> gui can assume graphical stuff
19:50:56 <TrueBrain> peter1138: map-double-buffer! :)
19:51:04 <TrueBrain> inf act does solve the problem
19:51:08 <TrueBrain> only needs twice the memory ;)
19:51:22 <Bjarni> I wondered about something like that
19:51:25 <peter1138> you also need to buffer all the other data structures...
19:51:28 <Noldo> some copy on write thing there?
19:51:32 <peter1138> well, those changable ones
19:51:33 <TrueBrain> peter1138: okay okay, true :)
19:51:40 <Bjarni> but I didn't really figure out if it were worth the time coding it
19:51:56 <TrueBrain> so we end up with a game twice the memory, just so you can scroll with 1000 trains :p
19:52:01 <TrueBrain> You know what... let's forget it :)
19:52:28 <Zr40> something completely different... any progress on FS#104? :)
19:52:39 <Bjarni> "..., so I discarded that idea" I already said that
20:11:18 <Zr40> or put differently, anything I can do to have it move forward? :)
20:13:37 <Belugas> some lost tabs again, Zr40
20:14:04 <Zr40> where? I was sure I've removed them all
20:15:07 * Zr40 can't find any stray tabs
20:15:43 <Belugas> well.. i have this text editor called notepad2. it shows tabs as red long arrows...
20:15:53 <Belugas> i'l give you a screenshot. wait
20:16:11 <Zr40> I have this text editor called vim. It doesn't locate any \t within the widget array
20:18:02 <Zr40> if you take a look at the original file, you'll see tabs
20:18:38 <peter1138> the lines which are just tabs, of course
20:18:38 <Belugas> i never said anything about the widget array, didn't i?
20:18:39 <Zr40> according to grep, tabs are _everywhere_
20:18:45 <Belugas> yeah! the had to go :D
20:19:04 <Belugas> don't trust tools, trust your eyes
20:19:14 <Zr40> I don't see any tab-only lines in your screenshot
20:19:15 <peter1138> strange, 'my' vim automatically removes those lines
20:19:39 <Zr40> the green highlighting threw me off guard
20:20:36 *** boekabar1 has joined #openttd
20:21:23 <glx> tabs in empty lines (aka trailing whitespaces)
20:26:24 *** skidd13 has joined #openttd
20:27:05 * Belugas forgot he is running Delphi and not some interpreted languages... code does not magically recompile :S
20:27:50 *** Vikthor has joined #openttd
20:55:45 *** G0D_aw is now known as |2rB
20:56:38 <Sleepie> skidd13: just a question have you made any progress on your HQ's?
21:17:39 <skidd13> @sleepie: read your question just yet... The coding of the current ones is fininshed. The coder will releas them in the near future.
21:21:11 <Sleepie> skidd13: nice to hear, will be fun to see them in game
21:29:32 <Wolf01> what about the adjoin stations patch?
21:51:52 <Eddi|zuHause3> hmmm... some recent update changed the IO scheduler from anticipatory to cfq, since then, the system is often unresponsive when copying large files...
21:57:04 <yeti_> Eddi|zuHause3: strange, i've always had better results with cfq, especially regarding responsiveness (responsivity?)
22:01:37 *** Osai is now known as Osai^zZz
22:05:45 *** Osai^zZz is now known as Osai
22:24:34 <Bjarni> 4 is found by using a die and we all know that dies are random, right?
22:24:49 <SpComb> int getRandomNumber() { /* generated via fair dice throw */ return 4; }
22:25:18 <Eddi|zuHause3> Bjarni: you mean "dice"?
22:25:51 *** Twofish has joined #openttd
22:26:26 <SpComb> hmmmm.... 1980s 'intelligent pop'
22:27:19 <Bjarni> and do I want to know? :)
22:28:11 <Bjarni> sounds like a cat fight outside
22:29:22 *** mikegrb has joined #openttd
22:29:40 <Sacro|Laptop> Eddi|zuHause3: no he doesn't
22:29:58 <Sacro|Laptop> he only threw the one, hence "die" is correct
22:31:03 <peter1138> Sacro|Laptop: dies -> dice
22:31:12 <Sacro|Laptop> peter1138: i know
22:31:22 <peter1138> hence "dies" is incorrect ;p
22:31:43 <Sacro|Laptop> he used both in the sentance
22:31:57 <Bjarni> then again Sacro|Laptop might be more used to dies than dice
22:32:04 <peter1138> enough reimplemnting pbs for yapf ;p
22:32:06 <Bjarni> he lives in Hull after all
22:32:31 <Sacro|Laptop> network is killing my tv
22:35:18 <Bjarni> you have the porn channel?
22:35:32 <Eddi|zuHause3> "the" porn channel?
22:35:49 <Bjarni> well, in Denmark we have one porn channel
22:36:09 <Eddi|zuHause3> in germany, distributing porn over TV is illegal
22:37:03 <Bjarni> TV is broadcasting. You have no control of who watches what you transmit, so it kind of make sense
22:37:50 <Bjarni> here a TV station had a technical issue, so the computer decided to show hardcore porn instead of the children show
22:37:57 <Eddi|zuHause3> often you'll find them broadcasting softporn during the night
22:38:00 <Bjarni> a whole lot of people got mighty upset
22:38:37 <Bjarni> that is what you get for trusting a computer to broadcast TV and then go home
22:38:39 <Eddi|zuHause3> you might classify those more as "gymnastic exercises" than "softporn" :p
22:39:12 <Bjarni> so kind of like naked pole jumping?
22:39:33 <Eddi|zuHause3> with synchronised "oh"s and "ah"s :p
22:41:05 <Rubidium> they synchronize that too?
22:41:23 <Bjarni> so they are broadcasting adult comedy at night then
22:42:07 <Eddi|zuHause3> they synchronise _everything_ :p
22:42:39 <Eddi|zuHause3> i'm actually surprised they did not synchronise on MTV :)
22:43:30 <Eddi|zuHause3> Bjarni: actually, there was a swiss-german movie, and they synchronised it in high-german :)
22:44:36 <Bjarni> here the only synchronisation is done to movies where the target audience are too young to read the subtitles
22:44:45 <Rubidium> it's not a worse as synchronising someone speaking Dutch with English and then subtitling it in Dutch
22:44:50 <Bjarni> it makes the experience of watching TV a whole lot better
22:45:07 <Rubidium> and you actually learn better english ;)
22:45:37 <Rubidium> ofcourse Germans are (or at least were) known to make their own words instead of taking words from other languages
22:45:50 <Bjarni> so you are saying that the Germans learn Engrish from their synchronisations?
22:46:26 <Rubidium> no, that unsynchronized is better for your English
22:50:46 * Rubidium understands pretty much of it ;)
22:51:05 <Bjarni> it's not like it's meant to be tricky
22:51:11 <Rubidium> really stupid that the replace all references to Vietnam with something else ;)
22:51:16 <Bjarni> you see, the Germans are meant to understand it :p
22:52:08 <Eddi|zuHause3> yeah, that's explained (a little) later in the article
22:53:17 <Bjarni> I read that in the German sync of Monsters Inc, the Yeti talks with a Swiss dialect and instead of whatever he says in English, he states that he wanted to be in the Alps instead because he has relatives there :D
22:53:55 <Eddi|zuHause3> i don't remember much about Monsters Inc
22:54:24 <Bjarni> I never saw the German version. I just read this on IMDB or somewhere like it
22:58:09 <Bjarni> anyway I better get some sleep now
22:58:17 <Bjarni> I didn't really sleep last night
22:58:26 <Bjarni> in fact I went to bed after sunrise :(
23:09:59 *** Sleepie_ has joined #openttd
23:10:32 *** Sleepie_ is now known as Sleepie
23:19:11 *** TheMask96 has joined #openttd
23:44:37 *** TheMask96 has joined #openttd
23:54:55 *** TheMask96 has joined #openttd
23:54:56 *** Ammller has joined #openttd
continue to next day ⏵
