IRC logs for #openttd on OFTC at 2012-07-21
⏴ go to previous day
00:07:17 <Sleepie> bedtime for me now, g'night
00:48:30 *** brambles has joined #openttd
00:57:15 *** brambles has joined #openttd
02:13:12 *** Supercheese has joined #openttd
04:13:22 *** HerzogDeXtEr has joined #openttd
04:22:29 *** Supercheese has joined #openttd
04:56:17 *** Eddi|zuHause has joined #openttd
05:56:46 *** sla_ro|master has joined #openttd
06:01:20 *** Zeknurn has joined #openttd
06:26:47 <Wolf01> uhm, I'm trying to open the in-line volume control of my stereo headpohones because it has a false contact, I can't find a weak spot, I think I'll resort to violence
06:39:00 <Wolf01> opened! the good new is that the cables are ok, the bad new is that the problem seem to be the trimmer :(
06:42:57 *** Alberth has joined #openttd
06:42:57 *** ChanServ sets mode: +o Alberth
06:45:39 <Alberth> it's quiet, no new posts in the openttd forums even
06:50:41 *** telanus has joined #openttd
06:57:23 <Rubidium> Alberth: then continue with zbase and let your computer make some noise
07:00:38 *** tokai|mdlx has joined #openttd
07:07:37 <Alberth> that would be mostly the fan, as building zbasebuild takes ages :)
07:08:09 *** Devroush has joined #openttd
07:09:36 *** LordAro has joined #openttd
07:40:37 *** linux_probe has joined #openttd
08:18:30 *** frosch123 has joined #openttd
08:19:04 *** telanus1 has joined #openttd
08:27:35 *** telanus2 has joined #openttd
08:46:20 *** Sleepie has joined #openttd
08:50:29 *** Chris_Booth has joined #openttd
08:51:45 *** TGYoshi has joined #openttd
09:04:22 *** Progman has joined #openttd
09:06:11 *** Markavian has joined #openttd
09:17:38 *** tokai|noir has joined #openttd
09:17:38 *** ChanServ sets mode: +v tokai|noir
09:56:39 *** FLHerne has joined #openttd
09:59:10 <frosch123> hmm, what debian package might contain /usr/include/mysql/mysql.h
09:59:33 <frosch123> libmysqlclient-dev maybe
10:00:02 <frosch123> hmm, oh, i already have that file
10:00:43 <frosch123> no, i do not... wrong window
10:03:27 <Alberth> just switch windows until you find it :)
10:07:27 <frosch123> does svn 1.7 support externals for single files meanwhile?
10:15:09 *** Biolunar has joined #openttd
10:17:55 *** FLHerne has joined #openttd
10:18:43 <Rubidium> though it doesn't seem to exist for sid; it does exist for wheezy and experimental
10:26:05 <frosch123> symbolic links are already supported by svn 1.4
10:26:20 <frosch123> but that's not an external
10:30:30 <Sleepie> yes more a kind of internal
10:32:17 <Alberth> If 1.6 didn't have it, I wouldn't expect it in 1.7, they have problems enough removing all .svn directores in all sub-root directories
10:33:51 <Sleepie> which is fortunately the case since 1.7
10:39:58 <frosch123> ottd 0.7 looks awkward
10:41:20 <Alberth> try 0.6, it has a funny AI
10:45:02 *** Arafangion has joined #openttd
11:06:21 <Alberth> FLHerne: your harbour picture in the CHIPS thread seems to suggest to me you want concrete under your rail tracks
11:14:20 <FLHerne> Alberth: Between the rails in CHIPS? Or is that a comment on NuTracks ballast? Or both? :P
11:15:14 <Alberth> under the diagonal tracks + crossings north of the ship depot
11:16:09 <FLHerne> Not really - that's a 3rd-rail electrified mainline :P
11:16:20 <Sleepie> maybe even only under the diagonals
11:16:26 <FLHerne> Would be useless as a loading pad :P
11:17:00 <FLHerne> I'd like half-tile triangles to fit between the station and mainline though :-)
11:17:03 <Alberth> you see often rail tracks fully embedded in the pavement in industrial areas
11:17:31 <FLHerne> Alberth: Yes, but not 90mph mainlines :P
11:17:43 <FLHerne> That would be dangerous :-(
11:18:02 <Alberth> the concrete may jump up :p
11:18:24 <FLHerne> People would step on the 3rd-rail, too...
11:18:34 <Alberth> yeah, not a good idea
11:20:04 * FLHerne considers triangle-overlapping tiles
11:20:49 <Alberth> I never make eye-candy, so don't ask me :)
11:20:55 <Sleepie> would look more pleasing imho
11:21:30 <FLHerne> Ok. Should be easy to draw. Was short of object ideas anyway :P
11:21:40 * FLHerne goes out to walk the rat
12:03:19 *** FLHerne has joined #openttd
12:06:13 <FLHerne> It's finally stopped raining :D
12:07:35 *** LordAro has joined #openttd
12:08:44 *** LordAro has joined #openttd
12:11:51 <Alberth> FLHerne: what I think it needs are stacks of things and stuff. The docks are way too clean
12:12:00 *** LordAro has joined #openttd
12:12:48 <FLHerne> Alberth: That's only because the train just left :P
12:13:49 <Alberth> it did not even leave empty pallets?
12:15:01 <FLHerne> Tell andythenorth to make it do that, then...?
12:15:30 <Alberth> I remembered someone had run out of objects to draw :p
12:16:41 <FLHerne> But objects can't measure cargo waiting at all, that's a station kind of thing
12:17:08 *** LordAro has joined #openttd
12:17:13 <Alberth> some stacks are part of the docks :p
12:17:30 <FLHerne> That's a dock thing, then :P
12:17:43 <Alberth> oh, fishing nets can be added too
12:21:47 *** Stimrol has joined #openttd
12:22:01 <Sleepie> and I would also overbuild the left shores in the habour
12:24:13 <FLHerne> Still a dock thing. I'm just using/modifying/drawing variations of various sprites from CHIPS & FIRS and recoding them as objects
12:24:29 <FLHerne> Sleepie Exactly for that reason :P
12:24:59 <FLHerne> CHIPS can't do that, that's why I need to finish doing the objects :P
12:25:12 <Alberth> 'various sprites' don't have cargo lying around?
12:26:43 <FLHerne> Alberth: They do. andythenorth has them for 'amount of cargo waiting' in CHIPS
12:27:11 <FLHerne> That picture demonstrates the need for NewObjects, not their use :P
12:28:13 <Sleepie> yep the only downside is that the list of newgrfs loaded gets longer and longer until you reach the current limit
12:28:56 <Sleepie> which afaik cannot be lifted, because it would break multiplayer
12:29:43 <FLHerne> Chill's PP has a much higher limit, and doesn't break multiplayer :P
12:30:17 <Sleepie> I don't know never played CPP
12:31:17 <frosch123> FLHerne: ofc it does
12:32:09 <frosch123> it will only transmit the first part of the grfs and drop the other ones
12:32:17 <frosch123> so, once you use too many grfs, multiplayer will fail
12:32:35 <Sleepie> because it only uses one packet for it iirc
12:32:56 <frosch123> more specifically a udp packet
12:33:15 <FLHerne> In that case, add separate limits for single-player games and servers? :P
12:33:17 <frosch123> though maybe it only breaks the server list
12:33:32 <Sleepie> so question is can this be changed "easyly"
12:33:58 <Alberth> FLHerne: you don't want to have different game files for SP and MP, I think
12:34:21 <frosch123> Sleepie: experience tells that everyone who reaches the newgrf limit does not know what he is doing with them
12:34:30 <frosch123> it's like the mapsize limit
12:34:42 <frosch123> there are some people requesting 16kx16k maps
12:34:48 <frosch123> while 2kx2k is already utterly useless
12:35:10 <FLHerne> Alberth: Keep the files identical, but forbid loading ones with >x NewGrfs on a server
12:35:35 <Alberth> FLHerne: how? a savegame contains references to all loaded newgrfs
12:35:36 <FLHerne> frosch123: I hit the limit, and I know what I'm doing with them :P
12:35:54 <Sleepie> frosch123: personally I think that was true in the past, but now there are a lot of little grfs popping up
12:36:21 <Alberth> FLHerne: as said, I don't think you want some magically unloadable files just because you use them in a different context
12:36:53 <Sleepie> and even if they are just eyecandy you can reach the limit quite fast without loading everything available
12:37:47 <FLHerne> Alberth: Add a parameter for 'max SP NewGRFs', at MP limit by default?
12:38:04 <frosch123> anyway, noone stepped up to even fix fs#5158
12:38:10 <FLHerne> Then it wouldn't confuse people, especially with these fancy description strings :P
12:38:43 <Alberth> FLHerne: x.sav by itself does not show whether it is loadable in MP
12:38:48 *** Stimrol has joined #openttd
12:39:49 <FLHerne> Alberth: It doesn't show that it needs obsure_unavailable_thingum_v283.grf either :P
12:39:50 <Alberth> But the problem is better solved by a fundamental fix than this hacking into different types of save games
12:40:33 <Alberth> FLHerne: yeah, it's bad, I'd like to fix that too, let's throw away newgrfs
12:40:49 <Alberth> which by the way solves your loading problem nicely too
12:40:59 <frosch123> just restrict loading of newgrfs to a single one
12:41:15 <frosch123> one baseset, one newgrf, one gs, one ai
12:41:27 <FLHerne> Alberth: Removes the developers' excuse of 'that can be done in a convoluted way by NewGRFs' though :P
12:41:29 <Sleepie> then the baseset must be extended massively :P
12:42:33 <Sleepie> for it would help if many of the little grfs could be combined just to a few ones
12:43:45 <Alberth> it probably gets terribly complicated, much more than fixing the limit would be my guess
12:44:38 <Sleepie> yeah probably, also all people involved must be willing to work together on a bigger thing and so on
12:45:36 <Sleepie> Alberth: what do you think how complicated the change in the network code would be?
12:46:47 <Alberth> You'd have to allow for receiving several UDP packets, with just some of the newgrfs. You can have some missing, and receive some double
12:47:36 <Alberth> you need a mechanism to know how many to expect, and a check whether you got all
12:48:21 <Sleepie> ok so there should be already things that work that way
12:48:57 <Alberth> I don't even know why it uses UDP instead of TCP
12:49:10 <Sleepie> maybe I should look myself at the network code
12:49:10 <Alberth> which would seem easier to me at first sight
12:49:39 <Sleepie> maybe because it works better with weaker connections
12:50:14 <frosch123> Alberth: it's about queryiing all servers without authentication
12:50:55 <Alberth> ah, ok, that explains using UDP :)
12:51:50 <Alberth> Sleepie: so a bunch of UDP packets gives nice options for a DOS attack
12:54:48 *** Stimrol has joined #openttd
12:58:11 <Rubidium> actually, I think a ChillPP MP server with too many NewGRFs will hit an assert, or if it doesn't, never advertise properly
12:59:35 <Rubidium> Alberth: the limit is fixed ;)
13:00:24 <Rubidium> and with the nature of UDP, receiving multiple packets from another party without packet loss is significantly less likely
13:00:25 <Alberth> I was not going to unfix it ;)
13:01:25 <Rubidium> and it uses UDP because a certain operating system has/had a limit of connections that could be made with a single binary (in the non-server version)
13:02:00 <Rubidium> which meant that a server wouldn't be visible anymore after a few hours
13:02:18 <Rubidium> but there is a relatively easy way to solve the whole issue. Just increase the MTU of the internet
13:02:26 *** TWerkhoven has joined #openttd
13:03:24 *** LordAro has joined #openttd
13:05:34 *** Chris_Booth has joined #openttd
13:07:43 <frosch123> Rubidium: submit a patch to the internet authority
13:42:58 <Sleepie> tea time see you later
14:13:43 *** andythenorth has joined #openttd
14:22:14 <andythenorth> can we script map gen?
14:26:29 <Alberth> would that give anything more sane results?
14:27:45 <andythenorth> 'generate 2 large cities for this GS'
14:43:23 <Eddi|zuHause> no, but you should be able to found towns afterwards if there are not enough
14:44:59 <andythenorth> can I remove towns as well? :P
14:59:20 <MNIM> sadly, not without the scenario editor
15:10:22 <Eddi|zuHause> andythenorth: your better bet is to just make a scenario for such a special case
15:10:52 <andythenorth> I made a scenario once
15:11:10 <andythenorth> the most boring thing....ever
15:11:46 <Alberth> at least it is done faster than coding a baseset :p
15:13:10 <andythenorth> coding is interesting
15:13:31 <andythenorth> I might have a few hours of coding time this weekened
15:13:48 * andythenorth considers: nml FISH, or start a GS project?
15:13:58 <Alberth> you know a base set is just real srpites, right? :)
15:14:13 <andythenorth> I have edited opengfx yes ;)
15:14:18 <andythenorth> GS, I need a buddy
15:14:52 <andythenorth> I have 3 or 4 GS ideas
15:15:55 <Alberth> that's too much for a few hours coding
15:16:17 <andythenorth> I was only proposing to do one of them :P
15:16:42 <Alberth> btw, I just read a page at the Django site, looks interesting to try one time
15:17:17 <planetmaker> bananasssssss. my preciousssss
15:24:54 <andythenorth> I have GS ideas like:
15:25:26 <andythenorth> - metro: build a city network capable of transporting x passengers per year
15:25:59 <andythenorth> - transcontinental: build west from a large east coast city to the other coast (large empty-ish map needed)
15:27:15 <andythenorth> - air mogul: become the biggest airline on the map by 1950 (starts before planes are are available, so have to begin with trains etc)
15:28:44 <Alberth> hmm, we could have lots of small islands that you have to bridge. Then you cannot have bridges over tracks.
15:29:46 <Alberth> how is "transcontinental" interesting, just a drag/drop to the other side is enough, isn't it?
15:29:59 <andythenorth> very tight time limit
15:30:09 <Alberth> build while paused :p
15:30:14 <andythenorth> start with only one large city in the east, few industries
15:30:29 <andythenorth> could limit amount of rail allowed to be built
15:30:44 <Alberth> could be interesting if there is a large gap in between where you cannot put much tracks
15:30:54 <andythenorth> there are quite a few transcontinental scenarios in railroad tycoon, they're fun
15:31:15 <andythenorth> I am most interested in scenarios that take 1-3 hours to play
15:31:56 * andythenorth ponders some more
15:34:26 * andythenorth should learn what GS can do, instead of guessing :P
15:35:33 * andythenorth thinks 'less' might be the way to go
15:35:40 <andythenorth> the vehicle sets and industry sets have become huge
15:35:47 <andythenorth> but bigger != more fun
15:36:15 <andythenorth> none of the big newgrf games I've played have been as much fun as the first games I played with vanilla openttd
15:36:34 <andythenorth> but...bigger = more scope for a GS that only uses parts of stuff
15:49:28 <CIA-4> OpenTTD: rubidium * r24427 /trunk/config.lib: -Change: allow passing C(XX)- and LDFLAGS to the compilation of helper binaries such as depend, strgen and settingsgen
16:12:37 *** andythenorth has joined #openttd
16:14:18 *** Stimrol has joined #openttd
16:33:32 *** FLHerne has joined #openttd
16:38:03 *** Chris_Booth has joined #openttd
16:50:41 * andythenorth tries to find docs for nml BITMASK()
16:50:50 <andythenorth> maybe I read the source
16:52:08 <Alberth> but it's not that clear :)
16:52:47 <andythenorth> trying to figure out bitmask(CC_LIQUID)
16:52:48 <Alberth> bitmask(0, 4, 5) == 0011 0001
16:53:07 <andythenorth> so bitmask(CC_LIQUID) sets the liquid bit
16:53:10 <Alberth> ie 1 << CC_LIQUID this
16:53:28 <andythenorth> in some limited cases, nfo is much easier :P
16:53:32 <andythenorth> no abstraction to deal with
16:54:00 <andythenorth> putting convenience around the bitmask makes it harder to remember / deduce what the bitmask is doing
16:54:02 <Alberth> but you have to encode to 0x31 :)
16:54:28 <Alberth> iirc nml also has <<, so use that instead
16:54:29 <andythenorth> so if I wanted to check a cargo had classe liquid, and only liquid?
16:54:48 <andythenorth> all other bits must be clear
16:55:26 <Alberth> value == bitmask(CC_LIQUID)
16:55:47 <Alberth> assuming 'value' is only cargo classes
16:56:20 <andythenorth> theoretical question atm
16:56:43 <Alberth> ie you want exactly the value with that one bit
16:58:37 <andythenorth> that explains why some cargos are refitting to tanker
17:00:51 <andythenorth> so in nfo I would 'just' mask the other bits out and check for liquid bit
17:00:51 <andythenorth> I need to do the same in nml
17:00:51 <andythenorth> the issue is vehicles that should be tanker if liquid class is present
17:00:51 <andythenorth> not iff liquid is the only class
17:02:50 <andythenorth> typo above * some cargos _aren't_ refitting to tanker :P
17:02:54 <Alberth> then you want (vaue and bitmask(CC_LIQUID)) != 0 , ie throw away all bits not in the bitmask, and then you should still have some bits left
17:03:30 <andythenorth> perhaps I need multiple switches in that case
17:03:33 <Alberth> and since you started with 1 bit, 'some bits left' is also 1 bit :)
17:04:00 <michi_cc> And 'value & bitmask(a,b) == bitmask(a,b)' just in case you want to test if two bits are set at the same time.
17:06:21 <Alberth> the 'and' operation forces all other bits to 0 (ie all bits that you mask away), so you can predict what to compare against
17:15:36 *** Chris_Booth has joined #openttd
17:34:08 <andythenorth> so masking in nfo is easy
17:34:14 <andythenorth> but I am baffled by how to do this in nml
17:34:29 <andythenorth> presumably in the switch somewhere
17:37:49 *** Progman has joined #openttd
17:42:27 <michi_cc> andythenorth: By literally writing a &
17:42:58 <andythenorth> cargo_classes_in_consist & some_value ?
17:43:14 <michi_cc> Yes. And that some value is your bitmask.
17:45:18 <CIA-4> OpenTTD: translators * r24428 /trunk/src/lang/ (brazilian_portuguese.txt korean.txt):
17:45:18 <CIA-4> OpenTTD: -Update from WebTranslator v3.0:
17:45:18 <CIA-4> OpenTTD: korean - 26 changes by telk5093
17:45:18 <CIA-4> OpenTTD: brazilian_portuguese - 6 changes by Tucalipe
17:46:09 * andythenorth seeks the & operator in nml
17:47:06 <planetmaker> value = variable & 0x00F0
17:47:27 <andythenorth> how do I create an expression?
17:47:35 <andythenorth> docs tell me only the operators
17:47:45 <andythenorth> I have to wrap it in parentheses?
17:48:15 <Alberth> the normal operator expression priorities apply
17:48:26 <Alberth> ie 1 + 2 * 3 == 1 + (2 * 3)
17:48:38 <andythenorth> using && gets me an nml error
17:48:47 <andythenorth> it expands it to an html entity
17:48:56 <planetmaker> && applies to boolean only
17:49:15 <planetmaker> that's your pre-processor which fails there then
17:51:12 <Alberth> as far as I can see, there seem to be mostly terrain and track sprites left, and stations
17:53:15 * andythenorth wonders wtf the templater is escaping non-templated things
17:58:00 <andythenorth> this is a showstopper :P
18:07:45 *** TWerkhoven[l] has joined #openttd
18:16:25 <andythenorth> ugly, but...meh :P
18:17:03 <andythenorth> grf_nml.write('&&'.join(master_template(vehicles=vehicles, repo_vars=repo_vars).split('$AND')))
18:19:22 <andythenorth> so what range is needed to check for 'true' with 'cargo_classes_in_consist && bitmask(CC_LIQUID)'
18:20:46 <andythenorth> I have tried 0, 1 and 255 as values
18:21:04 <Rubidium> doesn't that yield true whenever cargo_classes_in_consist is not 0?
18:22:12 <andythenorth> yes, it does yield true
18:26:00 <andythenorth> ok that works :)
18:26:19 <TrueBrain> I hate weekends; they are so boring
18:28:20 <TrueBrain> I am sorry, I speak no german
18:28:46 <Alberth> andy does not speak french, so that's a happy co-incidence :)
18:29:57 <Eddi|zuHause> we should all speak esperanto like we're supposed to
18:30:26 <andythenorth> someone here probably does
18:30:55 <Eddi|zuHause> we used to have an esperanto translator
18:30:57 <andythenorth> is it madness that I have to use $AMPERSAND in my nml templates to get an & char?
18:31:11 <andythenorth> should I rip out the entire templating engine and start again?
18:31:29 <Eddi|zuHause> andythenorth: you're probably just missing an escape character
18:32:29 <andythenorth> I tried all the ones listed
18:32:36 <andythenorth> I thought the same though
18:34:44 <andythenorth> I can't use string colours
18:36:42 <Eddi|zuHause> that doesn't look bad
18:37:07 <FLHerne> That second one looks much better :-)
18:37:37 <andythenorth> I could use cyan
18:37:45 <andythenorth> which would hide the vehicle info window problem
18:37:57 <andythenorth> can i have a guarantee that vehicle info window text colour won't change?
18:38:09 <FLHerne> Why is that a problem?
18:38:48 <Alberth> andythenorth: a tree-like list?
18:39:01 <FLHerne> Makes the 'in cabins' bit easier to ignore - which is fine since it's not relevant anyway :P
18:39:46 <andythenorth> it's wrong and ugly
18:39:51 <planetmaker> why do you need that guarantee?
18:40:09 <andythenorth> don't want to change the vehicle info window appearance unexpectedly
18:40:16 * andythenorth may be being too pedantic
18:41:01 <Eddi|zuHause> andythenorth: i don't see your problem...
18:41:23 * Alberth will ask permission for any vehicle window change
18:41:26 <andythenorth> the wrong colour sticks out like a sore thumb to me
18:41:36 <andythenorth> but seems I notice these things more than average
18:41:54 <Alberth> that's what we hired you for :)
18:42:33 <Eddi|zuHause> andythenorth: i think grey works well
18:43:09 <Eddi|zuHause> the hyphen needs a better idea though
18:43:22 <Eddi|zuHause> or at least another space :)
18:44:51 <FLHerne> andythenorth: It's not 'the wrong colour', it's 'a useful idea to make unneeded information less obtrusive' :P
18:49:45 <andythenorth> it's more intrusive
18:50:06 <andythenorth> making it a unique colour brings it forward as you try and parse the gestalt
18:50:22 <andythenorth> unique colour = higher significance in cognition
18:50:26 <FLHerne> andythenorth: It isn't, it blends into the grey background nicely. Less contrast
18:51:34 <FLHerne> The break in colour makes it easier to parse, because it marks the end of the bit I care about
18:51:51 <masch> is there a description of the algorithm thats used for map generation?
18:52:05 <FLHerne> Otherwise the information I needed would be in the middle of the string
18:52:12 <Rubidium> masch: what part of map generation?
18:52:56 <Rubidium> masch: there might be in tgp.cpp (source code); it has to do with Perlin noise
18:53:32 <masch> Rubidium: thanks, default libnoise perlin generates uncool maps that are impossible to display with minecraft
18:53:48 <masch> WTF why did i say minecraft? my head..
18:54:24 <Eddi|zuHause> that's the geek version of a freudian slip :p
18:54:44 <TrueBrain> Minecraft also uses perlin noise
18:55:01 <Eddi|zuHause> perlin noise is very boring for maps
18:55:21 <TrueBrain> I think OpenTTD does an epic job, does it not? :)
18:55:23 <Eddi|zuHause> we need a tectonics map generator
18:56:01 <TrueBrain> Eddi|zuHause: plate tectonics?
18:56:02 <Eddi|zuHause> TrueBrain: it has been improved by the overlay functions that form flat and hilly areas
18:56:09 <Rubidium> too bad TrueBrain didn't finish his map generator that uses very realistic tectonics ;)
18:56:16 <FLHerne> TrueBrain: Not that epic :P
18:56:19 <TrueBrain> Eddi|zuHause: 3D perlin noise, yes :)
18:56:28 <andythenorth> if you want a decent map, first turn off 'variety distribution'
18:56:35 <masch> wow there is nice short introduction how perlin noise works. great
18:56:53 <andythenorth> if you want a decent map, load a heightmap :P
18:57:00 <FLHerne> It needs better links with river generation :P
18:57:04 <TrueBrain> Rubidium: it is too bad there is no meta information which is correct, in regards of rivers and lakes :(
18:57:11 <Eddi|zuHause> TrueBrain: the "original" perlin noise makes very "uniform" hill-valley maps. really boring as you can never build a straight track on them
18:57:19 <FLHerne> At the moment, rivers make no sense whatsoever
18:57:41 <andythenorth> rivers make sense as eye candy
18:57:44 <TrueBrain> Eddi|zuHause: I guess it was a bit too noisy :P
18:57:46 <andythenorth> or things that you bulldoze
18:58:09 <FLHerne> They should have valleys, to make it easier to bridge them
18:58:27 <FLHerne> It looks silly to have one-tile-long hump bridges :P
18:58:31 *** Markavian` has joined #openttd
18:58:33 <TrueBrain> FLHerne: we will await your patch ;)
18:59:05 <Rubidium> FLHerne: there are barely any rivers around here where you do not need to gain some elevation to cross it
18:59:18 <Eddi|zuHause> FLHerne: no, they should have "flat" bridges (which makes them unpassable by ships)
18:59:50 <TrueBrain> nice going andythenorth :P
19:00:03 <andythenorth> I thought it was a nice idea
19:00:13 <andythenorth> then I remembered it will break gameplay :P
19:00:15 <FLHerne> Statemachine-controlled swing bridges? :D
19:00:20 <andythenorth> new ship prop: can pass under bridges
19:00:42 <Rubidium> new bridge prop: can open for ships that can't pass under bridges
19:00:47 <TrueBrain> would be weird: this ship cannot go under this bridge, but this ship can hide inside that other ship
19:00:47 <andythenorth> new ship cb: can pass under bridges; new var: height of bridge
19:01:13 <andythenorth> ships can load into ships? :o
19:01:23 <TrueBrain> as in: no collision :)
19:01:37 <Alberth> TrueBrain: nice implementation of multiple cargo bays :p
19:02:20 <FLHerne> Minimum bridge heights for ships might actually be a good idea?
19:02:24 <andythenorth> so what to do about the '-' char?
19:02:35 <FLHerne> andythenorth: No, that will be a nuisance :-(
19:03:13 <Eddi|zuHause> andythenorth: traditionally cargo subtypes are put in parentheses
19:03:28 <andythenorth> there we go, done
19:03:38 <andythenorth> I tried parentheses, thought it looked silly
19:03:41 <Eddi|zuHause> andythenorth: but if you use a hyphen, it must have spaces on both sides
19:03:52 <FLHerne> When I look at that window, I want to see the cargo and the capacity. I don't care about the 'fluff' :P
19:03:58 <andythenorth> but the font might not have rendered it very evenly
19:04:05 <Eddi|zuHause> andythenorth: that looks silly
19:04:21 <andythenorth> I'm adding parentheses back in
19:04:27 <FLHerne> Like that, it's difficult to parse either the subtype or the cargo, because both are in the same sentence and the same colour :-(
19:04:50 *** Sleepie has joined #openttd
19:05:21 <andythenorth> 'cargo holds' or just 'holds'
19:05:44 <andythenorth> 'cargo hold' is tautology?
19:06:00 <Eddi|zuHause> i wouldn't know what a "hold" is
19:06:08 <andythenorth> alternative suggestions?
19:06:39 <Eddi|zuHause> and i still like the grey better
19:06:44 <V453000> piece goods or something similar?
19:07:02 <andythenorth> has to be nothing that is a cargo class name
19:07:59 <V453000> then piece something :>.
19:08:00 <Hirundo> I agree with eddi that grey looked better, it's less obtrusive
19:18:02 <planetmaker> I think 'cargo hold' is a good description
19:18:22 <planetmaker> Eddi|zuHause: cargo hold = Frachtraum (somewhat)
19:20:28 <__ln__> i wouldn't have known what 'hold' is either, but is that a reason not to use the most appropriate english words?
19:22:01 <Alberth> people may even learn a word or two English from playing openttd :)
19:22:47 <Sleepie> I also didn't know it till now, but think its the best choose
19:24:29 <Sleepie> Alberth: or maybe not, because the translators are too fast
19:24:37 <__ln__> if "we" begin to choose "easier" words that are more suitable for e.g. non-native speakers, then it's Simple English by definition, isn't it.
19:25:21 <Eddi|zuHause> planetmaker: yes. i know that. but i meant just "hold" (without further context) wouldn't have a meaning to me
19:25:40 <planetmaker> ah. Yes. I agree.
19:25:45 <planetmaker> I misunderstood you
19:25:51 <Alberth> Sleepie: play without translation, I find wording in English better than the translation
19:27:41 <FLHerne> andythenorth: I still preferred grey :P
19:28:04 <FLHerne> Also, (holds) on its own would be confusing, I think
19:29:06 * andythenorth considers alternatives
19:29:13 <andythenorth> like '(refitted to tanker)'
19:29:26 <andythenorth> '(refitted to general cargo vessel)'
19:29:35 <andythenorth> '(refitted to passenger cabins)'
19:29:52 <FLHerne> That would be long and cumbersome :P
19:30:12 <andythenorth> might help player understand autorefit behaviour though
19:35:23 <andythenorth> poll options are?
19:37:43 <Sleepie> hmm ok maybe a poll isn't the best idea (maybe only for other colors), than maybe just ask for alternative in the forum
19:38:18 <Sleepie> based on the screenshot above
19:38:48 <andythenorth> forum rarely produces better suggestions than this channel ;)
19:38:53 <andythenorth> and more noise there
19:39:53 * andythenorth ships what he has so far
19:40:04 <andythenorth> so...cargoes that travel by tanker:
19:40:40 <Sleepie> wouldn't alcohol be in bottles
19:40:54 <andythenorth> this is my question ;)
19:41:11 <FLHerne> Both could. Depends on size of vessel, type of alcohol
19:41:30 <FLHerne> I can't imagine a supertanker full of milk :P
19:41:34 <Sleepie> true but which is more common
19:41:40 <andythenorth> small ships don't refit to tanker anyway, so moot point ;)
19:41:47 <FLHerne> But a tanker barge is a possibility
19:42:33 <FLHerne> Industrial alcohol might be tankered, pehaps in large quantities
19:42:46 <FLHerne> Wine/beer probably not?
19:42:50 <andythenorth> but not to hotels and shops...?
19:45:03 <Sleepie> with one/several big tanks?
19:45:44 <andythenorth> yes there are wine tankers
19:45:50 <andythenorth> also fruit juice tankers
19:47:55 <Sleepie> well not much different, at least the exist ;)
19:48:28 * andythenorth makes milk and alcohol use cargo holds, not tankers
19:49:08 <Sleepie> tankers would only make sense if delivered to other industry and not a town imho
19:52:00 <TrueBrain> be careful with that
19:52:05 <andythenorth> simpler if it only refits to tanker for cargos that have only liquid class
19:52:12 <andythenorth> less special case code
19:54:39 <andythenorth> if anyone adds a liquid cargo with extra classes set, they won't get tankers
19:56:11 * andythenorth considers removing subtype strings entirely
20:00:13 <andythenorth> that's better :)
20:04:13 <FLHerne> andythenorth: What about autorefitting?
20:04:35 <FLHerne> Subtypes without description strings would be confusing
20:06:34 <andythenorth> the subtype string is only there to help players understand autorefit rules
20:06:39 <andythenorth> it doesn't have any other use
20:08:04 <FLHerne> But in its absense, the players wouldn't understand the autorefit rules :P
20:09:10 <andythenorth> if (indirection > 2) { ABORT ABORT ABORT }
20:11:25 <FLHerne> Er...is 2.9 tonnes/second a stupid rate of fire for a Gatling gun?
20:17:15 * andythenorth adds more indirection after all
20:17:37 <FLHerne> What are you indirecting?
20:17:57 <andythenorth> this makes no sense to me:
20:17:59 <andythenorth> "0 .. 16377 as refit cost. Add CB_RESULT_AUTOREFIT if you want to allow autorefit."
20:19:49 <andythenorth> how do I add a cb result?
20:20:45 <fjb> That looks like a constant expression to me.
20:21:49 <fjb> E.g.: 295 + CB_RESULT_AUTOREFIT
20:23:30 <andythenorth> seems that setting a cost automatically allows autorefit
20:24:21 <Sleepie> so also cost zero works?
20:24:40 <andythenorth> seems that setting ROADVEH_FLAG_AUTOREFIT for ship misc_flags works :P
20:27:56 <andythenorth> there's no extra_callback_info1 for refit_cost
20:28:05 <andythenorth> how do I check the type being refitted to?
20:28:58 <andythenorth> is var 10 not implemented for nml?
20:33:19 <andythenorth> is there a way to check vars directly in nml?
20:35:10 <planetmaker> var 10 is implemented. extra_callback_info1 (or 2)
20:35:32 <andythenorth> did you look in src for that?
20:35:38 * andythenorth was about to try that
20:36:18 <andythenorth> just gets me the mapping to the cb number
20:37:05 *** Chris_Booth has joined #openttd
20:38:03 <planetmaker> but not sure it'll help. There's a subtype variable iirc
20:38:54 <andythenorth> nah, the refit_cost should have what I need
20:39:01 <andythenorth> nfo spec is explicit about it
20:39:37 <andythenorth> in this case I need to bitmask the classes
20:39:46 <andythenorth> but fortunately I know now how nml does that
20:44:36 <andythenorth> how / why does autorefit work at all?
20:44:36 <andythenorth> I don't see how it can guarantee deterministic refits for some orders if the previous orders are indeterministic
20:44:36 <andythenorth> i.e. order 1: refit available cargo
20:44:36 <andythenorth> order 2: refit [specific cargo]
20:44:36 <andythenorth> where [specific cargo] may or may not be permitted by the vehicle newgrf
20:44:36 <andythenorth> why doesn't this result in a lot of broken routes?
20:47:18 <planetmaker> andythenorth: available = cargo with biggest amount waiting
20:47:29 <planetmaker> refit to XY = refit to XY
20:47:33 <planetmaker> so you have both options
20:47:37 <planetmaker> in the orders menu
20:47:56 <andythenorth> and if available cargo = AB for order 1...
20:48:07 <andythenorth> but for order 2 grf does not permit AB -> XY ?
20:48:14 <andythenorth> what does the vehicle do? Stuck?
20:48:20 <planetmaker> The newgrf decides which refit is feasible
20:48:47 <andythenorth> is there error handling for the "can't refit" case?
20:56:14 <andythenorth> so for refit_cost, var 10 is in format ccccwwtt
20:56:26 <andythenorth> how do I get the first word out in nml?
20:56:35 <andythenorth> and mask it with 0s?
21:00:09 *** andythenorth is now known as Guest239
21:00:09 *** andythenorth has joined #openttd
21:07:56 <planetmaker> andythenorth: like usual: (var & 0xFFFF0000) >> 16
21:08:43 <planetmaker> will give you the high word of that dword
21:08:55 <planetmaker> you likely can leave out the masking even.
21:10:35 <andythenorth> angle brackets are escaped :P
21:16:37 *** LordAro has joined #openttd
21:19:12 *** mahmoud has joined #openttd
21:20:26 *** Devroush2 has joined #openttd
21:23:50 *** Stimrol has joined #openttd
21:24:05 <andythenorth> what's wrong with this as a switch result?
21:24:05 <andythenorth> return 0 + CB_RESULT_AUTOREFIT;
21:24:05 <andythenorth> doesn't bloody work :P
21:28:05 * andythenorth decides bed is nice
21:30:13 <andythenorth> "Bit 14: If set, the refit is allowed as an auto-refit as long as bit 4 of the miscellaneous flags is set as well."
21:30:13 <andythenorth> always read the nfo spec
21:30:13 <andythenorth> golden rule of nml coding :P
21:30:13 * andythenorth should edit the wiki
21:33:04 *** fjb is now known as Guest240
22:47:41 <planetmaker> masch: I notice the funny light direction. It should come from around 4:30pm. Not from 2pm.
22:48:27 <masch> planetmaker: O.o thats noticeable? The tiles are prerendered with blendern. Maybe i should fix that ..
22:48:40 <planetmaker> look at slope brightness
22:49:02 <masch> the last row is "buggy" :-/ need to be fixed
22:49:17 <planetmaker> it's later than 1:30 and earlier than 3:00
22:49:34 <planetmaker> it should be... around 4:30
22:50:37 <masch> okay.. thanks for that. I'll try to fix that blender file. That whole thing is for my bachlor thesis :-/
22:51:25 <planetmaker> there might be already light settings around for blender. Maybe in OpenGFX+Trains (though to me that seems directly 3pm) and in zBase maybe another. I need to look
22:52:38 <planetmaker> but I like the terrain for a change. Could make nice desert or so
22:52:49 <planetmaker> what's your goal, masch?
22:53:03 <masch> planetmaker: "openttd in browser & nodejs"
22:53:25 <masch> its project work & bachlor thesis so i can split it in two big parts
22:54:06 <planetmaker> o_O That's a BIG project
22:54:15 <masch> the "real" goal is to evaluate modern web technologies
22:54:31 <masch> performance, support in browsers, "what is possible"
22:54:39 <TrueBrain> then why nodejs? :)
22:54:39 <planetmaker> you're the guy from play-ttd ?
22:54:49 <TrueBrain> the fact that is uses javascript, doesn't make it a web technoligy ;)
22:54:55 <masch> planetmaker: play-ttd? whats that?
22:55:01 <planetmaker> openttd in the browser.
22:55:34 <masch> thats a emscript version
22:56:01 <TrueBrain> making its network to work would be a nice goal ;)
22:56:31 <masch> TrueBrain: thats the plan
22:56:40 <TrueBrain> I mean, with emscript
22:56:46 <TrueBrain> much more useful (to us) :P
22:56:55 <Eddi|zuHause> what's worse with an automatic conversion via emscript, that's not with automatically converting to your machine language with gcc (or similar)?
22:57:19 <masch> Eddi|zuHause: i dont say that emscript is bad
22:57:32 <Eddi|zuHause> you said "meeeh"
22:58:03 <Eddi|zuHause> and i didn't say "bad" either
22:58:25 <Eddi|zuHause> (referring to your "boring")
22:59:09 <masch> Eddi|zuHause: the difference between "using emscript" and "reimplementing everything" is that you learn a lot more about how things work and can implement things diffently
22:59:26 <TrueBrain> masch: just realise that "reimplementing everything" will take you several months if not years
22:59:45 <masch> TrueBrain: yes. My plan isnt "reimplementing everthing"
22:59:56 <TrueBrain> now you are just confusing us :D
22:59:58 <masch> my plan is "lets see how much i can get to work"
23:00:08 <masch> the target is not to have a complete game
23:00:08 <TrueBrain> so isn't emscript a better starting point?
23:00:13 <TrueBrain> as it already does 90%? :)
23:00:30 <TrueBrain> not wanting to spoil your fun etc, but I am just trying to get the most out of it of which OpenTTD benefits :)
23:01:22 <TrueBrain> the way of getting network to work in a browser is also a goal :D
23:01:28 <Eddi|zuHause> you did not even sketch a way...
23:01:28 <planetmaker> but not all ways equally interesting :-)
23:01:32 <masch> TrueBrain: i dont do this to contribute to OpenTTD, sry :D
23:01:50 <TrueBrain> planetmaker: btw, 200k lines, sry :)
23:01:51 <Eddi|zuHause> you just throw in a few buzz words
23:02:05 <TrueBrain> cpp: 184232 (90.52%)
23:02:06 <TrueBrain> ansic: 17135 (8.42%)
23:02:12 <planetmaker> TrueBrain: well, not bad for 12 months memory or so ;-)
23:02:17 <masch> Eddi|zuHause: i was just asking for comments. I just started work on this..
23:02:29 <TrueBrain> Development Effort Estimate, Person-Years (Person-Months) = 53.10 (637.21)
23:03:08 <Eddi|zuHause> masch: see... it would take you 53 years to reimplement openttd
23:03:21 <masch> it might take me over 100 years
23:03:31 <masch> but as i told you already, thats not what i want
23:03:35 <planetmaker> TrueBrain: ohloh gives me 225k LOC, 65k comments. Which is pretty close to 300k :-)
23:03:36 <TrueBrain> haha, you plan to become N+100 years old, and still code? :D
23:03:50 <TrueBrain> planetmaker: ohloh seems wrong ...
23:03:55 <TrueBrain> I used latest trunk ..
23:04:44 <TrueBrain> which is rather weird, as I don't see how counting lines can differ that much :P
23:04:56 * NGC3982 notices that measuring emf fields with an android phone was serious business.
23:05:15 <planetmaker> TrueBrain: a cat */* | wc -l gives me 290k
23:05:31 <Eddi|zuHause> maybe you get 30% more blank lines when you count \n and \r separately :)
23:05:43 <TrueBrain> planetmaker: pff, I get 300k then
23:06:10 <masch> there are a lot of things i'm doing differently by implementing them on my own. For example the "window system". The emscript version just draws anything in one big canvas
23:06:17 <TrueBrain> planetmaker: but when done right, that results more in the order of 600k
23:06:37 <planetmaker> I know that all subs are excluded there
23:06:53 <masch> For "UI" i'm using html elements & css. That should - in my hopes - improve performance
23:06:58 <TrueBrain> weird, that sloccount and ohloh give such a different value
23:06:59 <Eddi|zuHause> masch: that's what openttd does internally.
23:07:02 <TrueBrain> who cares .. a lot of lines :D
23:07:31 <Eddi|zuHause> masch: you have to dig waaaay deeper into openttd core to change that
23:07:33 <masch> But as i try to tell you: i dont try to copy OpenTTDs code
23:07:58 <masch> i try to build a game that looks like TTD and works like TTD
23:08:16 <masch> thats why i dont use emscript
23:08:29 <masch> if i want to copy openttd, i would use emscript
23:08:33 <masch> but thats not what i want
23:09:17 <masch> ohh and sry for my bad grammar. Its late & i'm not a navtive speaker
23:09:39 <Eddi|zuHause> if you don't want to reuse code, why be "like TTD" in the first place? TTD has so many shortcomings that you could easily avoid by designing a new game properly
23:09:41 <TrueBrain> please, never say those things ... it is one of the most annoying things to read every single time people post or talk
23:09:45 <TrueBrain> "sorry about my grammer"
23:09:56 <TrueBrain> if it would be bad, we would ask you every single time what the fuck you are saying :P
23:10:07 <TrueBrain> otherwise .. our humand minds can attach enough symbols together to make sense out of it :D
23:10:32 <TrueBrain> the fact that you do want to apoligize says you have enoguh understanding of the language tbfh :P
23:10:49 <Eddi|zuHause> masch: nearly nobody here is a native speaker... and usually people who say "sorry for my english. i'm not a native speaker" are way better than the native speakers anyway
23:11:43 <planetmaker> I think no-one who spoke in this hour is a native speaker
23:13:05 <masch> Eddi|zuHause: you ask why i want to build it "like TTD". Because i like TTD! I really love this game.
23:13:10 <planetmaker> masch: and, I don't think you have to re-invent the sprites. Many things become available in the zbase project I linked you to above
23:13:20 <planetmaker> under the terms of the gpl you may use them all
23:13:41 <planetmaker> ready-to-use blender files with ready-rendered sprites in 3 sizes
23:13:52 <planetmaker> and proper lightening :-P
23:14:05 <masch> planetmaker: yes, thanks. Thats what i looked for earlier. thanks a lot that will help me
23:14:16 <planetmaker> (but for a new game you should not care about openttd light direction anyway)
23:15:18 <masch> planetmaker: i'm still shocked that you can see the difference between 2:00 and 4:30 O.o
23:15:27 <TrueBrain> years of training :P
23:15:32 <TrueBrain> before he became a dev
23:15:34 <TrueBrain> we put him in a room
23:15:39 <planetmaker> masch: I'm the maintainer of the base graphics set ;-)
23:15:58 <planetmaker> and yes, TrueBrain is totally right ;-)
23:16:10 <TrueBrain> I totally not make that up :D
23:16:37 <planetmaker> masch: I'm not the only one who notices, though. Just read a bit grpahics forum
23:16:51 <planetmaker> it's like a religion :-P
23:16:53 <Eddi|zuHause> i've been hearing this talk for years, and i still don't notice a difference between 2:00 and 4:30
23:17:05 <planetmaker> Eddi|zuHause: brightness of slopes on terrain
23:17:06 <TrueBrain> Eddi|zuHause: maybe that is why you are not dev? :D
23:17:26 <Eddi|zuHause> that must be it :p
23:17:57 <planetmaker> tbh, it was andy who convinced me :-P
23:18:09 <Eddi|zuHause> and all these years i was thinking it's because i don't conctribute enough code :p
23:18:20 <TrueBrain> Eddi|zuHause: like planetmaker contributed code :P
23:18:36 <masch> TrueBrain: to be hones, these are the first working tiles and i havend done that blender file. Some guy on reddit postet it and helped me to add some missing tiles. One of my next steps is to get a proper, nice way to render those images
23:18:52 <TrueBrain> s/TrueBrain/planetmaker/
23:19:01 * TrueBrain hugs planetmaker; it was just too easy :D
23:20:39 <Eddi|zuHause> masch: one of the big shortcomings of TTD is the fixed steepness of slopes. if you design a new (tile-based) game, then you might want to allow for more varied slopes
23:20:51 *** Stimrol has joined #openttd
23:21:31 <masch> Eddi|zuHause: thats the only thing me engine does properly right now!!111 I will never change that xD
23:22:31 <masch> Map generation: ~1hour : learn to use libnoise, generate a noisemap; ~2hours: smooth that noisemap to be a valid openttd map!
23:23:38 <Eddi|zuHause> if i could redesign the slopes, i would change the current 8px (z-level) slopes into steps of 4,8 and 12. tunnels can only be placed on 12 slopes, rails only on 4 slopes, roads on 4 and 8 slopes. bridges over road/rail must have minimum height 12
23:26:36 <Eddi|zuHause> i recently started up SimCity 2000, man that game did not age as well as TT...
23:27:08 <planetmaker> please tell :-) Why not?
23:27:33 <masch> :D that one of the games i was to dumb for all these year. Never created something that worked
23:30:04 <Eddi|zuHause> dunno really, it's not one specific thing that is wrong... maybe the pale graphics. maybe the awkward user interface, maybe the really limited gameplay compared to SC4, ...
23:31:04 <Eddi|zuHause> the maps seemed smaller than 15 years ago :p
continue to next day ⏵
