Back to Home / #openttd / 2012 / 03 / Prev Day | Next Day
#openttd IRC Logs for 2012-03-09

---Logopened Fri Mar 09 00:00:44 2012
00:02-!-tokai|noir [] has quit [Ping timeout: 480 seconds]
00:14<Rhamphoryncus>Nat_aS: rotate it?
00:14*Rhamphoryncus hits 'x' futilely in firefox
00:15<Nat_aS>well I want a big station that trains can enter/exit in two directions
00:15<Nat_aS>I supose I could make it 10 platforms wide.
00:15<Nat_aS>with tracks spliting as they leave the station
00:15<Nat_aS>but that might add a lot of traffic
00:15<Rhamphoryncus>Hrm. Rotating would only help for traffic from the southeast. Not traffic from the northwest
00:15<Nat_aS>Satilite platforms going both ways could mitigate that though.
00:16<Nat_aS>I want to expand that station into an omega type on my chart.
00:16<Rhamphoryncus>Do you ever use bridges/tunnels?
00:16<Nat_aS>not usualy for stations
00:17<Rhamphoryncus>They give a great deal more flexibility in layout
00:17<Nat_aS>can you show me examples?
00:17<Rhamphoryncus>Sure, gimme a minute
00:18<Nat_aS>I like this type of station, because it looks natural, is relatively compact, and does not involve complex signaling.
00:18<Nat_aS>and can scale well
00:18<Nat_aS>but it has bottlenecks, and feeding two lines into a big station all at once might create bottlenecks with trains entering and exiting
00:19<Rhamphoryncus> and
00:20<Nat_aS>I don't like those designs
00:20<Rhamphoryncus>Yeah, it scales for waiting trains but not really overall throughput
00:20<Nat_aS>I'm thinking of making that station really wide (10 platforms) and then making two sets of satilite stations, one going N and the other going E
00:21<Nat_aS>hopefully trains will only visit there satilite stations and never bother ones going on the other route, UNLESS they need to go to the main station for overflow
00:23<Rhamphoryncus>The only way split/merge multiple tracks is with bridges/tunnels, otherwise you constrain your throughput to a single track
00:26<Nat_aS>I'm sure I could make a nice T intersection, but the problem is doing it without running out of room
00:26<Nat_aS>I also want to expand that station
00:26<Rhamphoryncus>Your existing stations rely on trains rarely crossing to a different track. If they did you'd have huge backlogs
00:28<Rhamphoryncus>So yeah, having what is essentially two stations might work, but it'd be fragile
00:29<Rhamphoryncus>Spreading out the station would probably fix your space problem
00:30<Rhamphoryncus>hmm, just had a thought..
00:40<Rhamphoryncus>That is like having two satellites as you suggest
00:40<Rhamphoryncus>I expect a problem though: trains won't be balanced
00:42<Rhamphoryncus>The prefer platforms that are closer, which means they'll often block each other
00:43-!-KouDy1 [~KouDy@] has joined #openttd
00:44<Rhamphoryncus>Nat_aS: still there?
00:45<Nat_aS> this abomination is what I just made
00:45<Nat_aS>most of the time trains will only visit the satellite stations, and it's like having two stations
00:46-!-KouDy [~KouDy@] has quit [Ping timeout: 480 seconds]
00:46<Nat_aS>but if it overflows, they go into the main station
00:46<Rhamphoryncus>Ah, didn't see the town in the previous screenshot
00:47<Nat_aS>it's one of the larger towns on this map
00:47<Nat_aS>is that short gap between the satilite station and the sawmill station dangerous?
00:47<Nat_aS>might trains get caught there?
00:47<Nat_aS>and clog shit up?
00:48<Nat_aS>oh, I forgot to add cross tracks, pretend they are there
00:48<Rhamphoryncus>Isn't that a factory?
00:48<Nat_aS>and signals
00:49<Nat_aS>but off the southeast corner is a sawmill station
00:49<Nat_aS>where wood comes from
00:49<Nat_aS>(in jungle maps sawmills make wood instead of consuming it)
00:49<Rhamphoryncus>Actually, they could clog going to/from the northeast satellite
00:50<Rhamphoryncus>Yeah, I've played them before
00:50<Nat_aS>yeah I have never attemtped anything like that angled setup
00:50<Nat_aS>In fact, I should probably have the corner happen after the satilite
00:50<Rhamphoryncus>Or remove the intermediate signals
00:51<Rhamphoryncus>Basically, assume that anywhere a train can wait it will wait
00:56-!-Eddi|zuHause [] has quit []
00:56-!-Eddi|zuHause [] has joined #openttd
00:58<Nat_aS>here we go
00:59*Rhamphoryncus nods
01:02<Nat_aS>This... is an abomination
01:03<Nat_aS>I have never made a station this large before
01:03<Nat_aS>I have made longer though.
01:03<Nat_aS>just not this wide.
01:06-!-ProfFrink [] has joined #openttd
01:09-!-andythenorth [] has joined #openttd
01:11-!-Prof_Frink [] has quit [Ping timeout: 480 seconds]
01:11-!-ProfFrink is now known as Prof_Frink
01:22<andythenorth>can I specift a directory, or some sort of manifest file as a target for make?
01:22<andythenorth>then it's coincidental that the scripts to build those also build 3,000 pngs?
01:24-!-Lackle [] has joined #openttd
01:25-!-Lackle [] has quit []
01:25-!-Pixa [~pixa@] has joined #openttd
01:28<Rhamphoryncus>andythenorth: that's what I was working at. The main thing I found is %.c and %.o
01:28-!-Prof_Frink [] has quit [Ping timeout: 480 seconds]
01:29-!-KingPixaIII [~pixa@] has quit [Ping timeout: 480 seconds]
01:30-!-LordPixaII [~pixa@] has joined #openttd
01:30<Rhamphoryncus>Oh, that's definitely not right. I hope I didn't corrupt FIRS by turning off industry closure and setting always 100% station rating after starting the game
01:31<Rhamphoryncus>My production isn't scaling to the production level.. and I just found a with 65534 crates (65535 max)
01:35-!-Pixa [~pixa@] has quit [Ping timeout: 480 seconds]
01:36-!-JVassie [~James@] has joined #openttd
01:36-!-Pixa [~pixa@] has joined #openttd
01:36<andythenorth>Rhamphoryncus: station rating won't corrupt FIRS, unless I'm wrong
01:37<Rhamphoryncus>Yeah, I assumed it'd be harmless when I did it
01:37<andythenorth>it should be
01:37<andythenorth>industry production is almost decoupled from stations, the only coupling is the amount distributed to stations
01:37<Rhamphoryncus>And likewise on the assumption that it's still harmless I'm looking into what could give bad values
01:38<andythenorth>industry closure should be harmless, but I'd have to read code to be sure
01:38<andythenorth>and it's too early for that, and I lack tea
01:38<Rhamphoryncus>It's 20 to midnight. Definitely too early ;)
01:39<Rhamphoryncus>var_supply_storage is at -2
01:40<Rhamphoryncus>or alternatively 65534. Max storage is only corrupted when displayed, as derived from var_supply_storage, so it's not involved
01:40-!-KingPixaIII [~pixa@] has joined #openttd
01:41-!-LordPixaII [~pixa@] has quit [Ping timeout: 480 seconds]
01:42<Rhamphoryncus> max(LOAD_PERM(var_supply_storage) - LOAD_PERM(var_supply_requirement), 0), // does this work in NML?
01:45-!-Pixa [~pixa@] has quit [Ping timeout: 480 seconds]
01:46-!-Pixa [~pixa@] has joined #openttd
01:48-!-KingPixaIII [~pixa@] has quit [Ping timeout: 480 seconds]
01:53<Rhamphoryncus>Found the bug. The supply ratio is calculated before subtracting the monthly consumption, but not acted upon until after.
01:57-!-LordPixaII [] has joined #openttd
01:57<Rhamphoryncus>So if it needs 6/month to maintain and it needs 18 to increase then it can use 6+18=24, putting it into the negative. Dunno why I didn't experience it before.
02:01-!-KingPixaIII [~pixa@] has joined #openttd
02:03-!-Pixa [~pixa@] has quit [Ping timeout: 480 seconds]
02:06-!-LordPixaII [] has quit [Ping timeout: 480 seconds]
02:14-!-TWerkhoven [] has joined #openttd
02:14-!-mahmoud [] has joined #openttd
02:15<Rhamphoryncus>The monthly production seems fine. Apparently I missed how low the per-month production is compared to base industries, so I was surprised they're not rising into the 2k/month range
02:17-!-JVassie [~James@] has quit [Ping timeout: 480 seconds]
02:18-!-sla_ro|master [slaco@] has joined #openttd
02:21-!-andythenorth [] has quit [Quit: andythenorth]
02:25-!-andythenorth [] has joined #openttd
02:50-!-cypher [] has joined #openttd
02:52-!-DDR_ [] has quit [Ping timeout: 480 seconds]
02:53-!-mahmoud [] has quit [Ping timeout: 480 seconds]
02:57*andythenorth broke his repo
03:01<andythenorth>case folding collisions
03:11-!-FLHerne [] has joined #openttd
03:16<andythenorth>oh ffs
03:16-!-Twofish [] has joined #openttd
03:16<andythenorth>why make a vcs that happily destroys your repo on non-case-sensitive filesystems, with no warnings
03:17<andythenorth>non-case-sensitive is hardly uncommon
03:17*andythenorth is fed up of the illusory nature of mercurial
03:18*andythenorth has lost the BANDIT repo
03:19-!-cypher [] has quit [Quit: Miranda IM! Smaller, Faster, Easier.]
03:19-!-Twofish [] has quit []
03:20<andythenorth>maybe I can fix it after some tea
03:20-!-namad7 [] has quit []
03:20<@planetmaker>andythenorth: what did you do?
03:20<andythenorth>tea reduces grumpiness and confusion
03:20<andythenorth>changed the case on some file names, some time ago
03:21<@planetmaker>and then...?
03:21<andythenorth>all appeared fine
03:21<andythenorth>but if I try hg up <prevrev>
03:21<andythenorth>case collissions
03:21<andythenorth>I've followed the instructions for fixing it, but now have multiple heads, and no fix
03:22<@planetmaker>that sounds not like destroyed
03:22<@planetmaker>and what does a case collision look like?
03:23<@planetmaker>(exact message)
03:25<@planetmaker>a random guess on my part is: you used like "hg up <rev>", got a case collision somewhere, ignored it. And the commited while hg didn't actually update due to the case collision as it does not want to overwrite files unless forced
03:25<@planetmaker>can that be?
03:26<andythenorth>not quite
03:27<andythenorth>you can't "hg up <rev>" if there are case collisions, it simply aborts
03:28<@planetmaker>if you then hg up -C it's dangerous for uncommited changes
03:28<andythenorth>I *did* notice that hg didn't seem to care when I changed the filenames
03:28<andythenorth>which I thought was odd, I expected to have a changeset from them
03:28<@planetmaker>how did you change the case of filenames?
03:29<andythenorth>rothrock_r504.psd became rothrock_R504.psd - for consistency with other files
03:29<andythenorth>the fix appears to be
03:29<andythenorth>which I've tried, to no success so far
03:30<andythenorth>I also have two unwanted extra heads in my local repo
03:30<andythenorth>and the project can no longer be cloned, which is why I think I've lost this repo :(
03:30<andythenorth>I might have to manually copy all files into a new repo, and start from r0
03:31-!-Zeknurn [] has quit [Read error: Connection reset by peer]
03:32<andythenorth>(fresh clone from devzone can't update to tip due to case-collisions)
03:33<@planetmaker>let's see
03:34<andythenorth>indeed, there must be a fix without drama ;)
03:34<@planetmaker>devzone has only one head
03:34<andythenorth>I didn't push any new ones :)
03:34<andythenorth>don't want to make it worse with a broken fix
03:35<@planetmaker>what's your changes relative to there?
03:35-!-DDR [] has joined #openttd
03:35<andythenorth>wrt tip in my repo and tip remote?
03:35<@planetmaker>your repo wrt devzone
03:36<andythenorth>is there a quick way to diff between the two?
03:36<andythenorth>or I can list manually
03:36<@planetmaker>hg outgoing might tell
03:37<andythenorth>I have two changesets, one to remove the .psd files and one the .png
03:37<andythenorth>I should have done both in same operation, but I didn't know about the .png
03:38<andythenorth>"One way to repair such a repository is to check it out on a case-sensitive Unix system, remove the problematic file, and commit it again."
03:39<@planetmaker>ok, the creation of heads is you commiting two different changesets based on the same revision. That primarily has nothing to do with cases, I'd say. Or?
03:39<andythenorth>yes, the heads are a consequence of the fix, not a cause
03:40<@planetmaker>do you have outstanding changes?
03:40<andythenorth>in short yes
03:40<andythenorth>but they could be sorted out manually
03:40<@planetmaker>in short?
03:40<@planetmaker>well, save them to a diff
03:41<andythenorth>they aren't outstanding for tip, but they are for whatever rev I'm currently on
03:41<@planetmaker>or commit them
03:42<@planetmaker>don't worry about creating yet another head ;-) We can merge later
03:43<@planetmaker>can you paste the output of "hg glog -l10" ?
03:43*andythenorth enables glog
03:44<@planetmaker>oh, that's an extension? :-)
03:45<@planetmaker>not here at least
03:45<andythenorth>"<command> | open -f" is very handy on os x
03:46<@planetmaker>and you're now at r387, right?
03:47<@planetmaker>then let's try to successively merge heads
03:47<@planetmaker>hg merge -r383
03:48-!-FLHerne [] has left #openttd []
03:48<andythenorth>pdq2s-macbook-3:pixel_generator andy$ hg merge -r383
03:48<andythenorth>abort: case-folding collision between src/graphics/rothrock_r504.png and src/graphics/rothrock_R504.png
03:48<@planetmaker>and the same when you merge -r382 ?
03:49<andythenorth>abort: case-folding collision between graphics_sources/rothrock_r504.psd and graphics_sources/rothrock_R504.psd
03:49<@planetmaker>then let's do that first :-)
03:49<andythenorth>this is why I was grumpy :)
03:49<andythenorth>the fix I've followed isn't a fix :)
03:49<andythenorth>it seems circular
03:49<andythenorth>I had a vision of doing it n times, spawning ever-more heads :P
03:50<@planetmaker>and you FS is not case sensitive, right?
03:50<andythenorth>"One way to repair such a repository is to check it out on a case-sensitive Unix system, remove the problematic file, and commit it again"
03:50<andythenorth>so I could make a disk image
03:50<@planetmaker>disk image?
03:50<andythenorth>with disk utility
03:51<andythenorth>make a case-sensitive volume, mount it, cp there, try to fix
03:51<@planetmaker>well, I could simply try to fix it...
03:51<@planetmaker>I have no volumes which are not case sensitive
03:51<@planetmaker>which btw, prevented me to install photoshop
03:51<andythenorth>so can you try to strip the files? I don't want to push my multiple heads
03:52<andythenorth>I don't mind if you simply delete both upper and lower case versions, I can put them back manually
03:53<@planetmaker>yes, I shall try. I only have r382
03:53<@planetmaker>src/graphics/rothrock_R504.png ?
03:53<@planetmaker>any other?
03:53<andythenorth>the equivalent psd in graphics_sources
03:54<andythenorth>I don't know of others, but there may be some, I didn't find a way to check yet
03:54<@planetmaker>ah. There's now both files, upper and lower case
03:54<@planetmaker>which version do you want to keep?
03:54<andythenorth>upper please
03:55-!-Zeknurn [] has joined #openttd
03:55<andythenorth>devzone sees both too
03:55-!-KouDy [~KouDy@] has joined #openttd
03:55<@planetmaker>any other files with the same problem?
03:55<andythenorth>ok, manual check on devzone, I don't see any
03:55<@planetmaker>ok, me neither
03:55<andythenorth>just the psd and png for r504
03:57*planetmaker authorizes himself for commit access :-P
03:58<andythenorth>I can discard local heads now?
03:58<@planetmaker>make it easy for you and a backup copy:
03:58<@planetmaker>cp -r bandit old.bandit
03:58<@planetmaker>or rather mv bandit old.bandit
03:59<@planetmaker>and then clone devzone
03:59<@planetmaker>only delete the old repo when you're sure you have everything
03:59<andythenorth>yup, works
03:59<@planetmaker>after all you could also still export all your changes or even pull from your old repo
04:00<andythenorth>oh how interesting
04:00<andythenorth>I have new clone
04:00<andythenorth>but still can't up 378
04:00<andythenorth>but I can now up tip
04:00-!-KouDy1 [~KouDy@] has quit [Ping timeout: 480 seconds]
04:01<andythenorth>planetmaker: can you up 378?
04:01<@planetmaker>yes, I guessed so
04:02<@planetmaker>rothrock_R504.png and rothrock_r504.png in r378
04:02<andythenorth>and the mac won't allow that on non-case-sensitive FS
04:02<andythenorth>so the fix is to rm from that rev?
04:02<andythenorth>as per
04:03<Rhamphoryncus>Hmm. 20 km/h bus. 80 km/h train. This sounds like a recipe for Fun.
04:03<andythenorth>or actually from the rev where I introduced the issue, which I don't know yet
04:03<@planetmaker>you could try that as described there, yes
04:04<andythenorth>r268 I think
04:04<@planetmaker>but that would for you after a commit not change r378, but create a new revision, r384 which has r378 as parent
04:06<andythenorth>also I just tried it, and it doesn't solve the issue
04:07<@planetmaker>what exactly are you trying now?
04:09<andythenorth>following instructions exactly as per link above, using r268 and removing all r504 and R504 files
04:09<andythenorth>doesn't work though
04:09<@planetmaker>ok, let me ask: why do you now want a revision which has this collision?
04:11<andythenorth>I'm trying to diagnose a bug which I think I introduced in r378, but can't verify
04:11<andythenorth>[so want to up 378 and build]
04:11<@planetmaker>did you switch file systems since then?
04:12<@planetmaker>should I commit a head based on r378 which removest he offending files?
04:15<andythenorth>clean clone, case-folding issues for up tip, up 384, up 378
04:15-!-pugi [] has joined #openttd
04:15<andythenorth>so maybe it's time to make a case-sensitive disk image
04:15<andythenorth>there's a lesson in this somewhere, not sure what yet :P
04:15<@planetmaker>well, 378 will not work for you on non-case sensitive
04:16<@planetmaker>384 should. Unless there's more files than the one png
04:18<@planetmaker>andythenorth: what hg version do you use?
04:19<andythenorth>I tried upgrading it earlier, but macports refuses to update the ports tree :P
04:19<@planetmaker>I wanna see what the hg say to this
04:19<@planetmaker>*hg guys
04:19<andythenorth>it was a frustrating morning ;)
04:19<@planetmaker>#mercurial on freenode, if you care
04:20<andythenorth>I should do later
04:20<andythenorth>I have to stop neglecting the kids now :o
04:21<andythenorth>ho my ports tree update succeeded
04:22-!-KingJ [~KingJ-OFT@] has quit [Ping timeout: 480 seconds]
04:23<SpComb>ya, case-sensitivity can be hard on VCS's
04:25<SpComb>I had a problem where I renamed some files and only changed their case, and hg didn't notice - then when I did a separate build checkout, it created a broken build (wrong filename case)
04:26<SpComb>then I sillyly fixed that `hg mv ...` in the build repo, and trying to merge that case-change commit back into the master repo was... hard
04:30-!-lugo [] has joined #openttd
04:31<andythenorth>SpComb: similar issue to mine :|
04:49-!-HerzogDeXtEr [~Flex@] has quit [Quit: Leaving.]
04:49-!-HerzogDeXtEr [~Flex@] has joined #openttd
04:50-!-smoovi [] has joined #openttd
04:55-!-Enorian [] has joined #openttd
04:56-!-Enoria [] has quit [Ping timeout: 480 seconds]
05:04-!-andythenorth [] has quit [Quit: andythenorth]
05:08-!-pjpe [] has quit [Quit: ajax IRC Client]
05:13-!-tokai|noir [] has joined #openttd
05:13-!-mode/#openttd [+v tokai|noir] by ChanServ
05:19-!-tokai|mdlx [] has quit [Ping timeout: 480 seconds]
05:29<Ammler>planetmaker: the histroy of mercurial is not case sensitive, they use _ for case
05:30<Ammler>but of course, the workcopy is
05:30<Ammler>but you could fix it and then andy can up to a fixed version
05:35<@planetmaker>we figured, yes
05:35-!-hbccbh [~hbc@] has joined #openttd
05:41-!-tokai|mdlx [] has joined #openttd
05:44-!-tokai|noir [] has quit [Read error: Operation timed out]
05:49<Rhamphoryncus>So.. on a lark I've discovered that the pathfinder does, in fact, care about track speed of nutrack.. which means you can have a dual mainline with both highspeed and lowspeed and the highspeed trains do prefer the highspeed land. The lowspeed lanes can then be pushed using a PF penalty (reversed path signal)
05:49<Rhamphoryncus>Not sure if it really improves network efficiency though, or if it's just a side effect of having a dual mainline
05:52<Rhamphoryncus>Anybody played with that before?
05:53-!-KingPixaIII is now known as Pixa
05:56-!-Prof_Frink [] has joined #openttd
05:59<Ammler>planetmaker: and what would you like to ask #mercurial ?
06:00<Ammler>ah, you are there too :-P
06:01-!-sla_ro|master [slaco@] has quit [Quit: DANGER is OFFLINE DANGER]
06:05-!-theholyduck [~holyduck@] has quit [Read error: Connection reset by peer]
06:08-!-ProfFrink [] has joined #openttd
06:09-!-Prof_Frink [] has quit [Ping timeout: 480 seconds]
06:09-!-ProfFrink is now known as Prof_Frink
06:22-!-DDR [] has quit [Quit: ChatZilla 0.9.88 [Firefox 10.0.2/20120216100510]]
06:23-!-ProfFrink [] has joined #openttd
06:23-!-Pikka [] has joined #openttd
06:28-!-Prof_Frink [] has quit [Ping timeout: 480 seconds]
06:30-!-Prof_Frink [] has joined #openttd
06:35-!-ProfFrink [] has quit [Ping timeout: 480 seconds]
06:40-!-ProfFrink [] has joined #openttd
06:46-!-Prof_Frink [] has quit [Ping timeout: 480 seconds]
06:48-!-Prof_Frink [] has joined #openttd
06:53-!-ProfFrink [] has quit [Ping timeout: 480 seconds]
06:57-!-brambles [brambles@] has quit [Remote host closed the connection]
07:06-!-dotwaffle [] has quit [Quit: WeeChat 0.3.6]
07:07-!-tokai|noir [] has joined #openttd
07:07-!-mode/#openttd [+v tokai|noir] by ChanServ
07:08-!-dotwaffle [] has joined #openttd
07:10-!-Pikka [] has quit [Quit: ChatZilla [Firefox 10.0.2/20120215223356]]
07:10-!-KingJ [~KingJ-OFT@] has joined #openttd
07:13-!-dotwaffle [] has quit []
07:13-!-tokai|mdlx [] has quit [Ping timeout: 480 seconds]
07:16-!-Prof_Frink [] has quit [Ping timeout: 480 seconds]
07:21-!-dotwaffle [] has joined #openttd
07:21-!-Prof_Frink [] has joined #openttd
07:22-!-Biolunar [] has joined #openttd
07:25-!-mahmoud [] has joined #openttd
07:27-!-ProfFrink [] has joined #openttd
07:28-!-dotwaffle [] has quit [Quit: WeeChat 0.3.6]
07:29-!-Prof_Frink [] has quit [Ping timeout: 480 seconds]
07:29-!-ProfFrink is now known as Prof_Frink
07:30-!-dotwaffle [] has joined #openttd
07:34-!-lofejndif [] has joined #openttd
07:35-!-lofejndif [] has quit [Max SendQ exceeded]
07:37-!-lofejndif [] has joined #openttd
07:38-!-Prof_Frink [] has quit [Ping timeout: 480 seconds]
07:46-!-Prof_Frink [] has joined #openttd
07:55-!-Muxy [] has joined #openttd
07:58-!-Muxy [] has left #openttd [PACKET_CLIENT_QUIT]
07:59-!-Prof_Frink [] has quit [Ping timeout: 480 seconds]
08:03-!-glx [glx@2a01:e35:2f59:c7c0:747f:3f81:7f24:a684] has joined #openttd
08:04-!-mode/#openttd [+v glx] by ChanServ
08:06-!-Prof_Frink [] has joined #openttd
08:07-!-KritiK [] has joined #openttd
08:12-!-Elukka [] has joined #openttd
08:14<Eddi|zuHause>nmlc: "src/../src/bav/18.gnml", line 115: All parts of the ternary operator (?:) must be integers. <-- is that an nml bug?
08:16<Eddi|zuHause> <-- the essence of the offending code
08:17<Eddi|zuHause>the property line is alright, the callback line is not
08:18<Rhamphoryncus>Try replacing 8 with 8.0
08:18<Rhamphoryncus>Wait, misread >.>
08:18<Eddi|zuHause>how should that help anything?
08:18*Rhamphoryncus makes up something about being tired. Which he should be, but isn't.
08:19<Eddi|zuHause>(besides... checking for vehicle_is_stopped is nonsense... running costs are 0 then anyway)#
08:20<Rhamphoryncus>does that include stopped at a signal?
08:20-!-ProfFrink [] has joined #openttd
08:20-!-ProfFrink [] has quit []
08:20<Eddi|zuHause>only when player clicks the stop button
08:20<Rhamphoryncus>Then why is being stopped at a depot's implicit signals treated specially? (vehicle_is_in_depot)
08:21-!-Prof_Frink [] has quit [Ping timeout: 480 seconds]
08:21<Eddi|zuHause>i'm not the author of this feature... that's a separate discussion :p
08:21<Rhamphoryncus>oh okay
08:24<Eddi|zuHause>bäh... devzone is so useless when javascript is disabled...
08:24<Eddi|zuHause>Ammler: i can't change the filter on the ticket list without javascript... wtf?
08:25<Ammler>Eddi|zuHause: you don't have javascript?
08:25<Ammler>please use another browser
08:25<Eddi|zuHause>Ammler: i disabled javascript.
08:25<Eddi|zuHause>makes for way smoother general browsing experience...
08:25<Ammler>so what is the issue?
08:26<Eddi|zuHause>Ammler: the issue is that something as simple as an "apply" button shouldn't need javascript
08:26<Ammler>maybe you can maek exceptions
08:28<Eddi|zuHause>Ammler: i can also eat a ton of caviar. doesn't mean i want to.
08:29<Ammler>not sure, you can :-P
08:29<Eddi|zuHause>depends on the timeframe :p
08:31-!-pugi [] has quit []
08:32-!-TGYoshi [~TGYoshi@] has joined #openttd
08:34-!-lugo [] has quit [Quit: ajax IRC Client]
09:08-!-pugi [] has joined #openttd
09:08<NGC3982>evening, duders.
09:14-!-andythenorth [] has joined #openttd
09:17<andythenorth>planetmaker: BANDIT is fine on a case sensitive disk image P
09:17<andythenorth>bah :)
09:19<@planetmaker>yes, of course
09:19<@planetmaker>that's why I had no trouble creating the revs I committed earlier today ;-)
09:20<@planetmaker>But I got so far no answer how to avoid your trouble on case insensitive FS
09:20<andythenorth>"just don't do the things that caused this issue" :P
09:21<andythenorth>also, /me concludes that captain rand may be an idiot
09:21<@planetmaker>well, yes. But that's a good advise but no good method to prevent the mishap
09:23-!-kkb110 [] has quit [Ping timeout: 480 seconds]
09:23<andythenorth>in that silly v2.0 thread. You upset him I think :)
09:23<andythenorth>hmm. statements like this: "If we would like TT to last another 20 years and bring in new players and contributors, the two halves need to become whole"
09:24<@planetmaker>oh. Well, who cares
09:24<andythenorth>always this fricking obsession with new players.
09:24<@planetmaker>well, new players are important. But you won't gain them by his proposed convergence.
09:24<andythenorth>yeah, because the reason people are playing WoW is because they checked out TTDP and OTTD and are cross that there are features different between them
09:24*andythenorth should have some tea
09:24<andythenorth>some days are grump days
09:24<@planetmaker>Rather by deviating more and embracing better (G)UI and ease of use
09:25-!-Pikka [] has joined #openttd
09:25<@planetmaker>not by adding custom bridge heads and programmable or restrictive signals
09:25<Pikka>boo, planetmaker, boo
09:25<andythenorth>oh it's pikka-bobble :)
09:26<@planetmaker>Pikka, it's not like I'm against those features. But... I doubt they are crucial for a person to like the game :-)
09:26<andythenorth>he always cheers me up
09:26<@planetmaker>or any significant amount of people
09:26<@planetmaker>indeed I think at least custom bridgeheads will be quite nice.
09:26<@planetmaker>Pikka, reference was this funny openttd v2.0 thread
09:26<andythenorth>silly thread
09:27<andythenorth>it should be in the suggestions forum, where I mostly don't go :P
09:27<andythenorth>suggestions forum is just /dev/null right?
09:27-!-zxbiohazardzx [] has joined #openttd
09:27<zxbiohazardzx>y hello thar :P
09:27<@planetmaker>nah, not exactly, andythenorth . But you can go wild there w/o penalty ;-)
09:27<Pikka>well the lack of custom bridgeheads is keeping SAC out of OpenTTD, so that's one argument against implementing it I suppose.
09:28*Pikka goes to make tea
09:28*planetmaker offers Pikka some freshly brewed tea
09:28<zxbiohazardzx>lol Pikka
09:28<andythenorth>Pikka: tea for two?
09:28<Ammler>what about good old coffee?
09:28*andythenorth needs tea
09:28*zxbiohazardzx needs coffe
09:28*andythenorth is staring into a nappy
09:28<zxbiohazardzx>e fail more
09:29<Ammler>planetmaker: you have a ppc mac?
09:29<zxbiohazardzx>anyway i had a question on the scenario-creator's rock tile placement feature
09:29<zxbiohazardzx>where is that in the sourcecode?
09:30<zxbiohazardzx>i assumed src/widgets
09:30<zxbiohazardzx>but that mainly has .h files
09:31<@planetmaker>Ammler, no
09:31<zxbiohazardzx>aka it only defines teh Enums in terraform_widget.h
09:32-!-kkb110 [] has joined #openttd
09:32<@planetmaker>zxbiohazardzx, what makes you believe that rocks are placed anywhere in the GUI functions?
09:32<@planetmaker>(aka widgets)
09:32<Pikka>I got halfway through making coffee before I remembered I wanted tea :) so I am having coffee after all.
09:32<@planetmaker>I'd try like terragen* or map_gen or similar
09:32<zxbiohazardzx>as far as i recall it was part of the landscape toolbar in the scenario generation?
09:32<zxbiohazardzx>well i didnt expect the placement to be there
09:33<@planetmaker>toolbar != feature itself
09:33<zxbiohazardzx>but i expected some sort of hint where to look for the feature
09:34<@planetmaker>consider looking at map_* and terraform*
09:35<@planetmaker>maybe also genworld*
09:35<zxbiohazardzx>ok thx
09:36<zxbiohazardzx>was trying to see if we can manually place buildings as well, similar to rock placement,(real buildings not newobject fake ones)
09:36<@planetmaker>for the SE that might be nice
09:37<@planetmaker>iirc some patch to that extend existed somewhen somewhere
09:37<@planetmaker>you might want a similar GUI like the station or new object placement offers
09:37<zxbiohazardzx>that would even be better then my idea
09:37<zxbiohazardzx>id just make it go random XD
09:38<zxbiohazardzx>being able to actually pick a building might really offer some nice features in the SE
09:38<@planetmaker>as you need a way to select the house (with 'random' being an option)
09:38<zxbiohazardzx>making a "downtown" area and "suburbs" etc
09:38<zxbiohazardzx>yeah sounds fair enough
09:38<zxbiohazardzx>ill check if i can find that patch, it might get me started a bit more :P
09:39<@planetmaker>but beware of pitfalls. Houses may decide to not be buildable everywhere :-)
09:39<zxbiohazardzx>maybe first run the check and only build it if allowed regardless of why it cannot be build (slopes etc i )
09:39<zxbiohazardzx>but lets start with the basics
09:39<zxbiohazardzx>then thik of fancy other stuff :P
09:40<zxbiohazardzx>think* omg whats wrong with my hands today
09:43-!-Markavian [] has joined #openttd
09:45<zxbiohazardzx>hate searching for old patches, specially if you dont know a name or forumpart, sofar on page 6 or 7 without it :P
09:45<zxbiohazardzx>and the second you say it, you find it
09:50<zxbiohazardzx>ill first try to just update patch to a more recent version, see how that works out :P
09:50-!-Markavian` [] has quit [Ping timeout: 480 seconds]
09:52<zxbiohazardzx>damn SVN errors :S
09:53<zxbiohazardzx>Compression of svndiff data failed
09:53<zxbiohazardzx>62.07MBytes transferred 1 min 45 seconds, the operation failed
09:53<zxbiohazardzx>but /care i got the stuff in :P
09:59-!-__mj [] has joined #openttd
10:00-!-__mj [] has quit []
10:02-!-cypher [] has joined #openttd
10:03-!-tokai|mdlx [] has joined #openttd
10:06-!-tokai|noir [] has quit [Read error: Operation timed out]
10:07-!-tokai [] has joined #openttd
10:07-!-mode/#openttd [+v tokai] by ChanServ
10:09-!-tokai|mdlx [] has quit [Read error: Operation timed out]
10:12<zxbiohazardzx>in window_type.h
10:12<zxbiohazardzx>i think you changed the WC_BUILD_OBJECT etc lines to add comments?
10:15<@planetmaker>zxbiohazardzx, for developing patches a distributed VCS like hg or git is easier to use
10:17<Ammler>why does openttd check for timidity?
10:17<Ammler>(the configure)
10:17<@planetmaker>it's a possible sound driver
10:18<@planetmaker>thus it can link against libtimidity
10:19<@planetmaker>configure checks for each library it can link against (on the build platform)
10:20<Ammler>hmm, maybe I should check if music works then
10:20<Ammler>or is that completely independent from openmsx, where timidity is required
10:21<@planetmaker>might be needed for any sound. thus music or sound
10:21-!-Pulec [] has joined #openttd
10:21<Ammler>but if openttd doesn't find it on building, openmsx will still work?
10:22<@planetmaker>depends whether it has another sound device. But iirc sound and music can in principle also play on the null device
10:23<@planetmaker>the enjoyment factor might be low, though
10:23<Ammler>timidity is there, just not while building openttd
10:23<Ammler>that is why I wondered about that
10:24<@planetmaker>then openttd will not link against it. Of course it would need its devel libs in order to do so
10:24<@planetmaker>like with every lib: you need it at build time. Or it will never be used
10:24<@planetmaker>(or rather you need it's development headers)
10:25<Ammler>since when is that the case?
10:25<Ammler>he :-)
10:25<Ammler>so suse openmsx never worked :-P
10:25<@planetmaker>it might use other sound devices
10:26<@planetmaker>did you never check the builds you put into suse?
10:26<Ammler>I speak about music... openMsx
10:27<Ammler>music works just fine
10:27<Ammler>fail alarm, that configure output was confusing
10:27<@planetmaker>how do openttd's sound driver and openmsx relate at all?
10:27<Ammler>so openttd does not need timidity on building to have openmsx working
10:27<zxbiohazardzx>Planetmaker yeah i know hg/git and i use that for the wow devment
10:27<zxbiohazardzx>but for now svn works
10:27<zxbiohazardzx>just wondering about that change in window_type.h
10:28<zxbiohazardzx>the old patch just added a line WC_BUILD_HOUSE,
10:28<zxbiohazardzx>between WC_BUILD_OBJECT and WC_INVALID = 0xFFFF
10:28<Ammler>btw. how do you control music on the openttd main menu?
10:29<zxbiohazardzx>but the current one has alot more comments there
10:29-!-Markavian` [] has joined #openttd
10:30<@planetmaker>Ammler, not.
10:30<@planetmaker>it's imho a missing UI thing
10:30<Ammler>could be added to the GameOptions beside the menu where you chose the music set
10:30<zxbiohazardzx>also planetmaker did nobody ever considered adding CMake to the project (i know they did that @ TrinityCore to make cross platform compilation less of a pain
10:30<zxbiohazardzx>sidetracked again :p
10:33<@planetmaker>zxbiohazardzx, obviously at least you did
10:33<Ammler> esound-daemon gstreamer-0_10-plugin-esd libao-plugin4-esd libaudiofile0 libesd0 openmotif-libs openttd-openmsx timidity <-- a lot additional package come with openmsx :-)
10:34<zxbiohazardzx>well adding it is a pain as my coding skills are very much limited
10:34<zxbiohazardzx>now the questions i had on the code so i can continue bumping that older patch to head
10:34<zxbiohazardzx>in window_type.h
10:34<zxbiohazardzx>old patch just added the WC_BUILD_HOUSE
10:34<zxbiohazardzx> WC_BUILD_OBJECT,
10:34<zxbiohazardzx>+ WC_BUILD_HOUSE,
10:34<zxbiohazardzx> WC_INVALID = 0xFFFF
10:35<zxbiohazardzx>but new code has way more comments there
10:35<zxbiohazardzx>and alot more stuff in between WC_BUILD_OBJECT and WC_INVALID that i assume is needed for all kinds of stuff
10:36<zxbiohazardzx>mostly wondering what comment to stick on top of it, i assume WindowFunction: %Window numbers:
10:36-!-Markavian [] has quit [Ping timeout: 480 seconds]
10:36<zxbiohazardzx>then - 0 = #NameWidgets?
10:37-!-tokai|mdlx [] has joined #openttd
10:37<zxbiohazardzx>and teraform_gui.cpp similar question, as that got revamed
10:38<zxbiohazardzx>i assume ETTS -> WID style there refering to the widget
10:40-!-flaa [] has joined #openttd
10:42-!-supermop [] has joined #openttd
10:42*zxbiohazardzx is now away - Reason : back in 2 hours...
10:43-!-tokai [] has quit [Ping timeout: 480 seconds]
10:44-!-lofejndif [] has quit [Quit: Leaving]
10:59<@planetmaker> <-- simply answering with "please no double postings" might be rude, but I'm quite tempted
11:03<andythenorth>I'd start by asking his age
11:03<andythenorth>well I wouldn't, because I'm a bad person to get into forum wars :P
11:04<andythenorth>I'd be surprised if he was >15
11:06-!-Alberth [] has joined #openttd
11:06-!-mode/#openttd [+o Alberth] by ChanServ
11:08<andythenorth>Alberth: :)
11:08<andythenorth>you have finished your week, done honest work, earnt honest money? :)
11:12-!-Strid [] has joined #openttd
11:12<Pikka>bongiorno Alberth
11:13<Eddi|zuHause>boingo boingo?
11:13<Pikka>says you, buddy
11:14<@planetmaker>bunga bunga
11:14<supermop>am i invited to that?
11:18<@Alberth>andythenorth: sort of, but 'honest' is somewhat tricky if you don't really like what you are doing :p
11:18-!-Prof_Frink [] has joined #openttd
11:18<@planetmaker>:-( @ Alberth you don't?
11:18<@Alberth>writing Eclipse extensions is not much fun
11:18<andythenorth>at least you keep the Eurozone productive :P
11:19*andythenorth got paid recently, so today has helped the economies of France, Denmark, and China
11:19<andythenorth>mostly Denmark :P
11:20<supermop>danish petrochemical solids
11:20<andythenorth>what comes from Denmark? Plastic bricks :P
11:20<supermop>my favorite commodity
11:22*andythenorth ponders what to code next
11:22<Pikka>what are the options?
11:22<andythenorth>(1) cleanup my python to not be technically invalid
11:22*Pikka codes Brush Class 92s
11:22<andythenorth>(2) fix some very boring positions of magic pixels
11:23<Pikka>not being technically invalid is always good.
11:23<andythenorth>(3) figure out how to actually configure and run pixa
11:23-!-ProfFrink [] has joined #openttd
11:23*andythenorth likes 3
11:23-!-bolli [~bolli@] has joined #openttd
11:24<bolli>i have a question for the linuxy people out there... :P
11:24<bolli>can i control a background process without bringing it to the foreground?
11:25<@planetmaker>if you refer to openttd and running it as a server: use screen and run it inside that
11:25<@Alberth>bolli: 'kill' is more versatile than its name suggests :)
11:26*planetmaker mostly uses -9 as option :-P
11:26-!-Prof_Frink [] has quit [Ping timeout: 480 seconds]
11:26-!-ProfFrink is now known as Prof_Frink
11:26<bolli>thanks :)
11:26<Ammler>planetmaker: you really should upgrade your system :-P
11:27<andythenorth>-9 ftw
11:27*Alberth tries to avoid -9 very much :)
11:27<bolli>so what would -9 d? :P
11:27*andythenorth wonders if sudo kill -9 * works
11:27<Ammler>if you need -9 that often, something is very broken
11:27<andythenorth>you need -9 that often ;)
11:27<@planetmaker>terminate process without any questions asked or notifications given
11:27<andythenorth>at least once a month or so
11:27<@planetmaker>Ammler, who says that I need it?
11:27<@Alberth>planetmaker: and without giving it the chance to cleanup in any way
11:28<@planetmaker>Alberth, I know
11:28<andythenorth>it doesn't deserve a chance imho
11:28<andythenorth>bad processes should be punished
11:28<@Alberth>better not even start bad processes :)
11:29*andythenorth could tell stories about printjobmgr using 200% of CPU while trying to print to a printer 5 miles away
11:29<andythenorth>but instead....pixel generator?
11:29-!-TWerkhoven [] has quit [Read error: Connection reset by peer]
11:29<Ammler>no andythenorth, tell the story, don't waste time with pixels
11:29-!-TWerkhoven [] has joined #openttd
11:31-!-smoovi [] has quit [Ping timeout: 480 seconds]
11:32<Ammler>Alberth: wanna update our hook to check for case sensitive issues?
11:32<@Alberth>Ammler: but a picture is worth a thousand words :)
11:33<@Alberth>Ammler: in what way 'case sensitive issues' ?
11:34<Ammler>mercurial is able to handle abc.nml and ABC.nml in same dir
11:34<supermop>planetmaker: better to buy a new lens with VR first, or a new body with better features?
11:34<@Alberth>Ammler: it should :)
11:34<Ammler>but if you have a case insensitive fs, you won't be able to update to that rev
11:34-!-_maddy [] has joined #openttd
11:34<_maddy>hi folks, anyone up for a multiplayer game?
11:35<supermop>assuming i have under 1000 usd to upgrade my slr so will have to do it piecemeal
11:35<@planetmaker>supermop, in case of doubt invest in lenses
11:35<Ammler>_maddy: join a existing server maybe?
11:35<@Alberth>Ammler: ok, I can have a look. I have completely forgotten how it works though, so no idea if it can be done
11:35<_maddy>Ammler: I'm looking for a coop game
11:35<@planetmaker>features on a camera body don't matter, if the lenses are bad.
11:36<@planetmaker>_maddy, why don't you then join one of our coop servers?
11:36<Ammler>Alberth: just in case the current script running on our server:
11:36<supermop>my lenses are ok now
11:36<@planetmaker>supermop, what you might consider though, is to avoid the lenses which won't fit a full-frame camera
11:37<supermop>just the nikkors that came with it in 2005
11:37<@planetmaker>one might want to upgrade to one later, maybe
11:37<supermop>but fx bodies are very expensive
11:37<_maddy>planetmaker: what coop servers? I only know of openttdcoop
11:37<@Alberth>Ammler: ok, thanks. Is that all my code? The final two functions are not documented
11:37<supermop>and fx lenses much more expensive than dx equivalents
11:38<@planetmaker>_maddy, so do I ;-)
11:38<Ammler>Alberth: I think so
11:38<supermop>but could use fx lenses on my 35mm
11:38<Ammler>the only thing I might have changed are config
11:38<@planetmaker>supermop, I'm not too familiar with nikon's naming scheme. dx=? fx=?
11:38<@Alberth>hmm, I was in a hurry perhaps, or it was clear enough at the time :)
11:39-!-_maddy [] has quit []
11:40-!-smoovi [] has joined #openttd
11:40<supermop>fx is a digital sensor roughly the same physical size as a fram of 35 mm film
11:41<supermop>dx is roughly the size of a frame of aps film but is pretty standard throughout digital camera bodies
11:41<@planetmaker>ok. So it's nikon's way to tell sensor size
11:41<supermop>the fx cameras are for the 'professional' market
11:42<supermop>the mounting ring and AF etc are all the same
11:42<@planetmaker>I'm kinda dreaming currently to buy an EOS 5 Mk III ;-)
11:42<supermop>so any nikor lense since the old nikon F will fit on any nikon body
11:42<supermop>and work reasonable well
11:43<supermop>(i use my old circa 1978 lenses on my d50 at times)
11:43<@planetmaker>friends of mine are very content with Nikon's 300 whatever
11:43<supermop>yeah, that is a dx one, as is my d50
11:44<supermop>but metering and sensor quality has gotten much beter since 2005, even on the entry level bodies
11:44<supermop>i am mostly dissatisfied with low-light
11:45<andythenorth>Alberth: so....reverse generating from png names?
11:45<supermop>the new bodies have isos up to the 10s of thousands that dont look noisy
11:45<supermop>whereass on mine i get a lot of noise even at lower isos
11:46<@planetmaker>yeah, sensor quality got a lot better
11:46<@Alberth>andythenorth: yeah, doesn't that sound like a good idea?
11:46<supermop>but my other problem is that i can't always have a tripod
11:46<supermop>and the new lenses can correct for vibration
11:46<andythenorth>Alberth: in principle
11:47<andythenorth>but I wonder chicken-egg? where are the pngs before they're generated?
11:47<@planetmaker>supermop, better performance at higher ISO "corrects" for that, too
11:47<@Alberth>andythenorth: you give a filename after they are generated :o
11:48<andythenorth>currently yes
11:49<supermop>the fx lenses also let more light through than the dx ones, typically 73 instead of 52mm filter size, at the expense of being heavy and bulky
11:49<supermop>what body do you use?
11:49<bolli>so would i use SIGPIPE with kill and pipe the command to it?
11:50<bolli>i'm a bit confused about how to send the command using kill...
11:50<@Alberth>kill takes a PID as parameter to send the signal to
11:50<@planetmaker>I still have a canon 30D
11:51<@planetmaker>but it gets time to upgrade that, too
11:51<@Alberth>bolli: type 'ps ux'
11:51<@Alberth>first column is the PID number, then type 'kill PID'
11:51-!-kkb110 [] has quit [Read error: Operation timed out]
11:51<@Alberth>for a friendly nudge to close down :)
11:52<bolli>but i just want to send a command to it..
11:52-!-sla_ro|master [slaco@] has joined #openttd
11:52<@Alberth>don't use the PID of the shell you are using :p
11:53<@Alberth>it's not an input stream, kill only sends a signal
11:53<Ammler>bolli: maybe you wanna use the admin interface
11:53*Alberth guess that too
11:54<bolli>admin interface?
11:54<bolli>i'm using the command line...
11:54<Ammler>but if you are on the same host as you plan to run openttd, screen is easiest
11:54<@planetmaker>bolli, 'screen' is also a command. Not a piece of hardware
11:54*bolli googles
11:54-!-Prof_Frink [] has quit [Ping timeout: 480 seconds]
11:55<Ammler>but fully valid to mention that indeed
11:55<Ammler>I guess, we never tried, if autopilot would work in background
11:57<Ammler>bolli: screen is quite certain installed on your system already, so you could check the man page instead a random website
11:58<@Alberth>andythenorth: this is one of the simplest forms, I think
11:58*Alberth is afk for some dinner
11:58<Ammler>for start run "screen" then run "openttd -D", then simply quit your ssh session and reconnect and run "screen -x"
11:59<bolli>and how would i disconnect from that session if its in openttd?
12:00<Ammler>you hit the X on your console window
12:00<Ammler>"simulate" a disconnect
12:01<Ammler>there are of course detach etc. for screen, which you will find in the manpage, but who uses that? :-)
12:02<andythenorth>Alberth: interesting. It would need to decode which gestalt to use
12:02<andythenorth>currently the filenames do encode a gestalt ID, so possible
12:04-!-Rhamphoryncus [] has quit [Quit: Rhamphoryncus]
12:05-!-flaa [] has quit [Ping timeout: 480 seconds]
12:10-!-kkb110 [] has joined #openttd
12:24-!-Devroush [] has joined #openttd
12:28-!-morphium [] has joined #openttd
12:28<morphium>Hey, me again. Is it possible to compile ONLY a dedicated server version of Openttd?
12:28-!-morphium is now known as Guest5650
12:30-!-supermop [] has left #openttd []
12:31-!-__mj [] has joined #openttd
12:32<__mj>hi, quick question, whats the standard procedure for releasing a patch on openttd ? just post it in the forums with a description ?
12:33-!-Progman [] has joined #openttd
12:34<andythenorth>__mj: yes
12:34-!-HerzogDeXtEr1 [~Flex@] has joined #openttd
12:35<andythenorth>if it's a big patch, break it into chunks
12:35<NGC3982>you know you live close to food when you can irc from your own wifi, at the pizza shop.
12:36<__mj>andythenorth: like a patch for each file ?
12:36<andythenorth>a patch for each logical segment
12:36<andythenorth>so if you add a new hook for feature x, then that's one patch
12:36-!-Guest5650 [] has quit [Quit: ajax IRC Client]
12:36<andythenorth>and the code that uses your new hook is another patch
12:37-!-morphee [] has joined #openttd
12:37<andythenorth>or if you patch timetables and cargo loading, it's probably two patches etc etc
12:38<__mj>well, its rather one large aspect of the game, but I will find some way to break it up somehow
12:38-!-flaa [] has joined #openttd
12:39<andythenorth>a set of small patches are more likely to be reviewed
12:39<andythenorth>and nothing gets included without review
12:39<__mj>:) good
12:40<__mj>its the template based train replacement patch I posted about, in about a millenium ago
12:40<andythenorth>sounds like
12:41-!-HerzogDeXtEr [~Flex@] has quit [Ping timeout: 480 seconds]
12:41<__mj>uhm, yes, you pre-define virtual trains and tell which of your current trains shall get changed in the depot once they enter it, like autoreplacement but for whole trains instead of single vehicles
12:42<__mj>based on groups for control which trains will get replaced and which don't
12:47-!-Chris_Booth [] has joined #openttd
12:52-!-Firartix [] has joined #openttd
12:53-!-Prof_Frink [] has joined #openttd
12:58<bolli>my fiddling with processes has created a "frozen" openttd copy :/
13:04-!-FLHerne [] has joined #openttd
13:06<@Alberth>'suspend process' is also a kill signal :)
13:08-!-ProfFrink [] has joined #openttd
13:08<@Alberth>__mj: basic rule is indeed: one logical change == one patch file
13:09<@planetmaker>__mj, second basic rule: after each patch, OpenTTD must continue to compile and work
13:11<@planetmaker>a good guide on how things can be broken down is found at various times within the commit history of OpenTTD for the majority of the non-trivial features
13:11<morphee>I just give up, can't make Autopilot+ to work
13:12-!-Prof_Frink [] has quit [Ping timeout: 480 seconds]
13:13-!-Prof_Frink [] has joined #openttd
13:16<andythenorth>Alberth: wrt make....I wonder whether to just do second-simplest thing I can think of:
13:16<andythenorth>- have each gestalt write out a manifest for all of its pngs
13:16<andythenorth>tell make the manifest is the target...
13:17<__mj>what is a reasonable number of LoC per patch ?
13:18-!-ProfFrink [] has quit [Ping timeout: 480 seconds]
13:19<@Alberth>loc is not that interesting, the number of conceptual changes is.
13:21<@planetmaker>__mj, e.g. see r23932 to r23936 for an idea of how splitting makes sense
13:23-!-Prof_Frink [] has quit [Quit: irssi received SIGKILL - going byebyes]
13:24<@Alberth>andythenorth: hmm, nice idea. May be easier to add a check to each gestalt whether it should regenerate each gestalt (based on time stamps), and if not have it skip its generation. Then run the 5 or so gestalts in parallel.
13:24<andythenorth>each gestalt should be entirely self-contained imho
13:25<andythenorth>I've tried thinking of config files for them etc
13:25<@Alberth>also gives better options for caching data :)
13:25<andythenorth>but I come back to always 'just put the config in the python' :P
13:25<andythenorth>I also considered the case of a gestalt that might generate ~200 output pngs, which will be slow
13:26<andythenorth>(for slow = ~10s)
13:26-!-Pikka [] has quit [Quit: ChatZilla [Firefox 10.0.2/20120215223356]]
13:26<@Alberth>one time slow is not that bad in general
13:26<andythenorth>in that case, if something changes that requires regeneration, it probably requires regeneration of all
13:26-!-morphee [] has quit [Quit: ajax IRC Client]
13:26<andythenorth>unless a cargo sprite changed...
13:27<andythenorth>but I'd rather throw more CPU at these occasional tasks
13:27<@Alberth>has checking up-to-dateness any use then?
13:27<andythenorth>only of the input files
13:27<andythenorth>e.g. a dep check or such
13:27<@Alberth>sounds likely
13:28<andythenorth>user shouldn't be modifying / deleting output
13:28<andythenorth>things go wrong if they do that :P
13:28<andythenorth>in the grf
13:28<andythenorth>shall I make pixa write a manifest?
13:30<@Alberth>make does need a list of targets for it to decide whether to run the command, so in that sense, yes
13:30<@Alberth>ie like the -M output of gcc/g++
13:37-!-Wolf01 [] has joined #openttd
13:38<@Alberth>hello Wolf01
13:43-!-Maarten_ [] has joined #openttd
13:45-!-Chris_Booth [] has quit [Quit: ChatZilla [Firefox 11.0/20120228210006]]
13:45<CIA-1>OpenTTD: translators * r24019 /trunk/src/lang/french.txt:
13:45<CIA-1>OpenTTD: -Update from WebTranslator v3.0:
13:45<CIA-1>OpenTTD: french - 24 changes by OliTTD
13:47-!-Maarten [] has quit [Ping timeout: 480 seconds]
13:50<andythenorth>Alberth: manifests are now generated, in a horrible way
13:50<__mj>Alberth: 'loc is not that interesting, the number of conceptual changes is.' << well, it's 3 gui componentes and one big conceptual theme in my case
13:50<andythenorth>it needs far too much code added to the gestalt currently :P
13:50*andythenorth -> baby -> bath
13:50<andythenorth>I've pushed
13:53-!-andythenorth [] has quit [Quit: andythenorth]
13:55*zxbiohazardzx is no longer away : Gone for 3 hours 12 minutes 37 seconds
13:55<zxbiohazardzx>in window_type.h, i see alot changed between this old patch and new code
13:55<zxbiohazardzx>mostly more comments
13:55<@Alberth>__mj: it's fine to consider adding 1 gui component a conceptual change :)
13:56<zxbiohazardzx>can i just add WC_BUILD_HOUSE anywhere in the file or?
13:56<@Alberth>zxbiohazardzx: I doubt people are that interested in how many seconds you went afk
13:56<zxbiohazardzx>i didnt put it for that
13:56<zxbiohazardzx>im on multiple channels
13:56<zxbiohazardzx>so my IRC client puts away and back statusses automatically
13:57<zxbiohazardzx>so sorry for the spamzorz :P
13:57<@Alberth>it's not a status change here
13:57*planetmaker wonders whether *anyone* could possibly interested in that
13:57*zxbiohazardzx slaps planetmaker around abit with a large rainbow
13:57*zxbiohazardzx take that you unicorn!
13:58<zxbiohazardzx>anyway back to the contentual questions :P
13:58-!-bolli [~bolli@] has quit [Quit: Leaving]
13:58<zxbiohazardzx>in window_type.h can i just add a line wih WC_BUILD_HOUSE somewhere in the file or not?
13:59<zxbiohazardzx>as the previous patch shows them as more of an ENUM styled list
13:59<zxbiohazardzx>and they added it on the bottom just before WC_INVALID
13:59<zxbiohazardzx>but now there is alot more WC_* stuff in between and comments on top of each
13:59<NGC3982>im recently introduced to timetables
13:59<NGC3982>i barely understand various bits of it
13:59<NGC3982>though, can i set conditions based on the content of stations?
14:00<NGC3982>would be neat to tell a train to visist station X only when it contains Y amount of Q.
14:00<@Alberth>never seen such conditions
14:00<NGC3982>i see.
14:00<@Alberth>I also have some doubts whether it is possible
14:01<NGC3982>what we are trying to achieve is a pack of trains, waiting in depot until a big ass boat arrives with stuff.
14:01<@Alberth>the next order is evaluated several times, and if it changes in the mean-time, things get very tricky
14:02<NGC3982>i see.
14:02<@Alberth>you should check the available conditional orders to verify whether I am right
14:03<zxbiohazardzx>also in terraform_gui.cpp, did we stop using ETTW_stuff and replaced it with WID_stuff?
14:03<@Alberth>NGC3982: why not make some tracks for trains to wait?
14:03<zxbiohazardzx>trying to update a patch that was up to date @ 22277 so yeah
14:04<@Alberth>zxbiohazardzx: stuff changed when goal scripts needed access to the gui buttons
14:04<zxbiohazardzx>yeah i figured
14:04<zxbiohazardzx>im just trying to get this patch back on head
14:05<zxbiohazardzx>but i ran into some issues on the window_type.h (new comments, i think i added the WC_BUILD_HOUSE line correctly, but im wondering what the comment above the others exactly is
14:05<zxbiohazardzx>most likely my comment will be Build house; %Window numbers:
14:06<zxbiohazardzx>- #BuildHouse = #BuildHouseWidgets (or 0 - #BuildHouseWidgets not sure)
14:06<@Alberth>links to goal script stuff
14:07<zxbiohazardzx>well old file didnt link, and thats all in comments so it will be ignored anyway?
14:07<zxbiohazardzx>old file had:
14:07<zxbiohazardzx> WC_GRF_PARAMETERS,
14:07<zxbiohazardzx> WC_BUILD_OBJECT,
14:07<zxbiohazardzx>+ WC_BUILD_HOUSE,
14:08<@Alberth>the docs are ignored, but there is another table somewhere for the purpose of GS
14:08<zxbiohazardzx>id assume i can just insert the WC_BUILD_HOUSE in after WC_BUILD_OBJECT however im not sure why all the /*comments with widgets exist, as they are ignored right?*/
14:08<zxbiohazardzx>ah ok
14:08<zxbiohazardzx>then ill just add it and the docs
14:09<zxbiohazardzx>hmmz ill have to check out the terraform stuff though, you really overhauled that one :P
14:10<@Alberth>I didn't :)
14:10<zxbiohazardzx>the rest did :P
14:15-!-JVassie [~James@] has joined #openttd
14:18-!-cypher [] has quit [Quit: Miranda IM! Smaller, Faster, Easier.]
14:20<zxbiohazardzx>also is there a list of Hotkeys i cannot use for the GUI?
14:20<zxbiohazardzx>current list is QWEDUIOP for the existing ones, id like to claim H, but is there some hotkeys we cant map or?
14:23*Alberth does not know
14:24<@Alberth>Ammler: what about enforcing a filename convention instead?
14:24<zxbiohazardzx>ill just leave it as a TODO like in original
14:25<@Alberth>Ammler: imho that would be what you want for people with case-sensitive file systems anyway
14:33<__mj>in general, is it frowned upon to skip checks for text-direction settings ? I mean if you draw your strings only in left to right direction ?
14:34<zxbiohazardzx>i think most services only offer left-right directional texts
14:34<zxbiohazardzx>i know twitter only swapped recently XD
14:35<__mj>I always skipped the rtl checks in my code but made a few markers to maybe include it later on, in case the devs complain ^^
14:35-!-Prof_Frink [] has joined #openttd
14:35<zxbiohazardzx>Hotkey<ScenarioEditorLandscapeGenerationWindow>('H', "house", WID_ETT_PLACE_HOUSE), // TODO: check if this is valid
14:36<zxbiohazardzx>anyone see an objection against that binding, as i think its only bound when the scenario editor terraforming window is active
14:37<@Alberth>entire trunk does both directions
14:38<@Alberth>and we will thus complain if it gets that far :p
14:42-!-mahmoud [] has quit [Read error: Connection reset by peer]
14:43<zxbiohazardzx>i reached end of patch o_o
14:43<zxbiohazardzx>so its time to eeehm go check and diff it out :P
14:44-!-frosch123 [] has joined #openttd
14:46<zxbiohazardzx>ow yeah building ottd stinks now i remember :P
14:46<zxbiohazardzx>no cmake, no VS10 solution XD
14:47-!-andythenorth [] has joined #openttd
14:48<@planetmaker>you're at least 50% wrong
14:48<@planetmaker>nor do I understand your obsession with cmake
14:48<zxbiohazardzx>its wintendo user proof?
14:48<zxbiohazardzx>and for me its quite fast on the world of warcraft emulation builds :P
14:49<@planetmaker>./configure && make && make run
14:49<@planetmaker>dunno what's missing there
14:49<@planetmaker>and VS has its own projects
14:49<@planetmaker>nor does cmake magically install missing deps
14:51<zxbiohazardzx>fair enough
14:52<@planetmaker>(and make is more cross-platform than cmake ;-) )
14:52<TinoDidriksen>CMake uses make
14:52*andythenorth adventures into the fridge
14:52<zxbiohazardzx>indeed it does
14:52<TinoDidriksen>Or whatever else is on the platform
14:52<zxbiohazardzx>oeee andythenorth, get me some icecream?
14:53<andythenorth>bavarian smoked cheese? brunswick ham? Bitburger beer? Definite germanic theme in today's fridge
14:53<andythenorth>saucisson sec? chorizo?
14:53*andythenorth has been propping up the eurozone some more it seems
14:54<@Alberth>andythenorth: perhaps you should come live here too :)
14:54<andythenorth>in the eurozone?
14:54<andythenorth>perhaps :)
14:55<zxbiohazardzx>euro = win
14:55<zxbiohazardzx>stupid greece stinks though
14:55<andythenorth>I think I'd prefer california - for the weather, not the food
14:56<andythenorth>Alberth: any inspiration wrt make and manifests etc?
14:57<@Alberth>your solution is a bit reversed w.r.t. what you want, I think :)
14:57-!-peteris [~peteris@] has joined #openttd
14:59<andythenorth>Alberth: quite possible
15:00<@Alberth>hmm, running bandit is not trivial :(
15:00<andythenorth>which aspects?
15:00-!-Sacro_ is now known as Sacro
15:00<@Alberth>ImportError: cannot import name PageTemplateLoader
15:01<andythenorth>easy_install chameleon
15:01<andythenorth>I have a plan for fetching deps - later though
15:02-!-mode/#openttd [+o orudge] by ChanServ
15:06-!-zxbiohazardzx [] has left #openttd []
15:06<@Alberth>I installed a chameleon python package which is some zpt template language, and the loader code looks dirty enough to be from easy_install. Unfortunately, it also does not explain where it got installed :(
15:10<andythenorth>is that bad?
15:12<@Alberth>if you want to look whether such a file exists in the package, it might be :p the manifest does not list it, so I assume it's the wrong one.
15:13<andythenorth>well it sounds like you got the correct package
15:13<andythenorth>tbh I have no idea how this is usually done
15:13<andythenorth>I live in a world of buildout for python
15:14<andythenorth>run buildout, get all deps needed, isolated to one local path, not spewed all over your system python
15:14-!-supermop [] has joined #openttd
15:14<@Alberth>I don't have write access there :)
15:14<andythenorth>usr python :P
15:15*andythenorth plans to provide a buildout for BANDIT
15:16<@Alberth> could be one step towards a more uniform .png name
15:16*andythenorth thinks supermop will be busy now
15:17<andythenorth>Alberth: surely you can that into one line :D
15:17<@Alberth>yes :)
15:17-!-__mj [] has quit [Remote host closed the connection]
15:18<@Alberth>return '_'.join(name if name else 'none' for name in raw) + '.png
15:19<@Alberth> -2 lines :)
15:26<@Alberth>from the gestalt, you can deduce the number of features, it seems
15:27-!-Pikka [] has joined #openttd
15:28<andythenorth>it's quite important to know that gestalts may differ wildly from each other
15:29<andythenorth>the only commonality is that they output n pngs
15:29<@Alberth>if you start with the gestalt in the .png name, you can use that to decode the other parameters
15:30<andythenorth>so 'tipping_trailer' etc
15:30<andythenorth>or 'tippingtrailer' depending on how much effort we want to put into splits
15:31*Alberth wonders whether 'generate' could be lifted out of the gestalts
15:32<@Alberth>tipping-trailer :)
15:34<andythenorth>Alberth: I tried to figure out lifting generate
15:34<andythenorth>it's too specific to the combination of variables
15:34<andythenorth>or configuring pixa gets too complex ;)
15:34<andythenorth>I think it could be decomposed into more functions though
15:36-!-pjpe [] has joined #openttd
15:37<@Alberth>what if you put the variables and their ranges in a dict?
15:37-!-mahmoud [] has joined #openttd
15:38<@Alberth>eg a string -> list of values mapping
15:38-!-Booth [] has joined #openttd
15:38-!-namad7 [] has joined #openttd
15:42<andythenorth>Alberth: there's an interesting overlap with BANDIT's
15:42<andythenorth>which has to know some stuff about available graphics
15:42<andythenorth>this is rather BANDIT specific though
15:48<@Alberth>nothing wrong with that
15:48<@Alberth>you'd use a class here, I think, where the basic process is defined in the base class, and you can refine some details of the process in a derived class
15:49<@Alberth>but it gets a bit tricky to explain that to people not used thinking in that way
15:49-!-Booth is now known as Chris_Booth
15:50<andythenorth>it's fine for me
15:50<andythenorth>I'm happy to have gestalts inherit
15:50<andythenorth>or have them use a generate class, with inheritance
15:51<andythenorth>I guess my concern is making this available to people who just want to replace load sprite colours or such
16:00-!-DDR [] has joined #openttd
16:04-!-lofejndif [] has joined #openttd
16:09-!-Devroush [] has quit [Ping timeout: 480 seconds]
16:10-!-Devroush [~dennis@] has joined #openttd
16:12<andythenorth>oh how interesting
16:12<andythenorth>openttd won't build
16:13<@planetmaker>your icu seems... strange or different
16:13<andythenorth>I did a port upgrade on everything today
16:13<andythenorth>not 100% intentionally
16:14<Rubidium>typical upgrade of icu
16:14<andythenorth>I should downgrade?
16:14<Rubidium>touch src/strings.cpp src/gfx.cpp shouldsolve it
16:14<Rubidium>+ a make
16:16<Rubidium>andythenorth: a downgrade is not needed
16:16<Rubidium>just a selective recompile of the stuff that is linked to ICU
16:22-!-valhallasw [~valhallas@] has joined #openttd
16:29-!-flaa [] has quit [Quit: leaving]
16:29-!-flaa [] has joined #openttd
16:31<andythenorth>Alberth: so the aspect I miss - if the pngs specify the to create the pngs? manually?
16:33<@Alberth>I'd say you have the ranges for the values already, you just need to link them to the variables. Then you can quite easily generate all combinations, without hooking it into the generator
16:36<andythenorth>so BANDIT generates the filenames?
16:37<@Alberth>with anything > 5 filenames, I'd prefer that very much :D
16:37<andythenorth>can it just do something like 'touch foo.png'
16:37<andythenorth>rather than opening PIL and writing actual pngs?
16:37<@Alberth>theoretically yes, except then it is up-to-date w.r.t. all existing files
16:38<andythenorth>I'd need a specific arg to do that I think
16:39<andythenorth>otherwise building the grf means running the generator :)
16:39<@Alberth>add a -M option :D
16:39<andythenorth>I'll keep this as notes
16:40<Ammler>Alberth: we don't have a case insensitive file name convention, not sure, we would want one
16:40<@Alberth>Ammler: I was refering to a case-sensitive filename convention
16:41<@Alberth>if you have that, you cannot make 'double' filenames any more
16:42<Ammler>so you would force people to have lower case only?
16:43<@Alberth>for example, yes
16:43<andythenorth>all my acronyms :(
16:43<@Alberth>otherwise you get foo.txt Bar.txt SPAM.TXT
16:43<Ammler>as said, I would not like that, of course if that is the only possibility, it needs considering...
16:44<andythenorth>most of my files are lower case
16:44<andythenorth>it's a habit from making the internets :P
16:44<Ammler>Alberth: i prefer README, others prefer readme.txt
16:44<@Alberth>all my filenames are lowercase, I hate needing the shift key to enter a filename :)
16:45<andythenorth>apparently we have a lot of though ;)
16:45<andythenorth>I have some FISH and BANDITS and HEQS
16:45<@Alberth>Ammler: you can have only one readme in a project
16:45<andythenorth>no FIRS though
16:45<andythenorth>readme.txt of course
16:46<andythenorth>upper caps fails readability
16:46<Ammler>Alberth: that's the point such a convention might be usefl per project
16:46<@Alberth>GNU said the name should be Makefile
16:46<Ammler>buton devzone there are 50? projects
16:46<andythenorth>GNU can pay wear and tear on my shift key :P
16:47<Ammler>and many individuals, which all like different file names :-)
16:47<@Alberth>Ammler: ah, you want one solution for all projects? ok
16:47<Ammler>well, I want a solution, we can enable per default,
16:48<@Alberth>I was just pondering about the problem, and after realizing you are asking for something completely new (namely looking at existing files in the project), I got wondering whether other solutions were possible,
16:49<andythenorth>do we cache file names for the dep check?
16:50<@Alberth>andythenorth: tbh, I am quite not sure you'd want to run pixa for generating just one .png file
16:50<andythenorth>Alberth: you wouldn't :)
16:50<andythenorth>the setup overhead wouldn't pay off
16:50<andythenorth>I am planning to generate ~3000
16:50<andythenorth>which is acceptable
16:51<@Alberth>so you need to be able to define sets of pngs to construct
16:52<andythenorth>I could create classes for them in the BANDIT code
16:52<andythenorth>but the classes would be effectively doing what gestalts do
16:52*andythenorth is scratching head still
16:55<andythenorth>I have the mapping of cargos to gestalts
16:56<andythenorth>but not the coloursets
16:56<andythenorth>but BANDIT will need to know about coloursets anyway, to code them
16:56<andythenorth>Alberth: I'll sleep on it, might wake up with the answer
16:56<andythenorth>unless you patch in the meanwhile :D
16:56<@Alberth>you could make a file with the names (simple, but not very maintainable), list the files on the command-line, or define it more globally in a constraint-form
16:58<andythenorth>I think there's an answer somewhere in that BANDIT has to know the file patterns
16:59<andythenorth>then gestalts aren't tied to BANDIT so much, and could maybe be reused easier by other people
16:59<@Alberth>you'll need that, I think when you want more control over what an invocation of Pixa will do
17:01<@Alberth>the question is however whether you can split the variables and their values nicely enough from the generate code
17:02<@Alberth>hmm, perhaps add some extra function parameters to handle the special cases?
17:02<@Alberth>(that is, give generate a function as parameter, which it can call)
17:02<Nat_aS>is there any reason why trains going diagonally appear larger?
17:03-!-tokai|noir [] has joined #openttd
17:03-!-mode/#openttd [+v tokai|noir] by ChanServ
17:03<@Alberth>which of course a framework in disguise :p
17:04<andythenorth>framework is fine ;)
17:05<andythenorth>I should write out a list of filename patterns
17:05<@Alberth>Nat_aS: there probably is.
17:06<andythenorth>some are very simple: e.g. tanker trailers have only colour variations. There's no difference per cargo.
17:06<andythenorth>others have lots of variations
17:06<Nat_aS>is there any people from the unrealistic trainset on right now?
17:06<andythenorth>gestalts could chain
17:06<andythenorth>stake-trailer is a flat-trailer with a parameter
17:07<andythenorth>tipping-trailer-4px is a tipping-trailer with parameter for 4px high body
17:07<@Alberth>Nat_aS: all trainsets are unrealistic by that standard :p
17:07<Nat_aS>V453000 Hanf Chris_Booth Mazur stephanbarth
17:07<Nat_aS>no the unrealistic trainset is the name of a specific newgrif
17:07-!-tokai|mdlx [] has quit [Read error: Operation timed out]
17:07<@Alberth>Nat_aS: not very nice doing that
17:08<Mazur>!kill Nat_aS
17:08<Chris_Booth>a few of us are here
17:08<Nat_aS>pinging people?
17:08*Chris_Booth saw Mazur doing it!
17:10<Nat_aS>what's wrong with pinging to get people's attention?
17:10<Nat_aS>anyways Chris, can I see that sheet of engine types?
17:10<Chris_Booth>what sheet of engines?
17:10<@Alberth>because you don't know whether they know the answer
17:11<Chris_Booth>everything for NUTS can be found in
17:11<@Alberth>at most you should only ping that person that knows the answer
17:11<Nat_aS>no, those are people who DO know the answer
17:11<Nat_aS>they are listed on the Unrealistic trainset's webpage
17:11<@Alberth>then you still don't need them all 5
17:11<Chris_Booth>Nat_aS: no we do not
17:11<Nat_aS>the question is "Are you on"
17:12<Nat_aS>only they can answer that question.
17:12<Nat_aS>and the answer can only be yes.
17:12<andythenorth>if you must do it, limit yourself to 2
17:12<Nat_aS>a negitive result comes from not answering.
17:12<andythenorth>or do @seen Nat_aS
17:12-!-Pikka [] has quit [Quit: ChatZilla [Firefox 10.0.2/20120215223356]]
17:12<andythenorth>@seen Nat_aS
17:12<@DorpsGek>andythenorth: Nat_aS was last seen in #openttd 37 seconds ago: <Nat_aS> a negitive result comes from not answering.
17:13<andythenorth>but too much of that is annoying too
17:13<Nat_aS>that's probably more spammy
17:13<@Alberth>just as annoying for the pinged person
17:13<Nat_aS>and makes more channel noise than just listing there names.
17:14<andythenorth>Nat_aS: why didn't you just look in the list of people in the channel?
17:14<Nat_aS>that wont tell me if they are AFK or not
17:14<Nat_aS>not even @seen will do that
17:14<Chris_Booth>Nat_aS: do you mean this: ?
17:14-!-lmergen [] has joined #openttd
17:14<Nat_aS>it's a perfectly normal way to get somebody's attention on irc
17:14<Nat_aS>and yes thank you!
17:14<Nat_aS>that's what I wanted!
17:15<Chris_Booth>you could have found that yourself where I told you on
17:15<Nat_aS>oh that
17:15<V453000>there are people from unrealistic people
17:15<Nat_aS>that is where i looked first
17:15<Nat_aS>and could not find it.
17:15<V453000>what do you need?
17:15<Nat_aS>I just wanted to look at the list
17:15<Chris_Booth>just to be annoying V
17:16<Nat_aS>I have my own ideas for a balanced fantasy trainset.
17:16<V453000>list ... is at the devzone, issue called engine table iirc
17:17<Chris_Booth>V read up ^ I posted it
17:17<Chris_Booth>Chris_Booth: do you mean this: ?
17:17<Nat_aS>I might be intrested in helping you guys (I used to make trains for simutrans, one of which is still in the current build) or just use your ideas as a springboard
17:17<Nat_aS>and yes I see it now, I just didn't look in the right place there
17:17<Chris_Booth>Nat_aS: please join #openttdcoop.devzone if you want to help with this set
17:18<andythenorth>Alberth: filenames might need some kind of delimiters :|
17:18<andythenorth>I don't trust delimiters :)
17:18-!-Chris_Booth [] has quit [Quit: Night Night all]
17:19<@Alberth>andythenorth: just check them to death (split, check how many you have, check whether they match with known values)
17:20<andythenorth>so if i in foo, otherwise raise...
17:20<@Alberth>I am a defensive programmer, so yes, I would do that :)
17:20<Nat_aS>this might be a rude question but what is it with open source communities and not liking people who ask questions?
17:21<@Alberth>it's mostly how you ask, I think
17:21-!-mahmoud [] has quit [Ping timeout: 480 seconds]
17:22<Nat_aS>it's because calling people by name makes the post show up in red, so it looks angry
17:22<Nat_aS>whoever programed pings didn't consider the implications of that.
17:27<andythenorth>I think it was mostly because your ping included 2 people not in the channel, which is silly :D
17:27<andythenorth>Alberth: so I'm thinking some kind of dispatcher script, that decodes the filenames (similar to your example paste)
17:28<@Alberth>and also being very pushy, imho
17:28<andythenorth>I could do the splits in make, maybe, but make syntax has zero appeal to me :P
17:28*planetmaker expects to be highlighted with a meaningful question / statement in the same line or only when I'm currently active in the channel
17:28<@planetmaker>otherwise I'll just ignore the highlight even when I'm online
17:29<@planetmaker>(and might ignore further ones from that person)
17:29*andythenorth is somewhat inspired by url dispatchers in web frameworks, which use patterns to find which class(es) need to handle the request
17:30<andythenorth>stuff like this:
17:31<Nat_aS>Chris_Booth, you aren't in that channel
17:32<andythenorth>hmm andythenorth has never seen pesto before, but it solves an equivalent problem for pixa: it will also generate a url for anything the dispatcher can handle
17:33<@Alberth>andythenorth: splitting in python seems to make most sense to me
17:33*andythenorth is thinking aloud, probably too much
17:33<@Alberth>we're used to you doing that :)
17:34-!-Pulec [] has quit []
17:34<andythenorth>so ideally the decoder would be a set of classes / methods that can both decode (and pass to pixa)
17:34<andythenorth>and also encode....called by BANDIT with params, when the -M arg is used (or similar)
17:34<andythenorth>then all the patterns can be kept in one location
17:35<@Alberth>-M could be useful, at least for debugging; not sure whether you actually need it, as you want to generate groups of them anyway
17:36<andythenorth>I only mention it because I'm short of ideas for generating initial pngs
17:36<andythenorth>the encoder / decoder could do it
17:36<@Alberth>ie I was thinking in terms of gestalt=X length=7 length=5
17:36<@Alberth>as command-line aruments
17:37<andythenorth>I think I need to understand the pipeline a little more
17:38<andythenorth>so: to generate rendered pngs: make -> dispatcher -> gestalt -> pixa -> png
17:38<andythenorth>to configure make?
17:40<@Alberth>hmm, tricky
17:41<andythenorth>BANDIT -> mappings in global_constants -> dispatcher -> write empty pngs?
17:41<@Alberth>maybe just use a single file as reference for building a whole group?
17:41<andythenorth>could be
17:42<@Alberth>which sort of implies you either generate all of them or none
17:42<andythenorth>I'm fine with that
17:42<andythenorth>if I've understood correctly though, it's similar to what a gestalt currently does...
17:42-!-peteris [~peteris@] has quit [Quit: Ex-Chat]
17:43<@Alberth>it does not have to be a real 'png' file even, it can be just a dummy empty file
17:43<@Alberth>good point
17:43*andythenorth wonders whats 'intrinsic' to a gestalt, and what's 'extrinsic'
17:43<andythenorth>cargo might be extrinsic
17:44<andythenorth>cargo also happens to be something BANDIT needs to specifically handle
17:44<andythenorth>sleep is something currently extrinsic to andythenorth
17:44<andythenorth>that might have to change :D
17:44-!-FLHerne [] has left #openttd []
17:45<@Alberth>do you want those groups being equal to the gestalts, or should that be more flexible
17:45<andythenorth>I think they're equal so far
17:45<andythenorth>it seems to work
17:45<andythenorth>but it was only done this way for convenience
17:46<andythenorth>one way might be to generate against cargo sprites
17:47*Alberth feels sleepy
17:47<andythenorth>which I think I'll generate in a separate project, then put into BANDITs repo
17:47<andythenorth>+1 to sleep
17:47*andythenorth -> bed
17:47<andythenorth>the answer usually presents itself when I wake up
17:48<andythenorth>if not I'll write out all the filename combinations I can think of in pencil on paper
17:48<andythenorth>that usually does the trick
17:48<andythenorth>good night
17:50-!-andythenorth [] has quit [Quit: andythenorth]
17:52-!-smoovi [] has quit [Quit: Verlassend]
17:55-!-Alberth [] has left #openttd []
18:00-!-lmergen [] has quit [Ping timeout: 480 seconds]
18:02-!-supermop [] has quit [Quit: supermop]
18:06-!-TGYoshi [~TGYoshi@] has quit [Quit: Popidopidopido]
18:14-!-JVassie [~James@] has quit [Ping timeout: 480 seconds]
18:16-!-morphium_ [~quassel@] has joined #openttd
18:20-!-morphium_ [~quassel@] has quit []
18:21-!-morph` [~quassel@] has joined #openttd
18:25<morph`>OK, there's this peanut-butter cookie on my desk
18:25<morph`>It tastes really good, but I am on a diet for last 10 months
18:25<morph`>Should I?
18:36-!-sla_ro|master [slaco@] has quit [Quit: DANGER is OFFLINE DANGER]
18:37<Wolf01>uhm, you reminded me I have a box full of biscuits hidden here around
18:41-!-Zeknurn` [] has joined #openttd
18:44-!-Zeknurn [] has quit [Ping timeout: 480 seconds]
18:46-!-cap [] has joined #openttd
18:46-!-flaa [] has quit [Quit: leaving]
18:47-!-Lurker [] has joined #openttd
18:47-!-cap [] has quit []
18:56-!-Lurker [] has quit [Quit: ajax IRC Client]
18:59-!-lofejndif [] has quit [Quit: Leaving]
18:59-!-Progman [] has quit [Remote host closed the connection]
19:03-!-Wolf01 [] has quit [Quit: Once again the world is quick to bury me.]
19:03-!-KritiK [] has quit [Quit: Leaving]
19:06-!-TWerkhoven [] has quit [Quit: He who can look into the future, has a brighter future to look into]
19:15-!-Devroush [~dennis@] has quit []
19:17-!-Biolunar [] has quit [Quit: All your IRC are belong to us]
19:20-!-valhallasw [~valhallas@] has quit [Ping timeout: 480 seconds]
19:25<morph`>Possible to compile and install JUST dedicated server of OpenTTD?
19:36<frosch123>yes, if you are on *nix
19:36<frosch123>configure has a --dedicated option or similar
19:37<frosch123>actually "./configure --enable-dedicated"
19:44-!-frosch123 [] has quit [Remote host closed the connection]
19:46-!-Firartix [] has quit [Ping timeout: 480 seconds]
19:53-!-Rhamphoryncus [] has joined #openttd
19:54<morph`>Well, compiling doesn't work for some reason on my Debian Squeeze - I get
19:54<morph`>I can't "make" after Configure ends
19:54-!-morph` [~quassel@] has quit [Read error: Connection reset by peer]
19:59-!-pugi [] has quit []
21:11-!-pjpe [] has quit [Quit: ajax IRC Client]
21:23-!-Maarten_ [] has quit [Read error: Connection reset by peer]
21:28-!-pjpe [] has joined #openttd
21:49-!-Elukka [] has quit []
23:02-!-glx [glx@2a01:e35:2f59:c7c0:747f:3f81:7f24:a684] has quit [Quit: bye]
23:07<Nat_aS>anybody awake?
23:08<Nat_aS> can anybody crit this work in progress?
23:08<Nat_aS>based on this crazy nazi train
23:09<Nat_aS>is there a max height for trains before they start hitting bridges?
23:14<Rhamphoryncus>nothing in the game mechanics. Just aesthetic
23:14<Rhamphoryncus>If you really look close you'll realize that all trains are several times taller than tunnels/bridges :)
23:15<Nat_aS>oh cool
23:15<Nat_aS>and lol
23:17<Rubidium>no try that in a rendered 3D world
23:22<Eddi|zuHause>Nat_aS: i think your side view is backwards
23:23<Eddi|zuHause>it looks in <- direction, but should be -> direction on that spot
23:23<Nat_aS>for the lead engine or the tail car?
23:24<Eddi|zuHause>all of them
23:24<Nat_aS>the one facing > is the lead engine, the one facing < is the tail car
23:24<Nat_aS>so they should be the other way around?
23:25<Eddi|zuHause>that. or your preview is backwards
23:25<Nat_aS>the preview is just copy paste
23:25<Eddi|zuHause>in depot, the train is shown in <- direction
23:25<Eddi|zuHause>so i just assumed your full train is like that
23:26-!-rickytaylor26 [~quassel@] has quit [Remote host closed the connection]
23:26<Nat_aS>I just made it facing the direction the picture it is based on is.
23:28-!-ricky26 [~quassel@] has joined #openttd
---Logclosed Sat Mar 10 00:00:44 2012