Back to Home / #openttd / 2011 / 11 / Prev Day | Next Day
#openttd IRC Logs for 2011-11-10

---Logopened Thu Nov 10 00:00:09 2011
00:57-!-Eddi|zuHause2 [] has joined #openttd
00:57-!-DorpsGek is now known as Guest16550
00:58-!-DorpsGek [] has joined #openttd
00:58-!-mode/#openttd [+o DorpsGek] by ChanServ
01:01-!-Guest16550 [] has quit [Ping timeout: 480 seconds]
01:03<Rubidium>ctibor: because good service only increases the chance the production will increase, which means it may very well not increase or even decrease
01:04-!-Eddi|zuHause [] has quit [Ping timeout: 480 seconds]
01:15<CIA-6>OpenTTD: rubidium * r23178 /trunk/src/ (5 files in 2 dirs): -Feature [FS#4780]: in-game readme.txt readmer (LordAro)
01:34-!-Prof_Frink [] has quit [Ping timeout: 480 seconds]
01:36-!-Celestar [] has joined #openttd
01:36-!-JVassie [~James@] has joined #openttd
01:44<@planetmaker>moin Rubidium :-)
01:44<@planetmaker>and also all others
01:53-!-Celestar [] has quit [Ping timeout: 480 seconds]
02:01-!-JVassie [~James@] has quit [Ping timeout: 480 seconds]
02:11-!-pugi [] has joined #openttd
02:12-!-Cybertinus [] has joined #openttd
02:23-!-cowsgomoo [~xren@] has joined #openttd
02:24-!-andythenorth [] has joined #openttd
02:24<andythenorth>Eddi|zuHause2: in the 'large' set scheme - containers?
02:25-!-KOPOBA [~xren@] has quit [Ping timeout: 480 seconds]
02:25<andythenorth>containers are interesting case - IRL they carry near everything, including bulk. In most sets they are piece / express
02:35-!-DayDreamer [] has joined #openttd
02:35*planetmaker has found another interesting thing and points at trunk head. HEQS is lacking there ;-) As is FISH and FIRS and ... every NeWGRF ;-)
02:40-!-HerzogDeXtEr [~Flex@] has joined #openttd
02:40<andythenorth>planetmaker: all I get is 504 :P
02:41<andythenorth>we've chosen to delete the repo and end the project? :o
02:41<andythenorth>it's 'done'
02:42<andythenorth>but yes, hg shows me tip, now I have to write text files :D
02:47-!-sla_ro|master [~slaco@] has joined #openttd
02:53-!-Progman [] has joined #openttd
02:54<@planetmaker>yes, we've got to write readme files now in a ledgible format
02:54<@planetmaker>Though, andythenorth, you're doing better than Pikka ;-)
02:54<@planetmaker>He supplies no readme whatsoever :-P
02:54<andythenorth>is the new feature announced in forums?
02:54<@planetmaker>nor do Georges' newgrfs supply readmes
02:54<@planetmaker>no(t yet)
02:54<andythenorth>I think I asked for it in my ponies thread
02:55<@planetmaker>I thought of making a short announcement
02:55-!-Celestar [~dax@] has joined #openttd
02:56<@planetmaker>TrueBrain, 504 on
02:56<andythenorth>I had 'bananas changelog field' as a pony
02:57<andythenorth>frosch got ahead of my idea about mapping equivalent cargo labels :(
02:57<andythenorth>and it's likely that his is a good idea and mine was silly :P
03:00<@planetmaker>I haven't yet quite understood what he wrote. My tea water is just boiling...
03:06<andythenorth>I haven't understood all of it either
03:06<andythenorth>my idea was that cargos could map equivalence
03:07<@planetmaker>equivalent cargos sounds like identical cargos which should have the same label ;-)
03:07<andythenorth>so if your cargo set introduces bauxite, you could say "copper ore is a known cargo, bauxite should be treated 1:1 like copper ore"
03:08<andythenorth>I think frosch is doing something slightly different
03:08<@planetmaker> <- readme announcement
03:10<andythenorth>planetmaker: so some of my grfs already have readme.txt....
03:10*andythenorth tests
03:10<@Terkhen>good morning
03:10<@Terkhen>nice, readmes can be read :)
03:12<andythenorth>I think I understand frosch's proposal
03:12-!-Celestar [~dax@] has quit [Ping timeout: 480 seconds]
03:12<andythenorth>HEQS readme isn't seen :P
03:13<andythenorth>nor FISH
03:13<andythenorth>they're not from bananas
03:14<andythenorth>now to make them readable *before* you download :)
03:14<@planetmaker>andythenorth, all of your NewGRF have one
03:14<andythenorth>maybe read-before-download is not needed
03:14<@planetmaker>but not all are very informative ;-)
03:15<@planetmaker>And good point. I have to adjust the 'make install' so that it tars the stuff
03:15<@planetmaker>such that the readme viewer works without hassle
03:15-!-DayDreamer [] has quit [Quit: Leaving.]
03:15<andythenorth>should I stop maintaining separate changelog? I would like the change information in the readme
03:16<@planetmaker>Yes, I thought about that, too. Especially with the ingame readme viewer
03:17<@planetmaker>except if we add a changelog viewer :-P
03:17<@planetmaker>which now would be very simple to add
03:17<@planetmaker>good idea? bad idea?
03:17<@Terkhen>IMO displaying a single file is enough
03:18<@planetmaker>then I have to merge all changelog.txt into the readme :-P
03:18<andythenorth>planetmaker: indifferent idea :)
03:18<andythenorth>changelog should show up before downloading - on a button called "what's new" :P
03:18<@planetmaker>would be just another button...
03:19<@Terkhen>that's my main problem with showing different text files: it needs buttons :)
03:19<@Terkhen>besides that I don't care
03:19<andythenorth>when you know the set, the stuff you want most is "what new stuff do I get?"
03:19<@planetmaker>and the other way around
03:19<andythenorth>but if it's new you want "what does it do"
03:19<@planetmaker>if you don't know it, the "what's new" doesn't give a picture
03:20<andythenorth>*short* changelog copied in - latest features / fixes only?
03:20<@planetmaker>possibly. for now
03:20<@planetmaker>That's probably what I'll employ. The last or last two changelogs
03:21<andythenorth>I now have an insane amount of newgrf work to do :)
03:22<@Terkhen>hi appe
03:22<appe>whats up?
03:22<@Terkhen>1) Short description of the newgrf
03:22<@Terkhen>2) Short changelog
03:22<@Terkhen>3) Rest of the stuff
03:22<@Terkhen>4) Full changelog
03:25-!-Celestar [~dax@] has joined #openttd
03:26<@planetmaker>sounds complicated, Terkhen
03:26<@planetmaker>especially as bananas supports a changelog.txt
03:26<@Terkhen>it works on the principle of tl;dr :P
03:27<@planetmaker>of what?
03:27<@Terkhen>too long; didn't read
03:27<@planetmaker>ah :-P
03:28<@planetmaker>I mean... OpenGFX' changelog has 458 lines at the moment
03:28<@planetmaker>which is too long to read :-P
03:28<@planetmaker>(attention span of 20 lines or so :-P )
03:28<andythenorth>tl;dr is pretty good
03:28<@Terkhen>4) should go in a different file then
03:29<@planetmaker>yes, that's all I wanted to say: keep 4) in the file as now. But make sure 2) is there
03:29<andythenorth>works for me
03:29<@planetmaker>I guess we all agree but didn't notice :-P
03:29<andythenorth>one more c+p before release :P
03:30<@planetmaker>shall I script that, andythenorth ? ;-)
03:30<@planetmaker>like {{new_changelog}}
03:30<andythenorth>you might go mad on edge cases
03:30<andythenorth>unless we fragment the changelog
03:30<@planetmaker>of course. I'll require a certain changelog format then. In the way I use it all the time ;-)
03:31<@planetmaker>Version 0.2.1
03:31<andythenorth>if you do it I'll try it
03:31<@planetmaker>^^ pretty unique and identifiable
03:31<andythenorth>if I break it, I won't ask for support :P
03:31<andythenorth>c+p is not too painful, releases are not frequent
03:31<@planetmaker>but it's not too high on my agenda right now, tbh. But I'll make a note
03:32<andythenorth>so much to do :o
03:32<andythenorth>if someone fixed newgrf vehicle smoke, I think my head might actually explode
03:34<@planetmaker>I'm afraid we can't allow that, andythenorth.
03:34<@planetmaker>a) who'll clean up all the debris?
03:34<@planetmaker>b) your head is better used where it is now
03:34-!-Neon [] has joined #openttd
03:34<@planetmaker>--> no new smoke now :-P
03:34<andythenorth>it's such a dumb situation, the patch was added, but is not usable for ships :)
03:35<andythenorth>it's extra painful to see that it's in the spec now :P
03:36<andythenorth>also....I'm thinking that Eddi|zuHause2 might have an unarguable point about two props for refittable labels
03:36<andythenorth>I think an exclude prop is wrong, but maybe the way it has to be done
03:40<andythenorth>lacks elegance :(
03:40<dihedral>greetings y'all
03:41<@Terkhen>hi dihedral
03:41-!-Brianetta [] has joined #openttd
03:41<dihedral> seems to be down <- TrueBrain, Rubidium ;-)
03:41<dihedral>or whoever takes care
03:42-!-|Jeroen| [] has joined #openttd
03:42<dihedral>good morning Terkhen
03:42<@planetmaker>helo dihedral
03:42<dihedral>and you too :-)
03:43<dihedral>interesting :-)
03:43<dihedral>thank you
03:44<dihedral>was interested in looking at the ingame readme viewer :-)
03:44-!-HerzogDeXtEr1 [~Flex@] has joined #openttd
03:44<Celestar>morning :D
03:45<appe>whats vcs and hg?
03:46<Celestar>how the hell do I git add all the changed files without using some damn awk magic?
03:47<dihedral>does it now work on a folder?
03:47<dihedral>or recursively
03:48-!-pugi [] has quit [Quit: I reject your reality and substitute my own]
03:49<Celestar>git add -u seems to do the trick
03:50-!-HerzogDeXtEr [~Flex@] has quit [Ping timeout: 480 seconds]
03:51-!-andythenorth [] has quit [Quit: andythenorth]
03:54<Celestar>I'm still amazed how fast the goddamn code is.
03:56<Celestar>on my comp, the tile loop takes 87 nanoseconds per tile.
04:00<Celestar>michi_cc: if the "tile stack" at one place has independent tile loop procs, couldn't we put them into seperate threads?
04:01<@planetmaker>would it remain deterministically?
04:02<Celestar>that's what I'm wondering.
04:02<@planetmaker>Think of industry, airport, object or house tiles querying the adjacent tiles
04:02<Celestar>planetmaker: yeah, adjacent.
04:02<@planetmaker>+- 0x0F
04:02<Celestar>planetmaker: I'm think of threading it purely in the "Z" direction
04:04-!-Eddi|zuHause2 is now known as Eddi|zuHause
04:04*dihedral is wondering if the readme viewer could be used to display a 'rules' text from a server one would like to join :-P
04:05<dihedral>or if that would not be total overkill
04:05<@planetmaker>that's not a readme viewer :-P
04:05<@planetmaker>Do you agree to the TOS [ ] yes [ ] no
04:06<@planetmaker>*<font size = 3>You agree to pay a fine of 50€ for the first kick and a administration fee of 500€ should you violate the rules and this violation result in a ban.</font>Please fill in credit card details here: ....
04:14<Celestar>planetmaker: the main question is, what happens if a tile is deleted during the Loop processing?
04:17<@planetmaker>I'm not entirely familiar with the involved loops: The tile loop only visits certain tiles per tick, right, like every 1/256 tiles?
04:17<@planetmaker>i.e. every tick some tiles?
04:18<Celestar>planetmaker: every tick 1/256th of the map.
04:18<@planetmaker>Then we have to make sure that those tiles cannot access eachother
04:18<@planetmaker>which is... difficult
04:18<Celestar>I still can't follow :P
04:18<@planetmaker>as e.g. deleting a house can change population of a town which again can be read by tiles MUCH more distant
04:19<Celestar>planetmaker: with newmap, there are multiple tiles at a single location.
04:19<Celestar>planetmaker: if this location is visited during the tile loop, each of the "subtiles" has its tile loop run.
04:19<@planetmaker>got me confused :-)
04:19<@planetmaker>not x/y processing parallel. only z
04:20<Celestar>planetmaker: those tile loops are (possibly) independent from one another.
04:20<Celestar>planetmaker: yes :D
04:20<@planetmaker>depends on whether they can access the previous state or not
04:20<@planetmaker>if they can: difficult
04:20<@planetmaker>or rather: the killer
04:20<Celestar>the previous state? you mean the state of the "lower/upper" tiles?
04:20<Celestar>from all I have seen: no.
04:21<@planetmaker>it could be threaded then, possibly, if the answer is "no and never will" ;-)
04:21<Celestar>the main concern I have is what happens if the tile loop removes a tile from the stack.
04:21<Celestar>is the vector we store the stuff in thread-safe? :P
04:22<dihedral>why would the tile loop remove a tile??
04:22<@planetmaker>dihedral, removing the subway
04:22<Celestar>dihedral: removal of the last tree on a tile.
04:22<Celestar>dihedral: then MP_CLEAR + MP_TREE would become MP_CLEAR
04:22<@planetmaker>how would that remove the tile, Celestar ?
04:23<Celestar>planetmaker: it would remove the "tree" part.
04:23<@planetmaker>hm, I see. I wasn't aware of that kind of "level"
04:24<Celestar>planetmaker: well according to michi_cc, it's experimental anyway.
04:24<dihedral>lets have it committed to trunk then :-P
04:24<@planetmaker>and call it openttdpatch
04:25<Celestar>anyway. The removal of the last tree is the only thing that can remove a tile during the tile loop.
04:26<dihedral>how come?
04:26<dihedral>what if the tile only contains a building?
04:27<Celestar>the removal would need to be thread safe. and I'm not sure it can be.
04:27<Eddi|zuHause>how about: memcopy the entire map at start of tile loop, lock that copy and use it for all further read accesses. fire up the whole tileloop as one thread per tile, and write changes to the original map. after joining all threads, remove the copy, and redirect read access back to the original map
04:27<Celestar>Eddi|zuHause: the read access isn't the problem.
04:28<Celestar>this is:
04:28<Celestar>you have, at one place, a MP_CLEAR and an MP_TREE.
04:28<Celestar>MP_CLEAR has a .. bit that sais that another tile follows.
04:28<TrueBrain>planetmaker: fixed; tracd decided to segfault over and over and over and over again :) He likes segfaulting :)
04:28<Celestar>if MP_TREE is removed, MP_CLEAR needs to be changed.
04:29<Eddi|zuHause>Celestar: yes, but if you have a locked copy of the original map, every other tile doesn't care if these two tiles change.
04:29<Celestar>Eddi|zuHause: the question is: how long does a memcpy of the entire map take?
04:30<Celestar>if this is more than a maybe two-digit figure of microseconds, it will not help.
04:30<Eddi|zuHause>Celestar: all tiles other than the one processed currently have the state from _before_ the tile loop was run
04:30-!-TWerkhoven [] has joined #openttd
04:31<Eddi|zuHause>i have no clue about nanoseconds
04:31<Celestar><Sheldon>a nanosecond is 1e-9 of a second </Sheldon>
04:32<@planetmaker>Celestar, call it MP_VOID then
04:32<@planetmaker>and leave it in place
04:33<Celestar>and then have a cleanup loop run less often that does GC?
04:34<@planetmaker>it might also just be re-populated
04:34<@planetmaker>so the "surface" layer might not need deletion at all
04:34<@planetmaker>it's not like we are short on memory really
04:38<Celestar>first question is ... how much does it gain :P
04:39<Eddi|zuHause>that's something that cannot possibly be answered from an abstract point of view :p
04:39<Celestar>hence I'm coding it :P
04:40<Eddi|zuHause>the same approach might be taken by pathfinding, but i think that needs some further refactoring first
04:40<@planetmaker>the copying the whole map is expensive.
04:42<Eddi|zuHause>but you can't copy only parts of the map, because the tile loop really touches all of the map at once
04:46<@planetmaker>well 1/256 at once
04:47<Eddi|zuHause>yes. spread like cancer throughout the whole map
04:47<@planetmaker>yes. Actually there was recently a patch which made that a) random and b) faster
04:47<@planetmaker>random as in not regular
04:47<@planetmaker>not really random
04:49<@peter1138>morning again
04:49<@Terkhen>hi peter1138
04:56<@planetmaker>salut peter1138
05:02-!-Neon [] has quit [Quit: Python is way too complicated... I prefer doing it quickly in C.]
05:09-!-DDR_ [~chatzilla@] has quit [Quit: ChatZilla 0.9.87 [Firefox 7.0.1/20111008085652]]
05:11-!-blotek___ [] has joined #openttd
05:16<Celestar>hacking threads into this is ... hacky :P
05:18-!-pjpe [] has quit [Quit: ajax IRC Client]
05:21<Eddi|zuHause>which is why nobody has done it yet :p
05:22*Celestar starts over :P
05:28<Eddi|zuHause>"Equivalent cargo labels" <-- now where do i remember that from? :p
05:46-!-hanf [] has joined #openttd
05:46-!-DayDreamer [] has joined #openttd
05:50-!-andythenorth [] has joined #openttd
05:52<@peter1138>i'm so hacky, hacky hacky hacky...
05:52<Eddi|zuHause>you are so old, old old old...
05:56<MNIM>hacky wacky jacky!
05:59*andythenorth seems to have high boredom threshold :P
06:03<andythenorth>I find myself agreeing almost 100% with MB
06:04<@Terkhen>so... this week long discussion is nearing a conclusion? :P
06:04<andythenorth>it needs at least one decision to be made
06:04<andythenorth>wrt explicit exclusion prop for labels
06:04<andythenorth>I don't want it
06:05<andythenorth>but if we only did what I want, it would be a poor world :P
06:05<andythenorth>my opinion on that prop has developed as far as 'meh' and 'shrug'
06:05<andythenorth>as long as the XOR goes away and we get a list of included props as indices to CTT, I'm ecstatic
06:06-!-KenjiE20 [] has joined #openttd
06:07-!-KenjiE20 [] has quit [Remote host closed the connection]
06:08-!-KenjiE20 [] has joined #openttd
06:08-!-|Jeroen| [] has quit [Quit: oO]
06:11<@peter1138>i still don't get Eddi|zuHause's objects
06:12<@Yexo>wasn't it just about the amount of work?
06:13<@peter1138>i still don't get Eddi|zuHause's objection
06:13<@peter1138>yes :)
06:13<andythenorth>he wants to be able to use classes
06:13<@Yexo>your set contains 300 engines. You learn of a new cargo type ABCD and want to add graphics support for that in one wagon.
06:13<andythenorth>explicitly excluding things in the CTT makes work
06:13<andythenorth>he might be right
06:13<andythenorth>it's not elegant though :P
06:13<@Yexo>Now you have to add it to the CTT, which in turns means that for every wagon you either add it explicitely to the "include list" or don't include it
06:14<andythenorth>he should create an internal define in his set, which composes a set of n labels
06:14<andythenorth>'groups' which are private to his set
06:14<andythenorth>does the world end if the same index appears twice in the list for hte refit prop
06:15<@peter1138>you'll need computers able to store 2 in a single bit ;)
06:15*andythenorth doesn't want ottd to cause end of world
06:15<jonty-comp>quantum openttd!
06:16<TrueBrain>2012 takes care of that, no worries
06:16<andythenorth>I hate the exclude property, but if it gets this issue moved to a conclusion, there would be happy people
06:16<jonty-comp>professor bigglesworth!
06:16<@peter1138>doing the exclude property is actually a bit easier to cod e;)
06:17<andythenorth>do that then :P
06:17<@peter1138>does Eddi|zuHause use raw nfo, preprocesser nfo, or nml?
06:17<@peter1138>andythenorth, i don't want it though ;p
06:17<andythenorth>I'll define private groups in my sets for excludes :P
06:17<Eddi|zuHause>worse. a google spreadsheet
06:17<andythenorth>find a way to make it optional
06:17<Eddi|zuHause>and a python script
06:18<Eddi|zuHause>that is then generating preprocessed nml
06:18<@peter1138>shouldn't be much hassle then :p
06:29-!-Elukka [] has joined #openttd
06:35<Celestar>bah :P
06:36<Celestar>stupid friggen threading :P
06:38<Celestar>wtb [new brain
06:39<Celestar>maybe I should first not use our implementation
06:39-!-hanf [] has quit [Read error: Connection reset by peer]
06:39<Celestar>but try it with pthread directly
06:45<Eddi|zuHause>peter1138: the point is that adding graphics to one wagon needs changes in totally unrelated places
06:46-!-MNIM [] has quit [Remote host closed the connection]
06:49<andythenorth>Eddi|zuHause: but you've changed the context of your set
06:49<andythenorth>your set has known and unknown cargos, and by adding one, you change that
06:50<andythenorth>so now you need to tell your set what to do with that
06:50<Eddi|zuHause>andythenorth: it's still bad from a coding point of view
06:50<andythenorth>writing code isn't bad
06:50<andythenorth>letting possible errors pass silently is bad
06:51<andythenorth>it's bad that when you add a parameter to a function, you might have to update every single place that's called
06:51<andythenorth>why not just auto-magic that?
06:51<andythenorth>but it's not how it's done
06:51<Eddi|zuHause>i can add optional parameters to a function, and only use it in places that actually need it
06:52<andythenorth>I tried to figure out a way to make this optional
06:52<@Yexo>andythenorth: making it optional is easy. Just go back to the two properties, one include list and one exclude list
06:52<Eddi|zuHause>that's exactly what i said
06:53<andythenorth>which is why I think Eddi|zuHause might be right
06:53<andythenorth>see above somewhere ;)
06:53<@Yexo>one advantage of the single property is that as soon as everybody starts using it changing cargoclasses for known cargoes doesn't break those vehicle sets
06:54<andythenorth>Yexo: it doesn't 'break' them in either case
06:54<Eddi|zuHause>which we can't do anyway, because of backwards compatibility
06:54<andythenorth>but in one case the spec is a huge mess
06:54<andythenorth>and in the other it's explicit
06:54<@Yexo>Eddi|zuHause: that's theory, you can't influence everyone not to do it
06:54<@Yexo>ie even george changed cargoclasses of some cargos in ECS not too long ago
06:54<andythenorth>MB made class change requests
06:54<andythenorth>no-one is immune
06:54<Eddi|zuHause>"Dec 31, 2010"
06:55<andythenorth>with the exclusion property, here is my problem
06:55<andythenorth>I have cargo 'foo'
06:55<Eddi|zuHause>but changing newgrf cargos is less problematic than changing original cargos
06:55<andythenorth>it's not in my include or exclude list. Does my vehicle carry 'foo' ?
06:55*Celestar is attempting to understand Foundations :P
06:55<@Yexo>andythenorth: depends on cargoclasses
06:55<andythenorth>I have no fricking idea if my vehicle carries 'foo', but it's in my CTT
06:56<@Yexo>you can't plan for that, since you didn't know about foo when you created your set
06:56<andythenorth>but it's in my CTT - so I know about it
06:56<@Yexo>than you should hav added it to either the include or exlucde list if you wanted to be sure
06:56<Elukka>that is a pretty amazing station
06:56<andythenorth>but that's not what Eddi|zuHause wants to do
06:56<Eddi|zuHause>andythenorth: imagine a container wagon. why should it know whether it carries "foo", when the open wagon displays special graphics for "foo"?
06:56<andythenorth>Eddi|zuHause wants to be sure on the basis of classes - unless I misrepresent him
06:56<@Yexo>it all boils down to: will cargoclasses get changed again, and do we care if that means some vehicles temporary transport some "wrong" cargoes
06:57<Eddi|zuHause>anyway, gtg
06:57<andythenorth>Eddi|zuHause: it shouldn't know. But vehicle set authors think it should :P
06:58<@Yexo>andythenorth: that was not the point. If you want to make sure whether your vehicle can refit foo you can do that either way (just include list or both include and exclude list). With just an include list you're forced to let the vehicle know whether it can refit foo
06:58<Celestar>do we have some link to decoded foundation sprites?
06:58<andythenorth>the single prop is the 'correct' route imho
06:58<andythenorth>but it doesn't meet the needs of vehicle set authors who want to avoid writing code
06:59<@Yexo><Elukka> that is a pretty amazing station <- Celestar somewhere in that repo :p
06:59<Eddi|zuHause>Celestar: in the opengfx repo?
06:59<@Yexo> <- ehm, there
06:59<Celestar>Yexo: rofl :P
07:00<andythenorth>Yexo: exclude property: +1 or -1 ?
07:00<@planetmaker>Celestar, OpenGFX has them decoded
07:00<Celestar>that's a shitload of parameters you need to store the (slope+foundation) in a tile.
07:00<@Yexo>no big opinion either way
07:00<andythenorth>me neither
07:01-!-sla_ro|master [~slaco@] has quit []
07:02<andythenorth>I guess it gets decided by whoever turns up with working code
07:02<Elukka>are those passenger coaches refit for cargo in the upper left O_o
07:07<Eddi|zuHause>never seen those before
07:07<Eddi|zuHause>but could be repainted old mail/baggage cars
07:09<Elukka>something like that probably
07:10<Elukka>i've never even seen the models before
07:10<Elukka>and that talgo trainset there is weird as hell, but that's talgo for you
07:11-!-Elukka [] has quit [Read error: Connection reset by peer]
07:11-!-Elukka [] has joined #openttd
07:16-!-Biolunar [] has joined #openttd
07:36<Celestar>aren't those ICE cars a short?
07:36<Celestar>they don't look like 25m to me
07:37<@planetmaker>indeed, they look short compared to the heads
07:46-!-arie- [] has joined #openttd
07:46<arie->I've got a possible bug, but if I remember correctly this one has been mentioned before:
07:47<arie->strange track reservations: there's a possible path for a train but it just won't take it
07:47<arie->possibly happened during track reconstructions
07:47<arie->I think I've seen the issue before on the forums but cannot find it
07:48<arie-> this one maybe
07:48<arie->I'll add a post to that one with a save game
07:50<Celestar>now this is a "user interface"
07:50<@peter1138>needs tooltips
07:51-!-DayDreamer [] has quit [Quit: Leaving.]
07:52<andythenorth>needs a bigger GUI :P
07:52<Celestar>lol yeah
07:52<Celestar>storing all 4 corners of a tile is not sufficient to define the configuration of the tile.
07:52<andythenorth>it should be made in squirrel so anyone can remake it as they see fit
07:53<Celestar>peter1138: for example, half-tile-slopes.
07:53<arie->Ok, got two autosaves, two minutes apart.
07:53<Celestar>peter1138: those even give a headache what to store in two of the corners.
07:53<@peter1138>magic voodoo
07:53<Celestar>because it's undefined.
07:53<@peter1138>slope with rail on top :S
07:54<@peter1138>are you storing it as a slope value, i.e. in 5 bits?
07:54<Celestar>or just store the north corner and define the "half tile slope" as separate Slope value
07:54<Celestar>peter1138: yeah, but currently I'm using 8 bits for simplicity.
07:55<@peter1138>that's where i'm going
07:55<Celestar>"great minds think alike" ? :P
07:55<@peter1138>can you have half-tile stuff on steep slopes though?
07:55<Celestar>even two half-tiles :P
07:55<Celestar>but theoretically, with cliffs ....
07:56<Celestar>there is no clear definition how many levels those two half-tiles are apart, is there?
07:56<@peter1138>ooh, a graphical bug :D
07:56<Celestar>I'm not sure it's only graphical tbh :P
07:57<@peter1138>no, i mean in trunk :)
07:57<Celestar>oh :d
07:59<@peter1138>i suspect it doesn't matter though ;)
08:00<Celestar>wtf is that from
08:01<Celestar>well anyway.
08:01<Celestar>that line.
08:01<@peter1138>it's the lower foundation creeping through
08:01<Celestar>oh right :D
08:02<@peter1138>possibly the wrong sprite selected :)
08:02<@peter1138>i don't remember everything about all the foundation sprites though :)
08:02<Celestar>neither do I
08:03<Celestar>logically, the tile consists of 4 parts.
08:03<Celestar>4 triangles, with two corners in adjacent corners of the tile, and the third corner in the centre of the tile
08:08<Celestar>with those 4, it would be easy.
08:08<Celestar>... theoretically :P
08:12<Celestar>halftiles are a pita
08:22-!-Fuco [] has quit [Ping timeout: 480 seconds]
08:31-!-Fuco [] has joined #openttd
08:34<Celestar>bah dis sucks.
08:34*Celestar has to rethink terraforming completely
08:42<Celestar>if you wanna be able to terraform cliffs/foundations, the GUI needs some changing :P
08:44-!-andythenorth is now known as Guest16582
08:44-!-Guest16582 [] has quit [Read error: Connection reset by peer]
08:44-!-andythenorth [] has joined #openttd
08:55-!-Neon [] has joined #openttd
08:55<Celestar>$ ls -alh bin/openttd
08:55<Celestar>-rwxrwxr-x 1 vici vici 65M 2011-11-10 14:55 bin/openttd
08:56-!-Bluelight [] has joined #openttd
09:01<@peter1138>-rwxr-xr-x 1 petern petern 41M Nov 10 11:48 bin/openttd
09:01<@peter1138>big but not that big
09:01<@peter1138>debug & profiling
09:02<SpComb>lzma loader :)
09:05-!-Bluelight [] has quit [Quit: ChatZilla 0.9.87 [Firefox 7.0.1/20110928134238]]
09:09-!-TWerkhoven [] has quit [Read error: Connection reset by peer]
09:10-!-TWerkhoven [] has joined #openttd
09:10-!-Tintinfan [] has joined #openttd
09:12<Celestar>cliffs are hell.
09:13<Noldo>you are not going to like where that leads
09:14<@peter1138>looks cool ;)
09:15<Celestar>peter1138: but the Z drawing order is fucked.
09:16<Celestar>plus the engines can jump over cliffs :D
09:16<Noldo>well naturally
09:17<Celestar>reload the link
09:17<Celestar>and open cliff2 :)
09:19<Celestar>plus .. er I lack a user interface :P I just manually modded the map array :D
09:19<@peter1138>is it?
09:19<Celestar>peter1138: ?
09:19<@peter1138>z order fucked?
09:19<Celestar>peter1138: the wagons should be partially hidden
09:20<@peter1138>that was never a problem :)
09:20<Celestar>no it wasn't :D
09:20<@peter1138>yeah, vehicles are always drawn after the landscape. hmm.
09:20<@peter1138>at least, i think
09:20<Celestar>how is it done with trees?
09:21<Celestar>because trees can hide vehicles
09:21<@peter1138>landscape then all objects
09:21<Celestar>so for cliffs to be any useful, that has to be changed .....
09:21*Belugas cleans his glasses and stares again at Celestar's cliffs, speahless
09:22<@peter1138>i suspect that adding landscape to the sortable list would be another performance hit :(
09:22<Celestar>I guess so as well.
09:23<Celestar>Belugas: don't get too excited, i still have no UI to make them :P
09:23<andythenorth>cliffs :D
09:23<andythenorth>just make bridges with filled in sides?
09:23<Celestar>bridges .. don't exist
09:24<Celestar>peter1138: the question is .. how much of a performance hit ...
09:24<andythenorth>draw tunnel sprites if another route passes under your 'cliffs' bridge
09:24*Celestar pokes andythenorth with a rusty pole :P
09:24*andythenorth is a big fan of doing things wrong
09:25<andythenorth>except cargo refits
09:25<@Belugas>Celestar, who cares, it's the possiblity that such a feature can be done, no matter how it will be done in a gui (which , in fact, will never win approval of all players...)
09:25<Celestar>peter1138: doesn't GTTS return whether a vehicle can actually enter a tile?
09:25<Celestar>Belugas: true.
09:25<@peter1138>Celestar, allegedly
09:25<Celestar>Belugas: but this can be yet another option to disable.
09:25<Celestar>peter1138: well.
09:25<@peter1138>Celestar, you'll want to add a z parameter to it
09:25<Celestar>peter1138: next problem. This needs to be z-aware :P
09:26<@peter1138>i think i may have done that once
09:26<@peter1138>when i was working on my original bridges over stuff patch
09:26<Celestar>me too :P
09:26<Celestar>for ... some patch I cannot remember
09:26-!-andythenorth [] has left #openttd []
09:26<@Belugas>the initial newmap, maybe?
09:26-!-andythenorth [] has joined #openttd
09:27<Celestar>anyway I think that patch is 1) now useless, 2) lost on some of my backup disks at mom's :P
09:28<Celestar>can't build a tunnel inside a cliff either :P
09:29<Celestar>because there isn't a "tile" to end it :D
09:32<@peter1138>Celestar, viewport.cpp:23057
09:32<@peter1138>+ AddSortableSpriteToDraw(image, pal, x, y, 16, 16, 0, z, false, extra_offs_x, extra_offs_y, 0, sub);
09:32<@peter1138>+ return;
09:32<@peter1138>something like that :p
09:32<@peter1138>i read the revision, not the line :p
09:32<Celestar>rofl :D
09:33<Celestar>I can't do that yet, because saveload will fuck the cliff.
09:33<@peter1138>it's in AddTileSpriteToDraw() basically
09:33<@peter1138>ok :)
09:33<@peter1138>^B is ... different with that
09:34<@peter1138>the world is filled with bounding boxes ;)
09:35<@peter1138>view bounding boxes
09:35<@peter1138>hmm, bounding boxes might be wrong for slopes...
09:36*Celestar thinks cliffs open not one, but several can-o-worms
09:39<Celestar>1) We need a new UI to change individual corners.
09:39<Celestar>2) Z Order problem in the viewport
09:39<Celestar>3) Missing Z information in GTTS
09:39<TinoDidriksen>...give in and just go 3D.
09:40<Celestar>TinoDidriksen: and that helps exactly how? :P
09:40<Celestar>I think 2) is a minimal problem other than performance
09:41<Celestar>but BBs might need adaption
09:42<@Belugas>3D.. the new buzz word... after XML ;)
09:42<SpComb>3D XML
09:42<SpComb>plaintext is too 2D
09:42<@Belugas>with hazarai
09:42<SpComb>instead, we store the XML nodes in a sparse voxel space
09:44<Celestar>also, I'm not sure how difficult add Z to the GTTS stuff is.
09:44<Celestar>wtb [grammar]
09:46<Celestar>how is it solved with bridges ....
09:46<@Yexo>as you said yourself, bridges don't exist
09:46<@Yexo>so vehicles on the bridge are on either end of the bridge
09:47*andythenorth remembers an idea about terraforming
09:47<Celestar>Yexo: nah I meant something else.
09:47<@Yexo>the drawing order? or what
09:48<andythenorth>There's only so much mud in the world. If you want more mud, you have to take it from somewhere. If you want less mud, you have to put it somewhere.
09:48<andythenorth>would make small maps interesting
09:48*andythenorth suspects that small maps would end up flat at height level 8 :P
09:49<Celestar>Yexo: same link as above, last file (Bridge)
09:50<@Yexo>the vehicle suddenly appears at the bridge ramp?
09:50<Celestar>no it doesn't
09:50<Celestar>I'm wondering why
09:50<Celestar>apparently it's not Z based :P
09:51<Celestar>it goes purely by direction
09:51<@Yexo>see GetTileTrackStatus_TunnelBridge
09:51<@Yexo>you can only enter the tile from one direction
09:52<Celestar>I was faster ^ ^
09:52-!-glx [glx@2a01:e35:2f59:c7c0:e3:6096:592c:9d52] has joined #openttd
09:52-!-mode/#openttd [+v glx] by ChanServ
09:52<Celestar>and I think I know where I last try to use z awareness....
09:52<Celestar>when doing cbh :P
10:04<Celestar>bbl :)
10:04-!-Celestar [~dax@] has quit [Quit: leaving]
10:08<@peter1138>yeah, sometimes it just knows :p
10:14<z-MaTRiX>hi :)
10:15<z-MaTRiX>wanted to try out haskell programming, but i got demotivated when the library install 'cabal' thing attempted to count the spaces in the config file before the options (unsuccessfully)
10:33-!-supermop [] has joined #openttd
11:02<andythenorth>it's gone all quiet :o
11:03<z-MaTRiX>so why does openttd not have an fps meter?
11:03<@Yexo>because it's pointless?
11:03<z-MaTRiX>rendering time/processing time statistics?M
11:04<z-MaTRiX>sure its statistical...
11:04<z-MaTRiX>or might tell you how slow is your pc for Openttd
11:04<@Yexo>well, same answer as always: nobody has been interested enough to code it
11:05<z-MaTRiX>aham i see :)
11:05<__ln___>seriously, what purpose would such a meter serve?
11:05<z-MaTRiX>__ln___<< informational
11:06<z-MaTRiX>also you could check your overall performance if you upgrade some functions
11:07<@Yexo>that is way too unreliable if you actually use a gui (which you need to see such a fps meter)
11:07<@Yexo>we have profiling and the null blitter for that
11:08<z-MaTRiX>Yexo<< sure i know SDL forces a waitvretrace, but it is still possible to check the processing time, and rendering time seperately ;)
11:08<@Yexo>that is compeltely besides the point
11:09<z-MaTRiX>btw im making friends with mutexes right now to be able to update screen asnychronously
11:09<+glx>we just disable display when profinling
11:09<z-MaTRiX>hmm well thats a solution for testing
11:19-!-Strid_ [] has quit [Ping timeout: 480 seconds]
11:21<andythenorth>Eddi|zuHause: have you got classes all figured out yet? :)
11:35-!-Tintinfan [] has left #openttd []
11:40-!-TGYoshi [~TGYoshi@] has joined #openttd
11:41-!-hanf [] has joined #openttd
11:42-!-sla_ro|master [~slaco@] has joined #openttd
12:05-!-SpBot [] has joined #openttd
12:06-!-SpBot_ [] has quit [Remote host closed the connection]
12:06-!-valhallasw [~valhallas@] has joined #openttd
12:20-!-andythenorth is now known as Guest16606
12:20-!-Guest16606 [] has quit [Read error: Connection reset by peer]
12:20-!-andythenorth [] has joined #openttd
12:23-!-andythenorth is now known as Guest16608
12:23-!-Guest16608 [] has quit [Read error: Connection reset by peer]
12:23-!-andythenorth [] has joined #openttd
12:29-!-Strid_ [] has joined #openttd
12:48-!-Elu [] has joined #openttd
12:52-!-Brianetta [] has quit [Quit: Tschüß]
12:53-!-Elukka [] has quit [Ping timeout: 480 seconds]
12:56<CIA-6>OpenTTD: rubidium * r23179 /trunk/src/ (3 files in 2 dirs): -Codechange: use some tooltips that already existed (monoid)
12:58<CIA-6>OpenTTD: rubidium * r23180 /trunk/src/lang/english.txt: -Cleanup: remove traces of having to double click on the NewGRF for changing the parameters
13:00<CIA-6>OpenTTD: rubidium * r23181 /trunk/src/lang/english.txt: -Cleanup: remove some unused strings (monoid)
13:02-!-LordAro [] has joined #openttd
13:02<@Terkhen>hi LordAro
13:04<LordAro>@commit 23178
13:04<@DorpsGek>LordAro: Commit by rubidium :: r23178 /trunk/src (5 files in 2 dirs) (2011-11-10 06:15:03 UTC)
13:04<@DorpsGek>LordAro: -Feature [FS#4780]: in-game readme.txt readmer (LordAro)
13:06<@Terkhen>what are you going to code now? ;)
13:06<Eddi|zuHause><Celestar> aren't those ICE cars a short? <-- they almost fit into TTD scale then :p
13:06*Terkhen hopes for something way more scary
13:07<CIA-6>OpenTTD: yexo * r23182 /trunk/src/newgrf_config.cpp: -Feature: allow translatable readme files
13:07<LordAro>i dunno, i had some ideas while making the readme viewer, but i've forgotten them :L
13:08-!-DDR_ [~chatzilla@] has joined #openttd
13:09<@Terkhen>it's late for making them translatable I guess
13:09<CIA-6>OpenTTD: rubidium * r23183 /trunk/src/ (lang/english.txt town_cmd.cpp): -Codechange: merge BRIBE_FAILED and BRIBE_FAILED_2 messages (monoid)
13:09<CIA-6>OpenTTD: rubidium * r23184 /trunk/src/lang/ (59 files in 2 dirs): -Cleanup: remove the removed strings from the translations as well
13:11<LordAro>i think i'll probably have a go at readme viewers for AIs and base sets, etc
13:11<LordAro>should be easier :)
13:12-!-supermop [] has quit [Quit: supermop]
13:12<Eddi|zuHause>CIA-6> OpenTTD: rubidium * r23180 /trunk/src/lang/english.txt: -Cleanup: remove traces of having to double click on the NewGRF for changing the parameters <-- i really would like this behaviour back...
13:13<@Yexo>currently double-click moves a grf from active<>inactive, right?
13:13<@Yexo>I don't see how that is compatible with "open parameter window on double-click"
13:13<Eddi|zuHause>a GUI setting
13:14<LordAro>yay! moar settings :)
13:14<Rubidium>okay... I should make more cryptic messages
13:25<andythenorth>LordAro: \o/
13:26<LordAro>ty, although i would estimate that 60%+ of the work was done by Alberth and Rubidium though
13:27<LordAro>but i motivated them to do it! :)
13:27<Rubidium>well, then you've done the remaining 120% ;)
13:31-!-andythenorth [] has quit [Quit: andythenorth]
13:32*LordAro only just got that :)
13:32<LordAro>90%9-90% rule, yes?
13:35<@Yexo>running 14 AIs makes openttd really slow
13:35<SpComb>on a 14-core cpu?
13:35<@Yexo>that doesn't matter
13:35<@Yexo>all AIs run on the same core
13:36-!-TGYoshi [~TGYoshi@] has quit [Quit: Poof]
13:37-!-arie- [] has quit [Quit: ajax IRC Client]
13:38<@Terkhen>something to fix :D
13:39-!-blotek___ [] has quit [Remote host closed the connection]
13:39<LordAro>umm... "Warning. Your computer is too useless to run with 14 AIs. Please give up now"
13:40<@Yexo>main problem is lack of multicore support
13:41<@Terkhen>yes, I meant multicore support for running AIs in parallel :P
13:42<LordAro>yes. bit advanced for me... :)
13:42<LordAro>see above solution :)
13:44<@Terkhen>ok, as long as you use all 14 cores for displaying the message
13:45<CIA-6>OpenTTD: translators * r23185 /trunk/src/lang/ (7 files in 2 dirs): (log message trimmed)
13:45<CIA-6>OpenTTD: -Update from WebTranslator v3.0:
13:45<CIA-6>OpenTTD: belarusian - 23 changes by Wowanxm
13:45<CIA-6>OpenTTD: english_US - 3 changes by Rubidium
13:45<CIA-6>OpenTTD: german - 24 changes by planetmaker
13:45<CIA-6>OpenTTD: spanish - 34 changes by Terkhen
13:45<CIA-6>OpenTTD: swedish - 45 changes by Zuu
13:47-!-TheMask96 [] has quit [Ping timeout: 480 seconds]
13:50-!-TheMask96 [] has joined #openttd
13:51-!-DayDreamer [] has joined #openttd
14:06-!-Alberth [] has joined #openttd
14:06-!-mode/#openttd [+o Alberth] by ChanServ
14:07-!-Devroush [] has joined #openttd
14:30-!-hanf [] has quit [Read error: Connection reset by peer]
14:39-!-Hyronymus [] has joined #openttd
14:40<CIA-6>OpenTTD: yexo * r23186 /trunk/src/3rdparty/squirrel/squirrel/sqbaselib.cpp: -Fix [FS#4830]: [Squirrel] replace custom qsort by std::sort to fix stack overflow
14:41*Rubidium ponders the irony in those emails "asking" to consider the environment by not printing it. Does one really think that it doesn't get printed that way? I'd argue you'd be printing even more (i.e. the request not to do it) and you'd be wasting a lot of energy by adding, transporting, processing and reading it
14:41-!-lobstaroooo [~michielbr@] has joined #openttd
14:42<TyrHeimdall>Rubidium: don't forget the added electricity needed for the transfer of said bulk of text :P
14:42<Rubidium>that's the transporting and processing I'm talking about ;0
14:42-!-Cybertinus [] has quit [Ping timeout: 480 seconds]
14:47*Eddi|zuHause never got such an email
14:48-!-TWerkhoven[l] [] has joined #openttd
14:48-!-TGYoshi [~TGYoshi@] has joined #openttd
14:49-!-pugi [] has joined #openttd
14:54-!-supermop [] has joined #openttd
15:03-!-Zuu [] has joined #openttd
15:04-!-Cybertinus [] has joined #openttd
15:07<CIA-6>OpenTTD: yexo * r23187 /trunk/src/3rdparty/squirrel/squirrel/sqbaselib.cpp: -Fix (r23186): MSVC allowed non-const where const was mandatory
15:07-!-JVassie [~James@] has joined #openttd
15:16-!-Sacro [~ben@] has quit [Quit: leaving]
15:17-!-Sacro [~ben@] has joined #openttd
15:19-!-brundlefly [] has quit [Remote host closed the connection]
15:20<Zuu>LordAro: Have you any plans to generalize the NewGRF readme reader to a Readme Reader for other content types that also use tar files?
15:23-!-Mucht [] has joined #openttd
15:26-!-KritiK [] has joined #openttd
15:33<@Yexo><LordAro> i think i'll probably have a go at readme viewers for AIs and base sets, etc <- yes
15:34<Zuu>Good to hear
15:35<Zuu>I've just submitted Swedish translations for the NewGRF readme feature :-)
15:36<Zuu>I kind of thinks that "NewGRF" in "NewGRF readme of {STRING}" could be omitted.
15:37<@Alberth>there are also 'NewGRF parameters' and 'NewGRF Settings' :)
15:37-!-Hyronymus [] has quit [Remote host closed the connection]
15:38-!-Hyronymus [] has joined #openttd
15:39<@Yexo>but those don't have the grf name in the title
15:40-!-ptr_ [] has joined #openttd
15:50<Zuu>Also you need to click first on NewGRF settings to be able to view the readme.
15:59*peter1138 idly ponders the purpose os 30_PaintingVoidTiles.patch
15:59<@peter1138>(more height levels)
16:00<@peter1138>why is there geometry outside of the map... :S
16:05-!-LordAro [] has quit [Quit: ajax IRC Client]
16:05-!-LordAro- [] has joined #openttd
16:06-!-LordAro- is now known as LordAro
16:07<Eddi|zuHause>peter1138: afair that has something to do with the "outside" needing to be black. with the current method, that doesn't extend well beyond height level 16
16:08<@peter1138>i don't see the relevance :S
16:08<LordAro>hai Alberth
16:08<@Alberth>hi LordAro
16:08<LordAro>i think congratulations are in order for the both of us :)
16:09<LordAro>(but mostly you :) )
16:09<@Alberth>I was somewhat surprised tbh :)
16:09<LordAro>likewise :)
16:10<Eddi|zuHause>peter1138: if the "void" tile is painted flat, it must be painted at level 0 and repeated for each height level, up to the level of the map border
16:11<Eddi|zuHause>peter1138: if you have sloped black tiles, you can just draw them like normal tiles
16:14-!-TWerkhoven2 [] has joined #openttd
16:16-!-andythenorth [] has joined #openttd
16:17-!-DDR_ [~chatzilla@] has quit [Ping timeout: 480 seconds]
16:18<LordAro>greets andy
16:20<@peter1138>Eddi|zuHause, turn it into a cliff :)
16:20<andythenorth>Eddi|zuHause: run classes off a cliff
16:20<Eddi|zuHause>peter1138: that would be very simcity-y
16:21<andythenorth>seems I upset wallyweb :(
16:21<andythenorth>who else can I upset :P
16:21<@peter1138>Eddi|zuHause, also wouldn't solve anything
16:21<@peter1138>because that's what it does currently
16:21<b_jonas>simcity isn't the only game doing that
16:21-!-TWerkhoven [] has quit [Ping timeout: 480 seconds]
16:21<@peter1138>except it just draws the shadow sprite
16:21<@peter1138>i guess drawing up to 255 of those shadow sprites gets a bit... repetitive
16:21-!-TWerkhoven[l] [] has quit [Read error: Connection reset by peer]
16:23<@peter1138>disallow freeform edges, that'll solve it
16:24-!-TWerkhoven[l] [] has joined #openttd
16:26-!-DDR_ [~chatzilla@] has joined #openttd
16:29<@peter1138>with ez
16:29<@peter1138>i can't see how to make scaled-up tile edges match native edges :S
16:31<andythenorth>Eddi|zuHause: how would you feel about abandoning cargo classes and substituting wagon classes?
16:31<andythenorth>I had some pm chat with MB
16:33<@peter1138>vehicle classes :S
16:33<@peter1138>scale2x is not right :S
16:35<andythenorth>I'm only continuing to pursue classes because I get kicked in the teeth every time they're changed for a FIRS cargo, and I want to get it sorted out :P
16:35<Eddi|zuHause>where was that ancient double mode filter patch...
16:35<andythenorth>from a vehicle set point of view, I'm ecstatic if we lose the XOR on labels + gain a list instead of bitmap
16:36-!-Kurimus [] has quit []
16:37<andythenorth>"NO you may not change classes, it breaks sets"
16:38<andythenorth>"NO you may not change labels, it breaks sets!"
16:39<Eddi|zuHause>i think that last one is rubbish
16:40<@peter1138>well yes, simple pixel doubling looks great
16:40<Eddi|zuHause>it may cause a few inconsistencies, but it's not "breaking"
16:40<andythenorth>nah, I'm exagerating
16:40<@peter1138>but the still has the issue of mixing highres and original sprites ;p
16:40<andythenorth>pixel doubling? :o
16:41<andythenorth>that would be useful now my eyes are getting old
16:41<Eddi|zuHause>peter1138: there was a patch that had several "filters" to get to double size, not only simple doubling
16:41<@peter1138>you didn't see ? heh
16:42<andythenorth>how cute
16:42<andythenorth>please don't ship that
16:42<andythenorth>it will reveal all my mistakes
16:44<Elu>i wish we could get an extra zoom level in trunk
16:44-!-Elu is now known as Elukka
16:44<Elukka>even with just the current graphics
16:44<Elukka>basically all good sprites look better at double size to me..
16:45<Elukka>because they're so damn small in the game by default
16:45<@peter1138>yeah, 2x looks nice
16:45<Elukka>that sure does
16:46<@peter1138>the 32bpp crowd (all 2 of them?) would probably go mad if it wasn't 4x ;)
16:47<andythenorth>4x also looks nice
16:47<andythenorth>they all look nice :)
16:47<andythenorth>what a lot of exciting stuff is getting done
16:47<Elukka>is that zoom thing something that we might one day see in trunk?
16:48<andythenorth>it's more exciting even than an episode of In The Night Garden :o
16:50<@peter1138>^ scale2x applied twice
16:50<@peter1138>looks mushy
16:50<@peter1138>road surface looks ok
16:51<andythenorth>pixels should be nearly-square
16:52<@peter1138>opengfx version
16:53<@peter1138>paint lines go weird :S
16:54<@peter1138>this is 8bpp, of course
16:54<@peter1138>opengfx looks nicer than ttd there
16:58<@peter1138>anyway, it has performance issues :(
16:58<Eddi|zuHause>the old patch applied some magic so that 30° lines get more emphasised
16:59<Eddi|zuHause>peter1138: apply scaling on a per-sprite basis, and put the high-res sprites into the sprite cache
16:59<@peter1138>Eddi|zuHause, it is
16:59-!-Neon [] has quit [Quit: Python is way too complicated... I prefer doing it quickly in C.]
17:01-!-LordAro [] has quit [Quit: ajax IRC Client]
17:02-!-Elukka [] has quit []
17:02-!-Elukka [] has joined #openttd
17:05-!-Hyronymus [] has quit [Remote host closed the connection]
17:17-!-Mucht [] has quit [Remote host closed the connection]
17:25-!-Alberth [] has left #openttd []
17:25-!-Zuu_ [] has joined #openttd
17:25<andythenorth>hi planetmaker
17:26<@planetmaker>hi andythenorth
17:29-!-DayDreamer [] has quit [Read error: Connection reset by peer]
17:29<andythenorth>my steel sprites suck
17:30<andythenorth>cba to fix them
17:32-!-Zuu [] has quit [Ping timeout: 480 seconds]
17:32<@peter1138>bah bah bah
17:35<@Terkhen>good night
17:35<andythenorth>tra la la
17:36<andythenorth>good night Terkhen
17:37<andythenorth>what larks
17:38<andythenorth>fixing lighting :P
17:38<@peter1138>way too much effort to draw sprites zoomed out :(
17:38-!-KritiK [] has quit [Quit: Leaving]
17:46-!-Prof_Frink [] has joined #openttd
17:54-!-lobstaro0o [~michielbr@] has joined #openttd
17:59-!-Cybertinus [] has quit [Remote host closed the connection]
18:00-!-lobstaroooo [~michielbr@] has quit [Ping timeout: 480 seconds]
18:00-!-Zuu_ [] has quit [Ping timeout: 480 seconds]
18:03<andythenorth>@calc 40 + 40 *0 +1
18:03<@DorpsGek>andythenorth: 41
18:09<Eddi|zuHause>even i could have calculated that...
18:10<V453000>:D nice one andy
18:11<TGYoshi>@calc 0/0
18:11<@DorpsGek>TGYoshi: Error: float division
18:13<__ln___>DorpsGek: float division what?
18:13<Mazur>Sleep well.
18:13<TGYoshi>And I thought 0/0 == 0
18:14<Mazur>0/0 can be anything.
18:14<andythenorth>Eddi|zuHause: I got that one wrong
18:14<__ln___>TGYoshi: it certainly isn't.
18:14<andythenorth>annoyingly my wife got it right
18:14<__ln___>Mazur: wrong.
18:14<Mazur>Just try with limits.
18:14<andythenorth>I never learnt proper order of operations before, in engineering maths we were taught that it's not a valid calculation if you don't include the brackets
18:14<TGYoshi>__ln___: Why not? Zero pieces of ??? devided over zero people, how much get each guy?
18:15<Eddi|zuHause>TGYoshi: 23
18:15<andythenorth>you can't divide by zero
18:15<andythenorth>it's not a thing
18:15<TGYoshi>Possible too :P
18:15<andythenorth>you get at best oo
18:15<TGYoshi>Err, 1/0 = infinite, for me
18:15<Eddi|zuHause>andythenorth: there's a whole branch of mathematics just about 0/9
18:15-!-MNIM [] has joined #openttd
18:15<Mazur>Of the denomimator > 0 you get something infinite, yes.
18:15<__ln___>TGYoshi: the question is not valid because there are zero people, so there's no one to divide to.
18:16<Eddi|zuHause>__ln___: if there are zero trees in the wood, and one falls down, does it make noise?
18:16<@planetmaker>TGYoshi: what's lim(x->0) x/(x^2) ?
18:16<@planetmaker>it's 0/0 in the limit. But is 0 the right answer?
18:17<@planetmaker>or lim(x->0) x/(2x) ?
18:17<TGYoshi>I think it depends on the situation the division happens
18:17<@planetmaker>there 0/0 = 0.5 is "right"
18:17<@planetmaker>it doesn't depend. It's never defined
18:18<__ln___>(it's defined on PowerPC)
18:18<@planetmaker>and if you define it, it's that: a definition which is individual to that very case you define it
18:18<andythenorth>philosophy before bed time :P
18:19<Eddi|zuHause>i have a feeling we have this discussion every week
18:19<TGYoshi>A discussion about 0/0?
18:19<andythenorth>we also discuss classes
18:19<@planetmaker>anyway. I was about to head off :-) g'night again
18:19<andythenorth>bye planetmaker
18:19<TGYoshi>Classes are discussed daily here >_>
18:23<__ln___>yes, this must be the second 0/0==0 discussion within two weeks.
18:25<TGYoshi>Maybe I should write (0/0) in every place any number can take place
18:26<Eddi|zuHause>you can draw an apple in every place where a number can apperar
18:26<__ln___>should there be something about 0/0 in the topic? ... on the other hand, that would make the matter on-topic
18:27-!-TWerkhoven[l] [] has quit [Remote host closed the connection]
18:27<Eddi|zuHause>just not a bitten one, you might run into legal troubles
18:27<TGYoshi>Why an apple? :P
18:28<Eddi|zuHause>why not an apple?
18:28<Eddi|zuHause>numbers are just completely random symbols
18:28-!-lobstaro0o [~michielbr@] has quit [Quit: COCKBUSTER SLEEP MODE]
18:30<TGYoshi>I´d prefer an orange instead
18:30-!-Brianetta [] has joined #openttd
18:34<Mazur>Here, have a bananananana.
18:37-!-TGYoshi [~TGYoshi@] has quit [Quit: Poof]
18:38<andythenorth>Eddi|zuHause: wrt to excludes - remind me - when you add a cargo to the CTT, why do you need to explicitly exclude it? You had a convincing case
18:38<andythenorth>but I forget
18:39<Eddi|zuHause>not sure what the question is
18:39-!-TWerkhoven2 [] has quit [Quit: He who can look into the future, has a brighter future to look into]
18:40<andythenorth>so if a prop is added for included refits using indices to CTT
18:40<andythenorth>you had compelling case to also add an exclude prop
18:40<andythenorth>again, indices to CTT
18:42<Eddi|zuHause>the reason was: when i add an entry to CTT, e.g. for one wagon to show separate graphics, then i shouldn't need to touch any other vehicle in the set
18:42<andythenorth>that is a compelling aim
18:42<andythenorth>what causes you to need to touch the others?
18:43<andythenorth>you want them to still be refittable to the new cargo?
18:43<andythenorth>I recall now
18:43<Eddi|zuHause>if you automatically exclude all entries of the CTT from refitting, i need to re-add it to every vehicle that previously decided based on classes
18:44-!-sla_ro|master [~slaco@] has quit []
18:46<andythenorth>and time for sleep for me
18:48-!-andythenorth [] has quit [Quit: andythenorth]
18:49-!-DOUK [] has quit [Ping timeout: 480 seconds]
18:52-!-devilsadvocate_ [~quassel@] has quit [Quit: - Chat comfortably. Anywhere.]
18:52-!-devilsadvocate_ [~quassel@] has joined #openttd
19:02-!-JVassie [~James@] has quit [Ping timeout: 480 seconds]
19:08-!-SpBot [] has quit [Ping timeout: 480 seconds]
19:09-!-Biolunar [] has quit [Quit: All your IRC are belong to us!]
19:10-!-enr1x_ [] has quit [Quit: going to work]
19:11-!-SpBot [] has joined #openttd
19:14-!-ptr_ [] has quit [Quit: Zzzzzz]
19:23<@Yexo>Eddi|zuHause: "that a vehicle set author has a way to ensure that every cargo, known or not." <- aren't you mssing something like "can be transported" at the end?
19:24-!-Devroush [] has quit []
19:38-!-Progman [] has quit [Remote host closed the connection]
19:41-!-supermop [] has quit [Quit: supermop]
19:48<@peter1138>23:43 < Eddi|zuHause> if you automatically exclude all entries of the CTT from refitting, i need to re-add it to every vehicle that previously decided based on classes
19:48<@peter1138>well no,
19:48<@peter1138>cos if the vehicles decided based on class, they wouldn't have the CTT refit property
19:48<@peter1138>so it would still be based on class
19:53<Eddi|zuHause>peter1138: but the vehicle might have needed the refit property for other cargos
19:53<Eddi|zuHause>peter1138: like the default cargos WOOD and STEL, which have "wrong" classes, so need to be overridden manually
20:00<@peter1138>bah, backups
20:11-!-Elukka [] has quit []
20:18-!-Brianetta [] has quit [Quit: Tschüß]
20:19<+michi_cc>peter1138: I fixed you EZ patch :)
20:19<+michi_cc>peter1138: on top of your ez2.patch
20:21<+michi_cc>No idea if the second hunk is needed or good, but I figured that this way leads to the same viewport area restriction.
20:28<z-MaTRiX>i've bought a god of knives ;>
20:28<z-MaTRiX>a ceramic knife
20:31<z-MaTRiX>cuts apple including seeds like butter
20:32-!-pugi [] has quit [Quit: I reject your reality and substitute my own]
21:06-!-Eddi|zuHause [] has quit [Remote host closed the connection]
21:32-!-DDR_ [~chatzilla@] has quit [Quit: ChatZilla 0.9.87 [Firefox 7.0.1/20111008085652]]
21:37-!-valhallasw [~valhallas@] has quit [Ping timeout: 480 seconds]
21:51-!-pjpe [] has joined #openttd
22:17-!-lobstaroooo [~michielbr@] has joined #openttd
22:26-!-rhaeder [] has joined #openttd
22:29-!-lobstaro0o [~michielbr@] has joined #openttd
22:31-!-rhaeder1 [] has quit [Ping timeout: 480 seconds]
22:34-!-lobstaroooo [~michielbr@] has quit [Ping timeout: 480 seconds]
22:35-!-lobstaroooo [~michielbr@] has joined #openttd
22:39-!-lobstaro0o [~michielbr@] has quit [Ping timeout: 480 seconds]
22:53-!-lobstaro0o [~michielbr@] has joined #openttd
22:57-!-Snail_ [] has joined #openttd
22:58<Snail_>hi there
22:59-!-lobstaroooo [~michielbr@] has quit [Ping timeout: 480 seconds]
23:08-!-glx [glx@2a01:e35:2f59:c7c0:e3:6096:592c:9d52] has quit [Quit: bye]
23:09-!-lobstaroooo [~michielbr@] has joined #openttd
23:15-!-lobstaro0o [~michielbr@] has quit [Ping timeout: 480 seconds]
23:15-!-lobstaro0o [~michielbr@] has joined #openttd
23:15<Snail_>hi there
23:19-!-lobstaroooo [~michielbr@] has quit [Ping timeout: 480 seconds]
23:55-!-lobstaroooo [~michielbr@] has joined #openttd
---Logclosed Fri Nov 11 00:00:13 2011