#gemrb@irc.freenode.net logs for 29 Nov 2011 (GMT)

Archive Today Yesterday Tomorrow
GemRB homepage


[00:07:03] --> lord_jeremy has joined #gemrb
[00:07:20] <lord_jeremy> brad_a: that Xcode file you gave me is missing a precompiled header
[00:07:43] <brad_a> jsut scrap it
[00:07:57] <lord_jeremy> did you merge your changes yet?
[00:08:06] <brad_a> no
[00:08:13] <brad_a> im quite busy with school work atm
[00:08:21] <lord_jeremy> think it makes sense to have a separate Xcode project for the Mac build?
[00:08:28] <brad_a> why?
[00:08:47] <brad_a> you can just remove the ios target if you dont want it
[00:08:58] <lord_jeremy> well your iOS target has a ton of specific dependancies
[00:09:06] <lord_jeremy> frameworks/libraries, etc
[00:09:12] <brad_a> so?
[00:09:29] <brad_a> like i said this is how i want it. you are welcome to remove tht target
[00:09:42] <lord_jeremy> ok
[00:44:31] <lord_jeremy> if there are any Mac builders that are familiar with the make process, the 'make install' script has a problem where if it is run as root the stuff that's put in ~/Library/Application Support is owned by root:wheel
[00:45:08] <lord_jeremy> it should be owned by the user that ran the install, despite the elevated privileges
[00:45:31] <lord_jeremy> unfortunately because it also installs a manpage and a couple things to /Applications it has to be run as root anywaty
[00:46:28] <brad_a> why are you running as root?
[00:46:50] <lord_jeremy> you can't run make install as a regular user
[00:47:02] <brad_a> use sudo?
[00:47:04] <lord_jeremy> because it tries to install the manpage to /usr/share/ and gemrb.app to /Applications
[00:47:08] <brad_a> thats what i do and have no problem
[00:47:27] <tomprince> Well, those paths should be configurable.
[00:47:30] <lord_jeremy> if you do 'sudo make install' then the GemRB folder and all of its contents in ~/Library/Application support will be owned by root
[00:47:32] <lord_jeremy> they're not
[00:47:34] <brad_a> i should probably jsut remove the man page alltogether
[00:47:38] <lord_jeremy> I checked the cmake file
[00:48:14] <lord_jeremy> well ideally there should be a readme beyond the manpage
[00:48:46] <lord_jeremy> the install script also put extend2da.py in /Applications
[00:49:04] <lord_jeremy> if the user doesn
[00:49:04] <lord_jeremy> '
[00:49:09] <brad_a> well the permissions are fixed easy enough. im guessing the remain readable by the user since i havent ran into any permission denied issues
[00:49:26] <lord_jeremy> they're readable but not writable
[00:49:30] <brad_a> right
[00:50:16] <lord_jeremy> if the user doesn't have write access to /Applications (which by default no non-system user does) then it should put the binary in ~/Applications
[00:50:40] <brad_a> yes i have a rather long list of things that the mac install should do
[00:50:45] <lord_jeremy> yay
[00:51:01] <brad_a> I wasnt aware of the permissions tho
[00:51:03] <lord_jeremy> I've got a passing knowledge of make so I was able to change my install locations
[00:51:17] <lord_jeremy> but I don't think I could add the necessary stuff to the configure process
[00:51:37] <lord_jeremy> it'd be nice if the Application Support target directory was configurable too
[00:51:38] <brad_a> the thinng is that untill i did this there was no mac install at all
[00:51:49] <lord_jeremy> well in that case <3
[00:51:49] <brad_a> gemrb used to do a unix install on mac which was terrible
[00:52:05] <lord_jeremy> what /usr/local/bin?
[00:52:08] <brad_a> i think
[00:52:13] <lord_jeremy> >.<
[00:52:23] <brad_a> it would build as a terminal application too
[00:52:32] <brad_a> among other things
[00:52:35] <lord_jeremy> yeah I saw you did the bundle stuff
[00:52:54] <lord_jeremy> are there any other Mac devs?
[00:53:05] <brad_a> no official ones
[00:53:51] <lord_jeremy> it looks like it's hardcoded to look at ~/Application Support/GemRB/gemrb.cfg for a configuration right?
[00:54:07] <brad_a> i cant remember
[00:54:21] <brad_a> i know gemrb itself looks in about 4 places
[00:54:30] <brad_a> im pretty sure it also looks at ./
[00:54:56] <lord_jeremy> yeah I see that when it launches
[00:55:33] <brad_a> i think in the end im going to have the override and guiscripts in the app bundle
[00:55:52] <brad_a> i think they arent there now because i didnt feel like learning the cmake syntax to make it happen
[00:56:02] <brad_a> i was quite sick of cmake at the time
[00:56:39] <brad_a> like i said i have a long list of things to do regarding building on mac
[00:57:16] <brad_a> one of which is to get a mac buildbot slave...
[00:57:56] <lord_jeremy> wish I could help more, I know a lot more about actual programming than I do about build scripts
[00:58:30] <brad_a> well i knew nothing of cmake before i did the gemrb mac stuff
[00:58:47] <brad_a> hece the "being quite sick of cmake at the time"
[00:59:25] <lord_jeremy> indeed
[01:03:00] <brad_a> ill try to have a patch for the permissions this week, but the other things im not so inclined to do in the near future
[01:04:41] <brad_a> i should probably add a cmake build target to that xcode project too.
[01:05:17] <lord_jeremy> if you just get rid of the manpage and move gemrb.app and extend2da.py (why is that going to Applications?!!) ~/Applications/ then the permissions problem is fixed because you don't need to install with sudo
[01:06:36] <brad_a> i have no idea what extend2da.py is
[01:06:40] <brad_a> that seems like a mistake
[01:07:00] <lord_jeremy> is that maybe an override thing?
[01:07:03] <lord_jeremy> or a plugin piece?
[01:07:22] <brad_a> well regardless of what it is it shouldnt be there
[01:07:48] <brad_a> and its not referenced directly in the cmake script so im guessing it was mistakenly placed somewhere
[01:07:51] <lord_jeremy> it's in /admin in the gemrb root
[01:08:18] <brad_a> ah
[01:08:23] <brad_a> it is referenced in the cmake
[01:08:28] <lord_jeremy> it's in the root level cmake_install.cmake
[01:08:28] <lord_jeremy> yeah
[01:08:32] <brad_a> i typoed while searching :(
[01:08:44] <brad_a> well i can remove that easy
[01:08:52] <lord_jeremy> ah
[01:08:59] <brad_a> tomprince: what is that py script for?
[01:08:59] <lord_jeremy> it's destination is BIN_DIR
[01:09:02] <brad_a> yes
[01:09:05] <brad_a> i see that
[01:09:12] <brad_a> but i dont think it is needed
[01:09:42] <brad_a> ill just cram it in app support like everything else
[01:10:06] <brad_a> as for ~/Applications vs /Applications id like to make that optional somehow
[01:10:26] <brad_a> either with parameters or by detecting if you are root or not
[01:11:57] <lord_jeremy> root detection makes sense
[01:12:04] <tomprince> It does what it says in the title ...
[01:12:18] <brad_a> its just many people like myself prefer everything to be in /Applications
[01:12:58] <lord_jeremy> tomprince: do people need to use it?
[01:12:58] <lord_jeremy> then you need to make sure that the ~ stuff is user owned
[01:13:05] <brad_a> i will
[01:13:24] <lord_jeremy> cool
[01:14:07] <tomprince> Should be useful for modding
[01:14:15] <brad_a> i see
[01:14:24] <brad_a> looks like it just appends rows to 2da
[01:14:38] <brad_a> oh or columns
[01:15:20] <lord_jeremy> how well is PS:T supported?
[01:15:43] <brad_a> better than IWD2 less than BG2
[01:15:58] <lord_jeremy> oh man IWD2 must be a clusterfrack
[01:16:14] <lord_jeremy> IIRC it's got new file formats for practically everything
[01:16:21] <brad_a> well its just the3rd ed rules are completely unimplemented etc
[01:16:35] <lord_jeremy> ah...
[01:16:37] <brad_a> i think we handle all the file formats
[01:16:39] <tomprince> I think it is more the rules, than the data that is the issue.
[01:17:22] <lord_jeremy> is there any difference between the Mac and Windows versions of the data files?
[01:17:23] <lord_jeremy> endianess?
[01:17:28] <brad_a> no
[01:17:42] <brad_a> well i guess im not actually using mac data tho
[01:18:15] <lord_jeremy> mmm I've got Mac BG1,BG2,IWD covered in dust somewhere
[01:18:26] <brad_a> i think there are some endianess diffrences but i believe we handle them accordingly
[01:18:27] <lord_jeremy> bought GoG versions of everything a while ago
[01:19:05] <brad_a> all i know is gemrb works with mac data and runs on PPC macs so it seems there is no problem regardless of data format
[01:19:20] <lord_jeremy> good to know
[01:19:40] <brad_a> the funny thing is tht windows versions are easier to install on modern macs
[01:19:52] <lord_jeremy> yep…
[01:19:54] <brad_a> since wine/crossover is easy to get and rosetta is gone
[01:20:11] <brad_a> and BG1 needed classic. *shudders*
[01:20:12] <lord_jeremy> yeah they work well in CXG
[01:20:17] <brad_a> yes they do
[01:20:30] <lord_jeremy> though I bootcamp and run them in that
[01:20:31] <lord_jeremy> for simplicities sake
[01:20:49] <brad_a> i havent had a problem running them in crossover
[01:22:08] <lord_jeremy> me neither, but most of my other games are in Windows anyway
[01:22:32] <brad_a> i dont really game much. i havent actually ever played in gemrb other than testing
[01:23:09] <lord_jeremy> I occasionally play Team Fortress 2 and Battlefield 3
[01:23:22] <lord_jeremy> and I play some single player games when they come out
[01:23:41] <lord_jeremy> back in the day I played the IE games when they came out
[01:24:09] <lord_jeremy> and I'm currently playing Dragon Age: Origins
[01:24:09] <lord_jeremy> definitely not as good as BG2
[01:25:36] <brad_a> thats what i hear.
[01:25:38] <lloyd> lord_jeremy:nope def not
[01:25:49] <brad_a> next gen of MBP im gonna finally buy a mac
[01:26:08] <brad_a> lloyd: i have been trying to aask you stuff!
[01:26:17] <lloyd> I know :D
[01:26:21] <brad_a> ha ha
[01:26:24] <lloyd> check pm
[01:26:37] <lord_jeremy> brad_a: don't own a Mac? hackintosh?
[01:26:37] <brad_a> i dont remember what...
[01:26:48] <brad_a> i have a MBP supplied by work is all
[01:27:00] <lord_jeremy> ah ok
[01:27:03] <lloyd> brad_a: you fixed stuff and wanted me to test
[01:27:26] <brad_a> i think i wanted to know if that SDL i gave you was patched to work with our font palettes
[01:27:35] <brad_a> i dont think it was
[01:27:59] <brad_a> but yes feel free to test that i fixed the mac build
[01:28:03] <lloyd> oh..? is this to do with my choices not being red?
[01:28:17] <brad_a> i should learn to check my old private repo next time somebody has problems
[01:28:26] <brad_a> yes
[01:28:28] <brad_a> that is why
[01:28:33] <lloyd> ah k how 2 fix :P
[01:28:59] <brad_a> well if they have fixed the problems building SDL from source i would try a new build from there
[01:29:13] <brad_a> but it doesnt look like they have fixed their drag and drop issues yet
[01:29:28] <brad_a> so it can give you a patch for it but you probably will have to manually apply it
[01:29:43] <brad_a> you can try applying it automatically tho
[01:29:58] <lloyd> there's no way to build the latest sdl on mac...?
[01:30:28] <lord_jeremy> it was broken at some point
[01:30:31] <brad_a> i assume if you simply revert that drag and drop commit
[01:30:33] <lord_jeremy> maybe fixed now
[01:30:36] <brad_a> its not
[01:30:45] <brad_a> but that drag and drop commit is what broke it
[01:31:17] <brad_a> http://dl.dropbox.com/u/13866402/SDLPatches.txt
[01:31:31] <brad_a> you only need the one that says it fixes palettes or whatever
[01:31:35] <brad_a> the other are for iOS
[01:31:47] <lloyd> hmm ok I'll give it a shot
[01:32:56] <brad_a> you may as well try building SDL from source again
[01:33:09] <brad_a> since that patch needs to apply to SDL source
[01:33:23] <brad_a> even if they didnt fix it you will need to rebuild
[01:35:50] <lloyd> then where do I get source that will build ? the one you gave me is just binary and header files irrc
[01:36:06] <lloyd> but anyway let me try anyway
[01:36:14] <lloyd> maybe its better to wait till they fix it
[01:38:26] <brad_a> the one i gave you is compiled
[01:38:35] <brad_a> you have the source frombefore tho
[01:38:48] <brad_a> onen terminal and cd to the sdl dir
[01:39:17] <lloyd> /Library/Framework?
[01:39:23] <brad_a> no
[01:39:29] <brad_a> to where you donwloaded the source
[01:39:37] <brad_a> probably ~/SDL
[01:40:15] <brad_a> once there run hg update -r 61fa4f3238c5
[01:40:40] <brad_a> then open the xcode project and try to build again
[01:40:52] <lloyd> ok what did that do?
[01:41:08] <lloyd> brb
[01:41:12] <brad_a> i think it reverts the broken drag and drop comit
[01:41:22] <brad_a> i dont know mercurial tho :9
[01:42:26] <lloyd> 0 files updated, 0 files merged, 0 files removed, 0 files unresolved
[01:42:33] <lloyd> sounds like it didn't work :P
[01:43:16] <lloyd> http://www.galaxygameworks.com/products.html
[01:43:23] <lloyd> can I just download that one?
[01:45:32] <lloyd> I tried that one but I get Invalid value '4.0' for GCC_VERSION
[01:47:14] <brad_a> no point in using that one
[01:47:23] <lloyd> oh ok
[01:47:39] <brad_a> you already have one setup to build it jsut has that infernal drag and drop stuff
[01:48:02] <brad_a> so since i dont know mercurial and im too lazy to google right now you can jsut comment out that code
[01:48:04] <lloyd> ya so can I still compile it somehow?
[01:48:09] <lloyd> ah ok
[01:48:27] <lord_jeremy> 61fa4f3238c5 ?
[01:48:41] <brad_a> yes
[01:48:44] <lord_jeremy> that sounds like the current rev
[01:48:54] <brad_a> hmmm
[01:48:58] <lord_jeremy> one sec
[01:49:02] <lloyd> ^^ that would make sense :P
[01:49:19] <brad_a> well we need to rool back3
[01:49:30] <brad_a> roll
[01:49:33] <lord_jeremy> one sec, I'm finding mine
[01:49:53] <lord_jeremy> I'm on revision ab97e8ce122a
[01:50:07] <brad_a> how far back is that?
[01:50:13] <lord_jeremy> checking their site
[01:50:33] <brad_a> we need to be on 687400b595ab
[01:50:57] <lord_jeremy> http://hg.libsdl.org/SDL/rev/ab97e8ce122a
[01:51:11] <lord_jeremy> http://hg.libsdl.org/SDL/rev/687400b595ab
[01:51:11] <lord_jeremy> ?
[01:51:38] <brad_a> yes that was the last commit before the drag and drop
[01:51:43] <lord_jeremy> ok
[01:51:48] <lord_jeremy> lets see if it builds
[01:51:57] <brad_a> well yours may need other love
[01:52:06] <brad_a> but lloyd had his environment ready
[01:52:36] <lloyd> 4 files updated, 0 files merged, 2 files removed, 0 files unresolved!
[01:52:40] <lloyd> ok
[01:53:14] <lord_jeremy> mmm mine built, you had a patch that fixes the font issue?
[01:53:28] <brad_a> yes scroll up for link
[01:53:29] <lloyd> huh it built for you ..
[01:53:39] <lord_jeremy> what were your errors?
[01:54:08] <lloyd> mine?
[01:54:21] <lloyd> "_SDL_SendDropFile", referenced from:
[01:54:22] <lloyd> -[SDLAppDelegate application:openFile:] in SDL_cocoaevents.o
[01:54:41] <lloyd> undefined symbol
[01:54:46] <lord_jeremy> that sounds like the drop problem?
[01:54:47] <lord_jeremy> did you change your SDL to rev 687400b595ab?
[01:54:53] <lloyd> ya
[01:54:57] <lord_jeremy> cd into the SDL directory and run hg update -r 687400b595ab
[01:54:57] <lloyd> oh wiat
[01:54:59] <lord_jeremy> oh ok
[01:54:59] <lloyd> right
[01:55:09] <lloyd> this is left over from last upate
[01:55:14] <lloyd> build*
[01:55:30] <brad_a> yeah xcode will do that sometimes :-/
[01:55:38] <lord_jeremy> >.<
[01:55:59] <brad_a> so yeah id check that new build to see if the font problem is still there
[01:56:08] <brad_a> then if it is apply that patch i posted
[01:56:17] <brad_a> but only that first chunk
[01:56:27] <lloyd> wtf I get the same thing >.>
[01:56:40] <lloyd> is there a way to clean targets etc in xcode
[01:56:45] <brad_a> yes
[01:56:47] <lloyd> I think its keeping something from the last thing
[01:56:49] <lord_jeremy> command shift K
[01:56:51] <lord_jeremy> probably
[01:57:02] <brad_a> type clean in the help menu
[01:57:02] <lloyd> ya worked :D
[01:57:06] <brad_a> i love that help menu
[01:57:15] <lord_jeremy> best thing about snow leopard
[01:57:19] <brad_a> indeed
[01:57:25] <brad_a> and NOBODY knos about it
[01:57:41] <lloyd> ffuuuuu I still get the SDL_SendDropFile
[01:57:43] <brad_a> it makes me sad that one of the best features is a nobody
[01:58:09] <brad_a> what does hg say whn you type that command?
[01:58:24] <lord_jeremy> well now it should say nothing
[01:58:25] <lloyd> last time I did it
[01:58:27] <lloyd> <lloyd> 4 files updated, 0 files merged, 2 files removed, 0 files unresolved!
[01:58:29] <brad_a> o you have 2 copies of the source for some reason
[01:58:38] <lord_jeremy> because he already changed revs
[01:58:57] <lord_jeremy> weird
[01:59:03] <lord_jeremy> well I'll be back later
[01:59:04] <brad_a> well i know but if he is on the right rev then he cant possibly get that error on a clean build
[01:59:34] <brad_a> so are you working from an xcode project in a diffrent dir?
[01:59:37] <lord_jeremy> unless the build process is wrong
[01:59:39] <-- lord_jeremy has left IRC (Quit: lord_jeremy)
[02:00:14] <brad_a> i also have to go. but if you cant figure it out ask lord_jeremy to post his hopefuly patched build
[02:00:28] <lloyd> kk thanks
[02:00:41] --> brad_a_ has joined #gemrb
[02:01:08] <-- brad_a_ has left IRC (Client Quit)
[02:04:40] <-- brad_a has left IRC (Ping timeout: 260 seconds)
[02:22:26] --> brad_a has joined #gemrb
[02:24:43] <-- brad_a has left IRC (Client Quit)
[02:34:58] --> joneirik has joined #gemrb
[02:52:48] --> joneirikb has joined #gemrb
[02:55:53] <-- joneirik has left IRC (Ping timeout: 245 seconds)
[04:25:27] <-- joneirikb has left IRC (Remote host closed the connection)
[04:38:15] --> Beh0lder has joined #gemrb
[04:38:16] <-- Beh0lder has left #gemrb
[06:31:35] <-- pugvadr_ has left IRC (Remote host closed the connection)
[06:31:54] --> pugvader has joined #gemrb
[07:30:19] <-- edheldil has left IRC (Remote host closed the connection)
[07:42:24] --> CJS|2 has joined #gemrb
[07:59:00] --> lynxlynxlynx has joined #gemrb
[07:59:00] <-- lynxlynxlynx has left IRC (Changing host)
[07:59:00] --> lynxlynxlynx has joined #gemrb
[07:59:00] --- ChanServ gives channel operator status to lynxlynxlynx
[07:59:25] <fuzzie> morning, lynx
[08:00:55] <lynxlynxlynx> oj
[08:01:35] <-- pugvader has left IRC (Quit: leaving)
[08:32:59] --> edheldil has joined #gemrb
[08:32:59] --- ChanServ gives channel operator status to edheldil
[08:46:31] --> SiENcE has joined #gemrb
[08:46:59] <-- CJS|2 has left IRC (Read error: Connection reset by peer)
[08:47:06] <-- _CJS_ has left IRC (Read error: Connection reset by peer)
[08:55:31] <xrogaan> Modron sensors ?
[08:58:31] <lynxlynxlynx> bots
[09:03:08] <edheldil> good morning
[09:03:46] <edheldil> lynxlynxlynx: no, it's modrons. They are more reliable than bots
[09:05:04] <lynxlynxlynx> at least exultbot
[09:37:52] <lloyd> urgh I can't figure out this searching for xxxxx.2da [error]
[09:38:17] <lloyd> whenever it succeeds in loading a .2da thing there is a file with that .2da somewhere
[09:38:41] <lloyd> or at least when its a gemrb override there is a file with that name when it succeeds in getting a 2da
[09:38:50] <lloyd> in overrides
[09:39:15] <lloyd> but when it fails its always looking for something that's in avatars.2da
[09:39:31] <lloyd> eg chfb1.2d
[09:39:50] <lloyd> chfb1 is an entry in avatars.2da
[09:39:55] <lloyd> not a file
[09:40:42] <lloyd> fuzzie told me that these are sprites of some sort
[09:41:00] <lloyd> which makes sense because it happens whenever I talk to someone at a certain angle :P
[09:41:24] <lloyd> It happens only in BG1 so far
[09:41:32] <lloyd> and in BG trilogy in BG1
[09:41:54] <lloyd> I haven't found a random npc in bg2 yet though
[09:46:42] <lloyd> also it happens in BG1 in both on both linux and mac
[09:54:59] <edheldil> did not have BG1 avatars less directions than the BG2 ones?
[09:56:25] <lloyd> maybe. It also matters where their head is looking etc
[09:56:43] <lloyd> but on a cow i get it %100 as long as its standing in the right place
[09:56:56] <lloyd> error error errror no moo sound
[09:57:01] <lynxlynxlynx> those 2das have some other animation info in them
[09:57:28] <lynxlynxlynx> http://iesdp.gibberlings3.net/files/2da/2da_tob/asounds.htm
[09:58:30] <lloyd> am I right in thinking that it should be looking for avatars.2da rather than chfb1.2da or whatever
[09:58:36] <fuzzie> no.
[09:59:13] <lloyd> all of the other .2da files resource manager tries to open are actual files in override
[09:59:21] <lloyd> at least when it doesn't get an error
[09:59:32] <fuzzie> avatars.2da is loaded at startup and kept in memory.
[10:00:40] <lloyd> ok -- so it should be looking in memory for chfb1.2da shouldn't it?
[10:01:03] <fuzzie> no.
[10:01:07] <lloyd> I mean looking for CHFB1 or whatver in avatars.2da
[10:01:21] <fuzzie> i think you have completely the wrong idea here
[10:01:34] <fuzzie> if the avatars.2da lookup failed, you'd get no graphics
[10:01:50] <lloyd> sometimes that happens :P
[10:02:02] <lloyd> like this gibbering I was fighting totalyl disapeared
[10:02:06] <fuzzie> and a lot of loud complaining
[10:02:14] <fuzzie> but that is nothing to do with your 2da stuff.
[10:02:52] <lloyd> you don't think that chfb1.2da refers to the CHFB1 entry in avatars.2da?
[10:03:34] <fuzzie> well, it probably does
[10:03:49] <fuzzie> when it wants a sound, it says 'ok, this is graphic CHFB1, let's see if there is sound associated with it'
[10:04:39] <lloyd> when I said no sound -- the sound does work if the person is at the right angle
[10:04:50] <lloyd> like looking the right way when you talk to them
[10:05:42] <lloyd> so your saying it looks to see what kind of graphic it is displaying for certain avatar and then finds a sound associated with it?
[10:06:17] <lloyd> oh and btw when the sound does work it never searches for a 2da
[10:06:45] <lynxlynxlynx> yes, these cblalba are for characters btw (c- character h- human f- female + suffix)
[10:07:07] <lloyd> ah interesting
[10:08:09] <lloyd> so does this idea make sense: Its fails to look/find CHFB1 in avatars.2da (in memory) and then appends .2da and asks resource manager for it?
[10:08:11] <lynxlynxlynx> maybe we're mixing different systems together
[10:08:43] <lynxlynxlynx> pst has sounds in separate ini files for example, then there are the verbal constants
[10:09:21] <lynxlynxlynx> CHFB1 is not missing from avatars
[10:09:32] <lloyd> no its there
[10:10:05] <lloyd> but I'm saying maybe something goes wrong or something and it then decides to look in resource manager?
[10:11:51] <lloyd> how can I debug this?
[10:12:17] <lloyd> I've looked at everything resources manager down and can't find anything wrong
[10:12:58] <lloyd> it just seems to be searching for something that only exists in avatars.2da and is instead searching for that entry with .2da appended
[10:16:13] <lloyd> is it GameData::LoadTable
[10:16:28] <wjp> that's just the generic 2da loader
[10:16:55] <lloyd> that wouldn't be used in this case?
[10:17:10] <lloyd> I mean i looks in some sort of table
[10:17:16] <lloyd> and if it doesn't find it looks in resource manager
[10:17:37] <lloyd> it fits in with what I thought :P
[10:18:04] <lynxlynxlynx> you could break in the manager and check the backtrace
[10:18:21] <wjp> you mean inside LoadTable? That just checks if the 2da is already loaded, and if not, looks for it
[10:18:35] <lynxlynxlynx> i think you're interested in Actor::GetSoundFrom2DA
[10:19:06] <lloyd> that sounds promising
[10:19:59] <lloyd> so chfb1 would be passed as sound?
[10:20:29] <lloyd> I guess I'll try a debugger
[10:21:31] <wjp> but I'm guessing if it gets there it might already be too late? (Assuming that the 2da is not where it should be looking, since that apparently doesn't exist?)
[10:22:10] <lloyd> ya I think you're right
[10:23:25] --> pugvader has joined #gemrb
[10:32:26] <lloyd> yup its LoadTable calling GetTableIndex which fails and then goes to res manager
[10:48:33] <lloyd> actually its not...something goes wrong earlier..
[11:11:23] --> Yoshimo has joined #gemrb
[11:17:25] --> Beholder has joined #gemrb
[11:17:31] <Beholder> hi all
[11:23:14] <Yoshimo> hi
[11:27:17] <-- lynxlynxlynx has left IRC (Read error: Connection reset by peer)
[11:31:43] --> _CJS_ has joined #gemrb
[11:35:14] --> lynxlynxlynx has joined #gemrb
[11:35:14] --- ChanServ gives channel operator status to lynxlynxlynx
[11:45:00] <-- SiENcE has left IRC (Quit: @all: cya)
[12:41:09] --> lord_jeremy has joined #gemrb
[12:45:24] <lord_jeremy> lloyd: get your SDL to build?
[12:45:35] <lloyd> yep!
[12:45:55] <lord_jeremy> did it need the font patch or was that already fixed?
[12:45:55] <pugvader> all countries are national-socialist
[12:46:09] <lloyd> initial gemrb doesn't build with 1.3 for me though
[12:46:17] <lord_jeremy> really?
[12:46:18] <lloyd> oh I didn't do the font patch
[12:46:43] <lloyd> ya because some newthread function or something has 3 args in 1.3 but only 2 in 1.2
[12:46:44] <lord_jeremy> the most recent trunk *should* build, I think brad submitted his fixes
[12:46:59] <lord_jeremy> huh, I haven't run into that problem
[12:46:59] <lloyd> in gemrb it uses the 2 thread
[12:47:03] <lloyd> hmmm
[12:47:05] <lloyd> one sec
[12:47:09] <lord_jeremy> can you tell me exactly what it is?
[12:47:13] <lloyd> ya
[12:48:27] <lloyd> player = SDL_CreateThread(&play,"derp",(void *) this);
[12:48:36] <lloyd> line 68 AmbientMgrAL.cpp
[12:48:48] <lloyd> in OpenAL plugin folder
[12:48:56] <lloyd> I had to put derp in to get it to work
[12:50:05] <lloyd> also line 189 of OpenALAudio.cpp
[12:50:10] <lloyd> has one
[12:50:13] <lloyd> musicThread = SDL_CreateThread( MusicManager,"herp", this );
[12:50:29] <lloyd> they are thread names
[12:50:57] <lloyd> or maybe I compiled it wrong?
[12:51:53] <lord_jeremy> and the error is specifically the function expects 3 parameters?
[12:52:48] <lord_jeremy> well I'm looking at SDL_thread.h from my SDL 1.3 build
[12:53:24] <lord_jeremy> the function SDL_CreateThread actually takes 4 parameters, but it's got a redefinition to take 2 parameters as well
[12:53:37] <lloyd> my redefinition takes 3 params
[12:53:40] <lloyd> :P
[12:53:57] <lord_jeremy> that's with the 1.3 you just compiled?!!
[12:54:05] <lloyd> yep
[12:54:16] <lloyd> unless I goofed it
[12:54:21] <lloyd> but I don't think so
[12:54:40] <lord_jeremy> gimme a sec
[12:54:50] <lloyd> SDL_CreateThread(SDL_ThreadFunction fn, const char *name, void *data);
[12:55:15] <lord_jeremy> alright
[12:55:15] <lord_jeremy> yeah
[12:55:17] <lord_jeremy> this is FUBAR
[12:55:23] <fuzzie> yeah, shouldn't the name param be in any recent 1.3 builds?
[12:55:34] <lord_jeremy> between the revision I was using a few days ago and the revision that you and I just built yesterday
[12:55:40] <lloyd> :D
[12:55:49] <lord_jeremy> they changed the number of parameters that SDL_CreateThread takes
[12:55:50] <fuzzie> changed 2 months ago
[12:55:50] <lord_jeremy> what the hell
[12:55:52] <fuzzie> :P
[12:56:11] <lord_jeremy> lol
[12:56:59] <lord_jeremy> I'm trying to find their damn documentation
[12:57:22] <fuzzie> relying on their docs is not a particularly great idea
[12:57:36] <fuzzie> if you really want to use 1.3, you rather need to keep up with the list/repository
[12:57:40] <lord_jeremy> yeah their docs say it takes two parameters
[12:57:42] <lord_jeremy> evidently
[12:58:03] <lloyd> they wrote a long paragraph in comment above function :P
[12:58:09] <lord_jeremy> well this is problematic though because older builds of 1.3 will have 2 parameters and newer will have 3
[12:58:11] <lloyd> * Create a thread.
[12:58:11] <lloyd> *
[12:58:11] <lloyd> * Thread naming is a little complicated: Most systems have very small
[12:58:11] <lloyd> * limits for the string length (BeOS has 32 bytes, Linux currently has 16,
[12:58:11] <lloyd> * Visual C++ 6.0 has nine!), and possibly other arbitrary rules. You'll
[12:58:13] <lloyd> * have to see what happens with your system's debugger. The name should be
[12:58:17] <lloyd> * UTF-8 (but using the naming limits of C identifiers is a better bet).
[12:58:19] <lloyd> * There are no requirements for thread naming conventions, so long as the
[12:58:21] <lloyd> * string is null-terminated UTF-8, but these guidelines are helpful in
[12:58:23] <lloyd> * choosing a name:
[12:58:25] <lloyd> *
[12:58:27] <lloyd> * http://stackoverflow.com/questions/149932/naming-conventions-for-threads
[12:58:29] <lloyd> *
[12:58:31] <lloyd> * If a system imposes requirements, SDL will try to munge the string for
[12:58:33] <lloyd> * it (truncate, etc), but the original string contents will be available
[12:58:35] <lloyd> * from SDL_GetThreadName().
[12:59:55] <lord_jeremy> so… 1.2 definitely has a 2 parameter function
[13:00:05] <lord_jeremy> and as of ~2 months ago? 1.3 has a 3 parameter function
[13:00:10] <lord_jeremy> #if guards?
[13:00:28] <lord_jeremy> it doesn't really matter what the thread names are
[13:00:31] <lloyd> its only used in two places atm
[13:00:53] <lord_jeremy> yeah, both for initializing sound threads...
[13:01:08] <lloyd> how can you differentiate between different SDL builds
[13:01:15] <lloyd> 1.3 builds*
[13:01:33] <lord_jeremy> I doubt you can with just #defines
[13:02:34] <lord_jeremy> yeah, there's only version number guards
[13:02:37] <lloyd> maybe just don't support older 1.3's and if its 1.3 have 3 params?
[13:02:47] <lord_jeremy> yeah that's the way to do it
[13:03:06] <lord_jeremy> maybe with a comment nothing that older 1.3 builds still take two parameters
[13:03:13] <lord_jeremy> I can write the patch
[13:03:35] <lloyd> noting*?
[13:03:47] <lord_jeremy> yeah
[13:03:52] <lloyd> cool
[13:04:31] <lord_jeremy> the OpenALAudioDriver thread looks like it's only created once
[13:04:44] <fuzzie> should be, yes.
[13:04:45] <lord_jeremy> assuming we don't create multiple instances of OpenALAudioDriver :P
[13:04:51] <lord_jeremy> what about AmbientMgrAL?
[13:05:05] <lord_jeremy> the thread is created in setAmbients()
[13:05:32] <lord_jeremy> actually, heh
[13:05:32] <lloyd> (obviously I just put static strings in there and nothing bad happened)
[13:05:39] <lord_jeremy> there's an assert(player == null)
[13:06:06] <lord_jeremy> ok
[13:06:11] <lord_jeremy> no need for unique thread names
[13:08:17] <lord_jeremy> oh wait
[13:08:23] <lord_jeremy> in SDL_version.h
[13:08:30] <lord_jeremy> there is a GetRevisionNumber function
[13:08:42] <lord_jeremy> Returns a number uniquely identifying the exact revision of the SDL
[13:08:43] <lord_jeremy> * library in use. It is an incrementing number based on commits to
[13:08:43] <lord_jeremy> * hg.libsdl.org.
[13:08:56] <lord_jeremy> that *should* work for this
[13:09:14] <lloyd> yep unless they change their minds!
[13:09:26] <lord_jeremy> someone want to go find what commit this was changed in?
[13:10:36] <lord_jeremy> act though that function is a function not a define
[13:10:40] <lord_jeremy> ach*
[13:10:53] <lord_jeremy> oh well
[13:11:08] <lloyd> can u just call it then?
[13:11:40] <lord_jeremy> well yeah but it won't be a define then
[13:11:45] <lord_jeremy> it'll be a runtime determination
[13:11:58] <fuzzie> let's not support old SDL 1.3
[13:12:09] <lord_jeremy> ok
[13:12:16] <fuzzie> that sounds like a terrible idea
[13:12:41] <lloyd> ya it would be ugly
[13:12:44] <lord_jeremy> question, what is the "official" editor standard?
[13:12:51] <fuzzie> editor?
[13:12:54] <lord_jeremy> I see some files with 4-space indents and some with 2
[13:13:25] <lord_jeremy> although for the most part everything just seems to be tabs, so I guess it doesn't matter?
[13:15:24] <lord_jeremy> does anyone know at around what revision this changed?
[13:16:09] <fuzzie> in 1.3?
[13:16:21] <lord_jeremy> yeah
[13:16:32] <fuzzie> in 3a041d215edc
[13:17:12] <lord_jeremy> http://hg.libsdl.org/SDL/rev/3a041d215edc
[13:17:26] <lord_jeremy> am I right that 5969 is the actual revision number?
[13:17:30] <fuzzie> no
[13:17:38] <CIA-44> GemRB: 03lynxlupodian * r4e5ef8f05988 10gemrb/gemrb/core/Game.cpp: CheckForReplacementActor: also copy the area resref in case the data is bad
[13:17:43] <fuzzie> you will break people's heads if you start using hg revision numbers, right?
[13:18:05] <lord_jeremy> I want to note in a comment which rev you need for it to work
[13:18:07] <fuzzie> > Revision numbers referring to changesets are very likely to be different in another copy of a repository. Do not use them to talk about changesets with other people. Use the changeset ID instead. \
[13:18:15] <lord_jeremy> >.<
[13:18:17] <lord_jeremy> ok
[13:18:24] <lord_jeremy> stupid mercurial
[13:18:37] <fuzzie> well, git solves this by making it very difficult to get a revision number
[13:18:50] <lynxlynxlynx> lord_jeremy: tabs
[13:19:02] <lord_jeremy> thanks
[13:19:09] <fuzzie> lynxlynxlynx: maybe we should put a note in CodingStyle.txt?
[13:19:31] <lord_jeremy> worked on some projects where the devs were hugely anal about their 7-space indents o.O
[13:19:39] <lynxlynxlynx> ok, i'll hash out a HACKING
[13:20:12] <fuzzie> well, as you can see, no-one has been particularly consistent in the past
[13:20:16] <fuzzie> not about *any* coding style
[13:20:35] <lynxlynxlynx> variable naming is my favourite :)
[13:21:16] <lord_jeremy> what's up with the proper case variables names, btw?
[13:22:04] <fuzzie> well, it's CamelCase.
[13:22:05] <lord_jeremy> e.g. int Count = 0; int Max = Distance(x,y);
[13:22:06] <lord_jeremy> etc
[13:22:16] <lord_jeremy> you mean camelCase
[13:22:20] <lord_jeremy> thisIsCamelCase
[13:22:24] <lord_jeremy> ThisIsProperCase
[13:22:25] <fuzzie> pretty sure i don't :)
[13:22:31] <lord_jeremy> wait
[13:22:49] <fuzzie> lowerCamelCase vs UpperCamelCase!
[13:23:00] <lord_jeremy> yeah ok
[13:23:09] <fuzzie> i see that wikipedia has a whole variety of them on the CamelCase page in case you would like your head to explode.
[13:23:27] <lord_jeremy> again, I see some files have lower and other have upper
[13:23:55] <fuzzie> well, the gemrb standard is upper, and most of us work generally on open source stuff which uses lower, so we tend to forget
[13:24:05] <lord_jeremy> ah
[13:24:20] <fuzzie> at some point we were pondering renaming everything and moving to m_ and g_ while we were at it
[13:24:26] <fuzzie> but, we have bigger problems
[13:24:27] <lord_jeremy> when I heard your lead dev was hooked to msvc6 I figured I'd be seeing hungarian
[13:26:36] <pugvader> you ever break something and let it heal wrong just to see how it would heal?
[13:27:04] <lord_jeremy> alright it builds with 1.3 now
[13:27:12] <lord_jeremy> I'm going to try it with 1.3
[13:27:14] <lord_jeremy> 1.2*
[13:27:48] <lord_jeremy> still builds with 1.2
[13:35:59] <lord_jeremy> I've made a git patch
[13:36:21] <lord_jeremy> I don't have write access so one of you will have to commit it
[13:39:37] <lord_jeremy> http://dl.dropbox.com/u/17088718/fix_sdlcreatethread.patch
[13:42:36] <lynxlynxlynx> sure
[13:42:53] --> SiENcE has joined #gemrb
[13:42:54] <gembot> build #274 of cmake g++-4.5.2 is complete: Success [build successful] Build details are at http://buildbot.gemrb.org/builders/cmake%20g%2B%2B-4.5.2/builds/274
[13:52:50] <lord_jeremy> can someone tell me where spell effects are handled?
[13:52:57] <lynxlynxlynx> which stage?
[13:53:00] <CIA-44> GemRB: 03lynxlupodian * r5b3122e8b224 10gemrb/HACKING: added a basic HACKING note
[13:53:02] <CIA-44> GemRB: 03jeremyagost * r2be4c28b7bf6 10gemrb/gemrb/plugins/OpenALAudio/ (AmbientMgrAL.cpp OpenALAudio.cpp): As of SDL 1.3 changeset 3a041d215edc SDL_CreateThread takes three parameters.
[13:53:02] <lord_jeremy> specifically, where the projectiles are spawned for spell effects
[13:53:34] <lynxlynxlynx> in Scriptable.cpp
[13:53:39] <lynxlynxlynx> GetProjectile iirc
[13:53:41] <lord_jeremy> thank you
[13:53:56] <lord_jeremy> oh that reminds me
[13:53:58] <-- lord_jeremy has left IRC (Quit: lord_jeremy)
[13:54:10] <lynxlynxlynx> Scriptable::CreateProjectile
[13:54:25] --> jeremyagost has joined #gemrb
[13:54:38] <jeremyagost> much better
[13:54:39] <lynxlynxlynx> Scriptable::CreateProjectile
[13:55:05] <lynxlynxlynx> thanks for the patch btw
[13:55:50] <jeremyagost> I pushed my gemrb clone to github and am using that at the moment
[13:56:02] <jeremyagost> easier to deal with since I've got multiple work machines
[14:01:00] <edheldil> jeremyagost: what about writing a configure check and set a define based on the results? That's the way to do it, IMO
[14:01:27] <fuzzie> edheldil: for every time an unstable API changes, though?
[14:04:08] <edheldil> ok, if it's still unstable, than ditch the previous api
[14:05:03] <fuzzie> i assume that is what got committed :)
[14:05:20] <fuzzie> (can't look from here right now
[14:05:21] <fuzzie> )
[14:06:01] <jeremyagost> yup, 1.2 considered stable with the 2-parameter function, 1.3 has 3-parameter
[14:08:49] <fuzzie> btw about previous conversation: upstream SDL 1.3 is indeed still trusting eglChooseConfig
[14:09:15] <fuzzie> so on android, whether you get good performance will depend on the drivers on the device
[14:10:59] <-- Beholder has left #gemrb
[14:13:36] <edheldil> <lord_jeremy> when I heard your lead dev was hooked to msvc6 I figured I'd be seeing hungarian.
[14:14:06] <edheldil> ^^^ you were closer than you think ;-)
[14:14:16] <fuzzie> heh :)
[14:23:38] <jeremyagost> uh oh
[14:23:53] <jeremyagost> I just got a crash in AmbientMgrAL while testing something else
[14:25:47] <jeremyagost> http://pastebin.com/H3c2W51
[14:26:11] <jeremyagost> whoops, http://pastebin.com/H3c2W51n
[14:26:43] <jeremyagost> crashed thread trace starts at 85
[14:29:08] <jeremyagost> doesn't seem like it's related to the patch I just made, but it's quite a coincidence
[14:33:54] <pugvader> http://www.youtube.com/watch?v=nt86Rtw3Z7M Carl Craig - My Machines
[14:59:52] <-- lynxlynxlynx has left IRC (Ping timeout: 255 seconds)
[15:01:05] --> lynxlynxlynx has joined #gemrb
[15:01:05] <-- lynxlynxlynx has left IRC (Changing host)
[15:01:05] --> lynxlynxlynx has joined #gemrb
[15:01:05] --- ChanServ gives channel operator status to lynxlynxlynx
[15:50:44] <-- jeremyagost has left IRC (Quit: jeremyagost)
[15:56:04] <pugvader> the holocaust is a lie
[15:56:48] <pugvader> the holocaust is a lie
[15:58:04] <-- pugvader was kicked from #gemrb by wjp (enough already)
[16:20:40] --> Beholder has joined #gemrb
[16:25:09] <-- Beholder has left #gemrb
[16:33:29] --> jeremyagost has joined #gemrb
[16:33:48] <jeremyagost> does anyone know what the original engine conditions are for how long a lightning bolt travels?
[16:34:11] <jeremyagost> is it a certain number of bounces or a timer?
[16:36:38] <fuzzie> > 23:04 <Avenger> i think it goes this way:1, lightning bolt which explodes on target. 2. once it explodes, it continues on the previous path, until it hits a wall. 3. once it hit a wall it creates another type 2 lightning or grounds itself.
[16:37:15] <tomprince> jeremyagost: If you put a repo up on github, you should fork from fuzzie's repo.
[16:38:28] <jeremyagost> fuzzie has a repo?
[16:38:40] <jeremyagost> dur guess I should've looked first
[16:39:07] --> Maighstir has joined #gemrb
[16:39:07] <jeremyagost> fuzzie: yeah as far as I can tell it creates a bolt that flies to the target, then after it hits it continues damaging in its path until it hits a wall
[16:39:15] <jeremyagost> with the bouncing
[16:39:32] <jeremyagost> it seems like it just dies after a certain amount of time
[16:40:40] <lynxlynxlynx> yes
[16:41:25] <fuzzie> distance-based?
[16:41:29] <fuzzie> i have no idea where to look
[16:41:51] <jeremyagost> distance-based makes sense
[16:41:52] <lynxlynxlynx> atleast from the spell description, the length is level dependant
[16:41:55] <lynxlynxlynx> http://www.gamebanshee.com/showshot.php?/baldursgateii/spells/images/lightningbolt.jpg
[16:42:44] <jeremyagost> so we presume the start point to be the targeted creature
[16:43:28] <jeremyagost> you mean level-independant?
[16:43:47] <lynxlynxlynx> no
[16:43:53] <lynxlynxlynx> the 40
[16:44:00] <lynxlynxlynx> check the top of the description
[16:44:08] <jeremyagost> oh I see
[16:44:46] <lynxlynxlynx> that's usually the range from which you can cast a spell, but it may have been abused
[16:45:13] <jeremyagost> huh the units are mixed
[16:45:21] <jeremyagost> that might just be copied from a DnD rulebook
[16:45:41] <jeremyagost> doesn't seem like the game adheres to the range per level thing
[16:45:55] <lynxlynxlynx> i suggest you look up the projectile data
[16:45:56] <fuzzie> never trust the spell descriptions :P
[16:46:27] <jeremyagost> you think that would be in the 2da?
[16:46:27] <jeremyagost> not hardcoded?
[16:46:29] <fuzzie> unfortunately this is almost certainly hardcoded and I doubt Avenger added a field
[16:49:19] <jeremyagost> ok range is definitely level-independant
[16:58:51] --> brad_a has joined #gemrb
[17:10:52] <jeremyagost> is there an integer sign function defined somewhere?
[17:14:48] <brad_a> looks like i will finally have to get around to updation iOS SDL ;-)
[17:15:32] <brad_a> hope they fixed all the things i had to hack in there originally. that has been what is keeping me.
[17:34:16] <-- SiENcE has left IRC (Quit: @all: cya)
[17:56:45] <jeremyagost> does anyone know if there's a technical reason that the Point class in Region.h doesn't have operators like +.-.assign implemented?
[17:57:42] <fuzzie> I imagine just because no-one's ever thought it worth the hassle.
[17:57:51] <jeremyagost> ok
[17:59:35] <tomprince> And there has been a bit of tendency to avoid C++ features. :)
[18:00:31] <-- Yoshimo has left IRC (Read error: Connection reset by peer)
[18:01:17] <fuzzie> Well, it doesn't really help when the new C++ things we gain are things like CObject that are inevitably incomprehensible to people who aren't that familiar with C++.
[18:01:36] <tomprince> :)
[18:02:18] <jeremyagost> isn't that MFC-specific?
[18:02:22] <fuzzie> I would *hope* that things like operator+/- for Point are going to be a lot better.
[18:03:00] <fuzzie> jeremyagost: Our CObject, which is a Python/C++ helper shared pointer class.
[18:03:10] <jeremyagost> oooooh ok
[18:03:15] <brad_a> without operator+/- how is a class better than a struct for those?
[18:03:28] <jeremyagost> I only asked because I had assumed that operator= existed and then was surprised when it didn't
[18:03:41] <fuzzie> brad_a: Well, it's effectively a struct, since it's a class with only public members.
[18:04:06] <fuzzie> But a struct with operator +/- would also effectively be that..
[18:04:10] <tomprince> jeremyagost: I would say, go ahead and add them.
[18:04:16] <brad_a> right so i also am suprised it doesnt have those operators overloaded
[18:04:28] <wjp> jeremyagost: only if you actually use them though :-)
[18:04:34] <fuzzie> Well, as I said, I imagine no-one's ever thought it worth the hassle of adding them.
[18:04:43] <fuzzie> I can't imagine we add points very much.
[18:04:56] <jeremyagost> indeed
[18:05:02] <fuzzie> And as tomprince said, there was at least historically a tendency to avoid C++ features.
[18:05:24] <wjp> we may recently have overcompensated in some places :-)
[18:05:44] <tomprince> That would be me ;)
[18:06:13] <brad_a> im gonna have to go hide in a corner in a bit when i post this TTF patch that usses malloc etc instead of vectors and maps :(
[18:07:06] <jeremyagost> for the lightning bolt "continue" functionality there's going to need to be some way of delivering damage to whatever the bolt touches
[18:07:55] <jeremyagost> I was thinking to add a proximity check to DoStep but there's the small issue of dealing with slow projectiles dealing damage multiple times to one creature
[18:08:12] <brad_a> thats how ligtning works in the original
[18:08:18] <brad_a> it was very annoying
[18:08:34] <jeremyagost> it never seems to deal damage twice before a bounce
[18:08:44] <brad_a> i assure you it does
[18:08:58] <brad_a> well i think anyway
[18:08:59] <jeremyagost> but I think the original engine got around that by making the bolt proximity very small, which is kind of hacky
[18:09:21] <brad_a> well im not going to complain about being better than the original
[18:09:23] <brad_a> :)
[18:09:43] <jeremyagost> well my thought was to implement it such that a lightning bolt won't damage the same creature twice in one "bounce"
[18:10:01] <brad_a> thats good wether or not its how original works
[18:12:18] <fuzzie> well, this is why i quoted Avenger above
[18:12:38] <fuzzie> it's meant to hit *once* and then be another form
[18:13:03] <fuzzie> and while enhancements are fine, it still has to work the original way by default
[18:13:24] <brad_a> except the original way seemed like a bug...
[18:13:30] <brad_a> to me anyway
[18:14:24] <tomprince> I think the discussion is regarding the other form.
[18:15:07] <fuzzie> i mean, it's meant to hit target and then head for the wall, if set to bounce, no? rather than continuing to hit
[18:15:16] <fuzzie> i am not wanting to get distracted going to look
[18:16:17] <brad_a> thats what i think its meant to do. ie not hit multiple times before a bounce like i *think* ihappens in the original
[18:18:09] <jeremyagost> from casting to first target it is a normal project and doesn't hit anything else
[18:18:25] <brad_a> that doesnt sound right either tho
[18:18:38] <brad_a> i though it should hit anything between caster and target too
[18:18:39] <jeremyagost> that's what happens
[18:18:47] <jeremyagost> nope, that's not the way the original engine works
[18:18:51] <brad_a> really?
[18:18:53] <jeremyagost> yep
[18:18:59] <tomprince> So it only hits one thing on each bounce then?
[18:19:01] <brad_a> are you sure?
[18:19:15] <jeremyagost> after it hits the target it then continues in a line, damaging anything it hits from here on
[18:19:21] <jeremyagost> with it bouncing for a few seconds
[18:19:42] <jeremyagost> after it has hit the original target it can hit as many things per bounce as it comes into contact with
[18:19:45] <brad_a> is it diffrent behavior in BG1 vs BG2 then?
[18:19:55] <jeremyagost> no, same AFAIK
[18:20:02] <fuzzie> lightning projectile is different to bouncing lightning projectile
[18:20:03] <brad_a> I think we should test
[18:20:05] <fuzzie> is the bouncing one in bg1?
[18:20:19] <jeremyagost> I just now tested BG2
[18:20:34] <jeremyagost> I don't know about BG1, I haven't played the BG1 engine in a long time
[18:21:04] <fuzzie> i forget.
[18:21:10] <brad_a> i guess we all do :(
[18:21:33] <Maighstir> I'm quite sure there is bouncing lightning in BG1. I think there's a trap somewhere in a gnome village using it.
[18:21:41] <brad_a> oh maybe im remembering when i cast lightning not at a target but just a floor tile???
[18:21:59] <brad_a> if you do that then it hits anything between caster and target right?
[18:22:15] <jeremyagost> you can't target the floor with Lightning spell
[18:22:20] <jeremyagost> it only targets creatures
[18:22:30] <jeremyagost> what I was talking about with limiting the number of hits, in practice in BG2 it doesn't seem to ever hit the same creature twice in one line
[18:22:43] <brad_a> bah i have all these PTSD nighmares about using lightning! yes apaprently it doesnt work the way i recall
[18:23:02] --> Yoshimo has joined #gemrb
[18:23:27] <fuzzie> bg1 lightning bolt is one hit per bounce.
[18:23:44] <jeremyagost> it will only hurt one thing per bounce?
[18:23:47] <jeremyagost> are you certain?
[18:24:02] <jeremyagost> I tested with BG2 by filling a tight space with lots of creatures
[18:24:11] <fuzzie> well, i am not certain
[18:24:18] <fuzzie> only one hit on the first bolt, anyway
[18:24:29] <brad_a> i think BG2 function is more important if it came down to difficulty having it both ways
[18:24:42] <fuzzie> which is really odd actually
[18:24:47] <jeremyagost> as far as I can tell, the mephit lightning bolts work the same way in regard to damaging others only after hitting the primary target
[18:24:55] <fuzzie> there's a youtube video with it hitting two chars on the second bounce, in bg1
[18:25:07] <jeremyagost> yeah that makes ense
[18:25:24] <jeremyagost> what I'm talking about might just be gemrb specific
[18:25:24] <fuzzie> coincidence in numbers, i think; it only hits one char on the first bounce but maybe they're not in line enough in the video
[18:25:33] <fuzzie> i am not good with bg1
[18:25:58] <jeremyagost> but the projectile update steps are so close together, I think the same char might be hit multiple time by just one bolt going through him
[18:26:14] <brad_a> that is one of my PTSD nightmares ;-)
[18:26:19] <fuzzie> i don't think it's meant to go off for every char anyway
[18:26:29] <fuzzie> i don't know, i'd have to have more time to play with it.
[18:26:29] <jeremyagost> well as far as I can tell, the IE solved this problem by making the lightning "hitbox" really small
[18:26:45] <fuzzie> but it seems the lightning doesn't necessary hit two chars if it goes through them both.
[18:26:56] <jeremyagost> that way it'd be rare for a character to be damaged multiple times when they were only "hit" once
[18:27:02] <jeremyagost> that's just a visual inconsistency fuzzie
[18:27:16] <fuzzie> right, but it means the hitbox can't possibly be the full char
[18:27:27] <fuzzie> i mean, playing with this in original engine
[18:27:32] <jeremyagost> related to the fact that the lightning BAM has a limited number of turn angles as compared to the actual positioning of the projectile
[18:27:56] <brad_a> yeah that was annoying too
[18:28:03] <brad_a> silly sideways flying lighning
[18:28:22] <jeremyagost> in the original engine, I guarantee that the lightning damage area is a relatively small circle around the projectile coordinates
[18:28:46] <jeremyagost> but the lightning graphic is an approximate line with a limited number of turn frames
[18:28:49] <fuzzie> hm
[18:29:10] <jeremyagost> hence it missing when it looks like it should hit and visa versa
[18:29:36] <jeremyagost> well I'm going to get this working first disregarding multiple accidental hits
[18:29:41] <jeremyagost> to see if it's really a problem
[18:29:41] <fuzzie> it isn't doing anything like that
[18:30:35] <fuzzie> it looks like it's just checking the search squares
[18:30:55] <jeremyagost> what is? engine or GemRB?
[18:30:58] <fuzzie> original engine
[18:31:10] <jeremyagost> decomp?
[18:31:13] <fuzzie> yes
[18:31:21] <jeremyagost> ah ok
[18:31:40] <jeremyagost> heh so the hit is pretty rough
[18:31:51] <fuzzie> i don't have IDA or anything here
[18:33:02] <fuzzie> hmm
[18:33:24] <jeremyagost> so how should I implement this?
[18:33:42] <jeremyagost> does the Map class have a search square querie?
[18:34:19] <fuzzie> maybe you shouldn't listen to anything I say actually :)
[18:34:48] <jeremyagost> well search squares would be a clean way to check for hits
[18:34:52] <fuzzie> what I *do* see is that CProjectileLightningBolt::DeliverEffects() constructs a new CProjectileLightningBounce before self-destructing
[18:35:01] <jeremyagost> that makes sense
[18:36:35] <jeremyagost> if a Projectile creates a child with CreateIteration() does the original Projectile have to stay alive for the child to stay alive?
[18:36:58] <fuzzie> but it the Bounce variant does not seem to create any more.
[18:37:05] <fuzzie> but then, even
[18:37:22] <fuzzie> no, the CreateIteration ones are independent
[18:37:29] <jeremyagost> a lightning instance of Projectile doesn't have an extension so I was just planning on reusing the original Projectile
[18:37:34] <fuzzie> but you can't use it because it has hard-coded ids
[18:37:41] <jeremyagost> ?
[18:38:00] <fuzzie> CreateIteration creates projectiles with an id one less than the current id
[18:38:04] <fuzzie> i don't think you need it anyway
[18:38:22] <jeremyagost> yeah I've added a step to ChangePhase which repurposes the existing one
[18:39:02] <fuzzie> don't we have both projectiles?
[18:39:20] <jeremyagost> what do you mean?
[18:39:59] <fuzzie> i mean, the PEF_BOUNCE should only be set on the bouncing version
[18:40:13] <jeremyagost> there is no bouncing version
[18:40:23] <fuzzie> there is, it's lightb.pro
[18:40:27] <jeremyagost> when the spell is cast it has the PEF_BOUNCE flag set at the begining
[18:40:49] <fuzzie> *that* doesn't seem right
[18:40:58] <jeremyagost> I tested it myself
[18:41:13] <jeremyagost> as soon as the spell is cast it has both PEF_BOUNCE and PEF_CONTINUE
[18:41:16] <fuzzie> yes, but probably our data is wrong in that case
[18:41:30] <jeremyagost> mmm
[18:41:46] <jeremyagost> I don't know enough about GemRB to comment on that
[18:41:48] <lynxlynxlynx> deja vu :o
[18:42:02] <fuzzie> well: all of this is hardcoded completely in original
[18:42:07] <fuzzie> PEF_BOUNCE is GemRB-specific flag
[18:42:20] <jeremyagost> if the current setup for this is wrong
[18:42:25] <fuzzie> we have gemprjtl.ids in override/<gametype>/ which maps ids onto projectile files
[18:42:27] <jeremyagost> then I need to radically change my implementation
[18:42:29] <fuzzie> and then the projectile files also there
[18:42:36] <fuzzie> yes, well, i don't know what i'm talking about, mind
[18:42:48] <jeremyagost> what a coincidence, neither do I!
[18:42:59] <fuzzie> i'm just thinking that it doesn't seem right, since original is using *two* projectiles
[18:43:02] <jeremyagost> well at least the continuation code I've written makes ense
[18:43:03] <jeremyagost> sense
[18:43:33] <jeremyagost> if PEF_CONTINUE is set then when the Projectile hits ChangePhase it'll turn into a continuer
[18:43:37] <fuzzie> or at least, original is using two projectiles in some cases, it's also creating the Bounce one directly..
[18:44:07] <fuzzie> brad_a: btw there is a CProjectileLightningBoltGround which also auto-converts into a CProjectileLightningBounce :)
[18:44:19] <jeremyagost> if we're spawning a second projectile for the purpose of bouncing, then there still needs to be a flag in the original to identify that
[18:44:38] <fuzzie> jeremyagost: i was under the impression there some some existing functionality for that
[18:44:50] <jeremyagost> not as far as I can see here
[18:45:12] <jeremyagost> there's the PEF_BOUNCE define
[18:45:16] <jeremyagost> and the PEF_CONTINUE
[18:45:20] <fuzzie> i think it's SecondaryTarget
[18:45:32] <jeremyagost> that makes no sense
[18:45:43] <jeremyagost> "//area projectiles (circles, cones)"
[18:45:59] <jeremyagost> "//secondary projectiles target all in the explosion radius"
[18:46:09] <jeremyagost> that doesn't make sense for lightning
[18:46:20] <jeremyagost> that's fireball stuff
[18:47:02] <jeremyagost> and the lightning projectile doesn't have an Extension, though from what you're saying that is defined by the GemRB overrides?
[18:47:33] <lynxlynxlynx> yes
[18:47:49] <lynxlynxlynx> you can use dltcep to explore them wholly
[18:48:08] <fuzzie> yes, 'you must use dltcep' is perhaps good advice to go first here
[18:48:46] <jeremyagost> well where are those defined then?
[18:48:46] <fuzzie> but this code is not my friend, ever
[18:49:05] <jeremyagost> gemprjtl.ids?
[18:49:14] <fuzzie> jeremyagost: gemprjtl.ids maps projectiles onto .pro files
[18:49:25] <fuzzie> then we override the .pro files, and they have the extension data etc
[18:49:37] <fuzzie> are the overridden ones the ones you look at?
[18:49:58] <fuzzie> none of this should be in original engine projectile files i guess
[18:49:59] <jeremyagost> lightb.pro?
[18:50:06] <jeremyagost> lightbnb.pro?
[18:50:08] <fuzzie> lightb.pro is bouncing one, lightbnb.pro is non-bouncing one
[18:50:10] <fuzzie> apparently
[18:50:12] <jeremyagost> those are in override/bg2
[18:50:30] <fuzzie> yes
[18:50:39] <fuzzie> obviously they're just made up by Avenger though..
[18:50:51] <jeremyagost> not based on the engine files?
[18:50:55] <fuzzie> no
[18:51:00] <fuzzie> this is all hardcoded in original
[18:51:13] <jeremyagost> gotcha
[18:51:20] <fuzzie> so if you think you need a flag or something then maybe you do
[18:51:27] <fuzzie> hrmph
[18:51:38] <jeremyagost> but then why does each version have their own lightning .pros
[18:52:23] <fuzzie> because hardcoding differs per-game :)
[18:52:42] <jeremyagost> ah hell
[18:52:47] <jeremyagost> NI needs a complete game, right...
[18:52:55] <fuzzie> NI doesn't understand any of our flags
[18:53:16] <fuzzie> see, isn't this heaps of fun?
[18:53:21] <jeremyagost> wahh
[18:53:21] <fuzzie> i would recommend just continuing as you are
[18:53:24] <fuzzie> as long as it's not too invasive
[18:53:41] <fuzzie> code which mostly works is much better than code that doesn't exist
[18:53:54] <brad_a> yes :)
[18:53:58] <fuzzie> as long as it's not too painful to fix it up later when someone knows what they're doing
[18:53:59] <brad_a> just dont give up
[18:54:04] <jeremyagost> well it means that anything with both PEF_CONTINUE and PEF_BOUNCE set will behave like the lightning bolt
[18:54:09] <jeremyagost> which I guess is a good thing :D
[18:54:09] <fuzzie> right
[18:54:11] <fuzzie> yes
[19:00:06] <Yoshimo> i'm quite confused by all this lightning bolt talk^^
[19:03:57] <jeremyagost> Orientation = (Orientation + 8) &15;
[19:04:00] <jeremyagost> dangit
[19:04:02] <jeremyagost> Orientation = (Orientation + 8 ) &15;
[19:04:10] <jeremyagost> can anyone tell me what that does?
[19:04:24] <lynxlynxlynx> there are 16 orientations
[19:04:32] <fuzzie> adds 8 to orientation and then wraps it to stay inside 0-15
[19:04:41] <fuzzie> because, what lynx said :)
[19:04:41] <jeremyagost> oh gotcha
[19:04:45] <lynxlynxlynx> so it looks like mirroring of some kind
[19:04:49] <jeremyagost> yeah
[19:04:56] <jeremyagost> that makes some sense
[19:05:26] <jeremyagost> those are animation orientations right?
[19:05:35] <fuzzie> yes
[19:05:56] <jeremyagost> clockwise?
[19:06:41] <fuzzie> i always forget. i always forget where it starts from too.
[19:07:39] <fuzzie> ah right, you are looking at GL_REBOUND
[19:07:49] <jeremyagost> yeah
[19:07:54] <jeremyagost> that's a 180-degree rotation
[19:07:59] <fuzzie> be warned that has probably never been tested
[19:08:00] <jeremyagost> which isn't what we want
[19:08:18] <jeremyagost> I rewrote it
[19:08:20] <fuzzie> i think it's just there as a 'stub' because we didn't add actual rebound for the lightning bolt yet
[19:08:22] <jeremyagost> the movement stuff at least
[19:08:32] <jeremyagost> but I wanted to deal with the animation now
[19:08:51] <fuzzie> ok. bear in mind you really shouldn't change anything other than that single case of course :-)
[19:09:16] <fuzzie> but orientations start south and go counter-clockwise, it seems.
[19:09:26] <jeremyagost> oh ok
[19:09:27] <jeremyagost> >.<
[19:09:45] <fuzzie> i figured this would be documented soemwhere, hrm
[19:09:59] <jeremyagost> I had to make a few changes to the GetLine function
[19:10:05] <jeremyagost> but functionally it's the same
[19:10:05] <brad_a> oh btw jeremy, feel free to add documentation to the wiki ;-)
[19:10:26] <jeremyagost> ah wikis, how I hate thee
[19:10:38] <jeremyagost> yet another @#$%ing markup language to learn...
[19:10:58] <brad_a> well ours has an RTF edirot
[19:10:59] <fuzzie> ah in fact, if i grep for 'clockwise': "orientations go counter-clockwise, starting south", in Projectile.cpp
[19:11:22] <jeremyagost> ah there we go
[19:11:34] <jeremyagost> well here goes nothing
[19:11:59] <fuzzie> although i'm not sure you should trust our comments any more than me
[19:12:04] <fuzzie> but i guess you will find out!
[19:12:31] <lynxlynxlynx> you don't need to use markup if you don't want to
[19:12:31] <lynxlynxlynx> the animation thing is documented in one comment, i think in charanimations.cpp (with an ascii chart)
[19:13:02] <fuzzie> you're right, at the end
[19:13:10] <fuzzie> i read half way through the block comment and gave up
[19:17:27] <jeremyagost> have you guys noticed Unrecognized SDL event type 303 when you press any buttons
[19:18:05] <brad_a> jeremy: ignore that
[19:18:29] <brad_a> unrecognized event just means its an event type we dont handle/care about
[19:19:06] <brad_a> tho i am curious what event type that is if its on every keystroke....
[19:19:29] <brad_a> i should wrap that in a debug macro
[19:19:37] <jeremyagost> I looked in SDL_event.h but couldn't figure it out
[19:19:55] <brad_a> i will wrap it in a debug macro
[19:20:04] <brad_a> i still use it
[19:20:07] <jeremyagost> unless
[19:20:20] <jeremyagost> I bet it's SDL_TEXTINPUT
[19:20:20] <brad_a> but it is useless for a release build
[19:20:24] <brad_a> oh probably
[19:20:25] <jeremyagost> 0x303
[19:25:06] <fuzzie> the idea being that you might be typing in some way that doesn't have keypresses for text
[19:25:12] <fuzzie> like a mobile keypad and T9 or whatever
[19:25:29] <fuzzie> it's kinda clever, but i hope irrelevant for us
[19:27:13] <brad_a> yes
[19:27:59] <brad_a> i should see if the ios soft keyboard is sending keyevents… if not then it may be usefull after all
[19:28:28] <jeremyagost> how do you bring up the console in GemRB?
[19:28:34] <brad_a> ctrl+space
[19:29:41] <jeremyagost> cheat keys aren't implemented?
[19:29:46] <brad_a> yes
[19:29:52] <brad_a> i think you have to enable in config
[19:29:53] <jeremyagost> is there a way to get the x/y at cursor?
[19:30:02] <brad_a> should be just the x key
[19:30:09] <brad_a> without console
[19:30:26] <brad_a> but like i said you need to enable cheat keys i think
[19:30:30] <jeremyagost> uhhhh wtf
[19:30:39] <jeremyagost> shift key isn't doing anything
[19:30:47] <jeremyagost> I type shift + 9 I get a 9 not a (
[19:30:54] <brad_a> in the console?
[19:30:56] <jeremyagost> yup
[19:31:01] <fuzzie> i guess maybe you do need to handle SDL_TEXTINPUT :-p
[19:31:09] <brad_a> lol
[19:31:11] <fuzzie> did someone sabotage the unicode handling appropriately for 1.3?
[19:31:23] <brad_a> nope
[19:31:47] <brad_a> the only thing using 1.3 is ios and its keyboard is not standard
[19:31:57] <brad_a> so it didnt need any hacking
[19:32:14] <jeremyagost> anyone know how to type a ( without shift?
[19:32:24] <jeremyagost> and a ) and : for that matter
[19:32:30] <brad_a> not in a way that gemrb would recognize
[19:32:31] <jeremyagost> or better yet enable cheat keys without the console command
[19:32:38] <fuzzie> hm, i misremember, not unicode handling which is relevant here. nm about that.
[19:32:40] <brad_a> use the config file
[19:33:19] <jeremyagost> oh it is enabled
[19:33:20] <brad_a> but yes maybe we should implement that SDL event
[19:33:36] <brad_a> i probably need to for iOS anyway
[19:33:56] <brad_a> because i doubt that hitting a on the soft keyboard will send an sdl key event
[19:34:01] <brad_a> but we shall see
[19:34:26] <jeremyagost> so it seems projectile coordinates don't correspond to area coordinates
[19:34:28] <jeremyagost> good to know
[19:34:43] <brad_a> well you may want to be sure what coordinate system is in
[19:34:58] <brad_a> i recall it being wrong but i cant remember if it got fixed
[19:35:06] <brad_a> ^x
[19:35:21] <fuzzie> jeremyagost: the projectile coordinates should be searchmap coordinates?
[19:35:32] <jeremyagost> maybe?
[19:35:36] <jeremyagost> ok
[19:35:43] <fuzzie> you should be able to pass them directly to GetLine
[19:35:52] <fuzzie> and also use them directly to access the searchmap
[19:36:32] <jeremyagost> yeah it looks like that by the time ChangePhase is hit, the Origin and Destination members of Projectile are meaningless
[19:36:58] <fuzzie> but i guess you need to know that searchmap squares are 16 pixels across and 14 pixels high if you want to poke at actors
[19:37:34] <fuzzie> oh, really?
[19:37:39] <jeremyagost> wait no
[19:37:40] <fuzzie> is ClearPath sabotaging them?
[19:37:41] <jeremyagost> those are ok
[19:37:43] <jeremyagost> Pos is
[19:37:45] <fuzzie> it really shouldn't
[19:37:56] <fuzzie> that should also be fine :-/
[19:38:01] <jeremyagost> but but...
[19:38:02] <jeremyagost> rrr
[19:38:07] <jeremyagost> confusion
[19:38:22] <jeremyagost> I've gotta take a break from this
[19:38:27] <jeremyagost> probably made a stupid mistake
[19:38:55] <fuzzie> you already sound exactly like me when I'm trying to do this kind of thing, very reassuring :-)
[19:39:05] <brad_a> i think we all do that.
[19:39:12] <fuzzie> lots of "oh this is broken .. oh wait .. oh argh .. ok, coffee time" moments
[19:39:21] <fuzzie> as typed into IRC for maximum looking stupid value
[19:39:25] <brad_a> im out of coffee :(
[19:39:43] <Yoshimo> how can you let that happen?
[19:40:12] <brad_a> i dont have a coffee pot at work so i only get what i bring
[19:40:34] <brad_a> 0,75 l is not enough
[19:41:12] <fuzzie> that is a mean, mean workplace.
[19:41:40] <tomprince> jeremyagost: We also have a tiny bit of doxygen documentation.
[19:41:48] <tomprince> not *at all
[19:41:56] <tomprince> * complete though
[19:42:57] <Yoshimo> me as trainee in our office has made the observation, the most important task is : having enough water and coffee, otherwise nobody can work^^
[19:48:39] <lynxlynxlynx> chocolate > coffee
[19:49:17] <brad_a> you can have chocolate coffee :)
[19:49:33] <fuzzie> chocolate unfortunately a bit distressingly calorie-laden for an office
[20:01:07] <lynxlynxlynx> well don't get white or milk, but dark chocolate
[20:01:17] <lynxlynxlynx> much lower on sugars
[20:01:31] <fuzzie> i thought it was still pretty bad
[20:02:00] <fuzzie> perhaps you're not quite so bothered though, with all your adventuring around to burn them off :)
[20:11:53] --> SiENcE has joined #gemrb
[20:31:44] <lynxlynxlynx> even without that it doesn't stick
[21:02:22] <Yoshimo> there are people that only have to see the sweets and turn into an elephant, and some can eat and eat and eat and still dont get fat
[21:06:55] <-- jeremyagost has left IRC (Quit: jeremyagost)
[21:08:06] <tomprince> And some of us don't generally like them anyway.
[21:08:49] <brad_a> ^ this
[22:07:40] <lynxlynxlynx> uhh, bg1 is boring to run through
[22:08:38] <lynxlynxlynx> and playing for more than a few minutes makes that 'boring' bug so much more annoying
[22:08:53] <lynxlynxlynx> good for motivation
[22:09:56] <brad_a> we could just let yoshimo do it :)
[22:10:07] <brad_a> "yoshimo is willing"
[22:13:49] <Yoshimo> it is annoying to the point where i just disabled sound lynx
[22:14:43] <lynxlynxlynx> i'll make it reset on any party action
[22:14:48] <lynxlynxlynx> now it is all individual
[22:15:04] <lynxlynxlynx> the roll ranges are big too, but i think that's straight from RE
[22:15:24] <Yoshimo> i mean i can run around with a thief and disarm traps and the whole party is bored so much that their "bored sounds" overlap each other ;)
[22:28:06] <-- Maighstir has left IRC (Quit: .)
[22:33:51] --> brad_a_ has joined #gemrb
[22:33:51] <-- brad_a_ has left IRC (Client Quit)
[22:37:18] <-- brad_a has left IRC (Ping timeout: 258 seconds)
[22:41:25] --> edheldil_ has joined #gemrb
[22:45:52] <-- edheldil_ has left IRC (Ping timeout: 244 seconds)
[22:48:49] <CIA-44> GemRB: 03lynxlupodian * rdea6cac31a5c 10gemrb/gemrb/core/Scriptable/Actor.cpp: don't trigger the bored comment when moving
[22:48:50] <CIA-44> GemRB: 03lynxlupodian * r2fab81bdcc51 10gemrb/gemrb/core/Scriptable/ (Actor.cpp Actor.h Scriptable.cpp):
[22:48:50] <CIA-44> GemRB: actor: added ResetCommentTime as a shortcircuit shortcut and a few
[22:48:50] <CIA-44> GemRB: extra invocations to try to keep the spam down
[22:50:57] <Yoshimo> is this "bored" problem in the original engine as well or just a problem of gemrb?
[22:58:55] <lynxlynxlynx> we just don't know how exactly it works
[23:01:20] <Yoshimo> ah ok
[23:07:59] <Yoshimo> all i see is a baldur.ini entry Bored Timeout=3000
[23:28:01] <lynxlynxlynx> if you set it to 0, the comments will be completely disabled
[23:29:33] <-- lynxlynxlynx has left IRC (Remote host closed the connection)
[23:30:49] <Yoshimo> there are 3 functions that include the string bored, but they are totally weird, so not much help from them
[23:34:39] <-- SiENcE has left IRC (Quit: cya)
[23:36:29] <-- Yoshimo has left IRC (Quit: Yoshimo)