#gemrb@irc.freenode.net logs for 5 Jul 2009 (GMT)

Archive Today Yesterday Tomorrow
GemRB homepage


[01:55:24] <-- Gekz has left IRC (Read error: 104 (Connection reset by peer))
[01:55:27] --> Gekz has joined #GemRB
[03:23:08] <-- Gekz has left IRC (Read error: 110 (Connection timed out))
[06:28:23] <-- Lightkey has left IRC (Read error: 104 (Connection reset by peer))
[06:29:16] --> Lightkey has joined #GemRB
[08:35:02] --> lynxlynxlynx has joined #gemrb
[08:35:02] --- ChanServ gives channel operator status to lynxlynxlynx
[08:46:56] <Edheldil> Sourceforge's download page is STILL broken, as far as I can tell
[09:32:35] <-- Edheldil has left IRC (Remote closed the connection)
[10:02:47] --> n00bystance1 has joined #GemRB
[10:03:42] <n00bystance1> awesome screenshots, huh?
[10:08:42] --> Avenger has joined #gemrb
[10:09:02] --- ChanServ gives channel operator status to Avenger
[10:17:00] <n00bystance1> do you have any requirements for these screenshots or just they have to be awesome and that's all? :d
[10:18:35] <wjp> I think being awesome is the most important thing :-)
[10:18:55] <fuzzie> I think really 'awesome' is our only requirement :) Our current ones are old and too boring-looking!
[10:19:22] <Avenger> the screenshots should show gemrb in action too :)
[10:19:38] <n00bystance1> I see ;)
[10:20:26] <Avenger> well, while it is not a requirement, it is suggested that you use some recent version of gemrb
[10:21:49] <n00bystance1> I'll try playing with it today maybe :)
[10:21:57] <fuzzie> I keep meaning to make a forum thread sometime and mention which bits might be particularly impressive, but I didn't find the time.
[10:40:11] <-- pupnik has left IRC (Read error: 101 (Network is unreachable))
[10:44:09] <-- n00bystance1 has left IRC ("Leaving.")
[10:45:41] --> pupnik has joined #gemrb
[10:52:42] <fuzzie> also, good morning all :)
[10:53:00] <pupnik> moin fuzzie
[10:55:42] <pupnik> i lost my new camera
[10:55:49] <pupnik> no more youtubes
[10:55:54] <pupnik> of gemrb
[11:03:49] <fuzzie> oh dear :(
[12:16:50] --> Gekz has joined #GemRB
[12:25:13] <-- Gekz has left IRC (Remote closed the connection)
[12:34:36] --> Gekz has joined #GemRB
[12:41:02] --> Gekz__ has joined #GemRB
[12:41:03] <-- Gekz__ has left IRC (Remote closed the connection)
[12:52:19] <-- Gekz has left IRC (Read error: 104 (Connection reset by peer))
[12:53:51] --> Gekz has joined #GemRB
[13:58:53] <-- dawid has left IRC (Read error: 110 (Connection timed out))
[14:01:19] <-- Avenger has left IRC ("ChatZilla 0.9.85 [Firefox 3.5/20090624025744]")
[14:04:00] --> dawid has joined #GemRb
[14:16:02] <CIA-19> gemrb: 03avenger_teambg * r6702 10/gemrb/trunk/gemrb/override/how/ (19 files): new projectiles
[14:29:53] --> Avenger has joined #gemrb
[14:29:59] --- ChanServ gives channel operator status to Avenger
[14:30:40] <Avenger> so fuzzie, do you want to implement some kind of reverb effects?
[14:30:54] <fuzzie> well, it is possible :) i don't volunteer
[14:31:06] <Avenger> ahh, i thought you do :P
[14:31:28] <wjp> with the new openal-soft it should be possible
[14:32:22] <fuzzie> yes, i just posted on the forum mentioning that; the old reference implementation seems to do AL_EFFECT_REVERB, even, but perhaps that is a bit too limited
[14:32:24] <Avenger> i can cover the engine part, i know, or i can find out how the various engines try to set up eax. But i cannot do the openal part
[14:33:13] <fuzzie> it's very trivial to apply reverb in openal if you know the EAX parameters
[14:33:25] <Avenger> hmm, i might know them
[14:33:46] <fuzzie> But I never had EAX, so I have no idea how it's meant to sound.
[14:33:50] <Avenger> at least i saw files covering these parameters in all engines
[14:34:11] <Avenger> i just don't know how to supply the parameters to openal, and what it expects
[14:34:54] <fuzzie> there's an AL_EFFECT_EAXREVERB effect in the effects framework, and i think it just takes an EAX parameters struct, but it's been a while since i looked
[14:35:34] <wjp> the number of parameters is rather huge
[14:35:53] <fuzzie> i think a lot of them are only supported by their modern EAX hardware anyway
[14:36:15] <fuzzie> so definitely wouldn't be used by BG-era games
[14:36:35] <Avenger> there are a few 2da and ids files dealing with this
[14:36:46] <Avenger> for example: sndchann.2da
[14:37:14] <Lightkey> can EAX thingamabobs be used when there is no EAX support on Linux/Mac?
[14:37:16] <Avenger> and reverb.2da
[14:37:19] <fuzzie> hm iesdp helpfully doesn't have sndenvrn.2da which sounds most likely :)
[14:37:36] <Avenger> that is in bg2
[14:37:55] <Avenger> luckily i know how it works
[14:37:55] <fuzzie> Lightkey: i think it's a lost cause on OS X, but there's a limited software version on Win/Linux
[14:38:43] <Avenger> fuzzie sndresrf.2da is also needed
[14:38:51] <Lightkey> good to know, is that new in OpenAL Soft?
[14:39:05] <wjp> looks like we'd need a fairly recent version of openal-soft
[14:39:14] <fuzzie> Lightkey: the original reference implementation seems to have a limited 'reverb' option which maybe could be a fallback..
[14:39:24] <wjp> maybe it should be compiled conditionally
[14:39:29] <Avenger> well, we can always fall back to limited or no eax
[14:39:36] <Avenger> even the gui has an 'eax' button
[14:39:48] <fuzzie> wjp: the effects call will just error out if the support is missing
[14:40:02] <Avenger> we won't even call them if there is no eax
[14:40:04] <wjp> oh, the relevant structs were already declared earlier?
[14:40:08] <Avenger> i'm sure it could be queried
[14:40:15] <fuzzie> Avenger: you query it by trying the effect call :)
[14:40:30] <fuzzie> wjp: i don't know, but we could always ship the struct in a header if we must
[14:42:37] <fuzzie> i mean, i have no objection to compile-time detection, it's just maybe not necessary
[15:04:50] <-- Avenger has left IRC ("bye!")
[15:55:48] --> Canageek has joined #GemRB
[15:56:19] <Canageek> Hey, how playable is GemRB right now with BG1 and its expansion?
[15:56:33] <Canageek> I'm playing with TuTu (All 3 with the BGII engine)
[15:56:59] <Canageek> but its got REALLY long load times (Over a minute sometimes, worse then back on my PIII)
[15:57:59] <fuzzie> GemRB isn't playable enough, still.
[15:58:20] <pupnik> it is fun to play *with* though :)
[15:58:30] <fuzzie> Did you try the standard recommendations like the bigg's Generalized Biffing?
[15:59:05] <Canageek> No, this is just something my Dad set up on his comp, its too addictive so I've not tossed it onto my laptop yet
[15:59:15] <Canageek> we both fell in love with BG when it came out 10 years ago
[15:59:19] <fuzzie> ah :)
[15:59:40] <Canageek> and I introduced him to Bittorrent and our old disks were scratched sooooo
[15:59:48] <Canageek> one thing lead to another
[15:59:58] <fuzzie> Well, it is worth looking into - it repacks the override into a faster archive file which helps a lot, although then you can't add/remove mods without reversing the process.
[15:59:59] --> Avenger has joined #gemrb
[16:00:05] --- ChanServ gives channel operator status to Avenger
[16:00:24] <Canageek> What is GemRB coded in? C++?
[16:00:36] <Avenger> c++ and Python
[16:00:37] <fuzzie> The core engine is C++, the GUI/game rules scripts are in python.
[16:00:43] <Canageek> ohhh
[16:00:53] <Canageek> Might have a look into helping with the rules scripts then
[16:01:02] <Canageek> Once I've got some time
[16:01:07] <Avenger> got iwd2?
[16:01:24] <Canageek> I'm doing python coding at a Quantum Chemistry volenteer position right now
[16:01:34] <Canageek> I don't *think* so
[16:01:45] <Canageek> BGI, BGII, IWD
[16:02:07] <Avenger> well, iwd lacks the level up scripts too
[16:02:21] <Avenger> and a lot of the code from bg1/bg2 could be used
[16:02:27] <Avenger> so it is good practice :)
[16:02:49] <fuzzie> I think our bg1 still lacks level up. :/
[16:02:58] <Avenger> a very easy thing to do in iwd: character import
[16:03:02] <Canageek> So is this meant to replicate the original games exactly or are improvments from the newer games going to be around in the older ones
[16:03:20] <fuzzie> The idea is that you can pick whichever features you want :)
[16:03:25] <Canageek> Nice
[16:03:29] <-- lynxlynxlynx has left IRC (Remote closed the connection)
[16:03:37] <Avenger> we try to replicate, but some small beneficial improvements are working for older engines
[16:03:48] <Avenger> for example moving around the game screen by arrow keys
[16:04:08] <Avenger> we simply couldn't take the efforts to disable that good feature in bg/iwd, for example :)
[16:04:16] <Canageek> So, are you coding the spells in BGI using the BGII versions or the BGI versions?
[16:04:16] <fuzzie> Heh, that drives me completely crazy when using original bg1, now.
[16:04:19] <fuzzie> Although gemrb doesn't handle double-clicking so that doesn't help either..
[16:04:42] <Avenger> the spells (i guess you meant the projectiles) are bg1 compatible for bg1
[16:04:58] <Avenger> bg1 has not the same graphical resources
[16:05:07] <Avenger> but we are capable of running bgtutu, of course
[16:05:13] <Canageek> No, spells in BGII were updated to have diffrent durations
[16:05:23] <Canageek> and such rules wise
[16:05:23] <fuzzie> the durations are specified in the game data
[16:05:24] <Avenger> we use the original game data
[16:05:32] <Canageek> ahhh
[16:05:43] <Canageek> SO you need the original game to do anything
[16:05:44] <Avenger> but you can always mod them or copy the spells from bg2
[16:05:47] <Avenger> yes
[16:05:50] <Canageek> I think I still have my old disks
[16:05:56] <Avenger> we don't copy the original data
[16:06:05] <Canageek> Well, there is IWD anyway
[16:06:48] <Avenger> i recently repurchased the complete iwd set, because my old cd's were ruined
[16:06:50] <pupnik> the castle aaaargh. Our quest, is at an end.
[16:06:55] <Avenger> they are still being sold
[16:07:04] <Canageek> I don't think we ever got IWD
[16:07:18] <Canageek> II, we were much less impressed with its linearness and lack of plot
[16:07:31] <pupnik> iwd2 needs the most script work, right?
[16:07:37] <fuzzie> yes, you can get an iwd/how/iwd2 set very cheaply, but iwd2 is pretty linear :)
[16:07:41] <Avenger> well, i liked iwd2 the first, but the end was a disappointment
[16:07:59] <fuzzie> pupnik: it needs the most everything, really..
[16:08:19] <Canageek> I've never finished BGI, going to try that this summer
[16:08:20] <fuzzie> last time i tried playing it i found crashes within the first few minutes, although those are all fixed now
[16:08:35] <Canageek> I was REALLY close, just Candlekeep and Durlags tower
[16:08:40] <fuzzie> but that gives you an idea how much love it has received :)
[16:08:51] <Canageek> when that comp died, and we never got it running on XP
[16:09:25] <fuzzie> well, bgtutu and BGT are not so awful for playing it
[16:09:39] <Canageek> I've got tutu, what is the other one?
[16:09:46] <fuzzie> they could both be better
[16:10:26] <Avenger> bgt is just another bg1 in bg2, no?
[16:10:47] <Avenger> i have to admit i never tried either of those ;)
[16:10:59] <fuzzie> BGT is the "Baldur's Gate Trilogy"
[16:11:13] <fuzzie> well, i get annoyed that neither of them admit that they have missing things
[16:11:24] <fuzzie> they both act like their conversions make everything better
[16:11:25] <Avenger> ahh and bgtutu is just bg1 in bg2 engine?
[16:11:58] <fuzzie> Avenger: i think they're just two different ways of doing the same thing
[16:12:28] <Avenger> ahh i forgot why i came here
[16:12:38] <Avenger> fuzzie, wjp. some problem with wavc sounds.
[16:13:04] <Avenger> i noticed that my projectiles play some sounds and don't play some others
[16:13:13] <Avenger> for example, tra_06 isn't played
[16:13:16] <Avenger> i don't know why
[16:13:36] <Avenger> dltcep plays it
[16:14:32] <fuzzie> oh, this is the horrible non-gemrb code :)
[16:14:32] <Avenger> rng_m03 is played
[16:14:52] <Avenger> well, dltcep has the same horrible code :)
[16:14:59] <Avenger> or some iteration of it
[16:15:29] <Avenger> the difference is that dltcep's code wasnt' touched by others :P
[16:15:43] <fuzzie> well, i guess dltcep's code also only works for dltcep :)
[16:15:57] <Avenger> well, yes
[16:16:01] <fuzzie> is there a simple way to play these sounds from within gemrb?
[16:16:19] <Avenger> GemRB.PlaySound("file") ?
[16:16:22] <Avenger> or something like that
[16:16:27] <Avenger> in the console
[16:16:29] <fuzzie> and this is bg2?
[16:16:32] <Avenger> HoW
[16:16:49] <fuzzie> i don't have that installed, so that is inconvenient :)
[16:17:08] <fuzzie> i'll have to install it later if wjp doesn't get there first
[16:17:46] <Avenger> ok, i will try to find similar sounds in bg2
[16:18:14] <fuzzie> bg2's tra_06 plays fine here, not sure how to search for others
[16:19:04] <Avenger> odd, now i got no sound at all
[16:19:40] <Avenger> the original game still has sound, and the plugin order is correct
[16:19:57] --> D_T_G has joined #gemrb
[16:20:12] <fuzzie> "Starting up the sound Driver..." has no errors?
[16:20:20] <Avenger> heh, cespenar has sound
[16:20:43] <fuzzie> if you're on windows then you could have a bad openal driver, some sound cards have them
[16:21:27] <fuzzie> for another project i ended up writing code to blacklist non-Creative drivers
[16:22:00] <Avenger> ok, every xth sound plays
[16:27:47] <Avenger> i don't think this is openal specific
[16:28:02] <Avenger> something with the sound position, i think
[16:28:18] <Avenger> the odd is, some sounds are working almost always, some are not
[16:28:23] <Avenger> i cannot find the rule
[16:28:55] <Avenger> but even if a sound doesn't work, if i move the game area, i still catch part of the sound
[16:29:32] <wjp> interesting
[16:29:52] <Avenger> tra_06 on bg2 (at least now) always gets played
[16:29:56] <Avenger> but tra_01 doesn't
[16:30:02] <Avenger> it is very annoying
[16:30:46] <Avenger> i can trigger the sound playing by simply moving the mouse a bit, to move the game area, as soon as it moves, the sound activates
[16:31:17] <wjp> regardless of direction?
[16:31:25] <Avenger> seems so
[16:31:51] <wjp> so something might be messing up the current listener location?
[16:31:56] <Avenger> yes
[16:32:05] <Avenger> that's a good explanation
[16:32:26] <Avenger> btw, double clicking on an actor did center the screen on it, no?
[16:32:37] <wjp> I've seen that work
[16:32:40] <Avenger> or just pressing '1' twice
[16:32:41] <fuzzie> double-clicking anywhere centers the screen
[16:32:47] <Avenger> not for me now
[16:32:57] <fuzzie> oh, you mean in gemrb
[16:33:00] <Avenger> gemrb
[16:33:07] <Avenger> i saw this working before
[16:33:08] <fuzzie> last time i tried it didn't work for me in gemrb
[16:33:20] <fuzzie> well
[16:33:21] <Avenger> the selection works, but not the centering
[16:33:28] <fuzzie> if you are not paused it works, because the selection code sets th flag
[16:33:38] <Avenger> oh s..t
[16:33:39] <Avenger> yes
[16:33:42] <Avenger> pause was the problem
[16:34:14] <Avenger> i think it should still center, even in pause
[16:34:30] <fuzzie> yes, it should
[16:34:36] <Avenger> ok, lemme test more sounds :)
[16:34:41] <fuzzie> but i am nervous about side-effects if i move the flag, didn't test it yet :)
[16:36:32] <Avenger> moving the viewport in any direction does activate the sound. (More precisely, i catch the end of it), the sound is played, i just don't hear it before moving
[16:39:22] <Avenger> btw, the sound stuff doesn't seem to be relative
[16:39:36] <Avenger> when i move it completely offscreen, and far away, i hear the sound just as well
[16:40:12] <Avenger> even when activated by projectiles, those call PlaySound with coordinates, and the relative flag
[17:11:38] <D_T_G> hi
[17:12:39] <D_T_G> is this normal that all those *.pro files Avenger uploads to svn trunk they don't land in installation directory of gemrb?
[17:13:06] <fuzzie> hm, no :)
[17:13:07] <Avenger> hmm,
[17:13:19] <fuzzie> someone needs to add fix Makefile.am in all those directories
[17:13:26] <Avenger> yeah
[17:13:36] <Avenger> not just .pro, there is a .spl too
[17:13:48] <fuzzie> same for the base CMakeLists.txt too
[17:13:55] <fuzzie> *.spl is already in both lists
[17:14:17] <D_T_G> I'm used to autogen.sh
[17:14:20] <Avenger> i'm fightning with openal windows right now
[17:14:22] <D_T_G> not cmake
[17:14:31] <fuzzie> Avenger: you tried openal soft?
[17:14:42] <fuzzie> D_T_G: yes, but fixing one and not the other would be silly :)
[17:14:52] <Avenger> i couldn't compile it first, i think
[17:15:10] <fuzzie> you can do it if you want, change override/bg2/Makefile.am to include *.pro as well (and all the other override subdirectories)
[17:15:23] <fuzzie> Avenger: hm, i thought there was a drop-in binary
[17:15:26] <D_T_G> but for now one needs to copy it manually?
[17:15:40] <fuzzie> well, i suggest you fix it and provide a patch :-)
[17:15:53] <fuzzie> i don't ever do an install so i don't test
[17:15:55] <D_T_G> I have no ieda about makefiles
[17:15:59] <D_T_G> I would broke it :P
[17:16:04] <fuzzie> but if you look at the Makefile.am then it's very clear
[17:17:38] <D_T_G> well it looks so :)
[17:17:40] <fuzzie> although i don't have any idea about automake so it's quite possible that adding another wildcard would cause some kind of large nuclear explosion
[17:18:47] <fuzzie> i should really fix gemrb saving sometimes
[17:19:18] <D_T_G> yeah, gemrb do not see all my saves :/
[17:19:41] <fuzzie> it drives me a bit mad when it just inserts savegames into the middle of the list where they fit the numbering
[17:19:50] <fuzzie> it should see all your original saves, i think
[17:19:57] <fuzzie> but i don't use the original saves much, i just make new ones
[17:21:28] <D_T_G> It does not see my 2 tob saves
[17:21:40] <D_T_G> in tob mode of course
[17:22:28] <Avenger> wjp the listener position isn't corrupted
[17:22:51] <Avenger> but it seems openal forgets to play it, until there is an update
[17:22:54] <Avenger> it is very odd
[17:23:13] <Avenger> i will try openal soft
[17:23:44] <fuzzie> it looks like a 'realtek HD' driver is the one i found buggy
[17:24:02] <fuzzie> people kept on reporting that their sound was very faint and in the wrong place
[17:24:17] <fuzzie> but i never found a machine with it to test
[17:25:13] <fuzzie> what is SPIN876 in bg2?
[17:26:10] --- D_T_G is now known as D_T_G_away
[17:28:48] <Avenger> i just looked at it, it seems to change a creauture into another
[17:29:04] <fuzzie> for some reason it just changed this creature into about 20 other creatures
[17:29:15] <fuzzie> i guess i'll have to look in DLTCEP myself later
[17:30:04] <Canageek> Avenger: As if for polymorph spells or druids shapechange?
[17:30:37] <fuzzie> well, in this case it's for a werewolf transformation
[17:31:07] <Avenger> named ANATH_WEREWOLF in scripts
[17:31:11] <fuzzie> specifically ANATH_WEREWOLF
[17:31:32] <Avenger> it could be bg1
[17:32:33] <Avenger> hmm no , it is not in bg1 spell.ids
[17:33:04] <Avenger> i will need a spell searcher which looks for spell ids too, not just resrefs
[17:34:57] <fuzzie> it's bg2
[17:35:02] <fuzzie> i mean, it is used in a script
[17:35:10] <Avenger> the ones in firkraag's dungeon?
[17:35:12] <fuzzie> but it's only meant to spawn one creature
[17:35:23] <fuzzie> but it spawned a whole area full
[17:35:31] <Avenger> it is in rngwlf01
[17:35:33] <Avenger> oh
[17:35:45] <Avenger> but it should destroy the creature it is applied on
[17:36:11] <Avenger> the script runs again and again?
[17:37:16] <Avenger> it calls target->destroyself only if parameter2 is 0
[17:37:23] <Avenger> otherwise it calls target->die
[17:37:28] <fuzzie> ah
[17:37:46] <Avenger> this is important, because of xp
[17:38:06] <fuzzie> ok, i'll look at scripting
[17:39:12] <Avenger> hmm, in the original it explodes with chunky death?
[17:39:36] <Avenger> the parameter is 1
[17:39:47] <Avenger> which means it calls ->Die instead of ->DestroySelf
[17:40:57] <fuzzie> yes, this is a missing scripting feature
[17:41:13] <fuzzie> it should stop running scripts unless a Died() trigger is encountered
[17:41:30] <fuzzie> but i didn't see code for that anywhere, so it is perhaps missing..
[17:41:44] <Avenger> i never thought about it
[17:41:53] <Avenger> so, that's how the original does it?
[17:42:07] <Avenger> only blocks with Died() in it will run?
[17:42:56] <fuzzie> some forum posts say so
[17:43:00] <fuzzie> i didn't check yet
[17:43:21] <fuzzie> oh, i guess the trigger is Die()?
[17:43:29] <-- Gekz has left IRC (Read error: 104 (Connection reset by peer))
[17:43:34] <fuzzie> i don't remember and i must go now
[17:43:35] <Avenger> well, you said: it should stop scripts unless a died trigger is encountered, that means it will reevaluate the script after every died(), so you basically can keep it alive
[17:43:51] <Avenger> oki, see you later
[17:44:24] <Avenger> die() yes, died() checks the death variable
[17:44:42] <Avenger> err that is dead()
[17:46:33] <Avenger> i wonder if died(myself) is the same as die()
[17:47:24] --> Gekz has joined #GemRB
[17:50:44] <fuzzie> no, i checked that
[17:51:05] <fuzzie> but, yes, we should probably also make sure that after death there is only one chance to run scripts
[17:53:14] <fuzzie> easy to fix but i must test it first
[17:54:08] <-- Gekz has left IRC (Read error: 60 (Operation timed out))
[18:05:36] --> Gekz has joined #GemRB
[18:16:20] <-- Gekz has left IRC (Connection reset by peer)
[18:23:12] --> Gekz has joined #GemRB
[18:42:51] <fuzzie> the conclusion in forum posts seems to be that Die() can only trigger once and a Die() block is always the last block executed (but there can be a Continue() in there)
[18:45:20] <Avenger> back
[18:45:34] <Avenger> i downloaded the openal soft source
[18:46:20] <Avenger> heh bz2
[18:47:30] <fuzzie> the binary is no good?
[18:47:34] <wjp> did you have the problems in windows or linux, by the way?
[18:48:40] <fuzzie> i tried the reverb thing on windows with just their binary
[18:48:57] <Avenger> i thought the binary is useless for compile?
[18:49:04] <Avenger> wjp: windows
[18:49:06] <fuzzie> you don't have to recompile
[18:49:16] <fuzzie> you just take the dll in their binary, rename it to openal32.dll, and put it in the gemrb dir :)
[18:49:38] <Avenger> it is binary compatible with openal32.dll?
[18:49:41] <fuzzie> yes
[18:49:46] <fuzzie> it is just another openal driver
[18:53:30] <Avenger> wow
[18:53:37] <Avenger> all button sounds work
[18:53:43] <Avenger> instead of every 4th
[18:53:46] <fuzzie> hehe
[18:54:07] <Avenger> ok, lets see the how projectile stuff
[18:54:51] <Avenger> works
[18:55:08] <Avenger> ok, now the projectiles sound right
[18:55:26] <Avenger> now just the relative sound stuff is needed to be fixed
[18:56:59] <Avenger> oops
[18:57:05] <Avenger> 0 pointer in startup
[18:57:17] <Avenger> starting up the sound drider
[18:57:19] <Avenger> driver
[18:57:27] <Avenger> *boom*
[18:57:58] <Avenger> if (ambim) ambim->deactivate();
[18:58:05] <Avenger> but ambim is 0xcdcdcdcdcd :)
[19:00:13] <fuzzie> strange, ambim should be set in the AudioDriver Init()
[19:00:53] <Avenger> yes, but init failed
[19:01:01] <Avenger> it should be set to 0 in the constructor
[19:01:32] <fuzzie> well that is an easy fix :)
[19:01:34] <Avenger> for some reason the openal manager didn't manage to init, and this cascaded into a segfault in the bail out code
[19:01:56] <Avenger> yes, added an ambim=NULL; into the constructor
[19:02:08] <Avenger> now i wonder why it didn't init
[19:02:10] <Avenger> once it worked
[19:02:22] <Avenger> now it runs
[19:02:24] <Avenger> meh
[19:03:21] <fuzzie> it would be nice if we printed the errors there, if we can
[19:03:40] <Avenger> again
[19:04:11] <Avenger> crashed again
[19:04:22] <Avenger> but now a bit later
[19:04:58] <Avenger> it should pick null sound if it cannot init openal
[19:05:31] <fuzzie> the OpenALAudioDriver looks like a potential disaster if init failed
[19:05:50] <fuzzie> destructor is going to try handling garbage
[19:06:04] <fuzzie> but it should be printing errors in Init too
[19:06:15] <fuzzie> so it needs some checks and a valgrind run, i guess
[19:06:42] <Avenger> now it dies in the release() of Openalaudio : OpenALAudioDriver::release() line 83 + 30 bytes
[19:06:53] <Avenger> delete this, but it calls into SDL?
[19:06:55] <Avenger> why
[19:07:18] <fuzzie> it needs threading
[19:07:22] <fuzzie> and SDL is the simplest way to get it
[19:07:32] <fuzzie> otherwise you have to write a lot of native code
[19:07:34] --> barra_away has joined #gemrb
[19:08:10] <fuzzie> but as i said, it is a disaster if init failed, the destructor doesn't do any checks
[19:08:38] <Avenger> i don't understand why it fails too
[19:08:47] <Avenger> i never noticed this before
[19:08:51] <Avenger> only with this soft openal
[19:09:06] <fuzzie> add some 'checkALError("", "");' calls before the 'return false;' in Init()
[19:09:36] <fuzzie> well, also before the alcCloseDevice or alcDestroyContext calls, i guess
[19:10:20] <Avenger> ok just checkALError("", ""); at the beginning of each bailout block
[19:10:37] <Avenger> caught it
[19:10:38] <fuzzie> if you were randomly losing all sounds before, i wonder if your sound driver is refusing to give access occasionally
[19:10:53] <Avenger> 40964
[19:11:06] <Avenger> does that tell you anything? or wjp?
[19:11:40] <fuzzie> AL_INVALID_OPERATION
[19:12:08] <fuzzie> or ALC_INVALID_VALUE
[19:12:12] <Avenger> device = alcOpenDevice (NULL);
[19:12:14] <Avenger> if (device == NULL) {
[19:12:15] <Avenger> checkALError("X", "");
[19:12:17] <Avenger> return false;
[19:12:18] <Avenger> }
[19:12:20] <Avenger> i got it here
[19:12:26] <fuzzie> so ALC_INVALID_VALUE i guess
[19:12:38] <Avenger> no default device available?
[19:12:45] <-- dawid has left IRC ("Leaving.")
[19:13:56] <Avenger> i think i had some loop with trying to open the device a few times
[19:14:03] <Avenger> but smeone removed that :)
[19:14:16] <fuzzie> it sounds like it would cover up bugs
[19:14:38] <Avenger> well something is buggy here
[19:14:55] <Avenger> but this is the first call to openal
[19:14:59] <Avenger> it isn't our bug
[19:15:21] <Avenger> it could be the multithreading in some way?
[19:15:29] <fuzzie> that happens before then
[19:18:30] <Avenger> i added a check error before alcopendevice and it returned the same errorcode
[19:18:31] <fuzzie> ALC_INVALID_VALUE there means that no devices worked
[19:18:44] <Avenger> even if it worked later
[19:18:47] <Avenger> so...
[19:19:01] <fuzzie> but i guess alcGetError if what we want there
[19:19:03] <Avenger> does checkalerror clear the error number?
[19:19:29] <fuzzie> so see if that gives a more useful error; i don't think it will
[19:19:31] <-- Canageek has left #GemRB ("As I walk through the valley of the shadow of death/I take a look at my life and realize there's nothing left")
[19:19:57] <Avenger> but why i get an error before any device opening
[19:20:51] <fuzzie> because you call an openal function without init :)
[19:21:20] <fuzzie> ALC_INVALID_VALUE is the only error you'll get from alcOpenDevice anyway, from the source
[19:22:14] <fuzzie> alcOpenDevice simply tries opening all devices (only 'dsound' for playback on Windows) and then returns that if it can't open it
[19:22:50] <Avenger> and how can i clear the error status?
[19:23:16] <-- barra_library has left IRC (Connection timed out)
[19:23:52] <fuzzie> if there is truly an error, and openal is initialised, then retrieving the error should clear it
[19:24:13] <Avenger> i thought so too, and the doc says too
[19:24:27] <Avenger> hmm, the problem is openDevice will always try with invalid values
[19:24:29] <fuzzie> but since you do not successfully initialise openal, i think you will always get the same error
[19:24:39] <Avenger> i get it for successful opens too
[19:25:07] <fuzzie> initialisation happens at alcMakeContextCurrent
[19:25:31] <Avenger> but i don't get a device about 20% of the times :)
[19:25:39] <fuzzie> which is why i said above you probably want alcGetError
[19:25:48] <fuzzie> which will give you the error for the device open
[19:26:02] <fuzzie> but since it is always no error or ALC_INVALID_VALUE it is a predictable value..
[19:26:23] <Avenger> it doesn't clear the error status
[19:26:52] <Avenger> device = alcOpenDevice (NULL);
[19:26:54] <fuzzie> alcGetError should clear the error status always, initialised or not.
[19:26:54] <Avenger> checkALError("X", "");
[19:26:58] <Avenger> checkALError("X", "");
[19:27:00] <Avenger> i did this
[19:27:04] <fuzzie> but checkALError doesn't call alcGetError
[19:27:05] <Avenger> i got error for both lines
[19:27:10] <Avenger> huh
[19:27:27] --- D_T_G_away is now known as D_T_G
[19:27:36] <D_T_G> http://wklej.org/id/116652/
[19:27:40] <Avenger> sure it does
[19:27:47] <fuzzie> it calls alGetError
[19:27:52] <Avenger> oh
[19:27:55] <D_T_G> *.pro files at place
[19:27:58] <fuzzie> which is only valid after you did alcMakeContextCurrent
[19:28:01] <fuzzie> it is very confusing
[19:28:14] <fuzzie> D_T_G: ok, that works fine?
[19:28:19] <Avenger> ok, is there an alcGetError then?
[19:28:22] <fuzzie> yes
[19:28:37] <D_T_G> yes, after make install I have all that *.pro files in installation directory
[19:28:53] <fuzzie> and a ALC_NO_ERROR to check against (it is also 0)
[19:29:07] <Avenger> ok, you said i should call this checkallerror :D, i will make an CheckALCError clone from it
[19:30:03] <Avenger> alcGetError wants some parameter
[19:30:43] <fuzzie> oh, it wants the device
[19:30:46] <Avenger> lol
[19:30:46] <fuzzie> you can just pass NULL here
[19:31:02] <D_T_G> odd, I lost bottom of gui in gemrb
[19:31:09] <D_T_G> when testing projectiles
[19:31:41] <D_T_G> eh, it just returned :)
[19:32:19] <fuzzie> yes, passing a NULL device when you have no device seems 'supported' by all existing implementations
[19:32:36] <fuzzie> D_T_G: did you save a game first?
[19:33:01] <D_T_G> no, didn't save anything
[19:33:05] <fuzzie> ok
[19:33:12] <fuzzie> there are some strange disappearing-gui bugs :(
[19:34:06] <D_T_G> Displaying string on: none
[19:34:06] <D_T_G> [KEYImporter]: Searching for chfb1.2da...[ERROR]
[19:34:06] <D_T_G> Displaying string on: imoen
[19:34:18] <D_T_G> i think it happened on that
[19:34:36] <fuzzie> did you get part of another window appearing instead?
[19:34:44] <D_T_G> no
[19:35:03] <fuzzie> just black? or the game view there?
[19:35:45] <D_T_G> the window in center were area was supposed to be displayed was black
[19:36:07] <D_T_G> and I thought it was misplaced on the map to somewhere with war fog
[19:36:23] <D_T_G> and so I clicked to see a map
[19:36:39] <D_T_G> and on back the bottom part of gui returned
[19:37:28] <D_T_G> and to tell, more irritating to me is gemrb not seeing my saves
[19:37:43] <D_T_G> is this a known bug?
[19:37:49] <fuzzie> Avenger: your bug seems very strange, other games should also fail
[19:37:58] <fuzzie> D_T_G: i think it is a known bug where gemrb looks in mpsave/ for tob saves
[19:38:00] <Avenger> which bug
[19:38:08] <D_T_G> again bottom gui dissapeared !
[19:38:09] <fuzzie> Avenger: the openal soft device failure
[19:38:13] <D_T_G> will do a screenshot
[19:40:51] <fuzzie> openal does a standard directsound init, apart from one call to CreateThread which can surely not fail
[19:42:39] <Avenger> now it doesn't want to die
[19:42:41] <D_T_G> what are the *wc files in override?
[19:42:43] <fuzzie> and bg2/nwn at least only try one directsound open, so if it fails in openal it should fail for them too
[19:42:49] <D_T_G> they seem to be binary
[19:43:07] <Avenger> they are vvc, now wc
[19:43:21] <D_T_G> oh :)
[19:43:44] <Avenger> vvc are graphic resources, similar to a .pro
[19:43:46] <-- Gekz has left IRC (Read error: 104 (Connection reset by peer))
[19:44:04] <Avenger> animations and sound
[19:44:29] <D_T_G> I see
[19:44:47] --> Gekz has joined #GemRB
[19:47:20] <D_T_G> now gemrb sees my tob saves :)
[19:47:39] <D_T_G> rmdir mpsave && ln -s save mpsave
[19:49:07] <Avenger> hehe
[19:49:32] <Avenger> i told them it isn't good
[19:50:44] <Avenger> now i got this on exit: AL lib: alBuffer.c:1079: exit(): deleting 10 Buffer(s)
[19:52:24] <fuzzie> that is normal, i think
[19:52:49] <Avenger> never seen it before
[19:52:55] <fuzzie> i get "AL lib: alBuffer.c:1097: exit() 10 Buffer(s) NOT deleted" but fixing that just led to segfaults everywhere
[19:52:56] <D_T_G> the button for releasing mage's familiar from inventory is not implemented
[19:53:45] <fuzzie> well, the releasing is just not implemented, i think :)
[19:54:09] <Avenger> how is that done? an innate?
[19:54:20] <D_T_G> but taking them to inventory is
[19:54:24] <D_T_G> but not to well
[19:54:40] <Avenger> doh
[19:54:51] <Avenger> the moment i took out the debug code, openal failed again
[19:54:53] <D_T_G> the first pc in team got it, not the mage, the owner of familiar :)
[19:55:38] <Avenger> only the protagonist may have familiar anyway
[19:56:31] <D_T_G> iirc noone beside the owner could
[19:56:48] <D_T_G> take it to inventory
[19:57:11] <Avenger> is that script?
[19:57:36] <D_T_G> no idea how it worked internally :|
[19:58:35] <Avenger> you should have some :)
[19:58:49] <Avenger> i never seen it
[19:58:58] <Avenger> but i guess you talk to the familiar
[19:59:13] <D_T_G> yes, I talked, I was last in the team
[19:59:34] <D_T_G> but the familiar appeared in leader's invent
[19:59:53] <fuzzie> gemrb is not very good at remembering who is the protagonist
[20:00:06] <-- Gekz has left IRC (Read error: 54 (Connection reset by peer))
[20:00:43] --> Gekz has joined #GemRB
[20:01:18] <Avenger> inpartyslot is probably wrong
[20:01:25] <fuzzie> if you do SwapPCs then the game will happily stomp over InParty and then other code checks InParty==1 to find the protagonist
[20:01:47] <fuzzie> i think it is the InParty==1 check that is bad
[20:02:16] <Avenger> yes, InParty is the actual order
[20:02:52] <Avenger> i think the protagonist is the physically first PC
[20:02:58] <D_T_G> the leaving of candlekeep cut scene with gorion in bgt with *.pro now looks much better! good work
[20:02:59] <Avenger> physically in the .gam
[20:03:17] <fuzzie> yes, that sounds right
[20:04:01] <fuzzie> but then where is party order stored?
[20:04:29] <Avenger> InParty
[20:04:31] <fuzzie> oh, in pcInfo.PartyOrder.
[20:05:31] <fuzzie> which is then modified and stored in InParty by SetPersistent
[20:05:44] <Avenger> yes
[20:06:04] <fuzzie> ok, so this is easy to fix
[20:06:09] <Avenger> InParty is the party order field in of the .gam, but the protagonist is always the physically first one in the .gam
[20:06:45] <fuzzie> the 'PCs' array should be in the right order
[20:06:59] <fuzzie> as in, the .gam order
[20:07:06] <Avenger> GetPC(0) is the protagonist, yes
[20:07:25] <Avenger> FindPC(1) is the first PC in the portrait list
[20:07:30] <fuzzie> so all this stupid code doing FindPC should be fixed
[20:07:43] <Avenger> hmm
[20:07:45] <Avenger> not ALL
[20:07:50] <fuzzie> well, the stupid ones :p
[20:07:56] <Avenger> or you will screw up the portrait list
[20:08:02] <Avenger> just the scripting code
[20:08:04] <fuzzie> source = game->FindPC(1); //protagonist
[20:08:08] <fuzzie> ^- this one looks a bit suspicious too
[20:08:09] <Avenger> huh
[20:08:11] <Avenger> yes
[20:08:25] <Avenger> that IS more than suspicious ;)
[20:08:31] <fuzzie> :-)
[20:08:41] <fuzzie> that one has been annoying me for a while
[20:08:45] <fuzzie> but i didn't know how this worked
[20:09:16] <D_T_G> yes, it's a bit annoying indeed :)
[20:09:23] <Avenger> well, the problem is that everything is related to everything in this complex mess
[20:09:38] <fuzzie> yes, sometimes it's amazing anything works
[20:09:49] <D_T_G> when the whole party you initiate a dialog my last mage starts it :O
[20:09:50] <Avenger> and sometimes 'fixing' something means it breaks something else
[20:10:44] <D_T_G> did I mention chapter screens in bgt are totally broken: http://img517.imageshack.us/img517/3610/gemrb.png ?
[20:10:49] <fuzzie> yes, GetFirstSelectedPC is broken
[20:10:56] <fuzzie> or maybe being misused
[20:11:15] <D_T_G> the sound stops very early
[20:11:20] <D_T_G> and the text too
[20:11:33] <fuzzie> D_T_G: yes, chapter screens are very broken everywhere
[20:12:18] <fuzzie> i think our IncrementChapter() is broken among other problems
[20:12:19] <Avenger> well, i cannot reproduce the openal bug when i add debug code :)
[20:12:27] <Avenger> it is annoying
[20:13:16] <Avenger> success :)
[20:13:23] <Avenger> [Core]: Starting up the Sound Driver...[OpenALC]: X: 40964 []
[20:13:25] <Avenger> Tries:9
[20:13:27] <Avenger> [OpenALC]: X: 40964 []
[20:13:28] <Avenger> Tries:8
[20:13:30] <Avenger> [OpenAL]: Allocated 30 streams.[OK]
[20:13:42] <fuzzie> that is not code we want in trunk, though
[20:14:00] <Avenger> i think it happens when i click away from the screen, putting it in the background
[20:14:01] <fuzzie> unless we want a BrokenAudioDriver=true option
[20:14:10] <Avenger> but only if i do it in the right moment
[20:14:27] <fuzzie> presumably your driver is only allowing one piece of audio to play at once, and it must be an app in the foreground?
[20:14:32] <Avenger> i added Sleep(200) into the code, which gave me time to bring it front again
[20:14:38] <Avenger> yes
[20:14:50] <Avenger> so, the device open failure is now clear
[20:15:02] <Avenger> the crash bug when it cannot open the device is a real problem
[20:15:09] <Avenger> it should use the nullsound device
[20:15:14] <Avenger> err plugin
[20:15:15] <fuzzie> i can fix the crash bug
[20:15:17] <Avenger> but it doesn't
[20:15:20] <fuzzie> it is very difficult to make it use nullsound
[20:15:26] <Avenger> ?
[20:15:28] <fuzzie> but making it error out nicely is trivial
[20:15:37] <fuzzie> the plugins are already fully loaded by the time the audio is initialised
[20:15:43] <Avenger> you can reproduce my problem by returning NULL from the init
[20:15:54] <fuzzie> so nullsound already got ignored
[20:16:11] <Avenger> oh Init is called after all plugins got loaded?
[20:16:23] <Avenger> well, that sucks a bit
[20:16:23] <fuzzie> yes
[20:16:49] <Avenger> ok, but a clean exit is doable?
[20:16:59] <Avenger> you need ambim=NULL; plus some magic
[20:17:00] <fuzzie> that is easy, the code is just stupid right now :)
[20:17:45] <Avenger> i will do some more projectiles now that this is cleared :)
[20:17:59] <fuzzie> maybe commit your error checkign code, or is it too messy?
[20:18:33] <fuzzie> otherwise i'll add it
[20:21:03] <Avenger> i will scrap my stuff, it was only to find out what's wrong
[20:21:12] <fuzzie> ok
[20:25:32] <D_T_G> the *.pro-s will be done for all games?
[20:26:10] <fuzzie> it's probably sensible to try copying them in every makefile anyway
[20:26:14] <fuzzie> i'll patch it up in a minute
[20:26:25] <Avenger> yes
[20:26:31] <Avenger> .pros will be needed for all games
[20:26:49] <D_T_G> yes, the compilation gives an erro if you add *.pro to Makefile.am where there are no such files yet
[20:27:12] <Avenger> well, you can safely copy arrow.pro to all
[20:27:49] <Avenger> most likely it is all the same :)
[20:28:30] <D_T_G> I don't have installed any games that didn't get pros yet ;)
[20:28:30] <Avenger> i've added my openal solution to the known_problems page
[20:28:30] <fuzzie> i thought that you'd found a timing difference or similar
[20:28:47] <Avenger> fuzzie yes, that could be
[20:28:51] <D_T_G> I wanted to install iwd/how to see it though
[20:29:56] <Avenger> i think my projectiles are still not going with the right speed
[20:30:08] <D_T_G> too fast imo
[20:30:09] <Avenger> and they dont' have the same brightness either
[20:30:27] <D_T_G> yes that too
[20:30:31] <Avenger> there are some rendering flags and such
[20:30:50] <Avenger> but i first want to shovel in all projectiles, we can refine them later
[20:31:19] <D_T_G> maggic missile the first to test uncovers projectiles not really finished ;)
[20:31:51] <fuzzie> that is my fault
[20:32:25] <D_T_G> my first thought when read spell projectiles got implement: lets see, hmmm, magic missile
[20:32:53] <CIA-19> gemrb: 03fuzzie * r6703 10/gemrb/trunk/gemrb/plugins/OpenALAudio/OpenALAudio.cpp: OpenAL: better error handling
[20:32:56] <D_T_G> very common, but indeed seems complicated to implement
[20:33:10] <fuzzie> well, magic missile has two problems
[20:33:20] <fuzzie> one is that they don't take curved paths, and that is on my todo list
[20:33:38] <fuzzie> and the other is that the missiles launch at different times, but that is to fix *after* the paths
[20:33:41] <fuzzie> did you find other bugs?
[20:34:34] <D_T_G> with mm's , not really
[20:34:35] <Avenger> magic missile actually works, with those 2 'small' flaws
[20:35:12] <D_T_G> Gorion killed the ogres with mm's this time :)
[20:35:20] <Avenger> if you give the game to someone knowing only d&d, they wouldn't even complain about it
[20:35:26] <D_T_G> one survived last time I tried :)
[20:35:48] <Avenger> gorion kills the ogres sometmes in the original game too
[20:36:07] --- barra_away is now known as barra_library
[20:36:08] <D_T_G> well, always killed the for me
[20:36:35] <Avenger> i always root for the old bugger, sadly Sarevok has some invulnerability. it would be fun if he could somehow kill sarevok
[20:36:38] <D_T_G> when they survived in gemrb it was a huge suprise for me
[20:36:59] <fuzzie> i think it is vaguely possible for the ogres to survive
[20:37:11] <fuzzie> but very uncommon
[20:37:20] <fuzzie> gorion's spells are all hard-coded into the cutscene i think
[20:38:15] <D_T_G> there are howtos on the net how to kill Gorion being yet in Candlekeep...
[20:38:42] <fuzzie> but not Sarevok :)
[20:39:21] <D_T_G> there are too, for final fight :P
[20:39:33] <Avenger> oh finally i found this elusive portal open graphic
[20:39:39] <fuzzie> we should really fix all of the final fight abusive tactics
[20:39:41] <fuzzie> just to be mean
[20:39:54] <fuzzie> Avenger: for pst?
[20:40:02] <Avenger> no, how
[20:40:14] <Avenger> they got some projectiles i call prtl_OP/prtl_CL
[20:40:22] <Avenger> the sound resource was easy to find
[20:40:35] <Avenger> but it does some hardcoded graphic on an area object
[20:40:41] <Avenger> it is very very very very hardcoded
[20:41:05] <fuzzie> ah, boring :)
[20:41:17] <fuzzie> i was going to say, i thought it all worked for pst
[20:41:47] <Avenger> well, it is almost perfect in pst
[20:41:50] <Avenger> sans sparks
[20:46:56] <Avenger> they themselves call these animations as 'portal animation flipping hack open/close'
[20:47:58] <D_T_G> do you plan to implement interrupting of pause when fighting on inventory entrance for bg1?
[20:48:15] <fuzzie> yes
[20:48:22] <Avenger> that's one line of guiscript
[20:48:50] <fuzzie> we should make it a config option somewhere, truly
[20:49:00] <D_T_G> :) well, so for bg2 it would be so easy too
[20:49:18] <D_T_G> yes, the option config would be cool
[20:49:53] <fuzzie> i think we should stop armour change too? i forget
[20:49:55] <fuzzie> must go now
[20:50:13] <D_T_G> bye
[20:52:26] <D_T_G> the blocking of armour changing during fight it in bg2
[20:52:37] <D_T_G> in bg1 it should stop the pause
[20:53:04] <D_T_G> but it would be nice for bg2 mode to set the bg1 behaviour
[20:54:02] <-- barra_library has left IRC ("Verlassend")
[20:54:37] <D_T_G> do you collect such ideas somewhere?
[20:59:55] --> barra_library has joined #gemrb
[21:00:30] <Avenger> these are very cosmetical changes, we are still using the axe :)
[21:01:15] <-- tombhadAC has left IRC (Network is unreachable)
[21:02:06] --> tombhadAC has joined #gemrb
[21:02:10] <D_T_G> I see :)
[21:12:29] <-- Gekz has left IRC (Read error: 104 (Connection reset by peer))
[21:16:28] <fuzzie> if you make a wiki account, you can put it on todo
[21:16:39] <fuzzie> or on sourceforge in the Feature Requests
[21:16:43] <-- tombhadAC has left IRC ("Verlassend")
[21:18:06] <Avenger> haha, this portal open/close stuff is at the very end of HoW, in the cutscene
[21:18:21] <Avenger> ctrl-y ing the demon ends the game
[21:18:51] <Avenger> lets say, this isn't top priority for me now :)
[21:19:10] --> tombhadAC has joined #gemrb
[21:20:20] <D_T_G> I registered on wiki :)
[21:21:20] <D_T_G> but I don't think it would fit to todo
[21:21:32] <D_T_G> maybe something like 'possible extenstions'?
[21:22:03] <fuzzie> well, the missing functionality of bg1 and bg2 belongs on todo :)
[21:22:17] <fuzzie> making it work for all games doesn't fit there though
[21:22:42] <fuzzie> although i would just add a 'New Features' section at the bottom
[21:24:53] <Avenger> new feature i need: setting looping sounds, setting moving sounds, and ability to kill a sound
[21:25:13] <Avenger> that would improve projectiles quite much
[21:25:50] <Avenger> i will add these to TODO
[21:26:19] --> Gekz has joined #GemRB
[21:27:57] <wjp> hm, moving sounds?
[21:28:05] <wjp> how would you specify the movement path?
[21:28:27] <wjp> or would the engine update the position from time to time?
[21:29:14] <Avenger> well, that is an option
[21:29:49] <wjp> sounds like Play() should return some kind of handle that can be used to later modify/stop the sound
[21:29:54] <Avenger> yes
[21:30:12] <Avenger> it would be the best if i could set the loop, coordinates on the handle
[21:30:18] <fuzzie> "sound handles" being at the top of our wiki todo :-)
[21:31:54] <fuzzie> might be a cleaner way to handle GEM_SND_SPEECH, too
[21:34:15] <Avenger> hehe spike growth is so weird
[21:34:28] <Avenger> and it is ALMOST like the original
[21:34:54] <Avenger> some projectiles succeed at the first time, without even seeing the original :)
[21:44:28] --> pupnik_ has joined #gemrb
[21:45:12] --> pupnik__ has joined #gemrb
[21:45:16] <wjp> wow, the BG2 start sequence runs a lot smoother than, say, a month ago
[21:46:14] <wjp> both wrt features and speed
[21:46:49] <Avenger> i didn't notice there were missing features in the first cutscene
[21:46:58] <Avenger> probably some hardcoded missile?
[21:47:32] <wjp> I could be misremembering things too, though :-)
[21:47:33] <Avenger> the escape cutscene improved
[21:48:01] * wjp tries that
[21:48:04] <Avenger> and the traps
[21:48:16] <Avenger> you know where the dwarf trips over a scorcher trap :)
[21:48:56] <fuzzie> i am wondering if there are script bugs or similar in the escape one
[21:49:27] <fuzzie> i didn't try debugging it yet, but i know lots of potential bugs
[21:50:58] <fuzzie> for example that silly SetWait in the spell casting
[21:51:05] <wjp> the timing of the escape sequence still feels off
[21:51:16] <fuzzie> which is presumably going to throw the timing right off
[21:51:55] <fuzzie> hm, that is probably the cause of the timing bugs in the beginning too
[21:52:05] <fuzzie> i shall move that up my priority list..
[21:55:06] <D_T_G> bye ppl
[21:55:16] <-- D_T_G has left IRC ()
[21:56:32] <wjp> did the speed improve? Things seem a lot more fluent than before
[21:57:18] <Avenger> yes fuzzie did a lot for that
[21:57:22] <-- Avenger has left IRC ("ChatZilla 0.9.85 [Firefox 3.5/20090624025744]")
[21:59:13] <fuzzie> well, it depends what you mean by speed, i guess :)
[21:59:40] <wjp> general responsiveness and fluidity of movement
[22:01:18] <CIA-19> gemrb: 03avenger_teambg * r6704 10/gemrb/trunk/gemrb/plugins/Core/Projectile.cpp: improved projectile graphics (speed differences) and sound
[22:01:28] <fuzzie> I don't think there have been any real changes in general responsiveness. There have been pathing fixes of various kinds though, which helped a lot for me.
[22:01:28] <wjp> hm, formation pathfinding will be an interesting challenge
[22:01:59] <wjp> right now when moving the party members in the back will try to walk around those in the front
[22:02:01] <fuzzie> The pathfinder right now is breadth-first, which is not such a huge challenge.
[22:02:22] <fuzzie> You can do all kinds of cleverness by brute-forcing there, even if you include multiple actors.
[22:02:28] <wjp> (very noticable on the promenade where there are narrowish stairs with alternative routes)
[22:02:33] <fuzzie> But really we should do A* and there it is painful :)
[22:03:10] <fuzzie> I am hoping we'll get it 'good enough' by having actors pause momentarily on collison and repath every few seconds..
[22:03:47] <CIA-19> gemrb: 03avenger_teambg * r6705 10/gemrb/trunk/gemrb/override/how/ (11 files): more projectiles
[22:03:53] <wjp> maybe also ignoring currently moving actors
[22:04:02] <fuzzie> yes, that is essential
[22:04:16] <fuzzie> and we need to set all actors to moving /before/ we path a formation
[22:04:30] * wjp nods
[22:04:33] <fuzzie> otherwise moving short distances in formation is impossible, you just collide with the party itself
[22:05:00] <CIA-19> gemrb: 03avenger_teambg * r6706 10/gemrb/trunk/gemrb/override/ (bg1/Makefile.am bg2/Makefile.am how/Makefile.am): fixed makefiles to copy .pro files too
[22:05:05] <wjp> that might mean some nice searchmap trickery
[22:06:18] <CIA-19> gemrb: 03avenger_teambg * r6707 10/gemrb/trunk/TODO: added sound handles to local TODO
[22:08:41] <fuzzie> The searchmap code needs a decent review anyway, lots of strangeness.
[22:09:15] <wjp> I wish each bit had a refcount :-)
[22:30:34] <fuzzie> hm, fixing the casting times just serves to make our lack of aura residue more irritating :)
[22:32:06] <Lightkey> so many projectiles..
[22:32:27] <wjp> aura residue?
[22:32:44] <fuzzie> wjp: inability to cast a new spell until 6 seconds have passed total
[22:33:05] <wjp> ah, the thing that alacrity removes?
[22:33:09] <fuzzie> mhm
[22:33:13] <fuzzie> that effect is 'aura cleansing'
[22:34:53] <fuzzie> all very dorky :)
[22:35:01] <fuzzie> i have something which i think works but i'll have to check it against original
[22:35:03] <fuzzie> ninight for now
[22:35:22] <wjp> good night
[23:37:58] <-- tombhadAC has left IRC (Remote closed the connection)
[23:46:28] <-- xrogaan has left IRC ("Why ?")