#gemrb@irc.freenode.net logs for 2 Aug 2009 (GMT)

Archive Today Yesterday Tomorrow
GemRB homepage

[00:49:06] <-- Gekz has left IRC (Read error: 110 (Connection timed out))
[01:21:55] --> Gekz has joined #GemRB
[02:52:24] <-- dawid has left IRC ("Not leaving, quitting.")
[03:00:34] --> dawid has joined #GemRb
[04:02:31] <-- barra_away has left IRC ("Verlassend")
[04:06:12] --> sney has joined #GemRB
[05:37:59] <-- Gekz has left IRC ("leaving")
[06:56:07] <-- sney has left IRC ("I don't think I'm better than you. I just want to share my thoughts")
[09:02:09] --> Gekz has joined #GemRB
[09:19:36] --> lynxlynxlynx has joined #gemrb
[09:19:36] --- ChanServ gives channel operator status to lynxlynxlynx
[09:36:04] --> zefklop has joined #gemrb
[09:40:43] <fuzzie> morning
[09:40:56] <zefklop> morning
[09:57:19] <pupnik_> p/
[09:57:38] <pupnik_> anybody want to come over and help me clean the house?
[09:59:45] <fuzzie> if you'd like to do my week of CS exams for me :)
[10:00:47] <zefklop> meh, haw comes that I have the mageaction bar when playing a fighter?
[10:00:55] <zefklop> in BG1
[10:02:03] <zefklop> well, in inventory too, I hav only two quick weapon slots
[10:02:12] <zefklop> will investigate later.
[10:02:23] <zefklop> In the meantime, could you test this
[10:03:25] <fuzzie> hm
[10:04:10] <fuzzie> hm
[10:04:23] <fuzzie> interesting :)
[10:04:46] <fuzzie> i get a mage action bar also
[10:05:22] <fuzzie> it works fine if i load another save, though
[10:05:37] <fuzzie> so it is maybe the chargen?
[10:05:45] <zefklop> seems so
[10:06:05] <fuzzie> the bg1 chargen got rewritten recently, and we had a bit of a battle with the spells
[10:06:19] <zefklop> great!
[10:06:31] <CIA-22> gemrb: 03zefklop * r6791 10/gemrb/trunk/gemrb/plugins/ (ACMImporter/CMakeLists.txt Core/Map.cpp):
[10:06:31] <CIA-22> gemrb: SDL includes are not needed in acmimporter.
[10:06:31] <CIA-22> gemrb: Try to fix sight. Tested in BG1, please test other games.
[10:06:34] <zefklop> please test that!
[10:07:31] <zefklop> especially walking near walls
[10:07:37] <pupnik_> hmmm
[10:14:14] <fuzzie> hm, rebuilt the whole thing and found a failed merge bug at the end, oops
[10:15:54] <zefklop> ah...
[10:16:12] <fuzzie> not your fault, mine! but now i have to rebuild again :)
[10:18:31] <fuzzie> wow, that is hugely improved
[10:18:40] <fuzzie> how on earth did you manage that?
[10:19:15] <fuzzie> PST is suddenly playable!
[10:19:30] <zefklop> I remembered what I did months ago, and decided to have a better fix :-)
[10:20:53] <fuzzie> doors are still buggy, alas
[10:21:21] <fuzzie> but i assume that is other code
[10:21:40] <zefklop> no, it is the same
[10:22:15] <zefklop> to be honest, I wonder why we don't use IsVisible() in this part of code
[10:22:18] <fuzzie> http://fuzzie.org/nfs/gemrb/pst_doors.jpg
[10:22:48] <zefklop> meh
[10:23:24] <fuzzie> i wouldn't worry, but if you ever have any ideas about how to fix those, it would be nice to hear
[10:24:03] <zefklop> let me DLTCEP this :-)
[10:24:14] <Gekz> PS:T is the ugliest of the Infinity games
[10:24:15] <Gekz> I must say
[10:24:21] <fuzzie> yes :) but it is also the best!
[10:24:56] <zefklop> meh; DLTCE¨P is not on this computer?
[10:25:21] <Gekz> I think BG is the best
[10:26:25] <zefklop> BG begins to feel its age to me
[10:26:58] <fuzzie> SoA is a beautiful masterpiece also
[10:28:14] <Gekz> ToB was a bit rushed
[10:28:19] <zefklop> yeah, very nice to look at, good story, good side quests... lacks subtlety in the end though
[10:28:20] <Gekz> but the completion of the romances in ToB was grand
[10:28:32] <Gekz> I cant wait for Dragon Age
[10:28:34] <zefklop> I never finished ToB :-/
[10:28:34] <Gekz> but omg
[10:28:38] <Gekz> have you seen the recommended specs?
[10:28:41] <Gekz> 20GB of space
[10:28:43] <Gekz> quad core
[10:28:45] <Gekz> 4GB of ram
[10:28:50] <Gekz> for a fucking computer game?!
[10:28:54] <Gekz> that's recommended
[10:29:02] <Gekz> minimum is dual core, 2GB of RAM, 20GB space
[10:29:11] <zefklop> Yeah, this game sounds promising
[10:29:14] <Gekz> oh, graphics, 128MB 7800GT
[10:29:17] <Gekz> upwards
[10:29:17] <fuzzie> it is one of those games which I will only end up seeing once it is old :)
[10:29:19] <Gekz> it looks awesome
[10:29:28] <Gekz> I have a computer that can play it at Medium
[10:29:32] <Gekz> I bought this to play NWN2
[10:29:38] <Gekz> I never played NWN2 because it's a disgrace
[10:29:42] <fuzzie> yesterday I upgraded the x86 desktop here with a Radeon 9800 Pro
[10:29:47] <fuzzie> this is my technology level :-)
[10:29:48] <Gekz> if Dragon Age sucks, I give up on Bioware forever
[10:30:02] <Gekz> fuzzie: that's an upgrade? wtf.
[10:30:17] <zefklop> fuzzie: 9800 pro was the best graphic cards for a long time
[10:30:19] <fuzzie> it had a Geforce 2 MX before :p
[10:30:37] <zefklop> which wa a very good one too :D
[10:30:48] <fuzzie> my brother bought both when they were new, and then gave them to me when he upgraded
[10:30:57] <Gekz> well
[10:31:24] <Gekz> fuzzie: I have an oldish VIA C7 mini-itx motherboard sitting in the other room
[10:31:26] <fuzzie> so now I can catch up on Portral and things like that, maybe :)
[10:31:36] <Gekz> 1GB DDR RAM
[10:31:43] <Gekz> 1GHz CPU
[10:31:49] <Gekz> pretty low power
[10:32:00] <Gekz> popped in a PCI graphics card, an MX4 or something
[10:32:04] <Gekz> 64MB
[10:32:06] <fuzzie> I live in a student building and don't pay for the electricity, hooray!
[10:32:13] <Gekz> if I were you fuzzie
[10:32:15] <Gekz> I'd buy a Core i7
[10:32:22] <Gekz> a 6-core one
[10:32:27] <Gekz> and rent it out as a vserver
[10:32:41] <Gekz> especially to the computer science students
[10:32:58] <Gekz> nothing like a testpad that can compile KDE in 2 mintues
[10:33:05] <fuzzie> the CS students are scary, though
[10:33:11] <Gekz> you dont have to talk to tehm
[10:33:13] <Gekz> you put up fliers
[10:33:16] <Gekz> with and email address
[10:33:17] <Gekz> lol
[10:33:18] <Gekz> an*
[10:33:26] <Gekz> make them pay through paypal :P
[10:42:35] <zefklop> fuzzie: I think I hav the door bit :-)
[10:42:43] <zefklop> let me test before committing
[10:46:28] <zefklop> err, no
[10:48:17] <fuzzie> I should fix trap disarming, I think it is the last game-breaking thing in the SoA dungeons..
[10:50:14] <fuzzie> it is terribly annoying to fix properly, but a hack to just pop nodes off the path seems viable
[10:50:31] <fuzzie> does anyone else see other game-breaking bugs in the dungeon?
[10:50:59] <zefklop> I have only BG1 here :/
[10:51:06] <zefklop> time to lunch
[10:51:17] <fuzzie> i think you can play bg1 quite far..
[10:51:48] <fuzzie> far enough that i got bored :)
[10:54:03] <Gekz> lol
[10:54:20] <Gekz> fuzzie: do you have OS9?
[10:54:39] <fuzzie> yes, in Dutch :)
[10:54:52] <Gekz> BG2 SoA and ToB are in OS9 and OSX versions
[10:54:55] <Gekz> on the same .dgms
[10:54:56] <Gekz> dmgs*
[10:54:59] <Gekz> just so you know
[10:55:00] <fuzzie> i don't think it boots on this machine though
[10:55:21] <Gekz> so you could debug the networking code on either one
[10:55:21] <Gekz> :P
[10:55:37] <fuzzie> "will start up only in Mac OS X", says the box
[10:55:41] <fuzzie> apple are big meanie heads
[10:56:08] --> xrogaan has joined #gemrb
[10:57:30] <fuzzie> anyway, networking is a distraction :)
[10:57:36] <fuzzie> single-player first
[11:10:13] <Gekz> fuzzie: maybe I have an older version :D
[11:10:45] <fuzzie> i mean, my Mac will only start up in Mac OS X :)
[11:11:11] <fuzzie> i should try the heat gun trick on one of the older machines.
[11:14:01] <-- Gekz has left IRC ("Reconnecting")
[11:14:13] --> Gekz has joined #GemRB
[11:14:21] <Gekz> heat gun trick?
[11:15:23] <fuzzie> I have a few dead G3 iBooks .. their GPUs failed because they'd disconnect from the board
[11:16:08] <fuzzie> apparently you can fix them by heating the GPU up to 230C or so, which will melt the solder enough that it'll reconnect
[11:22:00] <Gekz> haha
[11:22:01] <Gekz> weird
[11:22:08] <Gekz> wait
[11:22:11] <Gekz> you said a few
[11:22:14] <Gekz> how many is a few?
[11:23:04] <fuzzie> well, apple stopped bothering to collect the broken ones at some point.. :p
[11:23:33] <fuzzie> i think we have three which might be fixable with the heat trick
[11:30:52] --> Gekz_ has joined #GemRB
[11:30:57] <Gekz_> fuzzie: well
[11:31:03] <Gekz_> if you fix them and you have a spare one
[11:31:12] <Gekz_> let me know, we might be able to come up with some kind of deal
[11:31:22] <Gekz_> should .nl not have prohibitively expensive postage costs
[11:31:42] <-- Gekz has left IRC ("Reconnecting")
[11:33:19] --- Gekz_ is now known as Gekz
[11:35:49] <fuzzie> well, i have no idea how stable they'll be, but here's hoping :)
[11:35:58] <Gekz> haha
[11:36:12] <Gekz> are they the old clamshells?
[11:36:30] <Gekz> oh and they're EU too, this is good
[11:36:35] <Gekz> 220V is ~= 240v
[11:37:14] <fuzzie> they're newer white ones, although I keep seeing clamshells on ebay, i want one :)
[11:38:16] <fuzzie> i am pretty happy with gemrb's combat now, it works ok even in small tombs etc
[11:42:26] <fuzzie> hm, no res spells :(
[11:43:03] <fuzzie> but I found Mazzy, hoorah
[12:03:52] <fuzzie> zefklop: do you know if scripts continue during fading? you say they don't in bug #1837747, but i find that they do continue
[12:04:11] <fuzzie> and i wonder if i should look at it more, or if you just guessed
[12:10:50] <fuzzie> i'm pretty sure fading works when paused, too
[12:18:53] <Gekz> fuzzie: oh! that's even better
[12:19:01] <Gekz> I'm not fond of the clamshells, being male
[12:19:02] <Gekz> lol
[12:20:25] <Gekz> I've never had anything other than x86, and m68k
[12:20:59] <fuzzie> there's all kinds of different old hardware around here that I could play with
[12:21:12] <fuzzie> very pretty blue Indy, i think that's mips?
[12:22:37] <Gekz> lol
[12:22:46] <Gekz> I'm not interested in MIP
[12:22:46] <Gekz> S
[12:22:54] <Gekz> anything without an MMU makes me sad
[12:22:57] <fuzzie> it's pretty!
[12:23:08] <Gekz> I have an iPod
[12:23:09] <Gekz> it's ARM
[12:23:15] <Gekz> and it's not encypted :D
[12:23:18] <Gekz> encrypted*
[12:24:48] <fuzzie> do you have rockbox on it? :)
[12:24:51] <Gekz> yes
[12:24:53] <fuzzie> hooray
[12:24:54] <Gekz> and had ipodlinux
[12:24:59] <Gekz> but
[12:25:09] <Gekz> if you get one of these G3s working stably
[12:25:11] <fuzzie> rockbox is cooler!
[12:25:12] <Gekz> I'd happily trade :P
[12:26:15] <fuzzie> an arm device which had enough resolution for gemrb would be nice :)
[12:26:45] <fuzzie> even the shiny iPhone only does 320x480 :(
[12:28:15] <fuzzie> i am a bit doubtful about the heat thing working, but maybe after some practice
[12:28:43] <fuzzie> i have someone's thinkpad here which needs the same treatment too
[12:28:52] <fuzzie> but after exams :)
[12:30:58] <Gekz> lol
[12:31:07] <Gekz> the freaking ipod is worth more than the G3 these days
[12:31:08] <Gekz> that's sad
[12:31:45] <fuzzie> the resale value of the G3s kind of collapsed after they all started failing with this problem, and now everyone's moved to Intel anyway..
[12:35:13] <Gekz> my mini-distro
[12:35:23] <Gekz> I'm building toolchains for common architectures
[12:35:32] <Gekz> and making them into squashfs files
[12:35:50] <Gekz> cross-compilation is hard :O
[12:35:58] <Gekz> making it find the correct headers mostly
[12:37:19] <Gekz> brb
[12:37:20] <-- Gekz has left IRC ("leaving")
[12:37:24] --> Gekz has joined #GemRB
[12:37:38] <Gekz> hmm
[12:37:53] <Gekz> but yeah
[12:37:58] <Gekz> it's fun nonetheless
[12:47:03] <fuzzie> :)
[12:47:31] <Gekz> I can't wait to learn maths properly next year
[12:50:26] <fuzzie> properly?
[12:50:54] <Gekz> lol
[12:50:55] <Gekz> yes
[12:50:57] <Gekz> I failed last year
[12:51:02] <Gekz> and had to drop the course altogether
[12:52:42] <fuzzie> ah :( similar for me, i have mostly failed the second semester of maths
[12:53:04] <Gekz> is it because you're bad at it
[12:53:11] <Gekz> or just not completely the homewok
[12:53:13] <Gekz> like I wasn't
[12:53:18] <Gekz> and thus not studying hard enough
[12:53:38] <fuzzie> i fall asleep in all the lectures :) some kind of sleep problem, they didn't diagnose it yet
[12:53:59] <Gekz> << Thousands of people have been placed in quarantine in north-western China after a man died of pneumonic plague. >>
[12:54:09] <Gekz> this is bad.
[12:54:29] <fuzzie> scary
[12:55:46] <Gekz> lol
[13:40:30] <-- dawid has left IRC (Read error: 110 (Connection timed out))
[13:45:18] --> dawid has joined #GemRb
[13:52:39] --> zefklop_ has joined #gemrb
[13:59:31] <-- zefklop has left IRC (Read error: 113 (No route to host))
[14:27:33] <-- xrogaan has left IRC ("Why ?")
[15:05:44] <-- zefklop_ has left IRC (Read error: 113 (No route to host))
[15:05:54] --> Avenger has joined #gemrb
[15:06:05] --- ChanServ gives channel operator status to Avenger
[15:06:10] <Avenger> hi
[15:07:02] <Avenger> fuzzie what do you think, when could we call gemrb more beta than alpha?
[15:08:40] <Gekz> Avenger: I'll throw in my opinion, and that's when you can walk to the end of the game
[15:08:49] <Gekz> ie, go through all areas witohut it collapsing into a puddle of hate
[15:08:52] <Gekz> in one gay
[15:08:55] <Gekz> game*
[15:09:12] <Avenger> hmm
[15:09:26] <Gekz> and kill some stuff
[15:09:32] <Gekz> but
[15:09:35] <Gekz> in official terms
[15:09:43] <Gekz> a beta is when all features are in place
[15:09:46] <Avenger> yesterday i started a bg2 game, and it was fine until a dialog started while i opened a ground pile
[15:09:56] <Gekz> and it's quite complete
[15:10:00] <Gekz> but just buggy as fuck
[15:10:08] <Avenger> i could kill stuff
[15:10:13] <Gekz> alpha is when it's still incomplete in many places, stubs galore
[15:10:19] <Avenger> sometimes targeting was screwed up
[15:10:29] <Avenger> well, i wouldn't remove the alpha tag
[15:10:29] <Gekz> gibberlings giving you -65535 damage
[15:10:30] <Gekz> that kinda thing
[15:10:36] <Gekz> :D
[15:10:46] <fuzzie> we have a plan for the dialog starting bugs, hopefully fixed in next few days
[15:12:02] <Avenger> well, i plan to set the beta tag at the end of the month :)
[15:12:08] <fuzzie> i think beta for gemrb is when SoA is playable the whole way through without any annoying bugs :)
[15:12:20] <fuzzie> i have the last two weeks of august for that, i hope
[15:12:29] <Avenger> me too
[15:12:43] <Avenger> if everything goes right, i'll have 2 weeks out
[15:12:52] <fuzzie> i downloaded a big walkthrough of SoA, and i am going through it quest-by-quest
[15:13:08] <fuzzie> today i got all the way to rescuing Mazzy
[15:13:09] <Avenger> hmm, targeting was bugged recently
[15:13:20] <Avenger> i'm pretty sure it worked before
[15:13:25] <Gekz> what
[15:13:27] <Gekz> so spells work now?
[15:13:35] <fuzzie> sure
[15:13:38] <Gekz> wow
[15:13:38] <Avenger> now, when i stop a fight, and try to cast a healing, sometimes it targets the dead guy
[15:13:40] <Gekz> i missed that
[15:13:52] <Avenger> gekz-->spells work for 2-3 months now O_o
[15:13:57] <Gekz> right, I have to throw this in before I forget
[15:14:05] <Gekz> is there a flag to disable permadeath
[15:14:18] <Gekz> nothing worse than exploding party members
[15:14:20] <fuzzie> for party members? i think it is a difficulty level thing
[15:14:21] <Avenger> to disable chunky death?
[15:14:24] <Gekz> yes
[15:14:29] <Avenger> yes fuzzie is right
[15:14:33] <Gekz> yes but
[15:14:38] <Gekz> some might want to play on very hard
[15:14:41] <Avenger> that's a difficulty thing, not implemented by us
[15:14:41] <Gekz> without getting exploded
[15:14:43] <Gekz> haha
[15:14:56] <fuzzie> someone could mod it, but i think it is not for us :)
[15:15:02] <Gekz> ok
[15:15:18] <Avenger> there is some flag to prevent chunky death, no?
[15:15:31] <Gekz> that was my question
[15:15:53] <fuzzie> it's complicated - there is MC_KEEP_CORPSE
[15:16:15] <fuzzie> but it only works sometimes in the original
[15:18:55] <Avenger> i could add a new effect, which sets it, apparently there is no one
[15:19:07] <Gekz> this could be useful
[15:19:10] <Gekz> yay, I'm helping
[15:19:10] <Gekz> lol
[15:19:24] <Avenger> SetCreatureAreaFlag action does it
[15:19:54] <Avenger> but that is normally only in iwd2
[15:20:14] <Avenger> gemrb is already moddable (in a very awkward way) to have it, at least for the players
[15:20:42] <Avenger> just add the action to the action list, and modify the dplayer script
[15:25:52] --> zefklop has joined #gemrb
[15:26:37] <fuzzie> i wonder how long it'll be until we find a mod that doesn't work due to gemrb's action list hack :/
[15:27:54] <Gekz> lol
[15:30:35] <fuzzie> i really need to make a post about things that are completely wrong in iesdp, too
[15:31:30] <Gekz> lol
[15:31:36] <Gekz> you're doing a good job fuzzie
[15:31:38] * Gekz nods
[15:32:18] <fuzzie> just things like the description for EscapeAreaObject. :(
[15:33:39] <Gekz> what an ambiguous name
[15:33:51] <Gekz> is it an object that allows you to escape the area on click?
[15:34:46] <fuzzie> it's an action which tells someone to leave the area via the specified object (usually a door)
[15:36:30] <fuzzie> and IESDP says some nonsense instead, despite a correction from over three years ago.. :P
[15:37:02] <fuzzie> so the online copy is presumably that old
[15:37:23] <fuzzie> it says "04.05.2008" but that is clearly not true :(
[15:37:58] <Gekz> rage
[15:38:00] <Gekz> alright
[15:38:02] <Gekz> must sleep
[15:38:03] <Gekz> night night
[15:38:05] <fuzzie> night
[15:38:07] <Avenger> maybe it was 'corrected' back :)
[15:39:13] <fuzzie> any luck on finding the pst projectile code?
[15:39:58] <Avenger> i have it, i'm just not happy with it
[15:40:50] <Avenger> in pst, the projectile randomly contains spell effects
[15:41:08] <Avenger> almost all new projectiles
[15:41:35] <fuzzie> things like the spell 'turning'?
[15:42:07] <Avenger> no, things like damage, hit point transfer, coloration, strength bonus
[15:42:37] <Avenger> do you follow the thread about Power of One/Strength in pst, on SHS?
[15:42:51] <fuzzie> no, let me look
[15:43:08] <Avenger> the code that calculates how much points to give in the strength effect, is coded inside the strength of one projectile
[15:44:14] <Avenger> gemrb is almost capable of these features, except it has no feature for arbitrary spell effects. I guess, i will need to add another field
[15:44:37] <fuzzie> modders will never be short of fields to use :)
[15:44:55] <Avenger> i just added a new field, btw
[15:45:04] <Avenger> sparks in PST are very sparse
[15:45:13] <Avenger> only one particle per spark group
[15:45:20] <Avenger> so, i added a field for the group size
[15:46:13] <Avenger> strangely, gemrb still renders some 5, but at least not dozens as before
[15:46:28] <Avenger> and the sparks still fall all the way down across the screen
[15:46:33] <Avenger> they should stop :(
[15:46:54] <fuzzie> there's just no code which ever stops them, i think
[15:47:52] <fuzzie> but i don't have any clue how they look in the original game
[15:48:55] <Avenger> well my intention was to have the sparkle region height as 80
[15:50:05] <Avenger> the core which sets the width, seems to be working
[15:50:10] <Avenger> *code
[15:57:53] <zefklop> fuzzie: to me, the fade action is like a wait() one
[15:59:10] <fuzzie> zefklop: but did you check? :)
[15:59:34] <zefklop> no, it's only remembrance of what I did at this moment
[15:59:41] <fuzzie> ok.
[15:59:53] <fuzzie> i'll have to go check it in all the games sometime
[16:00:08] <zefklop> does it exist in BG1? I'm stuck with it for the next two weeks
[16:00:30] <zefklop> (which is good btw, as we don't work very uch on it)
[16:00:38] <fuzzie> yes, it does
[16:00:56] <fuzzie> i checked it in pst and bg2, and it just starts the fade and then continues running actions
[16:01:32] <fuzzie> but we should really work more on bg1, it is a lot simpler to make work
[16:01:55] <fuzzie> did you work out the action bar problem?
[16:02:03] <zefklop> that's strange, because when starting BG2, there is no such things. Irenicus seems to wait for the fade to finish
[16:02:29] <zefklop> no, I will do it now that I'm back
[16:02:50] <fuzzie> the Irenicus scene does Wait() after the fades, i think
[16:03:27] <fuzzie> if you remove them, so that the script simply does FadeToColor() and then FadeFromColor(), then it only does the 'fade from' bit
[16:03:28] <zefklop> this could be that it was a hack at this time, as timing was fixed since then
[16:03:37] <fuzzie> yes, maybe so
[16:03:51] <fuzzie> i think there are quite a few bits where your round fixes would be needed for things to work, in fact..
[16:04:34] <zefklop> yup
[16:04:45] <zefklop> so, let's fix this action bar thing
[16:05:27] <zefklop> oh, I had this door fog thing too
[16:05:59] <fuzzie> well, it is correct that the door should block all sight behind the door, but it should not block the door itself
[16:06:11] <zefklop> that's it
[16:06:21] <zefklop> so I'll treat the door as wall
[16:06:24] <fuzzie> it is good that you can see it :)
[16:08:54] <Avenger> ok, i think i will be able to fix the sparkles
[16:10:02] <Avenger> fuzzie: there is already a searchmap type for that, just do the same as that one
[16:10:38] <Avenger> PATH_MAP_SIDEWALL
[16:10:54] <fuzzie> how are normal walls done?
[16:11:07] <zefklop> avenger: this is for walls, but does nt work with doors
[16:11:38] <Avenger> well, doors be handled like sidewalls
[16:11:56] <zefklop> isn't int PATH_MAP_DOOR_OPAQUE ?
[16:12:16] <fuzzie> i think Avenger is saying, handle that the same as SIDEWALL
[16:12:21] <fuzzie> i don't know if that's correct though
[16:12:29] <fuzzie> sight for doors is definitely not the same as sight for walls
[16:12:34] <zefklop> yes, I tried to do it, but it does not work
[16:13:10] <Avenger> well, a door is just a sidewall :)
[16:13:24] <fuzzie> maybe the door polygons change things, then?
[16:13:47] <fuzzie> anyway, i will let zefklop work it out :)
[16:14:02] <Avenger> what about: #define PATH_MAP_DOOR_OPAQUE 16|PATH_MAP_SIDEWALL
[16:14:04] <Avenger> :)
[16:14:09] <Avenger> dunno if it works
[16:14:48] <zefklop> I tried to do if(type & (PATH_MAP_SIDEWALL|PATH_MAP_DOOR_OPAQUE))
[16:14:49] <Avenger> oops, no , that wouldn't work
[16:15:18] <zefklop> in map.cpp, line 2759
[16:15:25] <zefklop> mine or yours?
[16:15:49] <zefklop> mine does not
[16:15:50] <Avenger> mine
[16:16:04] <Avenger> in GetBlocked: if (ret&PATH_MAP_DOOR_OPAQUE) {
[16:16:05] <Avenger> //ret=PATH_MAP_NO_SEE;
[16:16:07] <Avenger> ret=PATH_MAP_SIDEWALL;
[16:16:08] <Avenger> }
[16:16:10] <Avenger> now i try this
[16:16:41] <Avenger> where is a door in pst
[16:17:02] <fuzzie> a good test door is the one in the area *above* the start area
[16:17:16] <zefklop> uh, I did not see that GetBlocked modified the value...
[16:17:23] <fuzzie> because you can see through the walls and not the door, maybe you can work out why
[16:17:24] <zefklop> I think that will work
[16:20:03] <fuzzie> Avenger's change doesn't work for me
[16:21:11] <Avenger> when nameless dies, he loses stuff?
[16:21:11] <zefklop> nor for me
[16:21:27] <fuzzie> loses what?
[16:22:26] <fuzzie> oh, i guess CheckOnDeath is going to make him drop his inventory
[16:23:09] <fuzzie> i guess we need a special-case there..
[16:25:17] <zefklop> avenger's ghange + changing 48 to 60 in map.cpp at line 2761 works for me
[16:25:23] <zefklop> again, BG1
[16:25:47] <fuzzie> well, changing the number seems like not a good fix :)
[16:25:52] <fuzzie> there are some pretty big doors some places
[16:25:54] <zefklop> no...
[16:26:25] <zefklop> yes, in fact we consider that only three tiles are visible
[16:26:39] <zefklop> even for sidewalls this is false
[16:26:53] <Avenger> well, the sidewall should only block things BEHIND it
[16:26:59] <zefklop> yes
[16:27:17] <Avenger> so, as long as it remains sidewall, it should stay visible
[16:27:31] <fuzzie> so i guess someone should make Avenger's change, and then the rest is a bug
[16:27:39] <Avenger> that means, you should always be able to see the door
[16:27:42] <fuzzie> i am more intrigued by the action bar :)
[16:27:48] <zefklop> yup
[16:28:31] <Avenger> heh, i couldn't close a door, which has its locked flag set, i guess that's right, in pst :(
[16:28:47] <fuzzie> yes, i think i deliberately changed that
[16:29:36] <Avenger> its ok, i just couldn't get back to the mortuary by pretending dead, and getting killed removed the key :)
[16:29:53] <zefklop> if (block) {
[16:29:55] <zefklop> Pass--;
[16:29:56] <zefklop> if (!Pass) break;
[16:29:58] <zefklop> }
[16:29:59] <zefklop> Avenger, what's that?
[16:30:10] <zefklop> I think this why we can't see more than three wall tiles
[16:30:33] <fuzzie> yes, it is the 'forced range' i think
[16:30:41] <fuzzie> the portion you can always see
[16:30:43] <Avenger> pass=2 is how many tiles you see behind the sidewall
[16:30:59] <zefklop> ah ok
[16:32:04] <fuzzie> maybe you don't want sidewall to block at all, but instead set something which does 'break' immediately when it stops being sidewall?
[16:32:27] <Avenger> oh the sidewall is breaking see through immediately
[16:32:38] <Avenger> it should just set 'block'
[16:32:59] <Avenger> not break out
[16:33:21] <Avenger> it looks like someone removed by code for sidewalls
[16:34:11] <Avenger> err by->my
[16:37:15] <zefklop> it might be me
[16:37:30] <zefklop> look at rev 5800 and earlier
[16:38:53] <Avenger> yep, there was some sidewall flag
[16:41:33] <fuzzie> but that code was just as bad, i guess
[16:41:36] <Avenger> sidewalls should block when the type changes
[16:41:46] <Avenger> it is a kind of delayed change
[16:41:49] <fuzzie> so i guess we don't need that 48 at all, we just need to check if sidewall was previously set and it changed
[16:42:10] <Avenger> yes, my old code was trying to do that
[16:42:24] <Avenger> doors were not considered as sidewalls
[16:42:29] <Avenger> that was one bug for sure
[16:42:39] <zefklop> the fact is that sidewalls are there even if the actor is behind it
[16:42:46] <fuzzie> well, before zefklop changed it, it just did 'block=true' on sidewalls
[16:43:08] <fuzzie> i mean, after
[16:43:19] <zefklop> that was the purpose of the 48, to avoid blocking sight when being beind the side wall
[16:43:28] <zefklop> but that was as buggy after all
[16:44:38] <Avenger> i think my code was ok about that, but maybe you need to set Pass = 3 or 4 too
[16:45:05] <fuzzie> no, i don't think changing Pass works
[16:45:18] <fuzzie> if you have very long range, that very long range should still work if you are behind a side wall
[16:45:33] <Avenger> well, reverting back to sidewall flag is a must :)
[16:45:35] <zefklop> yes
[16:45:46] <fuzzie> well, the current code is better than the one with the sidewall flag :/
[16:45:47] <Avenger> that was good
[16:45:48] <zefklop> I'll revert 5800
[16:46:06] <fuzzie> i wouldn't revert 5800 without a fix, it makes bg2 really difficult to play sometimes
[16:46:10] <zefklop> and make some code to see if we are behind the wall
[16:46:13] <Avenger> why fuzzie? it will keep visibility as long as you stay in sidewall
[16:46:35] <zefklop> don't worry, I have some ideas to fix t properly :-)
[16:46:46] <Avenger> it will set block as long as it stops being sidewall
[16:46:52] <fuzzie> Avenger: but if the actor pos is in a SIDEWALL tile, the actor can never see
[16:46:59] <fuzzie> because the block happens as soon as you go outside the sidewall
[16:47:03] <Avenger> hmm
[16:47:05] <fuzzie> so your actors are blind :)
[16:47:09] <Avenger> true
[16:47:18] <fuzzie> that is why zefklop changed it originally, because we couldn't see in some areas
[16:47:22] <Avenger> ok, then sidewall should be a 3 way field
[16:47:39] <zefklop> or just check if pos is in the tile
[16:47:57] <fuzzie> but we also don't want to be able to see through a wall, even if we're in the wall tile..
[16:48:00] --> tombhadAC has joined #gemrb
[16:48:05] <fuzzie> so i'm confused as to how this could work in the original game
[16:48:14] <Avenger> hmm
[16:48:16] <zefklop> some direction check?
[16:48:58] <Avenger> something blocks you walking through the wall, and you cannot walk into the sidewall itself, can you?
[16:49:14] <zefklop> you can walk behind it
[16:49:30] <Avenger> but that is not a sidewall searchmap tile :)
[16:49:42] <fuzzie> it must be, otherwise it would have worked before zefklop changed it
[16:49:44] <Avenger> if you walk behind it, and the wall covers you, that is a wallgroup thing!
[16:50:03] <Avenger> you cannot step on a sidewall tile
[16:50:07] <Avenger> normally, at least
[16:50:19] <fuzzie> maybe we are calculating the wrong tile?
[16:50:22] <Avenger> maybe one of the personal space tiles went over it
[16:50:27] <zefklop> hmm... then when drawing we shoud chack for wallgroups and not tiles?
[16:50:36] <Avenger> no zefklop, searchmap is ok
[16:50:46] <zefklop> so?
[16:50:56] <Avenger> i think the problem is about that you occupy 4 searchmap tiles
[16:51:10] <Avenger> and probably the wrong one was picked for visibility
[16:51:10] <fuzzie> ExporeMapChunk is called with actor->Pos but we are probably picking the wrong tile position
[16:51:43] <fuzzie> yes, GetBlocked divides the value without adjusting it
[16:51:47] <Avenger> part of the actor can bump into the wall, as you are able to walk across 1 tile wide paths
[16:52:01] <fuzzie> i tried fixing that sometime but that causes other bugs, i think
[16:52:03] <zefklop> so calculate with Pos+1tile?
[16:52:31] <zefklop> pos.x+16, pos.y+12 ?
[16:52:34] <fuzzie> it is not personal space, because we call GetBlocked without a size
[16:52:37] <Avenger> not sure, but put back the sidewall code and the door as sidewall change
[16:52:45] <zefklop> yes
[16:52:47] <fuzzie> yes, i guess you must revert the sidewall code
[16:52:49] <fuzzie> not all of 5800 though!
[16:52:55] <fuzzie> just that bit
[16:53:03] <Avenger> and then check which tile is taken as origin for visibility
[16:53:19] <Avenger> maybe it tries to look out from inside the wall :)
[16:53:22] <fuzzie> zefklop: it is -8 and -6
[16:53:41] <Avenger> - ?
[16:54:08] <fuzzie> well, that is how you reverse the calculation of the pathfinder
[16:54:08] <zefklop> we should be very pessimistic... what about an actor being in the corner of the tile?
[16:54:36] <fuzzie> that only happens when actors are moving, which has other problems
[16:56:03] <Avenger> well, maybe we should pick ALL tiles of the personal space
[16:56:04] <zefklop> eh, walking should not mmake us blind :-)
[16:56:38] <fuzzie> i don't think it ever would make you blind, due to personal space
[16:57:11] <Avenger> even the original game lets you see yourself, when you are blind
[16:57:20] <Avenger> i didn't try to do it with a dragon, though
[16:57:28] <fuzzie> i guess we should, sometime
[16:57:42] <Avenger> but i suspect you would see all tiles covered by the dragon body :)
[16:58:33] <Avenger> this will cause some slowdown, i guess
[16:58:42] <zefklop> oops, I forgot the door bit in commit log...
[16:58:47] <fuzzie> i think this code is already the slowest thing outside SDL
[16:59:23] <Avenger> hmm, we got commit log? CIA is sleeping
[16:59:35] <CIA-22> gemrb: 03zefklop * r6792 10/gemrb/trunk/gemrb/plugins/Core/Map.cpp: Revert part of r5800.
[16:59:35] <zefklop> no, my connexion is sloooow
[16:59:36] * Avenger kicks CIA-22
[16:59:36] <CIA-22> ow
[16:59:37] <fuzzie> because SearchMap is a bmp, and the GetPixelIndex of BMPImp is really slow
[16:59:53] <Avenger> hehe, woke up
[17:00:01] <zefklop> asm coding would be great there
[17:00:10] <fuzzie> asm coding doesn't help, it is a stupid design problem
[17:00:27] <Avenger> yea, right, asm :) that would kill our portability completely
[17:00:32] <fuzzie> we should copy the searchmap into a structure that we can do fast lookups in
[17:00:32] <zefklop> avenger: poor cia was ok, it is my upload rate which is to blame here
[17:01:01] <zefklop> with C++ backend, of course :)
[17:01:05] <fuzzie> otherwise it does multiplications and shifts and etc
[17:01:11] <Avenger> yep, i thought you already committed something before
[17:01:26] <fuzzie> zefklop: you'd have to code asm for x86, x64, arm, powerpc and sparc, at minimum :p
[17:01:37] <zefklop> meh :-'
[17:01:55] <zefklop> do we have any sparc user?
[17:01:56] <fuzzie> but it would be fine in C++, if it were in a fast array
[17:02:11] <fuzzie> oh, we have a mips user too :)
[17:02:19] <fuzzie> so that is 6
[17:03:02] * zefklop is glad to see how cosmopolitan our user base is
[17:03:52] <fuzzie> well, people on x86/x64 can just use wine, so that makes other people much more interested, i guess
[17:04:19] <zefklop> so, a pixel matrix with just the good size, without any shifts?
[17:04:35] <Avenger> well fuzzie, if you want, you can convert the searchmap to some internal format
[17:04:47] <Avenger> it will take over some memory, but not much
[17:04:51] <fuzzie> yes, an internal pixel matrix of 'unsigned char's is probably best
[17:04:58] <zefklop> just like getpixel(x,y){return matrix[x*rows+y];} ?
[17:05:00] <fuzzie> but i don't think optimisation is to worry about yet :)
[17:05:45] <zefklop> for searchmamp, it is 4 bits values I think
[17:05:56] <fuzzie> usually, yes, which is why it's so slow
[17:06:38] <Avenger> hmm, no
[17:06:40] <zefklop> I don't see waht can be faster
[17:06:49] <Avenger> i think we convert the searchmap to 8 bits
[17:06:59] <Avenger> that's how we have space for the dynamic part
[17:07:01] <Avenger> no?
[17:07:03] <fuzzie> oh, interesting
[17:07:17] <fuzzie> i didn't look at anything except the profile, to be honest..
[17:07:26] <fuzzie> i just saw that GetPixelIndex used all the cpu
[17:08:25] <zefklop> or use some external library which is already uberoptimised
[17:08:33] <Avenger> it could be faster if it is read into a byte array
[17:08:34] <zefklop> do we realy want it?
[17:08:40] <Avenger> no zefklop
[17:08:53] <fuzzie> i don't think there is anything faster than getpixel(x,y){return matrix[x*rows+y];} :)
[17:08:55] <zefklop> that's what I think too :)
[17:09:02] <Avenger> we could simply read it into a specialised storage, and don't use bmp
[17:09:15] <fuzzie> but, yes, we should read it into a byte array and simply do that
[17:09:26] <zefklop> yeah, once we have the matrix, we don't need the header stuff
[17:09:28] <fuzzie> but this is just a speed thing, and we probably don't want to add bugs at the moment :)
[17:09:36] <zefklop> :-)
[17:11:31] <zefklop> avenger: where is the wallgroup code?
[17:11:49] <fuzzie> i don't think the wallgroups matter for this, they're just used for drawing
[17:11:54] <Avenger> spritecover is doing it
[17:11:55] <fuzzie> or i hope so, anyway :/
[17:12:11] <Avenger> but i agree with fuzzie, don't touch wallgroups for visibility :)
[17:12:13] <zefklop> Avenger: thanks
[17:12:40] <zefklop> no no, just knowing that we are behind it would help.
[17:13:06] <zefklop> I don't know how, but that is a thing to think about
[17:15:13] <fuzzie> strange, GetActionButtonRow seems to just check IE_CLASS
[17:16:39] <zefklop> this should be good no?
[17:19:19] <fuzzie> well, i guess that means that IE_CLASS is wrong .. it's '2', i don't know what that means
[17:21:35] <fuzzie> oh, maybe i found it
[17:22:13] <zefklop> dltcep says fighter
[17:22:28] <fuzzie> yes, the core makes a bad function call, it turns out :)
[17:24:56] <Avenger> where?
[17:25:06] <fuzzie> you broke it in r6744
[17:25:11] <fuzzie> "don't overwrite customised action buttons"
[17:25:19] <Avenger> you shouldn't :)
[17:25:28] <fuzzie> because if we don't overwrite customised action buttons, then the buttons never change..
[17:25:32] <fuzzie> even if chargen changes the class!
[17:25:36] <Avenger> but they should
[17:25:43] <Avenger> by default the first value is -1
[17:25:55] <Avenger> it should notice that, and initialise the buttons
[17:26:06] <fuzzie> but thern the chargen changes it again..
[17:26:10] <fuzzie> .. and the core doesn't fix the buttons
[17:26:45] <Avenger> hmm, you mean, the buttons get somehow initialised before the chargen sets the class?
[17:27:34] <fuzzie> no, but you can go back and change the class in chargen
[17:27:43] <Avenger> ahh, i see
[17:28:31] <Avenger> well, maybe we should set the class to 0, and somewhere the code should set the button back to -1 if the class is 0?
[17:28:56] <Avenger> no class, no buttons
[17:28:59] <Avenger> maybe
[17:29:11] <fuzzie> there's a default set of buttons for class 0, i think
[17:30:04] <fuzzie> but pcf_class has the oldValue, so maybe we can simply check if that was 0
[17:30:09] <Avenger> yes
[17:30:13] <Avenger> that's a good idea!
[17:30:24] <Avenger> but pcf is only about modified values?
[17:30:31] * Avenger sighs.
[17:30:52] <fuzzie> i think it works anyway :) i'll try it
[17:30:56] <Avenger> k
[17:31:25] <fuzzie> it is nice to have the commit message so I can see why it changed!
[17:31:52] <Avenger> well, i knew it might break something, as it was first working, then not working :)
[17:32:21] <Avenger> i suspected it was broken because of fixing something else
[17:33:08] <Avenger> heh, next time i will write a shooter
[17:33:27] <Avenger> i guess that's simpler :)
[17:34:11] <Avenger> testing pst projectiles is difficult, some projectiles seem to crash when used in items, but they work in spells
[17:34:24] <Avenger> totally crappy code
[17:34:39] <zefklop> meh, this wall thing is a brain killing one
[17:34:48] <Avenger> i suspect it looks back into the spell struct
[17:35:25] <zefklop> too bad pst wasn't so modded
[17:35:30] <Avenger> zefklop stop before you smash your head into it :)
[17:35:37] <zefklop> we would have a lot more info
[17:35:48] <Avenger> well, pst isn't modded because it is hard to mod
[17:35:48] <fuzzie> there is a lot of info in the last week or two, on the g3 forums
[17:35:55] <zefklop> yeah, but it's areally annoying bug
[17:37:03] <Avenger> it is difficult to have a new spell, when the projectile code itself has half of the effects, i don't think it is easy to add a new npc either
[17:37:12] <Avenger> so, not much to do
[17:37:46] <zefklop> yeah, that is where gemRB will beat it! pst modding :-)
[17:38:06] <fuzzie> well, the pst fixpack people just started patching the exe heavily, heh
[17:38:35] <zefklop> meh, you have competitors
[17:39:52] <zefklop> wait, walking behind walls work with BG1
[17:40:10] <zefklop> why the f*ck does'nt it work with BG2???
[17:40:51] <zefklop> BTW, do you have contact with the pst fixpack team?
[17:42:57] <zefklop> so, it's time to have dinner
[17:42:59] <fuzzie> well, we post on their forum sometimes
[17:43:00] <zefklop> see you
[17:46:44] <fuzzie> damn, the pcf is never called because the base stats aren't initialized
[17:47:17] <fuzzie> the buttons are initialized before we even get a chance to set IE_CLASS, because the core does JoinParty immediately when making the actor for chargen
[17:50:54] <fuzzie> perhaps i can simply check InternalFlags&IF_INITIALIZED but i don't know how the iwd2 thing works
[17:51:37] <Avenger> hehe fun
[17:51:53] <Avenger> the initialisation order is a difficult thing
[17:57:47] <fuzzie> i think any ordering to do with pcfs and stats and things is difficult
[19:25:01] --> kettuz has joined #gemrb
[19:28:16] <zefklop> Avenger: is it possible to remove the associated function of a button in guiscript?
[19:28:54] <Avenger> you don't want to disable the button?
[19:30:14] <zefklop> no, I want it to do nnothing
[19:30:55] <zefklop> if I could remove the var associated that would be perfect too
[19:30:56] <Avenger> well if you disable the button, it will do nothing
[19:31:10] <Avenger> but you can assign "" as function
[19:31:16] <zefklop> will it be clickable?
[19:31:30] <zefklop> will try that
[20:02:44] <lynxlynxlynx> i'm home :)
[20:03:27] <lynxlynxlynx> re beta: no bugs/no pain -> infinity
[20:05:58] <zefklop> could someone tell me how many frames have animations cycles in BG2?
[20:06:10] <zefklop> I believe it is 8 or 9
[20:08:23] <lynxlynxlynx> i think it varies by type
[20:09:06] <zefklop> for characters animations
[20:09:07] <lynxlynxlynx> check a few different anims in dltcep (when you get it back)
[20:09:22] <zefklop> meh, I lacks BG2 now...
[20:09:25] <lynxlynxlynx> like a dragon and an elf
[20:09:28] <Avenger> i don't understand the question
[20:09:29] <fuzzie> there is no fixed amount
[20:09:56] <fuzzie> it differs by animation, and for instance speed depends on this
[20:09:57] <zefklop> hmm... I think for PC this is 9
[20:10:01] <fuzzie> we should really move that code into the right place sometime
[20:10:11] <fuzzie> zefklop: if you don't shapeshift or anything..
[20:10:19] <Avenger> ahh, i understand now
[20:10:20] <fuzzie> there is really no fixed value you can rely on
[20:11:38] <fuzzie> the standard PC anims are indeed 9
[20:11:39] <zefklop> when doing the code for launching projectiles, I had claculated that it was launched at GetFramesCount()/2
[20:11:48] <fuzzie> but you can't hardcode that, i mean
[20:11:54] <zefklop> but then I had worked on BG2
[20:12:18] <zefklop> on BG1 the projectile is launched at fourth animation
[20:12:28] <fuzzie> ah, you think it is simply hard-coded at 4?
[20:12:33] <zefklop> yes
[20:12:34] <fuzzie> that might make sense
[20:12:57] <zefklop> that would be nice if you could check on PST and IWD
[20:13:18] <fuzzie> i can't right now, but i could extract bam files from them
[20:14:18] <fuzzie> i would just commit and comment it, and someone can check later
[20:14:26] <fuzzie> pst animations are completely broken anyway
[20:15:49] <zefklop> ok... if someone is unhappy with it, tou can undo the change
[20:16:03] <zefklop> tou->you
[20:27:28] <zefklop> hmmm, I'm quite of tired... it seems that it is the eighth frame in BG1
[20:27:49] <zefklop> and I can't remember how many frames the BG2 launching projectiles have
[20:28:34] <fuzzie> if you can give me a bam name, i can extract one
[20:31:39] <zefklop> CEFC1SA is a good start :-)
[20:33:42] <-- Gekz has left IRC (Read error: 104 (Connection reset by peer))
[20:33:47] --> Gekz has joined #GemRB
[20:33:48] <fuzzie> ok
[20:35:06] <fuzzie> no such file, but http://fuzzie.org/nfs/gemrb/bam/cefc4sa.bam exists..
[20:36:46] <zefklop> thanks, this one is perfect
[20:37:43] <zefklop> ok, it is launched at the 9th frame in BG2
[20:37:46] <zefklop> as in BG1
[20:38:16] <Avenger> didn't you say 4?
[20:38:31] <fuzzie> we were calculating on the wrong animations
[20:39:03] <zefklop> I was mistaken, this is why in our BG2 this seems broken
[20:39:29] <zefklop> and in fact, it is
[20:39:30] <Avenger> iwd2 has the same file
[20:39:45] <Avenger> cfc4sa
[20:40:20] <zefklop> the only problem is that anim->GetCurrentFrame() returns the current frame in term of real frame
[20:40:31] <fuzzie> hm, yes, iwd2's is identical
[20:40:37] --> zipi has joined #GemRB
[20:41:09] <zefklop> and I'd want the absolute frame count (meaning the frame number + the nuber of times it has been repeated)
[20:41:16] <zipi> hi
[20:41:21] <zefklop> hi
[20:42:27] <zipi> can anyone contribute to the project?
[20:42:46] <-- kettuz has left IRC ("Leaving")
[20:42:48] <zipi> with some code...
[20:42:50] <Avenger> this depends mostly on them
[20:42:58] <Avenger> i mean, if the code is good, sure
[20:43:30] <zipi> i've tried to implement the use of keymap.ini
[20:43:38] <fuzzie> oh, interesting :)
[20:43:43] <Avenger> yeah :)
[20:43:56] <Avenger> do you have a sourceforge account?
[20:44:00] <zipi> yes
[20:44:04] <fuzzie> that one is difficult to get right for all the games, maybe
[20:44:09] <fuzzie> but anything would be a start
[20:44:18] <zipi> yeah, i have some ideas, it is not yet "finished"
[20:44:46] <fuzzie> oh, right, you filed the videos bug yesterday too :)
[20:44:52] <zipi> :)
[20:44:55] <Avenger> if you think it is good enough, then please submit it as a patch on sourceforge
[20:45:09] <zipi> ok, thanks
[20:48:51] <fuzzie> that action bar bug is annoying
[20:49:38] <Avenger> hmm, hacking on the pcf didn't help?
[20:49:54] <fuzzie> the pcf never gets called, because IF_INITIALIZED never gets set during chargen
[20:51:03] <fuzzie> I can add a forced call to InitButtons somewhere (for instance, when IF_INITIALIZED is set), but thatg seems like a hack
[20:51:09] <Avenger> but you don't really need the buttons in chargen
[20:51:49] <Avenger> the buttons shouldn't be initialised with class=0, maybe
[20:52:11] <fuzzie> apparently class=1 by default, i don't know why
[20:53:42] <fuzzie> well, i guess it is that in charbase
[20:53:55] <fuzzie> but that is not very helpful
[20:55:58] <Avenger> ok, try to reinit it in FillPlayerInfo
[20:56:04] <Avenger> that should work
[20:56:41] <fuzzie> ok
[20:56:45] <Avenger> fillplayerinfo is a good place to do chargen specific hacks
[20:57:05] <fuzzie> well, i think the iwd2 thing is probably the hack here
[20:57:17] <fuzzie> because if the class changes, the action bar should be updated - for instance when multiclassing
[20:57:49] <Avenger> iwd2 doesn't change it
[20:57:55] <fuzzie> but i don't know how iwd2's action bar works, so i can't fix that
[20:58:00] <Avenger> when you start the game, it sets the buttons up
[20:58:07] <Avenger> then you are on your own to customise it
[20:58:26] <fuzzie> so maybe we need a GF_
[20:58:40] <Avenger> there is a problem with dual classing
[20:58:49] <Avenger> yes, probably it needs one
[21:00:36] <zefklop> fuzzie: if it can help you, buying weapons and equipping them restores the actions bar
[21:01:22] <fuzzie> that is just weirder
[21:01:48] <zefklop> :/
[21:02:42] <fuzzie> it happens in all the other games too, by the way
[21:02:45] <fuzzie> it is not bg1-specific
[21:04:08] <CIA-22> gemrb: 03zefklop * r6793 10/gemrb/trunk/gemrb/plugins/Core/Actor.cpp: Launch projectile at the right time for BG1, BG2 and probably others. This is set to the 9th frame... so be it, you hardcoded thing. Thanks bioware, this seems to be shared across all of the games.
[21:04:26] <zefklop> so I guess that you're searching on the C++ side
[21:04:43] <fuzzie> yes, it happens before the chargen even gets the actor setup
[21:05:23] <fuzzie> but if equipping weapons restores the action bar, perhaps i have the wrong idea
[21:08:03] <zefklop> well, buy bow and arrows, try to equip them, anjoy
[21:08:21] <zefklop> you noticed that the inventory was broken too?
[21:09:59] <fuzzie> still broken for me after i equip new weapons
[21:11:30] <zefklop> which ones? bows have class restrictions
[21:11:35] <fuzzie> and the quick weapons in the inventory are broken because that code just checks the quickslots
[21:11:49] <fuzzie> i equipped a bow, arrows and an axe, and the action bar is still the mage one
[21:12:00] <zefklop> weird
[21:12:04] <fuzzie> it works if you save/load, but that is no surprise
[21:12:51] <zefklop> yes, my mistake... I think it is time to sleep for me
[21:13:10] <fuzzie> thankyou for help :)
[21:15:13] <lynxlynxlynx> zipi: nice
[21:15:19] <lynxlynxlynx> & gn
[21:15:30] <-- lynxlynxlynx has left IRC (Remote closed the connection)
[21:16:04] <zefklop> meh, I thought that the 9th frame thing would work for spell too
[21:16:31] <zefklop> and no, they don't even have that number
[21:16:42] <zefklop> so, good night!
[21:16:47] <-- zefklop has left IRC ("ChatZilla 0.9.85 [Firefox 3.0.8/2009032609]")
[21:24:11] --> pupnik has joined #gemrb
[21:24:30] --> tombhad-AC has joined #gemrb
[21:38:21] <-- tombhadAC has left IRC (Read error: 110 (Connection timed out))
[21:40:22] <-- pupnik_ has left IRC (Read error: 101 (Network is unreachable))
[21:51:40] <Avenger> bye
[21:51:41] <-- Avenger has left IRC ("ChatZilla 0.9.85 [Firefox 3.5.1/20090715094852]")
[23:35:41] <-- dawid has left IRC ("Not leaving, quitting.")