#gemrb@irc.freenode.net logs for 4 Oct 2009 (GMT)

Archive Today Yesterday Tomorrow
GemRB homepage


[01:09:33] <-- barra_away has left IRC ("Verlassend")
[01:40:04] <-- Gekz has left IRC (Read error: 104 (Connection reset by peer))
[02:10:23] --> Darki has joined #gemrb
[02:10:23] --- ChanServ gives channel operator status to Darki
[02:20:59] <-- Dark-Star has left IRC (Read error: 110 (Connection timed out))
[02:41:15] --> Forgetful_Lion has joined #GemRB
[03:47:57] <-- Maighstir has left IRC (Read error: 110 (Connection timed out))
[05:21:23] --> Dark-Star has joined #gemrb
[05:21:24] --- ChanServ gives channel operator status to Dark-Star
[05:34:11] <-- Darki has left IRC (Read error: 110 (Connection timed out))
[05:48:21] <-- Forgetful_Lion has left IRC (" HydraIRC -> http://www.hydrairc.com <- Po-ta-to, boil em, mash em, stick em in a stew.")
[07:31:43] --> barra_library has joined #gemrb
[09:48:29] <fuzzie> huh, what's an fx_ignore_dialogpause_modifier? madness
[09:57:07] <-- |Cable| has left IRC (kornbluth.freenode.net irc.freenode.net)
[09:57:07] <-- wjp has left IRC (kornbluth.freenode.net irc.freenode.net)
[09:57:32] --> wjp has joined #GemRb
[09:57:32] --> |Cable| has joined #GemRb
[10:10:52] --> Avenger has joined #gemrb
[10:11:38] <fuzzie> hi Avenger
[10:13:18] <Avenger> hi
[10:13:24] --- ChanServ gives channel operator status to Avenger
[10:14:20] <Avenger> i'm fixing all GoNearAndRetry stuff, using MoveNearerTo. But it doesn't work for some cases, i don't know why.
[10:14:38] <Avenger> it works for picklock/toggledoor/disarm
[10:14:45] <fuzzie> cool
[10:14:47] <Avenger> it doesnt' work for escapearea
[10:14:54] <Avenger> i don't know why
[10:15:00] <fuzzie> didn't i deliberately break escapearea?
[10:15:09] <fuzzie> yes, i did
[10:15:11] <Avenger> i dropped that stuff out
[10:15:24] <fuzzie> that whole EscapeAreaCore function needs deleting
[10:15:26] <Avenger> that was using GoNearAndRetry
[10:16:00] <Avenger> why?
[10:16:12] <fuzzie> because we should never be making new actions
[10:16:22] <fuzzie> and if we don't make new actions, there's not much it can do
[10:16:39] <Avenger> sure there is :)
[10:16:48] <fuzzie> you just have to set a timer in CurrentActionState, start walking somewhere (maybe to an exit?) and then give up after a while, and almost all of that code is specific to one action
[10:17:17] <Avenger> there are some 3-4 escapearea actions
[10:17:43] <fuzzie> EscapeAreaObject should go to a scriptable, not coordinates
[10:17:47] <Avenger> yes
[10:18:13] <fuzzie> i guess it's slightly nicer if you share the code, but it seems so .. silly
[10:18:25] <fuzzie> anyway
[10:18:36] <fuzzie> your problem is probably that the EscapeArea actiosn aren't AF_BLOCKING
[10:18:43] <Avenger> Oh
[10:18:44] <fuzzie> see the big array in GameScript.cpp, i guess you wrote all that :)
[10:18:55] <Avenger> !%^^^$*$%&*$#
[10:19:35] <Avenger> i should have come in earlier
[10:20:18] <Avenger> it is not sure this will work, but the result of missing that flag is exactly what i experienced :)
[10:20:23] <fuzzie> do you think you can get rid of GoNearAndRetry completely, now?
[10:20:30] <fuzzie> that would help a lot
[10:20:30] <Avenger> i'm working on it
[10:20:41] <Avenger> i think i removed almost all or all
[10:20:50] <fuzzie> i want to fix the visibility for script objeccts, but i have to check how it works for non-actors in the original game first
[10:21:05] <fuzzie> but that is today's plan, unless you already know how that works
[10:21:16] <Avenger> there is a forced GoNear in dialog
[10:21:22] <Avenger> that is not exactly the same
[10:21:44] <Avenger> i don't know what's wrong with the visibility
[10:22:03] <fuzzie> we just don't check it right now
[10:22:19] <fuzzie> it's simple to fix for actors, but i don't know how it works for scriptables
[10:23:27] <fuzzie> Avenger: do you know if VVCs get drawn after death?
[10:23:59] <fuzzie> lynx was wondering whether that is the bug with effects drawn after death, or whether effects should self-destruct and remove the VVCs
[10:24:21] <fuzzie> oh, also, CreateVisualEffectObjectSticky never goes away
[10:24:30] <fuzzie> it just loops forever
[10:24:32] <Avenger> some effects go away
[10:24:55] <Avenger> blur and mirror image surely go away, but those are not vvc based
[10:25:14] <fuzzie> if that is intentional, then "stickysinisterpoof" shouldn't be pointing at it.
[10:25:38] <Avenger> hehe, that sounds so PST
[10:25:43] <fuzzie> it is, of course :)
[10:25:55] <fuzzie> i try and fix pst a bit
[10:26:02] <fuzzie> it is difficult, because stances are so broken
[10:26:09] <Avenger> yep
[10:26:16] <Avenger> btw, try embalm
[10:26:22] <Avenger> it had no visual before
[10:26:25] <Avenger> i didn't know why
[10:26:33] <Avenger> now i fixed something with the embalm projectile
[10:26:37] <Avenger> but didn't test
[10:26:51] <Avenger> so it could have a visual now
[10:26:58] <fuzzie> hm
[10:27:03] <Avenger> kiss too, but i guess you don't have grace in the party
[10:27:09] <fuzzie> i have grace
[10:27:24] <fuzzie> dammit, more guiscript broken
[10:27:24] <Avenger> hmm, then try that too, any projectile in the end of the list would do
[10:27:32] <fuzzie> people should test pst when changing guiscripts :(
[10:27:46] <Avenger> i don't remember when i did that
[10:27:53] <Avenger> or is it some global stuff?
[10:28:10] <fuzzie> i don't know
[10:28:18] <fuzzie> lynx broke some stuff inside pst/, i had to fix that
[10:28:23] <fuzzie> let me work out who is to blame this time :)
[10:28:35] <Avenger> well i try to fix the GoNear stuff
[10:28:39] <fuzzie> thanks!
[10:28:41] <Avenger> now with the flag turned on, it works
[10:28:46] <fuzzie> that has been on my todo for a while, but i never got to it
[10:28:50] <Avenger> there is no wait time
[10:28:57] <Avenger> that's the worst now
[10:29:04] <Avenger> but that's much better than before
[10:29:06] <fuzzie> well, i can fix that, certainly
[10:29:16] <Avenger> i still call generate action though :P
[10:29:22] <fuzzie> heh, i can also fix that :-)
[10:29:37] <fuzzie> i have some savegames for test cases for EscapeArea
[10:29:50] <Avenger> i test with glaicus
[10:29:52] <fuzzie> ok, so lynx broke this one too
[10:30:02] <Avenger> charm him in nalia's keep
[10:30:08] <Avenger> that shows charm is buggy too
[10:30:41] <fuzzie> anyway, do you have an idea where to start with pst stances?
[10:30:44] <Avenger> ahh and i got a government district test case too
[10:30:56] <Avenger> no, sorry, i think we shouldn't focus on pst too much yet :)
[10:30:58] <fuzzie> do i just increase the max stances, add some checks for non-pst, and then start adding to the anim code?
[10:31:11] <Avenger> i think the stances shouldn't have this fixed setup
[10:31:15] <fuzzie> well, pst is interesting for me :)
[10:31:15] <Avenger> it just doesn't work
[10:31:33] <Avenger> pst has some 70 stances or so
[10:31:40] <Avenger> we cannot cache all of them
[10:32:03] <Avenger> a lof of the stances are not used, or duplicate
[10:32:45] <Avenger> and the stance numbers are not the same in the engine versions
[10:32:52] <Avenger> so we need a translator table
[10:33:39] <Avenger> to be more precise: there are 78 stances in pst
[10:33:44] <Avenger> see animstat.ids
[10:34:28] <fuzzie> meh
[10:34:41] * fuzzie sighs
[10:35:00] <Avenger> iwd2 has 17: sequence.ids
[10:36:48] <Avenger> tob has 17 too, but they are differently numbered
[10:36:56] <Avenger> sucks
[10:37:11] <Avenger> tob has them in seq.ids
[10:38:20] <fuzzie> wow, pst segfaults when i try kissing
[10:38:29] <Avenger> sucks
[10:38:38] <Avenger> i guess somewhere in projectile
[10:38:47] <Avenger> or maybe the life drain opcode
[10:39:08] <Avenger> that was never tested before
[10:39:15] <fuzzie> heh, it could be anywhere, i will blame my code first, let's see
[10:40:01] <CIA-22> gemrb: 03fuzzie * r7354 10/gemrb/trunk/gemrb/GUIScripts/pst/GUICommonWindows.py: revert r7284 (global pdoll) for pst, since LoadCommonTables doesn't work
[10:40:10] <fuzzie> Avenger: so how do I make an action play a VVC only once?
[10:40:10] <Avenger> ok, commented out GoNearAndRetry, and now trying to Compile
[10:40:20] <Avenger> vvc->PlayOnce()
[10:40:39] <Avenger> that should work :)
[10:41:07] <fuzzie> ok, the kiss breaks in the bam play
[10:41:22] <fuzzie> 0x0eca09d4 in fx_play_bam_not_blended (Owner=0x1027e378, target=0x1027e378, fx=0x13cd4dc0) at PSTOpc.cpp:225
[10:41:25] <fuzzie> 225 sca->twin->ZPos+=9999;
[10:41:26] <fuzzie> ^- sca->twin is NULL!
[10:41:51] <Avenger> hmm
[10:41:59] <CIA-22> gemrb: 03avenger_teambg * r7355 10/gemrb/trunk/gemrb/plugins/Core/ (5 files): removed all (almost all) traces of GoNearAndRetry (kept in comments for documentary reasons only)
[10:42:02] <fuzzie> i don't understand a word of the code, maybe you do
[10:42:03] <Avenger> it shouldn't call into the twin
[10:42:21] <Avenger> twin vvcs are a weird business :)
[10:42:39] <fuzzie> i will just wrap it in a check for now, i guess
[10:42:46] <Avenger> they are sticky vvcs, with one half of the animation played before and one half played behind the actor
[10:42:52] <Avenger> like firewall
[10:43:08] <Avenger> all single vvcs have their twin set as 0
[10:43:11] <fuzzie> oh dear, the next bit makes no sense
[10:43:24] <Avenger> so, i don't quite see how it runs into the twin
[10:43:42] <fuzzie> i guess this code never worked
[10:44:21] <Avenger> oh i see
[10:44:32] <fuzzie> i change it to be sca->ZPos in both places it looks wrong?
[10:44:37] <Avenger> some code says it is a twin anim, but then it isn't
[10:45:04] <Avenger> huh i see what you mean
[10:45:09] <Avenger> and it is truly crap
[10:45:26] <Avenger> the 'both' stuff is definitely wrong
[10:45:56] <Avenger> that should be only one twin
[10:46:09] <Avenger> and the foreground probably needs a guard
[10:46:22] <Avenger> or no twin at all
[10:46:26] <Avenger> i don't know which
[10:46:32] <Avenger> no twin, i guess
[10:47:09] <Avenger> i fixed it
[10:47:21] <CIA-22> gemrb: 03avenger_teambg * r7356 10/gemrb/trunk/gemrb/plugins/PSTOpcodes/PSTOpc.cpp: too many twins
[10:47:24] <fuzzie> just changing both those twin lines to remove '->twin' stops the crash
[10:47:33] <Avenger> yep
[10:47:42] <fuzzie> i don't see any graphic rendered, though
[10:47:50] <fuzzie> but maybe i just don't see it
[10:47:53] <Avenger> well
[10:49:04] <Avenger> this is a complicated effect :)
[10:50:33] <Avenger> i tried to force the various animations into the vvc class
[10:50:45] <fuzzie> oh, lynx also modified some iwd summon opcode to fix the EA stuff a bit
[10:50:54] <Avenger> did it work? :)
[10:50:56] <fuzzie> but the same problem is present in the FXOpc summon opcode, at least for iwd
[10:51:07] <Avenger> hmm
[10:51:15] <fuzzie> some of the parameter2 options should preserve the CRE's EA, and we don't do that
[10:51:16] <Avenger> maybe the summon opcode works differently in iwd?
[10:51:20] <fuzzie> IESDP seems correct on this
[10:51:27] <fuzzie> but i only checked in iwd, so..
[10:51:37] <Avenger> well, we have options for ALL cases
[10:52:02] <Avenger> EAM_DEFAULT is preserving the original EA
[10:52:02] <fuzzie> yes, but the eamods[] array doesn't have EAM_DEFAULT anywhere
[10:52:08] <fuzzie> oh, it does!
[10:52:17] <Avenger> it is 2
[10:52:21] <fuzzie> huh, why doesn't it work for fx_summon_creature, then
[10:52:24] <Avenger> and 4
[10:52:29] <fuzzie> oh
[10:52:39] <Avenger> well, what parameter is it in the opcode
[10:52:44] <Avenger> err the effect
[10:52:46] <fuzzie> why did you add an unsummon effect to fx_replace_creature, btw?
[10:52:55] <fuzzie> you unsummon critical actors, i had to disable it
[10:53:22] <Avenger> hehe, i do?
[10:54:16] <fuzzie> i just wondered if you saw it in the original code
[10:54:19] <Avenger> that was crazy
[10:54:22] <Avenger> no
[10:54:25] <Avenger> i just checked it
[10:54:29] <fuzzie> there's one in fx_find_familiar too
[10:54:40] <fuzzie> although i hope that one is harmless
[10:54:52] <Avenger> familiars don't vanish either
[10:55:20] <fuzzie> ok, well, i point you at FXOpc.cpp some more :)
[10:55:31] <Avenger> but let me see it
[10:56:56] <fuzzie> but if anytime you are bored, i would appreciate some thinking about the stances thing
[10:57:06] <fuzzie> i can write code, i just don't know where to start
[10:57:11] <Avenger> the following bg2 effects use unsummon: 043_summon, 07f_randomsummon, 0ec_copyself
[10:57:26] <Avenger> all else in fxopc is most likely craziness :)
[10:57:34] <fuzzie> ok :) that makes it easy
[10:57:46] <Avenger> just replace them with newfx = NULL
[10:58:02] <Avenger> i don't know why i did it
[10:59:22] <fuzzie> oh, another thing: when DLTCEP opens a 2DA file and ignores a column, it calls it a 'harmless inconsistency'
[10:59:36] <fuzzie> it would be nice if it mentioned it was deleting data :)
[10:59:43] <Avenger> should be more like harmful inconsistency?
[10:59:45] <Avenger> :)
[10:59:45] <fuzzie> lynx and I had to export the 2da to work out what was happening
[10:59:50] <fuzzie> harmful would be nice :p
[11:00:03] <Avenger> hehe
[11:00:05] <fuzzie> turns out that iwd's worlde.2da has a broken line
[11:00:16] <Avenger> a broken line???
[11:00:23] <fuzzie> they used a space inside one of the entries :P
[11:00:25] <Avenger> how is that supposed to work in the original
[11:00:28] <Avenger> ahh
[11:00:32] <Avenger> inside a field?
[11:00:35] <fuzzie> so it doesn't work
[11:00:43] <fuzzie> or i assume not, anyway
[11:00:52] <Avenger> i take a peek :)
[11:01:17] <fuzzie> i only wanted the stupid Kuldahar lines to work, and they are fine :P
[11:01:35] <Avenger> destroyed easthaven
[11:01:42] <fuzzie> yes, that's it
[11:01:54] <Avenger> what is this 2da i dn't know it
[11:02:01] <Avenger> this is the missing flags?
[11:02:12] <fuzzie> when the second column variable is set, it sets the first three area flags in the worldmap for the third column
[11:02:21] <Avenger> wow
[11:02:31] <fuzzie> yeah, it is horrible :) i implemented it now
[11:02:34] <Avenger> another text based hack
[11:02:45] <Avenger> blackisle people are sooo.... weird.
[11:02:56] <Avenger> instead of adding an action
[11:03:10] <Avenger> they set a variable, which is handled by the core
[11:03:15] <Avenger> brr
[11:03:56] <Avenger> but they surely don't work with the destroyed easthaven entry?
[11:04:04] <Avenger> i hope
[11:05:35] <fuzzie> we'll see
[11:05:48] <fuzzie> that 2da uses tabs to seperate fields, so maybe they detect tabs, i would hate that
[11:05:50] <Avenger> iwd2 has this too
[11:05:54] <fuzzie> but i hope it doesn't work
[11:06:22] <Avenger> if they detect tabs only, it sucks, because i seen other games to detect spaces
[11:06:42] <fuzzie> other iwd files use spaces
[11:06:48] <Avenger> yep
[11:06:50] <fuzzie> so i think it's likely just broken
[11:06:55] <Avenger> that entry MUST not work :)
[11:07:06] <Avenger> or we will need to implement another hack on this hack
[11:07:34] <fuzzie> but, yes, iwd2 has this too, and i think it's the same thing
[11:07:42] <fuzzie> IESDP claims the opposite but i think it's wrong
[11:08:17] <fuzzie> i finally installed iwd2 on a Windows machine, so i could check the death variables, so i'll check it too later
[11:09:28] <Avenger> hehe, they got it backwards
[11:09:49] <fuzzie> the forums being down was so annoying!
[11:09:56] <fuzzie> i worked out all the death variables stuff, and committed it
[11:10:09] <fuzzie> and a few hours later, the forums come back up, and most of the death variables stuff was already worked out :)
[11:10:15] <Avenger> haha
[11:10:24] <fuzzie> but at least i confirmed it
[11:10:29] <Avenger> but this worlde thing is defintely not right
[11:10:34] <Avenger> in the iesdp
[11:10:38] <fuzzie> yes, someone should post to the IESDP forum about it
[11:11:35] <fuzzie> it is updated when the world map is updated (ie, when you use an exit and the world map appears)
[11:12:31] <Avenger> is that how it works in the original too?
[11:12:36] <fuzzie> yes
[11:13:02] <fuzzie> it is a bit different in iwd2, the flags get updated at startup somehow, i didn't work it out yet
[11:13:02] <Avenger> because i like it, it is not always polling those variables, and it isn't the setvariable which is hacked to work
[11:14:32] <fuzzie> iwd2 has horrible CURRENT_AREA and PREVIOUS_AREA globals which are numbers, though
[11:14:53] <Avenger> hmm, those globals are always set?
[11:15:10] <fuzzie> "lame iwd2 uses a numeric area identifier", like the comment you added to gemrb's CurrentAreaIs :)
[11:15:20] <Avenger> but that is not a global
[11:15:23] <Avenger> that is a trigger
[11:15:28] <fuzzie> the globals are set too
[11:15:33] <Avenger> O_o
[11:15:41] <fuzzie> i went through and worked out where all the globals in my iwd/iwd2 savegames came from, you see
[11:15:42] <Avenger> i asked especially because of that trigger
[11:15:54] <Avenger> if they got a variable, why they needed the trigger
[11:16:18] <Avenger> you can always do Global("CURRENT_AREA",123) no need of a CurrentAreaIs(123)
[11:16:20] <fuzzie> oh, and EXPANSION_DOOR is some horrible hack too
[11:16:23] <fuzzie> horrible horrible horrible
[11:16:40] <fuzzie> i don't understand why it exists
[11:16:41] <Avenger> hmm, i smell another action :)
[11:16:52] <fuzzie> we just hard-coded that one in the guiscript
[11:16:59] <fuzzie> it just has to be set by the engine
[11:17:10] <Avenger> expansion_door is a door hacked into the area, isn't it?
[11:17:19] <fuzzie> no
[11:17:40] <fuzzie> there's already a script for it, it just unlocks the door
[11:17:41] <Avenger> i thought it is an additional door not existing in kuldahar
[11:17:49] <fuzzie> the door is already there, thankfully
[11:18:19] <fuzzie> the reason is that the ARE is probably already in savegames, but the script gets re-loaded every time
[11:18:54] <Avenger> yes
[11:19:05] <Avenger> so you set that global from the script, ok, that's fine
[11:19:10] <Avenger> i didn't know the door existed
[11:19:13] <fuzzie> mhm, it's fine from gemrb side
[11:19:15] <fuzzie> it works fine
[11:20:08] <Avenger> so, back to currentarea, and previous area, anything uses it?
[11:20:26] <fuzzie> not as far as i can tell
[11:21:11] <Avenger> i see no such string in iwd2.exe
[11:21:19] <fuzzie> yeah, me neither
[11:21:21] <fuzzie> not 'current' either
[11:21:23] <fuzzie> hmm
[11:22:31] <Avenger> so, how did you find it?
[11:22:37] <fuzzie> it's in one of my GAM files
[11:22:47] <fuzzie> or, two, with different values
[11:22:52] <Avenger> probably only set by script
[11:22:54] <Avenger> which is fine
[11:23:01] <fuzzie> it's not in any of the scripts
[11:23:04] <Avenger> wow
[11:23:06] <fuzzie> maybe in a dialog, i didn't check those..
[11:23:19] <fuzzie> very weird
[11:23:20] <Avenger> dialog? but dialogs wouldn't be setting an area code
[11:23:27] <Avenger> yeah, weird
[11:23:32] <fuzzie> ok, i move that to the 'weird, must test' column
[11:24:19] <fuzzie> it's set in my pst saves also
[11:24:27] <Avenger> pst does that in scripts
[11:24:50] <fuzzie> yes
[11:24:58] <fuzzie> ick, GlobalSetGlobal?
[11:25:00] <Avenger> just grepped all iwd2 bifs no current_area
[11:25:11] <fuzzie> but in pst, it's always 202
[11:25:12] <Avenger> case insensitively
[11:25:16] <fuzzie> so i guess it's only scripted in the start area
[11:25:34] <fuzzie> ok, i have to restart iwd2 games and see where it comes from
[11:25:37] <fuzzie> maybe iwd2 is modded or something
[11:26:46] <fuzzie> one nice feature to have in DLTCEP would be to only display the GAM variables which are different from the default ones
[11:26:51] <fuzzie> i guess that would be too much work
[11:27:06] <Avenger> are there default ones?
[11:27:11] <fuzzie> yes, the var.var ones :)
[11:27:35] <Avenger> ahh in pst
[11:27:39] <fuzzie> it is just impossible to check the variables for a pst GAM, there are thousands of them in the list
[11:28:01] <Avenger> i can do that
[11:28:32] <fuzzie> it would also be nice if, when i click on 'Edit PC' from a GAM edit, if it detected whether it was edited or not
[11:28:44] <fuzzie> but i guess that is impossible
[11:29:19] <fuzzie> just the message box drives me mad sometimes :)
[11:30:11] <Avenger> it tries to detect it :)
[11:30:24] <fuzzie> but for GAM it is always wrong :p
[11:31:51] <Avenger> i seen it too, but didn't care
[11:38:49] <fuzzie> well, i just mention things which bug me
[11:38:58] <fuzzie> none of it is important
[12:50:18] --> Gekz has joined #GemRB
[12:53:16] <Gekz> I was looking at Boost today
[12:53:21] <Gekz> we were all looking at it the wrong way
[12:53:23] <Gekz> it isnt bloat at all
[12:53:32] <Gekz> and it's extremely portable
[13:09:12] <fuzzie> well
[13:09:15] <fuzzie> boost itself is not bloat
[13:09:21] <fuzzie> but 90% of boost is bloat
[13:09:46] <fuzzie> and when people say 'we should use boost' they mean 'we should use this thing from boost that will make compiles horror'
[13:10:23] <Avenger> when i needed regexp, i simply cut a lot of boost, and still a lot of crap remained in it. It is nice, but it is huge, and it is not easy to get rid of the unneeded parts.
[13:10:28] <fuzzie> i used a bunch of boost things for another project and it's made the whole thing a nightmare
[13:11:40] <fuzzie> just because it takes 5 times the time to compile it because the boost stuff is template-obsessed..
[13:12:25] <fuzzie> things like the filesystem library are nice, but then gemrb already implemented all of that..
[13:14:59] <fuzzie> and 'extremely portable' and 'template-obsessed' are not compatible concepts :)
[13:22:24] <pupnik> ty
[13:22:48] <pupnik> boost was not on my list of friends
[13:23:10] <fuzzie> yes, also a nightmare for embedding :/
[13:25:14] <fuzzie> pupnik: cool that you updated the youtube video to note things were fixed in 0.5.0, btw
[13:26:31] <pupnik> i just got dev env setup again and laptop got roasted by ffox 100% cpu browsing
[13:26:52] <pupnik> ffox will be sorry
[13:40:13] <Gekz> lolol
[13:42:01] <fuzzie> Avenger: i want a game flag for pst-style text, GF_NO_MESSAGEWINDOW or GF_ONSCREEN_TEXT or you have a better idea?
[13:42:19] <Avenger> no better idea yet
[13:43:31] <fuzzie> any preference? :)
[13:44:52] <Avenger> maybe onscreen text is better
[13:44:59] <fuzzie> that is my thought
[13:45:25] <Avenger> i thought about adding some invisible control to the game area
[13:45:31] <Avenger> by guiscript
[13:45:46] <Avenger> but probably that is not how it was done in the original game
[13:45:55] <fuzzie> do you think a control would be better?
[13:46:02] <Avenger> no idea
[13:46:20] <Avenger> i thought i will just ignore pst for a while :)
[13:46:33] <Avenger> but if you would like to improve, it is nice
[13:46:49] <fuzzie> well, i like pst more than anything else :)
[13:47:09] <fuzzie> i am happy fixing bugs for the other games, i just want to keep improving pst a little bit at a time
[13:47:18] <Avenger> well, yes, i just don't like its coding hacks
[13:49:01] <CIA-22> gemrb: 03fuzzie * r7357 10/gemrb/trunk/gemrb/plugins/Core/Actor.cpp: char -> signed char
[13:50:23] <fuzzie> well, i just ignore the projectiles :-) then it is not so bad
[13:51:26] <fuzzie> well, actually, looking at my notes, it's still horrible hard-coded hacks, but meh
[13:51:41] <fuzzie> the fixpack people just make it worse
[13:54:44] <Gekz> how so
[13:54:50] <fuzzie> they hardcode more things :)
[13:55:23] <fuzzie> there's only one hidden door in the game, so they hack the hidden door code to be wired to a specific variable, for instance
[13:56:46] <fuzzie> and they add a boost to bashing if you have the PRYBAR equipped, hard-coded
[13:57:51] <Gekz> lol
[13:58:05] <Gekz> fuzzie: so the programmers of the original engine were absolutely shit?
[13:58:38] <fuzzie> basically, they had to rush it, because the company was firing everyone else and rushing the pst people to get pst finished
[13:58:42] <fuzzie> then, they fired the pst people, of course
[13:59:42] <Gekz> lol nice.
[13:59:53] <Gekz> this was Interplay, right?
[14:00:13] <Gekz> and didnt they all leave and make Troika Games and Obsidian Studios
[14:00:19] <Gekz> whereby Troika sucked balls and died
[14:00:23] <Gekz> and Obsidian still exists?
[14:00:32] <fuzzie> wel, it was Black Isle, and owned by Interplay
[14:00:40] <Gekz> I thought Black Isle just published it
[14:00:44] <fuzzie> apparently Interplay didn't fire all of Black Isle until 2003
[14:00:44] <Gekz> and Bioware programmed it
[14:00:55] <Avenger> obsidian lives on, happily
[14:00:55] <fuzzie> Bioware coded the Infinity Engine itelf
[14:01:03] <Gekz> oh, just the engine
[14:01:04] <Gekz> ok
[14:01:14] <fuzzie> and then Black Isle used it for iwd, iwd2 and pst, changing it a lot for each one
[14:01:15] <Gekz> I thought they took an active role in making Baldur's Gate
[14:01:24] <Avenger> they did something
[14:01:29] <fuzzie> bg1 and bg2 were all Bioware
[14:01:39] <Gekz> ok
[14:01:40] <Avenger> bioware and blackisle/obsidian live in a kind of symbiosis
[14:01:48] <fuzzie> so this is why so many engine differences :)
[14:01:50] <Avenger> i think blackisle had some part in the bg part too
[14:01:55] <fuzzie> the obsidian people left before pst, i think
[14:01:56] <Gekz> they did
[14:01:59] <Gekz> they published it
[14:02:15] <Avenger> also tob got backported some stuff from iwd
[14:02:34] <fuzzie> so it's all a very complicated relationship :)
[14:02:42] <fuzzie> and you had several teams of programmers involved
[14:03:13] <fuzzie> and the pst ones didn't really get it :P
[14:03:14] <Avenger> it would have been a nice story if they don't just flush this nice code down the toilet
[14:03:35] <fuzzie> well, we come closer to reproducing it :)
[14:04:54] <Avenger> it is like a waste of 5+ years. Imagine if we had the source five years ago.
[14:05:21] <Avenger> it would still be nice to see
[14:05:41] <Gekz> wait until you hit version 1
[14:05:47] <Gekz> and you get a takedown request
[14:05:48] <Gekz> hahaha
[14:06:05] <fuzzie> well, they often have a hard time just working out who owns the license to the PS:T game itself
[14:06:12] <fuzzie> never mind anything about source code :(
[14:07:12] <Avenger> well, gemrb is around for years, and it isn't like bioware doesn't know about it
[14:07:30] <fuzzie> do Bioware own all the rights?
[14:07:43] <Avenger> for their own engine, probably
[14:07:50] <Avenger> there are parts, like bink video
[14:07:51] <fuzzie> maybe to the core engine, but i don't think they'd be allowed to release anything which was D&D-ish or was written by Black Isle
[14:08:44] <Gekz> well
[14:08:51] <Gekz> DnD are absolute bitches about their licenses
[14:08:57] <Gekz> so prepare for them to DMCA you
[14:09:14] <Avenger> but i recently read in the multimedia blog, that rad game tools knows about the reverse engineering of bink :)
[14:09:51] <Avenger> yeah, probably, that's why we should avoid coding anything dnd specific in the core :)
[14:10:06] <Avenger> they cannot dmca vvc's and .are format :) and such
[14:10:47] <fuzzie> well, they can't DMCA you based solely on rules, if you don't distribute any tables or anything at all
[14:10:51] <fuzzie> well, maybe in the US, i don't know :)
[14:11:17] <Avenger> if the core is capable of doing any game system, what is the point?
[14:11:34] <fuzzie> you do have to be careful about what you put in override/ though
[14:11:55] <Avenger> yes, nothing that appears in a dnd sourcebook :)
[14:14:58] <fuzzie> does gemrb's .pro format already have a place to specify a spell?
[14:16:04] <Avenger> there is one for failure, but i don't think there is one for normal course
[14:46:00] <fuzzie> hum, the PS:T rights situation really is quite bad
[14:46:06] <fuzzie> who owns the rights? no-one knows!
[14:50:41] <Gekz> lol
[15:00:17] <pupnik> bah
[15:02:05] <pupnik> wonder if wotc would like to sellà these games for fones
[15:02:11] <pupnik> bioware
[15:02:21] <fuzzie> they don't even want to sell them for PCs.
[15:26:47] <-- pupnik has left IRC (Connection timed out)
[15:51:45] <-- Avenger has left IRC ("bye!")
[16:25:02] <-- Gekz has left IRC (Read error: 131 (Connection reset by peer))
[17:10:33] --> barra_away has joined #gemrb
[17:15:22] <fuzzie> sigh, some of our code is so hacky
[17:22:00] --> lynxlynxlynx has joined #gemrb
[17:22:00] --- ChanServ gives channel operator status to lynxlynxlynx
[17:22:59] <fuzzie> lynxlynxlynx: i backed our your pdolls change from pst because it doesn't work because if i add a call to the global table loading function then it tries loading an on-existant 2da and dies ('kitlist')
[17:30:15] <-- barra_library has left IRC (Read error: 110 (Connection timed out))
[17:41:29] <fuzzie> and Avenger's 'FloatRebus' function does something .. odd
[17:42:26] <fuzzie> it does what would seem to be the right thing, but i don't see it in the original game
[17:45:33] <CIA-22> gemrb: 03fuzzie * r7358 10/gemrb/trunk/gemrb/plugins/Core/GameControl.cpp: add missing ClearPath call
[17:56:21] <CIA-22> gemrb: 03fuzzie * r7359 10/gemrb/trunk/gemrb/ (3 files in 3 dirs): add GF_ONSCREEN_TEXT for pst
[17:59:00] <lynxlynxlynx> fuzzie: i'll make it check the existance
[17:59:11] <fuzzie> ok, i just wasn't sure what to do about it
[17:59:47] <fuzzie> i wonder how many of our strrefs are good for pst
[18:01:25] <fuzzie> all the ones i need, i guess :) nice
[18:17:35] <lynxlynxlynx> was kitlist the only missing one?
[18:18:08] <fuzzie> i don't know, i just stopped at first error
[18:18:46] <fuzzie> let's try them in console, i guess
[18:19:04] <lynxlynxlynx> xplevel i am not sure about
[18:19:12] <lynxlynxlynx> the previous two are in the override
[18:19:33] <fuzzie> it has clskills, races, xplevel, pdolls
[18:19:40] <lynxlynxlynx> but i'm pretty confident ok :)
[18:20:04] <lynxlynxlynx> and classes
[18:20:33] <fuzzie> it's basically just bg1
[18:20:41] <fuzzie> is the easy way to think about it, i guess
[18:20:51] <lynxlynxlynx> without kits
[18:21:02] <fuzzie> well, the only kitlist for bg1 is the one we supply
[18:21:14] <fuzzie> that code is probably still in pst's engine, just useless
[18:26:29] <fuzzie> ugh
[18:26:36] <fuzzie> quicksave feedback is done by the savegameiterator?
[18:36:19] <lynxlynxlynx> yes
[18:36:24] <lynxlynxlynx> QSAVESUCCEED
[18:37:49] <fuzzie> pst is a bit frustrating with no feedback, so i just add pause/quicksave feedback for now
[19:04:35] <fuzzie> ugh
[19:04:47] <fuzzie> why does some SDLVideo stuff mess with the input coords and some not?
[19:04:55] <fuzzie> that is not nice
[19:08:48] <fuzzie> ah, this is 'anchor'?
[19:11:06] <CIA-22> gemrb: 03fuzzie * r7360 10/gemrb/trunk/gemrb/plugins/Core/ (GameControl.cpp GameControl.h SaveGameIterator.cpp): add pst global DisplayText to GameControl, use it when pausing/saving
[19:11:11] <fuzzie> this fixes one of Edheldil's bugs, anyway :)
[19:12:08] <Edheldil> thanks :)
[19:22:50] --> Maighstir has joined #gemrb
[19:28:45] <Edheldil> Hi, Maighstir
[19:33:45] <Maighstir> Greetings
[19:34:51] <Maighstir> Isn't Lionheart based on the Infinity Engine as well? Looks like it
[19:35:12] <fuzzie> i'm pretty sure it isn't
[19:35:40] <fuzzie> people were looking at trying to use the graphics from Lionheart in IE, and there was some problem where the buildings were seperate from the maps, even?
[19:36:46] <Maighstir> Ah, I was thinking it used yet another heavily modified version
[19:36:57] <Edheldil> unfortunately not
[19:37:05] <Edheldil> the file formats are different
[19:37:49] <fuzzie> http://www.shsforums.net/index.php?showtopic=32402 discusses it from an IE point of view, and notes it uses the 'Velocity Engine'
[19:37:59] <Edheldil> but I haven't looked into what they are using
[19:41:47] <fuzzie> ah, that's the thread with jcompton's polygonning stuff
[19:56:02] <Maighstir> Oh well, thanks for the insight
[20:00:16] <Edheldil> mdl16? Sounds like a MDL format to me, but maybe it's just a coincidence stemming from 'model'
[20:00:47] <fuzzie> i think later in the thread they find it's a standard rad games tools thing
[20:08:08] <fuzzie> hm, junk doorway does nothing in gemrb
[20:09:15] <fuzzie> oh boy, the infopoint talks to you
[20:10:35] <Maighstir> can you have a conversation with it?
[20:10:47] <fuzzie> it is refusing to cooperate :(
[20:15:14] --> Avenger has joined #gemrb
[20:15:18] --- ChanServ gives channel operator status to Avenger
[20:16:55] <fuzzie> oh great, it's that stupid GoNear call on the target
[20:16:58] <Avenger> talking infopoints? :)
[20:17:19] <Avenger> well, at least you are motivated to fix it
[20:17:23] <fuzzie> yes :)
[20:17:25] <fuzzie> hello :)
[20:24:41] <fuzzie> aha, i broke this
[20:24:56] <fuzzie> well, sort of
[20:25:05] <fuzzie> gemrb confuses 'no interrupt' and 'dialog no interrupt' states
[20:25:21] <fuzzie> and now i set 'no interrupt' on infopoints by default, which is correct
[20:26:48] <fuzzie> maybe the real problem is that we swap source and target..
[20:34:46] <Avenger> hehe, that dialog swapping is a touchy point
[20:42:30] <fuzzie> well, i fixed it, and got through
[20:42:38] <-- lynxlynxlynx has left IRC (Remote closed the connection)
[20:43:41] <fuzzie> the pst death animations are so broken
[20:43:43] <fuzzie> i don't know why
[20:48:28] <Avenger> i just know, they were better at one point :)
[20:49:55] <Avenger> i work on avatar animation ids, it will be possible to have an interval of ids, so when a .cre in HoW says 0xf7b1 our engine won't be confused. The original engine ignores the lowest nibble most of the times.
[20:51:30] <Avenger> well, that was a bad example, but 0xf77a is a good one :) i don't really know what the lowest nibble means. As far as i see 0xf770-0xff7f are all the same.
[20:53:05] <fuzzie> :)
[20:56:44] <fuzzie> damn, segfault
[20:57:42] <fuzzie> can't speedrun the game if it keeps segfaulting!
[20:58:54] <CIA-22> gemrb: 03fuzzie * r7361 10/gemrb/trunk/gemrb/plugins/Core/ (GSUtils.cpp GameControl.cpp): mess with InitDialog IF_NOINT check
[20:58:56] <Avenger> hehe
[20:59:30] <Avenger> squish the bugs
[21:04:55] <CIA-22> gemrb: 03fuzzie * r7362 10/gemrb/trunk/gemrb/plugins/Core/IniSpawn.cpp: fix for copy-and-paste error causing crash
[21:14:43] <fuzzie> Avenger: where are all the coordinates for infopoints in DLTCEP?
[21:14:58] <fuzzie> gemrb reads LaunchX/Y, PosX/Y and TalkX/Y
[21:15:05] <fuzzie> i only see two pairs in DLTCEP, and i have no idea which they are
[21:15:16] <Avenger> let me see
[21:16:13] <fuzzie> (gemrb reads them in that order - LaunchX/Y, PosX/Y then TalkX/Y)
[21:17:11] <fuzzie> i guess 'launch point' is LaunchX/Y, 'use point' is PosX/Y, but i wish for TalkX/Y :)
[21:17:39] <Avenger> the launch point is the first one, yes
[21:18:04] <Avenger> the talk point is called 'pstpoint' in the source
[21:18:40] <Avenger> ok, if it is pst then the 'operating point' is editing the talk point
[21:18:53] <Avenger> if it is not pst, then it is editing the second group
[21:19:05] <Avenger> so i figured, pst doesn't use the second pair
[21:19:07] <fuzzie> hm, that would be why this doesn't work, i guess
[21:19:38] <fuzzie> thanks
[21:19:41] <Avenger> you can take a peek at the values even in pst, if you click the area type to bg2 in the first page in ther area editor
[21:19:57] <Avenger> oh wait
[21:19:59] <Avenger> no
[21:20:02] <Avenger> damn
[21:20:02] <fuzzie> it won't let me :-)
[21:20:17] <Avenger> well, you cannot, it just knows you are in pst
[21:20:18] <fuzzie> but it's okay i think
[21:20:33] <fuzzie> the problem is: i was wondering why gemrb had a useless UsePoint
[21:20:38] <Avenger> well, this is a quick hack to let me see the talk coordinates
[21:20:39] <fuzzie> when in DLTCEP, it showed up fine
[21:20:56] <fuzzie> i guess the answer is, i must check TalkPos instead :)
[21:21:01] <Avenger> yes
[21:21:09] <Avenger> it looks like that's what i do too :)
[21:22:19] <Avenger> i'm not completely sure those fields (the second pair of coordinates) are unused in pst. It could even be something entirely different
[21:22:52] <fuzzie> well, i just added another hack to gemrb for this, just so i can play, and i put it on the todo list to check later
[21:23:18] <CIA-22> gemrb: 03fuzzie * r7363 10/gemrb/trunk/gemrb/plugins/Core/ActorBlock.cpp: add more special checks for travel points
[21:23:20] <fuzzie> i should work out why the pst death anims broke, meh
[21:24:11] <fuzzie> the Dead Nations intro works quite well
[21:24:30] <fuzzie> although the fade is far too slow, they always are in pst
[21:28:18] --> pupnik has joined #GemRb
[21:28:23] <fuzzie> oh, right, we have pst's FadeToBlack action calling gemrb's FadeToColor action
[21:29:42] <fuzzie> and this means that it gets a default time of 64 ticks, which is too long, i guess
[21:40:00] <fuzzie> Avenger: you have Taimon's notes and can make sense of the spawn notes?
[21:40:15] <Avenger> not yet
[21:40:26] <Avenger> i started to clean up our code a bit
[21:40:29] <fuzzie> his latest reply says "Check my notes", which is not so helpful for the rest of us :)
[21:40:32] <Avenger> i just didn't commit
[21:40:47] <fuzzie> well, spawns are a big game-breaker for bg1, you die fast
[21:41:03] <fuzzie> so i just want to know if you stay on top of it, or whether i should look
[21:41:12] <fuzzie> but if you're in the middle of cleaning up, i'll leave alone
[21:41:24] <Avenger> i don't know what he meant, because his notes are ingame structures
[21:41:44] <Avenger> not saved game structures
[21:42:01] <Avenger> but the two are related, so if i look hard, i could probably find it
[21:42:17] <fuzzie> i seem to remember that if you save/load in bg1, all the spawns trigger again
[21:42:25] <fuzzie> someone mentioned that earlier in the thread
[21:42:54] <Avenger> the last tob patch which no one uses did something about spawnpoints too
[21:43:27] <Avenger> maybe it tried to save these fields :)
[21:43:57] <Avenger> and earlier versions (basically all) didn't save this information, maybe they simply didn't save spawned creatures
[21:44:40] <fuzzie> i think they did, you simply ended up with more
[21:45:44] <Avenger> hmm, so the spawns didn't regenerate if you saved, but they crowded on you when you saved/reloaded?
[21:46:07] <Avenger> i mean, didn't regenerate if you rested and reentered
[21:46:33] <fuzzie> i don't remember about re-enterable areas
[21:46:49] <Avenger> well, if you leave the area it saves automatically
[21:47:32] <Avenger> i will check this later, now i have to sleep
[21:48:05] <fuzzie> night!
[21:48:10] <Avenger> i think i mapped all the animation prefixes in HoW
[21:48:24] <-- Avenger has left IRC ("ChatZilla 0.9.85 [Firefox 3.5.3/20090824101458]")
[21:50:21] <-- tombhadAC has left IRC (Read error: 101 (Network is unreachable))
[21:52:36] --> tombhadAC has joined #gemrb
[21:56:34] <pupnik> happy monday
[21:56:50] <fuzzie> :(
[22:18:46] <pupnik> it makes me very happy to watch your work
[22:19:06] <pupnik> it is like "this is an island of goodness"
[22:43:22] <CIA-22> gemrb: 03fuzzie * r7364 10/gemrb/trunk/gemrb/GUIScripts/pst/GUISAVE.py: hack around stupid focus issue
[22:47:00] <fuzzie> hm, i have Morte in TNO's tomb :)
[22:47:40] <pupnik> planescape was a bit depressing
[22:48:12] <pupnik> glad they didnt realize their early design ideas. a random city would have been tough
[22:48:21] <fuzzie> oh dear, the ar1700->ar1800 transition has a flag set marked only as '8' in DLTCEP
[22:55:14] <fuzzie> oh i see, the door is trying to initiate dialog with me
[22:55:50] <fuzzie> i can see already that pst may well over-use this feature
[22:57:55] <pupnik> :)
[22:58:23] <pupnik> soon you will be able to recogniye what employees worked on what games
[23:00:21] <fuzzie> :o, ielister has failed me
[23:01:21] <fuzzie> it says "Name strref 534e4152"
[23:01:38] <fuzzie> but that is 'RANS'!
[23:01:40] <fuzzie> it is text!
[23:03:24] <fuzzie> in fact it is DPTTRANS! a dialog! and yet it claimed there was no dialog! the lies, they hurt :(
[23:09:35] <pupnik> hahaha
[23:13:45] <fuzzie> really, i feel betrayed
[23:14:32] <fuzzie> mostly because all my gemrb savegames are probably corrupt, again
[23:24:13] <pupnik> :Ã/
[23:33:05] <-- Edheldil has left IRC (Remote closed the connection)
[23:36:35] <fuzzie> sigh, back to this tomorro
[23:36:37] <fuzzie> ninight
[23:38:24] <pupnik> cheers fuzzie