#gemrb@irc.freenode.net logs for 26 Dec 2013 (GMT)

Archive Today Yesterday Tomorrow
GemRB homepage

[01:40:07] <-- edheldil has left IRC (Ping timeout: 246 seconds)
[02:25:46] <-- jackhunter has left IRC (Quit: Page closed)
[05:02:52] --> brada has joined #gemrb
[05:05:14] --> Eli2 has joined #gemrb
[06:11:00] <-- brada has left IRC (Quit: brada)
[06:26:22] --> brada has joined #gemrb
[06:54:57] <-- brada has left IRC (Quit: brada)
[08:34:06] --> lynxlynxlynx has joined #gemrb
[08:34:06] --- ChanServ gives channel operator status to lynxlynxlynx
[10:08:33] --> edheldil has joined #gemrb
[10:08:33] --- ChanServ gives channel operator status to edheldil
[10:08:47] --> jackhunter has joined #gemrb
[10:08:52] <jackhunter> hello
[10:08:58] <jackhunter> i have just completed planescape;)
[10:09:34] --> Eli2_ has joined #gemrb
[10:10:45] <jackhunter> wtf
[10:10:52] <jackhunter> westwood studio closed long ago
[10:11:06] <jackhunter> and yet ea let in the game the "westwood chat application"
[10:12:25] <-- Eli2 has left IRC (Ping timeout: 245 seconds)
[11:30:20] <edheldil> jackhunter: congrats :). Which ending?
[11:30:37] <edheldil> and ... in GemRB? ;-)
[12:25:44] <jackhunter> not in gemrb, fusion ending;)
[12:27:01] <jackhunter> i wait gemrb to be more advanced cause i tend to make very long palythroughs
[12:27:14] <jackhunter> tgenerallyi do all secondary quests when i do a run
[13:50:23] --> brada has joined #gemrb
[14:06:14] <jackhunter> hue
[14:06:25] <jackhunter> Europe: 0.01m sales for planescape torment in 2013
[14:06:34] <jackhunter> it doesnt include digital sales but is very low
[14:09:50] <brada> if it doesnt include digital sales it is meaningless
[14:11:33] <fuzzie> yes, >99% of sales would be digital
[14:18:31] <lynxlynxlynx> m as in mega or million?
[14:26:09] <jackhunter> as thousand
[14:26:16] <jackhunter> million
[14:26:24] <jackhunter> it makes 10 thousand sales
[14:28:33] <jackhunter> but well they dont have a lot of numbers for thos eold games
[14:28:43] <jackhunter> they dont even know how many copies of baldur games were sold
[14:30:26] <edheldil> do you think it's little? 10k physical copies of a 14 years old game?
[14:31:20] <jackhunter> it is since its since it was released
[14:31:44] <edheldil> ah, I thought this year only
[14:31:58] <jackhunter> its suposed to be 10 k in 14 years which seems odd
[14:31:58] <edheldil> where are the numbers from?
[14:32:03] <jackhunter> vg chartz
[14:32:15] <edheldil> vg?
[14:32:37] <jackhunter> http://www.vgchartz.com/game/14253/planescape-torment/
[14:32:40] <Pepelka> Planescape: Torment (Microsoft Windows) - Sales, Wiki, Cheats, Walkthrough, Release Date, Gameplay, ROM on VGChartz
[14:32:42] <Pepelka> »Planescape: Torment is a video game developed by Black Isle Studios and released on Microsoft Windows. View game sales, statistics, release dates, characters, credits, discussion and related links here.«
[14:33:26] <jackhunter> oh well
[14:33:36] <jackhunter> actually they only put the numbers of three latest years
[14:33:47] <jackhunter> which makes around 3,5 k per year
[15:24:00] <brada> wow, gemrb performs terribly even on newish i7 :(
[15:24:52] <lynxlynxlynx> what's that?
[15:26:24] <brada> i7? intel i7 CPU
[15:26:45] <wjp> and terribly is?
[15:27:47] <jackhunter> i cantt ell you i still have an old duocore
[15:30:04] <lynxlynxlynx> works fine on my i3
[15:30:44] <brada> < 20 fps
[15:32:01] <Lightkey> turn off your Bitcoin miner
[15:33:07] <lynxlynxlynx> still above 30 here and a flash video is playing besides a bunch of other hoggers
[15:34:03] <wjp> in a particularly complex scene? Are you running at normal resolution? At the right bitdepth?
[15:34:15] <wjp> which OS? SDL 1.2? SDL 2?
[15:34:55] <brada> os x SDL 1.2
[15:35:02] <brada> it runs great in fullscreen tho
[15:35:14] <brada> running in a window is sloooowww
[15:35:27] <brada> 1024x768
[15:35:41] <brada> and on main menu
[15:37:16] <wjp> At the right bitdepth?
[15:38:08] <lynxlynxlynx> do you use drivers for your gfx? :D
[15:38:46] <brada> 32bits as always
[15:40:48] <brada> ill assume its an SDL 1.2 thing on newer mac os
[15:50:48] <brada> indeed i get 30 fps with SDL2
[15:50:58] <brada> in a window
[15:51:57] <brada> i should probably "upgrade" the mac build to SDL2...
[15:56:13] <brada> except i cant cuz the mouse doesnt work well :
[17:07:08] <-- Eli2_ has left IRC (Quit: Ex-Chat)
[17:12:59] --> raevol has joined #gemrb
[17:59:13] <brada> I wonder if i broke audio in IWD2 movies
[17:59:41] <brada> I get spammed with "Unable to buffer data" errorss from openal
[18:01:58] <brada> then again its only for some videos
[18:02:24] <brada> but i dont see the difference. they are both 2 channel 16bit 22050 samplerate
[18:02:52] <brada> maybe the problem is that info is wrong for the broken one
[18:06:07] <fuzzie> wouldn't it just be corrupt then?
[18:06:42] <brada> could be
[18:07:13] <brada> i wonder why i have to invert our "U" and "V" planes for video
[18:07:20] <brada> SDL_UpdateYUVTexture works if i do that
[18:08:44] <fuzzie> ..?
[18:09:23] <brada> i need to pass what I thought was the U plane to the v parameter and same with V
[18:09:54] <brada> I also tried other varients of YUV texture format
[18:10:05] <brada> in case that was wrong
[18:10:16] <brada> but they just made black :p
[18:10:22] <fuzzie> the 1.2 backend is using YVU order
[18:10:50] <brada> oh
[18:10:57] <brada> makes sense then
[18:11:37] <brada> its nice when one line of code replaces an entire complex block :)
[18:11:38] <fuzzie> the different formats are confusing
[18:12:08] <fuzzie> the SDL 1.2 code is annoyingly complex
[18:12:15] <brada> I cant see an appreciable performance difference, however
[18:12:30] <brada> they both stutter like crazy in the ios simulator :(
[18:12:32] <fuzzie> this whole showYUVFrame function is just munging it to SDL 1.2's expected input I guess?
[18:12:44] <brada> I wouldnt know
[18:12:56] <brada> when i wrote the SDL2 version i scraped trying to make sense of it
[18:12:58] <fuzzie> you're using a 32bpp surface and blitting to the screen?
[18:12:59] <brada> and wrote my own
[18:13:09] <fuzzie> I'm pretty sure it's munging to SDL 1.2 format
[18:13:10] <brada> not a surface
[18:13:17] <fuzzie> well, a 32bpp yuv texture :-p
[18:13:17] <brada> texture
[18:13:20] <brada> yes
[18:13:26] <fuzzie> well not 32bpp!
[18:15:08] <fuzzie> what is it, a YV12 texture?
[18:15:52] <brada> I no longer remember exactly, but its what i determined the data to be when I was working on it originally :p
[18:15:54] <fuzzie> hm
[18:16:03] <fuzzie> isn't it always going to be unaccelerated?
[18:16:20] <brada> I guess, but how could I accelerate it?
[18:16:22] <fuzzie> since the SDL2 people removed all the non-32bpp modes..
[18:18:51] <fuzzie> yes, the create texture path surely always calls SDL_SW_CreateYUVTexture
[18:18:52] <brada> hmmm
[18:18:54] <brada> guess what
[18:19:06] <brada> it still stutters like mad if i simply do a return at the top...
[18:19:11] <brada> the audio i mean obviously :p
[18:19:21] <fuzzie> oh
[18:19:25] <fuzzie> the audio stutters
[18:19:28] <brada> yes
[18:19:30] <fuzzie> but does play?
[18:19:44] <brada> the audio?
[18:19:44] <fuzzie> and, well, it's bink, all the work will be done by the decoding
[18:19:50] <fuzzie> or, well, not all, but a large amount of it
[18:20:01] <brada> i can tell its the actual audio
[18:20:10] <brada> not just random noise if thats what you mean
[18:20:27] <fuzzie> you said you get spammed with errors, I figured it didn't work at all :)
[18:20:37] <brada> only on some of the videos
[18:20:44] <brada> im trying with one that doesnt
[18:21:01] <fuzzie> so, yes
[18:21:17] <fuzzie> YUV is software-path-only for opengl es 1/2, due to them having removed all non-32bpp texture support
[18:21:30] <brada> Its just as if the timing were off :/
[18:21:50] <fuzzie> so I wouldn't expect performance to be great.
[18:22:19] <brada> but the CPU is only at 20% during playback
[18:22:29] <fuzzie> you're not blocking somewhere?
[18:22:53] <brada> let me try on mac
[18:25:10] <brada> he he
[18:25:31] <brada> plays back smooth on mac, but with surrounding artifacts from other windows
[18:25:38] <brada> not clearing something
[18:26:46] <brada> and i still get those openal error, but the audio plays fine
[18:27:43] <brada> I definately still have a small bug in the input code
[18:27:53] <brada> sometimes mouse up doesnt work
[18:28:06] <brada> presumably because this was designed for touch screen
[18:28:48] <brada> ha ha nevermind
[18:28:58] <brada> apparently some things just dont work in IWD2 :p
[18:31:35] <brada> anyway i do wonder where the bottle neck is on ios
[18:31:55] <brada> unless they made the simulator throttle itself to actual device capabilities... somehow
[19:02:40] <jackhunter> so once you played all black isle rpg, what can you still hope?
[19:10:55] <edheldil> you can hope that PoE is as good as it looks
[19:37:18] <Pepelka> [wiki] pst_bugs - more torment infopoint wonderfulness http://www.gemrb.org/wiki/doku.php?id=pst_bugs&rev=1388086450&do=diff
[20:42:59] <brada> grr
[20:43:16] <brada> i guess SDL thinks its a good idea for the apple trackpad to generate touch events
[20:43:27] <brada> no wonder input sometimes gets wonkey on me!
[20:43:51] <Pepelka> [wiki] pst_bugs - [Gameplay] more trap infopoint testing http://www.gemrb.org/wiki/doku.php?id=pst_bugs&rev=1388090485&do=diff
[20:46:41] <brada> I dont quite understand why the finger up/down occurs when you click, however. i would think is wrong
[20:53:32] <brada> why do we use SDL_EnableUNICODE?
[20:53:41] <brada> that is why my keyboard is half broken
[20:54:33] <brada> or the if (key < 32 || key == 127) { check
[20:54:54] <brada> because the keysym.sym value is correct
[20:55:06] <brada> but the key is higher than 127
[20:55:14] <brada> so it doesnt get caught
[20:55:36] <fuzzie> without SDL_EnableUNICODE, you get no .unicode
[20:56:03] <fuzzie> obviously that's SDL1.2-specific though, and someone has wrapped it in an #if for SDL version
[20:56:30] <brada> fuzzie: i am talking about 1.2
[20:56:44] <brada> I cant really use sdl2 on mac due to no mouse cursors
[20:56:49] <fuzzie> so what's the problem with the keys?
[20:56:52] <brada> i can test with it thats it
[20:56:58] <brada> half of them dont work :/
[20:57:09] <brada> because the "key" value is too high
[20:57:37] <brada> and removing the enableUNICODE call "fixes" it
[20:57:45] <fuzzie> right, but that breaks unicode :-p
[20:57:48] <brada> right
[20:57:49] <fuzzie> what keys are causing the problem?
[20:57:53] <brada> thats why i am asking
[20:57:59] <brada> what should actually be done
[20:58:20] <fuzzie> that's some pretty weird code
[20:58:46] <fuzzie> why isn't it just always switching on the keysym.sym?
[20:59:03] <brada> thats a more direct way of asking what i am asking :)
[20:59:09] <fuzzie> yeah so it should do that
[20:59:13] <fuzzie> then the 'default' case should grab the .unicode.
[20:59:20] <brada> ok
[20:59:32] <brada> so no need for the silly range check at all?
[21:00:02] <fuzzie> well as long as you fix it!
[21:00:31] <brada> right
[21:00:44] <brada> I *think* i know what that means :p
[21:00:49] <brada> ill post a patch
[21:02:12] <fuzzie> kind of weird if you get a too-high key though? what's the values?
[21:04:26] <brada> I was looking at the wrong key variable, they are actually 0 which is too low for that test
[21:04:35] <fuzzie> 0 is good, right?
[21:04:53] <brada> and another oddity is that they are arrow keys, but the "scancode" is completely wrong
[21:04:54] <fuzzie> wait, maybe I'm misunderstanding
[21:04:58] <brada> the sym is correct
[21:05:04] <fuzzie> I mean, for arrow keys, key should be 0
[21:05:09] <fuzzie> so 'key < 32' succeeds
[21:05:18] <brada> oh
[21:05:20] <brada> grr
[21:05:23] <fuzzie> so you hit the switch table which checks sym and gives you SDLK_DOWN/UP/etc
[21:05:33] <brada> well thats what i get for looking at the wrong key variable
[21:05:56] <fuzzie> there's a corner case here where there's not actually any guarantee that you get a key==0 though, especially not in SDL2 land
[21:06:32] <brada> yeah thats semi uncharted due to me developing on ipad
[21:06:39] <brada> where there isnt really a keyboard
[21:06:55] <brada> so i never tested things like arrow keys with it obviously
[21:08:39] <brada> fuzzie: I must apologize for being not very bright right now
[21:09:19] <fuzzie> no, this is good
[21:09:25] <brada> fuzzie: I forgot to reenable unicode >_<
[21:09:35] <fuzzie> sometimes I spend half an hour rambling about stupid things before someone points out the obvious
[21:09:36] <brada> so yes it doesnt work and the key value is super high
[21:09:47] <fuzzie> so it's nice when other people also make mistakes occasionally.. :)
[21:09:54] <fuzzie> what's the key value?
[21:09:57] <brada> down key is giving me 63233
[21:10:47] <fuzzie> oh look, the PUA
[21:11:10] <fuzzie> did you try glaring at apple engineers until they stop doing weird stupid stuff?
[21:11:20] <brada> it didnt help :(
[21:11:25] <fuzzie> that's a NSDownArrowFunctionKey apparently.
[21:11:29] <fuzzie> bless.
[21:11:50] <brada> i wonder if its due to this keyboard driver mod i have installed
[21:12:09] <fuzzie> well, it is a legitimate thing to have in the unicode field
[21:12:40] <fuzzie> and a good example of this 'corner case' I was thinking of
[21:12:52] <fuzzie> so if you're motivated to go ahead and fix it, please go ahead and fix it
[21:13:05] <brada> so we should remove that check and then have the default case do the else if bit right?
[21:13:15] <fuzzie> yes
[21:13:19] <fuzzie> (maybe)
[21:18:17] <brada> ok that seems to work (for what i tested)
[21:21:28] <brada> odd that we handle console input in the video driver :p
[21:21:53] <brada> guess we would need to associate it with a window to do it sanely
[21:25:24] <brada> problem is we need a break 2...
[21:25:37] <brada> er is that legal in c++?
[21:26:00] <brada> no
[21:26:44] <brada> well it looks like its safe to simply return here
[21:27:12] <brada> not really a fan of returning from a switch in a switch
[21:29:05] <brada> fuzzie: https://github.com/bradallred/gemrb/compare/gemrb:master...master
[21:29:08] <Pepelka> Comparing gemrb:master...bradallred:master · bradallred/gemrb · GitHub
[21:29:09] <Pepelka> »gemrb - Game Engine Made with preRendered Background«
[21:35:44] <fuzzie> yes, well, it's icky
[21:35:49] <fuzzie> but yes that looks like the right idea
[21:40:45] <brada> kinda wondering why we dont just have one keypress handler
[21:49:35] <edheldil> brada: in sdl, it's hard to split it
[21:50:38] <edheldil> you have all those SDL_K_* constants
[22:13:33] --> chiv has joined #gemrb
[22:14:02] <chiv> hi all
[22:15:03] <brada> hey
[22:15:12] <brada> did you figure out the trigger actions thinggy?
[22:17:03] <chiv> well, all the clickable ones are working
[22:17:23] <chiv> portals are a bit weirder
[22:18:26] <chiv> each trigger has two points, a launch point and a talk point - but for portal activation traps, the original game totally ignores them and makes it's own point in the polygon centre...
[22:19:48] <chiv> so where a portal has a dialog attached to it, gemrb dutifully tries to send the player to a location on the map that he can't ever reach
[22:20:46] <brada> i saw you got it working, but i meant the more optimal way fuzzie mentioned
[22:21:21] <chiv> oh no, I've been christmas-ing
[22:21:29] <brada> well merry christmas
[22:21:48] <brada> there is nothing wrong with trying to send a player to a point he cant reach tho
[22:21:56] <brada> doesnt it jsut get as close as possible?
[22:22:11] <brada> the problem i suppose it the reachable point is out of range?
[22:22:13] <chiv> no, because the designers put the points in totally illogical places
[22:22:16] <brada> or is it the wrong point?
[22:22:31] <brada> maybe we are assuming the point incorrectly
[22:22:55] <brada> such as the center of a region instead of its edge?
[22:23:02] <chiv> I've tried changing all the points to various coords- the only thing that made any difference was moving the whole polygon
[22:23:23] <chiv> and yeah, it seems to calculate the centre
[22:23:43] <brada> mind you i have no idea how any of this works
[22:24:47] --> kpederse1 has joined #gemrb
[22:25:35] <lynxlynxlynx> several choices in the structure itself
[22:33:58] <brada> chiv: reading the pst bug page I saw that one for actors dropping their gear on death and I agree it would be an enhancement to only drop gear when giblitized.... except what do you do when a PC dies and you remove them from the party?
[22:35:01] <brada> also did my recent changes break the floating menu even more?
[22:38:10] <chiv> i haven't recompiled recently - but the problem with the float menu is that gemrb over-rides the action on mouse up, so unless you changed that
[22:39:28] <chiv> as far as not dropping items, I always thought it would be better to turn the corpse itself into a container - but that's just a mod idea
[22:39:40] <brada> no i changed a bunch of drawing code tho :p
[22:40:00] <chiv> is it in gemrb/master?
[22:40:04] <brada> yes
[22:40:35] <chiv> ok ill see what happens
[22:44:09] <kpederse1> hi all. When I attempt a compile on OpenBSD, I get undefined references to stuff in libc (free, memmove, strncpy).
[22:44:20] <chiv> regarding the portal trigger - whatever I change in torment.exe, TNO always turns to face a point that is the centre of the 'polygon bounding box' fields when the dialog loads
[22:44:54] <kpederse1> I notice in the CMakeLists.txt you link in libgcc-shared on NetBSD, is this related? (I tried the same in OpenBSD but it didnt make a difference)
[22:46:24] <fuzzie> no, libgcc is some different netbsd-only issue
[22:47:02] <fuzzie> chiv: we miscalculate the talk pt?
[22:47:19] <chiv> torment.exe totally ignores the talk point
[22:47:33] <fuzzie> well, let me get some terminology right
[22:47:41] <fuzzie> the actual talk point should be calculated at load time from the polygon
[22:47:49] <lynxlynxlynx> kpederse1: nobody compiled it for openbsd in a while
[22:48:09] <fuzzie> do we talk about that?
[22:48:12] <edheldil> has anybody at all?
[22:48:41] <chiv> hmm.. the .are files seem to already have coordinates inside them according to dltcep
[22:48:43] <fuzzie> gemrb's TalkX/TalkY is something else related
[22:48:59] <kpederse1> ah right. I am pretty stumped as to the issue then. I imagine it will be a build issue rather than code, so I will keep on fiddling and see if I can get anywhere
[22:50:17] <fuzzie> chiv: dltcep doesn't say "operating point"?
[22:51:02] <chiv> actually yes, but that is the one that tno tries to walk to
[22:51:13] <fuzzie> and that's also wrong?
[22:51:29] <fuzzie> last time I think we concluded that TalkPos seemed to be where to walk *to*
[22:51:31] <fuzzie> for pst only
[22:51:56] <chiv> well if you look at ar0101, from_portal trigger has it somewhere the player can't walk to
[22:52:10] <fuzzie> what does original do?
[22:52:15] <fuzzie> it doesn't try walking there at all?
[22:52:15] <lynxlynxlynx> edheldil: maybe you're right; our commit history only has mention of bugfixes for free and net variants
[22:52:38] <chiv> fuzzie: it seems to calculate a point in the centre of the bounding box
[22:52:44] <fuzzie> chiv: so I'm confused :P
[22:52:50] <fuzzie> you said "turns to face a point"
[22:52:54] <chiv> you and me both
[22:53:00] <fuzzie> but now you talk about walking :)
[22:53:09] <chiv> well, he walks if he can
[22:53:10] <jackhunter> oups
[22:53:13] <lynxlynxlynx> kpederse1: likely just a missing header or linker flag
[22:53:27] <chiv> jackhunter: yes, spells are mostly broken
[22:53:39] <fuzzie> oh, I thought the pst spells were mostly dealt with
[22:53:55] <jackhunter> planescape torment spells caused many crashes in original version, with gemrb hopefully its gone
[22:53:58] <fuzzie> chiv: anyway I don't have pst here so I can't look
[22:53:59] <lynxlynxlynx> not if you're a perfectionist
[22:54:17] <jackhunter> or will be gone
[22:54:39] <jackhunter> god place to test is when you go in maudith undeground
[22:54:51] <fuzzie> chiv: but dltcep is kind of unreliable here, it doesn't display all the points :/
[22:54:56] <jackhunter> i noticed a lot of lagg happens there in original version, and crashes
[22:55:00] <chiv> arg
[22:55:19] <chiv> well, there's something weird anyway, so I'll try and get to the bottom of it
[22:55:27] <jackhunter> another place whicnh can cause crashes is maudith being invaded by demons
[22:55:42] <jackhunter> cause enemy keep respawning like crazy
[22:55:42] <fuzzie> chiv: gemrb should set the region 'Pos' to the center of the bbox, right?
[22:55:45] <jackhunter> and it causes crashes
[22:56:17] <chiv> i don't remember spells crashing the game ever in torment.exe
[22:56:30] <jackhunter> it depends on places
[22:56:55] <jackhunter> i noticed there is a special creen in maudith undergound(the jail of the deva) where it can happen a lot
[22:57:06] <jackhunter> just when you enter that map
[22:57:12] <edheldil> fuzzie: anyhing that could be listed w/ iesh instead?
[22:57:21] <fuzzie> edheldil: doesn't it already?
[22:57:48] <jackhunter> in fact my expereince of pst is strange
[22:57:56] <fuzzie> iesdp has the 'alternative use point' and the 'talk use point' and the 'trap launch location' documented, I think that's all of them
[22:58:02] <jackhunter> half of the game goes ok, then half of the game has crashes
[22:58:38] <fuzzie> I think dltcep displays always the 'talk use point' and not the 'atlernative use point' for pst areas
[22:58:46] <fuzzie> irc log agrees sort of
[22:58:48] <jackhunter> i think its basically because at the beginning you dont have screens overloaded with monsters
[22:59:04] <fuzzie> but I really have no time to hunt this down atm :/ will have a bit more after jan 6th.
[22:59:10] <jackhunter> and less members too
[22:59:19] <fuzzie> by then chiv will probably have it all worked out from the behaviour of the original, which is ofc most reliable
[22:59:49] <chiv> lol?
[23:00:20] <edheldil> we believe in you :)
[23:00:33] <fuzzie> if you check it and see how it works in original then .. that's how it works :)
[23:00:51] <chiv> well, I have been prodding it for a couple of hours and that is my best guess
[23:01:08] <fuzzie> as I said, we do set the Pos to the bbox center in pst mode (if I read it right)
[23:01:22] <chiv> so I guess I'll go and implement that and see if I can play more of the game...
[23:01:24] <fuzzie> so probably that's related
[23:01:39] <fuzzie> but in any case you shouldn't need to walk to the actual point to do the dialog?
[23:01:43] <fuzzie> since sometimes they're unreachable
[23:02:19] <chiv> I am actually talking about the walkable trap triggers, and most of them don't have dialog so it's only rare cases there is a problem
[23:02:27] <fuzzie> and you seemed to have the right idea with trying a walk action first..
[23:02:39] <fuzzie> walkable trap triggerwhatnow
[23:03:14] <fuzzie> I mean the trap regions which just fire when you enter the polygon are a different story..
[23:03:17] <chiv> actually that seemed in the end to be superflous : the actual dialog() action provokes a player move
[23:03:48] <fuzzie> ah
[23:03:54] <fuzzie> right yes :)
[23:04:13] <chiv> so I whittled it down to one line and so far everything seems good
[23:04:18] <fuzzie> :-)
[23:04:38] <chiv> but I got distracted by this new problem :/
[23:04:42] <jackhunter> what about ice wind dale games? i guess they are less complicated maybe?
[23:04:48] <jackhunter> i mean for gemrb
[23:04:53] <chiv> icewind dale is already well supported
[23:04:56] <edheldil> will probably have horrible repecussions elsewhere :)
[23:05:24] <chiv> elsewhere don't have 'dlg' files attached to traps, as far as I am aware
[23:09:05] <lynxlynxlynx> pocket plane spirit summoner would be a place to check
[23:09:14] <lynxlynxlynx> does it spawn an invisible actor for the talking?
[23:09:41] <lynxlynxlynx> probably it's visible in dltcep if it has a dialog set
[23:09:51] <lynxlynxlynx> if so, you can easily search in there
[23:15:48] <kpederse1> Ok, I got gemrb compiling on OpenBSD. Was quite a small tweak to CMakeLists.txt in the end.
[23:15:51] <kpederse1> http://pastebin.com/D3D4a5Fu
[23:15:52] <Pepelka> --- CMakeLists.txt.orig Fri Dec 27 00:14:11 2013 +++ CMakeLists.txt Fri De - Pastebin.com
[23:21:15] <edheldil> I will commit it
[23:23:37] <kpederse1> cool. if I run into any other issues I can fix, I will send some more patches your way.
[23:24:09] <kpederse1> Unfortunately OpenBSD doesnt have a port of Wine so getting hold of the game data files is proving to be tricky ;)
[23:26:58] <edheldil> it's insecure :)
[23:30:24] <lynxlynxlynx> does unshield work?
[23:49:30] <kpederse1> yes. We also had unshield in the ports tree ;)
[23:50:24] <lynxlynxlynx> we have some info on the wiki, there's some post extraction janitoring to be done
[23:51:05] <kpederse1> nice, yeah I am following through this now.
[23:52:01] <-- jackhunter has left IRC (Quit: Page closed)
[23:57:40] <chiv> The code comments say that torment has no 'trap_usepoint' flag. However, Infopoint::GetUsePoint() only checks for that flag. so GetTalkPositionFromScriptable() is only ever going to return the trap launch point in pst
[23:59:00] <chiv> happilly though, I've traced the problem for the broken portal in ragpicker square