#gemrb@irc.freenode.net logs for 13 Apr 2011 (GMT)

Archive Today Yesterday Tomorrow
GemRB homepage

[01:06:40] --> pupnik_ has joined #gemrb
[01:09:29] <-- pupnik has left IRC (Ping timeout: 240 seconds)
[01:17:51] <-- Bo_Thomsen has left IRC (Quit: Leaving.)
[02:10:44] --- _pickle is now known as _hotdill69
[02:13:50] --- _hotdill69 is now known as pickle_
[02:23:28] <-- pickle_ has left IRC (Remote host closed the connection)
[03:21:19] --> raevol has joined #gemrb
[04:14:36] --> tomprince|hermes has joined #gemrb
[04:39:09] <-- raevol has left IRC (Quit: Leaving.)
[06:04:59] <-- tomprince has left IRC (Quit: leaving)
[06:05:37] --- tomprince|hermes is now known as tomprince
[07:01:19] --> lubos has joined #gemrb
[07:07:43] --> adominguez has joined #gemrb
[08:01:28] <fuzzie> morning
[08:03:12] <edheldil> Hi, fuzzie and all
[08:07:12] <-- devurandom has left IRC (Remote host closed the connection)
[08:07:31] --> devurandom has joined #gemrb
[08:09:20] --> Bo_Thomsen has joined #gemrb
[08:32:51] <dhewg> moroning
[08:53:23] <-- |Cable| has left IRC (Remote host closed the connection)
[09:10:03] --> lynxlynxlynx has joined #gemrb
[09:10:03] --- ChanServ gives channel operator status to lynxlynxlynx
[10:16:27] <-- lynxlynxlynx has left IRC (Ping timeout: 258 seconds)
[10:29:59] --> lynxlynxlynx has joined #gemrb
[10:29:59] <-- lynxlynxlynx has left IRC (Changing host)
[10:29:59] --> lynxlynxlynx has joined #gemrb
[10:29:59] --- ChanServ gives channel operator status to lynxlynxlynx
[10:59:49] <-- lynxlynxlynx has left IRC (Ping timeout: 276 seconds)
[11:23:31] --> lynxlynxlynx has joined #gemrb
[11:23:31] <-- lynxlynxlynx has left IRC (Changing host)
[11:23:31] --> lynxlynxlynx has joined #gemrb
[11:23:31] --- ChanServ gives channel operator status to lynxlynxlynx
[12:33:44] --- pupnik_ is now known as pupnik
[13:20:18] --> SiENcE has joined #gemrb
[15:23:00] <-- Bo_Thomsen has left IRC (Quit: Leaving.)
[15:25:40] <lynxlynxlynx> huh, fighter kit selection is broken
[15:25:49] <lynxlynxlynx> & nobody noticed :s
[15:26:48] <fuzzie> well, there's some rather more substantial stuff been broken with bg2 chargen and no-one noticed, i thought
[15:28:39] <-- SiENcE has left IRC (Quit: @all: cya)
[15:48:01] <-- ar has left #gemrb
[16:18:02] --> Maighstir has joined #gemrb
[16:41:31] --> ubermad has joined #gemrb
[16:45:48] --> Bo_Thomsen has joined #gemrb
[16:49:29] <ubermad> hi
[16:59:32] <-- adominguez has left IRC (Quit: Saliendo)
[16:59:41] <lynxlynxlynx> oj
[17:16:42] <lynxlynxlynx> something sinister is uo
[17:18:13] <fuzzie> oh?
[17:23:15] <lynxlynxlynx> don't worry about it yet
[17:23:52] <tomprince> That sound promising ... maybe. :)
[17:31:56] --> |Cable| has joined #gemrb
[17:41:01] <lynxlynxlynx> we seem to be misreading 2das :s
[17:41:20] <lynxlynxlynx> [ResourceManager]: Searching for K_B_H.2da...[chitin.key] <-- so it is the original
[17:42:02] --> Cable_ has joined #gemrb
[17:42:15] <lynxlynxlynx> last line (3) is 15, but we have only 1 stored on lookup time
[17:42:23] <-- |Cable| has left IRC (Read error: Connection reset by peer)
[17:42:47] <ubermad> I have a question, which uses a file to PST file or Torment.ini gemre.cfg?
[17:42:58] <fuzzie> lynxlynxlynx: encrypted?
[17:43:39] <lynxlynxlynx> any easy way to find out which biff it is from?
[17:43:50] <lynxlynxlynx> ubermad: what do you mean?
[17:47:15] <fuzzie> i added a quick fix for encrypted 2DAs which unbroke the obvious stuff, but didn't check beyond that
[17:47:19] <lynxlynxlynx> looks like it is from default.bif
[17:47:25] <lynxlynxlynx> grep matched ;)
[17:47:41] <lynxlynxlynx> oh wait, that's kittable
[17:47:52] <lynxlynxlynx> i am much to eager
[17:48:57] <lynxlynxlynx> hmm, dltcep says it is in the same biff
[17:49:10] <fuzzie> weidu is no good?
[17:50:00] <lynxlynxlynx> why bother with that
[17:50:23] <lynxlynxlynx> the file is in default.bif, i'll go check further
[17:50:50] <tomprince> I should get around to importing all the games in git.
[17:55:26] <lynxlynxlynx> default.bif is clearly uncompressed, so i doubt that bit is relevant
[17:56:25] <lynxlynxlynx> the only possibly odd thing i notice is that the tables are without the final newline, but that shouldn't matter either (doesn't to dltcep)
[17:56:42] <fuzzie> but encrypted or not? :P
[17:57:43] <ubermad> lynxlynxlynx: i don't know with file's used by gemrb to play in gemrb file is path to catch folder and in toremnt.ini to
[17:58:23] <lynxlynxlynx> fuzzie: grep matched, i can read it with less -> hardly zlibbed
[17:58:37] <fuzzie> lynxlynxlynx: encrypted != compressed
[17:58:39] <lynxlynxlynx> ubermad: set the cache folder in gemrb
[17:58:46] <fuzzie> but yes, if you can see it in the .bif file in plaintext, not encrypted.
[17:59:07] <lynxlynxlynx> yep
[18:00:29] <ubermad> can I put game files in catch folder and change patch cd to catch folder?
[18:02:33] <fuzzie> lynxlynxlynx: which file is it?
[18:02:44] <lynxlynxlynx> k_b_h.2da
[18:03:09] <lynxlynxlynx> ubermad: the cache folder will be wiped all the time, so don't put anything in there
[18:03:37] <fuzzie> lynxlynxlynx: and you only get one line, or you're missing the last?
[18:03:54] <lynxlynxlynx> i'm missing the last printable character
[18:04:07] <fuzzie> the '5'?
[18:04:27] <fuzzie> DataStream.cpp:202/203, move to the bottom of the 'while'
[18:05:48] <fuzzie> i was saying when i fixed the encrypted thing that we should maybe have a testcase of an encrypted 2DA, better add '2DA without newline at the end' to the list
[18:10:51] <tomprince> add reading them to OnLoad of gemrb/GUIScripts/test/Start.py
[18:11:12] <tomprince> Although I am not sure how to then fail the test. Maybe sys.exit(1) would work?
[18:12:01] <fuzzie> or just print a big error
[18:12:20] <lynxlynxlynx> i didn't check, but isn't there probably a \0 in between?
[18:12:54] <fuzzie> in between what?
[18:13:12] <tomprince> Well, but my buildbot wouldn't catch that...
[18:13:36] <tomprince> I guess I could teach it to, but it would be nice if the exit code matched in that case.
[18:13:39] <ubermad> lynxlynxlynx so i can't change this? On pc when I change cd patch to catch, game is loading and change location faster
[18:13:41] <fuzzie> well, for other projects I use some perl script which greps for the pass/fail stuff and gives a nice result thing and an exit code
[18:14:14] <fuzzie> but sys.exit(1) is ok, just haven't thought whether it's practical
[18:14:29] <lynxlynxlynx> ubermad: i'm not exactly sure what you mean, but just do a full install
[18:18:31] <lynxlynxlynx> nope, no delimiter at all
[18:18:32] <ubermad> lynxlynxlynx nvm one more question, where can I change maximum frame rate, catche size and path search nodes? In game file or gemrb files?
[18:19:13] <lynxlynxlynx> all in torment.ini, but it will have no effect in gemrb, except maybe the cache size
[18:19:52] <lynxlynxlynx> btw, torment is the second least playable game in gemrb
[18:21:25] <fuzzie> lynxlynxlynx: did you try that change?
[18:22:28] <lynxlynxlynx> will soon, i'm right in there now
[18:23:49] <ubermad> lynxlynxlynx: why?
[18:24:27] <lynxlynxlynx> because the engine was changed so much
[18:24:45] <lynxlynxlynx> and we haven't implemented all of the needed changes yet
[18:26:14] <lynxlynxlynx> fuzzie: yep, that's all that was needed :)
[18:26:57] <-- Bo_Thomsen has left IRC (Quit: Leaving.)
[18:27:04] <lynxlynxlynx> there is one other thing though
[18:27:08] <lynxlynxlynx> recurring heh
[18:27:23] <lynxlynxlynx> now we give double equipment on the start of tob again
[18:27:34] <ubermad> lynxlynxlynx: what a pity
[18:28:24] <fuzzie> lynxlynxlynx: we set the wrong expansion level i guess?
[18:28:56] <lynxlynxlynx> probably
[18:29:27] <lynxlynxlynx> but then i don't understand how it could've been zero-granting before
[18:29:52] <fuzzie> me neither
[18:30:39] <fuzzie> while i'm looking - bg2/GUILOAD.py has a GemRB.GameSetExpansion() in it
[18:30:57] <fuzzie> can't be right, that takes a param
[18:31:59] <CIA-52> GemRB: 03lynxlupodian * re1c28a181ef0 10gemrb/gemrb/ (core/GUI/GameControl.cpp plugins/SDLVideo/SDLVideo.cpp):
[18:31:59] <CIA-52> GemRB: made the savegame preview fit into the target control
[18:31:59] <CIA-52> GemRB: thanks fuzzie!
[18:31:59] <CIA-52> GemRB: 03lynxlupodian * rf17f7e5eae59 10gemrb/gemrb/GUIScripts/pst/GUISAVE.py: pst: close the save window before generating the screenshot
[18:32:04] <CIA-52> GemRB: 03lynxlupodian * r2e8c41abbbd8 10gemrb/gemrb/GUIScripts/pst/GUISAVE.py: pst: invalidate the window on save game deletion
[18:32:10] <CIA-52> GemRB: 03lynxlupodian * r2fa882737c59 10gemrb/gemrb/core/System/DataStream.cpp:
[18:32:10] <CIA-52> GemRB: DataStream::ReadLine: don't ignore the last character in non-terminated
[18:32:10] <CIA-52> GemRB: tables, thanks fuzzie
[18:32:23] <fuzzie> but bg2 CharGen should be calling GameSetExpansion(3) for SoA and GameSetExpansion(5) for ToB
[18:33:11] <fuzzie> or maybe 4 for SoA with ToB installed, i'm not sure, there are notes somewhere
[18:33:35] <lynxlynxlynx> and that load looks like it should be 4
[18:33:56] <lynxlynxlynx> i think 4 is the ingame transition
[18:34:00] --> Bo_Thomsen has joined #gemrb
[18:34:03] <lynxlynxlynx> 3 and 5 being separate starts
[18:34:14] <fuzzie> the load shouldn't touch anything
[18:35:10] <fuzzie> EF_EXPANSION is set on every EnterGame to make sure it's checked at the right point
[18:36:12] <lynxlynxlynx> a relic from before that
[18:36:19] <fuzzie> i guess so.
[18:36:43] <fuzzie> the original engine has different flags for whether the upgrade came via chargen, old game or action
[18:37:19] <lynxlynxlynx> yep - new game, ingame transition, party import
[18:37:37] <lynxlynxlynx> you get different gear i think
[18:37:42] <fuzzie> we can't differentiate between those, so it's probably broken
[18:37:49] <lynxlynxlynx> they claimed the difficulty adjusts too
[18:38:08] <fuzzie> yeah, the gear thing is a different story
[18:38:28] <fuzzie> chargen is meant to grant the gear itself, then pass 'no new gear' flag to the upgrade function
[18:39:56] <fuzzie> we could do that easily enough, just call the function directly from CharGenEnd and then you can pass it params..
[18:40:30] <fuzzie> but at that point you might as well fix it so it doesn't call GiveEquipment at all
[18:50:22] <lynxlynxlynx> why not just call setexpansion from there instead of GiveEquipment?
[18:50:34] <fuzzie> from where?
[18:50:40] <fuzzie> i mean, GiveEquipment is chargen *only*
[18:51:41] <lynxlynxlynx> heh, no gear
[18:52:20] <fuzzie> i'm not quite sure how this is meant to balance out at all
[18:52:56] <fuzzie> i think you should call GameSetExpansion(4) in chargen, then GameExpansion() directly with some 'no give gear' flag, but i forget
[18:53:46] <lynxlynxlynx> we need to give the amulet of seldarine in the other cases and it seems natural to handle it all in the same function
[18:53:54] <fuzzie> well, can we?
[18:54:10] <fuzzie> i mean, i honestly have no idea
[18:54:17] <lynxlynxlynx> i'll play with it
[18:54:32] <fuzzie> i just want gemrb to stop giving me the starting gear when i move to the expansion
[18:56:51] <-- ubermad has left #gemrb
[19:00:04] <fuzzie> so Taimon notes that it's 3 in the base GAM, and the original engine sets it to 4 and does the upgrade immediately when you open chargen in ToB mode
[19:00:22] <fuzzie> erm, with ToB installed in SoA mode
[19:07:01] <dhewg> what resolutions do you guys support with gemrb?
[19:07:34] <fuzzie> anything your game data provides
[19:08:06] <dhewg> i think this is just vanilla+updates
[19:08:23] <fuzzie> standard game is just 800x600, 1024x768, etc
[19:08:25] <dhewg> cant remember, its bitrotting there for a couple of years :)
[19:09:09] <dhewg> k, im using 800x600, and there's a area hilighting issue on the world map
[19:09:11] <fuzzie> it's a bit fiddly to expand the resolution because the UI backgrounds etc are just one big bitmap, so we just support the widescreen mod, which gives you arbitary resolutions >800x600
[19:09:32] <fuzzie> you have to click in the wrong place?
[19:09:39] <fuzzie> we scale the mouse coordinates incorrectly there i think
[19:09:39] <dhewg> yes
[19:09:55] <dhewg> on 800x600 you get a border
[19:10:09] <dhewg> and the cursor behaves correctly for ouside and inside of that
[19:10:27] <dhewg> but the areas are off, and it looks like it just doesnt respect that border
[19:10:46] <fuzzie> oh, off horizontally too?
[19:11:10] <dhewg> its shifted to the right and down
[19:11:39] <dhewg> hm, maybe the overall scaling, hard to tell
[19:11:51] <dhewg> either way, both axis are off
[19:14:15] <lynxlynxlynx> mhm
[19:14:41] <lynxlynxlynx> we never had it correct
[19:14:41] <dhewg> another thing
[19:14:55] <dhewg> when i rest, the time doesnt advance i think
[19:15:24] <dhewg> the textbox reads "rested for 8 hours", but the tooltip of that pause thingy on the lower left doesnt change
[19:15:30] <fuzzie> the gui tooltip isn't updated
[19:15:40] <dhewg> but its "later"?
[19:17:40] <lynxlynxlynx> yes
[19:18:08] <lynxlynxlynx> it would be visible in the savegame times, plus the day/night cycle is active
[19:21:33] <dhewg> ah right, savegame string confirms it
[20:12:50] <dhewg> another gui issue: when renting a room in an inn, and a dream sequence starts (which isnt visible), there's no way to get out of that status
[20:15:50] <dhewg> selecting multiple items in the sell screen works, but only the last is hilighted
[20:16:11] <dhewg> and when identifying items in a sell screen, they're still selected when switching to 'sell'
[20:16:21] <fuzzie> that is, oddly enough, an original game data bug
[20:16:26] <fuzzie> i mean, the highlighting
[20:16:37] <dhewg> hah
[20:16:39] <dhewg> ok
[20:16:40] <fuzzie> we need to patch up the flags on load
[20:17:21] <dhewg> all those tiny issues i find are now all in our backlogs, should i maybe add items to the tracker for it?
[20:17:53] <dhewg> i mean, if noone looks at it, its pointless, but does anyone?
[20:18:00] <fuzzie> yes, people do use it
[20:18:37] <lynxlynxlynx> or the wiki todo list
[20:18:54] <dhewg> nah, that sucks, we have that at scummvm too
[20:18:57] <lynxlynxlynx> some of the stuff you mentioned is already there
[20:19:03] <dhewg> oh
[20:19:42] <fuzzie> the trouble with tracker is that you could add thousands of entries
[20:20:20] <dhewg> if that helps tracking em and maybe even hope for someone fixing em i would do that :P
[20:20:47] <fuzzie> but mostly bugs people report are stuff we know needs rewrite
[20:20:51] <fuzzie> your bugs are not like that so far
[20:21:34] <dhewg> i can start anytime about the pathfinding or the make-my-eyes-bleed sdl video plugin, if that makes you happy :P
[20:22:14] <fuzzie> sdl plugin due to scaling?
[20:22:40] <dhewg> i looked at it because of the scaling, yes
[20:22:51] <fuzzie> oh, the code
[20:22:53] <fuzzie> it is tricky
[20:23:01] <lynxlynxlynx> scaling to what?
[20:23:10] <fuzzie> 2x, i guess
[20:23:27] <lynxlynxlynx> oh
[20:24:12] <dhewg> yeah, it was an idea because on full hd screens its kinda tiny in windowed mode
[20:24:39] <fuzzie> the SDL driver is not actually so bad if you're just fiddling with the SDL side
[20:24:52] <fuzzie> it's just no-one wants to touch it due to the fragile house of cards that is the BAM blitter
[20:25:12] <dhewg> thats exactly what i was wondering about
[20:25:29] <dhewg> why does the video plugin have to know about the BAM format?
[20:25:57] <fuzzie> because otherwise everything ends up huge and slow
[20:26:16] <fuzzie> i mean, it doesn't really know much, even
[20:26:35] <fuzzie> the BAM flag just means 'paletted sprite which probably needs custom blitting'
[20:27:07] <fuzzie> but for RAM usage issues, it also knows about the RLE
[20:27:31] <dhewg> doesnt that backfire at the CPU usage?
[20:27:40] <fuzzie> nope
[20:27:56] <fuzzie> well, i mean, which bit? :P
[20:27:58] <dhewg> if i run bg2 at 1024x768 i see one core close to 100%, but i didnt check where that comes from
[20:28:19] <fuzzie> i mean, the custom blitting itself is a large CPU consumer
[20:29:34] <fuzzie> but not much to do about that.
[20:30:13] <fuzzie> well, other than implementing an opengl renderer and using the set of fancy 3D flags instead.
[20:34:27] <pupnik> opengles too
[20:34:54] --> Maighstir_ has joined #gemrb
[20:35:20] <pupnik> to the map, add the current kinetic scrolling of standard android / touch apps
[20:35:36] <pupnik> maybe pinch to zoom
[20:35:59] <fuzzie> i am not touching the hacky android libSDL port with a bargepole :p
[20:36:22] <dhewg> words from a wise
[20:38:58] <-- Maighstir has left IRC (Ping timeout: 276 seconds)
[21:50:27] <-- lynxlynxlynx has left IRC (Remote host closed the connection)
[22:01:20] <-- pupnik has left IRC (Ping timeout: 276 seconds)
[22:21:06] <-- Bo_Thomsen has left IRC (Quit: Leaving.)
[22:58:47] <-- Maighstir_ has left IRC (Read error: Connection reset by peer)