#gemrb@irc.freenode.net logs for 7 Aug 2009 (GMT)

Archive Today Yesterday Tomorrow
GemRB homepage


[00:10:57] <Edheldil> night
[00:11:00] <-- Edheldil has left IRC ("Really?")
[01:07:56] <-- Gekz has left IRC (orwell.freenode.net irc.freenode.net)
[01:07:56] <-- |Cable| has left IRC (orwell.freenode.net irc.freenode.net)
[01:10:17] --> Gekz has joined #GemRb
[01:10:17] --> |Cable| has joined #GemRb
[01:46:16] <-- Gekz has left IRC (Read error: 104 (Connection reset by peer))
[01:46:24] --> Gekz has joined #GemRB
[01:55:49] --> pupnik has joined #gemrb
[02:11:56] <-- pupnik_ has left IRC (Read error: 110 (Connection timed out))
[03:34:21] --> xrogaan has joined #gemrb
[08:37:16] --> lynxlynxlynx has joined #gemrb
[08:37:17] --- ChanServ gives channel operator status to lynxlynxlynx
[08:54:40] <-- xrogaan has left IRC (Remote closed the connection)
[09:40:18] --> Pygma has joined #gemrb
[09:41:00] <fuzzie> hello
[09:42:55] <Pygma> Hello there
[09:43:16] <fuzzie> oh, that was my daily hello in the morning. but also, hi :)
[09:43:27] <Pygma> Ah ok
[09:44:03] <fuzzie> lynxlynxlynx: zefklop was wondering another day about whether we gave minuses to attack rolls when using non-proficient weapons? i don't understand any of it, but i thought you might know
[09:45:06] <lynxlynxlynx> i'll check
[09:46:52] <lynxlynxlynx> nope
[09:47:29] <lynxlynxlynx> we only give the dualwielding penalty
[09:48:26] <fuzzie> ok, thanks
[09:54:25] <lynxlynxlynx> it's just a -4 thac0 penalty, right?
[09:56:26] <fuzzie> i have no idea, sorry
[09:56:56] <lynxlynxlynx> why couldn't they just make it trivial and have -4 in that column of wspecial.2da :(
[09:58:00] <lynxlynxlynx> nice!
[09:58:07] <Gekz> ?
[09:58:24] <lynxlynxlynx> due to code completion, I found a THACOBonus vs THAC0Bonus inconsistency
[09:59:02] <Gekz> lol
[09:59:36] <lynxlynxlynx> no bugs though :)
[10:02:38] <Gekz> o.o
[10:02:43] <Gekz> whose code has that bug
[10:06:00] <lynxlynxlynx> mattinm's, but as I said, it's not really a bug, the behaviour is fine
[10:07:00] <CIA-22> gemrb: 03lynxlupodian * r6824 10/gemrb/trunk/gemrb/plugins/Core/Actor.cpp: added non-proficiency tohit penalty, which is missing from the wspecial.2da
[10:08:39] <CIA-22> gemrb: 03lynxlupodian * r6825 10/gemrb/trunk/gemrb/plugins/Core/Actor.cpp: fixed a THACOBonus vs THAC0Bonus inconsistency
[10:14:52] <CIA-22> gemrb: 03lynxlupodian * r6826 10/gemrb/trunk/NEWS: NEWS update
[11:17:24] --> Cable_ has joined #gemrb
[11:23:35] --> barra_home has joined #gemrb
[11:25:39] <-- |Cable| has left IRC (Read error: 110 (Connection timed out))
[11:27:51] --> tombhadAC has joined #gemrb
[11:30:21] --- barra_home is now known as barra_library
[11:37:12] <-- Cable_ has left IRC (Read error: 110 (Connection timed out))
[11:45:19] <-- Pygma has left IRC (Remote closed the connection)
[12:03:58] --> |Cable| has joined #gemrb
[12:08:23] <-- dawid has left IRC (Read error: 110 (Connection timed out))
[12:09:23] --> dawid has joined #GemRb
[12:53:33] <-- barra_library has left IRC ("Verlassend")
[13:28:19] <-- Gekz has left IRC (Read error: 110 (Connection timed out))
[13:43:46] <-- lynxlynxlynx has left IRC (Remote closed the connection)
[13:49:54] --> lynxlynxlynx has joined #gemrb
[13:49:54] --- ChanServ gives channel operator status to lynxlynxlynx
[13:53:11] --> barra_home has joined #gemrb
[14:00:55] <fuzzie> that TlkOverride thing is confusing, i wonder how it works
[14:55:17] <fuzzie> okay, does anyone know immediately how the Die() trigger works?
[14:55:48] <lynxlynxlynx> nope
[14:55:58] <lynxlynxlynx> zipi: did you upload your work somewhere yet?
[14:56:24] <fuzzie> it's in the sourceforge tracker
[14:56:53] <fuzzie> but edheldil and zefklop think it should be handled elsewhere
[14:57:14] <fuzzie> so it probably needs some heavy editing before applying
[15:04:03] <fuzzie> and I was going to ask about the chunks animation type, but I guess that is 0x0100 since our avatars.2da has SPCHUNKS there
[15:04:26] <fuzzie> not there in pst, though..
[15:04:51] <fuzzie> i guess our pst avatars.2da is just missing huge amounts
[15:05:09] <lynxlynxlynx> could be
[15:05:35] <fuzzie> whenever i ask Avenger about it, he avoids talking, i think it must be horrible :)
[15:05:36] <lynxlynxlynx> dltcep can generate it/part of it, but this is probably what we already have
[15:10:09] <fuzzie> or maybe pst just simply doesn't have chunky death.
[15:34:13] --> xrogaan has joined #gemrb
[16:12:59] <pupnik> heh @ fuzzie
[17:51:11] <fuzzie> i am not having much luck with gemrb recently, everything i try fixing is a hopeless cause :(
[18:53:55] --> Avenger has joined #gemrb
[18:54:01] --- ChanServ gives channel operator status to Avenger
[18:54:03] <Avenger> hi
[18:56:04] <Avenger> fuzzie: pst has some traces of the bg1 avatar system, but i didn't try to research it yet
[18:56:32] <Avenger> i'm working on the tlk override code
[18:57:52] <Avenger> if you thought our current tlk override is confusing, then you won't like the one i'm working on :) The current one is barely enough to decode custom map notes, it couldn't even decode custom bios
[19:16:14] <fuzzie> well, if you fix it, then i am happy :)
[19:22:20] <Avenger> the bio string is already working, but i'm still working on altering or adding custom strings
[19:22:25] <Avenger> it is a wicked format
[19:22:58] <Avenger> default.toh is a mapping of overwritten strrefs + offsets in default.tot
[19:23:27] <Avenger> default.tot is a dump of intertwined double linked lists
[19:23:32] <fuzzie> ah
[19:23:38] <fuzzie> i was wondering why all the unknowns in iesdp :)
[19:23:57] <Avenger> if a text is longer than 512 byte, it is stored in separate segments
[19:24:13] <Avenger> each segment has 2 pointers, one to the previous segment an one to the next segment
[19:24:24] <Avenger> there is a 'free segments list' too
[19:25:03] <Avenger> i think the engine stores default.toh in the memory, because if you remove a string, its entry is vanished, not just zeroed out
[19:25:23] <Avenger> but default.toh is made in this segmented form to be able to load only used strings
[19:25:47] <Avenger> even if they are possible to change, or remove
[19:26:06] <fuzzie> ok, splitting segments up is kind of horrid :)
[19:26:25] <Avenger> yep, and maintaining the free list, on disc file :)
[19:27:19] <Avenger> also, 0xf240-0xf245 are special strings, even if default.tlk has entries for them, they are ALWAYS mapped through tlkoverride
[19:27:36] <Avenger> they are the bio strings for player1-5
[19:27:37] <fuzzie> 6 of them .. something for each party member?
[19:27:38] <Avenger> err 6
[19:27:38] <fuzzie> ah
[19:27:59] <fuzzie> this is bg2?
[19:28:06] <Avenger> and how and iwd2
[19:28:20] <Avenger> luckily all of them got the same hardcoded 0xf240
[19:28:55] <Avenger> f240 was probably the original SoA size of dialog.tlk
[19:29:17] <Avenger> it is a shortsighted hack
[19:38:38] <fuzzie> i had a go at the magic missile thing using a path, but it doesn't work so well
[19:38:46] <fuzzie> i think you were right, and it should just be adjusted on display
[19:39:37] <fuzzie> but the impact times are different for each missile, i think, because the curve seems to be affecting when they hit..
[19:40:33] <Avenger> they are created one after the other, not simultaneously
[19:40:45] <Avenger> i will fix that once the curve is done
[19:40:52] <Avenger> or you can do it :)
[19:40:53] <fuzzie> i mean, in the original game
[19:41:01] <Avenger> oh
[19:41:09] <fuzzie> the outside-most missiles hit at a different time to the inside ones
[19:41:11] <Avenger> didn't you say they are simultaneous?
[19:41:32] <fuzzie> i think because they hit as soon as they are inside the actor's circle
[19:41:46] <Avenger> yeah, that should be done too
[19:42:01] <fuzzie> but then the hit-checking has to adjust position too :) i guess that is not hard either..
[19:42:11] <fuzzie> but i thought i should ask if you think that's a good idea
[19:42:49] <Avenger> that is easy, and i agree
[19:43:30] <Avenger> though it wouldn't be a big problem if they need to go a step further
[19:43:38] <Avenger> at least, they will hit a bit delayed :)
[19:44:25] <Avenger> huh, at last, gemrb compiles again :)
[19:44:49] <Avenger> i had to do a complete clean&rebuild, virtual methods suck
[19:45:09] <Avenger> at least in msvc6
[19:45:13] <Avenger> it is a lazy compiler :)
[20:41:23] <fuzzie> heh
[20:41:39] <fuzzie> "We cannot detect a CD-ROM drive on your system. You cannot play Baldur's Gate without a CD-ROM drive."
[20:41:47] <fuzzie> i guess i need some fake cd mounting thing..
[20:44:48] <fuzzie> after that, it does seem to work very well under wine, neat
[20:48:57] <fuzzie> trying to take a more careful look at the magic missile behaviour, not very easy
[20:53:14] <fuzzie> they seem to go out as an arc, and then curve back in at the end?
[20:55:27] <Avenger> yes
[20:56:16] <fuzzie> trying to get the arc right is not so easy.. maybe it really is better to make a path
[20:57:17] <Avenger> i tried to calculate the deviation from straight line already
[20:57:31] <Avenger> it is the biggest when the projectile is midway
[20:57:49] <fuzzie> yes, the first half is not so hard
[20:57:52] <Avenger> it also depends on a seed value, which could be calculated from the projectile ID
[20:58:16] <Avenger> i store an Origin coordinate, and calculate the distance from it
[20:58:33] <fuzzie> but the second half doesn't look so simultaneous
[21:00:52] <fuzzie> if you pause it after half way with 5 missiles, then three of the missiles are in a line, but two on one side are in front..
[21:01:15] <fuzzie> in fact, those three don't move for several frames, while the other two are moving towards the target..
[21:01:33] <fuzzie> and then eventually the other three start moving again, finishing the arc
[21:01:46] <fuzzie> i wonder if i am just seeing stupid wine bugs here? someone with real windows should verify
[21:02:10] <fuzzie> well, i guess i will borrow a machine tomorrow
[21:08:25] <Avenger> well, gemrb doesn't even seem to render all the path positions
[21:09:11] <Avenger> maybe the speed isn't calculated correctly and i skip too many path nodes
[21:11:55] <fuzzie> well
[21:14:18] <fuzzie> i don't know about the speed, but it should really be a smooth path anyway
[21:14:57] <fuzzie> you think it doesn't render the positions just because it's going too fast?
[21:19:25] --> Edheldil has joined #gemrb
[21:19:25] --- ChanServ gives channel operator status to Edheldil
[21:20:11] <Edheldil> hi
[21:20:20] <fuzzie> hi
[21:42:46] --- barra_home is now known as barraAway
[21:49:25] <-- lynxlynxlynx has left IRC (Remote closed the connection)
[21:57:45] <Avenger> i don't know what's the cause, but only very few positions are rendered
[22:03:23] --> Gekz has joined #GemRB
[22:11:22] <Avenger> heh, i have to modify a lot of code to save these map notes
[22:12:23] <fuzzie> well, don't break pst :)
[22:29:45] <Avenger> i changed vfs, filestream, so this has a bigger potential than simply breaking pst :D
[22:31:27] <fuzzie> that is usually not too fragile
[22:33:27] <Avenger> well, it exploded a few times :)
[22:33:33] <Avenger> but now it seems fine
[22:33:35] <Avenger> uploading soon
[22:33:37] <-- Avenger has left IRC ("ChatZilla 0.9.85 [Firefox 3.5.2/20090729225027]")
[22:37:37] <CIA-22> gemrb: 03avenger_teambg * r6827 10/gemrb/trunk/gemrb/plugins/Core/VFS.cpp: allow concurrent read/write
[22:39:58] <CIA-22> gemrb: 03avenger_teambg * r6828 10/gemrb/trunk/gemrb/plugins/Core/ (FileStream.cpp FileStream.h): implemented Modify (the crossover of Open and Create)
[22:46:19] --> Avenger has joined #gemrb
[22:46:27] --- ChanServ gives channel operator status to Avenger
[22:47:53] <Avenger> i didn't really test the code, there are lots of possible crashes
[22:47:59] <CIA-22> gemrb: 03avenger_teambg * r6829 10/gemrb/trunk/gemrb/ (17 files in 6 dirs): custom mapnotes and bio strings are now saved and modifiable
[22:48:06] <Avenger> especially if the length of the mapnote changes
[22:48:22] <fuzzie> both of the earlier changes seem ok, except the 'opened = true' changed in Create is strange?
[22:48:30] <Avenger> it is needed, i think
[22:48:35] <-- barraAway has left IRC ("Verlassend")
[22:49:19] <Avenger> create should work exactly like Modify, except that it doesn't need an existing file
[22:50:03] <fuzzie> your formatting goes a bit wonky in places
[22:50:05] <Avenger> maybe we don't even need both. But the created file should be readable
[22:50:25] <Avenger> hmm, maybe i forgot to format some files?
[22:51:46] <CIA-22> gemrb: 03avenger_teambg * r6830 10/gemrb/trunk/gemrb/plugins/TLKImporter/TlkOverride.cpp: format!
[22:52:05] <fuzzie> that is a lot of code, anyway
[22:53:55] <Avenger> well, i worked on it for some time
[22:54:17] <Avenger> i hope it didn't break normal usage
[22:55:04] <fuzzie> well, it looks fine at a glance
[22:55:40] <Avenger> i'm not even sure it is able to modify a short string to longer
[22:55:46] <Avenger> i tried the other way
[22:56:02] <Avenger> equal segment counts, and less segment counts seem to be working
[22:57:08] <Avenger> the biography editor is not working yet
[22:57:24] <Avenger> the Edit control lacks the multiline edit feature
[22:57:58] <Avenger> and there is an ugly background image too
[22:58:20] <fuzzie> the windows one?
[22:58:51] <Avenger> yes. also, if you modify a map note, you won't notice the change in the files. The .tot/.toh is updated only when the area is swapped out
[23:00:06] <Avenger> the tot file should be loaded in memory too
[23:00:25] <Avenger> err. i mean toh. The one with the strref -> offset mapping
[23:00:45] <Avenger> it is a simple mapping, and it is easier to remove an element if the whole is in memory
[23:03:27] <-- tombhadAC has left IRC (Read error: 101 (Network is unreachable))
[23:07:10] <Avenger> bye
[23:07:11] <-- Avenger has left IRC ("bye!")