#gemrb@irc.freenode.net logs for 5 Mar 2011 (GMT)

Archive Today Yesterday Tomorrow
GemRB homepage

[00:45:20] --> Bo_Thomsen has joined #GemRb
[01:51:55] <-- edheldil_ has left IRC (Ping timeout: 248 seconds)
[01:57:57] <-- ar has left IRC (Read error: Operation timed out)
[02:03:45] --> ar has joined #GemRb
[02:03:51] <-- ar has left IRC (Changing host)
[02:03:51] --> ar has joined #GemRb
[02:05:19] <-- Bo_Thomsen has left IRC (*.net *.split)
[02:07:59] --> Bo_Thomsen has joined #GemRb
[02:16:07] <-- ar has left IRC (Ping timeout: 240 seconds)
[02:37:13] <-- Bo_Thomsen has left IRC (Quit: Leaving.)
[03:29:14] --> ar has joined #GemRb
[03:29:16] <-- ar has left IRC (Changing host)
[03:29:16] --> ar has joined #GemRb
[06:31:10] --> edheldil_ has joined #GemRb
[07:17:19] <tomprince> Looking in diced_effects2, there is an effect called "Charm", but that seems to be the only reference to it anywhere.
[07:19:48] <tomprince> Also diced_effects2 seems to not effect anything to do with dice, simply short-circuiting check_level.
[07:32:18] <-- edheldil_ has left IRC (Ping timeout: 255 seconds)
[07:54:51] <tomprince> https://github.com/tomprince/gemrb/commits/effect-desc
[09:04:22] --> Maighstir has joined #GemRb
[09:08:25] --> lynxlynxlynx has joined #GemRb
[09:08:25] <-- lynxlynxlynx has left IRC (Changing host)
[09:08:25] --> lynxlynxlynx has joined #GemRb
[09:08:25] --- ChanServ gives channel operator status to lynxlynxlynx
[09:18:24] <lynxlynxlynx> yeah, the name is unfortunate
[09:21:47] <tomprince> Do you know anything about "Charm"?
[09:22:50] <lynxlynxlynx> i'll go check
[09:22:57] <lynxlynxlynx> the commits look fine to me
[09:23:21] <tomprince> They should be tested.
[09:23:44] <tomprince> And I dropped the "Charm", since that was the only reference to the bare name.
[09:23:55] <tomprince> I was wondering if it should in fact be something else.
[09:24:46] <lynxlynxlynx> my guess would be State:Charmed
[09:25:02] <lynxlynxlynx> it was added recently for one of the hardcoded pst projectiles
[09:28:25] <tomprince> Well, I'll let one of you push it. And add the flag and check for target.
[09:29:13] <lynxlynxlynx> the charm has zero in both fields, so it should also be fine without being in the list
[10:07:54] --> Edheldil_ has joined #GemRb
[11:43:38] --> Bo_Thomsen has joined #GemRb
[11:51:02] <-- Maighstir has left IRC (Quit: ~ Trillian Astra - www.trillian.im ~)
[12:05:51] --> Maighstir|phone has joined #GemRb
[12:21:40] <-- Maighstir|phone has left IRC (Ping timeout: 255 seconds)
[12:48:06] --> Avenger has joined #GemRb
[12:48:06] --- ChanServ gives channel operator status to Avenger
[12:48:11] <Avenger> hi
[12:49:55] <Avenger> "State:Charmed" is a non-diced effect, the original keeps some other data in the dice fields
[12:57:18] <Avenger> the original engine simply overwrites the dice count/size fields by -1 (0xffffffff)
[12:58:00] <Avenger> then on first cast, it stores the caster's EA and the target's EA in them
[12:59:17] <Avenger> they could have stored them in param3/4, or whatever. but it seems this stuff is ancient, i mean, coming from bg1 when there was no effect v2
[13:07:03] <lynxlynxlynx> ok
[13:07:12] <lynxlynxlynx> just the wrong name was used
[13:07:43] <lynxlynxlynx> i'm playing with tomprince's patches, but still need to resolve this crash
[13:19:13] <Avenger> what's crashing?
[13:19:36] <lynxlynxlynx> more stuff needed to be changed to use EffectDesc
[13:19:42] <lynxlynxlynx> just got it to run
[13:20:19] <lynxlynxlynx> now i'll play around for a bit, to see if the no-actor-needed thing works as expected
[13:20:54] <Avenger> hmm
[13:21:35] <lynxlynxlynx> http://sprunge.us/BMDN?diff <-- this is what i have over his tree
[13:22:57] <Avenger> i see, a new field in the effect reference
[13:23:04] <Avenger> i thought about this
[13:24:25] <lynxlynxlynx> it was mentioned the last two days :)
[13:24:28] <Avenger> so you just add another bit to flags
[13:25:02] <Avenger> first 2 bits for the dice field handling, and one more for the no safe for actors bit
[13:25:41] <Avenger> i mean, safe for non-actors :P
[13:29:11] <lynxlynxlynx> https://github.com/tomprince/gemrb/commit/67e68be418a205d04d4697a30a74222b3de284e0
[13:29:19] <lynxlynxlynx> https://github.com/tomprince/gemrb/commit/38cd050a2075c7a4e35e2abf3a41d6483896c351
[13:29:24] <lynxlynxlynx> these are the parents
[13:32:08] <Avenger> uhm
[13:32:27] <Avenger> this is what i thought about, but i was always lazy to do :P
[13:34:27] <Avenger> i was lazy to separate the two Effect* structs, and so i didn't add the new flag because it would have meant a lot of writing to existing structs :)
[13:35:08] <Avenger> on the other hand it seems to be a faster thing
[13:35:45] <Avenger> i really like it
[13:39:26] <lynxlynxlynx> it's clean :)
[13:39:48] <lynxlynxlynx> the info is in the plugins where it belongs and no extra merging is required
[13:41:34] <Avenger> yes, that's what i like about it. You just have to clean up a few effect refs sprinkled across the files
[13:57:02] <lynxlynxlynx> already managed to hit it :)
[13:57:07] <lynxlynxlynx> bbl lunch
[14:01:14] <Edheldil_> hi, AM GLAD TO SEE TOMPRINCE IMPLEMENTED IT :)
[14:01:20] <Edheldil_> sorry
[14:24:14] <Avenger> hehe
[14:24:21] <Avenger> my ears ring
[14:59:53] <lynxlynxlynx> seems to be working fine, so i'll commit it (without the assert, of course)
[15:07:12] <CIA-91> GemRB: 03tom.prince * r67e68be418a2 10gemrb/gemrb/ (15 files in 8 dirs):
[15:07:12] <CIA-91> GemRB: EffectQueue: Split EffectRef into EffectRef and EffectDesc.
[15:07:12] <CIA-91> GemRB: EffectRef is used for caching opcode lookup.
[15:07:17] <CIA-91> GemRB: 03tom.prince * r38cd050a2075 10gemrb/gemrb/ (5 files in 4 dirs):
[15:07:17] <CIA-91> GemRB: EffectQueue: Change diced effect handling to use flags in EffectDesc.
[15:07:17] <CIA-91> GemRB: Rather than hard coding these properties, allow the plugins to control them.
[15:07:23] <CIA-91> GemRB: 03lynxlupodian * r3c5168fa5df3 10gemrb/gemrb/plugins/FXOpcodes/FXOpcodes.cpp: make State:Charmed ignore the dice/level fields again
[15:07:28] <CIA-91> GemRB: 03lynxlupodian * rc1b44a218ad8 10gemrb/gemrb/ (5 files in 4 dirs):
[15:07:28] <CIA-91> GemRB: added another opcode flag to distinguish those which can run without an
[15:07:28] <CIA-91> GemRB: actor target and tagged the relevant opcodes
[15:26:20] <Avenger> ahh finally i seen a lightspot generated by a vvc
[15:41:38] <MikeChelen> found another trap that crashes, might be related to the other one
[15:46:52] <Avenger> well, the previous commits are supposed to fix all of them
[15:47:38] <-- Avenger has left IRC (Quit: ChatZilla 0.9.86 [Firefox 3.6.13/20101203075014])
[15:51:35] <CIA-91> GemRB: 03avenger_teambg * r585bd29cd6a6 10dltcep/ (8 files): dltcep update (vvc fields)
[15:57:33] <CIA-91> GemRB: 03avenger_teambg * rf006a68a28db 10gemrb/gemrb/core/ (ScriptedAnimation.cpp ScriptedAnimation.h Video.h): added lightspot feature support to vvc's
[16:05:31] <tomprince> Stupid type-unsafe C library functions.
[16:07:01] --> Avenger has joined #GemRb
[16:07:01] --- ChanServ gives channel operator status to Avenger
[16:12:09] <lynxlynxlynx> hey, do you remember why we changed the scroll learning to not use learnspell directly?
[16:13:05] <lynxlynxlynx> now it's done via useitem and fx_learn_spell, so we don't grant any xp
[16:13:29] <lynxlynxlynx> the last part being due to our parameter1 extension
[16:15:04] <lynxlynxlynx> i was thinking of creating another ls_ flag for that purpose
[16:22:15] <fuzzie> because scrolls can have other effects attached
[16:24:44] <lynxlynxlynx> ok, then that should be fine
[16:26:36] <fuzzie> the bg2 tutorial is a good test case
[16:30:34] <CIA-91> GemRB: 03lynxlupodian * r550f472c856e 10gemrb/gemrb/ (4 files in 4 dirs): fixed learning from scrolls not giving any experience
[16:31:16] <lynxlynxlynx> now for something more annoying
[16:31:48] <lynxlynxlynx> we don't toggle the circus travel triggers properly
[16:32:53] <fuzzie> not just an obvious bug?
[16:35:06] <lynxlynxlynx> we'll see
[16:35:17] <lynxlynxlynx> need to dump a bunch of vars
[16:58:29] <lynxlynxlynx> looks like a Dead check fails
[17:00:02] --> SiENcE has joined #GemRb
[17:05:18] <tomprince> https://gist.github.com/856486 or https://gist.github.com/856503 or https://gist.github.com/856507
[17:18:05] <lynxlynxlynx> ifdef the intree building if you don't want it, it is a good default
[17:21:25] <tomprince> Yeah, I plan to, I just haven't gotten around to it.
[17:25:49] <lynxlynxlynx> fuzzie: it's odd, a block in the area script doesn't run at all
[17:26:12] <lynxlynxlynx> i'm breaking on the first trigger
[17:26:32] <lynxlynxlynx> the second condition is true if the order of execution was different
[17:27:10] <lynxlynxlynx> scripts/bg2/ar0700.baf:76: Dead("KalahIllusion") <-- this
[17:27:35] <lynxlynxlynx> in ar0607.baf the same trigger was checked normally
[17:33:10] <lynxlynxlynx> basically no Dead trigger in that script gets run, only the ones from baldur.bcs
[18:20:56] <-- Avenger has left IRC (Quit: bye!)
[18:41:25] --> _pickle has joined #GemRb
[18:43:37] <lynxlynxlynx> doooh
[18:43:50] <lynxlynxlynx> it's an old save, so the area script slots are unset
[18:43:58] <lynxlynxlynx> grrr
[18:58:44] <-- Bo_Thomsen has left IRC (Quit: Leaving.)
[19:49:17] --> Maighstir has joined #GemRb
[20:02:56] --> Avenger has joined #GemRb
[20:02:56] --- ChanServ gives channel operator status to Avenger
[20:03:31] <Avenger> fuzzie or wjp can you make a spiraling projectile path?
[20:04:29] <wjp> what parameters does it have? and can you point at the location in the code where it would go?
[20:04:52] <wjp> start point, end point, ... ?
[20:12:53] <Avenger> hmm
[20:13:04] <Avenger> i got a start point and an end point, yes
[20:13:17] <Avenger> Destination is the end point
[20:15:01] <Avenger> i don't know what would be better
[20:15:13] <Avenger> we have code for precalculating the whole path
[20:15:22] <Avenger> path = area->GetLine( Pos, Destination, Speed, Orientation, GL_PASS );
[20:15:43] <Avenger> but, the original engine doesn't precalculate it
[20:16:31] <Avenger> anyway, if you go by this route, GL_SPIRAL will be a new value
[20:16:41] <Avenger> and if you are there, you could implement GL_REBOUND too :)
[20:16:51] <Avenger> that would be for lightning bolts
[20:17:05] <Avenger> which bounce off of impassable searchmap tiles
[20:26:31] <wjp> which projectile(s) use(s) it?
[20:27:21] <Avenger> the spiral one_
[20:27:24] <Avenger> ?
[20:27:27] <wjp> yes
[20:27:32] <Avenger> speak with dead, so far
[20:27:54] <Avenger> i will mark them by curved path+incoming path in the file
[20:30:08] <Avenger> i think there are other projectiles that will need this, but this one is the easiest, because it is a single travel projectile
[20:30:28] <Avenger> there are others with multiple spiraling thingies, i think
[20:30:32] <Avenger> maybe even in HoW
[20:31:45] <Avenger> hmm the filename is 'stories.pro'
[20:33:40] <lynxlynxlynx> bouncing would be nice
[20:33:55] <Avenger> yep, that is missing even from bg1/2
[20:33:56] <lynxlynxlynx> it's easy to notice since the lightning spell exists in all games
[20:34:02] <lynxlynxlynx> yeah
[20:34:16] <lynxlynxlynx> plus the bg2 intro makes it very obvious
[20:34:32] <Avenger> where?
[20:34:41] <Avenger> ahh, the lightning machine?
[20:35:25] <lynxlynxlynx> yeah
[20:44:45] <lynxlynxlynx> Avenger: 3a827867a865ab5b928b0f8945a0d175f29a0a44
[20:45:00] <Avenger> what's that?
[20:45:06] <lynxlynxlynx> shouldn't also the map be changed in the plugin? or about the other games
[20:45:11] <lynxlynxlynx> one of your commit hashes
[20:45:15] <lynxlynxlynx> git show 3a827867a865ab5b928b0f8945a0d175f29a0a44
[20:45:35] <Avenger> yes i see it
[20:45:39] <Avenger> what's up with it?
[20:46:13] <lynxlynxlynx> since the name in the opcode list didn't change, bg2 gives a warning
[20:46:16] <lynxlynxlynx> [EffectQueue]: Couldn't assign effect: summondisable
[20:46:25] <Avenger> hmm, that sucks
[20:46:53] <Avenger> the effects.ids should be changed in bg2 then
[20:46:59] <lynxlynxlynx> { "AvatarRemoval", fx_avatar_removal, 0, -1 }, //unknown
[20:47:00] <lynxlynxlynx> or that
[20:47:18] <Avenger> 0x13b AvatarRemovalModifier
[20:47:19] <lynxlynxlynx> i'm not sure why you renamed it in the ids
[20:47:31] <Avenger> oh
[20:47:32] <lynxlynxlynx> the specific plugins have precedence
[20:47:40] <lynxlynxlynx> that's a different one
[20:47:47] <Avenger> they are different
[20:47:52] <Avenger> so they need a different name
[20:48:13] <Avenger> iwd loads the bg2 plugin
[20:48:28] <lynxlynxlynx> oh, they're also at different addresses
[20:48:36] <Avenger> yes
[20:48:52] <lynxlynxlynx> so i'll just rename it in the map and in the ids of the other games
[20:48:53] <Avenger> they are two different opcodes at this point
[20:49:09] <lynxlynxlynx> do you know if iwd2 uses the iwd one?
[20:49:15] <Avenger> yes it does
[20:49:50] <CIA-91> GemRB: 03lynxlupodian * r0e9d26469181 10gemrb/gemrb/core/GameScript/Actions.cpp:
[20:49:50] <CIA-91> GemRB: TakePartyItem*: also take undroppable items
[20:49:50] <CIA-91> GemRB: fixes the chateau irenicus' portal key not being removed on exit (or ever)
[20:50:00] <Avenger> i'm sorry for not completing the change, most likely i left out some files from a commit
[20:50:48] <Avenger> hmm looks like i didn't change the effectref in fxopcodes?
[20:52:07] <lynxlynxlynx> yes
[20:52:13] <lynxlynxlynx> i'm on it
[20:52:25] <lynxlynxlynx> turns out only bg1 is missing then, pst doesn't have this override
[20:52:46] <Avenger> it looks like bg2 has both types
[20:52:56] <lynxlynxlynx> under the modifier one
[20:53:16] <Avenger> yes
[20:53:28] <Avenger> the fx_avatar_removal should be renamed to fx_summondisable
[20:53:40] <Avenger> the fx_avatar_removal_modifier one exists in both games
[20:54:44] <Avenger> opcode 0x10f is the summondisable opcode for bg2, 0x13b is the avatar removal in bg2, and 0x104 in iwd/iwd2
[20:55:02] <lynxlynxlynx> ok, i'll do that too
[20:55:09] <-- SiENcE has left IRC (Ping timeout: 255 seconds)
[20:55:11] <CIA-91> GemRB: 03lynxlupodian * rc67037836cd8 10gemrb/gemrb/ (override/bg1/effects.ids plugins/FXOpcodes/FXOpcodes.cpp): complete the SummonDisable/AvatarRemoval swap in bg1/bg2
[20:56:26] --> Bo_Thomsen has joined #GemRb
[20:57:24] <CIA-91> GemRB: 03lynxlupodian * r1c569534b6d0 10gemrb/gemrb/plugins/FXOpcodes/FXOpcodes.cpp: renamed fx_avatar_removal to fx_summon_disable
[20:57:27] <lynxlynxlynx> nice progress today
[20:58:00] <lynxlynxlynx> need to look at that wild magic thing and the crazy ranged ammo usage
[20:58:20] <lynxlynxlynx> then check beholder's list for any obvious regressions
[20:58:20] <Avenger> we still have 2 hours from this day :)
[20:58:27] <Avenger> i'm working on pst projectiles :)
[20:58:52] <lynxlynxlynx> so i'm thinking about a release in two weeks time
[20:59:09] <lynxlynxlynx> what do you think?
[20:59:12] <Avenger> any visible progress? besides the regressions? :)
[20:59:37] <lynxlynxlynx> sure
[20:59:39] <Avenger> ok, i coded a bunch of projectiles for pst
[21:00:39] <lynxlynxlynx> wild magic
[21:00:45] <lynxlynxlynx> check the NEWS
[21:00:49] <lynxlynxlynx> almost time for another update
[21:00:51] <Avenger> ah yes, i didn't even try it yet
[21:00:56] <lynxlynxlynx> android stuff
[21:01:25] <lynxlynxlynx> pst maze, bik player
[21:01:28] <Avenger> the wild magic is good, it just swung from barely implemented to barely buggy :)
[21:01:31] <lynxlynxlynx> these are the highlights
[21:01:41] <lynxlynxlynx> :)
[21:01:47] <Avenger> the bik player :( that sucks for me, still
[21:02:00] <Avenger> and we didn't fix its audio
[21:02:04] <lynxlynxlynx> also did dead magic and stuff like that, so silenced casters can chant
[21:02:14] <Avenger> wjp just fixed some small artifacts, actually
[21:02:15] <lynxlynxlynx> so it will be a fine release
[21:02:33] <lynxlynxlynx> well, for me it is now almost perfect, before it was heavy on the artifacts
[21:02:40] <Avenger> oh?
[21:02:42] <Avenger> where?
[21:03:23] <Avenger> which movie was buggy for you? my intro movie is still bad
[21:03:44] <fuzzie> i thought we established that it worked for everyone except you
[21:03:46] <Avenger> but only here, it is perfect when i play it in my workplace computer
[21:04:02] <fuzzie> which is why i was asking whether you were sure it was loading the right plugin, it's so weird
[21:04:05] <Avenger> yes fuzzie, but i want to know what was fixed for lynx recently
[21:04:32] <Avenger> hmm, did the plugin name change maybe?
[21:04:43] <Avenger> maybe i still load the bad one?
[21:05:12] <fuzzie> i don't think so, but it should print the plugin paths at startup? so, you could see
[21:05:46] <lynxlynxlynx> i don't remember the details, movies are off unless i say so
[21:06:11] <Avenger> nope, i just recompiled it
[21:06:30] * wjp really didn't see or hear any remaining artifacts
[21:06:55] <Avenger> wjp: audio is good for you too?
[21:06:58] <wjp> yes
[21:07:01] <Avenger> whoa
[21:07:36] <Avenger> mine still cuts every blocks end resulting in a sped up audio
[21:07:58] <Avenger> i wonder if my video and audio artifacts are of the same cause
[21:08:12] <fuzzie> the source file you're compiling from is the right one too?
[21:08:14] <wjp> hm, that sounds rather different than the stuttering I used to get
[21:08:39] <Avenger> bikplayer.cpp 38372
[21:09:05] <wjp> same filesize here
[21:09:51] <lynxlynxlynx> tried make clean anytime?
[21:12:12] <Avenger> yes
[21:12:21] <Avenger> actually, i just recompiled everything in release mode
[21:12:28] <Avenger> to see if my exit crasher is still there
[21:13:42] <-- Demitar_ has left IRC (Quit: Ex-Chat)
[21:15:04] <Avenger> sadly, yes
[21:17:50] <fuzzie> exact same crash?
[21:19:21] --- ChanServ gives channel operator status to fuzzie
[21:19:40] <Avenger> the one caused by the missing releasing of the symbol tables is gone, that happened even in debug mode
[21:19:54] <Avenger> this one doesn't exist in debug mode
[21:20:00] <Avenger> which makes it harder to catch
[21:20:34] <Avenger> i think this one is older than the Holder thing
[21:20:40] <fuzzie> you don't get symbol names in debug mode?
[21:20:59] <Avenger> i don't get crash in debug mode anymore ;)
[21:21:22] <fuzzie> sure, but you can still have symbol names
[21:22:10] <fuzzie> i guess you probably have to enable them manually though
[21:22:38] <CIA-91> GemRB: 03lynxlupodian * r808d46fe52e3 10gemrb/gemrb/core/GameScript/GameScript.cpp: assign the iscreaturehiddeninshadows trigger, which is already implemented
[21:24:46] <Avenger> lol, i always leave some of those out
[21:26:18] <Avenger> wjp, fuzzie: http://www.mypicx.com/03052011/buggy_bik/
[21:26:37] <Avenger> the falling leaves leave snail trails
[21:27:08] <lynxlynxlynx> it was implemented in 2005 :P
[21:27:27] <Avenger> the color tilings are from converting to png, so ignore those
[21:27:37] <Avenger> hehe lynx
[21:27:45] <Avenger> that's hilarious
[21:28:22] <lynxlynxlynx> no such artifacts here
[21:28:38] <lynxlynxlynx> i'm pretty sure there were before the latest batch
[21:28:57] <Avenger> huh
[21:29:07] <wjp> I never saw those
[21:29:07] <Avenger> so it was fixed for everyone but me
[21:29:19] <Avenger> O_o
[21:30:14] <Avenger> wjp: besides these, i got only those minor artifacts with the pattern block where i got vertical stripes instead of horizontal because of the cycle variable swap
[21:31:11] <Avenger> so from 3 of my artifacts, i experienced only 1 fix. But it seems everyone else got much better results
[21:32:26] <fuzzie> but it works fine at your work, and you used the same .bik file there?
[21:33:52] --> Demitar has joined #GemRb
[21:38:39] <-- lynxlynxlynx has left IRC (Remote host closed the connection)
[21:48:16] <Avenger> yes
[21:49:45] <Avenger> i copied the movie file and gemrb source after it failed here, so i brought it to my work, where it was fine
[21:50:39] <Avenger> hmm and now the cone projectile doesn't work :(
[21:50:44] <Avenger> meh
[21:58:28] <Avenger> ooh
[22:34:59] <Avenger> ok, this took a bit long, but it is nice
[22:35:06] <-- Avenger has left IRC (Quit: ChatZilla 0.9.86 [Firefox 3.6.15/20110303024726])
[23:03:17] <CIA-91> GemRB: 03avenger_teambg * r0704b4de2fef 10gemrb/gemrb/ (4 files in 2 dirs): cone of cold projectile
[23:03:20] <CIA-91> GemRB: 03avenger_teambg * r3934550a4297 10gemrb/gemrb/ (4 files in 3 dirs): Merge branch 'master' of ssh://gemrb.git.sourceforge.net/gitroot/gemrb/gemrb
[23:27:12] <tomprince> :q