#gemrb@irc.freenode.net logs for 12 Oct 2012 (GMT)

Archive Today Yesterday Tomorrow
GemRB homepage


[06:26:56] --> lynxlynxlynx has joined #gemrb
[06:26:56] <-- lynxlynxlynx has left IRC (Changing host)
[06:26:56] --> lynxlynxlynx has joined #gemrb
[06:26:56] --- ChanServ gives channel operator status to lynxlynxlynx
[07:10:43] <edheldil> Hi, guys!
[07:11:30] <lynxlynxlynx> gmornin
[07:38:24] --> rocket_hamster has joined #gemrb
[07:47:35] --> Avenger has joined #gemrb
[07:47:59] <Avenger> > <lynxlynxlynx> avenger leans to that side, but i'm not sure what his general verdict would be
[07:48:23] <Avenger> well, i say, use c++ if you really need to, but don't use it just because you can
[07:49:34] <Avenger> extreme use of c++ features also works as code obfuscation, but that could be said about using c where c++ is due.
[07:51:07] <fuzzie> and always make sure you have a convenient wjp to write your templates for you, if you need them.
[07:54:08] <lynxlynxlynx> :)
[07:58:05] <edheldil> so, have you guys backed Eternity ? :)
[08:00:31] * wjp glares at fuzzie :-)
[08:01:47] <wjp> edheldil: for more $'s than I had originally intended :-)
[08:02:29] <Avenger> i'm not entirely sure about eternity. It uses a 3d engine. Wouldn't that make it more like NWN
[08:03:03] <edheldil> hehe, going to get cast miniatures and cut out diorama of the whole world? :)
[08:03:09] <Avenger> on the other hand, we should try implementing the 3d avatars finally :D
[08:03:49] <edheldil> they have a picture there that looks stunning - whatever its relation to the game is
[08:03:55] <edheldil> hehe
[08:03:59] <Avenger> What i don't like about 3d engines is the limited party size
[08:04:19] <Avenger> hmm i didn't see that yet
[08:04:30] <edheldil> well, you are mixing 3d with fps, aren't you?
[08:04:59] <wjp> Avenger: they said party of 6
[08:07:36] <edheldil> Avenger: what do you mean with 3d avatars? Rendering 3d character models on the fly like in Lionheart?
[08:08:47] <edheldil> that would mean 3d engine as well (for some definition of 3d engine, for sure)
[08:09:14] <rocket_hamster> why do they need 3d engine? they intented to make isometric game
[08:11:29] <rocket_hamster> oh i though you are refering to baldurs gate enhanced edition
[08:11:34] <lynxlynxlynx> easier to do nicer lighting&effects, one less avatar step, better use of the gfx card
[08:11:56] <lynxlynxlynx> eternity will have a locked isometric view afair
[08:14:04] <rocket_hamster> sounds strange to me
[08:15:32] <rocket_hamster> or at least complicated
[08:17:17] <rocket_hamster> so they will use 3rd models and rotate them with some weird angle to observer
[08:17:39] <edheldil> well, as lynx said, it makes lighting fx easier - e.g. shadows from a sun changing
[08:17:57] <edheldil> why weird?
[08:18:02] <rocket_hamster> yes but do you need 3rd coordinate for that or just gfx card?
[08:18:31] <rocket_hamster> i dont doubt using gfx accell just why 3rd coordinate except for textures overlay
[08:19:10] <rocket_hamster> it sounds strange to me because iosometric view implies some angle to the scene
[08:19:28] <rocket_hamster> if they do that they will have to compute actor position in space with some weird math
[08:19:38] <edheldil> If they go the route of a full 3d just with locked camera, which I think will be the case, those things are relatively straightforward
[08:20:18] <rocket_hamster> yes thats what i mean, 3d with locked camera
[08:20:24] <rocket_hamster> but will that rly be like BG2?
[08:20:31] <rocket_hamster> then you cant really reproduce the art side
[08:20:46] <lynxlynxlynx> hypothesis
[08:21:04] <lynxlynxlynx> the area shot they posted looks great
[08:21:07] <edheldil> Have you seen Lionheart?
[08:21:09] <rocket_hamster> imho iwd2 was top of computer graphics in rpgs
[08:21:24] <rocket_hamster> lynx that shot looks like fan art to me, if its ingame im impressed
[08:22:14] <lynxlynxlynx> at 3M pixels :)
[08:22:22] <rocket_hamster> lionheart is 3d?
[08:23:30] <rocket_hamster> i just saw on youtube, i woulnt say its bad, i just like iwd2 better
[08:24:02] <edheldil> Only partially. Avatars, items are 3d, the rest might be composed of smaller parts. http://www.abcgames.cz/simage.asp?s=8&id=6266&num=47779
[08:24:55] <edheldil> so possibly it's not relevant., sorry
[08:25:24] <rocket_hamster> well its easier to do 3d model, because of all the armor, items etc you can texture and dont have to art everything up
[08:25:30] <rocket_hamster> as well as rotation of character etc
[08:25:38] <rocket_hamster> once you have 3d model you have every angle
[08:26:04] <rocket_hamster> as someone who knows nothing about art i would say creating 3d mdoel is easier then drawing
[08:26:14] <rocket_hamster> at least for me so i imagine its far easier for artist
[08:26:23] <edheldil> I had a vague notion of it containing 3d models of environment, but they are prerendered images spliced on 3d positions or somesuch
[08:28:42] <Avenger> that "environment screenshot" is not isometric, more like first person, i guess
[08:32:09] <Avenger> edheldil: i know only toee which rendered 3d avatars on isometric area art
[08:37:33] <rocket_hamster> edheldil, you can create 3d model then export its angles to png
[08:37:34] <edheldil> Lionheart did as well, I believe (e.g. the trees are certainly 2d images) , but I would have to check to be sure of the ratio
[08:38:19] <edheldil> rocket_hamster: you don't need angles if you use isometric camera and lock it's angle
[08:38:28] <edheldil> its
[08:38:33] <rocket_hamster> you need if your character moves around :)
[08:38:39] <rocket_hamster> facing, backwards
[08:38:41] <rocket_hamster> etc
[08:39:04] <edheldil> yes, but that's only the avatars
[08:39:17] <edheldil> you don't need back side of houses, for example
[08:39:21] <Avenger> stances and orientation, yeah. That takes up space. And that's why i would support 3d avatar rendering
[08:39:41] <rocket_hamster> well thats probably only reason to use 3d
[08:39:54] <Avenger> it would remove lots of hassle and there could be better animations
[08:40:34] <edheldil> The main reason is IMO to render the avatar with correct equipment
[08:41:07] <Avenger> that too, though we all seen 3d avatars with weird equipment renderning :D
[08:41:42] <edheldil> e.g. you don't need to prerender knight in breaches - knight in mail - knight in mail with shield - ...
[08:42:18] <edheldil> well, I still haven't played toee :)
[08:42:21] <Avenger> layers are already supported even by the IE
[08:42:51] <rocket_hamster> IE uses alpha or colorkeys?
[08:42:57] <Avenger> still all orientations and stances need to be pre-rendered
[08:43:03] <edheldil> I know, but with more combinations it gets awkward, I think
[08:43:05] <Avenger> IE uses palette locations
[08:43:16] <Avenger> for coloring
[08:43:46] <edheldil> rocket_hamster: color keys
[08:43:53] <rocket_hamster> anyway I find it hard to back up project without any ingame footage
[08:45:45] <edheldil> fair enough, but it's going to be isometric and I believe the authors wouldn't release anything inferior
[08:47:06] <rocket_hamster> I hope so for their own sake :)
[08:47:31] <Avenger> i didn't see enough proof for the "isometric" i really hope for their own sake, yeah
[08:48:21] <Avenger> there will be a huge uproar if it suddenly reverts to some nwn style stuff
[08:49:06] <rocket_hamster> for the amount of money they should be able to do 2 games
[08:49:08] <rocket_hamster> :D
[08:49:12] <rocket_hamster> one isometric and one 3d
[08:49:23] <Avenger> lol 2M is not THAT much
[08:49:48] <rocket_hamster> when you consider its just 2M before anything happened at all
[08:49:49] <Avenger> artists kinda suck the money up
[08:50:01] <rocket_hamster> 2M for nothing basicaly
[08:50:11] <rocket_hamster> 2M for 5 guys making a video with one fanart
[08:50:15] <rocket_hamster> before any real sales
[08:50:25] <rocket_hamster> iwithout investor or publisher
[08:50:26] <Avenger> but that is their funding
[08:50:28] <rocket_hamster> is a amaizing start
[08:50:42] <Avenger> when i say not much, i meant funding, not income
[08:50:45] <Avenger> obviously
[08:51:03] <rocket_hamster> its funding from independent ppl who cant really force them to do anything
[08:51:11] <rocket_hamster> like publisher would
[08:51:16] <Avenger> on ohloh, gemrb is priced at 2M without any art :D mere programming hours
[08:51:34] <Avenger> a complete game takes much more
[08:52:09] <rocket_hamster> im aware of that, im just saying they have 2M before even one line of code was written from people who cant force them to deliver anything
[08:52:27] <fuzzie> yes, my worry would be that they suddenly realise they don't have the resources to do a nice isometric game
[08:52:41] <Avenger> i agree with fuzzie
[08:52:43] <rocket_hamster> or they get a publisher with other 2M and then it screwes up
[08:52:54] <Avenger> that is a possibility too
[08:53:12] <wjp> it'll be very interesting to see how these high-profile multi-million kickstarter games turn out :-)
[08:53:18] <fuzzie> well, the more resources of your own that you have, the more negotiating power you have with publishers
[08:53:47] <rocket_hamster> wjp it really will, with some of the projects im really in doubt
[08:53:47] <fuzzie> since you can get a lot further on your own, and the risk to publishers is hugely lower
[08:53:49] <Avenger> well, we'll see. So far i believe more in a bg3 than in project eternity
[08:54:04] <rocket_hamster> bg3? from whom?
[08:54:15] <Avenger> overhaul games
[08:54:23] <rocket_hamster> enhanced edition?
[08:54:48] <Avenger> their plan doesn't stop at bg1/2 ee
[08:55:12] <rocket_hamster> overhaul wont do more then you guys already did
[08:55:13] <wjp> I wonder how large a percentage of the total sales these 60K+ copies will be
[08:55:15] <rocket_hamster> + they add art
[08:55:47] <Avenger> +they add art :D
[08:55:52] <fuzzie> wjp: I would think quite likely only a fraction. Also reducing the risk, since the fanbase is already established as paying for things.
[08:56:00] <rocket_hamster> they have whole coded handed to them just add some fixed new additions and art
[08:56:53] <Avenger> well... i cannot talk about what they added already, but there are significant parts gemrb doesn't have currently
[08:57:07] <rocket_hamster> well networking
[08:57:19] <Avenger> no, more like in the gui department
[08:57:57] <fuzzie> right, but they *did* start with a complete codebase :)
[08:58:19] <rocket_hamster> and its probably in C, im sorry no fancy clases and encapsulation :D
[08:58:21] <fuzzie> but, proving that you can add new features and do ports is an important step to 'proving' yourself as a game developer
[08:58:26] <rocket_hamster> hardcore malloc for them!!
[08:58:27] <rocket_hamster> :D
[08:58:28] <fuzzie> no, it is all fancy C++ classes :P
[08:58:31] <Avenger> well, i don't care about the length of the way they had to make, but i care about the goal
[08:58:34] <rocket_hamster> damn!
[08:59:12] <fuzzie> who would you trust more, a random developer wanting to do bg3, or a developer who'd already made huge improvements to bg1/bg2 codebases and connected with the bg community?
[08:59:28] <Avenger> yeah
[08:59:47] <rocket_hamster> mind you to add that random guy showed one fan art and no game footage at all
[09:00:18] <Avenger> well, to the defense of Obsidian, they are not entirely new to the isometric rpg business
[09:01:52] <rocket_hamster> btw any of you guys are fans of space sim perhaps?
[09:02:03] <rocket_hamster> like freelancer/ wing commander?
[09:02:14] <Avenger> no, but i like Orion
[09:02:35] <Avenger> that is 4x, not sim, i don't like twitch games
[09:02:39] <rocket_hamster> i like Moo2 too
[09:03:04] <Avenger> freeorion is almost there
[09:03:11] <rocket_hamster> i dont like it :(
[09:03:28] <Avenger> it is developing slowly, but it is highly moddable
[09:03:29] <rocket_hamster> I just cant find taste for it, the interface etc
[09:13:29] <edheldil> so Eternity is a gamble, but at least I have a nice desktop wallpaper ;-)
[09:14:06] <edheldil> (sorry, I had to afk before completing the prev. sentence)
[09:22:45] <-- Avenger has left IRC ()
[09:44:41] <wjp> it's indeed a gamble, but I'm interested enough to want to see what they make of it
[09:52:28] <lynxlynxlynx> for fans of oldschool rpgs there's another kickstarter
[09:52:43] <lynxlynxlynx> and it has "make 2 games" as a stretch goal
[10:23:59] --> lynxlynxlynx_ has joined #gemrb
[10:24:17] <-- lynxlynxlynx has left IRC (Ping timeout: 246 seconds)
[10:39:46] --> DocMcCoy has joined #gemrb
[10:41:13] <-- fuzzie has left IRC (*.net *.split)
[10:41:13] <-- rocket_hamster has left IRC (*.net *.split)
[10:41:13] <-- tomprince has left IRC (*.net *.split)
[10:41:14] <-- jonask has left IRC (*.net *.split)
[10:41:14] <-- gembot has left IRC (*.net *.split)
[10:41:14] <-- Drakkar has left IRC (*.net *.split)
[10:41:14] <-- DrMcCoy has left IRC (*.net *.split)
[10:41:14] <-- kida_laptop has left IRC (*.net *.split)
[10:41:14] <-- |Blaze| has left IRC (*.net *.split)
[10:41:14] <-- duckpunc1 has left IRC (*.net *.split)
[10:42:50] --- DocMcCoy is now known as DrMcCoy
[10:45:08] --> rocket_hamster has joined #gemrb
[10:45:08] --> kida_laptop has joined #gemrb
[10:45:08] --> Drakkar has joined #gemrb
[10:45:08] --> duckpunc1 has joined #gemrb
[10:45:08] --> |Blaze| has joined #gemrb
[10:45:08] --> fuzzie has joined #gemrb
[10:45:08] --> tomprince has joined #gemrb
[10:45:08] --> jonask has joined #gemrb
[10:45:08] --> gembot has joined #gemrb
[13:29:01] <-- DrMcCoy has left IRC (Changing host)
[13:29:02] --> DrMcCoy has joined #gemrb
[13:32:47] --> WingedHussar has joined #gemrb
[13:37:19] <-- rocket_hamster has left IRC (Ping timeout: 246 seconds)
[14:06:23] --> avenger has joined #gemrb
[14:06:23] --- ChanServ gives channel operator status to avenger
[14:16:26] <lynxlynxlynx_> heh
[14:16:35] <lynxlynxlynx_> debian has kgb, a cia alternative
[14:18:22] <edheldil> they should upgrade to FSB, but Debian is, as always, hopelessly outdated ;-)
[14:20:50] <lynxlynxlynx_> :)
[14:21:20] <lynxlynxlynx_> esr also wrote one, but i haven't found if it can be used without running the daemon yet
[14:21:26] <lynxlynxlynx_> http://www.catb.org/esr/irker/
[14:30:50] <avenger> Lynx is it a public alternative?
[14:31:15] <lynxlynxlynx_> both are, but it seems nobody is running esr's bot yet
[14:31:22] <lynxlynxlynx_> we could use debian's directly
[14:31:29] <avenger> yes
[14:31:35] <lynxlynxlynx_> i'll write to the and we'll have a test
[14:31:44] <avenger> k
[14:33:36] <wjp> hm, I forgot (again) how much outgoing connectivity SF's git hooks can have
[14:34:38] <-- WingedHussar has left IRC (Ping timeout: 246 seconds)
[14:36:19] <fuzzie> in the worst case we can just have something pull every 5 minutes
[14:37:25] <wjp> we used email for cia I think?
[14:38:13] --> rocket_hamster has joined #gemrb
[14:41:14] <lynxlynxlynx_> yes
[14:41:29] <lynxlynxlynx_> we have the xml-rpc one almost set up too iirc
[14:41:56] <lynxlynxlynx_> kgb needs a local client and it's all perl with lots of deps, so it's not looking so good anymore
[14:42:08] <lynxlynxlynx_> irker is decentralised, so it's even worse
[14:42:56] <lynxlynxlynx_> maybe it would be easier to hack something up for gembot
[14:43:11] <lynxlynxlynx_> or exultbot, since it has better uptime
[14:45:40] <wjp> unless things changed in the last year, xml-rpc won't get through SF's firewall, IIRC
[14:46:06] <wjp> but it would certainly be possible to hack things up for bots too...
[14:46:38] <lynxlynxlynx_> is exultbot based on something or is it completely yours?
[14:47:28] <wjp> it's completely mine
[14:47:43] <wjp> it's a very basic perl script
[14:47:48] <wjp> (written 10 years ago or so)
[14:48:17] <lynxlynxlynx_> ouch :)
[14:48:34] <wjp> 13 years minus 2 months
[14:50:47] <fuzzie> we can always find a place to run bot code if necessary though
[14:51:40] <wjp> sure
[14:52:25] <edheldil> hehe, I have my perl infobot written about the same time
[14:53:21] <edheldil> although my one was piggybacking on xchat, iirc
[14:55:25] <lynxlynxlynx_> it's less maintenance if we manage to do it without that
[14:55:44] <lynxlynxlynx_> cia is maybe comming back eventuallly, but the site has more bad news
[14:56:30] <lynxlynxlynx_> there's also the option to move to github and use theirs
[14:58:57] <wjp> github's is... annoying
[14:59:09] <lynxlynxlynx_> how so?
[14:59:32] <wjp> (although nothing that some irssi commands can't fix)
[14:59:38] <wjp> it joins, reports, and parts
[15:00:27] <lynxlynxlynx_> true, but why is that annoying?
[15:01:08] <wjp> I'm not sure, but emperically it is :-)
[15:01:34] <wjp> but it's not a real problem of course
[15:04:28] <rocket_hamster> imo github is more user appealing
[15:04:54] <rocket_hamster> btw im learning c++ what do you think about putting classes and methods in one file for one class?
[15:11:30] <edheldil> i.e. merging *.cpp and *.h to ... *.h?
[15:15:08] <rocket_hamster> yes
[15:15:16] <rocket_hamster> class and its methods in one file
[15:16:16] <rocket_hamster> if you dont do that you have to look in two files at same time, or to specs, highly impractical if there is not workaround
[15:16:43] <lynxlynxlynx_> think what the preprocessor actually does with the includes
[15:16:56] <wjp> a more general programming or C++ channel would be a more appropriate place for this
[15:17:00] <lynxlynxlynx_> and your editor sucks if you can't handle 2 files
[15:17:44] <rocket_hamster> its not about editor, im working on some class and i have to look into two files
[15:17:48] <lynxlynxlynx_> i noticed sf also has some forking capabilites now, but that's justa start
[15:24:27] <lynxlynxlynx_> can't find any other alternatives
[15:24:35] <lynxlynxlynx_> short names suck
[15:30:48] <-- rocket_hamster has left IRC (Ping timeout: 240 seconds)
[15:35:29] <edheldil> btw, speaking of SF, looks like we will host SF mirror soon :)
[15:39:30] --> rocket_hamster has joined #gemrb
[15:41:20] <lynxlynxlynx_> you mean at work, right?
[15:42:50] <edheldil> yes
[15:44:44] <lynxlynxlynx_> cool
[16:07:48] <-- rocket_hamster has left IRC (Ping timeout: 248 seconds)
[16:51:29] --> rocket_hamster has joined #gemrb
[17:02:47] <-- rocket_hamster has left IRC ()
[18:02:31] <avenger> lynx there is a bug in todo: failing to learn from scrolls also gives xp
[18:02:35] <avenger> is that still there?
[18:02:49] <avenger> low int isn't even checked
[18:03:01] <avenger> because LS_STATS is never set
[18:03:06] <lynxlynxlynx_> likely
[18:03:36] <lynxlynxlynx_> but these two things are not related
[18:03:47] <avenger> but how do you make them fail :D
[18:03:54] <avenger> too many spells in the book?
[18:07:26] <lynxlynxlynx_> that or more commonly just the probability check
[18:07:35] <lynxlynxlynx_> chance to learn is defined by int
[18:07:52] <lynxlynxlynx_> even at 18 you still have a few percent chance of failure
[18:08:06] <avenger> but as i said, ls_learn is never set
[18:08:11] <avenger> i mean ls_stats
[18:08:21] <lynxlynxlynx_> unless, in some games, the difficulty is low enough
[18:08:40] <lynxlynxlynx_> ls_stats is probably for the minimum int requirement
[18:08:44] <avenger> i changed that part to always return false, and it isn't even going there
[18:09:02] <lynxlynxlynx_> ... in a spell level sense
[18:09:17] <avenger> no, it is the chance to learn roll based on int
[18:09:51] <lynxlynxlynx_> then you're right and it is buggy
[18:10:04] <avenger> though, i don't even see where LS_ADDXP is set
[18:10:31] <avenger> but i definitely get xp for learning
[18:10:38] <lynxlynxlynx_> in the data?
[18:10:55] <lynxlynxlynx_> we have a separate flag to disable it
[18:11:37] <avenger> well, as i said i don't see any place where it is set
[18:11:46] <avenger> having a separate flag to disable it wouldn't help :D
[18:11:56] <lynxlynxlynx_> we already have it :P
[18:13:40] <avenger> ???
[18:13:53] <lynxlynxlynx_> LS_NOXP
[18:14:07] <avenger> and how that helps?
[18:14:11] <lynxlynxlynx_> which is just further proof that it could come from the data originally
[18:14:14] <lynxlynxlynx_> it doesn't
[18:14:22] <lynxlynxlynx_> all the learnspell callers look fine too
[18:15:03] <avenger> ok i know
[18:15:18] <avenger> the scrolls learnspell has 1 as Parameter1
[18:15:58] <avenger> but the scrolls don't contain LS_STATS bit
[18:16:09] <avenger> so, we'll never check int
[18:18:49] <lynxlynxlynx_> learning from scrolls is the only ocassion where it needs to happen
[18:19:02] <lynxlynxlynx_> shouldn't we just change fx_learn_spell then?
[18:19:29] <avenger> yes, probably i will add a ^LS_STATS to it
[18:19:41] <avenger> so one could still hack their fx
[18:20:07] <avenger> it should be checked only at certain difficulty too, i guess?
[18:22:31] <lynxlynxlynx_> core and above
[18:22:39] <lynxlynxlynx_> so the bottom two skip it
[18:22:41] <lynxlynxlynx_> bbl
[18:23:13] <avenger> >DIFF_EASY is fine, i guess
[18:23:23] <avenger> or >DIFF_NORMAL ?
[18:23:33] <avenger> hmm i guess the latter
[18:23:51] <wjp> bg2 says spells are always learned on normal
[18:24:14] <avenger> ok
[18:25:31] <avenger> no strref printed
[18:25:37] <avenger> but no xp gained either
[18:25:39] <avenger> so it is fine
[18:25:47] <avenger> the bug in the todo doesn't exist
[18:26:13] <avenger> also if in pause, the script runs on exception
[18:26:38] <avenger> i guess learnspell should unpause the game
[18:26:53] <avenger> i can't find any other way for this to work
[18:33:43] <avenger> anyone knows what strref is printed when the int check fails?
[18:36:14] <avenger> probably 'You failed to copy the spell to your mage book.'
[18:43:49] --> rocket_hamster has joined #gemrb
[18:50:06] <avenger> ok, cia now would notify a commit
[18:55:23] --> Yoshimo has joined #gemrb
[18:58:32] <wjp> it would?
[18:59:13] <avenger> well, i think i did... or ... it didn't pass though?
[19:00:19] <avenger> meh, i swear i did a checkout
[19:00:33] <avenger> i miss cia :(
[19:01:38] <avenger> ok now it did
[19:02:36] <avenger> is there any way to see the paused game state from script?
[19:02:50] <avenger> i want to unpause only if it is paused
[19:03:02] <avenger> and i want to pause it again, if it was paused
[19:04:18] --> brada has joined #gemrb
[19:05:32] <brada> avenger: unless you want to manipulate the flag directly you would get paused/unpaused messages
[19:06:16] <brada> what is the behavior in the original?
[19:06:21] <brada> jsut curious
[19:06:59] <avenger> i don't know how it works. i think it doesn't really care about paused state, though
[19:07:19] <avenger> but we apply the effects only in unpaused state
[19:08:00] <avenger> or rather... we 'use items' only in unpaused state, and the scroll is learned via use item
[19:08:17] <avenger> i think it is more hardcoded in the original
[19:21:26] <brada> seems kinda hackish to unpase then repause
[19:21:41] <-- Yoshimo has left IRC (Ping timeout: 245 seconds)
[19:22:39] <brada> no way to have an argumet to apply effects regardless of pause state?
[19:42:46] <avenger> yeah, i guess it is hackish both ways
[19:43:35] <avenger> i didn't have to do much any hack in engine level this way
[19:44:07] <avenger> i don't know how much it would take to implement the equivalent of use item which is ignorant of the paused state
[19:44:23] <avenger> so, i went the easy (evil) way :)
[19:47:53] <avenger> it is not a simple apply effect, it also decreases the item charge
[19:48:16] <avenger> and probably does other stuff i don't even remember now
[19:49:14] <avenger> some of that may also want some unpaused time. like game script triggers on use item, or whatever
[19:49:44] <fuzzie> that is quite evil :)
[19:55:11] <avenger> well, i fixed a bug (and its friends) from the todo list :D
[19:55:41] <avenger> that can't be evil
[19:59:02] <fuzzie> I just wait for the next bug report, "I pause the game, read a scroll, and the protagonist dies".
[19:59:25] <fuzzie> actually I will be very impressed if anyone manages that.
[20:04:50] <brada> btw we have named constants for the pause state :)
[20:05:45] <brada> forgot we had the quiet flag on there
[20:05:56] <brada> probably because somebody added that after my initial work :p
[20:08:48] <avenger> fuzzie UpdateActorState is the right place to implement confusion?
[20:09:33] <avenger> if state&CONFUSION --> issue random actions like randomwalk/attack nearest/nothing
[20:16:36] <fuzzie> sounds like a good place to put it for now, anyway
[20:17:00] <fuzzie> I don't have anything here to check it, packed in boxes.
[20:24:34] <brada> should have at leased marked that as a hack :)
[20:25:32] <brada> im not as against hacks as others here *coughtomprincecough*, but its nice to document when you hack something and how it really ought to be implemented
[20:26:11] <avenger> well randomwalk works, but why i can select a confused actor O_o
[20:26:28] <avenger> someone changed selection for sure :D
[20:27:23] <brada> that sounds vaguely familiar...
[20:28:00] <avenger> yeah, i won't fix that now
[20:28:14] <avenger> i'm sure those who are responsible heard me :P
[20:29:49] <brada> did anybody ever fix the hold animation position?
[20:32:39] <avenger> i thought i did
[20:32:53] <avenger> but the confusion birds are definitely in the wrong position
[20:34:59] <avenger> ok, confusion committed
[20:36:26] <fuzzie> sf is so slooow.
[20:36:45] <fuzzie> but, nice
[20:37:46] <avenger> the attack doesn't work, i don't know why
[20:37:56] <avenger> [0] should be nearest anyone but not me
[20:38:15] <avenger> at least, i thought so
[20:39:04] <fuzzie> do we parse that properly?
[20:39:12] <avenger> no idea :)
[20:39:20] <avenger> that's a fuzzie part, hehe
[20:39:44] <fuzzie> we might well only support symbolic identifiers
[20:40:04] <fuzzie> (i.e. [ANYONE])
[20:40:06] <avenger> the symbols are resolved when you load the script
[20:40:31] <avenger> and ANYONE is not even a valid symbol, it is what 0 is called by weidu
[20:40:39] <fuzzie> oh well, sigh :P
[20:40:52] <fuzzie> so what did they use in their script compiler then?
[20:41:11] <avenger> well, i don't know. nowadays [anyone] is allowed
[20:41:17] <avenger> only dltcep chokes on it :D
[20:41:30] <avenger> [0] is fine
[20:41:40] <avenger> and should be fine by our parsing too
[20:41:49] <fuzzie> right, but *is* it? :)
[20:41:58] <avenger> yeah
[20:41:59] <avenger> :D
[20:42:06] <avenger> i mean, is it
[20:42:20] <fuzzie> bioware docs say, use ANYONE
[20:42:47] <fuzzie> which is interesting because it really isn't in ea.ids
[20:42:49] <avenger> yeah, i guess you could use WHATEVER too
[20:43:07] <fuzzie> not hardcoded into their compiler either
[20:43:12] <avenger> or RAINBOW_PONY
[20:43:28] <avenger> i guess unresolved symbols became 0
[20:43:28] <fuzzie> right, that sounds the most likely explanation :/
[20:43:37] <fuzzie> ugh, their compiler is all MFC, too
[20:44:10] <avenger> what did you expect
[20:44:29] <avenger> especially in '98
[20:45:37] <fuzzie> anyway, I don't see any reason it wouldn't work, other than parsing fail
[20:50:04] <avenger> yes, if the [0] would be stored properly, then DoObjectIDSCheck would make it match any object
[20:50:19] <avenger> i see only one problem, if it matches self
[20:51:32] <avenger> ahh that is defended against earlier
[20:51:45] <avenger> we got if (ac == Sender) continue;
[20:53:45] <avenger> hmm there is some return if no filters were applied
[20:54:12] <avenger> i guess that's wrong?
[20:55:31] <fuzzie> don't think so
[20:56:02] <fuzzie> [0] results in oC == NULL results in the filter check never being reached
[20:57:17] <fuzzie> so the question is, why does GetActorFromObject return NULL so early? that shouldn't be checked until after the tgts stuff
[20:57:50] <fuzzie> probably moving that will fix it, then
[20:57:54] <brada> did you mean to remove that state variable? because it is used in the check on the next line....
[20:58:07] <fuzzie> it's not removed
[20:58:10] <fuzzie> it's just moved
[20:58:14] <brada> ah
[20:58:15] <brada> yup
[20:58:20] <brada> ignore me!!!
[20:59:00] <avenger> if oC is null - (becomes null even from [0]) then getactorfromobject returns null immediately, that's your problem?
[20:59:08] <fuzzie> sorry, I mean: move the oC==NULL check in GetActorFromObject to after the 'if(tgts)' block
[20:59:14] <fuzzie> yes
[20:59:48] <avenger> but will GetAllObject not crash on a null?
[20:59:51] <fuzzie> nope
[20:59:59] <fuzzie> it does the right thing
[21:00:07] <avenger> cool
[21:00:20] <avenger> i still think that my finding will also cause problem
[21:00:31] <-- edheldil has left IRC (Remote host closed the connection)
[21:03:38] <fuzzie> which one?
[21:04:06] <fuzzie> the filter check is good, I'm pretty sure, I went through this before
[21:05:18] <fuzzie> rght, I also noticed this NULL return at the time but didn't fix it, I see in log..
[21:05:25] <fuzzie> does it work if you move it?
[21:05:31] <avenger> in EvaluateObject
[21:05:45] <avenger> there is a return null if no filtering was used
[21:05:55] <fuzzie> EvaluateObject is never called
[21:06:02] <fuzzie> because oC is NULL
[21:06:06] <fuzzie> in this case
[21:06:15] <avenger> ah ok
[21:06:17] <avenger> ok ok
[21:06:18] <fuzzie> I hope!
[21:06:27] <avenger> yeah, it would crash earlier
[21:06:29] <fuzzie> this code just sort of got added to and added to
[21:06:40] <avenger> yep, it is evolutionary
[21:07:08] <avenger> anyway, i did your change, and still no attacks, only randomwalk
[21:09:56] <fuzzie> hmph
[21:10:28] <fuzzie> you put the confused stuff above the stupid round attack hack
[21:10:39] <fuzzie> but it doesn't seem like that should be a problem
[21:10:51] <fuzzie> so does GenerateAction succeed?
[21:11:15] <avenger> not sure
[21:11:24] <avenger> i'm trying to chck that now
[21:12:31] <avenger> yes, the action is created
[21:12:41] <avenger> but it could be that it switches too fast
[21:12:50] <avenger> i will slow the switches down
[21:17:49] <lynxlynxlynx_> re confusion: confused actors aren't selectable, that's as expected
[21:17:58] <lynxlynxlynx_> it's similar to panic in that regard
[21:19:43] <brada> yes i think the problem is that he is somehow able to select confused actors
[21:22:36] <avenger> lol fuzzie, actually it is not a null target
[21:22:47] <avenger> i just did an action->dump()
[21:23:13] <avenger> second object is 2. IDS Targeting: 0 0 0 0 0 0 0 0 0 0
[21:23:13] <avenger> Filters: 0 0 0 0 0
[21:23:24] <fuzzie> ah, so, buggy parsing :P
[21:23:27] <avenger> so, i guess, i was wrong, still i'm right
[21:23:29] <fuzzie> ah damn, maybe that's what I fixed
[21:24:02] <avenger> you made [0] into null?
[21:24:06] <fuzzie> hm, no
[21:25:44] <avenger> ok, so what happens if i remove that !filtered return from EvaluateObject. Suddenly everything will be buggy?
[21:27:47] <lynxlynxlynx_> btw, don't you also have a problem due to switching actions all the time? UpdateActorState doesn't run just each round
[21:28:52] <avenger> it should run rarely
[21:29:10] <avenger> of course, it would help if confused actors couldn't be given action via gui :D
[21:29:18] <avenger> someone could fix selection
[21:29:28] <avenger> while i kill the other bugs
[21:29:31] <lynxlynxlynx_> do you have any uncommitted stuff in actor.cpp?
[21:29:36] <avenger> yes
[21:29:41] <fuzzie> removing the check is possibly harmless, but I'm really not sure
[21:30:09] <lynxlynxlynx_> in UpdateActorState or elsewhere?
[21:30:09] <avenger> well, it still didn't fix attack, meeh
[21:30:43] <avenger> heeeh
[21:30:49] <avenger> aerie hits Anomen
[21:30:50] <avenger> woo
[21:32:03] <avenger> cool
[21:32:20] <avenger> i thought i will see xxx attacks yyy
[21:32:24] <rocket_hamster> http://fpaste.org/n3wk/
[21:32:26] <rocket_hamster> ^^
[21:32:26] <avenger> but i didn't see that
[21:33:33] <lynxlynxlynx_> rocket_hamster: sounds familiar, but i don't know if i'll remember the troll
[21:33:57] <lynxlynxlynx_> keywords ftw
[21:34:43] <avenger> ok commit done
[21:35:44] <rocket_hamster> http://artlung.com/smorgasborg/Invention_of_Cplusplus.shtml
[21:35:48] <rocket_hamster> this one is gold
[21:35:50] <rocket_hamster> :)
[21:36:29] <lynxlynxlynx_> http://www.27bslash6.com/index.html <-- thought i read it here, but can't find it
[21:39:13] <rocket_hamster> lol lynx!!!
[21:39:15] <rocket_hamster> good find
[21:40:00] <avenger> lol that's fun
[21:40:31] <lynxlynxlynx_> where did you pull 0x3f from?
[21:40:42] <lynxlynxlynx_> i recommend missing missy
[21:41:01] <lynxlynxlynx_> that's how i found it again :)
[21:41:10] <avenger> lynx: it is just an arbitrary number of course it must be a power of 2 - 1 value
[21:41:42] <lynxlynxlynx_> why not align it to round borders?
[21:42:18] <lynxlynxlynx_> i was just about to consolidate that (no need to run detect each tick either)
[21:44:11] <avenger> this text feels like what i thought about c++ :D
[21:46:33] <rocket_hamster> exactly avenger
[21:46:41] <rocket_hamster> same with me, im beginning to hate it with passion
[21:48:53] <avenger> lol that was fun
[21:49:18] <avenger> if that's a real stuff, he is more negative about his language than I am
[21:50:20] <rocket_hamster> its fake
[21:50:36] <rocket_hamster> but funny and true
[22:04:10] <avenger> ehh, DayNight happened here
[22:04:21] <-- avenger has left IRC (Remote host closed the connection)
[22:24:13] <lynxlynxlynx_> fuzzie: do you know why confusion must be timed at 2^n-1?
[22:41:32] <-- rocket_hamster has left IRC ()
[22:44:06] <-- lynxlynxlynx_ has left IRC (Remote host closed the connection)
[22:52:07] <-- brada has left IRC (Quit: brada)
[22:52:55] --> brada has joined #gemrb
[22:56:58] <-- brada has left IRC (Client Quit)