#gemrb@irc.freenode.net logs for 21 Oct 2012 (GMT)

Archive Today Yesterday Tomorrow
GemRB homepage


[08:07:31] --> lynxlynxlynx has joined #gemrb
[08:07:32] <-- lynxlynxlynx has left IRC (Changing host)
[08:07:32] --> lynxlynxlynx has joined #gemrb
[08:07:32] --- ChanServ gives channel operator status to lynxlynxlynx
[08:09:56] <-- |Cable| has left IRC (Ping timeout: 248 seconds)
[08:22:51] --> |Cable| has joined #gemrb
[08:55:40] --> rocket_hamster has joined #gemrb
[09:14:15] <-- rocket_hamster has left IRC ()
[09:15:21] --> rockethamster has joined #gemrb
[09:21:57] <lynxlynxlynx> rockethamster: today i have plenty of time if you want to play
[09:28:48] <rockethamster> lemme check
[09:28:57] <rockethamster> btw what do u think about neverwinter 1,2
[09:29:00] <rockethamster> and icwd?
[09:30:42] <rockethamster> bah cant
[09:30:50] <rockethamster> maybe later
[09:53:46] <lynxlynxlynx> i only played nwn1 and the first two addons
[09:54:08] <lynxlynxlynx> started mask of the betrayer but got sidetracked
[09:54:17] <lynxlynxlynx> iwds are good
[09:57:28] <rockethamster> like iwd2 better
[09:57:30] <rockethamster> more classes
[09:57:31] <rockethamster> and chars
[09:57:37] <rockethamster> also combat seems more fluid
[09:57:40] <rockethamster> and faster
[09:58:29] <rockethamster> maybe we can try icwd2 with hamachi, maybe it breaks through router
[09:58:31] <rockethamster> what do u think
[10:00:22] <rockethamster> hmm have a look at this
[10:00:23] <rockethamster> http://www.gameranger.com/download/
[10:09:39] <rockethamster> lynx
[10:10:05] <lynxlynxlynx> nah
[10:12:32] <rockethamster> will u push my fix perhaps?
[10:13:36] <-- rockethamster has left IRC (Quit: Leaving)
[10:15:24] --> rocket_hamster has joined #gemrb
[10:18:10] <lynxlynxlynx> what's the benefit besides one warning less and a regression for pst?
[10:20:14] <rocket_hamster> faster loop
[10:20:17] <rocket_hamster> and its more logical
[10:20:30] <rocket_hamster> how can u perform attack on something you dont see
[10:21:01] <rocket_hamster> way it works now is it deleted target and then checks for it, seems like a waste
[10:22:16] <rocket_hamster> also way it works now actors will still try to attack and position themself towards invisible moving actor
[10:22:32] <rocket_hamster> clearly this is not what is wanted
[10:24:22] <lynxlynxlynx> ok
[10:24:38] <lynxlynxlynx> you need to make a few adjustments though
[10:25:03] <lynxlynxlynx> you removed the distinction of pst invisibility state
[10:26:34] <lynxlynxlynx> to be safe, better make the original block an assert, since the target isn't passed on directly
[10:28:08] <lynxlynxlynx> you can use Actor::IsInvisibleTo for a more thorough check and also avoid having to redo the pst switch
[10:30:43] <lynxlynxlynx> for perfection, maybe also the stance needs to be reset (like in the next getcombatdetails call)
[10:32:44] <rocket_hamster> yes
[10:44:03] <rocket_hamster> something like this? http://fpaste.org/80Yg/
[10:51:03] <lynxlynxlynx> yeah
[10:54:54] <rocket_hamster> cool
[10:56:44] <rocket_hamster> will u have time starting hour from now?
[10:57:05] <rocket_hamster> current time +1 to current time +3
[10:58:39] <lynxlynxlynx> sure
[10:58:53] <lynxlynxlynx> finish the patch by then and it can go directly into the tree
[11:04:06] <rocket_hamster> anything else to add there?
[11:06:25] <lynxlynxlynx> see what i said and also compare it with your previous patch
[11:11:09] <lynxlynxlynx> attention to detail is critical in programming
[11:11:27] <rocket_hamster> well dead flag is missing in invisible
[11:24:48] <rocket_hamster> http://fpaste.org/Ng0Z/
[11:26:55] <lynxlynxlynx> better
[11:29:04] <rocket_hamster> as for assert im not sure how is it used properly
[11:29:17] <rocket_hamster> i know it terminates program when it evaluates 0
[11:29:26] <rocket_hamster> but i dont see why to use it
[11:32:58] <lynxlynxlynx> they are useful when you want to make sure the program is in the state you need it to be
[11:33:27] <lynxlynxlynx> since they cause catastrophic failure, they're used to prevent catastrophic code path combinations
[11:34:29] <lynxlynxlynx> we shouldn't be losing targets in performattack, so eventually also that warning afterwards will be a candidate for an assertion instead
[11:35:33] <lynxlynxlynx> if a dead or invisible target gets to that point that means that the check you're now adding became bugged, some other code was added that caused a regression or even worse, global id lookup got broken
[11:36:25] <rocket_hamster> now it should ReleaseAction and return from ActionCore
[11:36:37] <rocket_hamster> so i dont see how it could possibly get to perform attack
[11:36:48] <rocket_hamster> AttackCore*
[11:37:13] <lynxlynxlynx> *now*
[11:37:16] <rocket_hamster> :D
[11:37:19] <rocket_hamster> i see..
[11:37:41] <rocket_hamster> so the thing is to add assert to PerformAttack
[11:37:58] <rocket_hamster> assert(target) right?
[11:38:04] <lynxlynxlynx> no
[11:38:21] <lynxlynxlynx> what the block does now
[11:38:48] <lynxlynxlynx> assert(target) is too greedy atm, would need a lot of testing
[11:52:10] <rocket_hamster> ok shall we begin? rebooting to win
[11:52:59] <-- rocket_hamster has left IRC (Read error: No route to host)
[11:54:34] --> rockethamster has joined #gemrb
[12:45:14] <rockethamster> lol
[12:45:15] <rockethamster> rejoin
[12:45:25] <rockethamster> they pwned me so hard game bugged
[12:45:28] <lynxlynxlynx> yeah
[13:52:34] --> kida_laptop has joined #gemrb
[14:40:27] <-- kida_laptop has left IRC (Remote host closed the connection)
[14:44:39] <-- |Cable| has left IRC (Ping timeout: 260 seconds)
[14:50:14] --> kida_laptop has joined #gemrb
[14:56:26] --> |Cable| has joined #gemrb
[18:03:41] <-- |Cable| has left IRC (Ping timeout: 252 seconds)
[18:16:26] --> |Cable| has joined #gemrb
[18:51:38] <-- rockethamster has left IRC (Quit: Leaving)
[18:55:08] --> rocket_hamster has joined #gemrb
[20:05:10] <-- rocket_hamster has left IRC ()
[21:01:20] <-- lynxlynxlynx has left IRC (Remote host closed the connection)