#exult@irc.freenode.net logs for 5 Jan 2013 (GMT)

Archive Today Yesterday Tomorrow
Exult homepage


[02:52:07] --> Kirben has joined #exult
[02:52:07] --- ChanServ gives channel operator status to Kirben
[12:11:22] <-- Kirben has left IRC ()
[12:20:21] --> Marzo has joined #exult
[12:51:01] --> Dominus has joined #exult
[12:51:02] --- ChanServ gives channel operator status to Dominus
[14:36:04] <-- Dominus has left IRC (Read error: Connection reset by peer)
[14:36:35] --> Dominus has joined #exult
[14:36:35] --- ChanServ gives channel operator status to Dominus
[17:25:50] --> Malignant_Manor has joined #exult
[17:48:43] <Malignant_Manor> Marzo: Do you have any suggestions on how I should change Wait_for_arrival or create a new function to make this fix less hackish? http://pastebin.com/itrSiQsg
[17:48:58] <Marzo> I am doing a less hackish fix right now
[17:49:23] <Marzo> That does away with Wait_for_arrival entirely for this case
[17:49:45] <Malignant_Manor> In the original, it could be interrupted by moving.
[17:50:19] <Marzo> So can the way I am doing, but it is entirely coincidental :-)
[17:50:20] <Malignant_Manor> I found some situations, even in 320x200 needed about 7 seconds to pathfind to the bed.
[17:50:50] <Marzo> The way I am doing, you could take a day to pathfind to the bed and it would still work
[17:51:49] <Malignant_Manor> The z coordinate bug should probably wait until after release. I'm not sure if it would cause issues, and is there even more than one case?
[17:54:19] <Marzo> Hrm. Although stopping by walking seems to cause the avatar to not be able to move afterwards
[17:56:25] <Marzo> Hrm. Spark's gavel is unreachable at the moment.
[17:57:11] <Malignant_Manor> Lovely
[17:57:54] <Marzo> Ah, he gets unresponsive because canceling sleep schedule does not seem to reset schedule correctly
[17:58:02] <Malignant_Manor> Well, that should be fixed before release. There really isn't much else unless you want to do ranged combat. Of course, that could easily break things.
[18:16:19] <Malignant_Manor> After these fixes and whatever else you want, will we have a release candidate or just do testing without one?
[18:35:11] --> i30817 has joined #exult
[18:38:00] <i30817> about the sorting problem, marzo said it was a problem with the 0 case, and not braking the symmetry property right?
[18:38:02] <i30817> How about taking the mid-point between the objects and using that as a seed for a 'repeatable' random flip (so fir trees on the same plane don't all win out over the other kinds of trees or something).
[18:45:53] --> DominusExult has joined #exult
[18:45:54] --- ChanServ gives channel operator status to DominusExult
[18:46:32] <i30817> supposing the midway point is equal, whichever of the two objects being compared is first on the sort.
[18:49:17] <-- Dominus has left IRC (Ping timeout: 272 seconds)
[18:49:17] --- DominusExult is now known as Dominus
[18:51:15] <-- Malignant_Manor has left IRC (Quit: ChatZilla 0.9.89 [Firefox 17.0.1/20121128204232])
[18:51:50] --> Malignant_Manor has joined #exult
[18:56:50] <i30817> Mmmph. Wouldn't probably work without a symmetry property is 'repainting' a object causes it to be repainted over other objects, because there could be two contradictory overlaps.
[19:03:08] <-- i30817 has left IRC (Quit: ChatZilla 0.9.89 [Firefox 17.0.1/20121129151842])
[19:39:57] <-- Malignant_Manor has left IRC (Quit: ChatZilla 0.9.89 [Firefox 17.0.1/20121128204232])
[20:14:45] <Dominus> Marzo, I'm here now if you still need me (replying to your request a couple of days ago :))
[20:16:00] <Marzo> I wanted you to try something in your snapshots back then (so I wouldn't have to compile an older version to compare), but I found out I had an existing older version I could use
[20:16:21] <Dominus> ah...
[20:16:29] <Marzo> But in any case, I am finishing a few things regarding render order, and I would like to test them before committing
[20:16:50] <Marzo> But first, I have to fix beds
[20:17:05] <Marzo> And fix the render order of bedsheets versus player
[20:17:40] <Dominus> while you are at it, did you notice that beds are currently always made/set when an NPC gets up?
[20:17:54] <Marzo> I am adding an animation for it
[20:18:28] <Marzo> This at least should make it look better
[20:18:34] <Dominus> I don't know what the original did right now but it seems strange that when you wake NPCs via a spell the bed is instantly made
[20:18:50] <Marzo> Hence the animation :-)
[20:19:00] <Dominus> :)
[20:19:14] <Dominus> looking forward to this
[20:23:18] --> Malignant_Manor has joined #exult
[20:33:56] <Marzo> Dominus: can you check in your snapshots if bedrolls remain unrolled if you cancel sleeping (e.g., by pressing esc or selecting 0 hours)?
[20:34:31] <Dominus> yes
[20:34:36] <Dominus> am checking
[20:37:36] <Dominus> it has been remained unrolled since April 2010
[20:38:49] <Marzo> It is weird; the same usecode function works for when you sleep, but not for when you cancel
[20:38:55] <Dominus> (ouch another OpenGL bug, the Britain castle roofstones are looking very greenish with it... OpenGL scaler will probably always be bad)
[20:42:09] <Dominus> ha, and OpenGL doesn't honor Game Area setting in video gump and will always output auto...
[20:42:29] <Dominus> time for some more OpenGL bug reports and mode them deep blue :)
[20:42:57] <Dominus> Marzo, the cannons are still acting a bit weird
[20:43:04] <Marzo> How so?
[20:43:11] <Dominus> test point the double cannon in LB castle
[20:43:41] <Dominus> when you stand next to the west one and double click it, it will take the powder and cannon ball from the east one
[20:43:57] <Marzo> It is probably an usecode bug
[20:45:44] <Marzo> It is: the cannon usecode searches for cannonballs and powder kegs and uses the first it finds, whether or not they are the closest
[20:46:15] <Dominus> yes, can confirm the original has the same bug
[20:46:34] <Marzo> I just tested that too
[20:46:39] <Dominus> :)
[20:46:46] <Dominus> so disregard
[20:47:33] <sh4rm4> is there a section/bugtracker for usecode bugs/bugs in the original ?
[20:47:49] <Malignant_Manor> Keyring/SI fixes
[20:47:50] <Dominus> yes, docs/usecode_bugs.txt
[20:47:55] <Malignant_Manor> Forums
[20:48:04] <Malignant_Manor> or I guess feature requests
[20:49:22] <Dominus> (and that usecode_bugs.txt file is probably not up to date - I think the pikeman has been fixed somewhat *or* I just think it has been fixed because it is in SI fixes)
[20:56:41] <Marzo> I think it is the latter
[21:04:58] <Dominus> hmm, now that I added some more OpenGL bugs, I have the urge to sanitize our tracker...
[21:05:30] <Dominus> I *could* add another tracker for Exult Studio and another one for OpenGL....
[21:06:17] <Malignant_Manor> It would probably be more of a pain for users.
[21:06:51] <Dominus> yeah, an Exult Studio one could still make sense, though
[21:07:09] <Malignant_Manor> There's a category tab for Exult Studio.
[21:07:10] <Dominus> but that's just me trying to categorize stuff over and over again
[21:11:16] <Malignant_Manor> We should probably not build with OpenGL support in the release.
[21:11:30] <Dominus> that's true :)
[21:22:04] <Dominus> marzo, I *think* bed rolls used to work though
[21:26:39] <Marzo> Well, I can get them to roll up when you cancel, but the avatar fails to animate
[21:27:38] <Dominus> lazy do-gooder
[21:27:41] <Marzo> This is happening because the usecode uses a path_run_usecode before ending the sleep schedule, which fails to reach the target destination
[21:28:19] <Marzo> The check for path failure happens before the avatar is repositioned
[21:30:30] <Marzo> Be back later
[21:30:40] <Dominus> I see
[21:30:57] <Dominus> I'm probably gone then :)
[22:09:39] <-- Malignant_Manor has left IRC (Quit: ChatZilla 0.9.89 [Firefox 17.0.1/20121128204232])
[22:48:36] --> Matt_O has joined #exult
[22:57:00] --> Kirben has joined #exult
[22:57:00] --- ChanServ gives channel operator status to Kirben
[23:57:34] <-- Dominus has left IRC (Read error: Connection reset by peer)
[23:57:39] --> DominusExult has joined #exult
[23:57:40] --- ChanServ gives channel operator status to DominusExult