#exult@irc.freenode.net logs for 23 Aug 2014 (GMT)

Archive Today Yesterday Tomorrow
Exult homepage


[00:13:42] --> Kirben_ has joined #exult
[00:13:43] --- ChanServ gives channel operator status to Kirben_
[00:17:13] <-- Kirben has left IRC (Ping timeout: 255 seconds)
[01:13:49] --> Kirben has joined #exult
[01:13:49] --- ChanServ gives channel operator status to Kirben
[01:17:50] <-- Kirben_ has left IRC (Ping timeout: 255 seconds)
[01:38:54] <-- Dominus has left IRC (Ping timeout: 246 seconds)
[01:39:26] --> Dominus has joined #exult
[01:39:26] --- ChanServ gives channel operator status to Dominus
[01:48:09] --> sh[4]rm4 has joined #exult
[01:50:32] <-- sh4rm4 has left IRC (Ping timeout: 260 seconds)
[02:55:40] --- sh[4]rm4 is now known as sh4rm4
[04:19:03] --> Rottingbeef has joined #exult
[04:27:20] <-- Rottingbeef has left IRC ()
[06:51:32] <-- Lightkey has left IRC (Ping timeout: 260 seconds)
[07:04:11] --> Lightkey has joined #exult
[12:58:45] <-- SHODAN has left IRC (Remote host closed the connection)
[13:00:43] --> SHODAN has joined #exult
[13:44:18] <-- Marzo has left IRC (Ping timeout: 250 seconds)
[13:45:02] <-- Kirben has left IRC ()
[18:35:11] --> ShamblerDK has joined #exult
[19:56:08] --> Firebrand9 has joined #exult
[19:56:41] <Firebrand9> is it a common problem for Iolo to get stuck while attempting to get on the magic carpet?
[19:58:05] <Firebrand9> actually, I just confirmed thsi does not happen in version 1.2
[20:07:14] <-- Firebrand9 has left IRC (Quit: ChatZilla 0.9.90.1 [Firefox 8.0.1/20111120135848])
[20:41:48] --> Marzo has joined #exult
[20:45:47] --> Firebrand9 has joined #exult
[20:46:20] <Firebrand9> marzo, I sent you and Dominus an email
[20:48:01] <Dominus> hi
[20:48:04] <Dominus> "but, being neck-deep in my own game project, freelance work, and about to launch a business, I can't really say I have the time to properly debug Exult myself"
[20:48:22] <Dominus> yeah, different reasons but the same applies mostly to us ourselves
[20:49:52] <Dominus> Firebrand9: you are on Windows, why not use the portable command line option -p to bypass this problem? this way everything is neatly packed in the Exult folder
[20:50:14] <Dominus> Also, I'd still say, don't bother with anything but actual snapshot
[20:50:43] <Dominus> everything else is less compared to the original *or* has game stopping bugs
[20:51:58] <Firebrand9> yeah, but Dominus, you guys oversee this project, so I'm going to you to address issues with it
[20:52:13] <Firebrand9> sadly the newest snapshot has some very obvious bugs, the first of which I discovered with the magic carpet
[20:52:43] <Dominus> can you please use the bug tracker for all of these obvious bugs?
[20:52:56] <Dominus> along with the cfg bug?
[20:53:14] <Firebrand9> happy to. Where can I do this?
[20:53:16] <Dominus> everything else will just result in the bugs getting forgotten.
[20:53:24] <Firebrand9> fair enough
[20:53:38] <Dominus> http://exult.sourceforge.net/faq.php#bug_report
[20:54:40] <Firebrand9> ok, when I have a free moment I'll detail these 2 issues there
[20:55:11] <Dominus> and the earlier comment about real life, I wanted to make clear why we are probably not going to solve this soon
[20:55:28] <Firebrand9> I understand this isn't a paid project
[20:55:28] <Dominus> I have no problems with the magic carpet, btw
[20:55:36] <Firebrand9> odd
[20:56:37] <Firebrand9> I'll have to see if I can grab a sceenshot out of it. Does Exult have the means to do this internally?
[20:57:02] <Dominus> yes, but I'd recommend another tool like IrfanView or so
[20:57:44] <Dominus> we use the really small unscaled picture and store it as pcx which oddly most modern graphic viewers have forgotten about
[20:58:03] <Firebrand9> let me see if my attempt with prntscn worked
[20:58:17] <Firebrand9> yes
[20:58:25] <Dominus> Firebrand9: when this happens is it happening with a brand new game started from the snapshot?
[20:58:46] <Firebrand9> no
[20:58:52] <Firebrand9> ok, so how can I submit this to you?
[20:59:11] <Firebrand9> email? How can I tie the bug report I plan to submit to this?
[20:59:28] <Firebrand9> I suppose I can put it up to Photobucket
[21:00:01] <Dominus> wait a bit. What savegame is it from then? With what version was this save done?
[21:00:42] <Firebrand9> I want to say 1.2, but it otherwise loads and plays properly. It just has issues with Iolo getting back on the carpet
[21:00:50] <Firebrand9> do you want the save file too?
[21:01:04] <Dominus> no, sorry.
[21:01:22] <Dominus> but this is another problem with mixing these versions
[21:01:40] <Dominus> you *will* run into other things like this
[21:02:33] <Dominus> so many things have been done to how Exult does things in the last decade since 1.2 that running such old saves is asking for problems
[21:02:36] <Firebrand9> how is this merging the behavior of the characters?
[21:03:14] <Firebrand9> the engine, again in theory, should just be taking character locations and then running them through new AI functions
[21:03:45] <Firebrand9> all the save should do is save item and character locations and states
[21:04:10] <Firebrand9> have new states been added?
[21:04:23] <Firebrand9> that's the only way I can see this happening
[21:05:20] <Firebrand9> if the switch call for the behavior in the FSM for the AI has some new case's, or has removed old case's, then this could happen
[21:07:02] <Firebrand9> but I'd imagine at this point U7 has a well-known set of possible states the characters can be in
[21:07:20] <Dominus> I have no idea what all changed since 1.2 but it's been 10 years and Marzo has found so many things and corrected them, that this just seems to happen
[21:07:35] <Dominus> again, using 1.2 saves is not safe for a good playthrough
[21:08:02] <Dominus> and bugs that creep up when using these old saves but are not there in new saves are just nothing we can fix
[21:11:57] <Dominus> Firebrand9: http://exult.sourceforge.net/faq.php#old_savegames we mention this in the FAQ but it's not standing out right away
[21:12:23] <Firebrand9> well, obviously it's interpreting the data and jamming up some where. It sounds like you're saying that the priority on such bugs is low. But, it's code and data. A discrete answer for why is there.
[21:12:24] <Dominus> aehm, actually http://exult.sourceforge.net/faq.php#saves_compatibility
[21:12:31] * Firebrand9 reads
[21:12:52] <Firebrand9> hmm
[21:13:02] <Dominus> I'm saying the priority of such bugs is 0
[21:13:17] * Firebrand9 laughs
[21:13:21] <Firebrand9> ok...
[21:13:35] <Firebrand9> due to aforementioned disclaimer
[21:14:16] <Marzo> The issue here is that not only there have been changes in save game data since 1.2, but there have been some bugs that corrupted the save data in old versions that have since been fixed
[21:15:38] <Firebrand9> but oddly in v1.2, the carpet issue for instance, does not happen. So, that implies that it's not corruption in that particular case
[21:16:22] <Firebrand9> what you're saying is a broader fix, but the evidence doesn't point to that in the one case I have on-hand
[21:16:43] <Dominus> "changes in save game data"
[21:16:56] <Dominus> seriously Firebrand9 you are beating a dead horse
[21:17:08] <Dominus> and it's not going to be limited to this very bug
[21:17:41] <Marzo> The carpet has also seen some changes; it has been recently changed how the NPCs pathfind to chairs in a way that almost always makes it better than in 1.2
[21:17:54] <Dominus> yep
[21:18:05] <Marzo> But in some edge cases, I imagine it could end up failing
[21:18:15] <Firebrand9> how would the data end up affecting that though?
[21:18:23] <Firebrand9> Dominus I'm a programmer
[21:18:32] <Firebrand9> I *must* understand
[21:18:45] <Dominus> Firebrand9: try a new game, add as many party members as possible and get on the carpet. then repeat with 1.2...
[21:18:51] <Marzo> Possibly by corrupting a flag in Iolo's data, but this is speculation
[21:18:52] <Firebrand9> that's my plan
[21:18:54] <Dominus> Firebrand9: then dive in our code
[21:19:28] <Firebrand9> I'd rather discuss this with experts for starters. Diving in someone else's code is a much bigger proposal
[21:19:50] <Firebrand9> this I can do in a spare hour. That I can not
[21:20:10] <Dominus> I think you have gotten as much insight into this as can be squeezed out of us :)
[21:20:50] <Dominus> but please do file a bug on the -c command line thing.
[21:21:19] <Firebrand9> what Marzo says makes sense. If a flag or flag location has changed, Iolo could be set to a indeterminate state
[21:21:24] <Firebrand9> I will :)
[21:21:36] <Dominus> and please try whether this bug applies to the current snapshot as well (the -c bug)
[21:22:02] <Firebrand9> it most definitely applies to the current snapshot, unless that's been changed in the past week
[21:22:20] <Dominus> as for Iolo, did you try the teleport chat alt-t to reset things? I found that this often helps such things
[21:22:32] <Firebrand9> August 13 was when I originally found this issue
[21:22:43] <Firebrand9> hmm no I didn't
[21:22:48] <Firebrand9> let me test this
[21:23:03] <Dominus> nope, snapshot stayed the same, I was just unsure whether it happened with an old version only or with the current as well
[21:23:18] <Firebrand9> alt-t does this?
[21:23:25] <Dominus> I only glanced at the previous conversation
[21:23:30] <Firebrand9> sure
[21:24:30] <Dominus> yes, alt-t but cheats need to be enabled
[21:24:46] <Dominus> http://exult.sourceforge.net/docs.php#exult_cheats
[21:26:29] <Firebrand9> crap, I started a new game as per your suggestion and now that's the only one it sees. I have to do some file-wrangling to get my old save to load in the new version
[21:27:35] <Dominus> that's odd, if you could load the savegame previously it can only mean you are either not pointing at the same savegame folder or you were only sharing the gamedat folder
[21:28:09] <Firebrand9> I think this is related to the .cfg bug
[21:28:34] <Firebrand9> where it was pointing to default DIR's, but saving apparently alters where it's looking
[21:28:46] <Dominus> no it doesn't
[21:29:15] <Firebrand9> then why would it do this?
[21:29:50] <Dominus> I have no idea without looking at all your cfgs and knowing the location of your games and the savegame locations etc...
[21:30:11] <Firebrand9> it broke the 1.2 save DIR
[21:30:20] <Firebrand9> now that version can't find its saves
[21:30:42] <Firebrand9> ugh... ok, let me see what thsi is doing
[21:31:09] <Dominus>
[21:31:50] <Dominus> everything is in the cfgs :) they should tell you what is happening...
[21:33:42] <Firebrand9> it wasn't necessary actually. The problem is the 2 versions are in different DIR's. Since, the universal .cfg attempt didn't work, they have their own respective save DIR off from the DIR that exult.exe is in. So, I just had to copy the .sav files from one to the other
[21:36:48] <Firebrand9> ok, tried the alt-t cheat, which did teleport the party elsewhere, but Iolo is in some limbo state where he can no longer move
[21:36:59] <Firebrand9> meh whatever
[21:37:10] <Dominus> did you talk to him?
[21:37:18] <Firebrand9> I can make him leave the party. That does break him out of his trance
[21:37:35] <Firebrand9> yeah, I can still talk to him
[21:38:00] <Firebrand9> but he won't follow the party any longer after attempting to get back on the carpet
[21:38:48] <Firebrand9> fortunately I'm not far in the game, so starting over isn't a huge deal
[21:39:19] <Firebrand9> just start over and make a b-line for the carpet, and see how that goes
[21:39:51] <Dominus> yeah, that's what is working fine here
[21:40:14] <Dominus> good luck and more fune, less bugs in a new game :)
[21:40:51] <Dominus> (I'd seriously not consider playing through with anything older than current snapshot)
[21:41:04] <Firebrand9> I understand
[21:41:12] <Dominus> I *do* have the old versions around but only to do regression tests
[21:41:15] <Firebrand9> one odd last note though
[21:41:23] <Dominus> sure, shoot
[21:41:51] <Firebrand9> when I get on the carpet in 1.2 it plays the super-epic fast "swooshing" music
[21:42:09] <Firebrand9> the new version, for whatever reason, does not do this. Is this known?
[21:42:47] <Dominus> that sounds bug report worthy :)
[21:44:16] <Firebrand9> it could be another save game data issue
[21:44:32] <Firebrand9> perhaps the current song played is saved and the location changed
[21:44:42] <Firebrand9> location in the file*
[21:44:42] <Dominus> ah, I thought with a new game :)
[21:45:04] <Dominus> can't test, no headphones anywhere near...
[21:45:08] <Firebrand9> I'd have to try it to confirm. I'll get back to you if that's the case
[21:45:29] <Dominus> (and I need to keep quiet, everyone is asleep and I should be too)
[21:45:55] <Firebrand9> but, it just continues playing the Britannia music as if you were walking (I'm in Britannia if that wasn't already otherwise clear)
[21:48:10] <Dominus> bah, just tested anyway in lowest volume, tune changes
[21:48:39] <Firebrand9> so, another version incompatibility in other words
[21:48:59] <Dominus> probably
[21:49:15] * Firebrand9 nods
[21:50:40] <Dominus> good night, off to bed
[21:51:04] * Firebrand9 heads out through a black gate
[21:51:10] <-- Firebrand9 has left IRC (Quit: ChatZilla 0.9.90.1 [Firefox 8.0.1/20111120135848])
[21:59:31] --> sh[4]rm4 has joined #exult
[22:02:00] <-- sh4rm4 has left IRC (Ping timeout: 260 seconds)
[23:34:12] --> Kirben has joined #exult
[23:34:12] --- ChanServ gives channel operator status to Kirben