#exult@irc.freenode.net logs for 30 May 2008 (GMT)

Archive Today Yesterday Tomorrow
Exult homepage


[00:57:03] --> Kirben has joined #exult
[00:57:03] --- ChanServ gives channel operator status to Kirben
[01:56:28] --> petermdodge has joined #exult
[01:57:27] <-- petermdodge has left #exult ()
[03:00:50] <-- Kirben has left IRC (Read error: 110 (Connection timed out))
[03:14:53] --> Kirben has joined #exult
[03:14:53] --- ChanServ gives channel operator status to Kirben
[04:36:28] --> Colourless` has joined #Exult
[04:36:30] <-- Colourless has left IRC (Read error: 113 (No route to host))
[04:36:38] --- Colourless` is now known as Colourless
[04:36:40] --- ChanServ gives channel operator status to Colourless
[08:03:25] --> shazza has joined #exult
[12:14:18] --> servus_ has joined #exult
[12:28:38] <-- servus has left IRC (Read error: 110 (Connection timed out))
[14:09:24] <-- Kirben has left IRC (Read error: 110 (Connection timed out))
[14:13:32] --> servus has joined #exult
[14:28:49] <-- servus_ has left IRC (Read error: 110 (Connection timed out))
[15:01:40] --> Fingolfin has joined #exult
[15:01:40] --- ChanServ gives channel operator status to Fingolfin
[15:24:19] <-- Fingolfin has left IRC ()
[16:25:46] --> ettin_ has joined #exult
[16:31:04] <-- ettin has left IRC (Read error: 110 (Connection timed out))
[16:33:56] --> Fingolfin has joined #exult
[16:33:56] --- ChanServ gives channel operator status to Fingolfin
[16:48:23] <-- Fingolfin has left IRC ()
[17:34:36] --- ettin_ is now known as ettin
[21:16:09] --> Marzo has joined #exult
[21:18:58] <-- Marzo has left IRC (Client Quit)
[21:24:02] --> Marzo has joined #exult
[21:59:28] <wjp> hi
[21:59:37] <Marzo> Hi
[22:00:06] <wjp> can you check if my last change to Storm_effect makes sense?
[22:00:20] <Marzo> Is it on CVS?
[22:00:35] <Marzo> (on the public servers, I mean)
[22:00:38] <wjp> you removed the initialization of Storm_effect::start at some point, and I was wondering if that was an accident or if you were planning something different with it
[22:00:57] <wjp> I committed it 5 minutes ago
[22:01:15] <Marzo> 9 according to SF
[22:01:46] <wjp> uh, yes
[22:01:57] * wjp wonders which timestamp he was looking at
[22:02:05] <Marzo> (I am just saying that it is already on the public servers :-p)
[22:02:14] <wjp> k :-)
[22:03:05] <Marzo> It seems to make sense, yes
[22:03:40] <wjp> still tracking down some other uninitialized memory reports from valgrind
[22:03:54] <Marzo> I did some work on that recently
[22:04:09] <Marzo> (with valgrind too, even)
[22:04:20] <Marzo> Although the logs are huge
[22:04:20] <wjp> it's reporting uninitialized values in the framebuffer during some weather effects, strangely
[22:04:57] <wjp> (hm, or maybe it's in the palette)
[22:05:54] <Marzo> (is valgrind or ggc 4.2 which complain about initializing variables out of order?)
[22:06:45] <Marzo> To be completely consistent, I suppose effects.cc:1478 could be changed to 'start = false;', but that is just nitpicking :-p
[22:06:55] <wjp> yeah, already changed that here too :-)
[22:07:04] <Marzo> Hehe
[22:07:08] <wjp> (just didn't commit it yet because I also have a ton of debugging code in effects.cc now)
[22:07:23] <wjp> gcc does warn about out-of-order initialization
[22:07:36] <Marzo> (I wonder why I removed the initialization of that variable... must have been a mistake)
[22:07:51] <wjp> usually it's a harmless warning, but it can affect things when initializers use member vars themselves
[22:09:07] <Marzo> I know it is harmless; it is just that I remember seeing lots of those when compiling in gcc 4.2, and some of the variable initializations you added differed from those I had done in the order
[22:09:19] <Marzo> (I was striving to keep warnings to a minimum)
[22:09:45] <Marzo> But then I noticed the "deprecated use of string literal" warnings and gave up :-)
[22:10:20] <Marzo> (BTW, what are you testing in effects.cc that you have tons of debugging code?)
[22:10:36] <wjp> hm, I guess I did forget to check the order of initializers earlier this week
[22:11:28] <wjp> debugging: some valgrind client requests to try to track down the unitialized memory earlier than it is currently reported
[22:11:45] <wjp> s/unit/uninit/
[22:11:55] <Marzo> k
[22:12:25] <wjp> not much luck yet, though :/
[22:12:32] <Marzo> lol
[22:15:52] <wjp> so, svn?
[22:16:01] <Marzo> I'd say so
[22:16:29] <Marzo> (out of all that said something on the list, only Jeff hasn't manifested yet)
[22:16:39] <wjp> the current trend seems to already be moving past svn again, though :-)
[22:16:50] <Marzo> Yep
[22:17:06] <Marzo> I've seen some projects moving to Git, among others
[22:17:25] <Marzo> But as is, unless we move out of SF, we will have to use svn
[22:17:28] <Marzo> (or cvs)
[22:17:43] <wjp> I feel no strong need to use a distributed vcs at this point
[22:18:40] <Marzo> Neither do I
[22:29:45] <Marzo> Well, I am going home now
[22:29:55] <Marzo> Be back in ~40 min or so
[22:30:11] <wjp> ok, see you later
[22:30:18] <-- Marzo has left IRC ("Marzo vanishes suddenly.")
[23:41:51] --> Marzo has joined #exult
[23:43:24] <Marzo> back
[23:43:31] <wjp> wb
[23:44:11] <wjp> figured out the valgrind warnings from the framebuffer: apparently some missing flats in some chunks were visible
[23:44:31] <Marzo> Strange
[23:44:44] <wjp> (or had a partially translucent object on top of it or something)
[23:44:55] <wjp> I couldn't visually spot it, in any case
[23:44:58] <Marzo> So you are testing in SI?
[23:45:05] <wjp> BG
[23:45:20] <Marzo> Even stranger
[23:45:38] <wjp> how so?
[23:46:04] <Marzo> Because SI has some transparent, invisible roofs
[23:46:13] <Marzo> It could have explained it
[23:47:02] <Marzo> (and also because I don't remebers seeing any framebuffer issues in my own tests in BG with valgrind)
[23:47:46] <wjp> could be there's just a couple of places in the world where this shows up
[23:48:01] <Marzo> Possible
[23:48:02] <wjp> (I was seeing them in the swamp N of Trinsic)
[23:48:41] <Marzo> It *migth* be the bubbles
[23:48:50] <Marzo> (when in doubt, blamme the bubbles :-p)
[23:48:55] <Marzo> *blame
[23:49:31] <wjp> :-)
[23:50:07] <Marzo> While I am at it, let me see if I implement the game browsing menu in Exult
[23:50:31] <Marzo> It will be nice to workin on something different for a while
[23:53:21] --- Lord_Nightmare is now known as LordNAway
[23:57:52] <Marzo> (hm. Lots of familiar names in ScummVM SVN)
[23:58:43] <wjp> :-)
[23:59:23] <wjp> ugh, Path_walking_actor_action::handle_event can indirectly delete itself without noticing
[23:59:43] <Marzo> That is bad