#pentagram@irc.freenode.net logs for 6 Jul 2004 (GMT)

Archive Today Yesterday Tomorrow
Pentagram homepage

[00:31:14] --> Kirben has joined #pentagram
[00:31:14] --- ChanServ gives channel operator status to Kirben
[06:44:09] --> GingerNinja has joined #pentagram
[07:51:30] <-- Sectus has left IRC ("Leaving")
[07:59:46] --> Ember has joined #pentagram
[08:09:45] <-- Darke has left IRC (Read error: 110 (Connection timed out))
[08:39:01] <wjp> watt: there are some compliments for your web design on the exult forum
[09:18:31] <-- Ember has left IRC (Remote closed the connection)
[09:19:58] --- sbx|afk is now known as sbx
[09:21:20] <sbx> interesting conversation
[09:34:03] --> Darke has joined #pentagram
[09:34:04] --- ChanServ gives channel operator status to Darke
[11:04:36] --> Colourless has joined #Pentagram
[11:04:36] --- ChanServ gives channel operator status to Colourless
[11:05:03] <Colourless> hi
[11:08:12] <wjp> hi
[12:05:36] <-- sbx has left IRC ()
[12:25:56] --> skateKing has joined #pentagram
[12:25:59] <skateKing> hehe
[12:26:02] <skateKing> hi!:D
[12:27:17] <-- Kirben has left IRC ("System Meltdown")
[12:34:33] --> Kirben has joined #pentagram
[12:34:33] --- ChanServ gives channel operator status to Kirben
[12:34:50] <skateKing> hi Kirben: how are you doin'?
[12:35:25] <Kirben> Hi, not bad.
[12:37:28] <skateKing> ok
[12:52:54] <watt> wjp:hehe... my eyes played tricks on me for a second since I just woke.. thought you said complaints... heh
[12:53:17] <wjp> :-)
[12:53:27] <watt> glad people like it :-)
[12:55:04] <wjp> http://exult.sourceforge.net/forum/read.php?f=1&i=19789&t=19741
[12:55:13] <wjp> http://exult.sourceforge.net/forum/read.php?f=1&i=19993&t=19993
[12:57:45] <watt> yup, I fairly regularly read the forum.. but I don't think I've posted anything to it in a long time.
[12:58:32] <watt> Saw the first link.. second is new to me
[12:59:58] <-- skateKing has left IRC ("food is good")
[13:01:18] <watt> hmm.. I really should update my gentoo distro........ emerge -up world is getting a little longer than I like
[13:02:39] <watt> hmm.. and change my compiler flags... -03 is too agressive....|
[13:32:17] * Darke doesn't seem to have a problem with -O3 on anything he uses. Most stuff will strip that out if they expect a problem.
[13:39:00] <wjp> I did some major emerging last weekend; down to about 5 entries in -up world
[13:42:27] --> sbx|afk has joined #pentagram
[13:42:37] --- sbx|afk is now known as sbx
[13:44:58] <watt> I've heard in the past that -Os actually is safer and faster, but I haven't tested it out yet
[13:47:12] <watt> hmm.. actually firefox is compiling with -O2 on its own... that's interesting
[13:48:13] <watt> and a number of other flags
[14:00:09] <-- Kirben has left IRC ("System Meltdown")
[14:25:01] <Darke> There's nothing special about -Os that's intrinsicly safer or faster. It's basicly the same as -O2, except no options are turned on that *shouldn't* increase code size, and a couple of other optimisations are used that should decrease code size.
[14:25:31] <Darke> There's no real use for it unless you're programming for a small memory device. Or you want a particularly small kernel to toss on a floppy, or something odd like that.
[14:26:13] <Darke> -Os can also signficantly slow things down too, from memory.
[14:31:10] <wjp> watt: quite a few of the "larger" ebuilds strip a lot of optimization options
[14:33:38] <Darke> The cannonical example being OpenOffice, which at one time I believe basicly stripped all optimisation flags, and added a couple in that it knew it would work with, and went from there. *grin*
[14:35:32] <Darke> IIRC, pretty much every standards conforming program 'should' compile with -O3 (the exceptions usually being bugs in the compiler). Ones that take liberties with memory layout will usually work with -O2, and the rest you'll be lucky to get compile, let alone get optimised. *grin*
[14:36:52] <sbx> do you need broadband for Gentoo?
[14:37:12] <wjp> it helps :-)
[14:37:13] <Darke> Yes.
[14:37:30] <Darke> Well... no, not really, I believe DrCode did it without it. But it'll take... quite a while. *grin*
[14:37:42] <sbx> im getting a new computer this week but I probably wont have cable for a month or two
[14:37:53] <sbx> maybe I'll try SuSE
[14:38:32] <Darke> You can download compiled+source distro cds, but you really need a decent connection if you want to keep current, unfortunately.
[14:39:56] * Darke suggests SuSE. He just saw an install of an almost-latest version of it, and it's *so* pretty! Even nicer to use then it used to be when I was addicted to it, before transferring my alligance to The Great Gentoo. *grin*
[14:41:39] * Darke crawls off to bed. Night all and sundry!
[14:42:35] <wjp> night
[14:45:15] <sbx> bye
[14:45:52] <sbx> thanks for the suggestion, I remember DrCode used SuSE
[15:25:39] <-- GingerNinja has left IRC (Read error: 104 (Connection reset by peer))
[15:54:29] <-- exultbot has left IRC (ping timeout: 15 minutes)
[15:56:02] --> exultbot has joined #pentagram
[15:56:02] --- Topic for #pentagram is: Spoon!
[15:56:02] --- Topic for #pentagram set by Darke at Sun Feb 15 13:15:28 2004
[15:57:01] <-- Darke has left IRC (kornbluth.freenode.net irc.freenode.net)
[15:58:20] --> Darke has joined #pentagram
[16:30:45] --> Sectus has joined #pentagram
[16:46:38] --> GingerNinja has joined #pentagram
[17:19:45] <-- Darke has left IRC (Read error: 110 (Connection timed out))
[17:19:49] --> Darke has joined #pentagram
[17:32:04] --> nathan_ has joined #pentagram
[17:32:09] --- nathan_ is now known as unlord
[17:32:19] <unlord> any chance that this project is still underway?
[17:32:49] <wjp> I'd say that chance is about 100%
[17:33:02] <wjp> what makes you think otherwise?
[17:33:53] <unlord> oh woah
[17:33:56] <unlord> you all have a new website
[17:33:56] <-- GingerNinja has left IRC (Read error: 110 (Connection timed out))
[17:34:22] <unlord> it looked like your info hadn't been updated in a long time
[17:35:09] <sbx> the topic?
[17:35:44] <wjp> the webpage wasn't really a priority :-)
[17:36:04] <unlord> wjp: understood
[17:36:07] <wjp> I'll try to update it from time to time now :-)
[17:37:05] <unlord> how close are you all going to try and get the original gameplay?
[17:37:11] <unlord> is it difficult to recreate that?
[17:38:04] <wjp> things like plot and conversations should get very close to 100%
[17:38:28] <wjp> combat and monster AI related things might be somewhat less true to the original
[17:38:44] <unlord> do you all try to reverse engineer that stuff from the binary code?
[17:38:47] <wjp> (not necessarily worse than the original, but different)
[17:39:04] <unlord> right
[17:39:26] <sbx> it will probably be better :)
[17:39:28] <wjp> most of our knowledge about U8 comes from staring at the data files long enough
[17:40:31] <unlord> I should get exult working sometime and play through that
[17:42:53] <wjp> it should be fairly easy to get exult working
[17:43:00] <unlord> yah, I have it installed
[17:43:09] <unlord> I use gentoo linux, and I think I'll make an ebuild for it
[17:43:13] <wjp> there is one
[17:43:20] <wjp> games-engines/exult
[17:43:23] --> GingerNinja has joined #pentagram
[17:43:23] <unlord> for the data files?
[17:43:29] <wjp> oh, no, not for the data
[17:43:51] <unlord> it might be nice, dunno if its useful or not
[18:16:13] <wjp> watt: surfaceSearch might be better for finding supporting items (instead of a sweeptest 8 units down)
[18:17:12] <Colourless> yes non recursive surface search for direct supports and a recursive surface search for all supports
[18:25:45] <-- Sectus has left IRC ("Leaving")
[18:36:35] <wjp> hm, let's see
[18:36:49] <wjp> when picking up an item, we should call 'release' on the supporting items, right?
[18:37:30] <wjp> PLATE::release() seems to confirm that
[18:38:11] <wjp> I'm guessing Item::grab() would be a good place for this
[18:39:19] <sbx> does that make Crusader traps activate?
[18:39:58] <wjp> traps?
[18:40:53] <sbx> floor plates
[18:45:05] <wjp> in that case it's probably more what de-activates them
[18:45:16] <wjp> unless they're activated when you leave them
[18:45:39] <wjp> although this specific use of 'release' I just mentioned is more for picking up items
[18:45:55] <wjp> it should already work when you step on and off them yourself
[18:46:50] <sbx> hmm ok
[18:58:48] <wjp> oh great, Hanoi is broken *again*
[18:58:52] <wjp> but the pressure plates at least work :-)
[18:59:01] <wjp> bbl
[19:00:58] <Colourless> Hanoi has been broken for age
[19:00:59] <Colourless> s
[19:01:32] <Colourless> tis collision detection realted
[19:08:33] <wjp> heh, the Hanoi usecode calls AccumulateIntelligence :-)
[20:58:37] --> Sectus has joined #pentagram
[21:36:01] <wjp> I_legalMoveToPoint is returning (collideMove() != 0); isn't that kind of weird? Wouldn't (collideMove() == 0x4000) make more sense?
[21:36:28] <Colourless> ow
[21:36:41] <Colourless> a screwup if i've ever seen one
[21:36:43] <wjp> I don't think it's causing Hanoi to fail, though
[21:37:13] <wjp> since AFAICT, the collideMove should return 0x4000 at this point, which is still != 0 :-)
[21:37:48] <Colourless> it's about time i left
[21:37:51] <wjp> night
[21:38:04] <-- Colourless has left IRC ("casts improved invisibility")
[22:10:03] <wjp> ok, I _think_ the problem with sweepTest is that it fails to detect 'touches' at the end of the interval
[22:10:25] <wjp> or it's misinterpreted by collideMove
[22:13:24] <wjp> yay, it works now
[22:14:56] <wjp> great, I got the door to open :-)
[22:15:08] <wjp> (and promptly got toasted by those nice killer beams)
[22:18:17] <wjp> time to go; g'night
[23:15:34] <watt> wjp: Checked out surfaceSearch, one problem is that it check against the x,y,z of an item (avatar in this case) and avatar is not actually in that position that I wish to check... I could change the position and then check and change it back, but is that a good idea?
[23:59:35] --> TzarSectus has joined #pentagram