#uwadv@irc.freenode.net logs for 15 Nov 2002 (GMT)

Archive Today Yesterday Tomorrow
Underworld Adventures homepage


[15:06:56] --> yot has joined #uwadv
[15:26:13] <-- yot has left IRC ()
[16:27:21] --> Fingolfin has joined #uwadv
[16:27:33] <-- Fingolfin has left #uwadv ()
[16:41:05] --> vividos has joined #uwadv
[16:41:05] --- ChanServ gives channel operator status to vividos
[16:41:24] <vividos> hi
[16:52:18] --> b-llwyd has joined #uwadv
[16:54:15] <vividos> hi b-llwyd
[16:54:35] <b-llwyd> hi guys
[16:55:37] * b-llwyd just updated his cvs copy of uwadv, now he falls down, he doens't warp to the floor.
[16:57:02] <b-llwyd> i love the way it feels taking a step up to slightly higher ground in the game, its almost The Real Thing :)
[16:57:19] <vividos> yeah :)
[16:57:31] <vividos> I just added "look" for objects and walls
[16:57:51] * b-llwyd nods.
[16:58:21] <vividos> already seen it?
[16:58:29] <b-llwyd> when can you throw things around? ,)
[16:58:33] <b-llwyd> say no, dont think so
[16:58:58] <vividos> throwing things around is a bit more difficult, but the release after this one perhaps
[16:59:06] * b-llwyd nods again.
[16:59:49] <b-llwyd> is there a way to quit the game...Alt-X used to work in DOS, i think
[17:00:13] <vividos> yes, Alt+X quits the game, and Alt+Q returns to the main menu (when playing full uw1)
[17:01:16] * b-llwyd just discovered he had 'exec xmms' bound to Alt-X. o.O
[17:01:26] <vividos> heh
[17:02:12] <vividos> you always can alter the uwadv keymap by having a custom ~/.uwadv/keymap.cfg
[17:04:19] <b-llwyd> is there a sample keymap.cfg in the cvs?
[17:04:37] <vividos> yes, in uadata/uw1/
[17:04:48] <b-llwyd> k
[17:04:56] <vividos> do you "make install" before playing?
[17:05:05] <b-llwyd> yeah
[17:07:33] <vividos> could you send me a typical console log (or the stdout.txt, when that is created)? I'm just curious
[17:07:48] <vividos> btw, Alt+PageUp and Alt+PageDown changes levels
[17:07:54] <b-llwyd> in a minute, compiling the latest cvs
[17:07:59] <b-llwyd> oh
[17:08:07] <b-llwyd> say have to try that
[17:10:45] <b-llwyd> ok...where do you want the dump, e-mail ?
[17:11:04] <b-llwyd> (50+ lines)
[17:11:50] <vividos> yes, vividos@users.sourceforge.net
[17:15:26] <b-llwyd> it's on its way
[17:15:41] <vividos> ok thanks!
[17:16:32] <vividos> sorry, I more meant the log of the game :)
[17:16:38] <b-llwyd> oh, darn
[17:16:52] <b-llwyd> i thought i was clever :)
[17:17:16] <vividos> was my fault to not specify more exactly what log I wanted
[17:17:40] <b-llwyd> you mean the output from when you start up the game?
[17:17:58] <b-llwyd> like
[17:18:03] <b-llwyd> uwadv > log.txt
[17:18:25] <vividos> yes
[17:18:47] <vividos> but good that you sent the "make install" log, too. Just found a bug
[17:18:54] <b-llwyd> you want me to create a char and enter the game, or is the startup screen ok?
[17:19:01] <b-llwyd> aha
[17:19:22] <vividos> yes, enter the game with a char, run around a bit, maybe even changing levels.
[17:19:31] --> Coren_ has joined #uwadv
[17:19:36] <vividos> is ua-lowercase-all.pl installed into /usr/local/bin/linux/ ?
[17:19:38] <vividos> hi Coren_
[17:19:39] * Coren_ waves.
[17:19:48] <Coren_> WooWoo! My thesis project has been accepted!
[17:19:56] * Coren_ prances around the channel.
[17:20:01] <vividos> cool! what's the subject?
[17:20:49] <Coren_> "Of new idioms in human-machine interfaces." Basically, I get to throw all sorts of experimental UIs at guniea pigs and see how well they cope.
[17:21:25] <vividos> sounds interesting
[17:21:36] <Coren_> I have this one application (an appointment book), and about six different new idioms of UI.
[17:21:55] <vividos> what do you take for the guinea pigs?
[17:22:20] <Coren_> I'll give it to 18 volunteers (3 per) who have no CS experience (Joe Ramdom Users) from the management faculty, and record their woes and successes learning the new idioms.
[17:22:57] <vividos> hehe
[17:23:17] <vividos> do you write diploma thesis?
[17:23:32] * Coren_ failed to parse that question.
[17:24:15] <vividos> well, is it the diploma thesis or a master thesis? don't know exactly how it is in canada
[17:24:28] <vividos> or is it just a project?
[17:25:12] <Coren_> Ah. Doctorate thesis. The thesis-alike for a Master's is a 'mémoire'. I'm unsure about the english translation. I think "dissertation" IIRC but I'm not sure.
[17:25:52] <vividos> ah ok
[17:26:15] <vividos> any new features coded for low yet?
[17:26:41] <Coren_> Mouselook in CVS, and configuration file so that no paths are hardcoded anymore and settings can be saved.
[17:27:03] <Coren_> My check-out version has preliminary conversation engine being worked on.
[17:27:14] <vividos> cool
[17:27:22] <vividos> did you use infos from the uw-formats.txt?
[17:27:30] <Coren_> uw2 convo have new primitives, and LOTS more builtins.
[17:27:51] <vividos> do you mean opcodes with primitives?
[17:27:54] <Coren_> Yes; uw-formats was a good help to begin to parse; I'll have lots to add to it by the time I'm finished though.
[17:28:11] <Coren_> Yes, opcodes. "primitives" is the forth lingo. :-)
[17:28:18] <vividos> :)
[17:28:31] <Coren_> I have written a decompile-to-forth which will help lots.
[17:28:34] * vividos didn't look at conversation yet
[17:28:51] * vividos didn't look at conversation from uw2 yet
[17:29:18] <vividos> do you already have some output to show?
[17:30:17] <Coren_> Most remains the same, except that cnv.ark is compressed (with the same scheme as lev.ark).
[17:30:31] <Coren_> Output... you mean from the decompiler?
[17:31:00] <vividos> yes
[17:31:55] <Coren_> Some; although I have yet to transform the 0branch/branch into control structures-- so unless you are familiar with the way threaded code usually looks control flow is a bit hard to follow.
[17:32:18] <Coren_> Ah; and opcodes have been renamed to their "real" names. :-)
[17:32:43] <Coren_> Want to see an excerpt?
[17:32:45] <vividos> yes
[17:32:49] <vividos> is it long?
[17:33:17] <Coren_> I'll trim it for just one conversation; that'll give you enough to get a good feel.
[17:33:30] <vividos> you could send it to the mailing list
[17:33:41] <Coren_> Which?
[17:34:35] * b-llwyd sent a new console.log
[17:35:17] <vividos> thanks, b-llwyd!
[17:37:13] <vividos> Coren: uwadv-devel
[17:37:30] <Coren_> ...@sourceforge.net
[17:37:30] <Coren_> ?
[17:38:06] * Coren_ usually uses the sf interface for this.
[17:38:24] <vividos> uwadv-devel@lists.sourceforge.net
[17:38:39] <Coren_> Here; I have a fairly small conversation decompiled ready. I'll mail it now.
[17:39:34] <vividos> good
[17:52:40] <Coren_> 'tis sent.
[17:53:05] <-- b-llwyd has left IRC ("ttfn")
[17:53:46] <Coren_> Actually, when I said "fairly small" I hadn't realized it was the absolute smallest. :-)
[17:54:06] <Coren_> But enough to see how things work.
[17:54:41] <vividos> heh
[17:54:48] <vividos> "drop" is an opcode, I guess
[17:55:11] <Coren_> Yes. The "real" name of "POP"
[17:55:35] <Coren_> Or I should say the /Forth/ name thereof.
[17:56:05] <vividos> ah ok. I must admit Forth is really new for me
[17:56:43] <vividos> do you think they originally wrote all the conversations in Forth?
[17:57:44] <Coren_> I'm not sure-- It seems more likely to me that someone there who knew forth translated some imperative language into forth as an intermediate pass-- the code smells like machine-generated.
[17:58:04] <vividos> yes
[17:58:21] <Coren_> For one, if it has been written in forth it would be about half the size.
[17:58:34] <vividos> I also did another shot at decompiling the code to a C-like syntax (don't know if you've seen my mail on the mailing list)
[17:58:38] <Coren_> But the interpreter is definitely threaded code.
[17:58:46] <Coren_> (I've just seen it)
[17:59:16] <vividos> ... which just reminds me that cnvdasm still misses some code :)
[17:59:43] <Coren_> In case you are curious, in "proper" forth, say, 'w7a' would be like:
[17:59:59] <vividos> what do you mean with threaded code? to me it looks like that the code just runs as one thread on a virtual machine
[18:00:58] <Coren_> No, threaded in the interpreter model sense; thread tying blocks of code together. Not in the more recent multitasking sense.
[18:01:33] <Coren_> Try to find "Starting Forth" by Leo P. Brodie. Probably the best primer on this class of languages there is.
[18:01:44] <Coren_> PostScript is another well-known example.
[18:03:01] <vividos> don't think that our local uni library has "Starting Forth" (but they have tons of books about win32 api and visual c++)
[18:04:07] <Coren_> If you see something like: "a b 0branch>xx c d (xx) e f" the 'right' code is: "a b if c d then e f" which in c-like would be akin to "a; if(b) { c; d; } e; f" but it's not so clear because the if might be more like "if(a, b) {..."
[18:05:05] <Coren_> Which is why I doubt you'll ever be able to translate to c-like syntax with any success; stack-based interpreters do not have well-defined statement boundaries, or even expression boundaries for that matter.
[18:05:23] <vividos> it's beause it's a stack-based machine, I guess
[18:05:44] * Coren_ points to his previous line.
[18:05:48] <vividos> :)
[18:06:01] <Coren_> Hrm. Gotta go.
[18:06:03] * Coren_ waves.
[18:06:07] <vividos> I more used the decompiler for guessing "builtins" parameters
[18:06:09] * Coren_ is away: school
[18:06:10] <vividos> ok bye!
[18:07:12] <QQtis> hallo
[18:07:24] <vividos> hi QQtis
[18:08:46] <QQtis> just took a test in german
[18:10:19] <QQtis> for essay - we had to write an application for a job
[18:10:48] <QQtis> so I wrote Ich will ein Programmierer sein. Ich habe keine gute Bildung, aber mein Ruf ist gut. Ich bin kein auslandicshes Agent. Ich will fiel gut fur die CIA tun.
[18:11:11] <QQtis> :)
[18:11:32] <QQtis> god, I'm going crazy after this week.... too many tests
[18:13:07] <vividos> heh
[18:14:37] <vividos> Coren: ah, found some forth primer: http://www.forthprimer.hothere.com/
[18:17:59] <QQtis> vividos
[18:18:04] <QQtis> new CVS doe snot work
[18:18:16] <QQtis> as soon as it creates a character it crashes
[18:20:22] <vividos> did you do cvs update in the main project dir? I added some stuff to uadata/... to
[18:20:24] <vividos> too
[18:20:45] <QQtis> I think I did
[18:20:58] <QQtis> I just did CVS update to the whole thing
[18:21:06] <vividos> ok I'll test
[18:21:25] <vividos> (may take a while)
[18:21:37] <QQtis> loaded Lua compiled script "uw1/scripts/createchar.lob"
[18:21:37] <QQtis> importing 9 level maps from data/
[18:21:37] <QQtis> caught std::exception
[18:21:44] <QQtis> level 9 ????
[18:21:48] <vividos> yes
[18:21:49] <QQtis> might that be the problem?
[18:21:54] <vividos> no
[18:21:58] <QQtis> hm...
[18:22:27] <vividos> is the uw1-path in uwadv.cfg still ok? Maybe it was overwritten by the cvs update
[18:26:36] <QQtis> no teh path is good
[18:27:37] <QQtis> hm... couldn't the output file give more detailed accounts of the error?
[18:28:40] <vividos> unfortunately not, because it's a std::exception, and not a ua_exception (used by uwadv)
[18:30:22] <QQtis> hm....
[18:30:45] <QQtis> but it would be useful to see what data caused teh exception
[18:35:46] <vividos> hmm, the whole thing works for me here
[18:36:26] <vividos> do you use the uw_demo?
[18:37:32] <vividos> it could also be that some changed files were not recompiled. In that case you would have to recompile the whole thing
[18:40:15] <QQtis> not demo
[18:40:19] <QQtis> full version
[18:40:24] <QQtis> hm... ok I'll try that
[18:41:49] <vividos> ok
[18:42:24] <vividos> I'll start a README.music.txt that goes into the music packs
[18:42:53] <QQtis> cool
[18:43:00] <QQtis> what info do you need from me?
[18:44:01] <vividos> you could write your "statement" or opinion or remarks about the tracks, as you are the artist that did everything
[18:44:30] <QQtis> hm....
[18:44:34] <QQtis> ok
[18:44:38] <QQtis> when do you need this by?
[18:46:52] <vividos> on Thursday is release, so on Wednesday or so
[18:51:21] <QQtis> ok runs now
[18:51:23] <QQtis> but crashed again
[18:51:27] <QQtis> for no apparent reason
[18:51:44] <QQtis> format_item_name: using string red lizardmen
[18:52:04] <vividos> that is the last string?
[18:52:35] <QQtis> that's what it was
[18:52:40] <QQtis> now here si teh strange thing
[18:53:08] <QQtis> in game
[18:53:24] <QQtis> it actually looked at objects every tiem i left-clicked to walk
[18:53:36] <QQtis> which it shouldn't do unless you are in "eye mode"
[18:53:38] <vividos> I know
[18:53:45] <QQtis> but what's stranger is...
[18:53:53] <QQtis> it did not log any of these
[18:54:12] <QQtis> it only logged the ones that i right-clicked for
[18:54:26] <QQtis> because there are only 3 things listed in the logfile
[18:54:47] <vividos> was there output on the text scroll?
[18:55:18] <QQtis> yes
[18:55:25] <QQtis> lots of it too
[18:57:51] <vividos> what do you mean with "it did not log any of these"? The log file didn't mention the exact object properties?
[19:00:04] <QQtis> nope
[19:00:14] <QQtis> only for a few objects
[19:00:21] <QQtis> the ones I remember right-clicking on
[19:00:45] <QQtis> but not all of the ones it looked at as I left clicked to move
[19:04:16] <vividos> strange, but I change that part of code soon, so let's look at it then again
[19:05:38] <QQtis> ok
[19:40:58] <vividos> got to go. bye!
[19:43:13] <-- vividos has left IRC ("Leaving")
[20:19:09] --> b-llwyd has joined #uwadv
[20:20:38] <-- b-llwyd has left IRC (Client Quit)
[20:22:13] --> yot has joined #uwadv
[21:08:48] --> Eldron has joined #uwadv
[21:44:09] <-- Eldron has left IRC (Read error: 104 (Connection reset by peer))