#exult@irc.freenode.net logs for 22 Nov 2011 (GMT)

Archive Today Yesterday Tomorrow
Exult homepage

[00:28:04] --> Matt_O1 has joined #exult
[00:29:14] <-- Matt_O has left IRC (Ping timeout: 276 seconds)
[02:41:51] <-- Darrenor64 has left IRC (Ping timeout: 248 seconds)
[04:37:12] --> Darrenor64 has joined #exult
[06:29:16] --> Colourless has joined #exult
[06:29:16] --- ChanServ gives channel operator status to Colourless
[07:48:06] --> Matt_O has joined #exult
[07:49:35] <-- Matt_O1 has left IRC (Ping timeout: 248 seconds)
[08:27:10] --> Dominus has joined #exult
[08:27:10] --- ChanServ gives channel operator status to Dominus
[09:44:03] --> SiENcE has joined #exult
[11:56:41] <-- SiENcE has left IRC (Quit: @all: cya)
[12:28:54] <-- Kirben has left IRC ()
[13:19:28] --> SiENcE has joined #exult
[14:01:48] --> TheCycoONE has joined #exult
[14:16:53] <Dominus> wjp, any idea what I screwed up with the hq4x implementation to screw combat?
[14:17:42] <wjp> when _using_ hq4x?
[14:18:01] <Dominus> actually it shows with point as well
[14:18:06] <Dominus> that's what baffles me
[14:19:16] <Dominus> I have a snapshot with marzo's changes and that works fine but the snapshot with my changes shows very uncombatitive enemies
[14:24:03] <wjp> I don't see any reason why it would influence things when you're not using it
[14:24:39] <Dominus> really really strange
[14:24:57] <Dominus> people on the forum are seeing it too (or rather they prompted me to check on it)
[14:36:31] <wjp> wasn't there some "lack of aggressiveness" issue some months back too?
[14:37:23] <Dominus> yes, but I think Malignant got it under control back then
[14:41:53] <wjp> any idea where and when?
[14:42:37] <wjp> and is it the same type of "uncombative enemies" now?
[14:43:40] <Dominus> hmm, got to look that up
[14:45:47] <Dominus> http://sourceforge.net/tracker/?func=detail&aid=3059581&group_id=2335&atid=102335
[14:46:06] <Dominus> revision 6329 broke it and it seems you fixed it
[14:48:11] <Dominus> you seem to have fixed it in rev 6360 and marzo did something with actors in rev7095
[14:48:53] <Dominus> he actually reverted your change :)
[14:49:19] <Dominus> strange that it seemed to work fine, amybe my snapshot is incomplete
[14:51:16] <Dominus> so what is it with the rands?
[15:04:53] <wjp> oh sheesh
[15:05:07] <wjp> there's a FIXME there and Marzo just blindly changed it? :-(
[15:05:30] <wjp> but thanks for tracking that down
[15:06:29] <wjp> the issue is that that code did not at all do what it appeared to do, and Colourless Marzo changed the code to actually do what it appeared to do
[15:06:50] <wjp> which breaks things completely, because it is probably just wrong
[15:07:49] <wjp> hence the fixme
[15:08:18] <wjp> so I guess we should just re-commit 6360, this time with an ever bigger FIXME :-)
[15:08:23] <wjp> s/ever/even/
[15:09:52] <Dominus> an explicit fix me warning what goes wrong if reverted again :)
[15:12:47] <TheCycoONE> knees will be broken?
[15:56:13] <wjp> reverted that again...
[16:37:01] --> ParuCodex has joined #exult
[16:37:16] <Marzo> I am thinking: since the '!rand()%3' and the '!rand()%4' are basically true almost all the time (except for one case out of every 4294967295 in a 32-bit system), maybe we should just get rid of those checks?
[16:39:24] <-- ParuNexus has left IRC (Ping timeout: 240 seconds)
[16:49:48] <sh4rm4> !rand()%4 should be true in 25 % of all cases...
[16:57:51] <Marzo> sh4rm4: you are forgetting operator precedence
[16:58:14] <Marzo> !rand()%4 is actually equal to (!(rand()))%4
[16:58:50] <sh4rm4> mmh... could be true
[16:58:56] <sh4rm4> i think i ran into that myself once
[16:59:32] <Marzo> !rand()%4 is then only true when rand() is zero -- !0 == true == 1 and 1%anything is 1
[17:04:41] <sh4rm4> yep, just checked
[17:09:35] <sh4rm4> ah... finally seeing the diff
[17:09:37] <sh4rm4> after about 10 clicks and 3 nginx timeouts in the web interface
[17:13:03] <sh4rm4> at least sf.net uses a proper git web interface nowadays...
[17:15:59] <wjp> Marzo: the reason I put a FIXME there and didn't just remove it, is that I don't know what that code was trying to accomplish
[17:19:17] <Marzo> I wrote one of those checks if memory serves; the other one was Jeff I think
[17:19:34] <Marzo> The intent was to not check for foes every time the code ran
[17:19:44] <Marzo> But operator precedence screwed us both
[17:19:59] <wjp> well, yes, but there's a rather large range of "not every time" :-)
[17:21:11] <wjp> from, say, two in three to 4294967295 out of 4294967296 ;-)
[17:21:40] <Marzo> I wonder if the blocks can be removed entirely and still work
[17:21:58] <Marzo> After all, checking too "often" by fixing seems to cause enemies not to respond
[17:22:20] <Marzo> (fixing *it* the naive way, I mean)
[17:23:10] <wjp> right... it seems to change things in the wrong direction...
[17:24:03] <wjp> I wonder which of the two is responsible
[17:25:22] <sh4rm4> it was like && 0 && 0
[17:25:35] <sh4rm4> so never executed
[17:25:50] <wjp> yes
[17:26:49] <Marzo> The interesting think is that both of them never executing result in more responsive foes
[17:26:50] <sh4rm4> meaning that the following code really slows things down
[17:27:06] <wjp> sh4rm4: yes, we got that for :-)
[17:27:09] <wjp> s/for/far/
[17:27:26] <wjp> the question is why :-)
[17:29:02] <Marzo> I am trying to investigate this, but it seems the Gnome folks are stuck into 'destroy Gnome' mode, and have been removing useful features left and right from Gnome tools
[17:29:21] <Marzo> (for example, they seem to have removed the 'find in files' feature from Anjuta)
[17:29:38] <Marzo> Wll, back to grep
[17:29:59] <sh4rm4> anyway, they seem to seek foes regardless of that stuff
[17:30:01] * wjp switched away from gnome for pretty much that reason about 8 years ago :-)
[17:31:00] <Marzo> What are you using nowadays?
[17:31:22] <Marzo> I am considering moving to xfce (as I hate KDE)
[17:31:28] <wjp> xfce
[17:31:31] * sh4rm4 lxde
[17:33:02] <Marzo> That is odd; other than Npc_actor::handle_event, only the patrol schedule calls Schedule::seek_foes
[17:33:27] <Marzo> I wonder what else may be causing the enemies to change to combat schedule
[17:37:06] <wjp> Actor::set_target maybe?
[17:37:55] <Marzo> Seems to be
[17:38:14] <Marzo> Also, Actor::fight_back
[17:38:51] <Marzo> (but only for party)
[17:39:58] <Marzo> Hm. Schedule::seek_foes also seems to be missing several valid foes possible
[18:11:40] <-- SiENcE has left IRC (Quit: @all: cya)
[18:23:47] <sh4rm4> heh http://thehairpin.com/2011/11/women-struggling-to-drink-water
[20:24:15] <-- sh4rm4 has left IRC (Ping timeout: 248 seconds)
[20:28:46] --> sh4rm4 has joined #exult
[21:26:40] <-- Dominus has left IRC (Read error: Connection reset by peer)
[21:26:51] --> Dominus has joined #exult
[21:26:51] --- ChanServ gives channel operator status to Dominus
[21:38:39] --> SiENcE has joined #exult
[22:06:25] <-- TheCycoONE has left IRC (Quit: KVIrc 4.0.4 Insomnia http://www.kvirc.net/)
[22:11:50] <-- SiENcE has left IRC (Quit: cya)
[22:24:54] --> Kirben has joined #exult
[22:24:54] --- ChanServ gives channel operator status to Kirben
[23:35:08] --> Strangerke has joined #exult
[23:35:34] <Strangerke> hi guys
[23:38:24] <Dominus> hi and mostly bye...
[23:38:28] <Dominus> :)
[23:41:26] <Strangerke> indeed, it's late :)
[23:41:53] <Strangerke> I read some post on vogons about Drake. Is he still sometimes around?
[23:42:20] <Dominus> Darke, and no haven't seen him in ages
[23:42:25] <Dominus> !seen Darke
[23:42:28] <Strangerke> oups, sorry
[23:42:41] <Dominus> ?seen darke
[23:42:41] <exultbot> darke left IRC around Tue Sep 8 11:15:26 2009 (GMT) (Read error: 110 (Connection timed out))
[23:43:00] <Strangerke> oups, yes.. really a long time
[23:43:22] <Dominus> and I think this was a one time stop over from a long time absence before that
[23:44:19] <Strangerke> :(
[23:44:54] <Strangerke> Well, thanks for those details...
[23:45:03] <Strangerke> And good night, if you're going to sleep :)
[23:45:21] <Dominus> good night, btw. his email is on the abaout us page...
[23:45:55] <-- Dominus has left IRC (Quit: Leaving.)
[23:45:55] <Strangerke> Thanks :)
[23:47:13] <-- Darrenor64 has left IRC (Ping timeout: 260 seconds)