| --- Log opened Wed Jan 30 00:00:21 2008 |
| 00:29 | | gnolam [lenin@Nightstar-10613.8.5.253.static.se.wasadata.net] has quit [Ping Timeout] |
| 00:30 | <+Attilla> | Yayvoice |
| 00:36 | | Vornicus [~vorn@Admin.Nightstar.Net] has joined #code |
| 00:36 | | mode/#code [+o Vornicus] by ChanServ |
| 01:09 | <@McMartin> | GSS-API? |
| 01:09 | <@Vornicus> | GSS-API? |
| 01:11 | <@McMartin> | 15:53 <+MyNetCatVerbs> Hrmn. Wonder why Ubuntu turns GSS-API on for SSH by default? |
| 01:19 | <+MyNetCatVerbs> | Generic Security somethingorother API. |
| 01:20 | <+MyNetCatVerbs> | As far as I understand it, the implementation you're most likely to run into is Kerberos. |
| 01:20 | <+MyNetCatVerbs> | It's a standard for writing programs to make use of Kerberos-ey systems. |
| 01:57 | | Serah [~Z@87.72.35.ns-26506] has quit [Quit: Don't try to read the quit message, that is impossible. Instead only realize the truth; "there is no quit message" and you will see it is not you who read the quit message but the quit message who reads you.] |
| 02:26 | | Syloqs-AFH [Syloq@NetAdmin.Nightstar.Net] has quit [Connection reset by peer] |
| 03:05 | | Derakon[AFK] is now known as Derakon |
| 03:33 | <@Vornicus> | Help, I have rolling-ball physics stuck in my head. |
| 03:33 | <@Derakon> | Oh, that reminds me - minigolf pool! |
| 03:33 | | * McMartin replaces it with the Super Mario Land theme. |
| 03:34 | <@Derakon> | Play pool on a table that has hills and valleys like a minigolf course. |
| 03:34 | <@Vornicus> | Heh. |
| 03:35 | <@McMartin> | And windmills! |
| 03:35 | <@McMartin> | And, um, water traps? |
| 03:35 | | * Vornicus recommends you look at least briefly at Kirby's Dream Course. |
| 03:35 | <@Derakon> | It'd have to be a big table. |
| 03:35 | <@McMartin> | platform? |
| 03:35 | <@Vornicus> | SNES |
| 03:36 | <@Derakon> | There's a TAS of it. |
| 03:36 | | * McMartin has a conference paper due in 28 hours, so. |
| 03:36 | <@Derakon> | Good luck. |
| 03:37 | <@Derakon> | Hmm...said TAS was made a bit over a year ago. I hope there's still someone hosting it. |
| 03:37 | <@Derakon> | The AVI file, that is. |
| 03:37 | <@McMartin> | There's always "half a seed" |
| 03:38 | <@Derakon> | ? |
| 03:38 | <@McMartin> | There's a slow server that manifests, if it's on TASvideos, if there are peers but no seeds. |
| 03:38 | <@McMartin> | It's really slow, but it means the torrent is always complete. |
| 03:38 | <@McMartin> | IME. |
| 03:39 | <@Derakon> | Ahh. |
| 03:39 | | * McMartin had a number do this. |
| 03:39 | <@Derakon> | Well, that's cool. |
| 03:39 | <@McMartin> | THey report on the download page as "1/2 seed" |
| 03:39 | <@Derakon> | Hah. |
| 03:48 | <@Vornicus> | ew, collisions. |
| 03:48 | <@Derakon> | Nani? |
| 03:49 | <@Vornicus> | Just, thinking about the collision math required for VornBall. And, while I'm at it, the level format. I fear that it will be something along the lines of beziers, mainly because it involves curves that I don't think I'll be able to make anywhere near enough tiles to handle that way. |
| 03:50 | <@Derakon> | What is VornBall? Is this the morphball-maze game? |
| 03:53 | <@Vornicus> | Yes. |
| 03:53 | <@Derakon> | Well, sphere-bezier collision isn't terribly difficult, if I remember from eight years back correctly. |
| 03:54 | <@Vornicus> | True. And it does offer a rasonably simple culling method - you can't collide with a bezier if you're not close enough to its convex hull. |
| 03:54 | <@Derakon> | How many dynamic objects do you want to have? |
| 03:55 | <@Derakon> | If it's just the ball, then sure...but you may want to just go ahead and make a quad tree. They aren't *too* complicated if you don't have to worry about memory management (hint: don't use C++). |
| 03:55 | <@Vornicus> | Not many. |
| 03:56 | <@McMartin> | Hint: Don't use C++ anyway~ |
| 03:56 | | * Derakon pats Niobium. |
| 03:57 | <@Derakon> | You may also want to consider how you're going to handle the powerbombs. |
| 03:57 | <@Vornicus> | I'm /really/ not a fan of quadtrees. |
| 03:57 | <@Vornicus> | Mainly because I do not see how they manage to work at all. |
| 03:58 | | * Derakon blinks. |
| 03:58 | <@Derakon> | You mean in terms of how they actually return accurate results, or in terms of how they save you effort? |
| 03:58 | <@Vornicus> | I mean, it looks to me like you end up hitting your threshold limit, repeatedly, for every object, and I don't see how that is at all useful. |
| 03:59 | <@Vornicus> | s/threshold/division |
| 03:59 | <@Derakon> | I'm not certain what you mean. |
| 03:59 | <@Vornicus> | I mean, you put objects into your quadtree, and you have to enumerate the things that the objects fall into. |
| 04:00 | <@Derakon> | This is where recursion comes in. |
| 04:00 | <@Vornicus> | One diagonal line, and poof! suddenly you have to divide and divide and divide, and you end up bottoming out, or recursing forever. |
| 04:00 | <@Derakon> | Erf? |
| 04:01 | <@Derakon> | You have a minimum cell size. |
| 04:01 | <@Vornicus> | Right, that's the bottoming out. |
| 04:01 | <@Vornicus> | I do not see how you manage to /not/ have every object bottom out. |
| 04:01 | <@Derakon> | I'm still having problems understanding your problem, so let me just briefly describe my collision-detection algorithm in Niobium. |
| 04:02 | <@Derakon> | Or rather, the algorithm that sets up pairs of objects to collide. |
| 04:02 | <@Derakon> | Start at the root node. The root node has a list of objects I was unable to push further. I collide all these objects against each other. Then I pass the list down to the four children. |
| 04:02 | <@Vornicus> | "push further"? |
| 04:02 | <@Derakon> | They don't fit into my child cells. |
| 04:02 | <@Derakon> | Because they cross a boundary. |
| 04:02 | <@Vornicus> | ...oh. |
| 04:03 | <@Vornicus> | Okay, so you /don't/ divide up objects that cross boundaries? |
| 04:03 | <@Derakon> | What? No! |
| 04:03 | <@Derakon> | That explains your concern, I think... |
| 04:04 | <@Derakon> | The bookkeeping involved in trying to intelligently have each object in two different deeply-nested cells is hideous. |
| 04:04 | <@Vornicus> | Quite. |
| 04:04 | <@Derakon> | So anyway. At the child level, I have two lists - the one passed down, and the one of objects I couldn't push further. I prune out the ones in the passed-in list that aren't in my cell, and then collide them against everything in my list (and everything in my list against its pals). |
| 04:04 | <@Derakon> | Then I pass the combined list to my children. |
| 04:05 | <@Vornicus> | This sounds like it works best for shittons of little objects. |
| 04:05 | <@Derakon> | Er, collide the remainder. |
| 04:05 | <@Derakon> | Well, it works well when your objects are reasonably evenly-distributed. |
| 04:06 | <@Derakon> | The version on Wikipedia also recommends having a maximum occupancy for each cell, so densely-populated areas are more heavily-subdivided. I should probably do that in Niobium, but it's not crucial. |
| 04:07 | <@Derakon> | But consider terrain, for example - pretty much your entire relevant map will have terrain in it, but you don't want to be considering every terrain tile for collision. Quadtrees allow you to toss out the ones you don't care about, and you'll be doing a *lot* of terrain collision tests. |
| 04:08 | <@Vornicus> | One nice thing is that terrain is stationery |
| 04:08 | <@Derakon> | Yes. Once you've put it into the tree, you don't have to waste time continually changing its home turf. |
| 04:09 | <@Derakon> | I need to go fix dinner. BBIAB. |
| 04:09 | <@McMartin> | Terrain is, idly, stationary. |
| 04:09 | <@McMartin> | Unless you can write on it. |
| 04:09 | <@McMartin> | Nord and Bert, awaaaayyyyyyy~ |
| 04:09 | | * Vornicus gives McM a spray-can. |
| 04:10 | <@Vornicus> | That's one I need to try, too. |
| 04:10 | <@McMartin> | arr |
| 04:14 | <@Vornicus> | Hrm. |
| 04:15 | <@Vornicus> | ...ARG! I have CODE TO WRITE that I HAVE ALREADY DESIGNED |
| 04:19 | <@Derakon> | Sucks, don't it? |
| 04:21 | <@Vornicus> | Yes. |
| 04:32 | | Doctor_Nick [~nick@Nightstar-23600.hsd1.fl.comcast.net] has joined #code |
| 04:32 | <@Vornicus> | Also, I need to refactor the code I have. |
| 04:34 | <@Vornicus> | because it's nasty and horrible. |
| 04:35 | <@Vornicus> | Granted, this is by Python standards, so the longest existing method is 20 lines, but |
| --- Log closed Wed Jan 30 05:23:15 2008 |
| --- Log opened Wed Jan 30 05:23:21 2008 |
| 05:23 | | TheWatcher[zZzZ] [~chris@Nightstar-29731.dsl.in-addr.zen.co.uk] has joined #code |
| 05:23 | | Irssi: #code: Total of 18 nicks [10 ops, 0 halfops, 6 voices, 2 normal] |
| 05:23 | | mode/#code [+o TheWatcher[zZzZ]] by ChanServ |
| 05:23 | | Irssi: Join to #code was synced in 36 secs |
| 07:27 | | You're now known as TheWatcher |
| 09:37 | | RBot [~Reiver@Nightstar-23041.xdsl.xnet.co.nz] has joined #Code |
| 09:37 | | DiceBot [~Reiver@Nightstar-21973.xdsl.xnet.co.nz] has quit [Ping Timeout] |
| 09:38 | | Reiver [~reaverta@Admin.Nightstar.Net] has quit [Ping Timeout] |
| 09:38 | | RBot is now known as DiceBot |
| 09:45 | | Reiver [~reaverta@118.90.67.ns-12549] has joined #Code |
| 09:45 | | mode/#code [+o Reiver] by ChanServ |
| 10:10 | | DiceBot [~Reiver@Nightstar-23041.xdsl.xnet.co.nz] has quit [Ping Timeout] |
| 10:11 | | Reiver [~reaverta@118.90.67.ns-12549] has quit [Ping Timeout] |
| 10:13 | <@TheWatcher> | HAAAATE PHP. HAAAAATEE |
| 10:13 | | * TheWatcher stabstabstabs it |
| 10:15 | | DiceBot [~Reiver@Nightstar-10761.xdsl.xnet.co.nz] has joined #Code |
| 10:15 | | mode/#code [+v DiceBot] by ChanServ |
| 10:17 | | Reiver [~reaverta@118.90.70.ns-4464] has joined #Code |
| 10:17 | | mode/#code [+o Reiver] by ChanServ |
| 10:19 | | ToxicFrog [~ToxicFrog@Admin.Nightstar.Net] has quit [Operation timed out] |
| 10:20 | | ToxicFrog [~ToxicFrog@Admin.Nightstar.Net] has joined #code |
| 10:20 | | mode/#code [+o ToxicFrog] by ChanServ |
| 10:27 | | * TheWatcher stabstabstabstabs PEAR for good measure, and for being utterly crap too |
| 10:28 | | * McMartin erases PHP from history |
| 10:28 | <@TheWatcher> | I wish... |
| 10:31 | <@TheWatcher> | I mean, I have the latest PEAR stable version installed (1.6.2-r1) and PEAR.php itself contains code that will cause errors if E_STRICT is enabled. I mean, WTF mate? |
| 10:35 | | RBot [~Reiver@Nightstar-22195.xdsl.xnet.co.nz] has joined #Code |
| 10:35 | | DiceBot [~Reiver@Nightstar-10761.xdsl.xnet.co.nz] has quit [Ping Timeout] |
| 10:36 | | Reiver [~reaverta@118.90.70.ns-4464] has quit [Ping Timeout] |
| 10:37 | | DiceBot [~Reiver@Nightstar-22102.xdsl.xnet.co.nz] has joined #Code |
| 10:37 | | mode/#code [+v DiceBot] by ChanServ |
| 10:38 | | RBot [~Reiver@Nightstar-22195.xdsl.xnet.co.nz] has quit [Ping Timeout] |
| 10:45 | | Reiver [~reaverta@118.90.71.ns-12006] has joined #Code |
| 10:45 | | mode/#code [+o Reiver] by ChanServ |
| 16:07 | | Doctor_Nick [~nick@Nightstar-23600.hsd1.fl.comcast.net] has quit [Quit: Konversation terminated!] |
| 16:24 | | gnolam [lenin@Nightstar-10613.8.5.253.static.se.wasadata.net] has joined #Code |
| 16:24 | | mode/#code [+o gnolam] by ChanServ |
| 17:00 | | You're now known as TheWatcher[afk] |
| 17:56 | | UndeadAnno is now known as Pete |
| 18:30 | | You're now known as TheWatcher |
| 18:33 | <+MyNetCatVerbs> | This actually makes for an unsurprisingly lean, wonderfully stable IRC client. |
| 18:34 | <+MyNetCatVerbs> | I'm tempted to add logging to it, after which comes a slippery slope - automatic login, followed by colour highlighting by abuse of escape codes... |
| 20:43 | | MyNetCatVerbs [~richard@Nightstar-16091.cable.ubr06.hawk.blueyonder.co.uk] has quit [Client exited] |
| 20:44 | <@MyCatVerbs> | Even quits nicely, heh. |
| 20:51 | <@ToxicFrog> | What does? |
| 20:53 | | * MyCatVerbs waves at hacked up netcat substitute. |
| 20:53 | <@MyCatVerbs> | Wrote a program that repeatedly grabs lines from stdin, sends them along socket. |
| 20:54 | <@MyCatVerbs> | Also grabs lines from socket, throws them to stdout. |
| 20:55 | <@MyCatVerbs> | The only difference from netcat is that it checks messages to see whether or not they're PING. If they are, it chucks a PONG line down the socket. |
| 21:15 | <@ToxicFrog> | Doesn't netcat come with a bash script to do that? |
| 21:15 | <@ToxicFrog> | (also, I hacked together something similar using awk to make the output more readable) |
| 21:16 | <@Vornicus> | That doesn't look like a nice quit |
| 21:16 | <@Vornicus> | A nice quit would say "Quit:" |
| 21:18 | <@ToxicFrog> | Hmm. Yes, it does. I remember now. Used that as the basis of my script, added support for a few other commands, then filtered it through eight lines of awk. |
| 21:18 | <@ToxicFrog> | Doesn't scale well, though. |
| 21:30 | | MyNetCatVerbs [~richard@Nightstar-16091.cable.ubr06.hawk.blueyonder.co.uk] has joined #code |
| 21:45 | | MyNetCatVerbs [~richard@Nightstar-16091.cable.ubr06.hawk.blueyonder.co.uk] has quit [Ping Timeout] |
| 21:58 | | gnolam [lenin@Nightstar-10613.8.5.253.static.se.wasadata.net] has quit [Quit: Z?] |
| 22:00 | | UndeadAnno [AnnoDomini@83.21.84.ns-3801] has joined #Code |
| 22:01 | | Pete [AnnoDomini@83.21.14.ns-2903] has quit [Ping Timeout] |
| 22:01 | | UndeadAnno is now known as Pete |
| 22:17 | | Pete is now known as UndeadAnno |
| 22:24 | | You're now known as TheWatcher[T-2] |
| 22:30 | | You're now known as TheWatcher[zZzZ] |
| 23:56 | | Vornicus is now known as Finerty |
| --- Log closed Thu Jan 31 00:00:20 2008 |