#bitfighter IRC Log

Index Search ←Prev date Next date→

IRC Log for 2013-10-16

Timestamps are in GMT/BST.

00:14:34BFLogBot Commit: a1337ffc87cf | Author: buckyballreaction | Message: Fix potentially fatal bug in building triangulation geometry
00:14:49kaenthat looks promising
00:14:55raptorit's not really...
00:14:59kaenoh
00:15:04kaendid it fix the insignia segfault?
00:15:07raptornope
00:15:26kaenoh wait
00:15:34kaenjust saw the diff
00:15:42kaenwell, anyway that's definitely no good
00:15:51raptorboost polygon library?
00:16:06raptori didn't know one existed...
00:16:13kaenprobably wasn't causing any problems though because it'll just shadow the outer j within that scope
00:16:20raptoryeah
00:16:33kaenbut yeah, boost has a rock solid poly library
00:16:55kaensponsored by intel, no less
00:17:01raptoroooo
00:18:59raptor3392.136161, 4086.776194
00:19:01raptor3392.154161, 4086.872194
00:19:09raptorthose are two points fed into poly2tri
00:19:20kaenyuck
00:19:40kaenI always thought our float -> int -> float transformation was icky
00:19:53raptoryeah..
00:19:53kaendefinitely thwart's poly2tri's tauted numerical stability
00:19:59kaenthwarts
00:20:07kaenman I've been going nuts with apostrophes lately
00:21:09kaen" Specific algorithms provided are the polygon set operations (intersection, union, difference, disjoint-union) and related algorithms such as polygon connectivity graph extraction, offsetting and map-overlay"
00:21:28kaenit also has voronoi generator, which requires a constrained delauney triangulation
00:21:36kaenalthough I don't see an api for it
00:21:39raptorhuh
00:21:54raptorwell I think clipper is solid... but our conversions are biting us
00:21:59kaenif it does support CDT, it can replace clipper, recast, and poly2tri
00:22:12raptorhmmm
00:28:08raptoroh interesting
00:28:18raptorthere is precision loss on saving points to the level file
00:33:32raptoryep
00:33:34raptorsure it
00:33:37raptor*is
00:56:25raptorok, so when points are snapped to each other there can be floating point precision problems...
00:59:46koda has joined
01:07:25Flynnn Quit (Quit: This computer has gone to sleep)
01:11:19Invisible1 has joined
01:16:18raptor Quit ()
01:34:42watusimoto has joined
01:34:42ChanServ sets mode +o
01:34:42tjcarter Quit (Read error: Connection reset by peer)
01:34:59tjcarter has joined
02:02:30tjcarter Quit (Ping timeout: 241 seconds)
02:26:01Platskies has joined
02:31:17Skybax Quit (Ping timeout: 256 seconds)
03:02:18tjcarter has joined
03:31:31Platskies Quit (Quit: Sleep time…)
03:36:45Invisible1 Quit (Ping timeout: 240 seconds)
03:39:11Platskies has joined
04:46:01Canseco has joined
04:58:13Canseco Quit (Quit: Leaving)
05:55:57Platskies Quit (Quit: Gotta go)
06:59:11Invisible1 has joined
07:12:08kaenwatusimoto, I've been doing a thorough search of boost.polygon docs
07:12:20kaenit looks as though we could replace clipper, recast, and poly2tri with it
07:12:47watusimotothe advantage?
07:12:51watusimotolack of crashes?
07:13:12kaenyeah, raptor found another poly2tri crash
07:13:20kaenreachable just by loading a certain level
07:13:21watusimotois it the one he fixed?
07:13:26kaenit's not fixed
07:13:35watusimotook, he fixed some related issue last night
07:13:46kaenyou mean renaming j -> k ?
07:13:49watusimotoyes
07:14:01kaenthat probably wasn't causing any problems at all
07:14:09kaenbecause j just shadows the outer j there
07:14:17watusimotoah, I see
07:14:24kaenthere's at least two problems with our triangulation dataflow:
07:14:42kaenfloat -> int -> float (by rescaling) definitely invites precision loss
07:15:03kaenand poly2tri and clipper both have separate fatal epsilon values
07:15:24watusimotoso what are the downsides? license is ok, performance is ok (probably) is it just importing boost libs?
07:15:30kaenand I guess a third, which is that poly2tri will recurse infinitely or segfault rather than doing a thorough check of its input
07:15:37kaenafaik, that's it
07:16:00kaen"The implementation outperforms most of the known commercial and non-commercial libraries in both efficiency and numerical robustness aspects"
07:16:05kaenit's sponsored by intel, too
07:16:21watusimotowell, I will always vote for stability. code base size is not a factor for me. have you discussed with raptor?
07:16:34kaenI mentioned it to him but got no response
07:16:54watusimotohow much extra code do you think we'd need?
07:17:16kaenI doubt we would need any extra, I think we'd actually reduce code
07:17:16watusimotoi.e. just the boost geometry libs, or does that import boost.*?
07:17:25kaenoh, not sure just yet
07:17:27watusimotook
07:17:46watusimotowell, you are right -- our new imports would be offset by getting rid of some stuff
07:17:57watusimotowould we still keep clipper?
07:18:07kaenI don't see why
07:18:12watusimotoI like clipper!
07:18:15watusimotobut
07:18:16kaenit does polyboolean, offsetting, etc
07:18:30kaenyeah but just using the boost.poly stuff saves us data transformation
07:18:39kaenand boost.poly can be made to operate on F32 directly
07:18:48kaenwhich makes it a clear winner in my book
07:18:54watusimotoso my vote would be to go for it, and if the performance isn't too much worse than what we've got, we'll all be happier
07:19:05kaenok I'll see what I can cook up
07:19:16watusimotocutting down on mindless data conversions would be a performance boost, for sure
07:19:31kaenthose intel guys
07:19:36watusimotoand boost is well vetted code
07:19:38kaenthey know a thing or two about performance
07:19:48watusimotoindeed
07:20:02watusimotostart cooking!
07:20:08kaenwill do!
07:20:18kaenI've got three days off so this is perfect timing
07:20:32watusimotoif that's how you want to spend your time off :-)
07:20:52Invisible1 Quit (Ping timeout: 245 seconds)
07:32:40kaencomputational geometry is fun work, it's challenging, useful, and you get to visualize the product of your effort :)
07:32:40tjcarter Quit (Read error: Connection reset by peer)
07:32:55tjcarter has joined
07:46:51watusimotoso should we create a test suite for our integration with the geom libs?
07:47:01watusimotoi.e. feed them levels we know cause problems?
07:47:17watusimoto(not talking about testing the lib iteself)
07:47:30kaenI think so
07:47:41kaeninsignia being a good first candidate :)
07:48:00kaenI think that if we do so we should add a "fixture" sub dir to our test suite
07:48:06watusimotoy
07:48:07kaenfor putting level code we use in testing
07:48:08watusimotoyes
07:48:29watusimotoI have a .h file for that purpose already, but it could perhaps be restructured
07:48:49watusimotomay
07:49:08watusimotomaybe for full level files, we should store them as files and just read them in
07:49:25watusimotothat would test the level reading code at the same time
07:49:46kaenthat's sensible
07:50:15watusimotowhat I've done is best suited for short compact levels
08:30:49tjcarter Quit (Read error: Connection reset by peer)
08:31:04tjcarter has joined
08:34:36Invisible1 has joined
08:37:25Watusimoto_ has joined
08:56:02tjcarter Quit (Quit: brb)
08:58:12HylianSavior has joined
09:02:44kaenwatusimoto, any reason why TNL::Vector doesn't just inherit from std::vector ?
09:02:54kaenrather than wrapping it via innerVector?
09:08:10Watusimoto__ has joined
09:09:20kaenWatusimoto__, any reason why TNL::Vector doesn't just inherit from std::vector ?
09:09:22kaenrather than wrapping it via innerVector?
09:10:58watusimotokaen: I'm not sure -- raptor coded that stuff long ago. He probably wrapped it because I mentioned wrapping, though I don't think we ever discussed the merits of one over the other. Inheriting sounds like a better option.
09:11:19kaenit'd make boost::polygon integration tons easier
09:11:32watusimotoalso, at the time, he may have just been getting up to speed with c++, so he probably did not make an explicit decision
09:11:34Watusimoto_ Quit (Ping timeout: 272 seconds)
09:11:37watusimotojust did what worked
09:11:46kaenok, mind if I change it?
09:11:50watusimotoyou can change it if you like
09:11:52kaenok thanks
09:11:54watusimotoyes
09:11:56watusimotook
09:12:01watusimotoI mean no
09:12:05watusimotoI mean change it
09:12:16kaenhehe
09:12:31kaen"mind if X" is a terrible English construct
09:12:46watusimotoit does lead to some confusion from time to time
09:13:53kaenit's especially bad in my area because colloquially people sometimes say "yes" when they mean "go ahead"
09:14:08kaenso you never know if "yes" means "yes" or "no"
09:14:29watusimotono it does, yes
09:16:41kaenI also get a lot of "them are" and "I seen"
09:16:51kaenmakes me want to rip my hair out
09:40:45raptor has joined
09:40:45ChanServ sets mode +o
09:40:49raptorgood morning!
09:40:54raptorquestions questions...
09:41:18raptorTNL::Vector not inheriting? I think it was because I was young in the c++ arts at the time...
09:41:20raptoralso...
09:41:28raptormaybe because of Vector<bool>
09:44:26raptoras for boost geom libraries - I'm OK with it, but if there is code that needs to be linked, I'm a bit more wary (i.e. non-header-only stuff)
09:46:16watusimotoI think the geom stuff will require linked code... but as kaen pointed out, we can get rid of some embedded code as a tradeoff
09:46:26raptorhi watusimoto
09:46:30watusimotohey there
09:46:46Invisible1 Quit (Ping timeout: 265 seconds)
09:47:28raptorso I found precision errors all over the place last night (which I think it part of our problem). I want to tell you what I found then ask you for ideas on how to handle
09:47:35watusimotokaen: beware the Vector<bool> issue... we want our own implementation there, not the vector one
09:47:59Watusimoto__ Quit (Ping timeout: 246 seconds)
09:48:00raptorour Vector<bool> is like 10 times faster than std's
09:48:02watusimotook, but be aware I am leaving in about 10 minutes
09:48:05raptorok
09:48:11raptorhere's the flow
09:48:37raptorprecision isn't always lost, but about 50%, I think
09:48:59raptor1. create polywall with one vertex not on the grid (using ctrl+space)
09:49:24raptor2. snap a vertex of another polywall (or any object with vertex) to it
09:50:43watusimotois there a 3 coming?
09:50:47raptor3. save -> precision differences between the exact same point are saved to the file like so:
09:50:51watusimoto:-)
09:50:51raptorPOINT 1: -0.667173 -0.162639
09:51:01raptorPOINT 2: -0.667176 -0.162643
09:51:21raptorthis is precision lost #1
09:51:40watusimotowe should create a test to demonstrate this
09:51:51raptor4. on level load, it upscales by gridsize and by 1000 (clipper upscale)
09:51:56watusimoto\should be easy enough
09:52:09raptorthis makes it upscale times 255,000
09:52:29raptorsince clipper only handles integer types, at that scale the same two points are already different points
09:52:59raptor5. after clipper, since the points were clipped as distinct points, send to poly2tri downscaled by 1000
09:53:16raptorwhich end up being very, very close floats
09:53:23raptorpoly2tri freaks out and crashes
09:53:29raptor^^ number 6
09:54:40raptorso the points going into #3 (saving to file using geometry->tolevelcode() or whatever) already has different floating point values between teh snapped points
09:55:45watusimotook, so we need a test to demonstrate 1-3; the rest may be fixed by kaen's exploration of boot::geometry. He thinks we can use our points as-is, without conversion, which will boost performance and fix the rounding issue you described
09:55:50raptoralso clipper is great! and doesn't have a 'fatal epsilon value' as you say, kaen, because it only uses integers
09:55:58watusimotoI do like clipper
09:56:27watusimotobut the integer/float issue is difficult to bridge, as long as we use floats
09:56:49raptorwell, i think it wouldn't be so bad if we didn't scale by 255 in the level file
09:57:03raptorand maybe truncate at 0.001
09:57:30raptorbut we can't truncate there with the gridsize scaling
09:57:50kaenI meant because of the float to int conversion, two sufficiently proximate float points become coincident when made into ints
09:57:55watusimotowe can at some point switch and start writing coordinates with a gridsize of 1
09:58:16watusimotochange the editor so the scaling is transparent
09:58:17raptorah ha! back the the Version 2 level code line idea!
09:58:28watusimotodoesn't even need to be version 2
09:58:40watusimotoif gridsize is 1, everything works as-is
09:58:55raptori'm thinking about supporting legacy files
09:59:08watusimotoyes -- those will have a non-1 gridsize
09:59:14watusimotoand will continue to work
09:59:33raptorah, ok, so check the gridsize and load if it's there/not 1
09:59:34watusimotowe might have a cue in the editor that if the gridsize is 1 to draw the underlying grid differently
09:59:44watusimotobut really, nothing would change
09:59:55raptorwell, I think gridsize is useful, but only as a client-side thing
10:00:03watusimotoyou could start creating levels with gridsize = 1 in the 018a editor, and everything would work
10:00:10watusimotowith no rounding
10:00:53raptorso those are my findings...
10:01:00watusimotogood work!
10:01:02raptormy 2 late-night ideas where:
10:01:18raptor1. somehow clone the point when doing the snapping to it, in the editor
10:02:08raptor2. on toLevelCode() truncate the floats at 0.001 before applying gridsize scaling (this might get like 90% - 100% of the rounding problems)
10:02:40raptoroh and 3. rewrite poly2tri to use integer points
10:02:52raptor(late-night)
10:03:16watusimotothe editor issue is curious... it should work. I can write a test to demonstrate the failure
10:03:36raptorI duplicated the issue three times
10:03:42raptorwill polywalls, at least
10:03:48watusimotoright, but the test would isolate the location of the problem
10:04:01raptorand I could not for the life of me see how the new vertex (when snapped) was written to
10:04:29raptori was thoroughly pretzelized in the editor code..
10:04:32watusimotoas for the int/float issue, I think we should see what kaen comes up with
10:04:55kaenit's going well so far after writing the adapter code
10:05:07kaenit'll operate directly on our Vector<Point> structures
10:05:16watusimotoand as for the gridsize issue, I think we can transition to a default 1 gridsize at any point... just have the editor detect that and render the grids differently. nothing else needs to change (I think)
10:05:34kaenalso, boost::polygon can in fact replace all three of the libs
10:05:48kaenalthough it does trapezoidization (?) rather than triangulation
10:06:22watusimotothat will probably work for our renderers if they are concave
10:07:24kaenwell, you can safely break a trapezoid into two triangles by slicing at the two nearest non-adjacent points
10:07:54kaenright?
10:08:26watusimotoprobably. I think trapezoids will work for us, though I don't know how you can trapazoidize a triangle
10:08:38kaenoh, these trapezoids are convex by definition
10:08:55watusimotobut there are shpaes that won't trapezoidize
10:09:04watusimotobut whatever.. problem for the boost guys
10:09:10kaenpresumably the triangular parts are trapezoids with two coincident points
10:09:14kaenanyway we'll see how it goes
10:12:51watusimotook out of here
10:12:59raptorgood 10 minutes!
10:13:03raptor:)
10:13:16kaenlater!
10:13:19raptorkaen: trapezoids??
10:13:35kaen shrugs
10:13:42kaenthat's what the boost guys decided
10:13:53kaenfunctionally equivalent for us, right?
10:14:10raptorI supposed... the end goal is to get convex polygons for zones
10:14:27kaenthey guarantee the trapezoids are convex
10:14:38kaenwhich apparently is part of the definition of a trapezoid
10:14:45raptorhaha, yeah
10:17:04watusimoto Quit (Ping timeout: 248 seconds)
10:17:18kaenshould be really fast because we won't have any data transformation
10:17:30raptorall that work I put into poly2tri and it still makes me angry...
10:17:39kaenyeah :/
10:17:46kaenyou did a really good job
10:17:48raptorI really thought it would be even slightly resilient...
10:17:54kaenpoly2tri just failed us
10:18:05kaenalright time to go grocery shopping, back soon
10:18:08raptorthere's gotta be a precision check somewhere...
10:18:12raptorok, later
10:18:58koda Quit (Ping timeout: 272 seconds)
10:23:12raptorwe're currently using boos 1.46.1 headers - looks like we should update
10:23:15raptor*boost
10:53:24Darrel has joined
11:12:15FruitCupsFruitCu has joined
11:12:33FruitCupsFruitCuFRUIT CUPS FRUIT CUPS FRUIT CUPS FRUIT CUPS FRUIT CUPS FRUIT CUPS
11:12:35FruitCupsFruitCuoh hi
11:12:59raptorYes.. a good cause to be passionate about
11:13:24FruitCupsFruitCuWhy do some people frequent the IRC and pretty much never play the game? lol
11:14:04raptorI am one of those 'some people'... I like to code the game :)
11:14:17FruitCupsFruitCuNah you play the game a fair amount.
11:14:21FruitCupsFruitCuMore than the rest of the devs, in fact!
11:14:28FruitCupsFruitCuThough kaen plays a fair bit too.
11:14:35raptorhmm... is this... Quartz?
11:14:41FruitCupsFruitCuYeah, it is.
11:14:44FruitCupsFruitCuI figured it was obvious, but...
11:15:08FruitCupsFruitCuGiven I was the guy who recently put "FRTUI CUPS FRUIT CUPS FRUIT CUPS ..." in his map :P
11:15:14raptorah yes..
11:15:23raptor(I'm a bit thick)
11:15:39FruitCupsFruitCuApparently you figured it out just from my way of typing and my opinions though so that's pretty neat.
11:16:03FruitCupsFruitCuNah I was talking more about Nothing_Much and HylianSavior
11:16:12FruitCupsFruitCuOh, and Darrel.
11:16:21HylianSaviorwho is what
11:16:24HylianSaviorhi
11:16:30FruitCupsFruitCuI keep seeing them here but I verrrrry rarely see them in-game, like wut
11:16:55HylianSaviori just got here man
11:16:58HylianSaviori don't know when people play
11:17:02HylianSaviorservers are always empty
11:17:35FruitCupsFruitCuWell tonight people are getting on at 10:00 PM EST apparently so
11:17:40FruitCupsFruitCuFor what that's worth.
11:17:40HylianSaviorcool
11:17:43HylianSaviori'll try and be there
11:17:49DarrelI personally am one of the students from GCI last year. #bitfighter is still in my favorites, so I join it by default.
11:18:08FruitCupsFruitCuOhh I see, I remember that thing. Been a while.
11:18:12HylianSaviorhaven't played bf in forever
11:18:44FruitCupsFruitCuWell, thanks for the answers. I was genuinely just curious, not trying to throw around accusations. While you guys are in the IRC more than the game, I'm making maps more than I play them, so ... ^^
11:19:03HylianSaviorbob found me on fb a few days ago and told me to come back
11:19:04HylianSaviorso
11:19:06HylianSaviorhere i am
11:19:45FruitCupsFruitCubob who now?
11:19:58FruitCupsFruitCuI mean are we talking bobdaduck, C. Bob, Bob ...
11:19:59FruitCupsFruitCuxD
11:20:04HylianSaviorbobdaduck
11:20:20FruitCupsFruitCuI could've sworn I recognized "HylianSavior." Edshin right?
11:20:21HylianSavioraren't the latter two long ngone
11:20:23HylianSavioryeah
11:20:30FruitCupsFruitCuhohoho shiiiiz
11:20:34FruitCupsFruitCuHey there buddy! :D
11:20:36HylianSavior:>
11:20:37HylianSaviorhi
11:20:55FruitCupsFruitCuBob is around. C. Bob yeah he's been gone for a long time
11:20:57HylianSaviori gotta go to class
11:20:57HylianSaviorc ya
11:21:00FruitCupsFruitCuPeace
11:25:47FruitCupsFruitCu...Although C. Bob popped his head in once recently. He went for the usual "anonymous" appearance and then made some stupid reference to his long he's been around ... and then bobdaduck and I figured out pretty fast he was C. Bob. He then proceeded to make snarky comments about how immature I am. You know, how immature I was 5 years ago. And apparently NOBODY ages and matures from the age 14 to 19, right? I must still be a
11:27:22DarrelOf course. Aging and maturing are just myths :)
11:27:58raptorwell maturing is, anyways...
11:28:31Darrellol
11:28:36raptor:)
11:29:39Darrelby the way guys, speaking of GCI earlier... do you plan on trying to participate again? doesn't matter if alone or under an "umbrella" like last year
11:29:50raptorI'm not sure
11:30:05raptorit would be fun to get another T-shirt...
11:31:02FruitCupsFruitCuHah! ^
11:31:30raptorwhat?
11:31:44raptormy wife stole it anyways...
11:31:59Darrel:))
11:32:07FruitCupsFruitCuSteal it back. Like, as she's wearing it. ;)
11:32:17Darrel-.-
11:33:11raptorDarrel GCI is coming quick! We've got like 3 weeks to decide to register!
11:33:34DarrelYep, they announced it with GSoC like a week ago, if I'm not mistaken
11:34:01DarrelWhich is a surprise - surprisingly late announcement of GCI, but an incredibly early announcement of GSoC 2014
11:35:17raptorI ownder if they're phasing it out
11:35:20raptor*wonder
11:36:08Darrelyeah.. also it starts earlier and ends earlier
11:37:40DarrelThey also (finally) made it easier to switch between program websites..
11:38:01DarrelIt was crazy when you had to click the small text in the upper right corner :)
11:43:55Nothing_MuchFruitCupsFruitCu, I discovered this game a while ago, and I'd love to see more players on it, so I'm making a trailer for it :D
11:44:09FruitCupsFruitCuNiice, can't wait.
11:45:05LordDVG has joined
11:47:42Nothing_MuchFruitCupsFruitCu, What I need though is some actual "frenetic" gameplay which I'm hoping you or anybody else can record :)
11:47:58FruitCupsFruitCuI'll keep that in mind.
11:48:48raptorwe have a group play tonight, right?
11:48:48Nothing_MuchMy own gameplay footage isn't very "frenetic" enough, so hopefully some of you guys can chip in.
11:49:20raptor10pm EST (if today is wednesday)
11:49:35Nothing_MuchYep, it's Wednesday
11:50:40FruitCupsFruitCuYeppers
11:50:42kaenall day long!
11:50:47FruitCupsFruitCukaaaaaaaen!
11:51:00kaenwhat up fruit cup?
11:51:03FruitCupsFruitCusup
11:51:07FruitCupsFruitCuthat rhymes
11:51:07FruitCupsFruitCugg
11:51:09kaen(I did it because it rhymed)
11:51:37FruitCupsFruitCuSo how do I actually register a name in this thing?
11:51:43FruitCupsFruitCuAlso guys, tell me what I should go by.
11:51:45raptorbetter movie: "the wrath of khaaaaaan" or "into dorkness" ?
11:51:49FruitCupsFruitCuObviously it should incorporate "Quartz" in some way.
11:52:25ShadowXloner has joined
11:53:02ShadowXlonerHoly Crap.
11:53:07ShadowXloner019 is smooooooth.
11:53:15FruitCupsFruitCuhow do u has 019 >:(
11:53:25ShadowXloner:P
11:53:31ShadowXlonerBeta testing.
11:53:35Nothing_MuchOh wait, was 019 released yet?
11:53:39ShadowXlonerNope.
11:53:44ShadowXlonerJust beta testing.
11:53:58ShadowXlonerThat and I wanted to help you acquire some good game footage.
11:53:59FruitCupsFruitCu Quit (Quit: Page closed)
11:54:20Nothing_MuchHmm.. yeah, if there's a possible Beta testing group play that'd be perfect, since this is a big update and the trailer is.. uh.. highlighting the big release.
11:54:24Quartzy has joined
11:54:24ShadowXlonerSince you've mentioned that you haven't had any footage of actuall people playing yet.
11:54:38Quartzyguhh
11:54:38kaenyeah we should totally do a beta playdate
11:54:38Nothing_MuchYeah, only bots
11:54:42ShadowXlonerHey, Quartz.
11:54:47QuartzyIs there really no way to register a name on here?
11:54:58ShadowXlonerNot that I can see.
11:55:04kaenQuartzy, http://freenode.net/faq.shtml#contents-userregistration
11:55:06Nothing_MuchQuartzy, On Freenode?
11:55:11Nothing_MuchYeah it's possible
11:55:22kaenit's kind of a PITA but I'm here to help
11:55:28ShadowXlonerAlso, guys, I'm freaking out. For some reason I've been accepted into Phi Theta Kappa at my college.
11:55:39Nothing_MuchNo clue what that is
11:55:44ShadowXlonerWell, I'm Qualified and have been invited.
11:55:59ShadowXlonerApparently it's the top tier for students.
11:56:05kaenit's an honors society, Nothing_Much
11:56:09kaenafaik
11:56:16ShadowXlonerIS SCURY>
11:56:23QuartzyThanks kaen
11:56:24Nothing_MuchWow, honors, that sounds good
11:56:28ShadowXlonerI don't even know how I got qualified.
11:56:31kaencongratulations ShadowXloner!
11:56:33ShadowXlonerI'm just a slacker ;~;
11:56:35kaenthat's really great
11:56:49kaenfunny, all the honors students I knew said that too :)
11:56:49Quartzylol grats.
11:56:51Nothing_MuchIf you're a slacker and got honors, then that's even more impressive lol
11:56:56Quartzy*Sigh* College is overrated.
11:56:58ShadowXlonerkjjiqgaga
11:57:10ShadowXlonerI don't know how to cope with this.
11:57:18Nothing_MuchCollege is alright, I'll be heading back to it this January
11:57:42ShadowXlonerAlso, has the music been redone for 019?
11:57:53ShadowXlonerIt sounds a little more Digital Droo -ish.
11:58:08raptori don't think so...
11:58:08ShadowXlonerKind of reminds me of the music for Active Lancer.
11:58:16raptorat least i don't remember changing anything
11:58:20QuartzyPlan is for me to go back to college next fall lol. Gonna be a while.
11:58:21ShadowXlonerWoo, old school Mac Addict games.
11:59:16Nothing_MuchShadowXloner, amgine and myself were listening to some good songs which could be merged, so far I have two used in the trailer, one for the gameplay and another for the level editor.
11:59:31ShadowXlonerOkie dokie.
12:00:12QuartzyLol I'm musical. Now if only Bitfighter called for hard rock rather than like, elevator house techno :P
12:00:31raptorelevator house techno... that's seems like a perfect description
12:00:53raptoronly a few of the songs I think are good for gameplay...
12:01:08ShadowXlonerHahah.
12:01:09Nothing_MuchI like Kick the Beat the best
12:01:32ShadowXlonerI just realized, voice chat is still a thing. Is anything going to come of that?
12:01:33QuartzyI like the music so much that I turn it off.
12:01:49QuartzyHah, hopefully. I'd really love to play this game with a mic with other peeps, really...
12:02:03Nothing_MuchQuartzy, That's perfect, if you'd like to record footage, you can - Oh you like microphones too?
12:02:06QuartzyMaybe not though. I'd probably curse even more than I do.
12:02:28QuartzyYeah, I'll be sure to record some footage at some point as long as I remember.
12:02:46Nothing_MuchAll I need is the sound effects recorded in plain fullscreen without music
12:02:56LordDVG Quit (Remote host closed the connection)
12:03:36ShadowXlonerHere's the guy I was talking about earlier. The in-game music sounds similar to his works http://www.digitaldroo.com.
12:03:58QuartzyThat's exactly how I play soooo coo!
12:04:52Nothing_MuchDigital Droo sounds pretty nice
12:05:08QuartzyI'm too busy listening to Tool to listen ...
12:05:12QuartzyOkay, maybe a bit later. xD
12:05:23ShadowXloner"Friendly turrets are mostly harmless"
12:05:29ShadowXloner"Mostly Harmless"
12:05:37ShadowXlonerI C WAT U DID THAR.
12:06:11ShadowXlonerSo Friendly turrets are basically the equivalent of Earth?
12:06:47QuartzyThat reminds me of a post in the "exploits" thread
12:07:06Quartzybobdaduck: oh gosh in the tutorial bobdaduck: "friendly turrets are mostly harmless" bobdaduck: /ragequits
12:09:45ShadowXlonerI love Douglas Adam's references.
12:13:09kaenI'm not sure that one was intentional
12:13:21ShadowXlonerMany references aren't.
12:14:13kaenI remember there was a commit that added that word "mostly"
12:14:33kaenback when wat was writing the tutorial code
12:17:15Skybax has joined
12:17:24SkybaxGoodmorning
12:17:28ShadowXlonerMorning Bax.
12:17:30kaenmorning!
12:17:47Watusimoto has joined
12:17:51SkybaxHow's everyone todai
12:18:09ShadowXlonerPretty good.
12:18:14ShadowXlonerA bit shaken.
12:18:21SkybaxWhy?
12:19:08ShadowXlonerI've qualified and been invited to join an Student's Honor society.
12:19:11Nothing_Muchbrb
12:19:13kaenShadowXloner, don't sweat it bud
12:19:15SkybaxNice!
12:19:15ShadowXlonerPhi Theta Kappe.
12:19:16kaenall you can do is try
12:19:21ShadowXlonerI guess.
12:19:30ShadowXlonerIt's just something I was never expecting to happen.
12:19:34kaeneven if you don't make it, you've still done better than 99% of everyone else who goes to college
12:19:41kaenjust by getting invited
12:19:49ShadowXlonerTrue enough.
12:20:10Watusimotowhat's to sweat?
12:20:18kaenthat's what I'm saying!
12:21:16Quartzy^
12:24:22SkybaxI think I'm gonna work on my BitTown house today
12:24:46ShadowXlonerOh yeah.
12:24:47kaenoh Quartzy
12:24:49kaendude
12:24:52QuartzySup.
12:24:54kaenI have an idea for the new BitTown
12:25:03QuartzyOh okay I thought you were having a moment of passion with me for a second there...
12:25:04kaenlike, I know how to automate adding people's houses
12:25:28QuartzySo I have heard.
12:25:34kaen[21:36:05] <kaen> have people make their houses on maps, upload them, have a supervisor script with some !command to submit your house, extract the structure and save it to a file, then load and aggregate all of those files when bittown proper is loaded
12:25:48QuartzyNice.
12:25:53QuartzyI made a house yesterday :P
12:25:56kaenit'd only work in 019 though
12:28:17ShadowXlonerI should probably get to work reconstructing my tower.
12:28:20Watusimotoraptor: with the steps 1-3 problem you describe earlier, did you try it with a point item, or only a wall vertex?
12:28:33Quartzyhahaha Shadow nice
12:28:44QuartzyI bet everyone's houses will be intelligible compared to mine.
12:29:15ShadowXlonerYours will most likely be the most intricate I'd think.
12:29:23QuartzyProbably, yes.
12:29:35QuartzyI certainly have my niche carved out don't I? XD
12:30:18ShadowXlonerOh, after that video? You OWN it.
12:31:09QuartzyThanks bud!
12:32:31QuartzyThat map is done, too
12:32:37QuartzyIt's in my next pack. Which I literally just wrapped up...
12:32:40Quartzy28 maps
12:32:50ShadowXlonerDamn.
12:32:55ShadowXlonerThat's a lot of maps.
12:32:59QuartzyIt's been fun ... time to send this pack on its way lol
12:33:10QuartzyCrazy thing? I still have a good 8 maps or so that are incomplete
12:33:20Quartzy3 of which are waiting on 019 features. Although I think I mentioned that to you earlier.
12:33:35Skybax Quit (Read error: Connection reset by peer)
12:34:01ShadowXlonerYep.
12:35:04QuartzyWell cool. I'm going to be writing up the uhh ... write up ... and then post the level pack. Woot!
12:36:19ShadowXlonerSweet!
12:39:01Watusimotoraptor: I can reproduce but only with polywalls -- not regular walls or point items
12:42:20QuartzyWhoa, sounds like somebody has the hots for polywalls!
12:44:05raptorhi Watusimoto
12:44:10raptori did it with polywalls
12:44:13raptormy tests
12:44:30raptorI (incorrectly) assumed it work with other polygon geometry, too
12:47:05QuartzyBam, it's up.
12:47:07Quartzyw00t
12:48:18Skybax has joined
12:51:25QuartzyFYI ShadowXloner the map in the video is "Inca"
12:51:37ShadowXlonerOkie Dokie.
12:51:48ShadowXlonerSo that's where we'll want to record the gameplay?
12:51:55Quartzyhahaha what? If you want xD
12:52:20QuartzySo people can be like "much gameplay. such arts."
12:53:01QuartzyLol I just figured you were curious more than anything. If gameplay recording happens there, then bonus.
12:53:54QuartzyPeace
12:53:57Quartzy Quit (Quit: Page closed)
12:56:21Skybax_ has joined
12:56:53Skybax Quit (Ping timeout: 245 seconds)
12:56:58Nothing_MuchHi again, what'd I miss?
12:57:32ShadowXlonerNot too much.
12:57:40Nothing_MuchAh
12:57:41ShadowXlonerBut I'm out for the time being later.
12:57:50Skybax_Bai lol
12:57:56Nothing_MuchLater
12:57:59raptorhi
12:58:00ShadowXloner Quit (Quit: ShadowXloner)
12:58:09Skybax_xD
13:00:11Watusimotoback later
13:03:00raptorkaen: are you using the newest boost? seems like lots of stuff has been fixed since the first inclusion of Boost.Geometry (in 1.47)
13:03:17raptormaybe I should update our boost headers..
13:04:35Watusimoto Quit (Ping timeout: 240 seconds)
13:05:06Flynnn has joined
13:16:06Flynnn Quit (Quit: This computer has gone to sleep)
13:17:52kaen Quit (Ping timeout: 248 seconds)
13:19:16Skybax_raptor, thanks for directing me to sams level archives, I found quite a few in there
13:20:18raptoroh good
13:20:22raptoryou're welcome
13:22:00Skybax_I actually found three different versions of my "Castles" level >.> I'm gonna go look and see what's different lol
13:24:21Skybax_That's odd.. the code is different but the levels are the same lol
13:24:37Skybax_Well I guess I'll just delete two of them
13:49:47Watusimoto has joined
14:02:38Flynnn has joined
14:06:05Invisible1 has joined
14:14:52raptoremscripten!
14:19:26Skybax has joined
14:21:23Skybax_ Quit (Ping timeout: 240 seconds)
14:36:23Darrel Quit (Quit: IRC client killed.)
14:41:37Skybaxraptor, what is the point of the "Zones"?
14:41:49raptortriggers for Lua scripting!
14:42:03SkybaxAh.. I thought they were decorative lol
14:43:01SkybaxHmm
14:46:47kaen has joined
14:48:01kaenraptor, yes I am using the latest boost
14:48:43SkybaxHeheheh.. fixed it
14:58:02thread_ has joined
15:03:39thread_Hello World!
15:06:58Flynnn Quit (Quit: This computer has gone to sleep)
15:07:25kaenint main() { printf("Hello, %s\n", "thread_"); return 0; }
15:07:43kaen^ the reason C is a dying language
15:08:11thread_:)
15:09:30raptor Quit ()
15:09:56thread_kaen: is there a beta build of 019 I can get my hands on?
15:10:18kaenlemme try to make you one again
15:10:33kaenit's non-trivial because I have to cross-compile it (on linux)
15:11:08kaenI got all the way to linking last night but I get some undefined reference errors
15:11:08thread_I see
15:11:32kaenwhich means I just need to fix the last 0.1%
15:11:39thread_The build I've been testing on has some bugs that I don't want to report unless I know they are still current
15:13:42kaenok
15:13:52kaenthanks again for doing thorough testing, seriously
15:14:03kaenas devs we sort of get tunnel vision regarding that sort of thing
15:14:27thread_uhhh, well yea. I know what you mean
15:14:52thread_I WILL TEST THE DICKENS OUT OF THE LEVELGENS!
15:15:00kaengood
15:15:11kaenbecause it is almost certain that I broke _something_
15:15:15kaenI just don't know what yet
15:15:38thread_by the by, I did make a portalgun in the 019 levelgen api. It's pretty fun
15:16:16thread_*with the 019 levelgen api
15:16:44kaenwhoa
15:16:55kaennow that sounds awesome
15:18:54thread_I want to make sure it doesn't bug out, then I will distribute it
15:20:56thread_Though I guess I have to wait for the general release of 019 as well
15:26:51kaenah yes, STL's templatized containers
15:27:04kaenauthor of such classic error messages as
15:27:04kaen/usr/include/c++/4.7/bits/stl_vector.h:1003:7: note: void std::vector<_Tp, _Alloc>::insert(std::vector<_Tp, _Alloc>::iterator, std::vector<_Tp, _Alloc>::size_type, const value_type&) [with _Tp = TNL::JournalEntryRecord*; _Alloc = std::allocator<TNL::JournalEntryRecord*>; std::vector<_Tp, _Alloc>::iterator = __gnu_cxx::__normal_iterator<TNL::JournalEntryRecord**, std::vector<TNL::JournalEntryRecord*, std::allocator<TNL::JournalEntryRecord*> > >; typena
15:27:05kaenme std::_Vector_base<_Tp, _Alloc>::pointer = TNL::JournalEntryRecord**; std::vector<_Tp, _Alloc>::size_type = long unsigned int; std::vector<_Tp, _Alloc>::value_type = TNL::JournalEntryRecord*]
15:27:55kaenconceptually simple, easy to use, and above all informative
15:28:18kaenby which I mean none of those things.
15:30:04kaeneven more so when you see the full 200-line error message it vomits up
15:31:55thread_:|
15:33:55kaenmakes perfect sense, right?
15:36:26kaen79% done, thread_
15:36:37thread_\o/
15:40:40Watusimotothat's a pretty horrid error message
15:41:10kaenall the g++ STL error messages are like that
15:41:17kaenI'm switching to clang for development from now on
15:41:51Watusimotoare the messages any better? vc++ looks like that too
15:42:53kaenyeah, they compress all of that gibberish into "type diffing"
15:43:17kaent.cc:4:5: note: candidate function not viable: no known conversion from 'vector<map<[...], float>>' to 'vector<map<[...], double>>' for 1st argument;
15:43:19kaenturns into
15:43:26kaencandidate function not viable: no known conversion for 1st argument;
15:43:26kaenvector<
15:43:26kaenmap<
15:43:27kaen[...],
15:43:27kaen[float != double]>>
15:45:20kaennote the gorgeous line breaks to make the error messages human-readable
15:46:09kaenit also helps avoid the haphazard line-wrapping you get from g++
15:51:38kaenthe first error message I posted above as presented by clang:
15:51:39kaen/usr/include/c++/4.6/bits/stl_vector.h:943:7: note: candidate function not viable: no known conversion from 'S32' (aka 'int') to 'iterator' (aka
15:51:39kaen'__normal_iterator<pointer, std::vector<TNL::JournalEntryRecord *, std::allocator<TNL::JournalEntryRecord *> > >') for 1st argument;
15:51:39kaeninsert(iterator __position, size_type __n, const value_type& __x)
15:52:03kaenon the console it also has color-coded output
15:54:02koda has joined
15:55:48kaensorry thread_, on my third attempt to rebuild...
15:55:52kaenthink I've got it this time though
15:55:58thread_hehe
15:56:12Invisible1 Quit (Ping timeout: 260 seconds)
15:56:30kaenoh wait
15:56:33kaenapparently it worked
15:56:52kaenbitfighter.exe: PE32 executable (console) Intel 80386, for MS Windows
15:56:56kaenwell, let's see
15:58:46kaenthread_, you already have a 019 folder right?
15:58:46kaenwith all the .dll's and junk?
15:58:46kaenWatusimoto, have we updated an of the dll's recently?
15:58:46kaenany*
15:59:15thread_yea
15:59:18Watusimotoyes, the sdl dll
15:59:32Watusimotobut it may not be a critical update
15:59:45kaenit's probably ABI incompatible though
15:59:54Watusimotooh, yes, I suppose it is
16:01:03kaennbd, I'll just add that one
16:01:52kaencool, the exe works
16:02:27kaen20% uploaded...
16:02:50kaenso glad I finally got this to work
16:03:24kaenthere's one known bug in my builds: no game clock
16:03:25kaensorry
16:03:34koda Quit (Quit: K Thx Bai)
16:06:02thread_"no game clocks" meaning what?
16:06:06koda has joined
16:07:28thread_welcome back koda!
16:10:33kaenthe clock in the lower right
16:10:35kaenyou won't see it
16:10:42kaenthread_, http://www.filedropper.com/bf_1
16:11:38kaenwow these compiler errors are beautiful
16:11:46kaenand decipherable
16:11:57kaenand vibrantly colored!
16:15:27thread_someone is enjoying their error messages
16:16:20Nothing_MuchHello everyone
16:16:45thread_Hello Nothing!
16:17:27Nothing_MuchWhat's up?
16:17:35Nothing_MuchOh thread_, thanks very much for the editor footage!
16:18:02Nothing_MuchI got it covered in the trailer, with the windows stuff cropped off.
16:21:31thread_good :) hopefully you don't show the error popup at the end :p
16:24:23Nothing_Muchthread_, I got everything co- well, that's a bad way to put it, what I mean is that I edited most of it and sped the whole process up by almost 1000%
16:24:56Nothing_Muchand the error popup is not seen
16:24:58Nothing_Much:)
16:25:10thread_yay!
16:25:17thread_can't wait to see my claim to fame
16:25:37thread_do you need any more footage?
16:26:01koda Quit (Quit: K Thx Bai)
16:30:36Nothing_MuchWell, if you can record in fullscreen mode for gameplay, that'd be good :) but you'd have to turn off the music.
16:30:51Nothing_Much*not fullscreen stretched
16:31:17thread_right... i'll see how camstudio handles the larger area
16:32:10Nothing_MuchWhat are your computer specs?
16:33:33thread_I don't remember. Its about 1.5 years old
16:34:14thread_At this point, I just say "it works"
16:35:25Nothing_Muchthread_, Ah, your computer should be able to handle it well then
16:35:31Nothing_MuchYou bought it new 1.5 years old?
16:35:44thread_I bought it 1.5 years ago
16:40:26Nothing_MuchUh.. what brand? Dell? HP?
16:46:00thread_I built it myself. not really branded
16:47:21thread_Anyway, I'll try to record gameplay in fullscreen
16:48:43ShadowXloner has joined
16:49:00ShadowXlonerI'm back!
16:49:33thread_hey
16:49:49ShadowXlonerWhat's up?
16:50:14thread_What's up? I'd have to say... Nothing_Much!
16:50:18thread_:D
16:50:26kaenoh, just rewriting all of Bitfighter's core geometry routines
16:50:30kaenNO BIG DEAL
16:51:16ShadowXlonerI apologize, I don't immediately see the profundity of that task.
16:51:34ShadowXlonerNon-programmer and what not.
16:51:38Nothing_Muchthread_, How can you not know the specs of your own pre-built computer?!
16:52:01Nothing_MuchShadowXloner, It's not a problem, I'm no programmer either. :P
16:52:05thread_Like I said... it was 1.5 years ago. I could have told you back then
16:52:20thread_I'd have to open it up to tell you parts it was
16:52:23Nothing_MuchGo to your control panel and have a look-see if ya'd like.
16:52:28ShadowXlonerYeah, but I don't want Kaen to feel like his efforts are falling upon deaf ears.
16:52:39thread_I'm not actually on my computer right now... school machine
16:52:39ShadowXlonerWhat he does is very important and significant for the game.
16:52:52kaeneh, I'm making out to be harder than it is
16:53:03thread_kaen is just being cool
16:53:05kaenthe most difficult part is overcoming the gaps in my own knowledge :)
16:53:09ShadowXlonerI like knowing the full context so I can fully appreciate what Bitfighter's Developers do for us.
16:53:11Nothing_MuchThat's nice ShadowXloner, but the fact that we play the game should leave an impact!
16:53:18kaen^
16:53:27ShadowXlonerFair enough c:
16:54:21kaenI wouldn't be rewriting anything at all if you guys weren't here to play with the results :)
16:54:49thread_Everything is a feature until we label it a bug
16:55:11kaennow that's profound
16:56:36thread_yess...! score for thread!
17:20:02thread_On another note, ShadowXloner, my little secret project is coming along great
17:22:25thread_My problem is that I want to build hype about it without saying what it is
17:29:18raptor has joined
17:29:18ChanServ sets mode +o
17:31:35thread_ has left #bitfighter
17:33:47raptorhello
17:35:17raptorkaen: I'm curious as to how you're going to handle the holes-in-holes etc problems with triangulation..
17:35:33raptorwith Boost.Geom
17:36:02kaenI'm almost do the point where I can test whether that's handled automatically
17:36:06kaenmy guess is that it will be
17:36:34kaenalso the "bot zone buffer" code is super simplified
17:36:35raptorI had no idea that Boost.Geometry would be as advances as it is
17:36:37raptor*advanceed
17:36:45kaendude it's so high-quality
17:37:07raptorwhy on earth did it never turn up when we did the initial study of doing botzones?
17:37:13kaenafter I plug in the object-derived polygons I get a connectivity map in one line of code
17:37:19kaenit was released in 2009
17:37:57raptorso just not around long enough?
17:38:19kaenthat's my guess
17:38:19kaenconnectivity_extraction(get_trapezoids(bounds - holes)).extract()
17:38:37raptoroh interesting - clipper uses the Boost license..
17:38:39kaenall operating on our native Vector<Point> structures
17:39:23kaenno data transforms, no precision loss, no piping between libraries, sane exception raising for invalid input, etc
17:39:34kaenbuildBotZones is reduced to about 15 lines
17:39:40raptorha!
17:39:44kaenyeah
17:39:47raptorman what monster we have now...
17:39:59kaenall with the commercial-level performance and numerical stability you'd expect from intel
17:40:05kaenI think this is our best attempt yet
17:40:24raptorwow, found this: http://rogue-modron.blogspot.com/2011/04/polygon-clipping-wrapper-benchmark.html
17:40:35raptorsays Boost.Geometry is faster than everything except clipper
17:40:41raptor(for clipping)
17:42:46kaenI think we'll see even better results in our case because of the data transforms in the old system
17:44:26kaennot to be a broken record, but that part was always pretty smelly to me
17:45:23raptorI suspect you're right Point (float) -> IntPoint (long) -> Point* (double) is so messy
17:49:19kaennot to mention vector -> polytree -> rcPolyMesh -> vector
17:49:36kaenit could save RAM paging too because most of the operations are done in-place
17:49:58kaenespecially with our new 16,000 vert inputs :)
17:50:26raptori hope boost.geometry is as robust as clipper (i'm sure it'll be more so than poly2tri)
17:58:17Watusimotowell, it sounds pretty awesome to me
17:58:31BFLogBot Commit: 7216ea48fa55 | Author: watusimoto | Message: Implement spinner
17:58:33BFLogBot Commit: 54d150a56b2b | Author: watusimoto | Message: Start spiffing up message boxes, add spinner to high scores loading message
17:58:34BFLogBot Commit: 82951459c462 | Author: watusimoto | Message: Merge
17:58:36BFLogBot Commit: 22099c30a5f9 | Author: watusimoto | Message: Whitespace
17:58:37BFLogBot Commit: c8dc6b1c2871 | Author: watusimoto | Message: More polish for error messages... looking pretty! Messages can now be auto-wrapped to fill horizontal space if we change fonts or whatnot. Doing that was easier than manually reformatting when I changed the font from Roman.
17:58:41Watusimotoand there is my endo-of evening push
17:58:45Watusimotosigning off
18:01:34raptorso with boost - I don't mind using it
18:01:48raptorbut I do dislike the never-ending header includes
18:01:58raptorand how it so drastically slows down compile
18:02:23raptorspinner!?
18:02:40ShadowXloner Quit (Ping timeout: 260 seconds)
18:05:23Watusimotospinner... |/-\| animation
18:05:50Watusimotofor when we're loading the high scores and it appears nothing is happening
18:05:56Watusimotoui element, not game element
18:06:04Watusimotogood night!
18:10:48Watusimoto Quit (Ping timeout: 268 seconds)
18:13:55kaenmaybe we can use it for the upload/download stuff too
18:22:04raptorwell it looks ugly
18:23:42Flynnn has joined
18:28:30kaenman, clang is so awesome
18:28:30kaennever switching back
18:32:46raptorcan we compile with clang without modification?
18:33:48kaenyep
18:33:55kaenCC=clang CXX=clang++ cmake ..
18:34:12kaenyou have to rm CMakeCache.txt and CMakeFiles in the build dir
18:34:38kaenbest part is clang is command-line compatible with g++
18:34:43kaener, gcc
18:34:50raptoroh interesting...
18:35:03kaenso our special flag setup works fine
18:35:07raptorseems like gcc is best for release, clang for debug
18:35:14kaenyes
18:35:17raptorbecause gcc can make fast code..
18:35:22kaengcc produces faster code still
18:35:24kaenindeed
18:35:57raptorCC=clang CXX=clang++ cmake -DCMAKE_BUILD_TYPE=Debug ..
18:36:03kaenyep
18:36:13raptorok here goes!
18:36:35raptornoisy! but readable!
18:38:22kaenit's been a life-saver in the TNL::Vector rewrite
18:38:36kaenand also in writing the adapter code for boost::polygon
18:40:20raptoroh so you did do the rewrite? how did you handle Vector<bool>?
18:54:07Flynnn Quit (Quit: This computer has gone to sleep)
19:08:39ShadowXloner has joined
19:14:19Nothing_MuchWhen's the playdate and is it for the Beta?
19:17:12kaenI just left it as an explicit instantiation
19:17:54kaenit's 40 minutes from now, and it's for 018a Nothing_Much
19:18:33kaenhopefully sky_lark doesn't miss his own party :P
19:18:40Nothing_MuchAh
19:18:48kaenoh, and don't forget to fill out his survey guys
19:19:02kaenmost of the parties that fail do so because of conflicting schedules
19:19:13Nothing_MuchAlright, then if there's gonna be recording, then I can figure out how to blend the old to the new version..
19:20:03Nothing_MuchWhere's the survey?
19:20:20kaenhttps://docs.google.com/forms/d/1H41gnYzS_I2w3I5OXSU6aMg3zZRTz_m9nAZ0Uc70k0k/viewform#start=openform
19:21:39Nothing_MuchHey, good survey!
19:26:34Nothing_MuchI'm curious if anybody has the ability to close down my BF account?
19:26:55Nothing_MuchSpecifically, my "NothingMuchHereToSay" account
19:27:19Nothing_MuchThat one's a bit long, so I made a new account with my freenode name.
19:29:36raptorNothing_Much: I can change your old account name for you
19:29:44raptorand you keep all of your posts
19:29:46raptorwant that?
19:30:30Nothing_MuchOh dear, uh.. sure, but then you'd have to delete the new account I just made, sorry 'bout that.
19:30:39raptoreasy to do
19:30:44raptoris that what you want?
19:30:56raptortell me your exact forum name you want, case-sensitive?
19:31:03Nothing_MuchYeah, change it to Nothing_Much (case sensitive)
19:31:08raptorokey doke
19:33:13Nothing_MuchThanks!
19:33:18raptorok Nothing_Much, try to join the forums with your new username (using old password) and also verify inside of bitfighter
19:33:20kaenman
19:33:20kaensniper rifles?
19:33:24kaensniper rifles?!
19:33:29kaenI thought they were trolling me
19:33:30raptorbitquakefortress
19:33:35kaenbut then they kept on posting...
19:34:22kaenanyway, the kibosh has been officially put on.
19:34:24Nothing_MuchWorks fine raptor :)
19:34:31Nothing_Muchkaen, wuchu talkin' about?
19:34:45kaenso sayeth the purple username
19:35:29kaenNothing_Much, http://bitfighter.org/forums/viewtopic.php?p=20622#p20622
19:35:34kaenfrom there on
19:35:54raptorheh
19:36:02kaennot that it's a bad idea in general, just that it's not really bitfighter-friendly
19:36:13kaenfor technical, conceptual, and stylistic reasons
19:38:45raptorI'm updating phpbb, nobody freak out...
19:38:55raptormaybe I should wait until later...
19:41:48raptoror maybe not!
19:45:11kaenupgrading a core component straight on a live production server without testing...
19:45:13kaendo it!
19:45:45kaenoh hey, uh, what version are you updating to raptor ?
19:45:54raptorthe ne
19:45:58raptorxt patch release
19:46:00raptor3.0.12
19:46:36fordcars has joined
19:46:59kaenoh okay
19:47:00fordcarsAre you releasing 019??
19:47:20kaenyes
19:47:23kaeneventually
19:47:26fordcars:P
19:47:32kaen:)
19:47:36fordcarssorry,
19:47:48fordcarsi saw bitfighter forums were down
19:47:56fordcarsthan checked irc logs :P
19:48:17raptorhey - maybe this is a great way to get the players to congregate in the channel!
19:48:27kaenhaha
19:48:37kaenor maybe we'll get some chicken littles
19:48:44kaen"THAT'S IT BITFIGHTER IS OVER"
19:50:41fordcarshah
19:55:07raptordone! with 2 min to spare!
20:08:21koda has joined
20:08:25koda Quit (Client Quit)
20:15:04raptorkoda!
20:15:59BFLogBot Commit: 89041a7c2b34 | Author: buckyballreaction | Message: Update to latest stb_truetype
20:16:02raptor!bugs
20:16:02BFLogBotTo enter a bug: http://tinyurl.com/bfnewbug -- To view all bugs: http://tinyurl.com/bfbugs
20:23:50Flynnn has joined
20:34:33kaenFlynnn, HylianSavior, play party tonight
20:34:35kaenright now
20:34:45kaenif you're interested
20:35:11HylianSavioroh
20:35:12HylianSaviorkay
20:43:34Nothing_Muchyikes
20:47:26fordcarsRaptor?
20:47:29fordcarsraptor
20:47:52fordcarsWhat did you use to make your Lan server Wan in Globulation2?
20:50:22raptorhi
20:50:38fordcarsHi
20:50:39raptorfordcars: I just opened ports in my router and directed them to my computer's IP
20:50:44fordcarsReally?
20:50:46fordcarsOh ok
20:50:50raptoryeah, nothing special
20:51:00fordcarsThanks!
20:51:10raptorbut if we used hamachi or remobo, we would have been on a 'proper' LAN and maybe it would have worked
20:51:26fordcarsif you use hamachi
20:51:33fordcarsDo other people need hamachi too?
20:51:59Nothing_Muchfordcars, afaik, my bro and his friends use hamachi
20:52:03Nothing_Muchif not then they can't join his server
20:52:04raptorfordcars: yes
20:52:07fordcarsokok
20:52:08Nothing_Muchbut that's all I know
20:52:11raptorthey all need the same client
20:52:11Nothing_Muchah okay
21:12:40fordcarsAnybody want to test something with me?
21:12:44Nothing_Muchfordcars, Sure
21:13:16fordcarsCrap it only works on Windows
21:13:24fordcarsLAN Bridge
21:13:25fordcarsgr
21:13:28Nothing_MuchLAN bridge?
21:13:32fordcars*LAN Bridger
21:13:33Nothing_MuchI'm sure I could try that on Linux
21:13:35Nothing_MuchOh
21:13:38Nothing_MuchBridger?
21:13:47fordcars3rd party program
21:13:50fordcars:/
21:13:53Nothing_MuchWait
21:13:56fordcarsJust a VPN
21:13:56Nothing_MuchIt's a VPN?
21:14:00fordcarsYep
21:14:01Nothing_MuchHmm..
21:14:07Nothing_MuchI think I should consider doing that
21:14:11fordcars?
21:14:18Nothing_MuchDo you know how to work with a VPN of any sort?
21:14:28fordcarsheh, uh maybe?
21:14:36fordcarsI am hosting one right now
21:14:43Nothing_MuchOh cool
21:14:45Flynnn Quit (Quit: This computer has gone to sleep)
21:14:49Nothing_MuchIs it using that bridger thing?
21:14:54fordcarsYeah
21:15:16fordcarshttp://pertino.com/?utm_source=google&utm_medium=CPC&utm_campaign=Competition&utm_cpckeyword=hamachi%20vpn&utm_cpcadgroup=Hamachi&utm_content=pertino-com&utm_program=AdWords-OnGoing&utm_sfcid=701d0000000ccOf&gclid=CP6ktIf0nLoCFQie4AodqnoAHA
21:15:19fordcarsTHIS
21:15:34Nothing_MuchConsider this: http://openvpn.net/index.php/open-source/downloads.html
21:16:11fordcarsIt works in linux?
21:16:30Nothing_MuchopenVPN yeah
21:18:09fordcarscoool
21:18:30Nothing_MuchIt's in the Ubuntu repo, so yeah
21:18:49Nothing_MuchAlso I think Compiz has been improving on Bitfighter's Windowed mode
21:21:06Flynnn has joined
21:27:03fordcars Quit (Ping timeout: 250 seconds)
21:37:19BFLogBot Commit: 2cc39238aeac | Author: buckyballreaction | Message: Remove now-unused file
21:37:20BFLogBot Commit: d3fcfd093538 | Author: buckyballreaction | Message: Build tooltip on first load if no one is online
21:37:22BFLogBot Commit: 43fa6233d198 | Author: buckyballreaction | Message: Fix crash with loading JSON if it had control characters in it
21:38:28raptorone bug left on the bug list!
21:38:33kaen\o/
21:38:48kaenI just had my first topic-splitting fiasco
21:38:49raptoroh... repair teleporter, right..
21:38:55raptorhooray!
21:39:19kaenonly took me four tries to get it right :P
21:39:43raptorman, I remember the first time I did it - someone was posting at the same time *and* I messed up the split
21:42:32kaenouch
21:42:34kaenthat was my fear
22:00:59raptorback later
22:01:01raptor Quit ()
22:01:25fordcars has joined
22:11:23Nothing_Muchy'all take care now
22:14:00fordcarsLater?
22:19:52Flynnn Quit (Quit: This computer has gone to sleep)
22:23:29Flynnn has joined
22:25:50Flynnn Quit (Client Quit)
22:29:17Nothing_Muchwell
22:29:22Nothing_Muchmaybe a half hour more
22:33:17Flynnn has joined
23:03:29ShadowXloner Quit (Ping timeout: 272 seconds)
23:11:11Flynnnkaeeeen
23:11:15Flynnnpartyyy?
23:11:21Flynnnwhennn?
23:12:29Flynnn Quit (Quit: Leaving)
23:12:31kaenlike three hours ago
23:12:33kaentoo slow
23:45:33raptor has joined
23:45:33ChanServ sets mode +o
23:50:15fordcarsnight guys!
23:50:18fordcars Quit (Quit: Page closed)
23:50:22raptornight
23:50:24raptorman
23:50:36raptormaking a teleporter repairable is so far not easy
23:50:51raptorI wonder if I should make a Repairable class and inherit from it...
23:54:49raptorOr maybe it should be HealthObject

Index Search ←Prev date Next date→

These logs were automatically created by BFLogBot on irc.freenode.net.