#bitfighter IRC Log

Index Search ←Prev date Next date→

IRC Log for 2013-11-07

Timestamps are in GMT/BST.

00:03:30YoshiSmb has joined
00:04:01raptorheading home! back later
00:04:11raptor Quit ()
00:07:47thread_ Quit (Ping timeout: 250 seconds)
00:21:22BFLogBot Commit: e30947945b10 | Author: watusimoto | Message: Line endings
00:21:24BFLogBot Commit: dcb2016fddd9 | Author: watusimoto | Message: Update fonts used in UI -- gets rid of controversial font selection
00:21:25BFLogBot Commit: cc356643c99c | Author: watusimoto | Message: Merge
00:24:14Watusimotohttp://img33.imageshack.us/img33/1575/9lpa.png
00:24:22Watusimotoa screenshot of the new font selections
00:25:40Watusimotois this readable? this font change was accidental; it looks good, but unsure about utility
00:25:40Watusimotohttp://img9.imageshack.us/img9/7585/z5ts.png
00:27:02Watusimotoscreenshot of front page with new font
00:27:03Watusimotohttp://img208.imageshack.us/img208/2002/mn15.png
00:48:01Watusimotogood night!
00:49:01Nothing_Much Quit (Read error: Connection reset by peer)
00:49:25Nothing_Much has joined
00:49:25Nothing_Much Quit (Changing host)
00:49:25Nothing_Much has joined
00:52:53Watusimoto Quit (Ping timeout: 248 seconds)
00:55:57Nothing_Much Quit (Read error: Connection reset by peer)
01:22:40Nothing_Much has joined
01:22:41Nothing_Much Quit (Changing host)
01:22:41Nothing_Much has joined
01:27:05YoshiSmb Quit (Ping timeout: 250 seconds)
01:51:53Nothing_Much Quit (Remote host closed the connection)
01:53:20Nothing_Much has joined
01:53:20Nothing_Much Quit (Changing host)
01:53:20Nothing_Much has joined
02:05:26amgine123 has joined
02:05:32amgine123sup
02:06:03amgine123kaen you afk
02:08:30Nothing_Much Quit (Remote host closed the connection)
02:09:08Nothing_Much has joined
02:09:19Nothing_Much Quit (Changing host)
02:09:19Nothing_Much has joined
02:09:44Nothing_Much Quit (Read error: Connection reset by peer)
02:10:15Nothing_Much has joined
02:10:16Nothing_Much Quit (Changing host)
02:10:16Nothing_Much has joined
02:10:38kaenalways
02:10:41kaenWatusimotoam
02:10:45kaenwhoa oops
02:10:48kaenwhat's up amgine123 ?
02:11:28amgine123i boke the code completly
02:11:35kaenlike a true programmer
02:11:39kaennow you just have to fix it :)
02:12:04amgine123well im not surehow
02:12:08amgine123it broke and raptor and sme tried to help but tey made it worse
02:15:12amgine123http://pastie.org/8461473
02:18:05kaenamgine123: what does the error say when you try to run it?
02:18:23amgine123osmthng about compliling hold on
02:19:17amgine123***PLUGIN ERROR*** Error compiling script editor_plugins/Move objects1.1.lua editor_plugins/Move objects1.1.lua:20: '=' expected near 'end' Client connecting to master [bitfighter.org:25955]
02:21:35kaenok
02:21:48kaenso, I'm going to teach you a little bit about error messages
02:22:01amgine123ok
02:22:10kaensee how the second time that filename is shown, there's a :20 after it
02:22:19amgine123yes
02:22:30kaenok, that's the line number it hit the error on
02:22:53amgine123hmm
02:23:01kaenbecause it's a syntax error ("error compiling"), you mistake actually comes before that line
02:23:10kaenand the computer just doesn't realize it until line 20
02:23:34amgine123soo
02:23:35kaenlet me ask you this, amgine123:
02:23:45kaenwhy is there a `do i` there?
02:24:02amgine123wait i thought that was part of the loop
02:24:05kaennope
02:24:12kaena for loop looks like this:
02:24:18kaenfor i=1,#objects do
02:24:20kaenend
02:24:30kaenand your code goes *between* do and end
02:24:57kaenso when lua sees your `do i` it has idea what you're talking about or what that even means
02:25:04kaenso try removing it and see how it goes :)
02:25:15kaenhas no idea*
02:25:39amgine123lol now the error is on line 19
02:25:52kaenafter you deleted do i ?
02:26:01kaenpastie your changes.
02:28:20sam686 has joined
02:28:20ChanServ sets mode +v
02:32:22amgine123hmm
02:32:33amgine123if there is to many plugins some dont show u
02:32:37amgine123up
02:35:47amgine123weird i dont see my plugin
02:35:50amgine123bbl piano
02:38:46Nothing_Much Quit (Read error: Connection reset by peer)
02:41:55Nothing_Much has joined
02:41:55Nothing_Much Quit (Changing host)
02:41:55Nothing_Much has joined
03:02:55kumool Quit (Read error: Connection reset by peer)
03:03:13kumool has joined
03:06:15kaenkumool Nothing_Much amgine123 sam686 HylianSavior game night tonight
03:06:19kaenright now :)
03:07:03Nothing_Muchyeah, can't play on this odroid yet :( it crashes when there's a lot of things going on
03:10:33kumoolodroid? crashing? OS? future? what!?
03:12:13Nothing_Muchkumool: lol
03:12:16Nothing_Muchgo and play dude
03:12:22Nothing_MuchI'll explain what it is later
03:19:50Nothing_Much Quit (Remote host closed the connection)
03:35:29amgine123back
03:35:37amgine123did i miss iy
03:35:39amgine123it
03:38:08raptor has joined
03:38:08ChanServ sets mode +o
03:38:58raptorgames games?
03:39:17fordcars has joined
03:45:32raptorsam686: wrote it down: https://code.google.com/p/bitfighter/issues/detail?id=287
03:46:10raptorkaen: are you playing on some hidden server?
03:59:23amgine123but i found it xd
03:59:28amgine123firt to have many amgines Xd
04:15:38BFLogBot Commit: feb96db92326 | Author: sam8641 | Message: Fix master leave chat, enable joystick input all the time.
04:18:42raptorsam686! you committed! :)
04:19:19sam686yes, the master problem was a bit dumb, server don't care about chat, clients do.
04:19:47raptoryeah, probably copy pasta in the refactor
04:40:30raptorsam686: did you update master?
04:41:01raptorI'll do it anyways, just in case
04:41:20sam686i haven't
04:41:27raptorok, i'll do it
04:47:41amgine123raptor my build of 19 is outdated
04:53:15BFLogBot Commit: b9f848743df2 | Author: sam8641 | Message: Replace supressScreensaver with SDL_DisableScreenSaver, added ini option
04:54:23raptorsam686: you used iniSettings->clientPortNumber for the option in config.cpp
04:54:34sam686problem with the old supressScreensaver was it is windows only, and it randomly interrupts my alt+tab. SDL_DisableScreenSaver so far don't have the problem I had with the old way
04:55:05raptorthat's good you got rid of it..
04:55:26sam686oope i see my mistake on loadTestSetting
04:56:30amgine123sam im havng mic issues again i cant hear my voice on voice cht on BF 19
04:57:39BFLogBot Commit: b16bc76e425d | Author: sam8641 | Message: Fix my small error with my new ini setting
04:58:54sam686mic appears to work fine for me.
04:59:23amgine123wonder if its abu with my mic i hear no feedbackfrom audio
04:59:28amgine123on BF
05:00:02sam686you should hear yourself when "Voice echo" is on.
05:00:08amgine123i dont
05:00:40sam686first of all, have a mic? (some laptops have mic built-in)
05:01:15raptorbe back later...
05:01:23amgine123i have a old fashion head phone mic combo
05:01:26raptor Quit ()
05:02:10sam686you might need to look at volume and input, in sound control panel
05:04:49amgine123hmm i dont see anything
05:05:03noneofmynickswor has joined
05:05:09amgine123do you still have team viewer
05:08:14kumool Quit (Ping timeout: 240 seconds)
05:08:46amgine123jsut like a dev to goafk
05:11:34fordcarsCheck the .inin
05:11:45amgine123??
05:11:55fordcars.ini
05:12:26fordcarsIf you want to give devs a break, I can compile the latest 019 for you if you want
05:12:52amgine123try it must it must be a stand alone ZIP
05:13:10fordcarssure
05:13:14fordcarsWindows?
05:14:14noneofmynickswor Quit (Quit: Leaving)
05:14:24amgine123yep
05:14:52fordcarsOk, compiling, it'll be ready in about 5 minutes
05:15:01bobdaduck has joined
05:15:10bobdaduck has left #bitfighter
05:15:11BFLogBot Commit: 051ece7674eb | Author: sam8641 | Message: Fix switching level after ship holding Nexus flags
05:16:06sam686oh, forgot the scroll down this IRC chat, i didn't see new lines till now..
05:18:58fordcarsamgine, you really should find a way to compile bitfighter on your computer!
05:18:58bobdaduck has joined
05:19:02bobdaduckkaen around?
05:19:03fordcarshi boby
05:19:05fordcarsYep
05:19:12fordcarsI think
05:19:14fordcarsI guess
05:19:25bobdaducklol
05:19:26fordcarshey bobdaduck, I finished your bot!
05:19:29bobdaduckREAL?
05:19:32fordcarsSIKE
05:19:40bobdaduckrofl
05:19:50fordcarsLol, I should really finish it
05:20:07fordcarsI just, urgh
05:21:41bobdaduckhttp://pastie.org/8461733
05:21:51bobdaduckI test if one angle is near another angle like so
05:22:10bobdaduckhowever if the ships angle is up and to the right, the angle will be close to 360
05:22:12bobdaduckand the check will fail
05:22:43bobdaduck@sam686
05:23:16sam686looking at it..
05:23:33sam686I need to put it into my code that tests some values..
05:23:42thomas has joined
05:23:43bobdaduckalright
05:24:06thomas is now known as Guest55152
05:24:07sam686its been a while since I last use LUA
05:26:41fordcarsoh I think I know
05:27:28fordcarsif the angle is 359, it will check if it is between 340 and 10 degrees , for example
05:27:32fordcars?
05:28:00bobdaduckwell if the target angle is 0 degrees for instance
05:28:13bobdaduckit checks if its between like 20 and -20 degrees
05:28:20bobdaduckand if the ship is at 350 degrees it fails the check
05:28:59fordcarsit depends if 360 and up you have 370 degrees or 10 degrees, i guess
05:29:20fordcars350 is not between 340 and 20
05:32:45fordcarsShip angle is probably 0-360, where your code is 0 and up
05:33:21bobdaduckuh
05:34:20fordcarsshipangle = 350
05:34:32fordcarsurghè
05:34:57bobdaduckyeah
05:35:03bobdaducktarget angle = between -20 and 20
05:35:11BFLogBot Commit: 417ab44a1fed | Author: sam8641 | Message: Fix renderMessageBox with instr=NULL
05:35:21fordcars350 + 20 = 370
05:35:29sam686once again I find an error while trying to use LUA on 019 (nothing related to your code)
05:35:38fordcarstarget = 10 degrees!!!!!!!!!!
05:35:44fordcarsshipangle = 350
05:35:45fordcars350 + 20 = 370
05:35:47fordcarsshipangle = 350
05:35:49fordcars350 + 20 = 370
05:35:51fordcarstarget = 10 degrees!!!!!!!!!!
05:35:57fordcarslol sorry
05:36:12fordcars370=10 degrees
05:36:16fordcarsSomething like that
05:36:36fordcarsamgine, I am making the .zip
05:37:34fordcarsuploading
05:37:41fordcarsWow, I spammed
05:38:10bobdaducklol
05:38:13bobdaduckYeah I know the problem
05:38:38bobdaduckI tried checking if the angle was within the range, OR if the angle + 360 was within the range
05:38:50bobdaduckand that basically made my entire code break
05:39:04bobdaduckstill ran, just behaved in such a ridiculous manner
05:39:20fordcarsbobdaduck, try= if(targetAng>360) then targetAngle = targetAngle - 360
05:39:26fordcarsLol k :P
05:39:57fordcarsamgine, http://www.filedropper.com/exe_1
05:39:59bobdaducklike it suddenly acted like the ship was ALWAYS within target angle
05:40:29fordcarsoh ok lol
05:40:36fordcarswhat are you doin^
05:40:41fordcars*doing?
05:41:54bobdaduckif shipAngle is within targetAngle by a margin of 1/16 tau
05:41:56bobdaduckthen do things
05:42:32fordcarstau
05:42:34fordcarshehe
05:42:59fordcarsamgine, you downloaded the file^
05:43:27fordcarsI have to go, but I might be back in a few minutes
05:43:32fordcarsLater!
05:43:40sam686p = point.new(x, y) ; angle = what? I forgot and don't know..
05:44:28bobdaducksam686: What?
05:44:34bobdaduckangle is probably shipAngle
05:44:46sam686I was trying to convert point into a angle
05:44:50bobdaduckand targetAngle will be different but the check only fails when its close to 0
05:45:14sam686will try p.angleTo()
05:45:27bobdaduckpoint.angleTo(p, q)
05:45:28bobdaduckthat?
05:45:50sam686will try that too, mine errored out
05:46:32bobdaduckhow do you convert a point to an angle... Sorry I don't know what you're trying to do xD
05:47:57fordcars Quit (Ping timeout: 250 seconds)
05:49:47sam686getting closer to testing
05:58:56amgine123bye
05:58:59amgine123 Quit (Quit: Page closed)
06:04:51Guest55152 Quit (Ping timeout: 250 seconds)
06:05:51HylianSavior Quit (Read error: Connection reset by peer)
06:09:35sam686http://pastie.org/8461809 so far I just created an easy test now, ctrl + k in editor with this levelgen script is easy way to test
06:10:48sam686angle = point.angleTo(point.new(0,0), point.new(x, y)) -- it looks like if the angle is going by radians instead of 360 degrees.
06:11:10bobdaduckyeah, I'm using radians
06:16:13bobdaduckSo the problem is that a shipAngle of 350 degrees fails the withinTargetAngle test when the target angle is 0
06:17:07sam686I see, like if the angle stops when it goes around back to zero
06:18:37bobdaduckyeah
06:19:48bobdaduckSo I need to adjust the logic for withinTargetAngle
06:19:54bobdaduckto include angles close to 360
06:20:01bobdaduckI don't know how though
06:20:13bobdaduckand my last try made it basically always return true for some reason
06:29:14sam686http://pastie.org/8461842 I changed withinTargetAngle, seems to work better
06:30:56sam686appears to work fine as long as both angle inputs is between -3.14 to 3.14
06:31:33sam686bobdaduck: got my message yet?
06:34:19bobdaduckno
06:34:25sam686look at my pastie
06:34:30bobdaduckk
06:34:45sam686go test it out and see if it works any better?
06:36:18bobdaducktesting
06:38:49sam686or better yet, http://pastie.org/8461847 works with any angle, not just between -3.14 to 3.14
06:38:58bobdaduckit still seems to be returning false if the ship is facing to the right and slightly up
06:40:20sam686what is both angle input, i can test it if you gave me both input numbers
06:41:23bobdaduckokay
06:42:07nothing_much has joined
06:42:23nothing_muchI borked my irc client
06:42:46nothing_muchkaen: if you're there, lemme know how to get autoconf not to screw up
06:44:01bobdaduckinput angle is 6.25ish
06:44:08bobdaduck(which is in radians)
06:44:36sam686the other input angle? and the margin?
06:47:32sam686you may make use of logprint("angle " .. shipAngle .. " angle2 " .. targetAngle) or similar
06:50:30bobdaduckmargin is 1/16 tau
06:50:43bobdaducktarget angle is 0
06:51:56bobdaduckany others you need to know?
06:52:35sam686shipAngle? in the withinTargetAngle, if you still have problems... Make use of logprint I typed above as well.
06:52:54bobdaduckshipangle was 6.25ish
06:52:58bobdaduck(which is radians
06:53:44sam686logprint(withinTargetAngle(0, 6.25, 3.14/16)) prints TRUE for me
06:54:22bobdaduckreverse 6.25 and 0
06:54:40bobdaduckactually
06:54:45bobdaduckthat second pastie seems to be working
06:54:46bobdaducktesting
06:54:48sam686logprint(withinTargetAngle(6.25, 0, 3.14/16)) is TRUE for me too..
06:55:32bobdaduckI think the second one works
06:55:37bobdaduckcan you explain what you did to fix it?
06:55:38sam686my first modifying needs angle to be -3.14 to 3.14, but changed it to be any number for my second (latest) pastie.
06:56:12bobdaduckYeah that one seems to work
06:56:24sam686A simple % (3.1415926535897932384626433832795*2) at the end is like it repeatedly subtracts (3.14*2) until it is below the value
06:56:38bobdaduckhuh
06:56:52bobdaduckjoin my server to see what you helped me do?
06:56:53sam686it like a remainder of a dividing
06:57:04sam686ok, in 018?
06:57:12bobdaduckyeah
07:00:08Platskies has joined
07:10:29bobdaduckthanks sam!
07:10:34Platskies Quit (Read error: Connection reset by peer)
07:10:49bobdaduck Quit (Remote host closed the connection)
07:21:06nothing_much Quit (Quit: Page closed)
07:54:33Invisible1 has joined
08:02:50koda has joined
08:16:41watusimoto has joined
08:16:41ChanServ sets mode +o
08:20:00watusimoto Quit (Client Quit)
08:20:46Skybax has joined
08:20:50watusimoto has joined
08:20:50ChanServ sets mode +o
08:55:52watusimoto Quit (Quit: Leaving.)
10:04:43Darrel has joined
10:27:55Skybax Quit (Ping timeout: 246 seconds)
11:17:39watusimoto has joined
11:17:39ChanServ sets mode +o
11:31:16Invisible1 Quit (Ping timeout: 246 seconds)
11:37:52destroyerimo has joined
12:22:14watusimoto Quit (Ping timeout: 240 seconds)
12:34:31watusimoto has joined
12:34:35ChanServ sets mode +o
12:34:47YoshiSmb has joined
12:36:38YoshiSmbMorning. (9:36 hs on argentina)
12:38:44BFLogBot Commit: 404290c12f35 | Author: sam8641 | Message: Add copy constructor to NetObject, which fixes crash on exit after adding hundreds of walls in editor.
12:52:18watusimoto Quit (Read error: Operation timed out)
12:52:59watusimoto has joined
12:52:59ChanServ sets mode +o
13:06:47Watusimoto_ has joined
13:09:45Invisible1 has joined
13:17:16kaenhere's a big one for you guys :)
13:17:17BFLogBot Commit: 7686737d5d16 | Author: kaen | Message: massive copyright/licensing changes. also removed some cruft from the zap directory
13:17:18BFLogBot Commit: 4c04ddb8f041 | Author: kaen | Message: fix compiling on gcc
13:18:06kaenand good morning
13:21:13YoshiSmb Quit (Ping timeout: 250 seconds)
13:56:54Canseco has joined
14:00:01Invisible1 Quit (Ping timeout: 246 seconds)
14:20:46Watusimoto_ Quit (Ping timeout: 265 seconds)
14:26:28BFLogBot Commit: 1d7f530272ed | Author: kaen | Message: zap/tnl warnings
14:29:21BFLogBot Commit: 72499eec8d44 | Author: kaen | Message: change some leftover license headers
14:36:27watusimoto Quit (Quit: Leaving.)
14:41:28watusimoto has joined
14:41:28ChanServ sets mode +o
14:50:25Invisible1 has joined
15:06:36Canseco Quit (Quit: Leaving)
15:16:10LordDVG has joined
15:18:25Invisible1 Quit (Ping timeout: 246 seconds)
15:19:40watusimotohi
15:20:14watusimoto4 people playing I've never seen on before
15:33:39kaenevery. single. time
15:33:46kaenI join with a group of new players
15:33:51kaenget through part of like one game
15:33:54kaenthey all ragequit
15:34:04watusimotoreally??
15:34:06kaenstart a new server with a password
15:34:25watusimotowell, at least they know how to do it!
15:34:27kaenyeah
15:34:30kaenheh
15:34:55kaenI'm like really competitive... so it's hard to go easy on them :/
15:35:11kaenI guess it's ultimately my own fault because I know the cause of the problem but do nothing to correct it
15:38:50watusimotoyeah, I don't cause people to ragequit, no matter how hard I play
15:40:01kumul has joined
15:53:22HylianSavior has joined
16:00:09watusimotokaen: what do you think of the font changes?
16:00:34watusimoto1) improvement 2) same same 3) revert!!
16:03:38kaenPretty neutral
16:03:59kaenI like its character, but I think it looks out of place in the weapon list
16:04:20kaener I mean the equipped weapon/module display
16:04:21watusimotoyou mean the list across the top, or when changing your module
16:04:28watusimotothe list across the top
16:04:29kaenthe top
16:04:31kaenyes
16:04:31watusimotoyes
16:04:41watusimotounderstood
16:05:20watusimotoit may just be what you (and I) are used to -- it is theoretically a better font. But I really like it in some places, less so in others
16:06:13watusimotoI want to revisit the fonts again before the release
16:06:27watusimotobut I'm going to give it a day or two to see how my eye settles
16:06:36kaenoh, I get an assert btw
16:06:43kaensomething about fonts
16:06:53watusimotooh great!
16:07:02kaenInvalid font id!
16:07:09watusimotowhere/when?
16:08:49kaenhttp://pastie.org/8462960
16:09:02destroyerimo Quit (Ping timeout: 240 seconds)
16:09:07watusimotogrrrr
16:09:42watusimotoah, do you have the new fonts in your fonts folder?
16:09:47watusimotoyou might need to do a manual copy
16:13:56kaenyep, I have all of resource symlinked in to exe
16:18:00HylianSavior Quit (Quit: Leaving)
16:23:49HylianSavior has joined
16:27:57HylianSavior Quit (Client Quit)
16:45:06watusimotolaterz
16:46:53Invisible1 has joined
16:47:03Invisible2 has joined
16:49:41watusimoto Quit (Ping timeout: 248 seconds)
16:51:23Invisible1 Quit (Ping timeout: 272 seconds)
17:03:01Invisible1 has joined
17:04:03Invisible2 Quit (Ping timeout: 272 seconds)
17:08:32LordDVG Quit (Remote host closed the connection)
17:44:25raptor has joined
17:44:35ChanServ sets mode +o
17:45:21raptoroh WOW
17:47:54raptorsam686: I'm worried about your TNL change... what does it do exactly? And how will it affect the rest of TNL with respect to performance?
17:50:39raptorlooks like it does the exact same thing as the default copy constructor, but sets mNetIndex to U32_MAX?
17:51:16raptorinvalid font id!
17:57:10thread_ has joined
17:57:52HylianSavior has joined
18:00:12thread_Heya!
18:01:08kaenhi thread_ !
18:01:50raptorkaen: is that the 'proper' way to do licenses? have the LICENSE.txt point to the COPYING.txt for the GPL?
18:01:57raptoroh and good morning!
18:02:21kaenyep, COPYING is where the GPL goes
18:02:39kaenLICENSE is where the author puts the statement that his software is GPL-licensed
18:04:12fordcars_win has joined
18:04:17raptorah ok
18:04:36kaenI stole gcc's COPYING.txt, too
18:04:46kaenbecause it was formatted much more nicely :)
18:05:43kaenit's center-aligned if you're reading it in the 1970's
18:05:53fordcars_win Quit (Client Quit)
18:05:59Skybax has joined
18:06:12fordcars_windows has joined
18:06:19kaenyeah, that NetObject copy constructor is dubious
18:06:37kaenit actually has the same logic as the default (non-copy) constructor
18:07:31kaenmaybe the default copy constructor did too thorough of a job copying?
18:07:44SkybaxGoodmorning
18:07:51kaenthat's just a guess, I don't know what issue sam actually fixed
18:08:01kaenmorning Skybax
18:08:07kaenhey, and it's actually morning, too!
18:08:16SkybaxYay!
18:08:25SkybaxCause I actually woke up today lol
18:09:01kaenoh, I also removed the (C) from the source files
18:09:18raptorkaen: looks good
18:09:23kaenbecause apparently only the actual encircled copyright symbol, and the english word "copyright" have legal significance
18:09:32fordcars_windowsgood after noon!
18:09:36raptoroh lovely
18:09:36kaen(C) means nothing in court it seems
18:10:03kaengood afternoon, fordcars_windows :)
18:10:20kaendid people actually show up for game night last night?
18:11:01kaenI went afk to nurse my sickly girlfriend back to health and then I got trapped
18:11:17fordcars_windows Quit (Quit: http://www.kiwiirc.com/ - A hand crafted IRC client)
18:11:44kaennice talking to you fordcars ...
18:12:47fordcars_ has joined
18:12:52fordcars_urgh
18:13:02SkybaxxD
18:14:03fordcars_:P
18:14:43kaenwhere are we even copying netobjects?
18:15:01kaenhmm... maybe in the editor
18:15:05kaenwe do a lot of cloning
18:16:23kaenI'm half-tempted to make that 'explicit' because we definitely shouldn't be passing by value or anything to copy them implicitly
18:22:19raptorcopy constructors have always bothered me...
18:25:57Darrel Quit (Read error: Connection reset by peer)
18:26:11fordcars_ Quit (Quit: http://www.kiwiirc.com/ - A hand crafted IRC client)
18:26:31Darrel has joined
18:30:02kumool has joined
18:33:18kumul Quit (Ping timeout: 252 seconds)
18:33:20kaenit's also kind of smelly that it has a user-defined copy constructor but no assignment operator
18:33:42SkybaxUh oh. This is bad
18:34:20kaenSkybax I'm interested in your opinion of user-defined copy constructors in non-trivial types with pointer data members.
18:34:25kaen:)
18:34:55SkybaxIf you could explain to me what that in in layman's terms I could probably provide you with an opinion xD
18:35:13kaenI'm just playing -- I don't fully understand it myself
18:35:16kaenbut what's bad?
18:35:42SkybaxBitTown has reached a size where I can no longer open it xD
18:35:56kaenah
18:36:22kaenthat usually has to do more with barriers/polywalls than anything
18:36:34SkybaxIt loads the level but then the screen just stays black
18:36:35kaenconvert anything you can into line items.
18:36:56raptorSkybax: is it taking up too much CPU?
18:37:18SkybaxIdk. I can't even open it when I host a server and then switch to it lol
18:37:32SkybaxI can send you the level if you want and you can see for yourself
18:37:49kaenwhat's the task manager say?
18:38:02Invisible1 Quit (Quit: Invisible1)
18:38:07kaen(does it still have CPU/RAM info?)
18:38:12SkybaxYeah one sec
18:38:22Invisible1 has joined
18:39:41BFLogBot Commit: 7137b9f984c5 | Author: buckyballreaction | Message: Remove unused fonts to stop triggering the asserts
18:40:06SkybaxBitfighter actually takes up less CPU when I'm trying to load the level than it does when I'm just sitting in the editor
18:40:13Skybax3% to 24%
18:40:35raptorgoodbye assert!
18:41:14SkybaxSo it's not that xD
18:41:31raptorwell, the editor does have performance 'problems'
18:42:20kaenthe entire rendering system is a performance problem :P
18:42:29SkybaxxD
18:42:33kaenso it RAM?
18:42:42kaenso is it*
18:43:06SkybaxI don't have a RAM thing to see. It's just CPU, Memory, Disk, and Network
18:43:11kaenmemory
18:43:14kaenis ram
18:43:17raptori doubt it's ram
18:43:27kaenthen... what is it?
18:43:52SkybaxThe memory goes up two MBs when I start loading the level lol
18:44:02kaeneh that's not it then ...
18:44:08SkybaxFrom 58 to 60
18:44:18kaenI guess you'd better send the level
18:44:20raptorok, Skybax, it loaded in my editor but is using up 85% of one CPU
18:44:30SkybaxWow
18:44:34raptorand I have a monster dev machine at work...
18:44:38SkybaxI wonder why it made my CPU go down
18:44:48kaenbad metrics probably
18:45:05raptorthe forest is a clipping nightmare
18:45:13kaenyeah, that's what I was thinking of
18:45:16SkybaxYeah I was just thinking about the forest
18:45:18raptorI actually think there's a clipper call made each frame in the editor...
18:45:31kaenno way ...
18:45:35SkybaxI could probably take some of the trees out in favor of just a wall
18:45:53raptoryeah... i vaguely remember this from a profile I did one time (hunting down a different issue)
18:46:03raptorI just thought to myself "that's odd" and left it, at the time
18:46:12kaen@_@
18:46:50SkybaxHow do I turn a wall into a lineitem?
18:47:15kaenit's non-trivial, unfortunately :/
18:47:25kaenI wrote a plugin that does it, but only in 019
18:47:26kaenoh!
18:47:27kaenI know
18:47:36kaensend me the level and I'll run it through my plugin
18:48:03raptorSkybax: it loads on my machine, what would you like me to do... put it on the server?
18:48:08SkybaxBut how will you know what I want to change into a lineitem lol
18:48:24kaenyou could... you know... tell me
18:48:35Skybaxraptor, idk if that would be a good idea since only epic machines can load it xD
18:48:37raptorhehe
18:48:42raptorSkybax: ok
18:49:21Invisible1 Quit (Ping timeout: 252 seconds)
18:50:26thread_WHAT DOES THE FOX SAY?!?
18:50:43raptorgive me more rabbit?
18:50:45SkybaxRING A DING DING DING DA DING DA DING DING
18:51:41thread_YA CHA CHA CHOW YA CHA CHA CHOW!
18:51:53kaenit's supposed to be frog right?
18:51:57kaenI'm hip right guys?
18:51:58kaenguys?
18:52:38SkybaxHeh
18:53:28SkybaxI scented you the level code
18:53:42kaenfunny story: I first heard that song on a conservative talk radio show's bumper music
18:55:03SkybaxNice
18:56:29thread_Considering that the song was written by a radio talk show...
18:57:11watusimoto has joined
18:57:11ChanServ sets mode +o
18:57:56raptoroh yay, watusimoto is here
18:57:58raptormaybe
18:58:47thread_watusimoto is maybe here?
18:58:56SkybaxMebbe
18:59:19thread_He might possibly maybe could with a chance be here?
19:01:07watusimoto Quit (Read error: Operation timed out)
19:02:45kaenSkybax: I don't understand your instructions
19:02:54SkybaxxD
19:03:03kaenyou want me to make all of the trees that touch that wall into lineitems?
19:03:20SkybaxIn the forest, I put a big wall that goes almost all around the forest. Every tree that it's touching can be a lineitem
19:03:48kaenand all the rest stay walls?
19:03:54kaenI have to ask why ...
19:04:11sam686as for raptor questioning me about TNL changes... normally, NetObject should never be copied, or that would mess up the linked list. It is being copied only in editor, so adding a copy constructor was needed (and easy way) to avoid such an error.
19:04:17SkybaxFor gameplay purposes lol
19:04:50kaenah, so my conjecture was correct sam686 ?
19:05:29kaenand do you know where/how the copying occurs, and would it be reasonable to make that copy constructor explicit to prevent doing it accidentally?
19:05:44sam686i haven't read what kaen said, so not sure..
19:06:24koda Quit (Ping timeout: 268 seconds)
19:06:57raptorhi sam686
19:07:57raptorsam686: how does the copy constructor differ than the default one that would be triggered when copying an object?
19:08:21sam686the problem was, starting a new editor, creating lots of walls, when exit the editor/game errored out.
19:08:44sam686maybe the default one might be missing from some objects? WallItem?
19:09:18sam686in EditorUserInterface::cleanUp(), removing clearUndoHistory(); makes the error far more obvious, until added NetObject copy constructor for a fix
19:09:48raptorcould it be that some of the editor objects are not children of NetObject?
19:10:25sam686it won't affect objects that isn't NetObject
19:11:35raptoris there some other way to fix other than adding that copy constructor? I get worried with modifying TNL code...
19:12:10kaenI'm fairly confident that's the correct fix
19:12:21kaenin fact I'm confident that this was yet another TNL bug
19:12:58sam686I didn't see an easy way around that, and it looks like if TNL wasn't designed with copying objects in mind.
19:13:21kaenbecause NetObject absolutely should not be bitwise (i.e. default) copied, yet there was no sure-throw copy constructor nor a correct copy constructor implemented
19:14:09kaenbasically TNL sucks and fixing rookie mistakes like this that crop up in it is the price we pay
19:15:56sam686I just basically make a copy constructor copy nothing in NetObject, like if it is a new object.
19:17:06kaen(just to be clear, I'm arguing in favor of your change, sam)
19:17:55sam686yes, and it appears I know more about several parts of TNL then others
19:18:28LordDVG has joined
19:19:12kaenI might have to take issue with that. I think I'm the only current dev who has actually implemented TNL in an engine other than bitfighter :)
19:20:35sam686one of a funny problem I saw in TNL was hundreds of connection to master (014 and earliear?) that fails to time out due to a millisecond counter overflow, that I fixed by testing timer overflow effects.
19:20:36raptorwell I sure no nothing! (but just enough to be worried when it changes)
19:20:46raptor*know
19:22:04sam686eventually the old 014 master died from running out of memory after having too many connection that is suppose to time out..
19:22:46raptorfun!
19:22:51kaenalthough I guess I couldn't have written RPCGuaranteedOrderedBigData, so you're probably right
19:24:12raptorok... this may be a bug
19:24:25raptoryou make a zero width polywall
19:24:38raptorsave the level, reload it... polywall is gone in editor
19:24:39sam686oh thats when trying to send one big stats that ordinary PCGuaranteedOrdered can't split to multiple packets
19:25:22raptorbut it still exists in level file and is loaded in-game (and ends up being an invisible barrier)
19:25:42raptorso... should the editor discard the object when you reload?
19:25:52raptorit gives warning: Invalid barrier detected (polywall with invalid geometry). Disregarding...
19:26:41sam686I have 2 different problems with level on 019
19:27:04sam686first, missing addLevelLine in levelgen? ../../levels/6357.levelgen:5: attempt to call missing or unknown method 'addLevelLine' (a nil value)
19:27:09raptorbecause then there is no way to delete your invalid wall other than editing the file directly
19:27:15raptorsam686: that method has been removed
19:27:41sam686second problem applies to many old levels I have, they become extremely small as those levels is missing GridSize
19:27:58raptorwait... no gridsize?
19:28:07raptori thought gridsize was always added?
19:30:34raptorsam686: we got rid of gridsize for 019, unless it was found in the level file
19:31:16raptorwhat older levels didn't have gridsize?
19:32:10sam686it could be that some users, including me, just remove the "GridSize 255" since it default to 255 on 018 and earliear..
19:32:25raptorok, so not standard
19:32:32raptorif so, i'm not too worried
19:32:42kaenI don't think modifying level code by hand is supported anymore
19:32:53raptorheh
19:33:18raptoryeah, huh - we've actually made it so you don't really need to do that anymore...
19:33:21kaenI say that because you can screw things up with at least the LevelDatabaseID and GridSize
19:33:33kaenexactly -- plugins cover all the old use cases
19:33:52kaenmaybe with one exception: time and score limits
19:34:12raptorhow so? I remember adding those for some gametypes that didn't have them
19:34:49kaenwell for instance if I need a score higher than 99 in zone control (when I have e.g. 52 zones) then I have to go diving
19:35:05raptorreally? that's still in place??
19:35:10kaenyeah
19:35:21raptorok, let me change that - what should be good... 1000?
19:35:21kaenwait, maybe
19:35:36kaeni dunno I might have been in 018a when that happened
19:35:49kaenanyway 1000 should be enough for everyone :)
19:35:54raptorhaha
19:38:35sam686and it breaks "MazeRacer" too...
19:39:55raptormazeracer works
19:40:04raptoryou might need to copy new resources
19:40:09raptoroh wait
19:40:14raptordouble checking...
19:41:01kaenmaze racer works
19:41:05kaenI run it all the time
19:41:08sam686I have about 30 levels so far without GridSize and thay are all tiny and unusable on 019, mostly level contest maps and dungeons
19:41:17Skybaxkaen you broke BitTown lol
19:41:36kaenoh, sorry
19:41:42kaenwell I guess that doesn't work
19:41:56SkybaxApparently not xD
19:41:58kaenkind of weird though
19:42:11kaenbecause it shouldn't have made any 019-specific code
19:42:13kaenoh well
19:42:37SkybaxCan I just go into the code and change the barrier to lineitem and it would put a lineitem in the same place as the tree?
19:43:00kaensam686: for i in ${levels} ; do echo 'GridSize 255' >> $i ; done
19:43:15raptorkaen: if the level was loaded in 019 editor and saved, all the point values are changed
19:43:23kaenohhhh that's right
19:43:46raptorman... I'm already feeling the growing pains from my change...
19:46:02raptorthis new font... is it missing subpixel hinting?
19:46:21raptorseems of lesser quality, even though it does seem to be a bit better in certain areas of the game
19:46:30raptor(than tenby)
19:47:05sam686my "Picture Frame" level says "our cure oes ere" it suppose to say "Your Picutre Goes Here", messed up TextItem font?
19:47:21raptorhaha
19:47:22raptorwhat
19:47:27raptorcan you send me that?
19:48:40sam686http://sam6.25u.com/bitfighter/levels3/picture_frame.level
19:49:43sam686works fine on 018, messed up on 019
19:49:57raptorthis is my 019: http://imagebin.org/276111
19:50:04raptormaybe you need a fresh recompile?
19:50:13raptoror maybe there
19:50:29raptor's some bug in rendering that only shows on your graphics?
19:51:15sam686mine looks like this http://sam6.25u.com/upload/screenshot_121.png
19:51:31raptoroh!
19:51:37raptoryou need the new 'fonts' directory
19:53:12sam686why??? http://sam6.25u.com/upload/screenshot_122.png odd font as usual...
19:53:23thread_ Quit (Ping timeout: 250 seconds)
19:53:34sam686but if I hit ESC and have menu, http://sam6.25u.com/upload/screenshot_123.png it is the old way of line font?
19:54:25raptori'd say full recompile, and double check you have all the new resources
20:02:44SkybaxDangit... I took away tons of the trees and it still won't load
20:02:54SkybaxAre you sure lineitems don't take up the same space as walls?
20:03:28raptorthey take up the same 'space' but don't require some of the extra processing
20:03:42SkybaxWell it's still broked x)
20:04:58sam686At first it back to normal, but later the weard font with missing letters returns...
20:05:06sam686the weard font problem appear to start once a scoreboard is shown in a 2 or more team game
20:05:27raptorok, i'll test scoreboard
20:05:38sam686and continues to be weard until bitfighter is shutdown/terminated
20:05:46raptorsounds like a memory error?
20:06:42raptorI changed a header in my last commit
20:06:50sam686I only see missing letters on my "Picture Frame" level, other levels don't have missing letters, just different font (after scoreboard shown)
20:06:55raptorand visual studio sometimes doesn't recompile everything
20:07:01raptorproperly
20:07:09raptorunless you do a 'clean'
20:07:42sam686Everything recompiled, i even updated to latest changes, though I can try release mode instead of debug build..
20:07:49raptorno
20:07:53raptorit should work in debug
20:08:04raptorI'll do another full recompile, too
20:08:31raptorwhen you start bitfighter, is there any font error? like a missing file?
20:08:58sam686visual studio is so easy to switch between debug/release build, as it is a seperate set of compiled temperory directory..
20:09:09raptorah yes
20:12:20sam686also happens on release build..
20:12:32raptorstill can't dupe
20:12:41LordDVG Quit (Remote host closed the connection)
20:13:20sam686and when building ZAP_DEDICATED I get Error unresolved external symbol _WinMain@16 referenced in function ___tmainCRTStartup LIBCMT.lib bitfighterd
20:13:41raptorI thought that was fixed...
20:14:04raptorI think we need kaen or watusimoto to duplicate before we hunt much further, because I have no issues
20:14:08sam686thats in ZAP_DEDICATED release build, as client or debug don't have problem
20:14:15raptormaybe you're using the wrong SDL2 headers?
20:14:25raptorsounds like you're in a bad state in visual studio
20:21:14kumoolmmmm
20:22:28kumooli dont see nothing much... never mind
20:22:49kumooli think i'm gonna post a couple of bitfighter links on reddit
20:34:13sam686ummm... /SUBSYSTEM:WINDOWS causes problem in "release" build in bitfighter dedicated build.. in Linker . Command Line
20:34:46sam686I first check on Linker.System.SubSystem not knowing the command line overrides it..
20:36:16sam686RelWithDebugInfo build works fine as it is... except TextItem font getting changed after pressing TAB in multiple team map
20:37:21raptorsam686: hah! i got the bug... i'll fix the font
20:40:43Watusimoto has joined
20:42:14raptornow the real question is where are themissing popFontContext()
20:43:07sam686Might be in UIGame.cpp GameUserInterface::renderScoreboard(), problem only happens if isTeamGame = true.
20:43:23raptorthat's one, yes
20:43:28raptorbut there's at least one more
20:44:01sam686maybe it was looping 2 times, once per team..
20:46:55Invisible1 has joined
20:47:23raptorinteresting... it doesn't fix the rendering
20:48:00raptorah, i see why
20:48:27sam686it looks like multiple FontManager::popFontContext(); pushFontContext mismatch, and the one at the end of function is not in loop
20:50:06raptoryup, that was it...
20:51:26BFLogBot Commit: 1eca448d6b5e | Author: buckyballreaction | Message: Fix scoreboard font push/pop mismatches
20:53:07raptorsorry I doubted you sam686
20:53:18BFLogBot Commit: ee56e3950d85 | Author: buckyballreaction | Message: Increase gametype score limits in the editor and remove magic number
20:53:33sam686Oh and it looks like there is a size limit on the new font, some letters disappear when they get too big
20:54:13raptorwhich new font?
20:54:38sam686the one that changes TextItem font from the scoreboard push/pop mismatch bug
20:55:15sam686although zooming in does not make font disappear, just get a little blurry
20:56:38raptorthat limit might be do to the fontstash size limit, but should matter as it's not being used for textitems anymore.. :)
21:01:06raptorWatusimoto: you around for questions?
21:01:28sam686I see a little problem with the TextItem, their line gets thin when text goes vertical no matter how big it is, when horizontal it is thicker as it should..
21:01:30Watusimotohi
21:01:41raptorhi
21:02:27raptorthe new font looks good in some cases, but it seems like it is of lesser quality. I think it is missing sub-pixel hinting like the other one had: http://imagebin.org/276114
21:02:59Watusimotohi
21:03:07Watusimotowhat am I looking at in that pic?
21:03:20raptorthe crispness (or lack thereof) of the new font
21:03:34raptori'm getting a screenshot of the old, as soon as I finish compiling...
21:03:35Watusimotoit looks ultra crisp!
21:03:40Watusimoto(perhaps overly so)
21:04:13Watusimotoso, in general, do you think the font is an improvement?
21:04:32raptorold: http://imagebin.org/276115
21:04:34WatusimotoI think it only works well in larger headlinish areas, because it can be hard to read
21:05:04raptorI like it, but it is poorer rendering quality
21:05:50raptoryou can see tenby has the pixel hinting becasue it is much easier to read a smaller sizes
21:05:51Watusimotodo you think that quality affects larger text much?
21:06:05Watusimotoah
21:06:12Watusimotoyou're talking about tenby vs. play
21:06:19sam686Maybe you could render font texture at fixed resolution and enable mipmapping that will smooth things out for small size text
21:06:21WatusimotoI thought you were talking about the larger font
21:07:04raptorthe larger font (modern-something?) looks pretty close to same quality, at least at those sizes
21:07:13WatusimotoI replaced tenby with a combination of two fonts (play and modernvision)
21:07:25Watusimotook, so what do you think of the modernvision?
21:07:52raptorsam686: we aren't going to change the font engine that much before release. Also, we already create the font at twice the width/height and downscale with AA
21:08:20raptorI like the modern vision as a header-font, better than Tenby
21:08:28raptorfeels much more 'right'
21:08:35WatusimotoI agree, mostly.
21:08:53Watusimotomedium text is ok with play
21:09:06raptorPlay looks ugly everywhere
21:09:08Watusimoto"blast as many ships as you can"
21:09:20Watusimotougly in the rendering, or the font in general
21:09:43raptorrendering, and because of the contrast to the other, higher quality fonts
21:09:57Watusimotook
21:10:13Watusimotoplay is probably easier to replace, as it's more generic
21:10:22sam686it mostly depends on the window size or if going fullscreen
21:10:35Watusimotoyes, that's true
21:10:57raptoryes, of course, but the fonts that are higher quality look good everywhere
21:11:01Watusimotobut text like "treat level" on the scoreboard is much harder to read with play
21:11:23raptordid you use Play in the SUPER SECRET menu?
21:11:25Watusimotothe help has been play for a while
21:11:28Watusimotoyes
21:11:33raptorbecause it's ugly, too
21:11:36Watusimotoha
21:11:41raptorhttp://imagebin.org/276116
21:11:42WatusimotoI thought it looked pretty good
21:11:54raptorso, i like the font
21:11:59raptorbut it doesn't render well
21:12:03Watusimotook
21:12:25raptorin fact, it may be slightly easier to read than the old droid one, i think
21:12:31raptorbut it's ugly
21:13:01raptoreasier to read as in i'm looking at it in a font viewer at the exact resolution it's supposed to be viewed
21:13:05Watusimotosam, what is mipmapping?
21:13:57raptormore opengl processing
21:15:33raptorit's pregeneration of the different sizes of textures
21:15:57raptorbut with a specific higher quality algorithm to have better antialiasing
21:16:22raptorbasically like an icon file with all the various resolutions
21:16:57raptorbut we don't want to pregenerate lots and lots of certain sizes, as people will be resizing to weird resolutions a lot
21:17:25Watusimotook
21:21:20raptorbut that's just my layman knowledge
21:25:14raptoralso Watusimoto, sam686 made this fix last night - might it do with some of the changes you made to try and fix the lingering-frame issue?: http://code.google.com/p/bitfighter/source/detail?r=051ece7674ebdd7f2a5ef847bf66fa3e609d5282
21:26:33sam686That was just the Ship carrying Nexus flags causing weard ghosts flags that can't be picked up after level changes
21:26:48WatusimotoI doubt that was related
21:26:49raptorohhh.....
21:26:53raptorI hit that once
21:26:59raptorok then
21:27:01raptorgood find sam686!
21:28:04Nothing_Much has joined
21:31:15Nothing_Much Quit (Changing host)
21:31:15Nothing_Much has joined
21:36:06Watusimotosam686: what was that joystick fix you did yesterday, adding an sdl line?
21:41:34sam686018 and earliear allows joystick to work all the time, even if bitfighter is not an active window, but the SDL version for our Bitfighter 019 seem to stop joystick from working, killing the 2 player in one computer idea,
21:41:43YoshiSmb has joined
21:41:55sam686what I added the line
21:42:01sam686fixed the idea
21:44:43Watusimotogood
21:44:47WatusimotoI like the idea :-)
21:49:44Darrel Quit (Ping timeout: 246 seconds)
21:54:53kaenjust got back from the public library
21:55:04kaenall three dozen machines are running ubuntu!
21:56:52Nothing_Muchkaen: holy crap seriously?
21:56:53Nothing_Much:D
21:57:02Nothing_Muchand hello
21:57:05Nothing_Mucheveryone
21:58:09kaenyep!
21:58:13kaenI was so pumped
21:58:20kaengonna try volunteering as a lab tech there
21:58:49Nothing_Muchdude!
21:58:53Nothing_Muchwhere do you live again?
21:59:27Nothing_Muchsomewhere in the mountain timezone I think, right?
22:00:55Nothing_MuchI would totally go to that library
22:01:48kaenwashington state
22:01:54kaenpacific timezone :)
22:02:01Nothing_Muchoh man
22:16:11Watusimotohow's the quality?
22:16:11Watusimotohttp://img543.imageshack.us/img543/7218/g9m0.png
22:16:54Nothing_Muchlooks nice
22:17:01raptorthat looks much better... despite your inferior openGL implementation :)
22:17:14Watusimotoit's also in a small window
22:17:17Watusimotolots of text
22:17:18Watusimotohttp://img543.imageshack.us/img543/1133/pisx.png
22:17:25Watusimotoit might be too heavy
22:17:38Watusimotothere is a lighter version, but I think it might be too light
22:18:14Nothing_Much Quit (Remote host closed the connection)
22:18:21raptorthat has horizontal line alias problems (notice the 'e's), but that might just be your poor opengl implementation.. :)
22:18:28raptorbut much more readable
22:18:34raptora little heavy, yes
22:18:42raptorbut cleaner
22:18:48Nothing_Much has joined
22:18:48Nothing_Much Quit (Changing host)
22:18:48Nothing_Much has joined
22:18:58raptorand I'm probably really picky..
22:19:04Watusimototenby looked like that too (the e's)
22:19:27raptorok, then it is your implementation (I think it's a poorer AA algorithm in your drivers)
22:19:58Watusimotoyes, got it
22:20:00Watusimoto:-)
22:20:48raptori also like the colors on that page... looks nice
22:21:17Watusimoto:-)
22:23:46Watusimotolooks better here:
22:23:47Watusimotohttp://img203.imageshack.us/img203/8830/7j9y.png
22:24:21raptorare you not taking screenshots with ctrl + q ?
22:24:28Watusimotobut not so good here
22:24:29Watusimotohttp://img21.imageshack.us/img21/4008/o4p0.png
22:24:38Watusimotothe screenshots of the help are with ctrl-p
22:24:38raptoryuk
22:24:40Watusimotoctrl-q
22:25:00raptorhow did you get two different sizes? they all should be 800x600
22:25:02Watusimotothe ones with the scoreboard are via system screen shot because ctrl-q won't work with tab down on my system
22:25:14WatusimotoI was wondering the same thing
22:25:25raptorone is 666x500
22:26:29Watusimotoyes, weird
22:26:38WatusimotoI think the resize isn't happening
22:27:52raptorha, you're right!
22:27:56raptorhow did that break??
22:28:43Watusimotonot sure, but found something else interesting
22:28:52Watusimotomany of these fonts are otf
22:28:58Watusimotoand I've been converting them to ttf
22:29:06raptoroh?
22:29:13Watusimotoand the tool I'm using seems to seriously degrade the quality
22:29:20raptorah ha
22:29:30WatusimotoI don't think our font stuff handles otf
22:29:41Watusimotoif it does, I've been doing a lot of extra work
22:30:05raptortry it
22:30:11raptorbut probably not
22:30:57raptoryeah i don't hink it does
22:31:03raptorstb_truetype only does ttf
22:31:23sam686 Quit (Read error: Connection reset by peer)
22:32:18bobdaduck has joined
22:32:27raptorhunting down the screenshot bug...
22:32:34bobdaduckSomeone available?
22:33:29raptoroh, haha - if it's smaller than 800x600, then it keeps that size
22:33:39raptorand that's always been there...
22:34:29raptorhi bobdaduck
22:34:45raptori'm not super available (about to leave for home), but may be able to help?
22:35:16sam686 has joined
22:35:16bobdaduckI have an ontick function that is running twice simultaniously while one returns true and the other returns false
22:35:16ChanServ sets mode +v
22:35:37raptornothing runs simultaneously
22:35:54bobdaduckIts supposed to be running once
22:36:05bobdaduckbut it is running through the code twice
22:36:12raptorah
22:36:49raptoryou could find the callback to track it down: http://stackoverflow.com/questions/10838961/lua-find-out-calling-function
22:37:14bobdaduckAnd when the second one runs it returns false, resetting things that aren't supposed to be reset
22:38:30bobdaduckwell obviously onTick is calling it
22:38:56raptoryes, but is this happening in the same tick?
22:39:01bobdaduckyeah
22:39:12bobdaduckor at least close enough together to count
22:39:21raptorthere's a huge difference
22:39:54bobdaduckNot in this situation, I don't think
22:39:58raptorif both calls are in the same onTick, then it can be found in a different way
22:40:06raptorhuge difference in finding the problem
22:40:36bobdaduckWell the way it works is onTick calls the function every tick, and then it only actually does things inside the function every 1000 milliseconds
22:41:01raptorah ok
22:41:07bobdaduckits the stuff inside the function that is going off funkey
22:41:25raptormaybe your setting the return variable back to false too early
22:42:30bobdaduckThe beginning of the loop looks like this
22:42:30bobdaduckhttp://pastie.org/8463972
22:42:41bobdaduckAnd then that last if statement
22:43:03bobdaduckIt is returning both the true if value, and the false elseif value
22:43:30bobdaduckLoop may be the wrong word. Chuck of code.
22:45:26raptorok, you need to find out if it is in the same onTick or not
22:45:42raptorsince it's in a loop, it could be the same onTick
22:46:07raptorand somehow working on the same 'val' in spellsArray
22:46:37bobdaduckif it was working on the same val it would return true twice
22:47:12Watusimototoo roundy?
22:47:13Watusimotohttp://img35.imageshack.us/img35/9585/mq1v.png
22:47:21bobdaduckno
22:47:25kaenyou should paste the whole function
22:47:25bobdaduckJust roudy enough
22:47:37kaenI can't fix your car if you only let me look at the spark plugs
22:47:41raptorhaha
22:47:47bobdaduckThis is in DnD.levelgen
22:47:53bobdaduckthe "whole function" is huge.
22:48:04kaenWatusimoto: that looks great!
22:48:07kaenis it the Ubuntu font?
22:48:15raptorthe 'm' 'w' and 'k' are better than tenby
22:48:16Watusimotoworks at small sizes
22:48:17Watusimotohttp://img820.imageshack.us/img820/9827/mb7f.png
22:48:36bobdaduck"threat level"
22:48:37WatusimotoNo comment on the source
22:48:40kaenyep, must be
22:48:43raptorhahaha
22:48:46kaenI can tell by the u and the d :)
22:49:07Watusimotowell, I'm running out of options...
22:49:11raptorif possible, don't use a converted OTF font...
22:49:13kaenhey, it's a great font
22:49:23kaenthey invested some serious capital in that whole set
22:49:24Watusimotoproperly licensesd, high quality, ttf format
22:49:29raptorfind the sources to the fonts
22:49:58raptori have to go now, though - back tonight
22:50:05kaenhave fun!
22:50:14raptorsorry bobdaduck
22:50:27YoshiSmbsee you raptor.
22:50:41bobdaduckbye
22:51:21raptor Quit ()
22:51:52bobdaduckOkay well
22:51:58bobdaduckI added break to the end of each statement
22:52:01bobdaduckand that seemed to fix it
22:52:03bobdaduckso whatever
22:52:16bobdaduckI probably used ipairs instead of pairs and that screwed everything up or something equally silly
22:52:18bobdaduck Quit (Remote host closed the connection)
22:54:57kaenI've never met a new programmer so reluctant to post his code
22:55:01Invisible1 Quit (Ping timeout: 252 seconds)
23:09:02Watusimotothis one has the advantage of packing lots of text into a small area, and still being readable
23:09:06Watusimotohttp://img197.imageshack.us/img197/89/sck3.png
23:09:29kaenI think it's way too thin
23:10:06Watusimotoyet, ironically, too thick
23:26:47kaenWatusimoto: what do you think about piggy-backing the test suite onto the bitfighter executable?
23:27:02kaenmaybe having a cmake option which controls whether the tests get built
23:27:44kaenI took another crack at making a libbitfighter and had the same useless result
23:28:18kaenI think it has to do with tnl's use of macros, but I haven't been able to pin down a cause
23:34:17Watusimotonot sure I understand
23:35:36Watusimotoyou mean build both at once?
23:36:02Watusimotoor creating a combined exe?
23:37:37kaencreating a combined exe
23:38:05kaenadd like -test flag that throws control over to the googletest main()
23:38:33Watusimotowhy is that better than what we have now?
23:38:54kaenbecause right now you have to compile the files in /zap twice -- once for bitfighter, once for the test suite
23:39:33Watusimototrue
23:39:36kaenand when I tried (twice now) to put the zap files into their own library you couldn't actually play a game because no gametypes were found
23:39:46Watusimotooops
23:39:50kaenwhich indicates a static intialization error
23:40:05kaenand I'd be surprised if that were the only one
23:40:15Watusimotobut how would that work with vc++ and other platforms where you don't build with cmake?
23:40:32kaenyou always generate the build files with cmake
23:40:41kaenso you'd do like: cmake .. -DBUILD_TESTS
23:41:28kaenand then it would build the tests into the executable
23:44:35Watusimotoso your idea was to pull out the zap stuff the way tnl is pulled out
23:44:46Watusimotoso it would build once instead of twice
23:45:44kaenright
23:45:51Watusimotoalso... I undid a couple of changes you committed earlier
23:46:02kaenok
23:46:20Watusimotothey seem to be based on assumptions that are true now but could (theoreticlly) change
23:46:20kaenlooking ...
23:46:29Watusimotochecking failed
23:47:05kaenah
23:48:27Watusimotothat looks like its working
23:48:31Watusimotoyes
23:48:32BFLogBot Commit: c589b09531a1 | Author: watusimoto | Message: Remove unneeded parameter/mVar
23:48:34BFLogBot Commit: 4b7a3a9ad12b | Author: watusimoto | Message: Undo revisions based on assumptions that are currently true but might change at some point in the future (though they probably won't)
23:48:35BFLogBot Commit: 0e4730231321 | Author: watusimoto | Message: Merge
23:48:36Watusimotojust a small thing
23:48:56kaenI thought it would be those two :)
23:49:14Watusimotowhat else could it be?
23:49:32kaenthey make warnings in gcc 4.8.1 and clang 3.2.7
23:49:39Watusimotoreally?
23:49:47kaenand debian requires compiling without warning with -Wall
23:49:53kaenyes, autological comparison
23:50:05kaena uint will always be >= 0 (the value of those constants
23:50:10kaenbut I have another fix, don't worry
23:50:27Watusimotoit will always be true, true
23:50:32Watusimotobut they aren't exactly constants
23:50:39kaenenums, right?
23:50:44Watusimotothey're more like variables set to constant
23:50:50kaenok
23:50:51Watusimotono, those aren't enum vals
23:51:07Watusimotothey're variables that are initialized to enum vals that happen to be 0
23:51:29Watusimotobut if the enum changes, they might point to the 3rd value in the enum instead of the first
23:51:48sam686 Quit (Write error: Connection reset by peer)
23:52:06Watusimotobut surely the compiler would optimize that check away
23:52:46kaenok, so what I'm going to do
23:52:48kaentry
23:52:56kaenis add a DUMMY first value to that enum
23:53:09kaenthat way DUMMY is zero, and it's comparing against 1
23:53:14kaenshould silence the warning
23:53:24Watusimotothat would probably work, but that just proves that the warning is bogus
23:53:33kaenthe warning is undoubtedly bogus
23:53:40kaen15:49 <kaen> and debian requires compiling without warning with -Wall
23:54:24Watusimotoanother option would be to make to make FIRST_PRINTABLE_KEY not be const
23:54:47kaentrue, but that seems less correct
23:55:21kaenunless I misunderstand the system.
23:55:38Watusimotowell, it is a constant
23:56:05sam686 has joined
23:56:05ChanServ sets mode +v
23:56:16kaenthat const might one day save us from if(FIRST_PRINTABLE_KEY = inputCode) // oops
23:56:34Watusimotoyes
23:56:46Watusimotonow that should get a warning
23:57:17Watusimotodoes gcc have a way of disabling specific warnings inside of a block?
23:57:23Watusimotolike #pragma?
23:58:04Watusimotohttp://stackoverflow.com/questions/965093/selectively-disable-gcc-warnings-for-only-part-of-a-translation-unit

Index Search ←Prev date Next date→

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