#bitfighter IRC Log

Index Search ←Prev date Next date→

IRC Log for 2011-10-09

Timestamps are in GMT/BST.

00:08:00sam686 Quit (Ping timeout: 258 seconds)
00:09:00sam686 has joined
00:09:00ChanServ sets mode +v sam686
00:25:00Zoombersam686 there?
00:25:00sam686hi
00:30:00Zoombercan you test something for me?
00:30:00sam686test what?
00:32:00Zoombervnc://home.maxhushahn.com/
00:33:00sam686that vnc:// link don't work when inserting in web browser (chrome)
00:33:00Zoomberthats because only safari knows what to do with vnc://
00:34:00sam686stupid smily..
00:34:00Zoomberbut just try typing the address with port 5900 in vnc
00:34:00sam686it may work i think, at least http://home.maxhushahn.com:5900/ is getting me RFB 003.889
00:35:00Zoomberyeah
00:35:00Zoomberi can get to it, but ii cant test for real as I am local
00:35:00Zoomberit works on my imac, but then it mirrors the image forever
00:36:00sam686password?
00:37:00sam686i think it works (or there is no problem with port forwarding)
00:37:00sam686but, can't see without password
00:37:00Zoomberi messaged you the pass
00:38:00Zoombersee if you can type into irc
00:39:00Zoombergot on?
00:45:00Flynnn has joined
01:19:00koda Quit (Quit: k thx bai)
02:24:00raptorargh!
02:24:00raptori can't for the life of me solve this bug...
02:33:00karamazovapyI have bugs
02:33:00karamazovapythey gather on my storm door
02:34:00BFLogBot - Commit 766846f474c1 | Author: buckyballreaction | Log: Make levelgen items use an EditorObjectDatabase. Why does script toggle still not work in editor??? Also, some minor clean-up
02:36:00raptorok, i have given up for now
02:36:00raptoras a bike mechanic, the rule was: if you've spent more than an hour on a bike, it's time to step away and let someone else look at it
02:37:00karamazovapynot a bad rule
02:37:00karamazovapyI should let someone else take a crack at my thesis
02:39:00raptoris that legal?
02:39:00raptor:)
02:42:00karamazovapyit's your rule, not mine...
02:43:00raptori suppose if I was stuck on a particular idea, i'd go talk with another person in the field
03:11:00raptornow that i have a little LUA coding practice... I should start on enhancing bots again
03:15:00raptoroh wait! I was also in the middle of soccer changes...
03:17:00raptorsam686: are you awake?
03:34:00BFLogBot - Commit 703865e7d263 | Author: buckyballreaction | Log: Add Editor option to disallow robots. Finishes issue 65
03:41:00raptornow to take a stab at that soccerball again...
03:43:00Pointblank Quit (Ping timeout: 260 seconds)
03:48:00Pointblank has joined
03:50:00karamazovapyzoomber hoards nazigold
03:50:00karamazovapyzoomber loves nazigold
03:50:00raptoryou crack me up
03:50:00karamazovapynazigold pays for zoomber's apple investments
03:51:00karamazovapyif we're going to have a robot that learns, we might as well be teaching it something worthwhile
03:53:00Pointblank Quit (Ping timeout: 244 seconds)
03:57:00Pointblank has joined
04:03:00raptorOk testing out some soccer ball velocity transfer adjustments...
04:12:00raptorit works!
04:13:00raptorhmmm, except transfer back to ship isn't working so much...
04:58:00karamazovapywhat are you doing with soccer?
04:59:00raptorgetting it to have more 'inertia'
04:59:00karamazovapyah
04:59:00raptorvia velocity transfer...
04:59:00karamazovapyI'll be curious to see how that pans out
04:59:00raptorI think i finally found the right spot to adjust the velocities
05:00:00karamazovapyhow does it feel in testing?
05:00:00raptorwell it only works with hitting with weapons right now...
05:00:00raptori kind of like it
05:01:00raptori used a velocity transfer ratio of 3
05:01:00raptor10 is too much
05:01:00raptori'm thinking 5 or 6
05:01:00karamazovapywhat was the equivalent originally?
05:01:00raptor1
05:02:00raptori'm being very lazy mathematically
05:02:00karamazovapyso if the ratio is 3, only 1/3 of the energy transfers?
05:03:00raptoryes
05:05:00karamazovapyif there's a relatively minor alteration that "fixes" soccer, I think this is it
05:05:00raptoryes, i already like it better
05:05:00raptori just have to get the transfer to work with ships...
05:06:00raptoruhh...
05:06:00raptor Quit (Remote host closed the connection)
05:09:00raptor has joined
05:09:00ChanServ sets mode +o raptor
05:09:00karamazovapywelcome back
05:09:00raptorhardware problems on my laptop...
05:09:00raptorso it looks like I found the right spot to transfer velocity to the ship
05:09:00raptoronly i did it backwards...
05:09:00karamazovapyahh
05:10:00raptorhahaha
05:10:00raptorthis is awesome
05:10:00raptori tap the soccerball and i go flying backwards now...
05:10:00karamazovapythat's pretty good
05:11:00raptori think an integer ratio may be too much...
05:12:00raptormaybe I should have a return vector...
05:12:00raptor*shouldn't
05:13:00raptoryeah 5 is too much...
05:14:00raptorok i need to think about this
05:14:00karamazovapywell I think there's more than one question
05:14:00raptorfor each collision there is a force transfer one way and an opposing force
05:16:00karamazovapywhat if we tweak velocity transfer to have different values for weapon > ball, ship > ball, and ball > ship?
05:16:00raptorwe can do that
05:16:00raptornot hard
05:16:00raptorit's the ship > ball and ball > ship that i'm trying to get right
05:17:00karamazovapywell ship>ball/ball>ship, I imagine a person pushing a weight on ice
05:17:00karamazovapyif the weight hit you, you'd both keep moving...you wouldn't get a newton's cradle effect
05:18:00raptorhmm yeah
05:19:00karamazovapyI know that's not how our physics work generally
05:19:00raptori think instead of adjusting the reverse vector with the ratio, i need to subtract an amount from it instead...
05:19:00raptorbased on the ratio..
05:21:00karamazovapyyou know, I think I'd need to play around with it to get a handle on what kind of behavior I'd really expect
05:22:00raptorok, i will make you a build as soon as I get the math right...
05:22:00karamazovapyno rush on that
05:22:00karamazovapyI'm curious, but not in any hurry
05:30:00raptorok, I think I got it, more or less, except the ball comes to a complete stop when hitting the ship...
05:31:00raptorwhich now that i think about it - does anything else in the game actually keep 'travelling through' as it moves?
05:32:00raptorand i like how the stars shine through the ball...
05:38:00karamazovapywe have newton's cradle physics
05:38:00karamazovapyeverything currently has 100% energy transfer with no drift
05:39:00raptoryeah... that's the hard part...
05:40:00raptoranyways
05:41:00raptorgive me 7 minutes and I can get you a build with the velocity transfer working...
05:41:00karamazovapyalright...I'm done writing for the evening anyway
05:45:00raptorugh, i may have to add a few minutes onto that ETA - windows compiling is a bit slower...
05:51:00raptorok karamazovapy
05:51:00raptorhere is your build: http://208.107.52.15/upload/Bitfighter-Installer-016-soccer-adjustment.exe
05:52:00raptorit is using a velocity transfer ratio of 4
05:54:00raptorargh, burst still sends the ball flying...
05:57:00karamazovapyokay!
05:57:00raptori think it's a little better than the original..
05:57:00karamazovapyso I'd leave ball > ship transfer at 1
05:57:00karamazovapyand I'd get rid of the ball/ship impact sound
05:58:00raptoroh, haha - i never do tests with sounds...
05:58:00karamazovapybut I think the ship > ball and weapon > ball transfers are better
05:58:00raptorok
05:58:00karamazovapythanks to the lower energy transfer, ship/ball contact fires the sound over and over, like running into the wall
05:58:00raptorlet me make some quick adjustments
05:59:00raptorreally? yuk
06:02:00raptoroh yeah, much better with ball > ship at 1
06:03:00raptorlet me see about that impact sound..
06:06:00raptorkaramazovapy: do you think I should get rid of the sound completely?
06:06:00karamazovapyyes
06:06:00raptorwhen ball hits ship or vice/versa
06:06:00raptorok, i think that's easy...
06:08:00sam686 Quit (Ping timeout: 258 seconds)
06:09:00sam686 has joined
06:09:00ChanServ sets mode +v sam686
06:14:00raptorkaramazovapy: still up?
06:14:00karamazovapyyep. I'm teaching AP Calc on friday, so I'm reviewing some stuff
06:17:00raptorok, here is another adjustment: http://208.107.52.15/upload/Bitfighter-Installer-016-soccer2.exe
06:17:00raptortell me what you think
06:22:00karamazovapyI like it
06:23:00raptorgreat!
06:23:00raptori do too
06:23:00karamazovapythe only thing I'd tweak is that the ball seems to stutter on continued contact
06:23:00raptorthe ball or the ship?
06:23:00karamazovapythe ball
06:24:00karamazovapyalthough it's hard to say by memory...let me look
06:25:00karamazovapyyeah, it's the ball that I notice, although the ship's motion is halted in the process
06:25:00raptordid you look at it via commander's map?
06:26:00karamazovapyno...I can give that a shot
06:26:00raptoryeah, do that - it actually looks like all ship from that view...
06:27:00karamazovapyyeah, you're right
06:27:00karamazovapythe ship might be the real source of the stutter
06:28:00karamazovapymaybe it's some kind of minor ball>ship pushback?
06:28:00raptoryes, there is still a minor vector pushback on all objects that move when they collide
06:28:00raptororiginally it was this that i multiplied by the ratio
06:29:00raptorbut i removed it
06:29:00karamazovapyyeah
06:30:00raptorhow do you disallow soccer pickup in 015a?
06:30:00karamazovapyI think it's soccerpickup=no in the level file
06:31:00karamazovapySoccerPickup no
06:31:00karamazovapy(fixed)
06:32:00karamazovapyyeah, the 015 ball reaction seems way overblown after that 016 test
06:32:00raptorhaha, sure does
06:32:00karamazovapyit just jumps off the ship
06:33:00raptoryep
06:33:00raptordo you think the current ratio is OK?
06:33:00raptorfor 016?
06:34:00raptoralso, i'm not quite sure how to fix that stuttering with the ship... yet
06:34:00raptori'll need to sleep on it
06:35:00raptoractually i'm slipping into unconsciousness now... have a good night
06:36:00raptor has left
06:36:00Zoomberand im back
06:36:00Zoomber Quit (Quit: Zoomber)
06:39:00Zoomber has joined
06:39:00ChanServ sets mode +v Zoomber
06:40:00karamazovapyholler
06:41:00Zoomberhi
06:41:00karamazovapyhow's it goin?
06:48:00Zoomberi just went to see this jazz concert (http://www.jasonleebruns.com/) in news, it shows he was having a jazz collective today
06:50:00karamazovapyI didn't realize Jamie Kennedy was a percussionist
06:50:00Zoomberoh, in other news, i finally turned my "qwop" netbook into shiny platinum by finally getting QuartzGL to work
06:58:00sam686 Quit (Ping timeout: 258 seconds)
06:58:00sam686 has joined
06:58:00ChanServ sets mode +v sam686
06:58:00Zoomberhi sam686
06:59:00Zoomber_k friends with blackbird
06:59:00Zoomberblackbird _k
06:59:00Zoomberzoomber nazigold
06:59:00karamazovapynazigold zoomber
06:59:00Zoomberblackbird karamazov bb
07:00:00Zoomber_k is acquaintance blackbird
07:00:00Zoomber_k karamazovapy batman karamazov
07:02:00Zoomberkaramazovapy billydeewillams bobsaget santana joedirt donaldtriumph snoopd
07:02:00Zoomberand more
07:02:00Zoomberpointblank points at blank points blanking out to point out that blanking points point blanks out
07:19:00Zoomberchecking out karamazovapy 's new zc
08:51:00Zoomberzoomberout goodnight karamazovapy karamazov batman pointblank
08:51:00Zoomber Quit (Quit: Zoomber)
08:57:00Flynnn Quit (Quit: This computer has gone to sleep)
09:28:00Pointblank Quit (Ping timeout: 248 seconds)
09:32:00Pointblank has joined
13:55:00watusimoto has joined
14:01:00raptor has joined
14:01:00ChanServ sets mode +o raptor
14:02:00raptorgood morning!
14:07:00raptorwatusimoto, sam686: is dynamic_cast an expensive operation?
14:08:00sam686not sure, speed test?
14:09:00raptorI don't want to spend the time to build test code at the moment... I'm almost done with these soccer changes
14:11:00sam686i don't know the speed to dunamic cast, though i guess it might be slow due to having to search for if it can be dynamic_cast
14:12:00raptorfound this: http://tinodidriksen.com/2010/04/14/cpp-dynamic-cast-performance/
14:19:00raptorthat's a neat blog - there are lots of c++ performance comparison tests on it
15:11:00raptorsam686: are you free to help me a bit?
15:15:00sam686|2 has joined
15:15:00sam686 Quit (Disconnected by services)
15:15:00sam686|2 is now known as sam686
15:15:00ChanServ sets mode +v sam686
15:15:00sam686pl
15:15:00sam686ok
15:16:00raptorlet me get you a diff
15:16:00raptorhere: http://208.107.52.15/upload/soccer.diff
15:16:00raptorit is my attempt at increasing the ball 'mass'
15:16:00raptorfor issue 64: http://code.google.com/p/bitfighter/issues/detail?id=64
15:16:00raptorworks relatively well
15:16:00raptorexcept there are a couple of physics problems
15:17:00raptorone is that constantly pushing the soccer ball makes the ship jump backwards
15:20:00raptorthere are three physical interactions with the ball that need to be accounted for
15:20:00raptorship hits ball, ball hits ship, weapon hits ball
15:20:00sam686|2 has joined
15:20:00sam686 Quit (Disconnected by services)
15:20:00sam686|2 is now known as sam686
15:20:00ChanServ sets mode +v sam686
15:21:00raptorship hits ball, weapon hits ball interactions have their velocity adjusted
15:21:00raptorball hits ship is left alone - other wise the ship goes flying in other directions
15:22:00sam686i think i have a stupid problem with my computer having 2 ethernet bridged, if one connect or disconnect, i get disconnected from IRC...
15:23:00sam686as usual.. compiling with that patch..
15:30:00sam686ok, i see ball mass is higher, but shooting won't affect from mass..
15:30:00raptorshooting does affect it, just less
15:31:00sam686it still look like shooting don't affect it at all..
15:31:00raptorhmmm, maybe i gave you wrong path..
15:31:00raptorpatch
15:31:00raptorone second...
15:32:00sam686in some cases, shooting burst at the soccer make burst blow up immediately
15:32:00raptoryes, that another minor problem
15:32:00raptoryeah i gave you th right patch
15:33:00raptorshooting still affects it but should be slower
15:34:00raptorcompare with 015a to notice the difference
15:34:00sam686i think you only did it on GrenadeProjectile, not phaser or bouncer (Projectile)
15:35:00raptori did in soccergame.cpp
15:35:00raptorin SoccerBallItem::damageObject
15:35:00sam686look at http://208.107.52.15/upload/soccer.diff
15:36:00raptoryes
15:36:00raptorProjectile is not a moveObject
15:36:00raptordo I had to add the velocity adjustment to SoccerBallItem::damageObject
15:36:00sam686- mMoveState[ActualState].vel += iv * dv.dot(iv) * 0.3f;
15:36:00sam686+ mMoveState[ActualState].vel += (iv * dv.dot(iv) * 0.3f) / VELOCITY_TRANSFER_RATIO;
15:36:00sam686(basically, nothing changed)
15:37:00sam686oh, now i see
15:37:00raptorwell, there is a noticeable change when shooting the ball compared to 015a
15:37:00sam686its just that that patch somehow hates soccerGame.cpp while patching and did nothing with it
15:37:00raptorah
15:37:00raptorhaha
15:38:00raptorbe right back...
15:39:00sam686should mines blow up when soccer ball get too close to mines?
15:40:00sam686right now, soccer just go over mines, and mines don't blow up
15:43:00sam686while it takes a lot to move, ships can often instantly stop the soccer from moving when soccer hits ship
15:43:00sam686part of it is due ship's very high acceleration and deceleration
15:47:00sam686it is likely "Mass" need to be created in MoveObject instead of soccerBall
15:59:00raptorok back
15:59:00raptoryes
16:00:00raptorthe 'instantly stop' piece is because we have what karamazovapy called 'newton's cradle physics'
16:00:00sam686there is mMass in Item class, whick all MoveObject use Item class
16:00:00raptoryes, but mMass isn't used anywhere
16:00:00raptori don't htink
16:00:00sam686can make it use mMass instead
16:01:00raptorhttp://en.wikipedia.org/wiki/Newton's_cradle
16:01:00sam686as, Mass can be useful to more then one object
16:01:00raptorthat would probably be ideal
16:01:00raptorbut there is no math or logic done with mass anywhere
16:01:00sam686can add some logic to mass..
16:02:00raptormaybe long term the mass solution is better
16:03:00raptormy code is really a hack
16:03:00raptorbecause i
16:03:00sam686that is a mess of dynamic_cast
16:03:00raptoram simulating mass with velocity
16:03:00raptorbig mess
16:03:00raptorand i don't like it
16:04:00raptori know how to solve it
16:04:00raptorwith a virtual method on MoveObject - at lease as far as the sound goes
16:05:00raptorbut the real reason I asked for your help is to make better code with the velocity transfer, or if needed introduce mass
16:05:00sam686mMass is often set in constructor
16:10:00sam686TEST_ITEM_MASS = 4 it seem like someone want more mass to huge bouncy object
16:10:00raptorhuh, i see varying masses throughout the code
16:10:00raptoryeah just saw that..
16:10:00raptorbut i see no math with the mass
16:10:00raptorlet me try increasing soccer mass to like 20
16:13:00sam686your math has a problem, a slow moving TestItem with mass=4 can send ResouceItem flying at 4 times the speed with Mass=1
16:14:00raptorwhere is the math that takes mass into account?
16:14:00sam686MoveObject.cpp line 458 (i added some things)
16:14:00sam686v2f = v2f * mMass / moveObjectThatWasHit->mMass; v1f = v1f / mMass * moveObjectThatWasHit->mMass;
16:15:00sam686v1f = v1f / mMass * moveObjectThatWasHit->mMass;
16:15:00sam686that is mostly the same as your soccer mass idea..
16:16:00raptoris there any math in the code that uses mass, that you have not written?
16:17:00sam686i don't see any mass code used in moveObject for calculation
16:17:00raptorok, so this is new
16:17:00raptorgood
16:17:00raptorthat means this will alter the fundamental physics of the game
16:17:00raptorso we have to be careful
16:17:00raptori'm trying your code now...
16:20:00raptori htink i like your solution with the mass better, but it still has the kickback to the ship...
16:22:00raptoryeah with these physics you can really get a galileo's cannon going
16:27:00watusimoto Quit (Ping timeout: 240 seconds)
16:46:00raptorbe back in 4 hours...
16:46:00raptor Quit (Remote host closed the connection)
17:32:00LordDVG has joined
17:44:00LordDVG Quit (Ping timeout: 252 seconds)
17:47:00LordDVG has joined
17:55:00koda has joined
18:00:00watusimoto has joined
18:24:00watusimotois dynamic_cast an expensive operation? >>> yes
18:24:00sam686 Quit (Ping timeout: 258 seconds)
18:24:00sam686 has joined
18:24:00ChanServ sets mode +v sam686
18:36:00watusimotohttp://stackoverflow.com/questions/332030/when-should-static-cast-dynamic-cast-and-reinterpret-cast-be-used
18:36:00watusimotoRecently, I've been using dynamic_cast when I'm going to check for NULL, c style casts when I;m pretty sure of the result... I think I'll start migrating teh c-style casts to static_casts
18:46:00BFLogBot - Commit cab25a8004e8 | Author: sam8641 | Log: Make Mass work, move mMass to MoveObject
18:48:00sam686 Quit (Ping timeout: 258 seconds)
18:48:00sam686 has joined
18:48:00ChanServ sets mode +v sam686
18:52:00watusimotoI've been looking at the bot stuff, and think we would improve server perofrmance if we didn't keep wiping out and recreating our L object every time a script runs
18:52:00watusimotohere is one possible solution
18:52:00watusimotohttp://www.keplerproject.org/venv/index.html#status
19:57:00LordDVG Quit (Remote host closed the connection)
20:11:00Flynnn has joined
20:18:00Zoomber has joined
20:18:00ChanServ sets mode +v Zoomber
20:19:00Zoomberhi
20:20:00raptor has joined
20:20:00ChanServ sets mode +o raptor
20:20:00raptorhowdy folks
20:20:00Zoomberhi raptor
20:23:00raptorgood clean-up sam686 with the mass code
20:28:00raptorhi watusimoto
20:30:00raptorso essentially instead of using (S32), we should be happily verbose with static_cast<S32>()
20:33:00raptoroh my goodness sam686, you made mass transfer work really nicely
20:33:00raptorwow, waaaaay better than before
20:33:00raptorgood job
20:35:00raptorinteresting, you added elasticity to the other moveObject that is hit
20:35:00raptorwas that the key?
20:36:00watusimotohi all
20:36:00raptorhello
20:36:00watusimotowell... (S32) seems like such a trivial case
20:36:00raptor:)
20:36:00watusimotowe don't want to take it too far, do we?
20:37:00raptoryeah - we don't want our code to look like boost headers
20:37:00watusimotobest to resolve those by changing the types where possible
20:37:00watusimoto:-)
20:37:00watusimotoso I'd say stick with c-style for int->int casts and such.
20:37:00raptorme too
20:37:00watusimotouse static_cast for pointers
20:37:00watusimotounless you want to check the results, then use dynamic_cast
20:37:00raptoragreed
20:38:00watusimotobut if you don't check, it seems pointless to use dynamic
20:38:00raptoryep
20:38:00watusimotoso here's a bot question
20:38:00raptorok
20:38:00watusimotodo we want to continue to support bots that have no script?
20:38:00raptor??
20:38:00watusimotothat is, that just sit there and do nothing?
20:38:00watusimotoor, if a bot has no script, should it just never be created?
20:38:00Zoomber?
20:39:00raptoryou mean those white ghost ships?
20:39:00watusimotowhite ghost bots, I think
20:39:00watusimotoI'm not sure how to create one, but we have code to allow them to exist
20:39:00raptorthose were never actually bots, technically... right?
20:39:00raptori thought they were 'Ship' in the level code or something like that..
20:39:00Zoomberhow about a re-entry of the "Botsperteam" feature?
20:40:00watusimotoit's what the variable isRunningScript is used for, I think
20:40:00Zoomberthey are SHIP 0 x y
20:40:00Zoomberor ship x y
20:40:00Zoomberdont remember
20:40:00watusimotoI'm not talking about ships, only bots here
20:40:00watusimotoships is a different conversation
20:40:00raptorah ok
20:40:00watusimotoI'd like to get rid of that variable
20:41:00raptorwell, i've never seen bots that actually don't do anything...
20:41:00watusimotowe have other checks that may prevent them
20:41:00raptorscrap it, i'd say
20:41:00watusimotobut I *think* that variable only supports scriptless ships
20:41:00watusimotopart of the problem is that I find it's usage a bit confusing
20:41:00Zoomberwhat exactly do you mean by scriptless ship. a ship thats added wihtout the use of scripts folder?
20:41:00watusimotoI may have added it for some reason when I first started implementing bots
20:42:00raptoryeah i ran in to that confusion when trying to build bot spawning function once...
20:42:00watusimotoI mean a bot tha thas an empty or invalid script can, theoretically, live on
20:42:00raptormaybe you thought that a bot could be implemented in c++?
20:42:00watusimotoNo, I never did, though Sam talked about that once
20:43:00watusimotoI'd argue for using a different class in that case, one with no in-built lua support
20:43:00watusimotobut I'm not in favor of a c++ bot anyway
20:43:00watusimotoI like ahving a default script, but why have 2 mechanims for supporting bots?
20:44:00raptoragreed
20:44:00raptoryeah try to get rid of it, i say
20:44:00watusimotoespecially if we can make lua bots more efficient
20:44:00raptoryes!!
20:44:00watusimotoby, say, resuing the lua environment
20:44:00raptoryes yes yes!!
20:44:00watusimotowhich I may take a crack at
20:44:00Zoomberso, there will always be a scripts folder required?
20:44:00watusimotomaybe... maybee... we could run all bots in one environment
20:45:00watusimoto@z yes, nothing we're discussing would change that one way or the other
20:46:00watusimotook, I'll try removing the variable and see what happens
20:46:00watusimotoMaybe I'll check in my cleaned up scripting first
20:46:00raptori spoke with the 'naev' developers about running all their AI in one environment
20:46:00watusimotoI've merged a lot of levelgen and bot stuff
20:46:00watusimotois that what they do?
20:46:00raptoryes
20:47:00watusimotoour bots are supposed to be autonomous; I could see AI elements cooperating
20:47:00raptorthey said they keep a container with information for each AI (or bot) and then use one LUA environment and switch the data pointers as they loop through each AI
20:48:00raptoror something to that effect...
20:50:00watusimotothat's similar to what I've been reading about
20:50:00watusimotohttp://www.lua.org/pil/14.3.html
20:51:00watusimotoI'm going to try implementing something like this
20:52:00watusimotocase 133 finished?
20:52:00raptordid i close that?
20:52:00watusimotooh wait...
20:52:00watusimotohaven't refreshed my case list for a while
20:52:00watusimotothat's what intermittent internet does... I keep everything open
20:53:00watusimotoand end up with a whole mass of crap
20:53:00raptorhehe
20:53:00watusimotoforget it... closed
20:53:00raptori closed a few in last couple days
20:56:00raptorwatusimoto: here was my conversation: http://colabti.org/irclogger/irclogger_log/naev?date=2011-07-18#l37
21:01:00Flynnn Quit (Quit: This computer has gone to sleep)
21:01:00BFLogBot - Commit 96238ae666c6 | Author: buckyballreaction | Log: Fewer magic numbers
21:02:00watusimoto Quit (Ping timeout: 252 seconds)
21:04:00watusimoto has joined
21:06:00watusimoto Quit (Client Quit)
21:06:00watusimoto has joined
21:06:00watusimotohrm...
21:06:00raptorwelcome back
21:06:00raptortwice!
21:09:00raptorhey karamazovapy, you've gotta try sam686's implementation of soccer; much better than mine, I think: http://208.107.52.15/upload/Bitfighter-Installer-016-soccer3.exe
21:09:00raptorif you've got the time..
21:10:00watusimoto Quit (Read error: Connection reset by peer)
21:12:00watusimoto has joined
21:12:00watusimoto!!!!!
21:13:00raptori suggest using an old coat hanger to boost the signal
21:15:00watusimoto Quit (Read error: Connection reset by peer)
21:16:00watusimoto has joined
21:22:00karamazovapyso the difference is that the ball pushes you a bit without launching you?
21:23:00karamazovapyI like it...but I still think the sound effect has to go if there's going to be sustained ship/ball contact
21:23:00raptoroh yeah...
21:23:00raptori can do that
21:25:00raptori'm just impressed because his code is so much cleaner than mine, and the 'mass' increase can be applied to any object tha tmoves now
21:29:00karamazovapyI think asteroids should have mass added to them
21:29:00raptorthey do
21:29:00raptordefault mass is 1
21:30:00raptortestitem soccerballitem asteroiditem have mass of 4
21:30:00karamazovapy...then mass should be *added*
21:30:00raptorunless you think they should have more?
21:30:00karamazovapyah
21:30:00karamazovapyI don't know if I'd change it for everything by default
21:30:00karamazovapyI mean, asteroids are "supposed" to be heavy, but testitems don't exist
21:31:00karamazovapyI dunno...maybe it'd be neat if they were heavy and resources were light
21:31:00karamazovapydo smaller asteroid bits have less mass?
21:31:00watusimoto Quit (Ping timeout: 244 seconds)
21:31:00raptorhmmm... not sure
21:31:00raptorprobably not
21:31:00karamazovapythat'd be cool
21:31:00raptorlet me check
21:31:00raptoryeah...
21:31:00karamazovapyif they went 4:2:1
21:31:00raptori agree
21:31:00raptorshould i reduce testitem mass?
21:32:00karamazovapyI'm open minded on that one
21:32:00karamazovapybut it'll be cool if the smaller asteroids are more deadly because they accelerate more readily
21:32:00raptorgood thinking
21:32:00raptorwell, test out the testitems in conjunction with resource items
21:33:00raptorto get a feel for their mass and affects
21:33:00karamazovapyI'm gonna get some food and then go see Moneyball for a little mental vacation
21:33:00karamazovapyI've heard it's good
21:33:00raptorhaha, ok
21:33:00raptorenjoy
21:47:00BFLogBot - Commit 8ab2beb24ab6 | Author: buckyballreaction | Log: Reduce mass on asteroid split
21:52:00BFLogBot - Commit 2299a5745141 | Author: buckyballreaction | Log: Make second asteroid piece have adjusted mass, too
22:03:00Flynnn has joined
22:31:00raptorsam686: your new zeros look nice
23:05:00raptori think soccor scores on both goals gives a point to team 1
23:05:00raptoror your team, i mean

Index Search ←Prev date Next date→

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