IRC log for #bzflag on 20120824

00:00.00blast007gonna do some forum maintenance quick
00:01.49Guest4003opencombat: 03jeffm2501 * r217 10/trunk/unit0/ (12 files in 6 dirs): move avatar drawing to it's own lib to make the world drawing code easier to use in an editor.
00:01.52JeffMdelete all the trolls
00:03.12Guest4003opencombat: 03jeffm2501 * r218 10/trunk/unit0/WorldDrawing/WorldDrawing.csproj: remove actor files
00:03.14blast007it seems like phpBB may be adding the ability to send PMs to a group
00:03.28blast007but it might be a 3.0.11 feature, and 3.0.11 isn't quite out yet
00:03.42I_Died_Oncesounds like a spammers dream
00:05.01*** join/#bzflag L4m3r (~l4m3r@bzflag/developer/L4m3r)
00:05.01*** mode/#bzflag [+v L4m3r] by ChanServ
00:06.46blast007*shrug*
00:13.08I_Died_Oncebzbb is down!
00:13.10I_Died_Onceducks
00:14.56Guest4003opencombat: 03jeffm2501 * r219 10/trunk/unit0/ (12 files in 3 dirs): Start new editor with a better view paradigm.
00:16.42*** join/#bzflag I_Died_Once (~I_Died_On@unaffiliated/idiedonce/x-1828535)
00:22.41blast007I_Died_Once: no it's not  ;)
01:08.56*** join/#bzflag jeffm (~jeffm@cpe-76-167-239-8.socal.res.rr.com)
01:08.56*** join/#bzflag jeffm (~jeffm@unaffiliated/jeffm2501)
01:08.56*** mode/#bzflag [+v jeffm] by ChanServ
01:22.27apeman_which bzflag map editor can handle arcs?
01:22.54jeffmprobably only bzworkbench
01:23.01jeffmif they are implemented
01:30.40jeffmI know bzedit dosn't
01:45.51*** join/#bzflag jeffm (~jeffm@cpe-76-167-239-8.socal.res.rr.com)
01:45.52*** join/#bzflag jeffm (~jeffm@unaffiliated/jeffm2501)
01:45.52*** mode/#bzflag [+v jeffm] by ChanServ
01:53.42Foo_man_chooplays
02:21.16Foo_man_chooFoo_man_choo pwnt to a crisp
02:21.42*** join/#bzflag guest2343 (~nb_j@host85.186-109-235.telecom.net.ar)
03:04.33*** join/#bzflag javawizard (~alex@opengroove.org)
03:04.33*** join/#bzflag javawizard (~alex@bzflag/contributor/javawizard2539)
03:24.38*** join/#bzflag JeffM (~JeffM@cpe-76-167-239-8.socal.res.rr.com)
03:24.39*** join/#bzflag JeffM (~JeffM@unaffiliated/jeffm2501)
03:24.39*** mode/#bzflag [+v JeffM] by ChanServ
03:34.24*** join/#bzflag heartnet (~heartnet@expr.prosbloom.net)
03:39.28*** join/#bzflag heartnet (~heartnet@expr.prosbloom.net)
03:49.14*** join/#bzflag heartnet (~heartnet@expr.prosbloom.net)
03:56.03Guest4003opencombat: 03JeffM2501@gmail.com * r220 10/trunk/unit0/PlanEdit/ (12 files): stub out files for controller and views
04:15.53*** join/#bzflag I_Died_Once (~I_Died_On@unaffiliated/idiedonce/x-1828535)
04:23.34*** join/#bzflag allejo (~allejo@cpe-76-90-186-178.socal.res.rr.com)
04:39.33Guest4003opencombat: 03JeffM2501@gmail.com * r221 10/trunk/unit0/PlanEdit/ (6 files): load prefs and pick data dir so views and editors can know stuff.
04:59.23*** join/#bzflag ashvala (~ashvala@unaffiliated/ashvala)
05:01.04*** join/#bzflag allejo (~allejo@cpe-76-90-186-178.socal.res.rr.com)
05:16.48Constitutionkills Guest4003
05:17.00Constitutionhm
05:22.05*** join/#bzflag Constitution (~josh@bzflag/developer/Constitution)
05:25.42*** join/#bzflag ashvala (~ashvala@unaffiliated/ashvala)
05:55.09*** join/#bzflag ashvala (~ashvala@unaffiliated/ashvala)
06:53.16*** join/#bzflag ashvala (~ashvala@unaffiliated/ashvala)
06:56.05*** join/#bzflag sirquine (~quine@c-76-120-71-113.hsd1.co.comcast.net)
07:05.56*** join/#bzflag TimRiker (~timr@bzflag/projectlead/TimRiker)
07:05.56*** mode/#bzflag [+o TimRiker] by ChanServ
07:09.29*** join/#bzflag ashvala (~ashvala@unaffiliated/ashvala)
07:26.05Guest4003BZFlag: 03Allejo 07http://wiki.bzflag.org * r8374 10/Functions(API): bz_killPlayer(int playerID, bool spawnOnBase) works also
08:27.28Guest4003BZFlag: 03Piepiepie154 07http://wiki.bzflag.org * r0 10/User:Piepiepie154:
08:36.02*** join/#bzflag apeman_ (~apeman@115-188-244-191.jetstream.xtra.co.nz)
09:02.02*** join/#bzflag Pimpinella (~frank@gondolin.pimpi.org)
09:47.58Constitution~allejo++
10:19.24*** join/#bzflag jcp (~alex@opengroove.org)
10:19.25*** join/#bzflag jcp (~alex@bzflag/contributor/javawizard2539)
10:25.13*** join/#bzflag javawizard (~alex@opengroove.org)
10:25.13*** join/#bzflag javawizard (~alex@bzflag/contributor/javawizard2539)
11:06.44*** join/#bzflag macsforme (~josh@bzflag/developer/Constitution)
11:16.39*** join/#bzflag EvilJStoker (jstoker@unaffiliated/jstoker)
11:17.12*** join/#bzflag ashvala (~ashvala@unaffiliated/ashvala)
22:20.50*** join/#bzflag ibot (~ibot@rikers.org)
22:20.50*** topic/#bzflag is http://BZFlag.org || http://cia.vc/stats/project/BZFlag || http://wiki.BZFlag.org/Getting_Help || Channel Logs: http://ibot.rikers.org/%23bzflag/ || 2.4.2 is released, deal with it, purple monkey dishwasher
22:20.51*** mode/#bzflag [+o ibot] by ChanServ
22:20.54JeffMand that is different for each player
22:21.06JeffMso is a bit of jumping worse then that?
22:21.16JeffMsince we have the same jumping now
22:21.32JeffMit's just jumping from one wrong place to another wrong place
22:22.11apeman__currently its not wrong, its just late. I can make mhy own accurate judgments about how that lateness affects things
22:23.40*** join/#bzflag ibot (~ibot@rikers.org)
22:23.40*** topic/#bzflag is http://BZFlag.org || http://cia.vc/stats/project/BZFlag || http://wiki.BZFlag.org/Getting_Help || Channel Logs: http://ibot.rikers.org/%23bzflag/ || 2.4.2 is released, deal with it, purple monkey dishwasher
22:23.41*** mode/#bzflag [+o ibot] by ChanServ
22:23.50apeman__right now DR correctly models old information, is what i'm saying
22:23.58apeman__there is no way to correctly model information you don't yet have, as in prediction
22:24.10JeffMDR is making a guess too
22:24.24JeffMthe problem is that a kill has to happen using current info
22:24.25JeffMnot that old info
22:24.29JeffMso it's a problem
22:24.46JeffMbut it's all moot, nobody will ever change it
22:25.26apeman__DR (more or less) correctly models position though
22:25.34apeman__in the past
22:25.44randomparticlenot really
22:25.47JeffMif lag is constant
22:25.58JeffMand frames are constant
22:26.12apeman__Prediction assumes a tank won't change direction within some time, but when a tank changes direction, the teleport may upset observers
22:26.16JeffMI can see how it could not have a port if the offset was constant
22:26.21randomparticlethe only positions that you know are right are those for which updates were sent
22:26.31JeffMyeah we removed the ability to change dir instantly ;)
22:26.40apeman__in 2.99?
22:26.48JeffM2.4 should have -a always on
22:26.54JeffMbut yes in 2.99
22:27.00apeman__no one likes -a
22:27.09randomparticlethe updates aren't sent very frequently
22:27.10JeffMyeah cus the like changing dir instantly
22:27.12apeman__and it doesn't solve an inherent flaw
22:27.33JeffMrandomparticle, if the updates are always send with the same delay the visual jump will be minimal
22:27.36apeman__updates are sent every time DR needs to be updated. So DR works IF lag is constant, frames are constant
22:27.49JeffMoh none of this solves the inherent problem :)
22:27.55apeman__Whereas with constant lag and constant frames, prediction does not work
22:28.18randomparticlebz used to have lag correction but it was commented out
22:28.26JeffMrandomparticle, no, that was in 2.9
22:28.37randomparticleit was there back in 2.0.2
22:28.41JeffMnope
22:28.47JeffMthe timestamp was not used
22:28.50randomparticleit was. i can show you the code :)
22:28.54JeffMI know the code
22:29.03JeffMit did not do what you think it does
22:29.03randomparticleobviously you don't
22:29.06JeffMok
22:29.07apeman__lag correction is just a buzzword i believe - a buzzword for prediction, which is not suited for tanks that change directions often
22:29.40JeffMthere are multiple forms of correction, preditction is just one
22:29.42apeman__timestamp could be used to improve current DR though
22:29.52JeffMserver side hit rollbacks can be used to compensate as well
22:30.08JeffMbassicly have the server say "did he think he hit that guy?"
22:30.26JeffMthen you can let everyone deal with delayed DR
22:30.31JeffMthat's what halflife does
22:30.36apeman__thats used in other FPS games, but its pretty bad because bullets are slow
22:31.00JeffMHL uses prediction for slow shots like missiles IIRC
22:31.05JeffMmost are instahit
22:31.32JeffMany real solution needs a server state to do anything consistent
22:31.55apeman__right, i have played grenades like that, but its definitely not intuitive in some aspects
22:32.21JeffMyeah you have to find the right method for what is fun for your game
22:32.30blast007if we just do web scale, won't that solve everything?
22:32.40JeffMgets his gun
22:33.22JeffMthe one hit to kill nature of bzflag makes latency and the client kill part seem very bad
22:33.25apeman__if you think lag in bzflag induces rage, try playing laggy chess
22:33.34JeffMwith health and the like you can use some randomess to mask lag hits
22:33.38JeffMheh
22:33.55JeffMat least that is turn based and the timelynes of the state dosn't affect the moves
22:34.03apeman__yes it does
22:34.20apeman__10 minute and blitz matches, lag does a lot to your state of mind
22:34.41JeffMsure I can see lag affecting your ability to get the state and be able to make your move
22:34.52apeman__it feels like the other guy spends 4 seconds for a move, but his clock decreases by 1
22:34.58JeffMbut that is a human agrivation problem, not a state discontinuity
22:35.11JeffMahh there is a clock, so yeah then the state will be affected
22:35.48apeman__intense 1-minute (per player) lightnign matches take 3 minutes
22:35.50JeffMif time is part of the state then latency will always have some effect
22:36.21JeffMthat seems like a rule that is being applied to a medium that dosn't work well with it
22:36.50JeffMbetter to call it, 5 seconds per turn or something
22:37.05JeffMbut then you are going to get longer to sudy the board due to latency anyway
22:37.19JeffMyou need to use nutrinos
22:37.27apeman__the internet games have to mirror the format of live games though
22:37.37apeman__face-to-face games
22:37.47JeffMthey don't HAVE to, your rules body just wants them to so they can be compared
22:37.59JeffMwanting to call an apple an orange dosn't make it an orange ;)
22:38.08apeman__close enough to HAVE to
22:38.18apeman__someone will flip the table if they don't :P
22:38.20randomparticlechanges in DR from 2.0.0 to 2.4.1: http://pastebin.com/TXFMzFFm
22:38.23JeffMheh
22:38.55JeffMare you looking at the segment loop and thinking it's lag comp?
22:39.49JeffMbut whatevers
22:39.56randomparticlei think you ought to actually read it
22:39.59JeffMnaw
22:40.02JeffMit dosn't matter
22:40.07JeffMwe let people do whatever they want
22:40.18JeffMso code changes all the time with out people thinking how it affects anything
22:40.21randomparticleconclusively proves what i was saying
22:40.30JeffMok, good for you, you win the internet
22:40.38apeman__how? thats just an extension to the existing DR philosophy
22:40.42randomparticlebit touchy? :)
22:40.49JeffMjust don't are
22:40.50JeffMcare
22:40.53JeffMI can be wrong :)
22:41.18JeffMI just don't see how it afffects anything now
22:41.21apeman__oh right
22:43.15apeman__thats not lag correction. Thats DR trying to be done more properly than it is done now
22:43.48JeffMIIRC it just used updates from it's own client, it dosn't try to normailize the timestamps across players
22:45.25JeffMthere have been a number of attempts at things throughout the past, mostly little things
22:45.32randomparticlethe critical section begins like so:  // adjust for the time offset
22:45.52randomparticlewhich disappeared in later versions of bz
22:46.02JeffMprobalby caused problems
22:46.09JeffMor someone didn't know what it was doing
22:46.17JeffMprobably the later
22:46.33JeffMor it coulda been that time someone pulled all the code they wrote
22:47.32randomparticleI think chesval knew what happened. vaguely recall talking to him about it several years ago
22:47.48randomparticle*chestal
22:47.52JeffMyeah he had done a lot of work on the latency problems
22:48.06apeman__it attempts to incorporate jitter adjustment to the DR model
22:48.08JeffMmostly showing how bad it was
22:48.24apeman__chesval :P
22:48.30JeffMhe was big on jitter :)
22:49.39Guest4003opencombat: 03jeffm2501 * r222 10/trunk/unit0/PlanEdit/ (5 files): Draw plan view of map.

Generated by irclog2html.pl Modified by Tim Riker to work with infobot.