| 00:00.03 | blast007 | could try a USB mouse too |
| 00:01.38 | Atluxity | yeah |
| 00:02.48 | myrk | unrelated question: does anyone know where i can find mrapple? |
| 00:04.01 | blast007 | myrk: uh, here? :) |
| 00:04.04 | blast007 | he's on IRC right now |
| 00:04.09 | blast007 | send him a /msg |
| 00:04.12 | blast007 | and then wait |
| 00:04.24 | myrk | i did but hes eway |
| 00:04.28 | myrk | *away |
| 00:04.30 | mrapple | wut wut? |
| 00:04.36 | myrk | hi apple |
| 00:04.40 | mrapple | *waves* |
| 00:04.40 | myrk | see pm? |
| 00:04.45 | mrapple | er yah |
| 00:05.06 | myrk | k |
| 00:05.07 | mrapple | your web thingy failed and made apache max out my cpu |
| 00:05.13 | myrk | ?? |
| 00:05.22 | mrapple | the web version of your bot |
| 00:05.28 | myrk | well i hadnt tested it yet........told you :P |
| 00:05.37 | myrk | ill try to fix it |
| 00:05.38 | mrapple | eh k |
| 00:07.09 | blast007 | the fix would be via rm |
| 00:13.55 | AAA_awright | heh |
| 00:36.57 | *** join/#bzflag myrk (~myrl@ip68-5-244-72.oc.oc.cox.net) |
| 00:51.39 | *** join/#bzflag BulletCatcher (~bc@bzflag/developer/BulletCatcher) |
| 01:06.57 | *** join/#bzflag L4m3r (~l4m3r@bzflag/developer/L4m3r) |
| 01:06.57 | *** mode/#bzflag [+v L4m3r] by ChanServ |
| 01:19.35 | *** join/#bzflag me1 (~ausom@bzflag/player/Me1) |
| 01:22.46 | *** join/#bzflag smoooth (~smoooth@cpe-098-024-240-002.ec.res.rr.com) |
| 01:53.19 | *** join/#bzflag jomojo (~jomojo@cpe-76-183-45-237.tx.res.rr.com) |
| 01:58.01 | *** join/#bzflag Wow (84b53360@gateway/web/freenode/ip.132.181.51.96) |
| 03:51.00 | *** join/#bzflag JeffM1 (~Jeff@adsl-75-50-183-235.dsl.lsan03.sbcglobal.net) |
| 04:44.38 | *** join/#bzflag FusionDude (~jwmelto@bzflag/developer/Flash) |
| 05:11.14 | *** join/#bzflag DTRemenak (~DTRemenak@about/essy/CrazyCoder/DTRemenak) |
| 05:11.14 | *** mode/#bzflag [+v DTRemenak] by ChanServ |
| 05:46.37 | *** join/#bzflag L4m3r (~l4m3r@bzflag/developer/L4m3r) |
| 05:46.37 | *** mode/#bzflag [+v L4m3r] by ChanServ |
| 06:58.06 | *** join/#bzflag Murielle (~Murielle@unaffiliated/murielle) |
| 07:03.19 | *** part/#bzflag Murielle (~Murielle@unaffiliated/murielle) |
| 07:03.57 | *** join/#bzflag Murielle (~Murielle@unaffiliated/murielle) |
| 07:07.51 | *** join/#bzflag Murielle (~Murielle@72.13.de.static.xlhost.com) |
| 07:08.37 | *** join/#bzflag Murielle (~Murielle@unaffiliated/murielle) |
| 07:09.51 | *** join/#bzflag Murielle (~Murielle@72.13.de.static.xlhost.com) |
| 07:10.13 | *** join/#bzflag Murielle (~Murielle@unaffiliated/murielle) |
| 07:15.16 | *** join/#bzflag dhouthoo (~dhouthoo@2001:6f8:1472:1:5ab0:35ff:feff:b710) |
| 07:17.09 | *** join/#bzflag DTRemenak (~DTRemenak@about/essy/CrazyCoder/DTRemenak) |
| 07:17.09 | *** mode/#bzflag [+v DTRemenak] by ChanServ |
| 07:43.49 | *** join/#bzflag Wyk3d (~Wyk3d@cl-86-125-185-198.cablelink.mures.rdsnet.ro) |
| 07:44.29 | *** join/#bzflag TutTankAhmon (~Olivier@LAubervilliers-151-11-10-35.w193-251.abo.wanadoo.fr) |
| 08:31.50 | *** join/#bzflag bier (~bier@p54A56F1F.dip.t-dialin.net) |
| 08:32.08 | *** join/#bzflag bier|tp (~bier@p54A56F1F.dip.t-dialin.net) |
| 09:01.29 | *** join/#bzflag L4m3r (~l4m3r@bzflag/developer/L4m3r) |
| 09:01.30 | *** mode/#bzflag [+v L4m3r] by ChanServ |
| 09:55.22 | *** join/#bzflag tupone (~alfredo@gentoo/developer/tupone) |
| 09:55.22 | *** mode/#bzflag [+v tupone] by ChanServ |
| 10:12.50 | *** join/#bzflag I_Died_Once (~I_Died_On@c-68-60-212-188.hsd1.ga.comcast.net) |
| 10:54.34 | *** join/#bzflag tupone (~alfredo@gentoo/developer/tupone) |
| 10:54.34 | *** mode/#bzflag [+v tupone] by ChanServ |
| 11:11.44 | *** join/#bzflag Atluxity (~atluxity@wikipedia/Atluxity) |
| 11:52.29 | *** join/#bzflag randomparticle (~randompar@about/essy/snick/randomparticle) |
| 13:55.13 | *** join/#bzflag randomparticle (~randompar@about/essy/snick/randomparticle) |
| 13:55.56 | randomparticle | hmm.. bzflag seems to apply inputs retrospectively |
| 13:56.11 | *** join/#bzflag Upsetter (~er@89.246.170.145) |
| 13:56.13 | randomparticle | you move the mouse and the movement is felt in the preceding frame |
| 13:56.41 | randomparticle | is there a reason for that? |
| 13:56.58 | brad | you notice that? O_o |
| 13:57.23 | randomparticle | yeah, i tried changing the code order and you can feel the difference. |
| 13:57.39 | randomparticle | i think it explains the sponginess at high framerates |
| 13:58.45 | randomparticle | generally speaking one expects forces not to apply backward in time :) |
| 14:03.28 | brad | what fps do you play with? |
| 14:04.58 | *** join/#bzflag spldart (~spldart2@bzflag/contributor/spldart) |
| 14:04.58 | *** mode/#bzflag [+v spldart] by ChanServ |
| 14:08.41 | *** join/#bzflag Flash (~jwmelto@bzflag/developer/Flash) |
| 14:09.40 | randomparticle | 50 normally |
| 14:10.19 | randomparticle | this is the situation in bzflag: |
| 14:10.20 | randomparticle | http://www.pixentral.com/show.php?picture=1MCznJZ56eX8H0rZHCZuzOTc5UJc0 |
| 14:11.12 | randomparticle | the calculation is done at the end of frame n, but the results are applied across the whole frame |
| 14:11.19 | randomparticle | which covers time preceding the calculation |
| 14:11.30 | brad | is it really that big a problem though? :p |
| 14:11.40 | brad | with "normal" fps |
| 14:12.02 | randomparticle | lots of players choose 30fps |
| 14:12.15 | randomparticle | a tank moves at 25m/s max on gu |
| 14:12.23 | randomparticle | that's about 4 tank lengths |
| 14:13.06 | randomparticle | so in 1 frame a tank can move roughly 0.8 tank lengths |
| 14:13.20 | randomparticle | i mean metres |
| 14:13.36 | randomparticle | about 1/5th of tank |
| 14:13.43 | randomparticle | so that's the difference it makes |
| 14:18.21 | Manu | randomparticle: DR makes things smoother no? |
| 14:18.30 | SportChick | howdy randomparticle |
| 14:18.43 | Manu | hi SportChick, long time no see :) |
| 14:18.47 | randomparticle | hi SportChick |
| 14:19.05 | SportChick | hiya Manu...yep, I've been buy |
| 14:19.10 | SportChick | err, busy* |
| 14:19.26 | Manu | yes, I follow you in fb :) |
| 14:19.34 | randomparticle | the DR doesn't really smooth it (in 2.0.x) |
| 14:19.43 | randomparticle | i'm not sure about the new bzflag though |
| 14:19.54 | SportChick | Manu: hehe |
| 14:22.59 | *** join/#bzflag jomojo (~jomojo@cpe-76-183-45-237.tx.res.rr.com) |
| 14:24.10 | *** join/#bzflag bryjen (~bryjen@63.147.94.149) |
| 14:24.10 | *** mode/#bzflag [+v bryjen] by ChanServ |
| 14:24.55 | Manu | randomparticle: fps have been always a problem in bzflag :( |
| 14:25.19 | randomparticle | grr.. the colloquy frownicon again :) |
| 14:25.33 | Manu | :) |
| 14:26.29 | randomparticle | i'm not sure to what extent it matters |
| 14:26.39 | randomparticle | the human brain is really weird with respect to time |
| 14:26.58 | randomparticle | it fools you into thinking things you sense are current, for example |
| 14:27.30 | randomparticle | if you see your finger tapping on a table, you feel it tapping when you see it tapping |
| 14:27.43 | randomparticle | but the signals take two routes to your brain, one route longer than the other |
| 14:28.08 | Manu | yes |
| 14:28.13 | Cobra_Slow | the brain takes the visual and acustic signals with higher priority - always |
| 14:32.25 | randomparticle | does bzflag 3.0 use Cristan's algorithm for clock sync? |
| 14:32.33 | randomparticle | *Cristian's |
| 14:33.58 | *** join/#bzflag Atluxity (~atluxity@122.79-160-7.customer.lyse.net) |
| 14:34.00 | *** join/#bzflag Atluxity (~atluxity@wikipedia/Atluxity) |
| 14:34.43 | *** join/#bzflag AAA_awright (~a3@ip24-251-157-63.ph.ph.cox.net) |
| 14:36.56 | BulletCatcher | randomparticle: client and server clocks are not really synchronized at all. |
| 14:37.33 | randomparticle | are you storing an offset? |
| 14:38.42 | BulletCatcher | They use deltas from their own clocks. |
| 14:39.33 | BulletCatcher | Not quite. |
| 14:39.56 | BulletCatcher | They work with deltas from both clocks, so absolute values aren't important. |
| 14:40.48 | randomparticle | ok |
| 14:41.33 | randomparticle | so is it the old RTT/2 to get the half roundtrip lag? |
| 14:41.57 | randomparticle | i.e., you assume symmetric connections |
| 14:42.02 | BulletCatcher | Yes, I don't think that has changed from 2.0 -> 2.99. |
| 14:42.25 | BulletCatcher | The halflag computation remains. |
| 14:42.32 | randomparticle | is there a way of testing for symmetry? |
| 14:42.56 | BulletCatcher | That requires synchronized clocks, I believe. |
| 14:43.03 | randomparticle | yeah |
| 14:43.24 | randomparticle | i'm wondering if you could connect back to a server on your own machine having routed the packets over the net somewhere |
| 14:44.16 | randomparticle | i'm not sure if that helps though :) |
| 14:44.36 | randomparticle | the problem might be theoretically insoluble. kinda feels like it (without really thinking) |
| 14:45.27 | BulletCatcher | NTP adequately solves the problem of synchronized clocks, but BZFlag can't rely on its use. |
| 14:46.15 | randomparticle | wouldn't you just, effectively, have your own time server in the bzflag server and your own ntp client in the bzflag client? |
| 14:47.55 | BulletCatcher | That could work, but even the best implementations of NTP can take many minutes (longer than a 30 minute GU match) to converge. |
| 14:51.06 | ts | Hmm, maybe the delayed mouse movement should use a scale so with less FPS more frames are skipped until it is processed. |
| 14:51.39 | *** join/#bzflag MindstormsKid (~msk@Wikipedia/MindstormsKid) |
| 14:52.01 | blast007 | ideally, the simulation would be run at a constant "FPS" independent of the graphics |
| 14:52.16 | blast007 | but that would be quite a chore to separate in the current codebase :) |
| 14:52.32 | ts | That would be the better way but more complicated :) |
| 14:52.59 | randomparticle | ts: it's not that movement is delayed. it's that it's applied to motion that occurred before your mouse moved! |
| 14:53.30 | randomparticle | it's like dialing someone's number and their phone rings yesterday |
| 14:53.46 | blast007 | BZFlag - jumping tanks, and time travel |
| 14:53.48 | blast007 | :) |
| 14:54.03 | ts | It knows where your mouse will move and does that in the past? |
| 14:54.23 | randomparticle | no |
| 14:54.33 | randomparticle | it has to work out what to show you in the current frame |
| 14:54.45 | randomparticle | to do that it calculates what the movement in the recent past was |
| 14:54.49 | randomparticle | but.. here is the thing: |
| 14:55.06 | randomparticle | when it calculates the movement it uses your current mouse position |
| 14:55.21 | randomparticle | so it's as if you had moved the mouse there 1 frame ago |
| 14:56.20 | randomparticle | movement = tank movement. mouse = pointer |
| 14:57.31 | randomparticle | you may find this diagram helpful: |
| 14:57.31 | randomparticle | http://www.pixentral.com/show.php?picture=1MCznJZ56eX8H0rZHCZuzOTc5UJc0 |
| 14:59.10 | ts | Ok, it's official: I'm confused :) |
| 15:03.00 | randomparticle | this is the relevant section of code: |
| 15:03.01 | randomparticle | http://pastebin.ca/1887765 |
| 15:03.23 | randomparticle | i've altered it bringing myTank->update() to the top |
| 15:03.38 | randomparticle | in the original it occurs near the end |
| 15:03.55 | randomparticle | myTank->update() calculates where the tank moved during the previous frame |
| 15:04.24 | randomparticle | i.e., between prevTime and getTick() |
| 15:04.57 | randomparticle | i'm typing too much :) |
| 15:06.11 | randomparticle | doMotion() is where the current mouse pointer position is grabbed |
| 15:07.47 | ts | yes |
| 15:16.00 | randomparticle | i'm going to change it to use the mean position |
| 15:16.10 | randomparticle | see what that feels like |
| 15:17.01 | ts | You better don't play in the leagues with that mod ;) |
| 15:18.25 | randomparticle | wouldn't want to. the original setting gives the biggest playing advantage :) |
| 15:26.21 | *** join/#bzflag spldart (~spldart2@bzflag/contributor/spldart) |
| 15:26.21 | *** mode/#bzflag [+v spldart] by ChanServ |
| 15:27.46 | BulletCatcher | It would be better to explore changing it in the trunk code. We won't make that kind of change in 2.0.x. |
| 16:01.50 | *** join/#bzflag temporalD (~temporal@bzflag/serverop/TemporalDistraction) |
| 16:49.46 | *** join/#bzflag xaver (~xaver@pD9ED696C.dip.t-dialin.net) |
| 16:53.49 | *** join/#bzflag Will07c5 (~william@c-67-173-127-196.hsd1.il.comcast.net) |
| 17:34.48 | *** join/#bzflag QuantumTim (~tim@62.249.230.206) |
| 18:13.19 | *** part/#bzflag Upsetter (~er@89.246.170.145) |
| 18:21.49 | *** join/#bzflag spldart (~spldart2@bzflag/contributor/spldart) |
| 18:21.49 | *** mode/#bzflag [+v spldart] by ChanServ |
| 18:47.08 | *** join/#bzflag me1 (~ausom@bzflag/player/Me1) |
| 19:16.57 | *** join/#bzflag R0b0t1 (~Enigma@unaffiliated/r0b0t1) |
| 19:29.47 | *** join/#bzflag mrapple (~mrapple@bzflag/player/MrAppleComputer) |
| 19:48.37 | *** join/#bzflag TimRiker (~timr@bzflag/projectlead/TimRiker) |
| 19:48.37 | *** mode/#bzflag [+o TimRiker] by ChanServ |
| 20:08.32 | *** join/#bzflag JefferyM (~JefferyM@unaffiliated/jeffm2501) |
| 20:08.32 | *** mode/#bzflag [+v JefferyM] by ChanServ |
| 20:53.53 | *** join/#bzflag TimRiker (~timr@bzflag/projectlead/TimRiker) |
| 20:54.04 | *** mode/#bzflag [+o TimRiker] by ChanServ |
| 21:25.53 | *** join/#bzflag temporalD (~temporal@bzflag/serverop/TemporalDistraction) |
| 23:05.37 | *** join/#bzflag temporalD (~ATD@bzflag/serverop/TemporalDistraction) |
| 23:30.17 | *** join/#bzflag Quol_away (~Quol@69.90.0.146) |
| 23:31.53 | mrapple | hey everyone nightmare is making the round tonight with the host sndg02.sbcglobal.net in the 74.243.*.* range |
| 23:32.38 | *** join/#bzflag Gabe_G23 (~gabe@bzflag/player/GabrielG) |
| 23:36.17 | *** join/#bzflag delusional (~delusiona@unaffiliated/delusional) |
| 23:44.58 | *** join/#bzflag I_Died_Once (~I_Died_On@c-68-60-212-188.hsd1.ga.comcast.net) |
| 23:45.33 | *** join/#bzflag Will07c5 (~william@c-67-173-127-196.hsd1.il.comcast.net) |