| 00:19.45 | *** join/#bzflag infobot (ibot@208.53.50.136) |
| 00:19.45 | *** topic/#bzflag is https://BZFlag.org || http://ohloh.net/p/bzflag || https://bzflag.org/help || Channel Logs: http://infobot.rikers.org/%23bzflag/ || 2.4.14 released! https://forums.bzflag.org/viewtopic.php?f=62&t=19953 |
| 00:19.45 | *** mode/#bzflag [+v infobot] by ChanServ |
| 01:02.00 | Zehra | wonders if Low Gravity and Cloaked Shot will be added to 2.5 |
| 01:03.48 | JeffM | no |
| 01:04.06 | JeffM | but you can do LG with a custom flag if I finish everything |
| 01:04.49 | Zehra | yay |
| 01:33.44 | blast007 | so should setting an FPS limit actually reduce your CPU usage? because on my Linux system with Intel graphics, turning on a 30 FPS limit still consumes a full CPU core, whereas setting it to vertical sync (so running a solid 60FPS) drops the CPU usage |
| 01:46.20 | *** join/#bzflag DTRemenak (~DTRemenak@2605:e000:141f:c028:9035:9ed0:a4d5:4c0f) |
| 01:46.20 | *** join/#bzflag DTRemenak (~DTRemenak@about/essy/CrazyCoder/DTRemenak) |
| 01:46.21 | *** mode/#bzflag [+v DTRemenak] by ChanServ |
| 01:54.22 | JeffM | FPS limit spins the cpu in the bz thread, vsync has it wait in the os, iirc |
| 02:03.33 | blast007 | wonder if there's any point to FPS limit on linux then |
| 02:04.04 | blast007 | certainly felt like the CPU was busier because a video stream kept skipping when I had it set to FPS limit, but not with vsync |
| 02:06.57 | blast007 | adds investigating that to his 'low priority' list |
| 02:25.30 | JeffM | Client could compute the next frame time and call sleep |
| 11:57.09 | blast007 | So this is what rrlog does. Is that useful? http://static.bzexcess.com/rrlog.png |
| 11:59.44 | blast007 | iirc, someone had a tool that would show replays from a top-down radar view so they could step through time and see the interaction of tanks/shots, but I'm not sure if it used this |
| 13:33.32 | *** join/#bzflag tupone (~alfredo@gentoo/developer/tupone) |
| 17:18.49 | blast007 | JeffM: managed to crash a stock 2.5 client connected to a stock 2.5 server while messing around joining a modified client. in playingLoop, like 7371, the call to targetdPlayer->getFlag(). flagType inside of the Player object has already been free'd somewhere (has a pointer value of 0xdddddddd, which is "used by ... Microsoft's C/C++ debug free() function to mark freed heap memory") |
| 17:18.56 | blast007 | line* |
| 17:20.25 | JeffM | Ok so it just needs a null check en, the smart pointer should take care of it |
| 17:37.26 | *** join/#bzflag bertman (~bert@unaffiliated/bertman) |
| 18:38.50 | *** join/#bzflag FusionDude (~flash@c-71-205-162-253.hsd1.co.comcast.net) |
| 21:10.38 | *** join/#bzflag Zehra (~Zehra@unaffiliated/zehra) |
| 23:29.34 | *** join/#bzflag bertman (~bert@unaffiliated/bertman) |
| 23:35.18 | Zehra | https://www.bzflag.org/documentation/developer/bzfs_api/bz_eplayerdieevent |
| 23:36.09 | Zehra | Could we get the shot type, since we have flag type for the death, but not shot type? |
| 23:38.59 | Zehra | I kind of would like to make a check for "SW" deaths, but can't seem to find a way to do it. |
| 23:40.39 | Zehra | I know 'flagKilledWith' can be used, but this fails in a few cases, if shot changing or anything involving non-player shots is used. |
| 23:42.05 | Zehra | Thinking about it a bit more, I might have found the solution. |
| 23:42.11 | Zehra | nvm the above. |
| 23:47.54 | blast007 | does flagKilledWith not populate correctly for world weapon shots? |
| 23:51.11 | JeffM | Plugins should not have to do shot changes anymore |
| 23:51.36 | Zehra | I made a bit of a mistake in a calculation. |
| 23:51.58 | JeffM | In fact i need to break shot changes |
| 23:52.28 | JeffM | They get the same out of sync |
| 23:52.57 | JeffM | And there is now a flag effect that lets the server geneate the shot |
| 23:55.21 | Zehra | Can the shots be changed that way without a flag? |
| 23:57.55 | Zehra | knows a few plug-ins which would break due to preventing shot changes |
