| 00:26.18 | *** join/#bzflag infobot (~infobot@rikers.org) |
| 00:26.18 | *** topic/#bzflag is https://BZFlag.org || http://ohloh.net/p/bzflag || https://bzflag.org/help || Channel Logs: http://infobot.rikers.org/%23bzflag/ || 2.4.10 released! https://forums.bzflag.org/viewtopic.php?f=8&t=19401 |
| 00:26.18 | *** mode/#bzflag [+v infobot] by ChanServ |
| 02:26.47 | *** join/#bzflag Periapsis (~nadir@unaffiliated/znadir) |
| 03:50.53 | *** join/#bzflag KaadmY (~kaadmy@2604:a880:1:20::1315:3001) |
| 03:50.53 | *** join/#bzflag Void7 (~Void7@2604:a880:1:20::1315:3001) |
| 04:21.50 | *** join/#bzflag Void7 (~Void7@2604:a880:1:20::1315:3001) |
| 04:21.50 | *** join/#bzflag KaadmY (~kaadmy@2604:a880:1:20::1315:3001) |
| 05:27.59 | *** join/#bzflag a_meteorite (~a_meteori@unaffiliated/ameteorite/x-000000001) |
| 06:28.48 | *** join/#bzflag blastis (~BLASTIS@c-24-8-43-250.hsd1.co.comcast.net) |
| 07:41.28 | *** join/#bzflag blastis (~BLASTIS@c-24-8-43-250.hsd1.co.comcast.net) |
| 10:32.57 | *** join/#bzflag blastis (~BLASTIS@c-24-8-43-250.hsd1.co.comcast.net) |
| 15:23.58 | macsforme | blast007, no I haven't (with bzflag specifically), but I do have retina resolution support in my other projects using that functionality |
| 15:25.03 | macsforme | and like I said before, that will give us an OpenGL canvas with a higher resolution than we asked for, which causes problems for bzflag |
| 15:27.53 | macsforme | macOS will already give us the "real resolution" we ask for in fullscreen mode (e.g., you can run bzflag at 2880x1800 on a MBP retina 13", rather than the "logical" resolution which is half of that) |
| 15:28.37 | macsforme | but it needs to work in windowed mode too |
| 15:31.34 | *** join/#bzflag nadir (uid134094@gateway/web/irccloud.com/x-penscuzlzgdqkslz) |
| 15:32.33 | JeffM | so there is no way for the mac to enumerate the real resolution from the start so you can ask for and get the real res? that plist option doesn't change the initial resolution the game gets back? |
| 15:34.01 | *** join/#bzflag Guest372 (05fe6147@gateway/web/freenode/ip.5.254.97.71) |
| 15:35.30 | macsforme | in fullscreen mode, there is... they are reported like real resolutions, and that is working fine already |
| 15:40.04 | macsforme | really, if we simplified our window initialization code, and went fully with SDL 2, we could use SDL_CreateWindow() with SDL_WINDOW_FULLSCREEN_DESKTOP, and then return the resolution we got (which could be high-dpi) and go from there |
| 15:40.47 | macsforme | but I wouldn't know how to re-implement that in all our legacy platform code for the other platforms |
| 15:44.30 | JeffM | do all the current publsihed platforms use SDL 2? |
| 15:45.16 | macsforme | I think so |
| 15:45.58 | afreshmelon | macsforme: Home and tested |
| 15:46.03 | afreshmelon | As far as I can tell it works now |
| 15:46.32 | macsforme | excellent... thanks much |
| 15:47.30 | JeffM | if all the main platforms we publish use SDL 2, it's acceptable to make the change for SDL 2 only. Then you leave it up to the other platform maintainers to deal with it |
| 15:47.30 | JeffM | techincaly the other platfroms can be broken at any time |
| 15:48.06 | JeffM | and if there is nobody to maintain an alternate platform, then it's effectivly dead and should be removed. |
| 15:48.41 | JeffM | as long as you can ensure that all the platforms we publish bins/packages for will work, that's all you need to worry about. |
| 15:49.04 | JeffM | the secondary platform market is just that.. secondary :) |
| 15:49.58 | JeffM | I agree that using the resolution from the context makes more logical sense. There may be other things later that also modify the context that the current method fails on (maybe some kind of new VR or something someone wants to do). |
| 15:50.58 | JeffM | does windows still have that old directX code for resolutions? |
| 16:09.52 | macsforme | is any code ever removed? :P |
| 16:17.45 | JeffM | it should be |
| 16:17.47 | JeffM | like the old mac platform stuff should be removed |
| 16:18.20 | JeffM | if it isn't that's a total failure of maintenance. |
| 16:44.48 | *** join/#bzflag infobot (~infobot@rikers.org) |
| 16:44.48 | *** topic/#bzflag is https://BZFlag.org || http://ohloh.net/p/bzflag || https://bzflag.org/help || Channel Logs: http://infobot.rikers.org/%23bzflag/ || 2.4.10 released! https://forums.bzflag.org/viewtopic.php?f=8&t=19401 |
| 16:44.48 | *** mode/#bzflag [+v infobot] by ChanServ |
| 16:44.48 | JeffM | wonders if he can even still make those files |
| 16:44.48 | blast007 | even on Windows, we're not "fully" SDL2. We're still using more nativish stuff for some/all input. |
| 16:44.51 | JeffM | well you could go full SDL2 for the display chain |
| 16:44.54 | blast007 | yeah |
| 16:44.56 | JeffM | that'd be a good start |
| 16:46.21 | JeffM | yeah, Photoshop removed that format. |
| 16:47.11 | blast007 | ... and switch to using libpng instead of our own reader/writer... and .... |
| 16:47.21 | JeffM | baby steps :) |
| 16:47.31 | blast007 | :) |
| 16:52.00 | macsforme | is there any reason we can't go fully SDL 2 on windows? that could simply our platform code considerably |
| 16:52.18 | JeffM | does SDL2 support the same force feedback stuff? |
| 16:52.23 | blast007 | I think the input handling wasn't as good |
| 16:52.53 | blast007 | I believe it does, but we didn't have it hooked up |
| 16:52.54 | JeffM | that'd be something to eval for sure, but if it's acceptably different, then why not? |
| 16:53.15 | JeffM | as long as you know the diffs and if they won't affect anyone/thing major |
| 16:53.21 | blast007 | and my move to SDL2 on Windows was more because I couldn't figure out how to fix the native Windows platform code. :) |
| 16:53.41 | macsforme | frankly, I doubt anyone plays using a joystick |
| 16:53.48 | JeffM | it would probalby be trival to move windows sound over to SDL2 since bz does it's own mixing. |
| 16:54.07 | JeffM | does SDL2 have a 3d sound API? |
| 16:54.11 | blast007 | no |
| 16:54.22 | blast007 | basic 2D sound API |
| 16:54.25 | JeffM | then eventualy you'd want to move everyone to like openAL |
| 16:54.38 | JeffM | but a basic 2d api would work for the current code, because bz does it's own 3d |
| 16:57.41 | blast007 | http://wiki.libsdl.org/CategoryAudio |
| 16:59.35 | JeffM | yeah that'd probalby work |
| 16:59.52 | JeffM | just queue up a single pre-mixed buffer |
| 17:03.53 | blast007 | there is mention of 4 and 6 (5.1) channel setups, but I think you'd still have to do the 3D yourself, and I'm unclear if you'd actually need a quad or 5.1 speaker system for those modes |
| 17:13.54 | JeffM | it has some mixing ability. you could go and proalby re-implment the mixing bzflag does with it's mixing functions |
| 17:14.06 | JeffM | but if you were going to do that may as well use a real 3d sound API |
| 17:14.46 | JeffM | and that's a larger project, because you have to decide what you want to keep/change from bz code. |
| 17:15.06 | JeffM | BZs "world" for audio physics is not exactly real world accurate |
| 17:15.18 | JeffM | the speed of sound is WAY off |
| 18:05.46 | JeffM | awww... unity 2017.1 added support for visual studio on mac. that's nice of them. |
| 20:30.38 | macsforme | if I've declared a variable extern in bzfgh.h, and it's declared in the ogl module but also used in the bzflag module, how should I declare it so that visual studio can find it when linking the client? |
| 20:31.04 | macsforme | I'm getting errors for "unresolved external symbol" |
| 20:31.30 | macsforme | oh, it was declared extern in *bzfgl.h |
| 20:34.50 | macsforme | for example, https://github.com/macsforme/bzflag/blob/better_multisampling/include/bzfgl.h#L37-L65 and https://github.com/macsforme/bzflag/blob/better_multisampling/src/ogl/OpenGLFramebuffer.cxx#L27-L57 |
| 20:43.44 | macsforme | nevermind, I guess actually building the file where it's defined would help... |
| 20:58.05 | *** join/#bzflag Guest731 (53250cf4@gateway/web/freenode/ip.83.37.12.244) |
| 21:02.43 | *** part/#bzflag Guest731 (53250cf4@gateway/web/freenode/ip.83.37.12.244) |
| 22:46.20 | *** join/#bzflag infobot (~infobot@rikers.org) |
| 22:46.20 | *** topic/#bzflag is https://BZFlag.org || http://ohloh.net/p/bzflag || https://bzflag.org/help || Channel Logs: http://infobot.rikers.org/%23bzflag/ || 2.4.10 released! https://forums.bzflag.org/viewtopic.php?f=8&t=19401 |
| 22:46.20 | *** mode/#bzflag [+v infobot] by ChanServ |
| 23:10.23 | allejo | it's not a bug, it's a feature to teach users how ot use the `kill` command |
| 23:29.32 | *** join/#bzflag macs4me (~macsforme@bzflag/developer/Constitution) |
| 23:29.32 | *** mode/#bzflag [+o macs4me] by ChanServ |