| 00:06.32 | *** join/#bzflag Delusional (~Delusiona@unaffiliated/delusional) |
| 00:30.02 | *** join/#bzflag I_Died_Once (~I_Died_On@unaffiliated/idiedonce/x-1828535) |
| 00:38.33 | macsforme | PR #86 looks good to me... if nobody else has comments on it I'll merge it in shortly |
| 01:45.37 | Delusional | Burgermeister Meisterburger |
| 01:46.18 | Delusional | I'm not evil I just wish to see a world without toys. |
| 01:48.10 | Delusional | wants to run his haloween map sooooooooo baaaad. |
| 01:48.14 | Delusional | but |
| 01:51.13 | Delusional | knows no bzflag.org licensing exceptions will be made to host images that are 20 year old animated gifs that were posted on every myspace page |
| 01:51.32 | Delusional | therefore⦠the name for the next version should beâ¦.. |
| 01:52.05 | Delusional | Burgermeister Meisterburger |
| 01:52.57 | Delusional | have yourself a merry toy burning my it be warm and bright |
| 01:55.48 | blast007 | macsforme: does building/installing the *nix way on macOS cause any problems? |
| 02:31.32 | *** join/#bzflag bertman (~bert@unaffiliated/bertman) |
| 04:24.47 | *** join/#bzflag FastLizard4 (fastlizard@wikipedia/pdpc.active.FastLizard4) |
| 06:54.42 | *** join/#bzflag DTRemenak (~DTRemenak@2605:e000:7f0c:c401:e966:a4bb:97ca:708c) |
| 06:54.42 | *** join/#bzflag DTRemenak (~DTRemenak@about/essy/CrazyCoder/DTRemenak) |
| 06:54.42 | *** mode/#bzflag [+v DTRemenak] by ChanServ |
| 08:37.44 | macsforme | blast007: nope... the only deficiency I found was that you have to have SDL installed as a library rather than as a framework (which was already a problem, and is really beyond the scope of the problem he was trying to solve)... for the homebrew/macports crowd, I think this will make things work for them |
| 08:41.52 | macsforme | we did have an SDL framework check for macOS in the make system at one point (2.0 branch, maybe?), so a sufficiently motivated person could probably add it back in easily |
| 09:02.11 | *** join/#bzflag macs4me (~macsforme@trinity.fairserve.net) |
| 09:02.11 | *** join/#bzflag macs4me (~macsforme@bzflag/developer/Constitution) |
| 09:02.11 | *** mode/#bzflag [+o macs4me] by ChanServ |
| 09:18.39 | *** join/#bzflag macsforme (~macsforme@trinity.fairserve.net) |
| 09:18.40 | *** join/#bzflag macsforme (~macsforme@bzflag/developer/Constitution) |
| 09:18.40 | *** mode/#bzflag [+o macsforme] by ChanServ |
| 09:27.41 | *** join/#bzflag macs4me (~macsforme@trinity.fairserve.net) |
| 09:27.42 | *** join/#bzflag macs4me (~macsforme@bzflag/developer/Constitution) |
| 09:27.42 | *** mode/#bzflag [+o macs4me] by ChanServ |
| 11:29.05 | blast007 | macsforme: okay, then I'm fine with it being merged |
| 11:35.38 | *** join/#bzflag BZnotify (~BZnotify@192.30.252.35) |
| 11:35.38 | BZnotify | [13bzflag] 15macsforme pushed 1 new commit to 062.4: 02https://git.io/v55br |
| 11:35.38 | BZnotify | 13bzflag/062.4 143e5524d 15Joshua Bodine: Changed how SDL 2 handles key down/text input events so that international keyboards can make ASCII characters with modified non-ASCII keys. |
| 11:35.38 | *** part/#bzflag BZnotify (~BZnotify@192.30.252.35) |
| 11:38.27 | macsforme | blast007: you may want to look at that to make sure I didn't undo any of your work in https://github.com/BZFlag-Dev/bzflag/commit/82250476d2aa5a4ce3d8dab741d8aefd7ce92235 |
| 11:39.34 | macsforme | but lowercase, uppercase, numbers, symbols, and numpad (with numlock on and off) keys seem to work in my tests so far, and I can make the '?' symbol now on a german keyboard layout |
| 11:40.53 | macsforme | aFreshMelon: if you get a chance, could you test this new build and make sure that the black screen/resolution scaling issue on retina macbooks is still working? https://download.bzflag.org/bzflag/osx/2.4.11/BZFlag-2.4.11-20170916-macOS.zip |
| 11:41.43 | macsforme | that has the SDL 2.0.6 pre-release framework in it, so if that's working for you now we should be good to go |
| 11:50.01 | aFreshMelon | macsforme: It sets a resolution that works |
| 11:50.33 | macsforme | great |
| 11:50.45 | macsforme | does it list all the resolutions in the correct order in display settings? |
| 11:55.08 | *** join/#bzflag BZnotify (~BZnotify@192.30.252.35) |
| 11:55.08 | BZnotify | [13bzflag] 15macsforme pushed 7 new commits to 062.4: 02https://git.io/v55Na |
| 11:55.08 | BZnotify | 13bzflag/062.4 14c1df307 15Mike Miller: Fix compilation using gcc on OS X... |
| 11:55.08 | BZnotify | 13bzflag/062.4 1432feed6 15Mike Miller: Update AUTHORS |
| 11:55.08 | BZnotify | 13bzflag/062.4 146361e24 15Mike Miller: Remove GLUT (mistake) and put a proper check for gl.h on darwin |
| 11:55.08 | *** part/#bzflag BZnotify (~BZnotify@192.30.252.35) |
| 12:10.58 | aFreshMelon | macsforme: No |
| 12:11.23 | aFreshMelon | The largest resolution is second, the native of the display that is. The resolution it sets is actually the scaled one from settings. |
| 12:12.20 | macsforme | hm, okay |
| 12:12.21 | aFreshMelon | macsforme: Want a screenshot? |
| 12:12.24 | macsforme | please |
| 12:13.31 | aFreshMelon | I changed it, it auto-set 1680x1050 |
| 12:15.43 | aFreshMelon | This is from settings, note the resolution on the left left https://usercontent.irccloud-cdn.com/file/KYoo0Cif/Bildschirmfoto%202017-09-16%20um%2014.13.51.png |
| 12:16.47 | aFreshMelon | In case you're wondering, at the place I'm staying I have an upload of less than 100 KBit/s |
| 12:16.58 | aFreshMelon | The other image is still loading |
| 12:17.13 | blast007 | BulletCatcher: I'd be curious if your client, when built with SDL2, repeatedly tries to create a window of one size when it actually creates it of another size. In my xrandr scaling test, it was repeatedly creating/destroying the window, triggered from a resize event. I have a patch that writes out the existing window sizes/flags and the new requested size/flags: |
| 12:17.18 | blast007 | https://gist.github.com/blast007/90e3ccc5186acd6f78545efdf3b8b1f4 |
| 12:18.30 | aFreshMelon | https://usercontent.irccloud-cdn.com/file/WdTQiBkM/Bildschirmfoto%202017-09-16%20um%2014.12.39.png |
| 12:18.38 | macsforme | very similar to the issue we were having on macOS with scaled retina displays |
| 12:18.53 | blast007 | BulletCatcher: in my test, it was trying to create a window of 1920x1200, but when it came around to getting the size of the existing window, it was 2400x1500 |
| 12:20.05 | aFreshMelon | By the way, blast007 and macsforme, is there an issue or settings that makes keystrokes register more "slowly" - I'm having an issue where often when I hit the key I don't get the full effect. Say I hit an arrow key, it immediately spins in that directions, but it starts out very slowly. And then randomly when I hit the arrow key it just goes full tilt immediately |
| 12:20.21 | macsforme | aFreshMelon: if you try to set it to 1680x1050, does it work? |
| 12:20.36 | aFreshMelon | macsforme: That is what the game set on it's own, it works |
| 12:20.40 | blast007 | arrow keys are treated special, I believe |
| 12:21.03 | aFreshMelon | Because if it doesn't register the press properly, it makes para jumping really painful |
| 12:21.06 | blast007 | I think they start out slower and quickly accelerate to full speed, which enables smaller jumps forward with the keyboard |
| 12:21.07 | aFreshMelon | Basically impossible |
| 12:21.42 | blast007 | may also depend if you're switching directions |
| 12:22.10 | macsforme | if it selected the second resolution on the list, that was probably a manual setting from before... I think the issue was that 1680x1050 wasn't actually a supported resolution for that display |
| 12:22.37 | aFreshMelon | I believe 1050 isn't actually a resolution, yeah |
| 12:22.40 | macsforme | so perhaps they didn't actually fix it... grumble |
| 12:22.55 | aFreshMelon | One sec |
| 12:24.04 | aFreshMelon | macsforme: I have a screenshot for you that you will just love |
| 12:25.20 | aFreshMelon | Nvm I won't upload the screenshot cause of my connection |
| 12:25.34 | aFreshMelon | But essentially every resolution I try to test with T goes <unloadable> |
| 12:26.47 | macsforme | sigh |
| 12:27.47 | macsforme | so if you temporarily move your bz config file, I take it the program will still fail to load? |
| 12:27.54 | aFreshMelon | No |
| 12:28.10 | aFreshMelon | BZ opens fine, when you gave me the new build I removed the resolution setting from my config to see if it'd work |
| 12:29.33 | macsforme | very odd... it should never select the second resolution on the list by default, I don't think |
| 13:42.47 | *** join/#bzflag DipoleHourglass (uid134094@gateway/web/irccloud.com/x-ldmwtiaafouyuyum) |
| 15:30.03 | *** join/#bzflag dngor (abuse@p3m/dngor) |
| 15:45.05 | BulletCatcher | After applying that debug patch, blast007, I ran "src/bzflag/bzflag -window 960x540". |
| 15:45.10 | BulletCatcher | It printed "Creating window of size 960 x 540 with flags 34" followed by 7 lines of "Existing window of size 960 x 540 with flags 38". |
| 15:45.15 | BulletCatcher | Then (with the resolution already set to 1920x1080 in my config file) I pressed F1 and it printed "Existing window of size 960 x 540 with flags 1574" and "Creating window of size 1920 x 1080 with flags 3" once each. |
| 15:45.17 | BulletCatcher | My screen went black with only a working cursor visible and I had to switch to a virtual tty to kill -9 the bzflag process. |
| 15:45.42 | blast007 | k, so a different problem then |
| 15:45.52 | blast007 | and you have to kill -9? not just normal kill? |
| 15:46.09 | BulletCatcher | Yes. Other kill numbers don't make it die. |
| 15:46.15 | blast007 | fun |
| 15:46.56 | BulletCatcher | In previous tests the desktop would reappear and typing CTRL/C would work as expected. |
| 15:47.22 | blast007 | ever tried running it from a debugger (ideally over SSH, but might be okay from a VT) and breaking execution to see if it's getting stuck somewhere? |
| 15:47.36 | BulletCatcher | This is using SDL2-2.0.5-3.fc25.x86_64. |
| 15:49.11 | BulletCatcher | Let me try that. |
| 15:49.59 | blast007 | I just wish I could reproduce it here. |
| 15:51.01 | aFreshMelon | macsforme: No no, it selected first |
| 15:54.37 | aFreshMelon | blast007: Interestingly it seems like the slow keys issue only affects the left and right keys |
| 15:54.50 | aFreshMelon | Up and down are not affected, which makes it even less sensical |
| 16:08.06 | blast007 | current code that does that: https://github.com/BZFlag-Dev/bzflag/blob/2.4/src/bzflag/LocalPlayer.cxx#L213-L225 |
| 16:08.23 | blast007 | looks like it's had that function since 1.10.0 |
| 16:08.40 | blast007 | functionality* or feature |
| 16:14.48 | blast007 | Looks like it was originally part of this patch: https://sourceforge.net/p/bzflag/patches/189/ and was applied with slightly different values https://github.com/BZFlag-Dev/bzflag/commit/66aa61e262feffdabca2d586f04512a57814f6a8 |
| 16:15.15 | blast007 | went through a couple other changes (values, making it just for keyboard input) before making it into 1.10.0 |
| 16:16.32 | blast007 | so, it seems it was to make it possible to aim better with low FPS when using the keyboard |
| 16:16.53 | blast007 | I guess I was thinking of the slow keyboard button for jumping onto platforms |
| 16:57.50 | BulletCatcher | https://pastebin.com/TCqd1Fzm shows the stack trace from when bzflag stops responding following F1. |
| 16:57.52 | BulletCatcher | When I do a plain kill I can see quitFlag set to true, but X11_SetWindowGrab() never exits to let that be noticed by the main loop. |
| 21:26.02 | macsforme | aFreshMelon: I think the resolution testing issue is a separate problem... my friend and I replicated it, and he's working on opening an issue for it now |
| 21:27.14 | macsforme | he was able to actually use most of the resolutions, though, including the first one listed, which was the main problem before... so I think the main problem is resolved |
| 21:29.36 | *** join/#bzflag bzdoug (4a825598@gateway/web/freenode/ip.74.130.85.152) |
| 21:31.50 | macsforme | https://github.com/BZFlag-Dev/bzflag/issues/88 |
| 21:32.17 | macsforme | that also exists in 2.4.10 |
