ScummVM IRC Log - 03 Sep 2017

[Back to Index]

[00:04] <-- Henke37 left irc: Quit: ERR_SHUTDOWN
[00:17] <-- Axy left irc: Ping timeout: 246 seconds
[00:53] --> Axy joined #scummvm.
[00:53] <-- Axy left irc: Changing host
[00:53] --> Axy joined #scummvm.
[01:17] <-- SylvainTV left irc: Read error: Connection reset by peer
[01:29] <-- D0SFreak left irc: Ping timeout: 264 seconds
[01:45] --> DJW|Home joined #scummvm.
[01:45] #scummvm: mode change '+o DJW|Home' by ChanServ!ChanServ@services.
[01:48] <-- DJWillis left irc: Ping timeout: 248 seconds
[01:54] <-- Dominus left irc: Ping timeout: 260 seconds
[01:54] --> Dominus joined #scummvm.
[02:00] --> DJW|Badger joined #scummvm.
[02:01] --> DJWillis joined #scummvm.
[02:01] #scummvm: mode change '+o DJWillis' by ChanServ!ChanServ@services.
[02:03] <-- DJW|Home left irc: Ping timeout: 240 seconds
[02:05] <-- DJW|Badger left irc: Ping timeout: 252 seconds
[02:10] <-- Lightkey left irc: Ping timeout: 246 seconds
[02:23] --> Lightkey joined #scummvm.
[02:39] <-- dreammaster left irc:
[02:40] <snover> overlay mode fixed&
[02:57] snover (snover@unaffiliated/snover) left #scummvm.
[02:57] --> snover joined #scummvm.
[02:57] #scummvm: mode change '+o snover' by ChanServ!ChanServ@services.
[02:59] <-- DrMcCoy left irc: Ping timeout: 248 seconds
[03:02] --> DrMcCoy joined #scummvm.
[03:02] #scummvm: mode change '+o DrMcCoy' by ChanServ!ChanServ@services.
[03:24] --> Begasus joined #scummvm.
[03:26] --> CuriosTiger joined #scummvm.
[03:27] --> Begas_VBox joined #scummvm.
[03:32] <-- Dominus left irc: Ping timeout: 260 seconds
[03:33] --> Dominus joined #scummvm.
[05:00] <snover> phew. ITS WORKING. now i just gotta clean up the commit history, which is going to be a bit of a chore since i did not hold off on fixing comments either, and ended up having to rename a few hundred instances of things like where OpenGLSdl was using _hwSurface and SurfaceSdl was using _hwsurface&
[05:02] <Deledrius> that part is always fun
[05:03] <snover> but, SDL1, SDL2, and OpenGL now all share code for handling the overlay and game layer dimensions, translating between the different coordinate systems, and managing aspect ratios (for the most part; i did not bother to harmonise the AR-detection because of PR#929)
[05:04] <snover> oh, and managing the window itself, of course.
[05:14] <Deledrius> awesome
[05:18] <snover> there are some incantations in here which i am nervous about getting rid of
[05:46] <bgK> less magic sounds good
[05:47] <snover> it looks like SDL2 is less awful than SDL1 about mouse warp
[05:58] <snover> maybe. ugh, i dont even know. how do i test this without doing something awful&
[06:01] <bgK> maybe use the testbed engine?
[06:05] <snover> thats the smart idea, but i am dumb, so i just chucked an SDL_Delay in there so i could unfocus the window to test what happens
[06:05] <snover> answer: sdl will still move your system mouse
[06:05] <snover> so thats fine, there is a check in SdlWindow for this though so there is no need to have more than one of those
[06:35] <snover> i seem to be doing a good job of finding existing bugs though i dont know why they exist
[06:36] <snover> displaying an osd message while another osd message is in the middle of fading out seems to start breaking future osd messages
[06:38] --> jamm joined #scummvm.
[06:38] <snover> oh, gosh, this is probably leaking surfaces.
[06:39] <snover> nope, nevermind. thats a good sign that its a good time for me to sleep.
[06:39] <snover> tomorrow& pull requests.
[07:18] <Deledrius> heh
[07:19] <-- jamm left irc: Ping timeout: 246 seconds
[07:30] <-- _sev left irc: Quit: This computer has gone to sleep
[07:35] --> Joefish joined #scummvm.
[07:35] #scummvm: mode change '+v Joefish' by ChanServ!ChanServ@services.
[07:39] <-- DJWillis left irc: Read error: Connection reset by peer
[07:41] --> m_kiewitz joined #scummvm.
[07:41] #scummvm: mode change '+o m_kiewitz' by ChanServ!ChanServ@services.
[07:55] --> _sev joined #scummvm.
[07:55] <-- _sev left irc: Changing host
[07:55] --> _sev joined #scummvm.
[07:55] #scummvm: mode change '+o _sev' by ChanServ!ChanServ@services.
[07:56] --> waltervn joined #scummvm.
[07:56] #scummvm: mode change '+o waltervn' by ChanServ!ChanServ@services.
[08:39] --> GitHub35 joined #scummvm.
[08:39] <GitHub35> [scummvm] sev- closed pull request #996: RISCOS: Add RISC OS support (master...riscos-port) https://git.io/v5fTO
[08:39] GitHub35 (GitHub35@192.30.252.34) left #scummvm.
[08:39] --> GitHub162 joined #scummvm.
[08:39] <GitHub162> [scummvm] sev- pushed 8 new commits to master: https://git.io/v50PO
[08:39] <GitHub162> scummvm/master a11985c cameron: RISCOS: Add RISC OS support
[08:39] <GitHub162> scummvm/master 7a00045 Cameron Cawley: RISCOS: Add networking files to application
[08:39] <GitHub162> scummvm/master 0e7d941 Cameron Cawley: RISCOS: Correctly set executable extension
[08:39] GitHub162 (GitHub162@192.30.252.42) left #scummvm.
[08:57] <-- LittleToonCat left irc: Remote host closed the connection
[09:47] --> Farmboy0 joined #scummvm.
[09:47] <-- Farmboy0 left irc: Changing host
[09:47] --> Farmboy0 joined #scummvm.
[09:58] <-- Farmboy0 left irc: Ping timeout: 248 seconds
[10:04] <-- Begasus left irc: Ping timeout: 252 seconds
[10:16] --> Begasus joined #scummvm.
[10:56] --> ccawley2011 joined #scummvm.
[10:59] --> criezy joined #scummvm.
[10:59] #scummvm: mode change '+o criezy' by ChanServ!ChanServ@services.
[11:04] --> ajax16384 joined #scummvm.
[11:04] #scummvm: mode change '+o ajax16384' by ChanServ!ChanServ@services.
[11:27] --> Henke37 joined #scummvm.
[11:50] --> jamm joined #scummvm.
[11:53] <-- Begasus left irc: Ping timeout: 255 seconds
[12:06] --> Begasus joined #scummvm.
[12:20] --> SylvainTV joined #scummvm.
[12:20] #scummvm: mode change '+o SylvainTV' by ChanServ!ChanServ@services.
[12:58] <-- ajax16384 left irc: Read error: Connection reset by peer
[13:05] <-- DrMcCoy left irc: Ping timeout: 246 seconds
[13:14] --> DrMcCoy joined #scummvm.
[13:14] #scummvm: mode change '+o DrMcCoy' by ChanServ!ChanServ@services.
[13:33] --> K4T joined #scummvm.
[13:33] <-- K4T left irc: Client Quit
[14:11] <-- vv222 left irc: Quit: WeeChat 1.9
[14:27] --> Littleboy joined #scummvm.
[14:27] #scummvm: mode change '+o Littleboy' by ChanServ!ChanServ@services.
[14:31] <-- TMM left irc: Quit: Ex-Chat
[14:32] --> vv222 joined #scummvm.
[15:27] <-- vv222 left irc: Quit: WeeChat 1.9
[15:27] <-- jamm left irc: Ping timeout: 246 seconds
[16:22] <-- Begasus left irc: Ping timeout: 255 seconds
[16:25] --> vv222 joined #scummvm.
[16:30] --> ny00123 joined #scummvm.
[16:34] --> Begasus joined #scummvm.
[17:22] <-- Begasus left irc: Ping timeout: 252 seconds
[17:28] --> Strangerke_ joined #scummvm.
[17:31] <-- Strangerke left irc: Ping timeout: 248 seconds
[17:31] Nick change: Strangerke_ -> Strangerke
[17:34] --> Begasus joined #scummvm.
[17:42] --> LittleToonCat joined #scummvm.
[18:16] --> dreammaster joined #scummvm.
[18:16] #scummvm: mode change '+o dreammaster' by ChanServ!ChanServ@services.
[18:40] --> ajax16384 joined #scummvm.
[18:40] #scummvm: mode change '+o ajax16384' by ChanServ!ChanServ@services.
[20:29] <dreammaster> Face slap. I'm finally making headway / realised on what the heck's going wrong with reverse playback.
[20:30] <dreammaster> The chicken dispensor video, and likely others like the Doorbot, don't have key frames for every frame. So after a frame is decoded, I use seekIntern to seek to the previous frame. But, seeking decodes every frame from the closest key frame to, now, the frame previous
[20:30] <snover> be careful with the face!
[20:30] <dreammaster> So it overrides the just previously decoded frame before it even has a chance to be used. :P
[20:30] <snover> ah, so it *did* have to do with keyframes.
[20:30] <dreammaster> Which is why I see earlier frames at the end of the chicken return animation then I should. Also why the Doorbot's frames are out of sync with the transparency track's
[20:30] <dreammaster> Yeh :P
[20:32] <dreammaster> Now I just need to decide how to fix the problem. Maybe make a seekIntern call at the start of decodeNextFrame when in reverse mode. And hope it doesn't screw up framerate timings
[20:53] <snover> i guess it is a good weekend for finally fixing previously intractable problems.
[20:54] <dreammaster> Seems like it. I saw in the logs that you were making headway with all the overlays/SDL/OpenGL stuff
[20:55] <snover> yes.
[20:55] <dreammaster> Darae I hope it'll be easier in future to implement alpha transparency for cursors? ;)
[20:55] <snover> unfortunately i am not working on cursorman.
[20:55] <dreammaster> Oh, just SCI then?
[20:56] <snover> well, SCI32 forced me to do this ;), but i had been wanting to do this for a while
[20:57] <dreammaster> I feel your pain, since it'll likely end up being "I've wanted to do this for Titanic for a while" eventually. Presuming no one else takes care of it
[20:57] <snover> with these changes, you can resize the window with the SDL2 surface renderer, the mouse cursor does not distort the coordinate system when you resize the window or run in fullscreen, a bunch of coordinate conversion hacks are gone, some code that is the same between opengl and sdl renderers is now moved to a common class
[20:58] <snover> i am forcing myself to stop deduplicating code. there is a *lot* of duplication.
[20:58] <snover> you can do mode switches in games after resizing the window and the window wont be forcibly resized back to the default size
[21:02] <dreammaster> Ah nice. I get you now. I remember all the prior discussions about mode change problems.
[21:04] <-- ajax16384 left irc: Quit: Leaving
[21:06] <-- criezy left irc: Quit: criezy
[21:06] <snover> i think i did that thing i accidentally do all the time when refactoring and made it too difficult to separate out various changes to different commits
[21:07] --> girafe joined #scummvm.
[21:12] <-- ny00123 left irc: Quit: Leaving
[21:12] <dreammaster> Yeh that sucks with it happens. Though I try to keep my commits as small and self contained as possible, I've occasionally run afoul of a similiar problem
[21:13] <snover> its just sort of a cascade and i seem to have a thing about half-finished commits
[21:16] <snover> someday i will figure out how to start working on a feature, discover refactoring is required, and then back out enough to refactor without implementing the feature simultaneously
[21:16] <snover> i hope.
[21:17] <dreammaster> We can all hope, for your own sanity, if nothing else :)
[21:21] <-- Begas_VBox left irc: Quit: Vision[0.9.8]: i've been blurred!
[21:22] <-- Begasus left irc: Quit: Ex-Chat
[21:24] <snover> whats sanity? ;)
[21:24] <snover> i think you gotta be a little crazy to choose programming as a career
[21:27] <dreammaster> I think you're right.. the lines can blur a bit between crazy and inspired at times :)
[21:27] <dreammaster> Huzzah. It works
[21:28] --> GitHub134 joined #scummvm.
[21:28] <GitHub134> [scummvm] dreammaster pushed 1 new commit to master: https://git.io/v5E3d
[21:28] <GitHub134> scummvm/master 22e24d7 Paul Gilbert: VIDEO: Fix reverse playback in AVIDecoder
[21:28] GitHub134 (GitHub134@192.30.252.42) left #scummvm.
[21:29] <dreammaster> And it fixes the sound system button too
[21:35] <-- Henke37 left irc: Quit: ERR_SHUTDOWN
[21:35] <snover> yay! :)
[21:35] <dreammaster> Let's see how the prologue Doorbot looks..
[21:40] <dreammaster> Hmmm.. not quite perfect yet. But close. I'm just wondering if I haven't accidentally stumbled on the reason for the 'frame - 1' calls. If the initial seek goes to frame - 1, then that leaves readNextPacket in a correct position to read the actual "next frame". Let me experiment a bit
[21:49] <-- girafe left irc: Read error: Connection reset by peer
[22:08] --> GitHub77 joined #scummvm.
[22:08] <GitHub77> [scummvm] csnover opened pull request #1006: COMMON: Add helper macro for printing rects (master...print-rect) https://git.io/v5EGX
[22:08] GitHub77 (GitHub77@192.30.252.40) left #scummvm.
[22:11] --> GitHub199 joined #scummvm.
[22:11] <GitHub199> [scummvm] csnover opened pull request #1007: SDL: Do not reset window size when engines update rendering surface (master...stickier-window-size) https://git.io/v5EG5
[22:11] GitHub199 (GitHub199@192.30.252.41) left #scummvm.
[22:18] --> GitHub117 joined #scummvm.
[22:18] <GitHub117> [scummvm] csnover opened pull request #1008: SDL: List supported 32bpp pixel formats when using SDL2 (master...add-32bpp-formats) https://git.io/v5EZO
[22:18] GitHub117 (GitHub117@192.30.252.34) left #scummvm.
[22:20] --> TMM joined #scummvm.
[22:20] #scummvm: mode change '+o TMM' by ChanServ!ChanServ@services.
[22:22] <-- waltervn left irc: Quit: Leaving
[22:30] --> GitHub27 joined #scummvm.
[22:30] <GitHub27> [scummvm] csnover opened pull request #1009: Graphics backends improvements (master...graphics-backends-improvements) https://git.io/v5EZh
[22:30] GitHub27 (GitHub27@192.30.252.34) left #scummvm.
[22:31] <-- Littleboy left irc: Read error: Connection reset by peer
[22:31] --> Littleboy joined #scummvm.
[22:31] #scummvm: mode change '+o Littleboy' by ChanServ!ChanServ@services.
[22:37] Nick change: Stormkeeper -> Storm-AFK
[23:00] <TMM> snover, your refactor looks great
[23:00] <TMM> snover, it may fix some of the issues I'm having playing games on my Linux tablet, quite a few engines have pointer movement scaling issues precisely for the reasons you're fixing
[23:11] <Lightkey> what tablet?
[23:16] <-- m_kiewitz left irc: Quit: technology isn't intrinsically good or evil. It's how it's used. Like the Death Ray.
[23:32] <-- ccawley2011 left irc: Ping timeout: 260 seconds
[23:37] <-- _sev left irc: Ping timeout: 246 seconds
[23:52] --> GitHub75 joined #scummvm.
[23:52] <GitHub75> [scummvm] dreammaster pushed 1 new commit to master: https://git.io/v5EW2
[23:52] <GitHub75> scummvm/master e72a1d7 Paul Gilbert: VIDEO: Simplify AVIDecoder reverse playback special handling
[23:52] GitHub75 (GitHub75@192.30.252.41) left #scummvm.
[23:59] <snover> TMM: thanks! i hope it does fix your tablet too.
[23:59] <snover> let me know if you get a chance to test it.
[00:00] --- Mon Sep 4 2017