#exult@irc.freenode.net logs for 27 Nov 2016 (GMT)

Archive Today Yesterday Tomorrow
Exult homepage

[00:01:04] <wjp> well, "Much of this code is adapted from the XMIDI implementation from the exult project" :-)
[00:03:32] <Marzo> Oh, wait, they fixed it
[00:03:45] <Marzo> There is a -1 there
[00:05:05] <Marzo> In any case, I sent a message to Colourless to see if he can help me figure out what he wanted to do in a part of the LowLevelMidiDriver
[00:05:17] <Marzo> It is unchanged since his first commit
[00:06:02] <Marzo> And without the formatting, it looks like this:
[00:06:03] <Marzo> if (phys != -3) continue;
[00:06:04] <Marzo> if (phys == -3) protectChannel(sequence_id, c, false);
[00:06:04] <Marzo> else lockChannel(sequence_id, c, false);
[00:08:45] <Marzo> I think I have done enough for today
[00:09:07] --> GitHub has joined #exult
[00:09:07] <GitHub> [exult] marzojr pushed 24 new commits to master: https://git.io/v1kGO
[00:09:07] <GitHub> exult/master d9c0477 Marzo Sette Torres Junior: Replaced several pre-checked dynamic_casts by static_casts. Deleted...
[00:09:07] <GitHub> exult/master 4089936 Marzo Sette Torres Junior: Allow reading fixed-size strings from DataSources.
[00:09:07] <GitHub> exult/master 5a9e0a4 Marzo Sette Torres Junior: Fixing data requirement error in makefile.
[00:09:07] <-- GitHub has left #exult
[00:09:30] --> GitHub has joined #exult
[00:09:30] <GitHub> [exult] marzojr merged master into coverity_scan: https://git.io/v1kG3
[00:09:30] <-- GitHub has left #exult
[00:12:41] <wjp> that first check looks like the broken one
[00:13:32] <wjp> checking if a channel is locked is probably (phys >= 0 && chan_locks[phys] == sequence_id)
[00:15:06] <wjp> but might need some closer look to see if there are corner cases
[00:33:13] <Marzo> Hm. I just noticed that there is a bug in the midi playback; the first few notes sound kind of rushed
[00:40:05] <Marzo> This only happens when changing songs, so it may be due to timbre caching
[00:40:17] <Marzo> Maybe timing is not being correctly accounted for in the event list
[00:52:50] <Marzo> Hm, the songs play back if the previous song has finished instead of being killed
[01:24:43] --> GitHub has joined #exult
[01:24:43] <GitHub> [exult] marzojr pushed 1 new commit to master: https://git.io/v1kng
[01:24:43] <GitHub> exult/master b514a51 Marzo Sette Torres Junior: Fixing a couple typos.
[01:24:43] <-- GitHub has left #exult
[01:30:23] <-- Marzo has left IRC (Ping timeout: 240 seconds)
[02:03:26] --> DominusExult has joined #exult
[02:03:26] <-- DominusExult has left IRC (Changing host)
[02:03:26] --> DominusExult has joined #exult
[02:03:26] --- ChanServ gives channel operator status to DominusExult
[02:07:25] <-- Dominus has left IRC (Ping timeout: 245 seconds)
[02:07:25] --- DominusExult is now known as Dominus
[08:33:03] <-- Lightkey has left IRC (Ping timeout: 240 seconds)
[08:46:07] --> Lightkey has joined #exult
[10:12:06] <Dominus> he he, the function that gets the available screen sizes is described as //Destroy window :)
[10:12:38] <Dominus> some copy/paste error since the *next* function is described the same and that actually does that :)
[10:57:19] --> Marzo has joined #exult
[11:08:43] <Marzo> I think I will add some enums to LowLevelMidiDriver to make for easier reading
[11:36:34] <Marzo> wjp: by the way, everything seems to sound fine with that suggestion you made; but I don't know if it is because it is correct or because nothing uses the xmidi events that lock/unlock or protect/unprotect channels
[11:36:52] <Marzo> ScummVM, at least, doesn't even implement these commands
[11:37:08] <Marzo> So it may well be that it turned out to never have been used
[11:58:23] <-- wjp has left IRC (Quit: reboot)
[12:00:53] --> wjp has joined #exult
[12:00:53] --- ChanServ gives channel operator status to wjp
[15:07:19] <Dominus> Hmm, Marzo that high-dpi vs pixel iOS problem should be solved by adding the flag SDL_WINDOW_ALLOW_HIGHDPI to SDL_CreateWindow https://wiki.libsdl.org/SDL_CreateWindow
[15:07:58] <Dominus> But it seems that only works if we use Opengl for rendering instead of surface
[15:08:01] <Dominus> I think
[15:08:48] <Dominus> Since we don't that has no effect on Exult (also not in OS X)
[15:08:59] <Marzo> Maybe it is not supported for surface yet, and you need to file a bug report
[15:09:47] <Marzo> Or maybe we need to add additional scaling in high-dpi mode in non-OpenGL
[15:15:09] <Dominus> Or we need an OpenGl output...
[15:15:12] <Dominus> ;)
[15:16:20] <Dominus> It's driving me crazy, especially as I can't sit long enough at the PC to try and figure things out before there is something else in our home to do
[16:13:32] --> GitHub has joined #exult
[16:13:32] <GitHub> [exult] marzojr pushed 2 new commits to master: https://git.io/v1kHM
[16:13:32] <GitHub> exult/master e5a5b8e Marzo Sette Torres Junior: Fixing data sent back for Exult_server::locate_shape (not that it...
[16:13:32] <GitHub> exult/master ed2ed9b Marzo Sette Torres Junior: Preventing more segfaults in ucc when handling classes.
[16:13:32] <-- GitHub has left #exult
[21:24:10] <-- Marzo has left IRC (Ping timeout: 268 seconds)
[21:28:43] <-- Rottingbeef has left IRC (Remote host closed the connection)
[21:28:57] --> Rottingbeef has joined #exult
[22:27:26] --> Marzo has joined #exult