#exult@irc.freenode.net logs for 17 Nov 2016 (GMT)

Archive Today Yesterday Tomorrow
Exult homepage


[02:21:08] <-- frob has left IRC (Ping timeout: 268 seconds)
[02:36:25] --> DominusExult has joined #exult
[02:36:25] <-- DominusExult has left IRC (Changing host)
[02:36:25] --> DominusExult has joined #exult
[02:36:25] --- ChanServ gives channel operator status to DominusExult
[02:36:54] <-- Dominus has left IRC (Ping timeout: 256 seconds)
[02:36:54] --- DominusExult is now known as Dominus
[04:27:29] --> frob has joined #exult
[04:55:19] <-- frob has left IRC (Remote host closed the connection)
[05:10:15] --> frob has joined #exult
[05:31:29] <-- frob has left IRC (Remote host closed the connection)
[06:16:18] --> frob has joined #exult
[06:16:45] <-- frob has left IRC (Remote host closed the connection)
[06:41:43] --> i30817 has joined #exult
[06:44:15] <i30817> Hey guys - I decided to cut down on my version of the mt32emu patch for exult and use the local version and only leave the hardcode of the mt32 rom to a 'system location' so it can be used by both it and dosbox.
[06:44:16] <i30817> I noticed that you can't install the previous version of exult 64bits version without uninstalling the 32bits dosbox. Which is a problem since dosbox 32 bits still has the fastest jit.
[06:44:58] <i30817> eh, it's a bit sad to not auto update but eh.
[06:47:19] --> frob has joined #exult
[07:01:43] <-- i30817 has left IRC (Quit: ChatZilla 0.9.93 [Firefox 49.0.2/20161025170400])
[07:22:23] <Dominus> Why what how? Why does 64bit Exult affect 32bit Dosbox? Makes no sense...
[08:35:27] <-- Lightkey has left IRC (Ping timeout: 240 seconds)
[08:48:27] --> Lightkey has joined #exult
[10:14:30] <Marzo> The only way Exult and DOSBox would interfere like that is if you are providing libmt32emu for 64-bit only; so installing the lib would uninstall the older versions and leave no 32-bit version for DOSBox
[11:27:43] --> i30817 has joined #exult
[11:34:02] <i30817> actually, i think i figured out at least part of the problem meanwhile.
[11:34:04] <i30817> the packaging was missing the Multi-Arch field on both produced packages (one a simple script to install the rom for the 'system location' the other the lib itself. I figured it out when i built the new version patch (that only searches the rom at that system location), and it still had the problem.
[11:34:05] <i30817> Although i'm not confident in enabling both 64bits and 32bits versions of a -dev package (the ubuntu wiki page makes it sound like it's 'planned in the future') so i'll still use the new version, only enabling 'multi-arch any' for that script for now and not having both version of libmt32emu installable at the same time.
[11:40:31] <i30817> actually it isn't even needed to enable multiarch for the -dev package doh, it's a buildbot requirement only, since it's statically linked dev library anyway. The error must have been only on the script.
[11:41:33] <-- i30817 has left IRC (Quit: ChatZilla 0.9.93 [Firefox 49.0.2/20161025170400])
[12:35:29] <Dominus> Marzo, another oddity, shape 874 is used in two instances to "cover" the hole in dungeon's 2nd floor
[12:35:59] <Dominus> dec coordinates 804x955 and 686x1191
[12:36:26] <Dominus> but it is messed up in Exult, but when I use map_patch to remove them it looks rather good :)
[12:37:39] <Dominus> Looks better than the original ;)
[12:38:05] <Dominus> I wonder why that shape was even used there
[12:40:02] <Dominus> We mess up some translucency there but even with correct translucency it doesn't look as nice in the original
[12:43:47] <Dominus> http://pastebin.com/G15SfaAT below the sawdust patch
[12:44:16] <Dominus> and I can't see any problem yet when playing with taht patch
[12:45:50] <Dominus> hmm, probably the original used that to have a floor above to pop the dungeon roof there.
[12:45:58] <Dominus> works in Exult without
[14:19:26] <Dominus> Marzo: I think our monster spawning should check for walls and not spawn in walls
[14:20:04] <Dominus> I have a map_patch for SI that gets rid of an egg in a wall that spawns spiders, but doesn't spawn at all in the original http://pastebin.com/n75HSCYZ
[15:13:05] <Dominus> odd, I tried deleting that floor piece (874) in the original and that has no effect as well on roof/upper floor popping out of view or not
[15:39:09] --> GitHub has joined #exult
[15:39:09] <GitHub> [exult] DominusExult pushed 1 new commit to master: https://git.io/vXHpT
[15:39:09] <GitHub> exult/master 2e9f9dc DominusExult: added several map_patches to BG and SI to circumvent Exult issues
[15:39:09] <-- GitHub has left #exult
[15:40:10] <-- frob has left IRC (Remote host closed the connection)
[15:49:40] <Dominus> Marzo, just now tried your experimental last commit and it fails on OS X
[16:10:12] <Marzo> Dominus: yeah, there is a reason why I said it was experimental
[16:10:47] <Marzo> I need to figure out how to configure it for clang
[16:20:14] <Dominus> Let me know if I can test something - just not in the next three hours ;)
[16:46:26] --> frob has joined #exult
[16:50:50] <-- frob has left IRC (Ping timeout: 260 seconds)
[16:59:20] --> frob has joined #exult
[16:59:25] <-- frob has left IRC (Remote host closed the connection)
[17:15:00] --> frob has joined #exult
[17:20:14] <-- frob has left IRC (Ping timeout: 268 seconds)
[17:36:07] <Marzo> Dominus: it makes no sense not working on OSX; the "missing" macro (AX_CHECK_COMPILE_FLAG) is used to select flags since a few commits back, and it is right there in m4/ax_check_compile_flag.m4
[17:37:48] <Marzo> O
[17:37:52] <Marzo> *OH
[17:38:19] <Marzo> It is actually AX_REQUIRE_DEFINED which is missing
[17:42:16] --> GitHub has joined #exult
[17:42:16] <GitHub> [exult] marzojr pushed 1 new commit to master: https://git.io/vXQcC
[17:42:16] <GitHub> exult/master 4548dff Marzo Sette Torres Junior: Fixing bug #1988 "Configure broken on OS X" by adding missing AX_REQUIRE_DEFINED macro (needed by fff311ba95d226cd35a2dc2ae777203fc9ecd076).
[17:42:16] <-- GitHub has left #exult
[17:46:08] --> frob has joined #exult
[17:50:44] <-- frob has left IRC (Ping timeout: 250 seconds)
[17:59:36] --> frob has joined #exult
[18:06:01] <Dominus> Marzo, confirmed to be working... 2-3 minutes for a build now...
[18:08:24] <Dominus> with the move to github, changelogs get neglected more and more...
[18:32:04] <Marzo> Dominus: that has more to do with git than github; we can generate a changelog easily from the git history with "git log"
[18:33:23] <Dominus> Yes, I did the now common mistake of declaring git == github ;)
[18:34:11] <Dominus> Seriously, I don't really mind the neglected changelog, it's just something that I noticed.
[18:34:37] <Dominus> You need to make more of a conscious effort to write a changelog entry.
[18:35:59] <Marzo> When I write the git log entry, I write hat I would have written in changelog
[18:36:36] <Marzo> When I as doing both I wrote the changelog and copied to the git commit message
[18:58:09] <Marzo> Dominus: git log <start-revision>...<end-revision> --pretty=format:'%h %aai%x09%an <%ae>%n%w(90,4,4)%B%n'
[18:59:03] <Marzo> By messing with the format string, we can make a colorful log, make it link to the github commit diff, and so on
[19:01:14] <Marzo> And if you omit both revisions nd the ellipsis, it defaults to all revisions
[19:21:36] <Dominus> ok, shape #874 annoys the hell out of me...
[19:22:24] <Dominus> Marzo, I'm puzzled in Exult Studio that shape #874 is not present with plain Black Gate but it is there for FoV
[19:26:13] <Dominus> really really odd
[19:26:34] <Dominus> it's not in BG, only in FoV
[19:37:18] <-- frob has left IRC ()
[19:45:06] <Dominus> it's in the German BG but with no transparency at all, looking in Dosbox just like Exult did it
[19:47:16] <Dominus> it's not in the Spanish version
[19:48:44] <Dominus> and not in the French version
[19:49:05] <Dominus> Seems like they had a problem with this shape...
[19:55:24] <Dominus> so the original already could use blanks in shapes.vga?
[20:06:01] <Dominus> Funny, I feel like I stumbled over something up to now unknown...
[20:07:26] <Dominus> I wonder where else shape 874 is getting used and whether it is as problematic... got to look through the isle of fire...
[21:39:58] <Marzo> Dominus: maybe the German BG was built off of an intermediate version between BG and FoV?
[21:40:09] <Marzo> Is shape 874 even used in plain BG?
[21:41:42] <Marzo> Shape 874 exists in BG, but it is empty — it has one empty frame
[21:51:22] <Dominus> Marzo if you use the fov shapes.vga with pure bg (I tried only with the french version) then the parts that I map_patched look like the German BG (and Exult)
[21:52:33] <Marzo> German BG has the same shapes.vga as FoV
[21:52:42] <Dominus> Ah
[21:52:47] <Dominus> And odd
[21:53:25] <Marzo> Spanish and French BG has the same as English BG
[21:53:51] <Marzo> Meaning German BG was probably built from a version in-between BG and FoV
[21:54:13] <Dominus> So definitely need to look through the FoV part to see whether #874 gets used and whether it looks ok or bad
[21:54:20] <Marzo> It is likely that map is plain BG, but maybe we can mine something interesting from the usecode
[21:56:00] <Dominus> So they definitely added some transparency code for that shape in FoV but not yet in German BG
[21:57:40] <Dominus> And we don't same as pure BG hence it looks so bad ;)
[21:58:57] <Dominus> Gonna add a bug report for that tomorrow- now I need to go to bed
[21:59:05] <Marzo> Maybe they just removed it from the map?
[22:00:26] <Dominus> ?
[22:01:00] <Dominus> It's there in the FoV map at the same spots
[22:01:12] <Dominus> Shame and Despise towers
[22:01:58] <Dominus> But parts of the shape are transparent so it doesn't look that bad as in German BG
[22:02:48] <Dominus> The brown part is still there to see which makes it slightly odd when you look through the stairways snd see dirt ground
[22:02:59] <Dominus> While the ground should be stone
[22:03:35] <Dominus> Good night ;)