#gemrb@irc.freenode.net logs for 27 Feb 2012 (GMT)

Archive Today Yesterday Tomorrow
GemRB homepage


[00:12:22] --> raevol has joined #gemrb
[00:12:42] <raevol> anyone around? wondering if there is a pre-built package of the latest release of gemrb?
[00:12:58] <-- haad1 has left IRC (Quit: Leaving.)
[00:14:03] <raevol> there doesn't appear to even be a package in ubuntu 11.10 :[
[00:19:05] <tomprince> raevol: I'd guess not for linux, unless it is a third-party one. Maybe playdeb?
[00:19:13] <raevol> hmm ok
[00:19:30] <raevol> i'll have to mess with it some other time
[00:19:38] <raevol> ttys!
[00:19:41] <-- raevol has left IRC (Quit: Leaving.)
[01:00:19] <-- edheldil_ has left IRC (Ping timeout: 260 seconds)
[03:10:56] --> brad_a has joined #gemrb
[03:11:34] <brad_a> i assume we are aware that gemrb isnt currently building
[03:15:02] <CIA-28> GemRB: 03bradallred * re33105e10c7d 10gemrb/gemrb/core/Spellbook.h: Spellbook: add missing method definition preventing compilation.
[03:20:26] <gembot> build #592 of cmake clang++ is complete: Success [3build successful] Build details are at http://buildbot.gemrb.org/builders/cmake%20clang%2B%2B/builds/592
[03:24:05] <gembot> build #203 of cmake g++-4.2 is complete: Success [3build successful] Build details are at http://buildbot.gemrb.org/builders/cmake%20g%2B%2B-4.2/builds/203
[03:24:31] <gembot> build #200 of osx-test is complete: Success [3build successful] Build details are at http://buildbot.gemrb.org/builders/osx-test/builds/200
[03:27:36] <gembot> build #206 of cmake g++-4.4 is complete: Success [3build successful] Build details are at http://buildbot.gemrb.org/builders/cmake%20g%2B%2B-4.4/builds/206
[03:30:57] <gembot> build #205 of cmake g++-4.6 is complete: Success [3build successful] Build details are at http://buildbot.gemrb.org/builders/cmake%20g%2B%2B-4.6/builds/205
[03:31:28] <gembot> build #179 of osx-xcode-binary is complete: Success [3build successful] Build details are at http://buildbot.gemrb.org/builders/osx-xcode-binary/builds/179
[03:31:33] --> joneirik has joined #gemrb
[03:34:40] <gembot> build #197 of cmake g++-4.5 is complete: Success [3build successful] Build details are at http://buildbot.gemrb.org/builders/cmake%20g%2B%2B-4.5/builds/197
[03:52:34] <gembot> build #251 of nmake-msvc++10 is complete: Success [3build successful] Build details are at http://buildbot.gemrb.org/builders/nmake-msvc%2B%2B10/builds/251
[04:08:36] <gembot> build #535 of mingw32 is complete: Success [3build successful] Build details are at http://buildbot.gemrb.org/builders/mingw32/builds/535
[04:12:05] <gembot> build #522 of msvc++6 is complete: Success [3build successful] Build details are at http://buildbot.gemrb.org/builders/msvc%2B%2B6/builds/522
[04:21:26] <gembot> build #295 of nmake-msvc++6 is complete: Success [3build successful] Build details are at http://buildbot.gemrb.org/builders/nmake-msvc%2B%2B6/builds/295
[05:05:08] <-- joneirik has left IRC (Remote host closed the connection)
[05:23:38] <CIA-28> GemRB: 03bradallred * rc1249eab9e2b 10gemrb/gemrb/plugins/SDLVideo/ (SDLVideo.cpp SDLVideo.h): SDL video: we dont actually need or use the lastEvent ivar. let's remove it and use a local variable in PollEvents.
[05:23:49] <CIA-28> GemRB: 03bradallred * r3b60568854c0 10gemrb/gemrb/plugins/SDLVideo/SDL20Video.cpp: SDL 2 video: we should be using the passed in event for processing and not lastEvent (even though they are the same)
[05:50:04] <-- brad_a has left IRC (Quit: brad_a)
[07:07:48] --> edheldil_ has joined #gemrb
[07:14:08] <-- edheldil_ has left IRC (Ping timeout: 244 seconds)
[07:57:14] --> lynxlynxlynx has joined #gemrb
[07:57:14] --- ChanServ gives channel operator status to lynxlynxlynx
[08:11:28] --> edheldil has joined #gemrb
[08:11:28] --- ChanServ gives channel operator status to edheldil
[09:11:19] <wjp> I wonder how SSE-friendly those inner loop blit operations are, and how good g++ is at auto-vectorization these days. I should have a look someday :-)
[09:12:14] <fuzzie> you're doing it with constants?
[09:12:44] <wjp> the shifting? yes
[09:45:02] --> SiENcE has joined #gemrb
[13:50:18] <fuzzie> our console output seriously needs some padding
[13:51:15] <edheldil> what do you mean? Indentation?
[13:51:34] <fuzzie> I guess? right now it's "[Section:] Text"
[13:51:48] <fuzzie> would be nicer if the 'Text' bits all (or mostly) lined up
[13:57:15] <fuzzie> ok, we really are back to the squirrel death thing
[13:57:23] <fuzzie> that is sad
[13:57:34] <fuzzie> who broke it?
[14:01:18] <fuzzie> lynxlynxlynx: j'accuse..!
[14:05:18] <CIA-28> GemRB: 03fuzzie * r9dd155bba4f0 10gemrb/gemrb/core/Scriptable/Actor.cpp:
[14:05:18] <CIA-28> GemRB: GetHpAdjustment: try harder not to kill squirrels.
[14:05:18] <CIA-28> GemRB: This adds the player class check back that was removed in 829c8f69
[14:05:18] <CIA-28> GemRB: ("GetHpAdjustment: try harder not to kill actors").
[14:05:37] <fuzzie> think of the squirrels.
[15:02:17] <lynxlynxlynx> didn't the min hp check work?
[15:07:30] <fuzzie> well, I didn't actually debug that, since this stuff should only apply to PC classes anyway, right?
[15:39:42] <fuzzie> in fact probably am not going to find time to debug that.
[15:55:29] <lynxlynxlynx> avenger said it shouldn't, that's why it was removed
[15:55:57] <lynxlynxlynx> but it is not nicely selfcontained
[16:10:47] <-- lynxlynxlynx has left IRC (Read error: Operation timed out)
[16:27:47] --> lynxlynxlynx has joined #gemrb
[16:27:47] <-- lynxlynxlynx has left IRC (Changing host)
[16:27:47] --> lynxlynxlynx has joined #gemrb
[16:27:47] --- ChanServ gives channel operator status to lynxlynxlynx
[17:22:25] <fuzzie> well, atm everything is expecting the actor to be a PC
[17:22:39] <fuzzie> but now I am unhelpfully not home amd cannot check original, sorry
[17:22:51] <fuzzie> will keep it in mind
[17:28:20] --> kida_laptop has joined #gemrb
[17:31:04] <-- SiENcE has left IRC (Quit: @all: cya)
[17:39:39] <CIA-28> GemRB: 03wjpalenstijn * r813d3d44e725 10gemrb/gemrb/plugins/SDLVideo/ (SDLVideo.cpp SpriteRenderer.inl): Add 16bpp support to templated sprite blitter
[17:39:42] <CIA-28> GemRB: 03wjpalenstijn * rbe49c4b16da3 10gemrb/gemrb/plugins/SDLVideo/ (SDLVideo.cpp SpriteRenderer.inl): Add grey/sepia effects to template sprite blitter
[17:39:42] <CIA-28> GemRB: 03wjpalenstijn * rb9ae630c6fa3 10gemrb/gemrb/plugins/SDLVideo/ (Makefile.am SDLVideo.cpp SpriteRenderer.inl):
[17:39:42] <CIA-28> GemRB: Add RLE BAM templated sprite renderer
[17:39:42] <CIA-28> GemRB: This temporarily disables non-RLE BAM rendering and 16bpp output.
[17:39:43] <CIA-28> GemRB: 03wjpalenstijn * ra0c06224f849 10gemrb/gemrb/plugins/SDLVideo/ (4 files):
[17:39:43] <CIA-28> GemRB: Merge branch 'blitter'
[17:39:44] <CIA-28> GemRB: This is a rewrite of the sprite blitter, turning it from
[17:39:44] <CIA-28> GemRB: preprocessor-magic-heavy into template-magic-heavy.
[17:39:45] <CIA-28> GemRB: 03wjpalenstijn * r34e81a638a07 10gemrb/gemrb/plugins/SDLVideo/ (SDLVideo.cpp SpriteRenderer.inl): Add non-BAM RGB templated blitter
[17:39:45] <CIA-28> GemRB: 03wjpalenstijn * r2ac260561a26 10gemrb/gemrb/plugins/SDLVideo/ (SDLVideo.cpp SpriteRenderer.inl): Add non-RLE BAM templated blitter
[17:39:51] <CIA-28> GemRB: 03wjpalenstijn * rd12a09b5728d 10gemrb/gemrb/plugins/SDLVideo/ (Makefile.am SDLVideoDriver.inl): Remove now unused SDLVideoDriver.inl
[17:39:58] <wjp> interesting commit order there
[17:41:22] --> brad_a has joined #gemrb
[17:42:02] <fuzzie> hm, definitely not ordered by hash
[17:43:07] <tomprince> I've noticed that CIA is "interesting" in that regard.
[17:43:26] <fuzzie> yes.
[17:43:37] <fuzzie> does it from github etc too.
[17:44:03] * wjp waves bye-bye to SDLVideoDriver.inl
[17:44:53] <fuzzie> :-)
[17:46:51] <lynxlynxlynx> any practical benefits from this rewrite?
[17:51:37] <wjp> well
[17:51:48] <wjp> depends on what you mean with practical :-)
[17:51:59] <wjp> it should be a lot more readable and maintainable
[17:52:44] <wjp> performance seems almost exactly the same in the scenes I've been trying
[17:52:53] <wjp> binary size is a bit bigger
[17:53:19] <wjp> haven't measured compile time
[17:56:46] <lynxlynxlynx> ok, nothing extra interesting for the changelog then
[17:57:21] <wjp> not really, except maybe "introduced a number of subtle rendering regressions" ;-)
[17:57:41] <wjp> (not that I'm aware of any, but...)
[18:01:41] <CIA-28> GemRB: 03lynxlupodian * rb82a3223d6c8 10gemrb/NEWS: regular NEWS bump
[18:01:58] <tomprince> Support 32 bit map tiles?
[18:02:41] <wjp> tiz?
[18:02:48] <tomprince> Yes.
[18:03:30] <wjp> you'll want to look at BlitTile and TileRenderer.inl for that
[18:05:56] <wjp> I'd suggest converting them to screen format on load, and then the rendering should be very similar to the current function
[18:09:25] <wjp> although keeping it 32 bit would probably make tinting cheaper
[18:56:16] --> Yoshimo has joined #gemrb
[20:19:00] <brad_a> any decision on GSOC?
[20:20:19] --> haad has joined #gemrb
[20:44:03] --> SiENcE has joined #gemrb
[20:46:14] <lynxlynxlynx> i'm still for it, just too swamped atm
[21:04:34] --> edheldil_ has joined #gemrb
[21:13:01] <lynxlynxlynx> oh, and feel free to edit the idea page
[21:13:27] <lynxlynxlynx> add more detail, maybe something new, etc
[21:31:17] <CIA-28> GemRB: 03avenger_teambg * r2e625d5e6d2a 10gemrb/gemrb/ (6 files in 2 dirs): adding stunning fist, reworked giving simple innates by guiscript
[21:39:05] <-- Yoshimo has left IRC (Ping timeout: 260 seconds)
[21:51:16] <-- wrotek has left IRC (Read error: Connection reset by peer)
[21:54:16] <CIA-28> GemRB: 03bradallred * r781b229cbed0 10gemrb/apple/osx/ (5 files in 2 dirs):
[21:54:16] <CIA-28> GemRB: Mac build: delete any previous plugins prior to installation to ensure that if
[21:54:16] <CIA-28> GemRB: the user had installed a previous version with an optional plugin then upgrades
[21:54:16] <CIA-28> GemRB: and opts not to install the optional plugin then no problems occur from GemRB
[21:54:16] <CIA-28> GemRB: trying to load and use a potentially out of date plugin.
[22:02:16] --> Avenger has joined #gemrb
[22:02:16] --- ChanServ gives channel operator status to Avenger
[22:02:17] <lynxlynxlynx> installing doesn't happen as root, right?
[22:02:30] <Avenger> anything i should know about the sdl changes?
[22:02:42] <brad_a> only if they build/install via cmake
[22:02:58] <brad_a> speaking of i probably should have added that script to cmake
[22:03:03] <Avenger> my msvc compile asserts on start
[22:06:00] <wjp> uh oh
[22:06:48] <wjp> any details?
[22:07:17] <Avenger> assert(cover)
[22:09:40] <lynxlynxlynx> brad_a: ok, just so i know where to symlink critical files >:)
[22:09:40] <wjp> bah
[22:09:54] <wjp> so it's calling the wrong instantiation? :-(
[22:09:57] <brad_a> bah I didn't realize rm would return an error if no plugins exist...
[22:10:12] <lynxlynxlynx> use -f
[22:10:13] <wjp> brad_a: rm -f
[22:10:18] <brad_a> ill need to fix that or install will fail if its not an upgrade :(
[22:10:22] <brad_a> ah
[22:10:24] <brad_a> thaks!
[22:10:38] <brad_a> very nice
[22:11:25] <wjp> hm, well, this sucks
[22:11:26] <brad_a> lynx wont rm jsut remove the symlinks :)
[22:11:34] <lynxlynxlynx> yes
[22:12:09] <wjp> so are we actually going to have to wrap true and false in classes?
[22:14:16] <Avenger> i guess, eventually, tom will wrap everything in classes
[22:14:27] <Avenger> even wrappers
[22:15:08] <wjp> I suppose I should have realized that if it fails on typenames it will also fail on bools
[22:19:22] <Avenger> this means, no gemrb for me today?
[22:19:53] <Avenger> well, nevermind, i gotta sleep then :) i won't have time to refactor the feat stuff
[22:20:01] <-- Avenger has left IRC (Quit: bye!)
[22:20:27] <wjp> the exact brokenness of msvc6's templates is only now starting to sink in
[22:20:53] <wjp> argh, this is just too annoying
[22:22:27] <-- haad has left IRC (Quit: Leaving.)
[22:23:07] <-- edheldil_ has left IRC (Ping timeout: 246 seconds)
[22:30:34] --> edheldil_ has joined #gemrb
[22:32:43] <brad_a> avenger: did that intel compiler not work?
[22:36:59] <fuzzie> intel compiler?
[22:37:56] <CIA-28> GemRB: 03wjpalenstijn * r0b675f6ea2ff 10gemrb/gemrb/plugins/SDLVideo/SpriteRenderer.inl: Fix templates for MSVC6, hopefully
[22:38:27] * wjp wasn't thinking, as usual
[22:38:34] <wjp> anyway, this should do the trick
[22:38:36] <fuzzie> :)
[22:43:08] <wjp> a lot of SSE instructions end up in those blitter loops, by the way, but I haven't looked more than superficially
[22:44:13] <wjp> but bedtime now; good night
[22:44:27] <brad_a> fuzzie: tomprince linked something about using an intel compiler though msvc6 that would hopefully make it less of a hastle
[22:46:41] <fuzzie> isn't it $500, though?
[22:48:33] <brad_a> i have no idea :) i didnt even think about $
[22:50:23] <fuzzie> i mean, one obstacle to trying a new msvc release is that that is *also* $500 or so
[22:50:54] <fuzzie> and "use msvc6 for some stuff, and msvc express for other stuff" is not very convincing argument
[22:52:27] <fuzzie> well, maybe if vs2010 wasn't sodding awful in general :/
[22:53:03] <fuzzie> so very glad that as a student I can just grab vs2008 without having to worry about paying for it
[23:02:17] <brad_a> im surprised to hear you use VS at all ;-)
[23:02:45] <fuzzie> yes, well, if only we lived in a world without Windows users.
[23:03:20] <brad_a> never done any non CLI windows programming
[23:03:37] <brad_a> i guess i used mingw with some widget something or other in school way back when
[23:05:12] <fuzzie> gcc on Windows, on the other hand, is an an abomination.
[23:15:24] <fuzzie> I see that I can still kill Jon in the dungeon escape cutscene, which I suppose is one way to cut bg2 short.
[23:19:08] <-- edheldil_ has left IRC (Ping timeout: 252 seconds)
[23:23:45] <brad_a> he he
[23:23:56] <brad_a> i actually thought that was fixed
[23:24:41] <brad_a> maybe i jsut forgot how to exploit it tho
[23:29:54] <lynxlynxlynx> ctrl+y if you pause soon enough
[23:32:12] <-- lynxlynxlynx has left IRC (Remote host closed the connection)
[23:47:56] <gembot> build #527 of msvc++6 is complete: Failure [4failed compile] Build details are at http://buildbot.gemrb.org/builders/msvc%2B%2B6/builds/527 blamelist: wjpalenstijn@users.sourceforge.net
[23:52:23] <gembot> build #300 of nmake-msvc++6 is complete: Failure [4failed compile] Build details are at http://buildbot.gemrb.org/builders/nmake-msvc%2B%2B6/builds/300 blamelist: wjpalenstijn@users.sourceforge.net
[23:58:53] <-- SiENcE has left IRC (Quit: cya)