Commit graph

5746 commits

Author SHA1 Message Date
Braden Obrzut
6a87c6cd19 - Fixed usage of uninitialized value in skybox rendering. 2016-01-14 00:49:57 -05:00
MajorCooke
629eaa35a3 - Fixed: A_Fade* could destroy live player bodies because it was calling the Destroy() function directly instead of going through P_RemoveThing. 2016-01-13 20:26:15 -06:00
Randy Heit
f9574a98fd Fix Timidity's DLS instrument loading:
- Envelope data needed to be converted to SF2 values.
- Fine tuning was ignored which made pretty much every instrument off tune.
- Despite this, it still sounds like shit compared to FMOD or Microsoft's
  wavetable synth. There are lots of missing notes and some instruments
  are still off tune. I'm not sure it's worth trying to salvage it. It'd
  probably be better to scrap it, since Timidity is very much oriented
  toward GF1 patches, which it handles perfectly fine.
2016-01-13 17:25:24 -06:00
Christoph Oelckers
d234a6af5d Merge branch 'master' of https://github.com/rheit/zdoom 2016-01-12 21:35:18 +01:00
Christoph Oelckers
37f18055af - turned sector_t::FloorSkyBox and CeilingSkyBox into an array.
- removed unused SpreadCeilingPortal function.
2016-01-12 21:34:41 +01:00
Randy Heit
dc66f4cb3f Make _WM_AdjustNoteVolume's pan_ofs a signed int 2016-01-12 06:35:05 -06:00
coelckers
db527c02f3 Merge pull request #472 from edward-san/wildmidi_fix
- Fixed Clang warning on wildMIDI code.
2016-01-12 11:21:07 +01:00
Edoardo Prezioso
b87c612aba - Fixed Clang warning on wildMIDI code.
Now that '_event_data.data' is changed to 'unsigned int', the 'unsigned long' cast is not needed anymore.
2016-01-12 11:12:19 +01:00
Randy Heit
46f04830e2 Update VC2005 project files 2016-01-11 21:11:36 -06:00
rheit
8408aec347 Merge pull request #456 from alexey-lysiuk/lzma_1514
Updated LZMA SDK to version 15.14
2016-01-11 20:53:51 -06:00
Randy Heit
141cd660eb Didn't need note_off_decay in the WildMidi sample struct 2016-01-11 20:07:14 -06:00
Randy Heit
27bf86e1eb Fix(?) dBm_pan_volume table
- The missing comma on the first line of dBm_pan_volume's definition
  looked suspicious, so I checked the MIDI specification at
  http://www.midi.org/techspecs/rp36.php and found the equations
  there gave different dB values than were in the table. So I
  rebuilt it using the equation given there:

    20*log (sin (Pi /2* max(0,CC#10 – 1)/126))
2016-01-11 20:06:07 -06:00
Randy Heit
268d2faaf6 Import volume handling changes from current WildMidi git
- Description there reads:
  > Major work done on “log” volumes. Calculations are now based on decibels.
2016-01-11 20:06:07 -06:00
Randy Heit
5b4a6483b5 Remove get_decay_samples. Not needed. 2016-01-11 20:06:06 -06:00
Randy Heit
7f01a6a30a Import envelope tweaks for current WildMidi git 2016-01-11 20:06:06 -06:00
Randy Heit
50ecec199f More stuff we don't need in WildMidi removed 2016-01-11 20:06:05 -06:00
Randy Heit
88beae55bd Add separate NRPN coarse/fine adjustment for WildMidi 2016-01-11 20:06:05 -06:00
Randy Heit
7f44bc6d0e Add Yamaha and GM reset sysex handling to WildMidi 2016-01-11 20:06:05 -06:00
Randy Heit
a74d352490 Update gus_pat.cpp's envelope fudging to current git version 2016-01-11 20:06:04 -06:00
Braden Obrzut
0c3b468e42 Merge branch 'osx_start_wnd' 2016-01-11 20:47:06 -05:00
Randy Heit
56f7ace9be include <limits.h> in the bin packers to make GCC happy 2016-01-11 17:01:29 -06:00
Christoph Oelckers
61e48013dc - fixed a warning in WildMidi code. 2016-01-11 23:46:12 +01:00
Christoph Oelckers
cd9e18a72b - fixed: operator<< (FArchive &arc, secspecial_t &p) erroneously read the old special in the compatibility handler as a short, but it must be read as an int.
- bumped savegame version to bring it in line with a GZDoom exclusive change.
2016-01-11 23:44:13 +01:00
coelckers
f539803d49 Merge pull request #471 from MajorCooke/FriendlyFireFix
Small LaxTelefragDmg Flag Fix
2016-01-11 22:45:37 +01:00
MajorCooke
7843ece01a Fixed: Friendly fire damage didn't take LAXTELEFRAGDMG into account. 2016-01-11 15:08:38 -06:00
Christoph Oelckers
011385a20e Merge branch 'master' of https://github.com/rheit/zdoom 2016-01-11 15:30:04 +01:00
Christoph Oelckers
7115590c1d - moved setup code for Eternity-style skyboxes to p_spec.cpp and consolidated most of it with nearly identical parts of the stacked sector portal setup code. 2016-01-11 15:29:44 +01:00
Randy Heit
42e9a6a711 Make atlas size a macro definition and default it to 512x512 (was 256x256) 2016-01-10 22:14:35 -06:00
Randy Heit
aaac5ac7a1 Use not so brilliant backgrounds for allocated atlas backgrounds 2016-01-10 22:01:29 -06:00
Randy Heit
670058fa3c Switch to the bottom-left picker for the Skyline Bin Packer 2016-01-10 21:54:28 -06:00
Randy Heit
4f0046105d Change texture atlas display to show allocated rectangles 2016-01-10 21:09:16 -06:00
Christoph Oelckers
918a2ed562 -fixed: Eternity-style skyboxes must check if some portal copy definitions have been set for them. 2016-01-10 22:26:08 +01:00
Christoph Oelckers
2c0f64cf9f - refactoring of R_PointToAngle2 when used to calculate direction between two actors. 2016-01-10 20:46:26 +01:00
Christoph Oelckers
1e2ce9a622 - some refactoring of P_AproxDistance calls into newly defined AActor method AproxDistance.
The main reason here is to reduce the number of instances where AActor::x and AActor::y are being referenced.
2016-01-10 17:52:41 +01:00
Randy Heit
1f34372abc Remove the upper limit on midi_voices.
- Black MIDIs can be brutal, and there's really no reason that this needs
  to be artificially limited to a paltry 256 voices.
2016-01-09 17:55:39 -06:00
Christoph Oelckers
68b4c23a82 Merge branch 'master' of https://github.com/rheit/zdoom 2016-01-09 21:10:46 +01:00
Christoph Oelckers
53d385a596 - need to validate SetSectorTerrain's 'plane' parameter, 2016-01-09 21:10:12 +01:00
coelckers
725408d9fe Merge pull request #468 from alexey-lysiuk/osx_fullscreen_shortcut
Shortcut to toggle fullscreen on OS X
2016-01-09 18:23:43 +01:00
alexey.lysiuk
b27fdcffb5 Added shortcut to toggle fullscreen on OS X
Press Command+F to toggle fullscreen mode in native OS X backend
Set k_allowfullscreentoggle CVAR to false to disable the shortcut
2016-01-09 17:12:58 +02:00
Christoph Oelckers
23cfd29dbb - let's save the terrain properties as names so that they survive a change in the definition files. 2016-01-09 12:28:42 +01:00
Christoph Oelckers
7f454358b9 - added ACS SetSectorTerrain function. 2016-01-09 12:16:41 +01:00
Christoph Oelckers
71c7f2b42c - added 'floorterrain' and 'ceilingterrain' sector properties for UDMF. These will take precedence over texture based terrain if used.
'ceilingterrain' is needed because the top of 3D-floors refers to the model sector's ceiling, so in order to give a 3D floor a terrain it must be assignable to the sector's ceiling.
Note that although it is basically the same property, its actual function bears no relevance to its use in Eternity.
2016-01-09 12:10:36 +01:00
Christoph Oelckers
4b3adf548a - fixed compile errors. 2016-01-09 11:44:05 +01:00
Christoph Oelckers
a5054b0882 Merge branch 'master' of https://github.com/rheit/zdoom 2016-01-09 10:11:46 +01:00
Christoph Oelckers
0b5e419593 - fixed friction logic for sectors.
Issues with the old code:
* when calculating friction for a 3D-floor - swimmable or not - the 3D floor's top texture must be used. The previous version always checked the sector's floor texture, even though the top might as well come from the sector's ceiling.
* 3D floors never checked for the SECF_FRICTION flag at all. According to Boom specs, sector-based friction must be ignored if this is the case.
* Terrain based friction had a higher priority than the sector's own value.

Changed the rules as follows:
* if the sector's SECF_FRICTION flag is set (i.e. something explicitly changed the sector's friction), this value is used regardless of terrain settings.
* if this flag is not set, the terrain's friction is used, if defined, using the proper plane for 3D-floors.
* otherwise the default is used.
2016-01-09 10:11:20 +01:00
Randy Heit
786caaf36b Execute disconnect scripts immediately before the player is destroyed.
- Disconnect scripts were previously run at some point after the player
  left. Now they are run immediately before destroying the player. Since
  the player hasn't actually been destroyed yet, the player also gets to
  be the script's activator. This gives you a chance to scrape whatever data
  you want from the player before they're history. Note that if you do
  anything to make the script wait, the script's activator will become the
  world, as it was before.
2016-01-08 22:41:23 -06:00
Randy Heit
bf31d66d31 Use a better packing algorithm for the texture atlases
- The old algorithm is something I threw together that produced decent,
  but not spectacular results since it had a tendency to waste space by
  forcing everything onto "shelves".
  The new packer is the Skyline-MinWaste-WasteMap-BestFirstFit algorithm
  described by Jukka Jylanki in his paper *A Thousand Ways to Pack the Bin - A
  Practical Approach to Two-Dimensional Rectangle Bin Packing*, which can
  currently be read at http://clb.demon.fi/files/RectangleBinPack.pdf
  This is minus the optimization to rotate rectangles to make better fits.
2016-01-08 22:37:06 -06:00
Randy Heit
f330a81909 Rename packing textures to texture atlases for the D3D code
- I figure since I now know the "proper" name for these things, I should
  call them by it.
2016-01-08 20:08:10 -06:00
Randy Heit
a43f5a8eca Remove now unused variable from SpawnPolyobj 2016-01-08 20:08:06 -06:00
Christoph Oelckers
cf564c60c2 - simplified setup of polyobjects defined from explicit lines
The old method had a problem with missing order numbers and aborted the level load and made many assumptions that no longer apply with BSP based polyobject rendering.
2016-01-09 01:23:00 +01:00