rogerman
d627a5b6aa
OpenGL Renderer:
...
- Fix bug where certain scenes would fail to draw when edge marking is enabled.
2015-04-28 00:39:48 +00:00
rogerman
e89544aa7b
OpenGL Renderer:
...
- Fix some more graphical bugs with fog and translucent fragments.
- Do some small optimizations to the fragment shaders.
2015-04-27 21:18:48 +00:00
rogerman
ecc22fb24d
OpenGL Renderer:
...
- Don’t output the depth to the secondary depth buffer if depth writing is disabled.
2015-04-25 19:07:54 +00:00
rogerman
74dbe2efdd
OpenGL Renderer:
...
- Fix compiling for platforms that aren't OS X. (Regression from r5169.)
2015-04-25 03:08:20 +00:00
rogerman
398cd31bf3
Cocoa Port:
...
- In the 3D Rendering Settings panel in the DeSmuME Preferences view, move the Enable Edge Marking and Enable Fog checkboxes to the General Settings section.
- Update tooltips to reflect the new behavior.
2015-04-25 02:52:57 +00:00
rogerman
d652249c34
OpenGL Renderer:
...
- Add support for edge marking.
2015-04-25 02:04:13 +00:00
rogerman
629da9c8e2
OpenGL Renderer:
...
- Fix compiling for platforms that aren't OS X. (Regression from r5166.)
2015-04-24 09:00:54 +00:00
rogerman
f66287d0e0
OpenGL Renderer:
...
- Fix some graphical bugs with fog and translucent fragments.
2015-04-23 22:46:37 +00:00
rogerman
2ae84828e5
OpenGL Renderer:
...
- Add fog support.
2015-04-23 20:01:51 +00:00
rogerman
4fdcffa126
OpenGL Renderer:
...
- Do a bunch of code cleanup.
- Prepare the code for doing multipass rendering for the implementation of fog and edge mark.
2015-04-22 21:38:31 +00:00
zeromus
db5b63fc2e
fix vs2012 vcxproj
2015-04-18 16:17:49 +00:00
rogerman
21fce2cab0
OpenGL Renderer:
...
- Bring back vertex draw batching from r4522, fixing the bug that caused it to fail on Metroid Prime Hunters. This gives a small performance improvement for users with older drivers.
2015-04-16 21:55:57 +00:00
zeromus
e0aa9eb56b
update to fex github.com/kode54/File_Extractor/commit/e9e7560df9ca9b25348e0bfb4012f6e570711bb3
...
should fix new format RARs that dont work, but havent tested it
2015-04-15 06:04:46 +00:00
zeromus
66e314ec65
add a stdint for old visualstudios
2015-04-15 06:00:23 +00:00
zeromus
c767066858
Merged revision(s) 5157-5158 from branches/release_0_9_11/desmume:
...
fix files missing from Makefile.am
........
oops this should probably be tabs
........
2015-04-14 23:57:20 +00:00
zeromus
6ca691914e
png crunch
2015-04-14 22:58:23 +00:00
rogerman
e526546bc0
OpenGL Renderer:
...
- Remove all of the state caching code. Things actually run a little faster without it.
- Do a bunch of other code cleanup.
2015-04-07 22:26:39 +00:00
zeromus
6af7350461
fix small mistake in r5121 that whacks all battery saves when making a savestate
2015-04-05 18:13:43 +00:00
rogerman
207c00df6b
Cocoa Port:
...
- Update all info .plist files to v0.9.12 for future SVN builds.
2015-03-31 01:34:11 +00:00
zeromus
e973f53e52
bump version number for trunk svn builds
2015-03-29 09:12:27 +00:00
zeromus
95ba5d8a97
clarify new version changelog
2015-03-29 09:11:11 +00:00
rogerman
ccbf85ed42
Cocoa Port:
...
- Have SoftRasterizer’s Fragment Sampling Hack be disabled by default.
2015-03-29 01:14:03 +00:00
rogerman
d5565bad72
SoftRasterizer:
...
- Do some code cleanup around GFX3D_TXTHack. Also make this check a little more robust.
2015-03-29 01:12:47 +00:00
zeromus
6697c7c0db
fix nand more
2015-03-28 20:26:47 +00:00
zeromus
c325eac5e3
winport - remove glitchy save type advanced menu thingy
2015-03-28 00:16:41 +00:00
zeromus
3289a3aecc
clean formatting of r4800, messing up otherwise clean code
2015-03-28 00:10:53 +00:00
zeromus
6fadb2c07f
improve previous commit
2015-03-28 00:05:29 +00:00
zeromus
b79bd5fd33
fix " #1454 NAND games won't save " by continuing my sloppy fixes from r5052 and r5053 while also removing some of r4880 "optimize read/writes to NAND" since that kind of logic should be handled in the file IO layer even if it isn't now, otherwise the logic is too redundant and complicated spread around the codebase. In the future we need to evaluate using EnablePositionCache() on all the mc.cpp operations
2015-03-27 23:46:53 +00:00
rogerman
0161ced7ff
OGLRender:
...
- When doing the depth buffer calculation, clamp the depth value to GL_DEPTH_RANGE {0.0, 1.0} in the fragment shader itself. Fixes 3D rendering on older drivers that won’t do the clamp for you. (Regression from r5133.)
2015-03-27 21:45:43 +00:00
rogerman
8a9295796d
Documentation:
...
- Add one more note to the Cocoa port changes in ChangeLog.
2015-03-27 20:28:43 +00:00
rogerman
16bfa74d9f
Documentation:
...
- Update release date for README.
- Update Read Me for Macintosh to reflect the current feature set.
2015-03-27 19:59:31 +00:00
rogerman
75bdce0942
OGLRender:
...
- Force cache alignment on the 5-bit LUT.
- Better localize OpenGL-specific LUTs.
2015-03-27 19:52:31 +00:00
zeromus
e9b919d792
update changelog
2015-03-27 05:21:40 +00:00
rogerman
4031ee211a
OGLRender:
...
- Revert depth buffer calculation change in r5133 for the z-buffer mode. Keep the w-buffer mode change, since that’s the one that works. Fixes the buttons in Blazer Drive. (Regression from r5133.)
2015-03-26 11:01:27 +00:00
rogerman
69ec85e624
OGLRender:
...
- Fix depth buffer calculations. Fixes lots of graphical glitches in “Harry Potter and the Order of the Phoenix” and many other 3D rendering cases.
2015-03-26 10:30:44 +00:00
rogerman
9b2d36df0c
GFX3D:
...
- Do workaround for possible NaN comparisons being done when Y-sorting. (Addresses bug #1461.)
2015-03-26 09:49:56 +00:00
rogerman
2fc39e3c3e
Documentation:
...
- Note Cocoa port changes in the ChangeLog.
2015-03-25 06:12:03 +00:00
zeromus
61df0246e6
add comments about y-sorting
2015-03-25 06:04:23 +00:00
rogerman
3a747f85ce
GFX3D:
...
- Fix bug where 4x4 compressed textures in Mode 2 would mix the incorrect colors. (Regression from r5128.)
- Do some minor code cleanup.
2015-03-25 03:48:15 +00:00
rogerman
59d33fddb7
GFX3D:
...
- Fix texture coloring bugs with 4x4 compressed textures on big-endian systems. (This should be the last of the texture coloring bugs.)
- Do small optimization to 4x4 compressed texture conversion.
- Do some minor code cleanup.
2015-03-23 20:58:37 +00:00
rogerman
b253fc44a7
Video Filters:
...
- Do some minor code cleanup on the xBRZ filter.
2015-03-23 17:52:33 +00:00
rogerman
4db72d4254
Cocoa Port:
...
- Fix bug when using dual display mode with a screen separation where the displays could mistakenly draw ghost lines at the top or bottom of each screen. (Partially addresses bug #1435.)
2015-03-23 07:57:18 +00:00
rogerman
f0e8760dc4
Cocoa Port:
...
- Fix compiling when using Xcode 3.
- Don’t set the output frame size with multiple glViewport() calls per frame. Just set it once for the output.
- Do some minor code cleanup.
2015-03-21 21:18:00 +00:00
rogerman
b906b233c5
Cocoa Port:
...
- Fix bug where the user can force activate the mic while the emulator is idle by manipulating its mute control.
- Fix bug where the mic icon remains black if the emulator resets while in execute.
- Further optimize mic icon updates.
2015-03-20 08:48:40 +00:00
rogerman
3088fd581e
Cocoa Port:
...
- Run the Display Preferences preview inside a single-buffered context.
- Do some minor code cleanup.
2015-03-20 07:42:13 +00:00
rogerman
1674170370
mc.cpp:
...
- Handle the case where the EMUFILE size is less than the DeSmuME footer size. Allows an in-memory BackupDevice to work with NAND flash saves.
2015-03-19 22:06:26 +00:00
rogerman
3331fb1890
Windows Port:
...
- Fix crashing bug where the app would crash if the SaveRam path is invalid or does not allow for read/write access. (Fixes bugs #1394 and #1426.)
- New behavior: If the SaveRam path is invalid or does not allow for read/write access, warn the user. After the warning, continue emulation as normal.
mc.cpp:
- Allow backup memory to operate inside RAM if read/write file access is unavailable.
2015-03-19 01:50:15 +00:00
rogerman
bfe105a188
GFX3D:
...
- Workaround a bug specific to GCC v4.9 where allocating the VERTLIST blocks would cause a std::bad_alloc exception. (Fixes bug #1438.)
2015-03-18 03:25:14 +00:00
rogerman
5bab1a14f3
Windows Port:
...
- Fix bug where using the 5xBRZ filter through the DirectDraw HW/SW display method would cause a crash. (Fixes bug #1444.)
2015-03-15 22:40:29 +00:00
rogerman
0ac489ade1
Cocoa Port:
...
- Fall back to using a legacy OpenGL context if the GPU driver rejects initializing a 3.2 Core Profile context.
2015-03-15 06:46:49 +00:00