4e7e7f2dd5
Nocturne, DDS1 and DDS2 CRC hacks to remove motion blur and smudge filter
...
Signed-off-by: i30817 <i30817@gmail.com>
2015-06-09 00:25:33 +02:00
2503d9698c
gsdx-ogl: re enable bilinear fitlering on palette for paltex mode
...
However keep it disabled for StarOcean 3 use cases
2015-06-08 21:01:18 +02:00
95c2622b7f
Gsdx: Typo'd!
2015-06-07 18:31:51 +01:00
b5171dfe5d
Gsdx: De-crapify Native resolution, potential fix for custom resolutions.
2015-06-07 16:57:55 +01:00
0946958acf
Gsdx: Height change on Gregory's recommendation :)
2015-06-07 16:12:14 +01:00
c3c29945b2
gsdx-ogl: only set a cst blend factor when Ad is used
...
Fix GT4 and potentially FFX
2015-06-07 16:10:59 +02:00
5daf7aa98f
gsdx-debug: keep s_n/pop in sync
2015-06-07 12:39:48 +02:00
5bf5b5bca4
gsdx-ogl: extend StretchRect to write in depth texture
...
It will allow to convert color texture to depth texture
2015-06-07 12:39:23 +02:00
0518aaedc9
gsdx-ogl: unattach palette to avoid noise in debug
2015-06-07 12:39:15 +02:00
23f8203a22
gsdx-ogl: add some shaders to convert depth <-> color
...
I'm afraid of rounding error
2015-06-07 12:39:00 +02:00
a5fe8478ae
Merge pull request #575 from PCSX2/gsdx-half-screen-snow-engine-v2
...
gsdx-texture-cache: Improve fix for snow engine game
2015-06-07 10:18:56 +02:00
0a467291f9
Gsdx: Height thing didn't work so well, using old buffer width test again
2015-06-07 03:50:34 +01:00
0e3349f953
Merge pull request #590 from PCSX2/OpenGL-printf-corrections-removed-shadow-hearts-crc
...
OpenGL printf wording corrections, removed Shadow Hearts CRC
2015-06-06 21:56:07 +02:00
1de58349c5
Bleh
2015-06-06 22:39:34 +03:00
8d61ab2af2
OpenGL printf wording corrections, removed Shadow Hearts CRC (not used anymore in GSstate)
...
Some minor grammar fixes to OpenGL prints and removed Shadow Hearts CRC
which has been left behind
2015-06-06 22:28:59 +03:00
1fa13163e4
GSDX: Fix for games which change height but keep the same buffer width
...
- Fix for games that specify a double height, like Pool Paradise
2015-06-06 20:24:10 +01:00
c3dc51826f
Merge pull request #541 from TheLastRar/master
...
DEV9ghzdrk - thread safer calls to DEV9irq
2015-06-06 10:47:32 +02:00
97467f9e30
Merge pull request #579 from PCSX2/drop-vs2010-support
...
Drop vs2010 support
2015-06-06 10:47:16 +02:00
68e44fd354
gsdx: reload the state when GSconfigure is called
...
It fixes the reloading of openGL option and it avoids
issue during pause/resume.
Fix issue #584
2015-06-06 08:51:53 +02:00
7e662c50d4
Revert "gsdx: always restore the full gs state object (s_gs)"
...
This reverts commit 2ea049b5a5
2015-06-06 00:22:15 +02:00
4feeaac7d1
Merge branch 'alignment' of git://github.com/xsacha/pcsx2 into xsacha-alignment
...
Gregory: Get all changes but keep C11 code path which will be the future
(if someone can find info on Visual Studio support)
* Keep the old posix api to use address sanitizer on gcc 4.9
Conflicts:
common/build/Utilities/utilities.vcxproj.filters
common/build/Utilities/utilities_vs2012.vcxproj.filters
common/build/Utilities/utilities_vs2013.vcxproj.filters
common/include/Utilities/MemcpyFast.h
common/include/Utilities/StringHelpers.h
common/src/Utilities/AlignedMalloc.cpp
common/src/Utilities/vssprintf.cpp
plugins/GSdx/stdafx.cpp
2015-06-05 20:31:14 +02:00
e774fecb77
gsdx-ogl: enable again geometry shader on Mesa
...
Please use a recent Mesa version (10.5 if possible)
2015-06-05 18:59:10 +02:00
6c8f7a0147
Merge pull request #580 from PCSX2/issue-565-option
...
gsdx: always restore the full gs state object (s_gs)
2015-06-05 17:51:02 +02:00
7d124e6a83
gsdx-texture-cache: Improve fix for snow engine game
...
Instead to hardcode the half-buffer size to 0x140, the new code
compute it.
To avoid regression, I limit the code to big target (>= 1024 pixels)
2015-06-05 09:06:12 +02:00
65b83a36a7
GSDX: Removed some printf's I left in.
2015-06-05 00:13:21 +01:00
91f775f3f0
GSDX: Fixed up some of the scaling height settings
...
-Last commit broke Snowblind games again
-Width now based on output circuit
-Try to use the smallest possible while being at least the size of the screen
2015-06-05 00:12:15 +01:00
6f4ec98bc3
gsdx: try to reduce the size of target as much as possible
...
I suspect too consume too much memory (issue #576 )
2015-06-04 22:15:52 +02:00
f8f6117ea9
gsdx-debug: allow to track texture move
...
Very painful for the cache and the performance
2015-06-04 22:10:04 +02:00
979ea92754
gsdx: make png++ optional
...
Some distributions (Fedora) doesn't have a png++ package...
2015-06-04 20:22:05 +02:00
bd2c04a0ec
ooups bad copy/past
2015-06-04 19:58:58 +02:00
2ea049b5a5
gsdx: always restore the full gs state object (s_gs)
...
It will force a reload of opengl option
2015-06-04 18:29:31 +02:00
c2e851b3a5
gsdx-ogl: don't use slow SW blend for nothing
...
When A == B, coeff is 0 so <= 1
When C == 2, just check the value of coeff
Barely faster in accurate_blend = 2
2015-06-04 18:18:47 +02:00
ba5b367ae8
drop support of Visual Studio 2010
...
* Severals PR uses C++11 features
* 2015 will soon be released
* Win 10 will soon be released
Time to upgrade
2015-06-04 08:55:22 +02:00
85266abd71
gsdx-ogl: use a 8MB vbo/ibo
...
Dirge of Cerberus draw 143,360 vertices in a single draw!
It is likely to emulate a particle effects
Hopefully improve issue #573
2015-06-04 08:28:40 +02:00
e34477a497
Merge pull request #547 from pgert/master
...
GSdx > HW-hacks > Rewording comments on duplicate CRC's and minor cleanup
2015-06-03 15:58:28 +03:00
0d5d94f56e
Merge pull request #570 from ssakash/patch-3
...
Three new Opengl settings for Windows [GUI]
2015-06-03 14:01:39 +02:00
3dd3bf6e2b
gsdx-ogl: new hidden option accurate_blend = 2
...
Debug option to emulate all blending draw call in the shader
Of course it is slow but it is very accurate
2015-06-03 09:56:07 +02:00
cd6c93cd9e
gsdx: silly mistake
2015-06-03 09:31:52 +02:00
f2d193edcf
Add three new Opengl options.
2015-06-02 16:02:56 +05:30
995ae51bf4
gsdx-ogl: disable a log message
...
Too verbose and code was fixed anyway
2015-06-02 09:28:35 +02:00
9fe13f7a6b
gsdx linux: fix a memory leak
...
Not important but pollute the valgrind log
2015-06-02 09:28:35 +02:00
4914620178
GSDX: Fix for 6x Native, fixed GH3 in native mode taking up quarter of the screen after the snowblind changes
2015-06-01 19:04:39 +01:00
2cbde89084
Merge pull request #555 from PCSX2/real-fb-format
...
GSdx: better framebuffer format
2015-06-01 11:48:07 +02:00
f81cf360bc
Merge pull request #545 from PCSX2/gsdx-half-screen-snow-engine
...
Gsdx half screen (most of) snow engine games
2015-06-01 11:47:40 +02:00
9fa473a57d
gsdx-ogl: glDebugMessageCallback is optional
...
Not even used in release build
2015-05-31 17:55:34 +02:00
9d69ff1f91
gsdx-debug: create the savef option
...
The idea is to easily extract bad frame from gs dump and to compare them
with the sw renderer.
Here the summary of the options:
savet => dump all input textures
save => dump the render target
savez => dump the depth buffer
savef => dump the frame
2015-05-31 17:38:52 +02:00
d301848848
gsdx-ogl: fix paltex on opengl
...
RT uses as palette must use the alpha channel
Palette texture uses the red channel
2015-05-30 19:01:09 +02:00
2f2ae2bc89
gsdx-ogl: enable gl tracing on windows
...
Trace info is enabled in dbg/dev build for the opengl renderer
Info is stored in GSdx_opengl_debug.txt when debug_opengl option is 1
2015-05-30 09:58:47 +02:00
009dd103ae
gsdx-ogl: protect the trace with if statement
...
And mark the function as optional
2015-05-30 09:58:46 +02:00
a639634769
gsdx-ogl: actually code use 4.0 GL_ARB_draw_buffers_blend
...
It is used for glBlendEquationSeparateiARB && gl_BlendFuncSeparateiARB
The 3.0 function could work too. It just need to be added
2015-05-30 09:58:46 +02:00
92d68b70d3
gsdx-ogl: add a performance note for later
...
The idea will be to replace StretchRect for standard case with a framebuffer
blit. Potentially it toggles less gl state.
Worth a test on Star Ocean 3 that uses a lots this function for stencil emulation
2015-05-30 09:58:46 +02:00
01e4a5148d
gsdx-ogl: avoid crash in debug/dev build
...
Don't be silly, don't put gl trace in EE thread.
2015-05-29 16:50:27 +02:00
0cb6c37f14
gsdx-tc: add an explanation of the magic number 0x140
2015-05-29 12:27:05 +02:00
97b237712c
gsdx: detect RT size based on fb size and gs output configuration
...
* It is required for snowbling games
* Targets of Texture cache are dropped.
2015-05-29 12:26:44 +02:00
02274601b3
gsdx-ogl: always copy texture into 0,0
2015-05-29 12:18:54 +02:00
e5326d1bd2
gsdx-tc: allow to reuse the right part of large RT as source
2015-05-29 12:18:54 +02:00
7ac533a4d1
gsdx-tc: invalidate half-RT src
...
snow engines uses the RT as 2 half RT.
2015-05-29 12:18:54 +02:00
da0b77e09b
gsdx-ogl: new logo
...
Thanks bmate for the logo.
Close issue #192
2015-05-29 11:05:52 +02:00
8f54da5fd3
GSDX: Removal of Nvidia hack.
...
- Hack no longer required as an option as a safe and simple workaround has been discovered which seems to work without modifying any vertices.
2015-05-28 21:41:02 +01:00
5d00c00826
GSDX: Fixed UserHacks_AutoSkipDrawDepth dialog creation on windows.
...
- CheckDlgButton was just wrong
- Changed tag to say "Auto-skip depth" as you could only see "Auto-skip draw" which was highly ambiguous on GSDX!
2015-05-28 21:21:19 +01:00
66b08c0b2f
Use Mutex to control writes to the RX_Frame counter
...
A user on http://obsrv.org had an issue with freezing, unfreezing than
dying on other players screen, but can continue playing anyway. This
comment didn't end up fixing the issue of becoming a ghost, but he
didn't freeze.
2015-05-28 20:22:38 +01:00
31ffdd8c53
Merge pull request #559 from ssakash/patch-5
...
Auto depth skip hack initialization.[GUI]
2015-05-28 19:21:45 +02:00
8e32d71c07
Merge branch 'patch-3' of git://github.com/ssakash/pcsx2 into ssakash-patch-3
2015-05-28 18:30:49 +02:00
5f7e09268c
change "D3D Internal Resolution" to "Internal Resolution"
2015-05-28 16:33:16 +05:30
bf743fec17
add logo for Opengl renderer.
2015-05-28 01:34:51 +05:30
c43ddaec4f
gsdx: add Castlevania hack explanation
2015-05-26 17:03:13 +02:00
419dfe0544
glsl: redo color/alpha management correction
...
Please test it!
GS supports 3 formats for the output:
32 bits: normal case
=> no change
24 bits: like 32 bits but without alpha channel
=> mask alpha channel (ie don't write it anymore)
=> Always uses 1.0f as blending coefficient
16 bits: RGB5A1, emulated by a 32 bits openGL texture. I think it will be more correct to use
a real 16 bits GL texture. Unfortunately it would cost several (slow) target conversions.
Anyway as a current solution
=> apply a mask of 0xF8 on color when SW blending is used (improve Castlevania shadow)
unfortunately normal blending mode still uses the full range of colors!
This commit also corrects a couple of blending factor. 128/255 is equivalent to 1.0f in PS2, whereas GPU uses 1.0f. So the blending factor must be 255/128 instead of 2
Note: disable CRC hack and enable accurate_colclip to see Castlevania shadow ^^
(issue #380 ).
Note2: SW renderer is darker on Castlevania. I don't know why maybe linked to the 16 bits format poorly emulated
2015-05-26 16:49:43 +02:00
ef3247c233
Update copyright license
2015-05-26 19:15:30 +05:30
99f9407c24
Auto depth skip hack initialization.
2015-05-26 19:13:56 +05:30
9ee3a173d0
gsdx-ogl: use a local ALPHA register
...
It would allow to easy tune the parameter to support 24 bits format
2015-05-26 15:36:48 +02:00
d31bd97d59
gsdx-ogl: add a variable to select FB output
...
Either 32bits/24bits/16bits
2015-05-26 14:59:07 +02:00
93c0d5a268
gsdx-ogl: add an error message when a RT is reused with a format cast
...
For example a 32 bits RT could be used as a 16 bits RT.
It is likely the source of various bug in HW renderer
2015-05-26 11:16:49 +02:00
a3bbe68d56
gsdx: throw an exception when we miss a mandatory gl feature
...
Gracefully quits when driver is not supported
(like nvidia legacy driver for GTX260)
Please test the impact on windows
2015-05-26 11:03:59 +02:00
3be5a6036b
gsdx-ogl: remove assertion + debug message
...
opengl use a different object to compute the vertex count
2015-05-26 11:03:59 +02:00
b84a462798
Unreachable code message was annoying me.
2015-05-26 00:30:12 +01:00
2704a65f57
gsdx: add some info for Star Ocean3/Valkyrie Profile 2/Radiata Stories
...
Feel free to test those games with CRC disabled on the openGL renderer (native resolution)
2015-05-25 12:37:48 +02:00
580d177951
gsdx: improve commit 11708486d8
...
This time linear filtering is disabled only for the bad draw call
(RT used as a palette texture).
2015-05-25 09:46:51 +02:00
99d7434671
gsdx-ogl: disable SSO on nouveau
2015-05-24 19:24:10 +02:00
11708486d8
gsdx: allow to control the filtering of texture cache
...
When the RT is used as an input texture, we need to rescale it.
Previous behavior was to always uses a linear filtering (more smooth).
Unfortunately it broke some games that expected an exact value like Star Ocean 3
This commit will disable the linear filtering in normal filtering mode (filter = 0
or filter = 2)
This way, shadow of Star Ocean 3 will appear correctly in upscaling (not
100% perfect but can't do better)
Note: SO3 only requires a nearest sampling of the alpha channel but
I don't know the behavior for others games.
2015-05-24 18:59:27 +02:00
b0af54d33e
gsdx-ogl: better support of palette
...
The purpose of the code is to support alpha channel
of RT uses as an index for a palette texure.
I'm afraid that code will likely break pure palette texture. Only used
if paltex is enabled
It fixes missing shadow in Star Ocean 3 (issue #374 ) in Native resolution
with filter = 0 (no filtering) or = 2 (normal fitering)
Rendering explanation:
The game emulates a stencil buffer with the alpha channel
The alpha channel of the RT can contains a palette texture index (format 4HH)
The idea is to have a gradient of value in the palette (16/32/48/...).
This way you can implement a +16/-16 and even wrap the alpha value every time
you hit the pixel.
Bilinear filtering breaks the rendering because it interpolates between counts
so you doesn't have the exact count
Upscaling breaks the rendering because the RT is reused as an input texture. It means
that we need to scale it down which again create some interpolations.
2015-05-24 18:07:16 +02:00
951a1fed94
gsdx-hack: enable auto skip of depth texture for God of War
...
Rendering is quite complex to emulate on the GPU due to several limitation.
Until we find a solution let's enable the old solution
You could find a description of the rendering here:
http://forums.pcsx2.net/Thread-God-of-War-Red-line-rendering-explained
2015-05-24 13:43:36 +02:00
a70c3bf5de
glsl: correct the alternate implementation of ps_main1
...
Still not yet enabled by default
Potentially it can be optimized with the dot product but special care
need to be taken to ensure float accuracy.
Bonus: it could work on old GPU (aka DX9)
2015-05-24 13:43:36 +02:00
c82fd94c6c
gsdx-debug: use GL_INS instead of printf
2015-05-24 13:43:36 +02:00
c01686d252
gsdx-png: add a fixme message for 16 bits by channel image
2015-05-24 12:51:20 +02:00
7f614401a6
gsdx-ogl: set accurate_blend as default
...
Rendering is better and I think speed impact remains small.
2015-05-23 15:10:04 +02:00
c507e816c2
gsdx-ogl: oups forget a break
2015-05-23 13:06:23 +02:00
b884e0c0c0
gsdx-debug: improve debug experience
...
* Dump context before the increase of s_n
=> aligned with the global call number
* Don't print colclip not supported when it is optimized away
* dettach the input texture when it is useless
=> avoid to show a wrong texture in the debugger
2015-05-23 12:23:05 +02:00
f665180d11
gsdx-debug: use hex number for texture pointer
...
Easier to grep
2015-05-23 12:20:26 +02:00
7609fdc576
gsdx: rename too much sr to sRect
2015-05-21 09:48:15 +02:00
b52ee6f1f6
gsdx-linux-gui: add an option for accurate color clipping
...
Note: it is only for debug purpose
Sotc testcase 187 fps => 125 fps (2x scaling)
2015-05-20 09:42:22 +02:00
183af4ece6
gsdx-ogl: don't enable SW blending when there is no blending
...
Mostly to avoid useless message
2015-05-20 09:36:01 +02:00
3d35e4f8ec
gsdx-ogl: implement all blends mode in shader
...
+ 250 lines !!!
2015-05-20 09:11:36 +02:00
2d54d59add
gsdx-ogl: update assertion ps sel is 8B now
2015-05-20 08:40:08 +02:00
d3d5a436ea
gsdx-ogl: add code to read back depth texture
2015-05-20 08:07:40 +02:00
a4c74ef872
gsdx-ogl: disable GL_DITHER
...
I really don't know the impact but it is not supported with integer texture.
2015-05-20 08:05:27 +02:00
8d3e3e6c5b
gsdx-ogl: more blend rework to support accurate_colclip
...
So far few blending equations are implemented in PS. It is only
for test the behavior on GoW
2015-05-20 08:00:40 +02:00
c5341a2711
gsdx-ogl: update blending management
...
This way it will allow to implement all blendings operartion in FS.
Of course it will be slow, but it would be nice for debug and quickly check
game error rendering.
2015-05-20 00:12:52 +02:00
358e0d445b
gsdx-ogl: add a shader to convert depth texture into uint
2015-05-19 18:56:32 +02:00
ac38ebbf29
Merge pull request #546 from ssakash/patch-3
...
enables Anisotropic filtering option for OGL (HW mode) [Windows UI]
2015-05-19 15:20:12 +02:00
ec45090358
glsl: implement an alternative of convert 1 shader
...
Code not yet enabled. Quick test show no heavy regression but I need
to carefully validate it.
2015-05-19 13:14:18 +02:00
a12acc0d24
gsdx-ogl: clean a bit Texture Cache OGL
...
Add a placeholder for depth format but nothing done yet.
2015-05-19 13:14:18 +02:00
1837001e75
gsdx: extend CopyOffscreen with a new shader parameter
...
Currently we're trying to infer the conversion shader based on the output format
It only works if the input data is RGBA8. It might not be true in the future
2015-05-19 13:14:18 +02:00
9554b5dd56
gsdx-lzma: drop CONCAT flag
...
I have some issue with it. Unlikely to use multi-stream gs anyway
2015-05-19 13:14:18 +02:00
fa761a129d
gsdx-linux: add an option to enable/disable recording
...
Avoid to kill the computer/application if the user press F12 by accident
2015-05-19 13:14:18 +02:00
34202933da
GSdx-linux-recorder: add various option to control it
2015-05-18 19:57:21 +02:00
388f46b577
gsdx-png: use array interface instead of set_pixel
...
Doc says it is "non-checking equivalent of set_pixel". Still slow as ass
At least syntax is nicer
2015-05-18 17:25:14 +02:00
a4d6722b26
gsdx-thread: add a CAPACITY parameter to GSJobQueue
2015-05-18 17:08:18 +02:00
e7665ee7dd
gsdx-linux-recorder: upgrade code to use n threads
...
Unfortunately it requires too much memory (easily 6GB)
2015-05-18 16:56:44 +02:00
cff168e002
gsdx-linux-recorder: implement multithread to save the image
...
Currently only a dedicated thread. It is still too slow!
2015-05-18 16:56:44 +02:00
012f1b4f05
gsdx-thread: remove deprecated boost code
2015-05-18 16:55:49 +02:00
19d9349b0b
gsdx-debug: remove old assert
2015-05-18 16:45:38 +02:00
c9508298b4
Update GSCrc.cpp
2015-05-18 16:37:28 +02:00
70b8772bb4
enables Anisotropic filtering option for Opengl (Hardware mode) [Windows UI]
2015-05-18 19:16:22 +05:30
503459798a
gsdx-linux: add a basic implementation to record the gameplay as png file
...
The idea was to merge them later with ffmpeg or other external tool
It kinds of work but current fps is around 2 !!! So not really usable
The speed issue is related to PNG.
So either the code to store pixels data must be optimized. Or maybe it misses some MT loves :)
2015-05-18 11:56:20 +02:00
2783da4a22
gsdx-ogl: use a local buffer to store offscreen texture
...
It will allow to read texture in // (and potentially could be useful
for recording)
2015-05-18 11:29:04 +02:00
79a9254894
gsdx-ogl: print some error messages if extenal shader is wrongly configured
2015-05-18 10:56:32 +02:00
540717b295
gsdx-ogl: don't force logz only on debug build
2015-05-18 07:24:26 +02:00
5c046180a1
cmake: properly add detection of lzma
...
Code is 100% optional.
Code will allow be enabled if lzma-dev is installed at compile time
2015-05-17 22:08:10 +02:00
98dde32dc0
gsdx-replayer-linux: use new code to allow loading of .gs.xz dump
...
A bit slower at startup but it saves a lots of disk space!
Reduce the size of my gs collection of 11GB :)
2015-05-17 22:08:10 +02:00
e93f780268
gsdx-replayer-linux: separate a bit parsing and executing code
...
It would ease the porting of the code to read xz file
2015-05-17 22:08:10 +02:00
86e3c18c33
gsdx-linux: add code to read xz file
...
Linux only to avoid the extra dependency on windows
2015-05-17 22:08:10 +02:00
5cfb496700
gsdx-ogl: only open debug file once
2015-05-17 14:43:56 +02:00
7979dec5b0
gsdx-ogl: optimize colclip 0
...
Currently colclip uses 2 passes to wrap the output of blending unit
However some blending mode are only a plain copy (of 0 or Cs or Cd).
So no overflow of [0:255], no need to wrap it
Note: I saw those cases in GoW.
2015-05-17 13:05:08 +02:00
6ced837360
gsdx-debug: add some PERF info in trace
...
The idea is to detect easily effect that are known to be slow on opengl
2015-05-17 13:05:08 +02:00
818412ba26
gsdx-debug: fix a crash in trace
2015-05-17 13:05:08 +02:00
b1ea081fc3
gsdx-debug: improve tracing interface
...
Basically move the format and c_str() in the macro
2015-05-17 13:05:08 +02:00
b8aa5f8f80
gsdx-debug: more trace for the texture cache
2015-05-17 13:05:08 +02:00
8a73849531
gsdx-ogl: enable multithread driver by default for nvidia
...
+ add a linux gui option to disable it (for test purpose)
2015-05-16 15:22:20 +02:00
c567198967
gsdx-ogl: replaced draw_count by s_n
...
This way error message is aligned with everything else :)
It is not perfect (normally it must be done in start of main draw)
2015-05-16 13:59:13 +02:00
62e0e6a067
gsdx: remove deprecated code
...
Core was fixed to call GSFifo in the good thread
2015-05-16 13:52:16 +02:00
2165c68938
oups forget a couple of break
2015-05-16 13:26:55 +02:00
02b478dfbc
gsdx: plug the new PNG wrapper
...
Drop various duplicated code :)
2015-05-16 12:47:28 +02:00
8cd533304c
gsdx: new dependency on linux => png++
...
Note: it is only header file that wrap the standard png (which was
already mandatory for wx/sdl)
2015-05-16 12:47:28 +02:00
cab59587e8
gsdx: add a PNG wrapper
...
Normally it supports all formats but I didn't have luck with 16 bits depth gray
2015-05-16 12:32:05 +02:00
3f278382a1
gsdx-debug: dump all drawing register
...
Thanks PERL
2015-05-16 11:16:33 +02:00
8341055f3e
gsdx-ogl: don't enable ogl_texture_storage on catalyst
2015-05-16 00:31:25 +02:00
a2d605be0b
gsdx: copy past mistake
...
Let's always increment the s_n counter
2015-05-16 00:31:25 +02:00
cfddcb7a93
gsdx-ogl: typo
2015-05-16 00:31:25 +02:00
0f01ba4c46
gsdx-ogl: mega boost
...
Enable Nvidia multi thread driver optimization
Enable ogl_texture_storage by default (requires for the speed boost
, later the option will be removed)
2015-05-16 00:31:25 +02:00
a1bfb237e6
Partially Revert "gsdx-ogl: drop support of GL_ARB_clear_texture extension"
...
This reverts commit 03e72781aa
2015-05-16 00:31:25 +02:00
28bb64aae8
gsdx: sed/dr/dRect/
2015-05-15 20:49:32 +02:00
445c28fe97
gsdx: sed/sr/sRect/
2015-05-15 20:47:14 +02:00
d566bb2a23
gsdx: sed /st/sTex/
2015-05-15 20:45:31 +02:00
6a9e425308
gsdx: sed /dt/dTex/
2015-05-15 20:44:15 +02:00
d870188d21
gsdx: sed/o/off/
2015-05-15 20:40:09 +02:00
27d485a940
drop a file that was wrongly commited
2015-05-15 20:34:09 +02:00
6166c95325
gsdx-ogl: protect PBO with fence
...
Safer and doesn't impact perf too much.
2015-05-15 18:32:47 +02:00
a5e424512c
gsdx-ogl: really avoid consecutive clean
2015-05-15 16:00:46 +02:00
84c3592fbe
gsdx-debug: more debug message/group
2015-05-15 16:00:45 +02:00
08291aed0c
gsdx-ogl: color state impact the clean command
2015-05-15 15:25:45 +02:00
613e215c73
gsdx-ogl: add some note for the persistent buffer + add a flush
...
Persistent is slower (at least on my gs dump) because data is put
in host instead of the video memory
I don't understand why upload the data directly to the video memory
is faster
2015-05-15 15:25:45 +02:00
5628bfb20c
gsdx-ogl: drop old code
...
I have group so it doesn't pollute anymore gl trace
2015-05-15 15:25:45 +02:00
ff4cf51494
use volatile
2015-05-14 13:23:09 +01:00
6d8695127f
gsdx-recorder: use recursive mutex
2015-05-14 11:56:07 +02:00
a0ec633c46
Provide a DEV9async and use it to call irq
2015-05-13 10:04:04 +01:00
bd96a73580
gsdx-debug: trace texture cache hit/miss/removal
...
Actually only a small part but I think it is enough to help debugging
tc issue.
Only enable it on debug because it is quite verbose
2015-05-13 09:01:04 +02:00
5e9a3a424e
Merge pull request #479 from PCSX2/gsdx-texture-cache
...
gsdx: value was wrongly overwritten
2015-05-13 08:08:46 +02:00
b7a9465963
gsdx-ogl: update the device to use the new texture flags
...
Hopefully it will increase a bit the speed
2015-05-12 18:18:20 +02:00
3e784d57e8
gsdx-ogl: add some flags to trace texture state
...
goal1: avoid 2 consecutives clean of the render target
goal2: only invalidate texture correctly
2015-05-12 18:03:06 +02:00
2e34d48e97
gsdx-ogl: add a virtual GetID method for texture
...
Much more readable
2015-05-12 17:41:41 +02:00
e0012811ae
gsdx-debug: more debug info in gl trace
2015-05-12 12:36:34 +02:00
4e222f18cd
gsdx-ogl: it was a bad idea to use DSA on fb
...
Actually I'm not sure we can mix both dsa/standard approach
2015-05-11 18:04:16 +02:00
f37f3cb3cf
gsdx-ogl: improve texture uploading
...
Initially we copy pitch by line in the PBO and tell the dma to only
use the first valid byte.
Now, we only copy useful data to the PBO. It reduce the copy and PBO memory requirement.
It seems a bit faster on native resolution
2015-05-11 16:32:13 +02:00
4e2e9aa56c
gsdx-ogl: always read the first attachment of the fbo
2015-05-11 16:28:34 +02:00
cb6e8a7d6a
gsdx-linux: Don't enable EGL by default
...
I was hoping that EGL become a standard much more faster. Currently it is an useless dep
so let's disable it by default.
2015-05-11 15:33:40 +02:00
643b1647c4
gsdx-gui-linux: add some tooltips for the hw options
2015-05-11 14:12:41 +02:00
921fa3bab8
gsdx-ogl/linux: drop fba option
...
It is useless as DX11
2015-05-11 13:57:26 +02:00
f0de7d6386
gsdx-debug: add 2 extra frames in GS dump
...
Initially I was afraid of the size increase but it can be
higly compressed.
Sometimes image was computed inside a RT but never outputed to
the screen which remains black. It is annoying to see the glitches
2015-05-11 13:45:24 +02:00
b3d8981a22
gsdx: link replayer to png to avoid link failure
2015-05-11 13:24:40 +02:00
eff4ac3e4a
gsdx-ogl: update UBO hint
...
On Nvidia, DYNAMIC is a bit faster.
2015-05-11 12:35:41 +02:00
59f7e4d2dc
Revert "gsdx linux: drop aligned stack option"
...
This reverts commit 2dcffe7fa4
2015-05-11 12:15:38 +02:00
2dcffe7fa4
gsdx linux: drop aligned stack option
...
I think the bug was related to a wrong declaration of exported function without stdcall
2015-05-11 11:35:34 +02:00
286fe4db8c
cmake/gsdx: check png++ is installed to support png image
...
Otherwise fallback to bmp file
2015-05-11 11:32:13 +02:00
30e3956ea7
gsdx-ogl: avoid issue with apitrace
...
The tool doesn't support NaN value so be sure the buffer is correctly initialized
2015-05-11 11:19:18 +02:00
1523b9534f
gsdx-debug: compact the code
2015-05-11 11:19:00 +02:00
51ed3e63e9
glsl: oups forget to remove a line
2015-05-10 10:39:23 +02:00
5565544ba6
gsdx-ogl: DATE with texture barrier
...
Much faster for small batch that write the alpha value. Code can
be enabled with accurate_date option.
Here a summary of all DATE possibilities:
1/ no overlap of primitive
=> texture barrier (pro no setup of stencil and single draw)
2/ alpha written
=> small batch => texture barrier (primitive by primitive). Done in N-primitive draw calls.
(based on GL_ARB_texture_barrier)
=> bigger batch => compute the first good primitive, slow but only 2 draw calls.
(based on GL_ARB_shader_image_load_store)
=> Otherwise there is the UserHacks_AlphaStencil but it is a hack!
3/ alpha written
=> full setup of stencil ( 2 draw calls)
2015-05-09 19:54:01 +02:00
390811f414
gsdx-debug: Dump in PNG on linux & dbg/dev build
...
Keep BMP for standard user of release build
2015-05-09 18:41:17 +02:00
3c7a3a3d83
gsdx-ogl: disable accurate option if driver doesn't support it
2015-05-09 18:41:17 +02:00
7fac25edb0
glsl: add shader implementation of impossible blend
...
Issue: I don't know if destination alpha value must be multiplied by 2.0f
2015-05-09 15:02:34 +02:00
f029e4763f
gsdx-ogl: add the alpha value in a constant buffer
2015-05-09 15:02:34 +02:00
472608b879
gsdx-ogl: add accurate blending implementation
2015-05-09 15:02:13 +02:00
1e8aea033c
gsdx-ogl: add a wrapper to control the drawing of primitives
...
No barrier => draw all primitives
Barrier but without overlap => draw all primitives
Barrier with overlap => draw primitive by primitve
It will ease the implementation of accurate blending and why not date too
2015-05-09 15:01:39 +02:00
625d65d4b4
gsdx-ogl: encode the bogus id as shader parameter
2015-05-09 14:55:44 +02:00
bc5de8df6b
gsdx-ogl: disable FF blend when blend is done in shader
2015-05-08 20:28:50 +02:00
380e420cdd
gsdx-ogl: add a blend parameter to shader
2015-05-08 20:28:50 +02:00
9023f5f0db
glsl: drop useless ifndef
...
I often mix them with the real value...
.
2015-05-08 19:28:17 +02:00
8e1db43431
gsdx-debug: debug stuff
2015-05-08 19:28:17 +02:00
3c66da4d82
gsdx: remove old code
2015-05-08 19:28:16 +02:00
1addae1993
gsdx-ogl: don't use extra shader for sprite hack
...
Atst == 2 && sprite_hack is equivalent to Atst == 1
It frees a bit in the shader selector, and reduces shader combinations.
2015-05-08 19:28:16 +02:00
b490085214
gsdx-ogl: add 2 options in the GUI to ease testing
...
accurate_date => use an emulated stencil to compute destination alpha
test (could be slow)
accurate_blend => do nothing (future)
2015-05-08 19:28:16 +02:00
4c91bab00e
gsdx: fix compilation error on release build
2015-05-08 19:28:00 +02:00
6d31124d3a
gsdx-debug: only enabled advance gl debugging on linux
...
It will avoid some crash on windows for driver that doesn't support
the extension.
Make debug support mandatory for debug build on linux.
2015-05-08 09:54:34 +02:00
e87d129b09
gsdx-ogl: more verbose debug
2015-05-08 01:01:01 +02:00
d6448183d7
gsdx-ogl-debug: insert error message in GL stream
...
This way, you can see your message in the GL debugger
2015-05-08 00:16:31 +02:00
544a923e32
gsdx: Add an option to control auto skip of depth texture
...
Default is 1 if the user enable hack.
It fixes GTA sun effect and likely others bugs.
Please someone add a windows GUI option.
2015-05-07 23:54:22 +02:00
51a67029cf
gsdx-ogl: add an option to print gl error messages
2015-05-07 23:54:22 +02:00
7518b2ef21
gsdx-gui-linux: add debug option in the gui
...
* Only for debug/dev build
* look awful (expand/fill) but otherwise it is nice for the debug ;)
2015-05-07 23:54:22 +02:00
d20501739f
Merge pull request #528 from ssakash/patch-1
...
sync with GSdx.ini variable.(fix earlier typo)
2015-05-07 22:15:41 +02:00
fda746ff93
gsdx-gui-linux: support 3 states hack options
...
Based on a Combo Box on linux. As good as 3 states check buttons
2015-05-07 22:13:49 +02:00
d69ffa9b8b
gsdx-gui-linux: move remaining table filling into small functions
2015-05-07 22:13:49 +02:00
b39b9c9f55
gsdx-gui-linux: use callback everywhere
...
Note: texture offset doesn't work nicely (you need to press enter
to validate the change)
2015-05-07 22:13:49 +02:00
d428e0d19d
gsdx-gui-linux: move table filling into separate function
...
It is much easier to read this way. I need to use callback for the remaining
object before I can move everything out of main function.
2015-05-07 22:13:49 +02:00
917a6c28e5
gsdx-gui-linux: factorize code
...
* Use wrapper and callback for Check Button/Spin Button/Combo Box
* Use TheApp.m_gs vector to create combo box
Remove nearly 100 lines of code ^^
2015-05-07 22:13:49 +02:00
ba21879059
gsdx-ogl: 1x aniso <=> off
2015-05-07 22:13:49 +02:00
e6560f348e
gsdx: add vector option for linux & openGL
2015-05-07 22:13:49 +02:00
8f1d00292b
gsdx-ogl: allow to control vsync on Windows
...
Issue #529
Completely untested
2015-05-07 18:41:10 +02:00
6095f40baf
gsdx-ogl: add the number of free bit in selector structs
2015-05-07 18:41:10 +02:00
9a083d5506
gsdx: crc hacks: few new crcs, few crcs region changes
...
Signed-off-by: Avi Halachmi (:avih) <avihpit@yahoo.com>
2015-05-07 15:57:50 +03:00
97fad356b8
gsdx-ogl: fix debug build for AMD/Intel HW
2015-05-06 19:09:13 +02:00
cc4713d379
gsdx-debug: extend ogl debug capabilities
...
Group opengl calls into a nice name.
Apitrace shows them in a tree format that support folding. Previously it
was a long flat list (10K-40K of lines by frame)
I align the call number with the internal s_n variable. This way it is
easy to map GSdx dump output with the GL debugger :)
2015-05-06 19:09:13 +02:00
530e4ce776
gsdx-ogl: drop hack that rescale primitive (to avoid upscale glitch)
...
Rendering is bad. It renders sprites at native resolution.
2015-05-06 19:09:13 +02:00
6d65867b26
gsdx-ogl: comment point_sampler
...
It is not enabled on the shader so I will reuse the bit
2015-05-06 19:09:12 +02:00
3fcac07120
gsdx-ogl: print some message when blending is not properly supported
2015-05-06 19:09:12 +02:00
6fca3b421d
gsdx-debug: dump register context in a .txt file
...
+ Realign s_n number if frame is skipped
This way it is easier to understand current draw. It also eases
comparaison between SW/HW.
2015-05-06 19:09:12 +02:00
24e37d2dea
sync with GSdx.ini variable.(fix earlier typo)
...
fixed the description for round offset hack.
fixed the new description for round offset hack.
2015-05-05 23:01:46 +05:30
5d4bc5c8fe
Merge pull request #516 from PCSX2-underground/master
...
round offset UI initialization
2015-05-05 17:48:53 +02:00
5f5b901bca
gsdx-ogl: alpha bending equation/function are constant
...
Drop useless variables/state checking
2015-05-05 11:20:25 +02:00
8032e2c369
gsdx-ogl: separate color mask state from the blending state
...
Unlike DX they're uncorrelated.
2015-05-05 10:26:01 +02:00
302ce51efe
round offset UI explanation
...
initialization of IDC_ROUND_SPRITE
set&get config for round offset.
three state checkbox for the hack.
define IDC for round offset.
Linux (GUI) refresh.
typo
2015-05-05 00:29:00 +05:30
7bfee7e377
gsdx-debug: stop the dump after a limited number of draw call
...
Otherwise it fills your disk in a short time with useless data.
2015-05-03 16:46:48 +02:00
c1ddad2d25
gsdx-linux: Refresh the GUI
...
* Use table everywhere (better alignment)
* Factorize Box/Frame creation
* Factorize table insertion
* Add Anisotropic Filtering support
2015-05-03 16:36:48 +02:00
ae70344fbc
gsdx-ogl: fix 2 passes DATE with GL42 (override_GL_ARB_shader_image_load_store = 1)
...
It fixes shadows issue with Persona 3
2015-05-02 16:56:18 +02:00
14a1925de0
gsdx ogl: date texture is signed to use i variant
2015-05-02 16:53:34 +02:00
f37ef105c5
gsdx-ogl: add support for anisotropy
...
Close feature request #447
2015-05-02 10:54:58 +02:00
b1ec8a5a42
gsdx-ogl: detect support of GL_EXT_texture_filter_anisotropic
2015-05-02 10:53:19 +02:00
73d04e33e9
gsdx ogl: clean various comment and old code
2015-05-01 20:04:23 +02:00
335695bd0e
purge GLES from GSdx !
...
mobile will use vulkan (or any new API) anyway
2015-05-01 20:02:17 +02:00
e17ba1c143
gsdx ogl: disable logz when ARB_clip_control is supported
...
Basically any recent drivers but Catalyst...
2015-05-01 15:01:08 +02:00
2ff3d17f67
gsdx-ogl: don't create override option for mandatory extension
...
Ini is alredy big enough
2015-05-01 14:18:05 +02:00
a0c17ae0a4
gsdx: use vertex storage by default
...
It is faster on games that draw lots of primitives and in low upscaling.
2015-05-01 14:05:45 +02:00
004fa7aea4
gsdx debug: allow to dump alpha channel as a gray texture
...
I would love to find an image viewer that allow to mask channel of the image
2015-05-01 13:38:58 +02:00
c8a3db114c
gsdx debug: create savet option
...
0: don't dump input texture
1: dump input texture
Now, you can do a first pass with only RT. When you find the wrong call, you can redump the input texture of the bad draw.
2015-05-01 13:35:21 +02:00
f96a653d05
gsdx-ogl: GSVertexArrayOGL uses ASSERT with condition
...
Remove useless indentation too
2015-05-01 12:16:26 +02:00
71c26a829e
gsdx-ogl: forget to remove this useless call
...
It is done once in Device Creation instead of per stencil object
2015-05-01 12:10:03 +02:00
c76e66f8d2
gsdx-ogl: fix read back of render target
...
Initial code use a PBO to do asynchronous transfer. It is silly because
GSdx doesn't use this free time. So let's use a sync read. Same speed but
no PBO to manage.
2015-05-01 01:26:44 +02:00
de52ce956a
gsdx linux: update gui
...
Clear texture was dropped but texture_barrier was added
2015-05-01 01:00:11 +02:00
7367b22e03
gsdx-ogl: reduce toggling of scissor state for DATE
2015-05-01 00:59:49 +02:00
71e517108d
gsdx-ogl: fix colclip
...
Sotc shadow and likely various games :)
Thanks to pseudo
2015-04-30 23:08:05 +02:00
25997647f2
gsdx-ogl: add ENABLE_OGL_PNG_OPAQUE to dump texture without alpha
...
Alpha is nice but fully transparent texture suck
The best will be an image viewer that can toggle the alpha channel
2015-04-30 23:06:54 +02:00
39a5d4c839
gsdx: add PNG support for SW renderer
...
Easier to compare
2015-04-30 20:13:24 +02:00
b65a62096f
gsdx-ogl: drop support of ENABLE_OGL_DEBUG
...
Stencil can be read by GL debugger due to correct mask configuration
2015-04-30 20:02:51 +02:00
8a52fdab57
gsdx-ogl: allow to dump texture as png file
...
-- slower (but that a debug feature)
++ smaller (40x-50x)
++ native support of alpha
Require libpng++ and the define ENABLE_OGL_PNG
Note: depth is not supported yet.
2015-04-30 20:02:50 +02:00
ee19a2789c
gsdx: move invalidation from GSDevice to GSTexture
...
Much cleaner this way
2015-04-30 19:55:57 +02:00
2bd9043657
gsdx-ogl: improve debug of stencil
...
Note: ENABLE_OGL_STENCIL_DEBUG could be dropped now that I can read the stencil properly
2015-04-30 19:55:57 +02:00
7887d7b5a5
gsdx-ogl: use less verbose code
2015-04-30 19:55:41 +02:00
f0181d98fd
gsdx-ogl: save the texture state
2015-04-30 09:57:30 +02:00
0ab0c6cfba
gsdx: verbose debug option
...
Print opengl error message on stderr
Rename Debug.txt into GSdx_opengl_debug.txt
2015-04-27 19:30:03 +02:00
eaa0dae305
glsl: drop invariant qualifier
...
Fix issue #510 and potentially it allows the compiler to better optimize the code
2015-04-27 18:52:55 +02:00
46ff4dc3d3
gsdx-ogl: hardware unit only support normalization of 4 bytes...
...
(At least on recent AMD GPU)
2015-04-27 18:51:59 +02:00
301fe869c7
glsl: improve the logz handling
...
This way only 0 and 1 will conflict
2015-04-26 17:51:33 +02:00
768ae342c3
gsdx: move buffer check after allocation
2015-04-26 17:51:06 +02:00
346ef1eec2
gsdx-ogl: ouch! Properly check the creation of vertex buffers
2015-04-26 11:01:55 +02:00
6a3250cad2
gsdx-ogl: fix logz management (FFXII)
...
The issue is that z+1 overflow if z is UINT max
log2(z) overflow if z is 0
So let's use an or 1 to avoid both overflow
2015-04-25 20:23:26 +02:00
ee244071fa
gsdx-ogl: use 64 bits counter + fix division factor
...
I also added a counter of the real size of the texture.
I have a bad overhead for pbo transfer
2015-04-25 14:18:21 +02:00
00e62919c5
gsdx-ogl: use countof macro instead to hardcode the size
2015-04-25 13:06:02 +02:00
47a0026b60
gsdx-ogl: print the bandwidth of uniform
2015-04-25 13:00:03 +02:00
f2f59544d2
gsdx-ogl: early bind vertex array object
...
Creation of buffer will generate a bind in BufferStorage mode. It is just to be safe.
2015-04-25 12:50:12 +02:00
15ae9996bb
glsl: format white space + comment
2015-04-25 12:50:12 +02:00
757726bb91
gsdx-ogl: allow to invalidate the texture
...
It just a hint to the driver to avoid any useless transfer
I don't expect any change but it is free so why not ;)
2015-04-25 12:50:12 +02:00
75817bb27b
gsdx-ogl: add a quick and dirty DSA layer emulation
...
The global idea is to use
1/ bind in tight loop
2/ DSA otherwise (to avoid any binding in tight loop)
2015-04-25 12:50:11 +02:00
eb257d9295
gsdx-ogl: add dsa function place holder
2015-04-25 12:50:11 +02:00
baf84b98c4
gsdx-ogl: add override_GL_ARB_texture_barrier option
...
To ease regression test.
2015-04-25 09:59:25 +02:00
b12eb45bb7
gsdx-ogl: try to avoid crash on fglrx windows
2015-04-25 09:50:19 +02:00
f0182f9a66
windows requires APIENTRY (=> stdcall)
2015-04-25 01:32:09 +02:00
7b0775d887
gsdx-ogl: add some fences to protect the upload of vbo buffer
...
This way ogl_vertex_storage must be safer to activate
And it brings a nice performance boost (game with lots of primitives and
reasonable upscaling)
SotC testcase 4x: 61fps => 78fps
2015-04-24 23:15:19 +02:00
36514bd95f
glsl: fog is a single byte
...
Give a chance to the driver to optimize if possible
2015-04-24 21:37:37 +02:00
c207632e49
gsdx-ogl: improve date performance for GL45
...
If there is no overlap, it is allowed to directly read from the render target.
On SotC testcase with 6x scaling: 30fps -> 40fps
Note: it requires GL_ARB_texture_barrier extension so be sure to have a recent driver
Note2: it requires a lots of testing too
Open question: in case of complex date (written alpha)
Will it be faster to split the draw call into multiple call with no
primitive overlap
2015-04-24 21:12:33 +02:00
795ae50ecd
gsdx-ogl: fix the recently broken advance date feature
...
Now it is really working with a 2 stages shaders but it is still slow.
2015-04-24 20:13:38 +02:00
672e3f9533
gsdx-ogl: use DSA for texture management
...
Yeah code is much nicer :)
2015-04-24 19:34:17 +02:00
6d31d1e0d0
gsdx-ogl: add a layer to emulate DSA for texture
...
Framebuffer function will be nice too
2015-04-24 19:32:00 +02:00
f71eb171cf
gsdx-ogl: add glTextureBarrier function pointer
...
Could be useful
2015-04-24 18:35:01 +02:00
6e386df535
gsdx-ogl: avoid to clean fully texture in DATE
...
Is is useless and it has a small impact on performance for big upscale
2015-04-24 18:32:08 +02:00
03e72781aa
gsdx-ogl: drop support of GL_ARB_clear_texture extension
...
Extension is a bit slower.
We use it to clear the RT but we generally use it right away so
we don't avoid the FB attachment.
2015-04-24 18:15:58 +02:00
89d5e5637c
glsl: use an explicit cast insead of notEqual function
...
If the compiler didn't optimize the code, it will be a bit faster
2015-04-24 18:01:25 +02:00
56836561f4
glsl: replace runtime condition by preprocessor condition
...
It might be an easier work for the compiler
I didn't replace all occurences to keep readability
2015-04-24 17:51:29 +02:00
4bb8d15228
gsdx: be more verbose on title and bandwidth debug
2015-04-24 17:13:56 +02:00
258b73409c
gsdx-ogl: update flags for buffer storage
...
Fix issue with Mesa driver
2015-04-23 21:10:43 +02:00
f6652e9a50
gsdx-ogl: disable slow and buggy code until I found a better solution
...
ogl_texture_storage 1 creates texture corruption.
Advance date is too slow, code need to be updated (properly) to uses 2 passes only not 3
Maybe one could be enough (sometimes)
2015-04-22 09:33:41 +02:00
b32f808fd4
gsdx-ogl: increase the number of pbo
...
It would cost 16MB of extra storage on the GPU but it might
reduce conflict of texture upload.
2015-04-22 00:40:38 +02:00
bd6ea17bdc
gsdx-ogl: speed improvement for DATE
...
DATE is implemented in 2 ways.
1/ with stencil
2/ purely in FS (sw)
I kept method 1 to reduce the work on method 2. It sucks for performance.
So it would be either 1 or 2.
Note: DATE has a big impact on higher upscaling
Note2: you can disable the 2nd method with this configuration parameter
override_GL_ARB_shader_image_load_store = 0
2015-04-22 00:36:34 +02:00
15dcf07b3b
revert previous commit
...
Not better, worst slower.
I'm afraid I will need proper fencing
2015-04-22 00:32:46 +02:00
8386b427ea
gsdx ogl: restore GL_MAP_COHERENT_BIT for texture upload
...
I hope to fix the texture upload corruption
I hope that impact on perf will remain small
2015-04-21 23:34:26 +02:00
19eb1f00d1
gsdx ogl: flush vbo range instead of barrier
...
For testing purpose. I don't know which one is better.
It seems flushing have less fps fluctuation than barrier.
2015-04-21 21:44:50 +02:00
ce98276322
gsdx-ogl: improve speed of vertex streaming
...
Note yet enabled because I'm afraid of data corruption but feel free to test it
The option:
ogl_vertex_storage = 1
Performance note (warm cache+gs replay on colin3)
60 fps -> 76 fps
2015-04-20 09:38:03 +02:00
62489f42f1
gsdx-ogl: add an optimization note for later
...
Only 1 byte of fog is useful
2015-04-20 07:18:09 +02:00
6d253c0b8f
glsl: fix debugging of tex coordinate in apitrace
2015-04-20 07:18:08 +02:00
31f8c065db
gsdx-ogl: implement a new hack UserHacks_UnscaleSprite for opengl
...
UserHacks_UnscaleSprite = 1 will unscale flat sprites
UserHacks_UnscaleSprite = 2 will unscale all sprites (don't work well so far)
The idea of the hack is to redo the interpolation of texture coordinate
based on the non-upscaled pixel position.
It avoids various glitches but sprites aren't upscaled anymore (so no
more anti-aliasing, potentially a coefficient can be added).
2015-04-20 07:18:08 +02:00
6124eb844e
gsdx-ogl: only compile useful VS
...
logz is a constant
wildhack is only compatbile with TME/FST
Compilation goes down from 64 to 20 vertex shaders.
2015-04-20 07:17:58 +02:00
16e6d0d305
glsl: move shader into a separate directory
...
Only keep glsl_source.h for clarity
2015-04-19 18:49:02 +02:00
55fdf26898
glsl: remove the older file tfx.glsl
2015-04-19 18:49:02 +02:00
15264c6c63
glsl: split the main shader
...
* separate VS/GS and FS
* separate subroutine part of the FS
It already complex enough without subroutine stuff. Besides I'm not sure
we will keep subroutine on the future.
2015-04-19 18:49:02 +02:00
6fc9afb175
Merge pull request #507 from PCSX2/stdcall-for-plugin
...
Stdcall for plugin
2015-04-19 18:48:32 +02:00
e605ed1d09
gsdx-queue: add a comment for the future
2015-04-17 19:12:36 +02:00
fa243afbab
gsdx SW: enable new queue && C++11 on linux/MSVC 2012+
2015-04-17 19:12:36 +02:00
d91e989abb
gsdx-queue: pass shared_ptr by reference
...
It avoids atomic +1/-1 of the reference counter
The counter is still incremented when the ptr is copyed into the queue
2015-04-17 19:12:36 +02:00
84b33d2ddb
gsdx-queue: plug the new queue as a drop-off replacement of previous boost queue
2015-04-17 19:12:36 +02:00
90794c302a
gsdx-queue: import spsc_queue of boost
...
I remove 80% of the file to only keep the ring buffer core
Same speed as boost but without the boost dependency
2015-04-17 19:12:36 +02:00
c9194301a0
gsdx-queue: (linux) add a GUI option to select the queue
2015-04-17 19:12:33 +02:00
0aac47ca59
gsdx-queue: add a new option "spin_thread" to select the queue behavior at runtime
...
If someone has a more elegant solution, feel free to share it
spin_thread = 0
spin_thread = 1 // the faster but GS thread will never stop, very bad for laptop
2015-04-17 19:03:21 +02:00
9682061472
gsdx-queue:add a new job dispatcher queue based on boost and C++11
...
It is faster on linux, it requires less code, and it is "portable"
It requires boost (only hpp files) + MSVC 2013 (for atomic) (seem doable by 2012 too)
Actually there are several queues that either use spinlock or full sleep
2015-04-17 19:03:21 +02:00
a75d78bd7e
gsdx: use standard lock_guard instead of GSAutoLock
2015-04-17 19:03:21 +02:00
9ad5933120
gsdx: Use composition insead of inheritance to support lock
...
To ease update to C++11
2015-04-17 19:03:21 +02:00
8deee6afbc
gsdx: include some C++11 define for later
2015-04-17 19:03:21 +02:00
9ce7f515bc
cdvdiso: add stdcall convention
2015-04-17 18:34:05 +02:00
5c8ea74cb9
null plugins: add stdcall convention
2015-04-17 18:33:10 +02:00
75a3abc225
lilypad: add stdcall convention
2015-04-17 18:32:39 +02:00
8e3dab7e11
zzogl: add stdcall convention
2015-04-17 18:32:10 +02:00
e16cb9a00e
onepad: add stdcall convention
2015-04-17 18:31:44 +02:00
c456764342
spu2x: declare exported function as stdcall
...
* Fix a stack buffer overflow on address sanitizer
* Might explain an old clang crash
2015-04-17 18:24:23 +02:00
545c1d387c
gsdx: add an include for MSVC
2015-04-16 22:28:09 +02:00
aff999cf69
gsdx linux: refresh the gui
...
* Use tooltip for hack
* Update string of previous hack
* Remove unused hack
Note: hack_sprite_check requires 3 states (and potentially others hack too) but
I don't know how to do it. It likely requires a "scale button"
2015-04-16 19:28:55 +02:00
85cbe285f0
gsdx dialog: move hack info into a separate file
...
This way it can be reused on linux.
2015-04-16 19:26:33 +02:00
670bcc1818
gsdx hack: use a dumb implementation for UserHacks_round_sprite_offset = 2
...
It works as bad as a "clever" implementation.
It seems to be enough for games such as venus/taisho-monoke/FFX
Note: it might creates glitches. Code will never be nice, so it is just
a trade-off
2015-04-16 19:26:33 +02:00
41e5be0673
gsdx hack: limit range of UserHacks_align_sprite_X
...
Just an extra check to avoid an overlap (avoid white line in tekken)
2015-04-16 19:26:33 +02:00
e8698a9c0c
gsdx hack: hole in vertex position also appears in !FST projection
...
Fix various black vertical line regression with UserHacks_align_sprite_X
2015-04-15 09:40:06 +02:00
65a84b9f64
gsdx hack: drop previous filtering hack
...
superseeded by UserHacks_round_sprite_offset
2015-04-15 09:26:01 +02:00
2da4a5e6bc
gsdx hack: use a 0 offset instead of -1
...
-1 is annoying because minimum value is 0. Instead to add more logic,
let's try to use 0 which seems to be good enough (fix regressions on DQ8/AT)
Unfortunately it causes a mini regression on taisho-mononoke. Rotation of sprite is done with 2 triangles.
Potentially previous value wrongly recover this section.
2015-04-14 18:39:26 +02:00
cde72300a5
gsdx hack: Correct formulae for reversed texture
...
It fixes issues on Ace Combat UI
2015-04-13 22:14:36 +02:00
418f2e69a8
gsdx-ogl: implement the wildhack on the GPU
...
Likely much faster for opengl and much easier to implement
Note: hopefully UserHacks_round_sprite_offset will replace it
2015-04-13 22:14:36 +02:00
50bac01d89
gsdx hack: disable upscaling hack at native resolution
2015-04-13 22:14:36 +02:00
815c99b6e4
gsdx-hack: remove an extra offset + duplicate linear computing
...
Initially code aligned texture on pixel boundary + 1/16th. It seems to work without it.
2015-04-13 22:14:36 +02:00
399ddbd74a
gsdx hack: reduce complexity of alpha* function
...
It will help the compiler to make a couple of optimizations (and it will help
a futur porting to GSVector)
2015-04-13 22:14:36 +02:00
1d0deb5ffe
gsdx hack: move round sprite offset into a separate function
...
Use a boolean template to separate the linear case. I'm afraid it will
cost extra computation but 90% of sprites use nearest.
2015-04-13 22:14:36 +02:00
1aebca3740
gsdx hack: m_userhacks_round_sprite_offset take precedence over m_userhacks_stretch_sprite
...
m_userhacks_stretch_sprite is only kept for debugging. Let's avoid bad mixing by users
2015-04-12 11:05:23 +02:00
c4fef0256e
gsdx hack: Better rounding fix letter in ace combat and line in tekken
...
Tekken is broken with setting UserHacks_round_sprite_offset = 2 use 1 instead.
=> game use upscaling internaly so my rounding of coordinate break everything
Yakuza uses float coordinate but this hack only correct the integer coordinate
=> the solution will be to add a dedicated hack for this game.
Colin3 got a regression but I don't know when...
2015-04-12 10:31:04 +02:00
7e89247bf4
gsdx hack: don't add extra correct for zero length texture
...
Otherwise t1-t0 is 16 instead of 0
2015-04-11 10:54:30 +02:00
d2156c86eb
gsdx: improve previous hack for bilinear sprite
...
PS2 uses a -0.5 offset before sampling so texels must be rounded to half-pixel boundary
If fixes glitches on Venus and taisho-mononoke
Note: I didn't fixed yet texture render in reverse because I don't have any test for it.
2015-04-10 09:42:39 +02:00
f5eeb2024e
gsdx: anti-glitches v3: Tester wanted
...
UserHacks_round_sprite_offset = 1 <= enable correction of flat sprites
UserHacks_round_sprite_offset = 2 <= enable correction of all sprites (better on a couple of dump but not sure of the consequence)
I completely redo the algorithm. This time I do the projection and
interpolation of the 2 extrem vertex. This way I can compute the min/max
valid texture coordinate.
It gives stronger guarantee that texture sampling will be done inside the texture.
However it might have a performance impact, likely reasonable because it
is limited to sprite vertex.
A big thanks you to all people that provide me GS dump and test reports.
2015-04-09 19:51:50 +02:00
05eb1641f2
gsdx: add glsl debug capabilities
...
Useful to debug bad texture sampling
2015-04-08 10:15:25 +02:00
98d8ad7484
gsdx: new anti-glitches upscaling hack: UserHacks_round_sprite_offset
...
It is replacement of the previous hack (UserHacks_stretch_sprite). Don't enable both in the same time!
The idea of the hack is to move the sprite to the pixel boundary. It
avoids most of rounding issue. It also rescales verticaly the sprite (avoid horizontal line on ace combat).
I don't like this rescaling maybe we can limit it to only 1 pixels.
On my limited testcase, results are much better with any upscaling factor.
I still have a bad line in Kingdom heart. If you have issue with others
game please provide us a GS dump.
2015-04-07 19:18:24 +02:00
3077de3637
Merge pull request #496 from 3descx9/master
...
crc hack for Ponkotsu Roman Daikatsugeki: Bumpy Trot ( Steambot Chronicles JP)
2015-04-07 12:04:25 +03:00
defb57178f
crc hack for Ponkotsu Roman Daikatsugeki: Bumpy Trot ( Steambot Chronicles JP)
2015-04-07 14:14:26 +05:30
506648bd6b
GSdx: Hacks UI: Add missing BREAK that I left out
2015-04-06 14:39:51 -05:00
c6fb098703
Add two new tick boxes and associated description
...
For Gregory's two new awesome hacks
Also make new hacks obey "Enable HW Hacks"
2015-04-06 13:48:59 -05:00
3c84e6848b
gsdx: don't enable previous hack if m_upscale_multiplier <= 1
2015-04-04 01:21:46 +02:00
183b5d4def
gsdx: improve UserHacks_align_sprite_X hack for ace combat
...
Game uses 1x1 texture sampling. In this condition we don't care
if texture is correctly aligned.
2015-04-03 23:00:54 +02:00
53ab390e72
gsdx linux: add a nice gui option for previous hack
...
I also added two nice tooltips :)
2015-04-03 21:25:34 +02:00
fbaa84e691
gsdx: anti-vertical-line hack UserHacks_align_sprite_X
...
This ack realign sprite to full pixel coordinate
It solves vertical line in ace combat/tekken/colin mac rally
2015-04-03 21:02:25 +02:00
e40a26be19
gsdx: add some helper to debug openGL renderer
2015-04-03 20:28:44 +02:00
5269e54f02
gsdx: tune previous hack
...
Only disable bilinear on the sprite that were forced by the user.
If the PS2 requires a bilinear filtering, there is likely a big enough texture
2015-04-03 20:09:02 +02:00
bb728ea186
gsdx linux: filtering label was in wrong oder
...
The standard mode is 2
2015-04-03 20:07:50 +02:00
d4b7410e3e
gsdx linux: add a checkbox to enable easily previous hack
2015-04-03 18:33:05 +02:00
e1a5736583
gsdx: anti-upscale-glitch hack UserHacks_SkipDraw
...
2x upscaling is pixel perfects. Bigger upscaling is better but not yet perfect
Feedbacks are welcomes (note it doesn't solve all upscaling issue, only wrong texture sampling)
For the history:
If you have a texture of [0;16[ texels and draws a primitive [0;16[
The formulae to sample last pixels of texture is
0.5 + (16*s-1)/(16*s) * 16
Native (s==1): 15.5 (good)
2x (s==2): 16 (bad, outside of the texture)
4x (s==4): 16.25 (bad, really outside of the texure))
2015-04-03 18:33:05 +02:00
22710d9b44
lilyPad: fix the linux crash if ini doesn't exist
...
All credits go to Arcum42
2015-04-03 18:33:05 +02:00
45e2fd9204
crc hack for Raw Danger! (Zettai Zetsumei Toshi 2 US)
...
thanks to ssakash for the patch
2015-03-29 15:41:06 +03:00
45930ecbf3
gsdx: value was wrongly overwritten
...
Need a lots of tests to ensure at least no regression on cache behavior
Issue #332
2015-03-17 09:39:50 +01:00
3b5367c5b7
Merge pull request #452 from turtleli/fix-linux-build-avx2-with-tsx-disabled
...
GSdx: Linux: Fix build failure on AVX2 processor with disabled TSX, enable SSE4.2, SSE4.1, SSSE3 builds
2015-03-01 17:53:21 +01:00
1222bcbf6e
GSdx: Comment out TransactionScope code
...
The TransactionScope class is compiled but never used. Comment it out
for now.
2015-03-01 16:40:48 +00:00
c1604c4b38
gsdx 🧵 only keep std::mutex/std::condition_variable
...
Keep pthread for thread until std::thread is really working
2015-03-01 17:29:40 +01:00
i30817