Commit Graph

2905 Commits

Author SHA1 Message Date
Gregory Hainaut 9d1674db99 gsdx-gui: remove useless linux ifdef 2015-08-04 22:08:55 +02:00
Gregory Hainaut 36554c3375 Merge branch 'hdr-colclip-32bits' 2015-08-04 21:55:40 +02:00
Gregory Hainaut 45bb27d6db gsdx-ogl: extend HDR colclip to 32 bits texture
Unfortunately 16 bits wasn't enough for Castlevania.
2015-08-04 21:54:27 +02:00
Gregory Hainaut c6ff7531fb gsdx-ogl: performance boost on virtuafighter 2015-08-04 21:26:03 +02:00
Gregory Hainaut d80aa0b0bd gsdx-ogl: remove 2 printfs
GL_INS is a better tracing solution
2015-08-04 20:47:02 +02:00
Gregory Hainaut 744f9ebc09 gsdx-ogl: rare corner case when both texture shuffle and date are enabled
In texture shuffle mode the texture data is either RG or BA. It means
that DATE must either checks MSB of G or A.

Close #693
2015-08-04 20:10:44 +02:00
gabest11 42f51591df gsdx: lupin 3 fix, texture addressing outside the limits, only for sw and opencl yet 2015-08-04 13:27:08 +02:00
gabest11 d631030608 gsdx: fixed some of the renames where it made no sense, also added a commend about lupin 3rd. 2015-08-04 03:36:42 +02:00
Maxxus c0f85de17f onepad: Minimum code needed to get Dualshock 3 working on recent linux and evdev releases.
Commit message edited by Gregory.
Comment below from Jonathan Li

On my system, /dev/input/event13 (evdev interface) and /dev/input/js0 (joystick interface) correspond to the DS3 on my system. User/group ownership are both root.

With /dev/input/event13 at 0640 and /dev/input/js0 at 0644 - SDL2 detects no pad, SDL1 detects a 19 button, 27 axis pad

With /dev/input/event13 at 0666 and /dev/input/js0 at 0644 - Both SDL1 and SDL2 report a 19 button, 4 axis pad.

SDL2 only uses the evdev interface, SDL1 uses the evdev interface but if that fails, it uses the joystick interface.
2015-08-03 19:34:08 +02:00
gabest11 e010004f1f gsdx: _isnan was not reliable, rewrote it as GSVector4::replace_nan, it uses cmpps and the result as the mask to blend the original value and FLT_MAX. No jumps or function calls. 2015-08-03 14:35:47 +02:00
gabest11 85117ecbdd gsdx: simplified isnan test, it compiles to an inline sse instruction and a flag test, silent hill dump still looks fine. 2015-08-03 11:27:14 +02:00
Gregory Hainaut 6e1cd077a9 gsdx-debug: push ubo in a group 2015-08-03 08:46:25 +02:00
Gregory Hainaut e972f4f4dd gsdx-ogl: extend device to support an offset for normal draw 2015-08-02 21:30:19 +02:00
Gregory Hainaut cabd7409e5 gsdx-ogl: don't validate the pipeline
Too noisy in debugger trace. Never catch any errors anyway.
2015-08-02 21:30:19 +02:00
Gregory Hainaut 59cdf77784 gsdx-ogl: create a new function to set the blending state 2015-08-02 21:30:19 +02:00
Gregory Hainaut 1da611fb75 gsdx-ogl: clean PS selector 2015-08-02 21:30:19 +02:00
gabest11 542dd33169 Merge branch 'master' of https://github.com/pcsx2/pcsx2 2015-08-02 20:27:33 +02:00
gabest11 a1a842b07f gsdx: date/datm fix for 16-bit frame buffer in sw rendering mode 2015-08-02 20:21:49 +02:00
Gregory Hainaut 8424c18e9f Merge pull request #688 from PCSX2/hdr-colclip
Hdr colclip
2015-08-02 18:13:28 +02:00
Gregory Hainaut 1f402b1b56 gsdx-ogl: fix bad detection of overlapping
avoid rendering corruption with SW blending
2015-08-01 13:29:25 +02:00
Gregory Hainaut ec007ac8d0 gsdx-ogl: support accurate blending without geometry shader
For the Mesa driver
2015-08-01 13:26:15 +02:00
Gregory Hainaut 4a3c145c72 gsdx-ogl: depth support: better support of 16 bits z buffer
Fix issue in socom2
2015-08-01 01:28:41 +02:00
Gregory Hainaut eb0fa8c7dc gsdx-ogl: fix bad detection of overlapping
avoid rendering corruption with SW blending
2015-08-01 01:27:22 +02:00
Gregory Hainaut 8452d2ccfe gsdx-ogl: fbmask regression! don't use bit operation with integer 2015-07-31 21:10:58 +02:00
Gregory Hainaut fff59f547d gsdx-ogl: fbmask regression! don't use bit operation with integer 2015-07-31 19:43:06 +02:00
Gregory Hainaut a0edcb58af gsdx-ogl: extend cclip blending level with destination alpha blending
The purpose is to emulate correctly destination alpha factor

An alpha channel of 128 is 1.0 in the GS but only ~0.5 in the GPU

I think few draw call use destination alpha so impact on perf must remains small.
2015-07-31 09:45:28 +02:00
Gregory Hainaut 97b38d9e1b gsdx-ogl: directly set impossible mode in the blending table
Avoid to hack it in the creation

Allow in the future to reuse the table directly instead of converting
in a blend object
2015-07-31 09:45:28 +02:00
Gregory Hainaut 8554f32086 gsdx-ogl: clean the blend table
Remove old shader define
Prefix macro with BLEND_
Add some notes to explain the special symbol
2015-07-31 09:45:28 +02:00
Gregory Hainaut cfd0fd6cc8 gsdx-ogl: remove old colclip algo 2015-07-31 09:45:28 +02:00
Gregory Hainaut 93c47feb7c gsdx-ogl: replace old colclip algo with the HDR algo
Similar speed but more accurate

Allow to clean the code
2015-07-31 09:45:28 +02:00
Gregory Hainaut 83f874db93 gsdx-ogl: remove bsel.ps
Just clear bsel.abe to disable blending
2015-07-31 09:45:28 +02:00
Gregory Hainaut 25298c70f7 gsdx-ogl: move blending management into a separate function 2015-07-31 09:45:28 +02:00
Gregory Hainaut 25bd5f5e85 gsdx-ogl: request texture barrier to emulate accurate date
Actually it can partially be done with GL_ARB_shader_image_load_store
extension. However all drivers that support shader_image have
texture barrier too.
2015-07-31 09:45:28 +02:00
Gregory Hainaut 2901e94ebc gsdx-ogl: always bind the RT as input texture
To avoid code duplication
2015-07-31 09:45:28 +02:00
Gregory Hainaut 1fe3e04ce3 gsdx-ogl: don't alias m_env/m_context variable
It is cumbersome to move code
2015-07-31 09:45:28 +02:00
Gregory Hainaut 8f27a5a92b gsdx-ogl: only enable fast accurate colclip in level3
Until we drop the old method
2015-07-31 09:45:05 +02:00
Gregory Hainaut 83dfc6b633 gsdx-ogl: clean a bit selector code
Use countof macro (avoid to duplicate the size)
Fix the size of array
Remove useless alpha_stencil case
2015-07-30 18:36:05 +02:00
Gregory Hainaut e026f1bac6 gsdx-ogl: implement a fast accurate colclip algo
The idea is to use a floating texture to accumulate the data and
then do a final postprocessing pass to apply the modulo

v2:
* use bounding box to
* fix vertex corruption issue
* use negative number in shader which allow to use half float (+12
  fps@4x)
2015-07-30 18:34:52 +02:00
Gregory Hainaut aa8f5848d1 gsdx-ogl: always issue a barrier when requested
Safer this way
2015-07-30 18:24:36 +02:00
refractionpcsx2 28fbae4791 gsdx-gui: Tidy up GUI options 2015-07-30 18:24:32 +02:00
Gregory Hainaut 88bd0996f5 gsdx-ogl: only print same tex/rt message when prims overlaps
Avoid most of the false positive
2015-07-30 18:24:32 +02:00
Akash 350e373e3f gsdx-gui: Remove Aggressive CRC hack.
The option is pretty much useless, CRC hack level controls the usage of hacks already.
2015-07-30 18:24:28 +02:00
Akash 68833e10d1 gsdx-gui: Accurate blending unit for Windows.
The following patch merges all the Accurate options related to the blending unit into a single one.
2015-07-30 18:24:19 +02:00
Gregory Hainaut d191ae4aba glsl: use rounding in convert 4 shader
It will avoid texture rouding error with negative number
2015-07-30 18:22:59 +02:00
Gregory Hainaut 7b9fa8fbe5 gsdx-ogl: add an assertion that will save me 2015-07-30 18:22:59 +02:00
Gregory Hainaut 46a1525668 gsdx-ogl: support various texture format creation
Such as GL_RGBA32F/GL_RGBA16F/GL_RGBA16UI/GL_RGBA16I/GL_RGBA16

Not yet used
2015-07-30 18:22:59 +02:00
Gregory Hainaut ee9edb0b19 gsdx-ogl: create a copy rect with conversion function
Previous CopyRect function does a memcopy without conversion.

This function will allow to use different format for input/output. Just a
possibility for the future
2015-07-30 18:22:59 +02:00
Gregory Hainaut ae8df002af gsdx-ogl: optimize Cs * As + Cd and Cs * Af + Cd blending
Basically the code does the alpha multiplication in the shader therefore
the blend unit only does a pure addition. This way the multiplication is
accurate and accurate_blending doesn't requires a costly barrier.

This code also avoid variable duplication to make the code more separated.
Hopefully blending can be done in a separated function

It is preliminary work to support fast color clipping with HDR

v2: fix assertion compilation failure

v3: fix regression in not accurate mode

v3: Cs * As/Af is not an accumulation

Those cases don't need the Cd addition and were already optimized anyway

Fix a regression on GoW2
2015-07-30 18:22:59 +02:00
Gregory Hainaut 12fdc37599 gsdx-ogl: reorganize blending in the renderer
Do DATE algo selection before blending. This way we can detect bad
interaction.

Regroup all blending/colclip in a single block.  Avoid to check abe &&
rt multiple times.

v2: only enable sw blending when abe is true
2015-07-30 18:21:01 +02:00
Gregory Hainaut caadc73e1b gsdx-ogl: add a new level for accurate blending
The updated medium level will run for all sprites. It helps sotc blooming effect and it remains
fast enough to be enabled by default (at least on 3D games)
The new high level will run for all sprites + color clipping
2015-07-30 18:21:01 +02:00