Next commit will include some cleanups, and I'll probably also remove the dynamic-tunning and related support code (not a lot of code though, but.. it's currently unused). I'm also considering completely removing the old stretcher from the code, such that both dynamic-tunning and the old stretcher will only live in SVN history.
Meanwhile, you can test this patch and report back here if you notice a difference. BTW, on my system I use the portaudio output since IIRC it offers the lowest latency.
As before, to display stretcher messages at the console, enable underrun/overrun messages at the SPU2X debug UI (which is also available on normal builds).
git-svn-id: http://pcsx2.googlecode.com/svn/trunk@5729 96395faa-99c1-11dd-bbfe-3dabce05a288
Performance note, it might be faster to replace the MODULO with an AND. Not sure on the impact
for the new time stretcher algo.
GSdx: fixed use-after-free (linux)
PCSX2:
* add a define to support address sanitizer (both rely on 0x20000000-0x3fffffff memory ranges..)
* sio_buffer out of bond (-1). Maybe we can move the flush in the 2 if previous branch. It would
avoid the extra test.
* wxGetEnv (linux) generates double free (maybe not thread safe). Cache the result so it doesn't crash
anymore when switching renderer
Comment/Review/Improvement are welcome :)
git-svn-id: http://pcsx2.googlecode.com/svn/trunk@5727 96395faa-99c1-11dd-bbfe-3dabce05a288
* GL_ARB_shader_subroutine for perf
fix for nvidia => add missing shader declaration. Nvidia got +4fps on colin3 :)
For the moment only 2 PS parameters are supported. Code need to be extended to support others games that often
switch shader program (like xenosaga).
require GL4 class hardware and the option override_GL_ARB_shader_subroutine = 1
Note: strangely on AMD linux it is slower!
* GL_ARB_shader_image_load_store for accuraccy (Date)
Use a signed integer texture and reenable color buffer writing
Current status: Amagami_transparency.gs & P3_battle_shadows.gs are now working on Nvidia with a small perf impact.
Current implementation detail:
1/ setup the standard stencil as before
2/ on remaining pixel, draw once to compute first primitive that will write a fail alpha value.
3/ final draw based on primitive id of step 2
Note: I think we would get a bad behavior if depth test&mask are enabled on step 2/3
Note2: on my limited testcase the perf impact was on CPU. It would be possible to merge step1&2 to nullifying it (could
even be faster actually), however it would require more GPU power.
Again require GL4 class hardware. And the option UserHacks_DateGL4 = 1
git-svn-id: http://pcsx2.googlecode.com/svn/trunk@5725 96395faa-99c1-11dd-bbfe-3dabce05a288
zzogl: fix memory leak (fix#1431, #1432)
GSdx ogl: disable geometry shader on Nvidia/Windows (I will wait a 3rd implementation to find which one is correct)
git-svn-id: http://pcsx2.googlecode.com/svn/trunk@5724 96395faa-99c1-11dd-bbfe-3dabce05a288
* redo most of the texture upload (PBO): colin3 benchmark: 32 fps now (vs 26 fps 2 weeks ago)
* use the cross vendor vsync extension on linux (previous wasn't supported by nvidia)
git-svn-id: http://pcsx2.googlecode.com/svn/trunk@5721 96395faa-99c1-11dd-bbfe-3dabce05a288
* some preliminary work to test/benchmark bindless texture in the future (glsl was not yet updated)
Bindless texture allow to get a GPU texture pointer and then set it directly
to the shader as a basic uniform.
=> no more texture unit selection/validation
=> no more texture validation neither texture hash lookup
3rdparty: update gl header to the latest gl4.4
git-svn-id: http://pcsx2.googlecode.com/svn/trunk@5720 96395faa-99c1-11dd-bbfe-3dabce05a288
Card that support gs:
remain only a gs to generate sprite from a line. Even dummy gs are costly for the GPU.
Card that don't support gs:
remove useless copy of color for line and triangle primitives
Note for dx: opengl 3.2 (maybe not gles) supports both flat interpolation
convention (GL_FIRST_VERTEX_CONVENTION or GL_LAST_VERTEX_CONVENTION). It might
be possible to shuffle vertex index to put the last vertex in first position.
- buff[0] = head + 0;
- buff[1] = head + 1;
- buff[2] = head + 2;
+ buff[0] = head + 2;
+ buff[1] = head + 1;
+ buff[2] = head + 0;
git-svn-id: http://pcsx2.googlecode.com/svn/trunk@5718 96395faa-99c1-11dd-bbfe-3dabce05a288
The idea was to replace shader program swith by pointer function calls inside
shaders. At least parameters that are often changed between draw call. So far
I only ported atst and colclip. Unfortunately code is "slower" (on GSdx standalone).
For the moment keep the code but disabled.
If I understand well the validation of program is done in the "driver thread"
but the additional call are done in the overloaded MTGS thread. Apitrace
profiling shows faster GPU draw calls. Another possibility is that the driver still
need to validate the draw call because of others state change.
Here some stats on colin3 (90 frames):
without subroutine: UseProgram 125246
with subroutine: UseProgram 2906, subroutine 125945 => 3605 extra calls overhead (not
all parameters are ported to subroutine)
git-svn-id: http://pcsx2.googlecode.com/svn/trunk@5715 96395faa-99c1-11dd-bbfe-3dabce05a288
Note: I think we can do the same on DX11
Perf wise: on colin mcrae 3 it reduces shader prog setup from 3005 to 2086 each frames. It saves 2 ms of CPU processing (27->29fps)
git-svn-id: http://pcsx2.googlecode.com/svn/trunk@5714 96395faa-99c1-11dd-bbfe-3dabce05a288
* move most of gl states into a separate namespace. Extend it to depth/stencil/blend micro state
=> save 10,000 opengl call by frame for colin mcrae 3
* Only setup blend state of first drawbuffer
* Don't request anymore a debug context on dev/release build
git-svn-id: http://pcsx2.googlecode.com/svn/trunk@5713 96395faa-99c1-11dd-bbfe-3dabce05a288
* clean extension management and fix compilation of previous gl44 code.
* Use pixel buffer object to upload texture data.
=> avoid crash on AMD driver
=> a bit faster and probably got some margins for the future
git-svn-id: http://pcsx2.googlecode.com/svn/trunk@5712 96395faa-99c1-11dd-bbfe-3dabce05a288
* preliminary work for GL4.4 extensions (ARB_clear_texture & ARB_multi_bind). Disabled until I got a 4.4 driver
Note: I plan also to use ARB_buffer_storage
* compute texture gl option in the constructor (avoid a couple of swith case)
* redo texture unit management. Unit 0-2 for shaders, Unit 3 for texture operations. MultiBind will allow to bind
shader input without disturbing texture binding points.
git-svn-id: http://pcsx2.googlecode.com/svn/trunk@5711 96395faa-99c1-11dd-bbfe-3dabce05a288
* add a non-working hack: UserHacks_DateGL4, goal was to replace UserHacks_AlphaStencil
+ Detection of good/bad samples is based on primitive ID variable. However I'm not sure
the behavior is always the same between draw call...Anyway let's keep a copy of the current
work
* Dump integer texture into text csv
* add gl4.2 ARB_shader_image_load_store extension (needed by UserHacks_DateGL4)
git-svn-id: http://pcsx2.googlecode.com/svn/trunk@5707 96395faa-99c1-11dd-bbfe-3dabce05a288
* allow to switch renderer with F9
* skip first frame in stat of the replayer
* drop msaa. Fxaa and internal resolution will do the job
* move texture attachment from texture object into device object (allow to keep sanely the state)
* split the write buffer and attachment setup
* completely split sampler and texture input setup
* redo GSDeviceOGL::CopyOffscreen to avoid an extra copy.
git-svn-id: http://pcsx2.googlecode.com/svn/trunk@5704 96395faa-99c1-11dd-bbfe-3dabce05a288
* use gles header file, disable opengl code (mainly GLX, ARB_sso, geometry shader)
* Define properly the function pointer, GLES use basic linking whereas GL must get the symbol dynamically
* cmake: properly search and set libglesv2.so
* don't use dual source blending => HW renderer work (only miss unimportant FBA)
git-svn-id: http://pcsx2.googlecode.com/svn/trunk@5701 96395faa-99c1-11dd-bbfe-3dabce05a288
* uniform was wrongly set before the activation of the program (free driver only)
* Always use only 1 drawbuffer. Easier besides previous setup was wrong for convert:ps_main1
GLES trial (v3):
* add the cmake option GLES_API. Note library (libgles) are hardcoded for the moment
* Disable opengl check
* Disable gl_GetDebugMessageLogARB not supported!
* Emulate gl_DrawElementsBaseVertex, add manually the index offset (surely slow but work)
* Fix hundred of shader error (no implicit cast of integer to float...)
Unfortunately GLES doesn't support dual blending so no blend in hardware renderer. Otherwise it is fine
git-svn-id: http://pcsx2.googlecode.com/svn/trunk@5700 96395faa-99c1-11dd-bbfe-3dabce05a288
* replace vertex interface with block interface. It avoid to depends on the ARB_sso extension.
* disable geometry shader on Nvidia & Linux. Slower but better than a black screen !
* default logz to 1, avoid some glitches.
git-svn-id: http://pcsx2.googlecode.com/svn/trunk@5699 96395faa-99c1-11dd-bbfe-3dabce05a288
* redo glsl2h.pl script to generate only one big glsl headers.
* fix gcc warning in GSVector.h
* fix memory leak of GSDeviceOGL.m_shader
* clean shader compilation function => split generation header & drop malloc stuff
cmake:
* only rebuild shader when asked by the use. Avoid perl dependency to build pcsx2
git-svn-id: http://pcsx2.googlecode.com/svn/trunk@5698 96395faa-99c1-11dd-bbfe-3dabce05a288
* remove the possibility to compile shader from file. Some people loads older shaders...
The feature might be readded later
git-svn-id: http://pcsx2.googlecode.com/svn/trunk@5697 96395faa-99c1-11dd-bbfe-3dabce05a288
* try to setup advance gl context attribute. If driver doesn't support it fallback to the default.
zzogl:
* use glsl2h to generate an header shader as GSdx. Much easier to install
* Get GSUniformBufferOGL & GSVertexArrayOGL speed improvement from GSdx
cmake:
* detect current gcc version. Yield a warning if < 4.7 or an error if < 4.5
glsl2h:
* support zzogl too
* compute md5sum to avoid useless relink
git-svn-id: http://pcsx2.googlecode.com/svn/trunk@5696 96395faa-99c1-11dd-bbfe-3dabce05a288
* remove unused variable
* move static function from h into cpp
* Initialized hw_by_page to 0xFFFFFFFF instead of -1 (number must be a positive integer)
* Use a s32 fore m_current_lsn instead of u32 (use -1 as error code)
Bonus: a couple of fix for clang compiler (doesn't mean that it fully compile with clang)
* remove useless __debugbreak on linux
* use short for 16bits atomic function
git-svn-id: http://pcsx2.googlecode.com/svn/trunk@5695 96395faa-99c1-11dd-bbfe-3dabce05a288
* remove iconvert.cpp not used anymore
* change copyrigh header from GPL to LGPL. Those 2 files are only basic .h. 1 licence by plugin is enough ;)
git-svn-id: http://pcsx2.googlecode.com/svn/trunk@5691 96395faa-99c1-11dd-bbfe-3dabce05a288
* fix shader compilation on Nvidia (broken on r5682)
* fix various memory leak thanks to Valgrind
Cmake: fix a minor typo
git-svn-id: http://pcsx2.googlecode.com/svn/trunk@5688 96395faa-99c1-11dd-bbfe-3dabce05a288
* use svnrev.h on linux too
* replace sprintf_s with snprintf (hope it still compile on Windows)
* init integer with 0 instead of NULL
* various int -> u32/uint32/uint on for loop index
* remove a couple of unused variable
* init few variable
* disable unused warning results
git-svn-id: http://pcsx2.googlecode.com/svn/trunk@5683 96395faa-99c1-11dd-bbfe-3dabce05a288
* Separate state and shader compilation into separate function
* replace various hash_map by basic array
* Compact VertexScale and offset into a single vec4
* add the new option "ogl_vertex_subdata": subdata is faster on FGLRX, test are welcome on Nvidia drivers
0 => use map/unmap
1 => use subdata
replay: add "linux_replay" option and compute some nice stat (mean, standard deviation)
cmake: recreate shader header at build time
git-svn-id: http://pcsx2.googlecode.com/svn/trunk@5682 96395faa-99c1-11dd-bbfe-3dabce05a288
* don't delete the wnd in GSclose. It can still be used later
* Properly detect the GL_ARB_gpu_shader5 extension for Fxaa
* A couple of fix in Create (GSopen1) of GSWndEGL/GSWndOGL
git-svn-id: http://pcsx2.googlecode.com/svn/trunk@5674 96395faa-99c1-11dd-bbfe-3dabce05a288
