Commit Graph

299 Commits

Author SHA1 Message Date
Gregory Hainaut c41cf6c444 gsdx: init variable in constructor
CID 168626 (#1 of 1): Uninitialized scalar field
uninit_member: Non-static class member m_end_block is not initialized in this constructor nor in any functions that it calls.
2016-08-12 22:45:17 +02:00
Gregory Hainaut fb2182dd9b gsdx small nitpick 2016-08-09 15:15:30 +02:00
Gregory Hainaut e9e1b33884 gsdx: log a nice string instead of the hex format of PSM 2016-08-03 18:07:54 +02:00
Gregory Hainaut 16affc9ef4 gsdx tc: potentially fix a regression
Update done on f712c5c6d0

Previous code use the size of the draw to compute latest block. I
don't know why I use .x/.y which are the origin offset so the start of the block.
2016-08-02 15:35:28 +02:00
Gregory Hainaut 242ac26299 Merge pull request #1485 from ssakash/HPO_Custom
GSDX-TextureCache: Port Half pixel offset hack for custom resolutions
2016-07-29 15:07:02 +02:00
FlatOutPS2 02b0451d3c GSdx: Star Ocean 3 depth issue fix
Fixes issue #1475
2016-07-29 12:13:26 +02:00
Akash 47f69f694a GSDX-TextureCache: Port Half pixel offset hack for custom resolutions 2016-07-28 18:45:23 +05:30
Gregory Hainaut 7fa55c39f1 Merge pull request #1479 from ssakash/custom_fixes
GSDX-TextureCache: Add proper rounding when unscaling texture size
2016-07-27 22:31:21 +02:00
Gregory Hainaut d6383e6c21 Merge pull request #1472 from PCSX2/gsdx-gta-depth-issue-1457
Gsdx gta depth issue 1457
2016-07-26 18:40:53 +02:00
Akash d3ebd4b318 GSDX-TextureCache: Add proper rounding when unscaling texture size
Fixes flickering issues on Dragon Ball Z Budokai Tenkaichi FMV's on custom resolution, might also help prevent some crashes on custom resolution.
2016-07-26 18:28:07 +05:30
ramapcsx2 c592ccb3e5 Merge pull request #1435 from ssakash/Custom_TextureCache
GSDX: Improve scaling of custom resolution
2016-07-26 14:07:19 +02:00
Gregory Hainaut aea75f51bd gsdx tc: partial support of pseudo depth for Dx
Fix #1457 (GTA)

The game uses a depth format for a pure color buffer (cokes do ravage
in gaming industry)

However I'm really afraid that it migth break another effect in other games.
2016-07-22 20:38:46 +02:00
Gregory Hainaut abc9f7d096 gsdx tc: log unsupported LookupDepthSource 2016-07-14 18:29:22 +02:00
Akash a5671f016a GSDX-TextureCache: Add proper scaling for custom resolution
* Improve frame buffer height management on custom resolution. Width seems to be fine with the same size as scaled image output.
* Prevent offset issues on Persona 3 based on the data from merge circuit.

Note: Fixes custom resolution upscaling on ICO 50Hz/60Hz mode when large frame buffer is enabled. previously 60Hz mode only displayed half of the screen and 50Hz mode only worked due to the scissor hack.
2016-06-30 00:35:10 +05:30
Akash ce20544a4f GSDX-TextureCache: Remove hacks which caused scaling issues
* Ignore Frame memory offsets for calculating dimensions value of display rectangle.
* Remove hack which limited scaling size based on the scissor value.

Note: With the following commit, SilentHill 2 now properly outputs the desired resolution by the users on custom resolution. Previously if we set 1024 x 1024 , it'll output a lower height value which was caused by the hack removed on this commit.
2016-06-26 12:14:38 +05:30
Gregory Hainaut a9c18c57e4 gsdx option: use the new GetConfig* function
v2:
add PSX stuff
ssakash review
2016-05-28 10:07:27 +02:00
Gregory Hainaut 247d33a627 gsdx ogl: always clear depth buffer in FFX2
It avoid depth issues and I'm sure it won't broke other games this way.

Issue #1340
2016-05-19 17:53:46 +02:00
Gregory Hainaut a4c7541092 gsdx ogl: tentative hack to make FFX go along with depth option
Issue1: Depth buffer is wrongly invalidated only the first page is detected.

Issue2: First page seems to be partially written. Could be a GSdx transfer bug.
Anyway, invalidation only support a page granularity.

So here a quick workaround that will clear depth buffer in case of very small partial write.

Might worth to check regression on nocturne/digital saga
2016-05-17 19:33:12 +02:00
Gregory Hainaut 96b5170d8c Merge pull request #1317 from PCSX2/gsdx-array-coverage
Gsdx array coverage
2016-05-16 18:16:15 +02:00
Gregory Hainaut 91c164ecca gsdx tc: avoid potential division by 0 2016-05-07 18:36:28 +02:00
Gregory Hainaut 265ea82780 gsdx tc: avoid to load data outside of the GS memory
Avoid crash in Kungfu panda
2016-05-06 21:46:29 +02:00
Gregory Hainaut f34e77f08c gsdx: fix regression on 24 bit format (gust game)
bpp is the size of the container format but not the size of the format itself.
For PSM24, bpp is 32, trbpp is 24
2016-05-06 13:49:24 +02:00
Gregory Hainaut e890ce989c gsdx: throw oom exception + error
Texture can't be NULL anymore, so we can remove a couples of check
2016-05-05 18:53:51 +02:00
Gregory Hainaut 8424eeb4ee gsdx tc: small hint for the compiler 2016-05-05 15:32:21 +02:00
Gregory Hainaut 055b9c57e7 gsdx tc: use unscaled size as clamping in Target::Update
Might avoid/reduce crash like #1340
2016-05-05 12:59:49 +02:00
Gregory Hainaut f712c5c6d0 gsdx tc: use GSLocalMemory::m_psm instead of hardcoded value 2016-05-05 12:19:25 +02:00
Gregory Hainaut 826b38c47a gsdx tc: use array in GSOffset to store texture coverage
GSOffset is already based on a lookup of PSM/BP/BW. Coverage only adds
the size parameters (so only 256 possibilities)

It replaces the hash lookup with a free array access.
2016-05-01 15:02:36 +02:00
Gregory Hainaut 5676acaef9 glsl: support channel effect on depth texture too 2016-04-30 16:07:45 +02:00
Gregory Hainaut 7d191ebf8e gsdx tc: load palette for pseudo 8 bits depth
Rendering is still broken but I'm afraid it will need a dedicated shader
2016-04-30 14:52:43 +02:00
Gregory Hainaut 7a5e1d4154 gsdx ogl: allow to use a separate texture for the channel effect 2016-04-29 17:34:17 +02:00
refractionpcsx2 0aa8fe0f20 GSDX: Avoid crash when TBW == 0 2016-04-28 21:36:21 +01:00
Gregory Hainaut 726b901dc5 Merge pull request #1304 from PCSX2/gsdx-faster-texture-cache
gsdx tc: reduce texture cache overhead
2016-04-26 16:46:44 +02:00
Gregory Hainaut 49d175b677 gsdx tc: allow to create a dummy Source without tons of memory allocation
It makes shared texture virtually free from the CPU PoV.
2016-04-24 22:30:56 +02:00
Gregory Hainaut ad08701cb1 gsdx tc: trick the texture cache to use a depth format for depth texture
This way we avoid the rescaling of the depth buffer and texture allocation.
2016-04-24 22:18:26 +02:00
Gregory Hainaut 1960d51e60 gsdx tc: properly support 16 bits depth conversion 2016-04-24 22:18:26 +02:00
Gregory Hainaut 583de1bf0b gsdx tc: add a dedicated function to lookup a depth source
The hypothesis is that game will use a depth (aka Z32/Z24/Z16/Z16S)
format when sampling depth texture as color.  Technically one could use
a standard color format but block/pixel order won't be the same.
(otherwise I'm screwed)

=> Hypothesis invalid on GoW. They just do a scrambled rendering...

Lookup info:
* The first searched list is the depth pool as we search a depth
  texture.
* 2nd one is the render target pool (if a depth was converted to a
  render target already)

To avoid any CPU overhead, the source will be a pointer to the real texture
* Conversion (if float texture) will be done on the fly by the shader (GPU).
* Relative rescaling won't be supported. Texture must be fetched with
  integral coordinate
2016-04-24 22:18:26 +02:00
Gregory Hainaut 9ee090a36e gsdx tc: add a shareable texture flag
Plan is to bypass shader conversion/rescaling/copy for depth texture
2016-04-24 22:18:26 +02:00
Gregory Hainaut fa1377a8ee gsdx tc: reduce texture cache overhead
Cache page coverage of texture into a hash map

Test done on Champion of Norrath (paltex + DisablePartialInvalidation)

Profiler:
Self of GSTextureCache::SourceMap::Add 5.39% => 0.23%
Self of GSTextureCache::LookupSource 15.27% => 10.82%

Hard to measure on CoN as it depends on memory transfer. Seem to be 5-10 fps faster.
2016-04-23 19:22:58 +02:00
Gregory Hainaut a149d6b26f gsdx: use psm.bpp flag to improve code reading 2016-04-23 18:29:39 +02:00
Gregory Hainaut 3709b5aecc gsdx: add a small comment to ease code reading 2016-04-21 09:28:44 +02:00
Gregory Hainaut cb8088216b Merge pull request #1281 from PCSX2-Alpha/NTSC_saturation
GSDX: Remove some unnecessary/dubious hacks
2016-04-21 09:28:30 +02:00
Gregory Hainaut 59ef668fd2 Merge pull request #1222 from PCSX2/gsdx-black-fmv
Gsdx black fmv
2016-04-21 09:25:36 +02:00
refractionpcsx2 dcb676765e GSdx: Clean up some warning under windows (#1289) 2016-04-14 11:00:58 +01:00
Gregory Hainaut 4611264ce3 gsdx: add CRC hack for Nocturne PAL
Reduce depth transfer to the real size of the game. Otherwise
we need to handle a mix of color/depth transfer.
2016-04-11 16:04:04 +02:00
Gregory Hainaut 9e6cb79f4e gsdx linux: move depth option as a speed hack
Someone ought to add the Windows option too (and DisablePartialInvalidation too)

It might break a couple of games but most of them run better with depth enabled.
2016-04-10 13:22:55 +02:00
Akash a1b0ae467e GSDX: Remove dubious/unnecessary hacks
* Silent Hill 2 doesn't need the CRC hack
* GSRenderer: no need to explicitly set bottom value for r.
* Texture Cache: Removed a check which couldn't possibly enter true
branch.
2016-04-07 15:15:39 +05:30
Gregory Hainaut 063d2e568a gsdx tc: re-implement frame lookup
Try to avoid random black screen frame

v2: don't force the preload hack on the frame
It creates a ghost image over FMV

v3: support offset within a frame
2016-04-04 21:30:37 +02:00
Gregory Hainaut 72d43b2e56 gsdx tc: add an invalidation procedure to drop sub rt
It will be used in next commit. However, it might not be the best solution
2016-04-04 21:29:32 +02:00
Gregory Hainaut d922225e69 gsdx tc:invalidate RT when a texture is written inside it
It often happens the game try to upload the FMV directly which typically
gave a black screen.

Commit fix rules of roses and I hope various black screen FMV

Performance impact must be tested, and I'm afraid of strange texture cache behavior.

V2: check the size of the transfer too
V3: add support of 16 bits format
V4: avoid division by 0
2016-04-04 21:29:31 +02:00
Gregory Hainaut 9aea76b0e8 gsdx tc: store a hint on the target size
The hint is based on the rendering
2016-04-04 21:29:31 +02:00
Gregory Hainaut c4ebdd1f37 gsdx ogl: allow to read the depth buffer
Unfortunately can't be tested on gs dump

Ought to impact #1276 (Incognito/Eat Sleep Play's games, Nocturne too)
2016-04-03 12:30:14 +02:00
Gregory Hainaut 00643dbdfb gsdx tc: report unsupported depth read path
Help to debug depth issue  (aka lighting effects pass through wall)
2016-04-03 11:22:11 +02:00
Gregory Hainaut c43b1ac201 gsdx tc: extend the fast invalidation hack to limit the size of target read
It actually removes the previous hack that read the full target.

Unfortunately snowblind engine game uses big target so the read is very big too (1280x448)
which is killer for the perf. Whereas the game requires only 24x12 texels

Give a 2x speed boost on Champion of Norrath !!!
2016-03-20 18:11:44 +01:00
Gregory Hainaut 46a4d2dc47 gsdx-tc: add UserHacks_DisablePartialInvalidation for snowblind engine game
Games uses very special texture with a lots of repeating.

It is much faster to send the full texture rather than trying to partially invalidate it.

On my gs dump:
FPS: 29 => 68 !
2016-03-19 00:25:58 +01:00
Gregory Hainaut 902e295089 gsdx: fix onimusha crash in custom resolution
I don't understand why but it seems it needs more than 8MB
2015-12-30 19:14:52 +01:00
Kingcom 191453d89a gsdx: fix windows compilation 2015-12-28 22:05:01 +01:00
Gregory Hainaut a595a09fbd gsdx: increase buffer to have crash in case of overflow
Avoid a crash on Onimusha3 (PAL 60HZ)

In theory it will be better to find the root cause of overflow. I.e. somewhere in this
code below. Dirty rectangle is too big.

***********************************************************************
if(rowsize > 0 && offset % rowsize == 0)
{
    int y = GSLocalMemory::m_psm[psm].pgs.y * offset / rowsize;

    if(r.bottom > y)
    {
        GL_CACHE("TC: Dirty After Target(%s) %d (0x%x)", to_string(type),
                t->m_texture ? t->m_texture->GetID() : 0,
                t->m_TEX0.TBP0);
        // TODO: do not add this rect above too
        t->m_dirty.push_back(GSDirtyRect(GSVector4i(r.left, r.top - y, r.right, r.bottom - y), psm));
        t->m_TEX0.TBW = bw;
        continue;
    }
}
***********************************************************************

So as a temporary solution (that will likely stay for a couple of
years), buffers were increased.
2015-12-28 21:40:06 +01:00
Gregory Hainaut c36fccdd8e gsdx: fix crash when preload data frame is enabled
Height of the dirty rectangle must be the GS size of the RT. Of course
RT doesn't have any height so we compute the max safest value.

Fix issue #987

Candidate for 1.4 release
2015-12-28 18:00:14 +01:00
Akash 8b6ad4ff85 GSDX-TC: check userhacks for preload data frame 2015-11-22 10:49:46 +05:30
Gregory Hainaut 7a69812b17 Merge pull request #927 from PCSX2/preload-rt
gsdx-tc: extend preload frame hack to load target too
2015-11-16 09:36:31 +01:00
willkuer 9245d3ed7e gsdx: RendererEnum
1. Add GS_Renderer Enum

Replace all instances of int/uint32 renderer identifier by a strongly
typed enum and appropriate casts.

Only instances in GS[*].cpp/h classes were touched. GPU[*].cpp/h classes
do not to follow the same convention.

2. Add default renderer according to OS

The default renderer is OS dependent (Win -> Dx9HW, others -> OGLHW).
Consequently one should always check againt the appropriate default
value on config load.

The old behaviour was only - if a at all - problematic if the respective
element in the gsdx.ini was missing and probably even then didn't create
issues. The current implementation is still more stable and does not
depend on the implementation of GS.cpp -> GetConfig()
2015-11-10 00:26:39 +01:00
Gregory Hainaut 5ba41306ea gsdx-tc: HasSharedBits uses PSM not address
Nice heap overflow. I'm surprised that it kinda worked.
2015-11-02 07:36:37 +01:00
Gregory Hainaut 03bf1ec7b3 gsdx-tc: extend preload frame hack to load target too
The goal is to check the impact on game that have wrong RT content.

It helps a bit Smash Court Tennis Pro Tournament 2 but the game suffers
another texture cache bug. (RT BW is 10 whereas texture BW is 8)

Note: Armored Core: Last Raven must be tested (only game so far
that rely on the option and I didn't want to add a new one).
2015-10-25 15:26:38 +01:00
Gregory Hainaut 295f867118 gsdx: invalid alpha after an EE write
Typical wrong draw:
1/ draw in 32 bits
2/ draw in 24 bits
3/ Use alpha as a texure. (Must reuse the GPU data)
4/ Write alpha from EE
5/ Use alpha as a texure. (Must upload new data)

This commit fixes the step 5.

Fix #917 (Conflict - Desert Storm)
2015-10-24 16:23:50 +02:00
Gregory Hainaut 5fb8c7e65c gsdx: initialize members in constructor of objects
A couple of useless members were removed too.

Also fix wnd initialization

Coverity:
CID 146955 (#1 of 1): Uninitialized pointer read (UNINIT)
18. uninit_use: Using uninitialized value wnd[i].
2015-09-23 09:46:53 +02:00
Jonathan Li 69d27dfb4b Merge pull request #829 from ssakash/rework_internalres
gsdx changes:
Remove native resolution checkbox from GUI and rework associated code
Small changes to Windows and Linux GUI
Support 8x native resolution
Fix custom resolution width less than native width use case
2015-09-21 20:12:47 +01:00
Akash a9b0cfc519 GSDX: Add 8X native to Internal resolution. 2015-09-20 07:49:32 +05:30
Gregory Hainaut cee0fde940 gsdx-tc: improve support of TEXA register
* Greatly reduce the number of clut read (factor 10x)
* Avoid to get wrong TEXA texture in the cache.
* Fix "jump depends on uninitialized variable" Valgrind warning.

Fix #748

I try my best to avoid any breakage of DX but please test it too.
2015-09-11 14:16:11 +02:00
Gregory Hainaut ddc03cbccf gsdx: add a nice enum to replace the hardcoded convert shader 2015-09-08 16:51:24 +02:00
Gregory Hainaut 1ae0fed83b gsdx: fix a regression due to the shader renaming 2015-09-08 16:06:15 +02:00
Gregory Hainaut ca9b5ce11d gsdx-ogl: move depth conversion shader
Add 2 new shaders:
* ps_main12: cast a 16 bit depth to a RGB5A1 color
* ps_main16: cast a a RGB5A1 color to a 16 bit depth

Shader might be used in future commit as it seems Silent Hill uses this
kind of format.

Fix tab/indentation too
2015-09-08 12:41:05 +02:00
Gregory Hainaut d822b6bd1a gsdx-ogl: add some comments for the future 2015-08-08 09:17:06 +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 f2c77ffa08 gsdx:tc: detect texture shuffle in previous frame
GoW2 uses the effect at the start of a new frame and therefore need the
state of the previous frame. (Note 1 issue remains)

Ricky need to be tested.
2015-07-26 10:14:07 +02:00
Gregory Hainaut 8be32e4661 gsdx-tc: fix a memory leak when depth isn't fully supported
Fix issue #642
2015-07-12 17:57:42 +02:00
Gregory Hainaut 5f8d3b9e4b gsdx-tc: remains in the memory
avoid a crash with kungfu panda
2015-07-11 14:35:35 +02:00
Gregory Hainaut 5ed45b6806 Merge pull request #644 from PCSX2/improve-texshuffle-detection
Improve texshuffle detection
2015-07-11 09:34:33 +02:00
Gregory Hainaut ec61ad3667 gsdx-tc: Add some code for the future
Partially invalidate RT when there is a write in the middle of it (actually 2 pages below)

Code is not yet enabled because
1/ I want to stabilize latest update
2/ not sure of the impact of the code
3/ maybe it need a more generic version
2015-07-10 22:35:26 +02:00
Gregory Hainaut 6f9a89dcf2 gsdx-debug: print memory usage of all textures pools
It is a bit crude but it allow to see the impact of code and options.
2015-07-10 22:35:26 +02:00
Gregory Hainaut ad86bb5faa gsdx-tc: clear 32 bits state after vsync
Avoid issue on game that uses only 16 bits RT
2015-07-09 23:03:55 +02:00
Gregory Hainaut 95e1cd9ea8 gsdx-tc: don't considerer frame as a 32 bits RT
Frame is always 32 bits but game can reuse it later as a 16 bits RT.

Fix half screen issue with Ricky Ponting Cricket

Unfortunately it triggers texture shuffle wrongly. I hope there is no
regression.
2015-07-04 11:00:32 +02:00
Gregory Hainaut 0a07762f6b gsdx-tc: only track 32 bits fmt if the texture is not a palette fmt
Fix a regression on ricky ponting cricket (texture shuffle was triggered without any valid reason)
2015-07-03 23:04:48 +02:00
Gregory Hainaut c6f026ee68 gsdx-tc: keep 32 bits status when converting from/to depth
Partially fix conan effect
2015-07-03 19:02:44 +02:00
Gregory Hainaut 28815db7cc gsdx: fix depth option 2015-07-02 19:13:38 +02:00
Gregory Hainaut 831c24de51 gsdx-dx: only enable new code when CRC is below FULL level
Code is not ready. It allow to test the new code without a massive
regression
2015-07-01 09:36:54 +02:00
Gregory Hainaut 074881228d gsdx:comment: small explanation of the half offset hack 2015-07-01 09:36:54 +02:00
Gregory Hainaut 6121677aa1 gsdx-tc: use a single shader pass to convert texture in 8 bits format
It might save a couple of fps

Add a define to test the perf if we keep only the blue channel. It brokes
the code in Prince Of Persia that use the Red/Green channel... Maybe the
speed hack :( Or find a way to replace all if with a lookup table

Note: it is only supported on OpenGL currently
2015-07-01 09:36:54 +02:00
Gregory Hainaut 2ecca529d1 gsdx-tc: log dirty target 2015-07-01 09:36:54 +02:00
Gregory Hainaut 2af7dcbda2 gsdx-tc: GPU accelerate 8 bits texture when upscaling is enabled
Code unscale the texture to ease the conversion. Quality is awful (same as before)
but I'm not sure we can support an upscaled texture

Maybe the quality loss is due to the reduction without mipmap

Maybe the best solution will be to add an hack to extract the blue channel
(with texture swizzle), and uses a "full page/screen" spirte instead.
(it would be faster too)

Note: won't be compatible with MSAA (but gl doesn't support it anyway)
2015-07-01 09:36:54 +02:00
Gregory Hainaut 6ca7a802bf gsdx-tc: add a big comment to explain the scaling of src in TC 2015-07-01 09:36:54 +02:00
Gregory Hainaut d29e375f72 gsdx-tc: GPU accelerate 8 bits texture conversion
Only native is supported currently
2015-07-01 09:36:54 +02:00
Gregory Hainaut a8bcc760b4 gsdx-tc: only clean RT in openGL
It is a debug feature for myself
2015-07-01 09:36:54 +02:00
Gregory Hainaut 61a717db14 gsdx-tc: clean the RT in debug build
// In theory new textures contain invalidated data. Still in theory a new target
// must contains the content of the GS memory.
// In practice, TC will wrongly invalidate some RT. For example due to write on the alpha
// channel but colors is still valid. Unfortunately TC doesn't support the upload of data
// in target.
//
// Cleaning the code here will likely break several games. However it might reduce
// the noise in draw call debugging. It is the main reason to enable it on debug build.
//
// From a performance point of view, it might cost a little on big upscaling
// but normally few RT are miss so it must remain reasonable.
2015-07-01 09:36:53 +02:00
Gregory Hainaut 99b03b4632 gsdx-debug: log all write (texture) from the EE
It is verbose but it might help to understand the invalidation of texture
2015-07-01 09:36:53 +02:00
Gregory Hainaut e66aac8ab7 gsdx-hw: add an hidden option to preload the final frame with GS mem data
Game can directly uploads a background or the full image in
"CTRC" buffer. Previous code was a full black screen.

It will also avoid various black screen issue in gs dump.

hidden option: preload_frame_with_gs_data

Note: impact on upscaling was not tested and it's likely broken
2015-07-01 09:36:53 +02:00
Gregory Hainaut 85543db094 gsdx-tc: trace the alpha channel update in TC
Improve the rendering in MGS3 (even if the game is still broken
due to others TC issues)

// Typical bug (MGS3 blue cloud):
// 1/ RT used as 32 bits => alpha channel written
// 2/ RT used as 24 bits => no update of alpha channel
// 3/ Lookup of texture that used alpha channel as index, HasSharedBits will return false
//    because of the previous draw call format
//
// Solution: consider the RT as 32 bits if the alpha was used in the past
2015-07-01 09:36:53 +02:00
Gregory Hainaut 9802ba6621 gsdx-tc: add support of pseudo 8 bits RT conversion
Code is obviously slow but at least it works.

It fixes the blur effect of VP2
2015-07-01 09:36:53 +02:00
Gregory Hainaut 77e718c61f gsdx: disable linear interpolation when converting texture in texture cache
It avoid various upscaling glitches on GS post-processing effect

// 1/ Palette is used to interpret the alpha channel of the RT as an index.
// Star Ocean 3 uses it to emulate a stencil buffer.
// 2/ Z formats are a bad idea to interpolate (discontinuties).
// 3/ 16 bits buffer is used to move data from a channel to another.
//
// I keep linear filtering for standard color even if I'm not sure that it is
// working correctly.
// Indeed, texture is reduced so you need to read all covered pixels (9 in 3x)
// to correctly interpolate the value. Linear interpolation is likely acceptable
// only in 2x scaling
//
// Src texture will still be bilinear interpolated so I'm really not sure
// that we need to do it here too.
//
// Future note: instead to do
// RT 2048x2048 -> T 1024x1024 -> RT 2048x2048
// We can maybe sample directly a bigger texture
// RT 2048x2048 -> T 2048x2048 -> RT 2048x2048
// Pro: better quality. Copy instead of StretchRect (must be faster)
// Cons: consume more memory
//
// In distant future: investigate to reuse the RT directly without any
// copy. Likely a speed boost and memory usage reduction.
2015-07-01 09:36:53 +02:00
Gregory Hainaut 87f54ae0ff gsdx: report error on unsupported case in texture cache
It seems to impacts lots of games that still have issue (VP2, MTG3, PoP)

The PSMT32 format is read a PSMT8. I think we need to convert it as PSMT8H (i.e.
unpack it to have only an alpha channel)
2015-07-01 09:36:53 +02:00
Gregory Hainaut 3b127f663b gsdx-tc: trace the texture format to detect texture shuffling
It fixes games that uses 16 bits RT (like snow engine games)
2015-07-01 09:36:53 +02:00