* Improve secondary dialogs in Onepad. I'll do gtk stuff in this branch, too, I promise...
* More fiddling with onepad's secondary dialogs.
* Rework config.inl to use wx, getting the rest of the null plugins.
* Remove some unnecessary includes, and convert SysMessage to wx in onepad.
* Add in tellowkrinkle's Mac OS fixes.
Updates the d3d device and swapchain creation to more modern methods.
- Use CreateDXGIFactory2 to create the factory and store in member
- Use CreateSwapChainForHwnd
- Add messages for failures to the console
- Some general formatting improvements
Convert sprites on the cpu instead of the gpu when applicable. Port from
Opengl.
Lines: GPU conversion.
Triangles: CPU conversion.
Disable Geometry shader when conversion is done on the cpu.
v1 It won't allow users to try to input a value that isn't supported
osd (size, timeout, max), tc offsets (x, y), rendering threads.
v2 Use const if applicable for variables throughout
GSSettingsDlg.cpp. Will cleanup some warnings.
DMC3, Onimusha 3 rely on this behavior.
They do fullscreen rectangle with scissor, then shift by 8 pixels, not done with recursion.
So we check if it's a TS effect by checking the scissor value.
Credits to Kojin for figuring out the issue.
Hack is obsolete/replaced with accurate date/date_one option/code.
There's pretty much no usecase for it remaining other than using it as
an alternative to skipdraw which removes/breaks transparency effects.
In GSRendererHW, reuse the method CanUseSwSpriteRender for OI_JakGames.
As in current master, 16x16 "flat" sprite draws are performed via CPU: those are palette rendering better emulated with the SwSpriteRender.
Jak games do also many 64x64 "flat" draws, but mostly for render to texture effects that are already correctly emulated in the TC, so those draws are not done via CPU (as on the contrary happens for DBZ BT games).
Improve texture cache Source objects invalidation logic taking into account overlapping with current draw.
Fixes eyes rendering for the Major in Jak 1.
Use SwSpriteRender to render sprites, which:
- fixes kaioken effect,
- fixes character outlines when upscaling,
- reduces frametime by up to 3ms with big upscaling factors.
Don't disable sparse texture by default before extension check, it will
display sparse texture always not supported. Do an ini check in
check_sparse_compatibility function instead.
Log should properly state if sparse texture is supported now.
Move Additional colclip info, dithering, FixedTEX0 to extra debug logs.
It will allow to keep track of more important stuff going on and they
can be enabled with ENABLE_EXTRA_LOG if needed.
Change context creation log type from stderr to stdout.
The behavior was verified on Dobie to be correct.
The code needs to be ported to SW renderer too to
improve rendering on SW side.
Current PR will fix plenty of games on HW renderer
that had/have zbuffer issues before.
v2. Set DepthMask to the maximum the current depth format allows.
Will properly clamp for 16bit and 24bit formats.
v3. gl: Fix uniform buffer upload/cache for VSConstantBuffer.
Add support for depthmasking to EmulateZbuffer, previous old code had
support but wasn't ported properly with the new code a few years back.
VS Constant buffer is now properly setup.
* Add Z clamping to GSdx SW mode
* Fix spacing
* Only clamp when max vertex z is greater than zfmt max
* Fix Z format switches
* Get rid of needless shuffle
* Whoops, missed a case
* Replace switches with a shift
* Disable triangle clamping for SSE2
* Implement clamping on GS Raster Interpreter
* Added SSE2 Triangle Z clamping by KrossX
- Always enable sw colclip for BLEND_NO_BAR.
Free blending and it's faster than standard hdr algo when Blending
Accuracy is disabled.
- Enable sw colclip for Safe FBMASK SW emulated.
instead of hdr mode with accumulation blend so we don't use a mix of sw
and hw blending. It is more accurate.
Test case: Superman Shadows of Apokolips, fixes shadows when blending is enabled.
- Use const to get rid of some warnings.
- Free blending and it's faster than standard hdr algo when Blending
Accuracy is disabled.
- Add assert when disabling hw blending, hw blending should only be disabled
on free/no barrier blending. D3D11 lacks other types of blending.
- Use const to get rid of some warnings.
Fast floating point behaviour can optimise out handling for special
floating point values e.g. NaNs. As GSdx needs to handle NaNs in a few
places, precise floating point behaviour should be used instead.
Fixes a flashlight regression in Silent Hill 2/3 that was caused by
VS2019 optimising out NaN handling.
ePSXe for example.
There is little to no benefit in keeping it, especially when nobody was
maintaining the code. If we ever want to reintroduce it we can do so
from older branches like 1.6.
Plus if we want to integrate plugins into pcsx2 itself this is a required step.
PS1 games can ofc still be played within pcsx2.
Leave it as an ini option only.
People should never use custom resolution as it causes many issues and
many features aren't supported with it.
On linux option was removed a long time ago.
This reverts commit 99f814d376.
The assert gets triggered on Linux and so debug build can't be used properly.
Would be nice to investigate why in the first place but until then revert the
entire commit. We have checks for texture limits anyway so it shouldn't be an issue.
Previous behaviour:
Anisotropic filtering was continuing to run even with Nearest filtering.
On opengl it doesn't run on nearest filtering. The gui for both
renderers also greys out aniso when nearest is selected.
Anisotropic filtering being able to run on palette texture. This caused
to break the rendering on palette textures for games that use it.
PR behavior:
The PR corrects both of these behaviors, Aniso won't run on palette
textures as well as nearest filtering.
Test cases for observing the issues were used: Star Ocean 3, Fifa
Street.
Automatic, Force Disabled, Force Enabled.
Automatic - automatically determinate if a game needs the half screen
fix.
Force Disabled - fully disables the half bottom fix. It will help
Xenosaga games.
Force Enabled - fully enables the half bottom fix. Will help Lord of the
Rings Two Towers, Superman Shadows of Apokolips, Midnight Club 3.
Automatic, same as before, automatically enable based on height delta.
Force Disabled, fully disables the half bottom detection.
Force Enabled: Fully enables the half bottom detection.
Separate and group hack types in to Rendering and Upscaling hacks. This
should help users identify better what some hacks do and make it a
little bit easier to use.
Rename Automatic CRC Level label back to CRC Hack Level.
Rename Accurate Date label to DATE Accuracy.
Rename Accurate Blending label to Blending Accuracy.
Remove 10x and 12x upscaling options, we don't want to hit the maximum
texture size limit and cause issues so better play it safe and allow
only up until 8x upscaling.
Reword the tooltip to make it a bit more fitting with it's current
behavior.
Add all games in the list that have some sort of aggressive state.
Update anisotropic filtering, autoflush sw, sparse texture, gl blending tooltips.
Xenosaga handles the half bottom as an vertex offset instead of a buffer
offset which does the effect twice.
Half bottom won't trigger a cache miss that skip the draw because it is
still the normal buffer but with a vertices offset.
Issue #3147
Previous skip value 6 wasn't enough and caused some visual issues on the
ntsc version. Use value of 8 instead.
Issue reported by ozzie212121 on discord.
Effect is properly rendered with depth emulation, so far only Burnout
Revenge triggered the crc hack from my testing.
Triggering the draw is done when the player car crashes.
Partially replace the Linux ini code so that it:
- works on both Windows and Linux
- is backwards compatible with older GSdx inis
- properly handles UTF-8 ini pathnames
Fix an issue with GSdx on Windows not being able to handle non-ASCII
ini pathnames.
Fixes eyes rendering in Jak games both ingame and in cutscenes.
Previous method yielded no eyes in Jak 2/3/X cutscenes.
- Generalized offset search logic with caching system (works for BW > 1
and any PSM, limited to PSMCT32 for now)
- CRC Flags mechanism for default behavior enabling in Jak games
Fixes eyes blinking in Jak games with tex in rt search
enabled.
- Target TEX0 in Source is Source TEX0 if the Source is
not build from Target, otherwise it is the Target TEX0,
allowing improved Source invalidation
Add XgKickHack gamefix for Dog of Bay., fixes corrupt textures. Issue
#3041
Add some crc ids. Big Mutha JP version(Bakusou Convoy Densetsu - Otoko
Hanamichi America Roman),
Jak X Public Beta v.1 US, Sly 3 EU Prototype.
Developed a new SW sprite render function, which is used to fix
Jak and Daxter colors in Jak games and to remove the old Jak games hack
which yielded wrong colors and clashed with TC search texture in RT function.
- Remove OO_JakGames hack for palette readback from framebuffer (slow,
clashes with texture in RT TC function, clashes with Target readback logic)
- Developed new method SwSpriteRender with CPU rendering logic
- Add OI_JakGames method using SwSpriteRender for CPU palette rendering (fast,
does not clash with texture in RT TC function, no readback required)
Bonus: fps also sees an improvement in Jak games 3-5 increase.
It will allow to render the shadows properly.
Rendering pattern is:
Save RG channel with a kind of a TS (replaced by a copy in this hack),
compute shadow in RG,
save result in alpha with a TS,
Restore RG channel that we previously copied to render shadows.
Seems like Texture shuffle/depth is not handled properly in all cases so
revert the change until we fix it properly. Move it to gl level too.
Commit:
11cd6b56cd
With the sprite hack removed it was no longer possible to toggle this
hack on nvidia gpus, the toggle has been moved to Disable Safe Features.
It some issues caused by the hack in sotc, fatal frame and border offset
issues.
The adapter ID was needed to be passed before for a legacy DX9 hack which is no longer present, I figured it could cause a potential conflict when the adapter value stored at INI is outdated. (which is now a possibility after
lightning removed reference device)
The other alternative would be to just force set the INI value on dialog initialization to avoid unavailable values in the INI file, but that would be a rough for a person transitioning from debug build to release build using the reference device
option, so just removing this outdated variable for now, I rather doubt we'd be needing it in the future, in case we need adapter passing to subdialogs, the former suggestion needs to be implemented. Not implementing it right now since there's no need for it.
It is quite complex to handle rescaling on ComputeFixedTEX0
so this function is less stricter than GetSizeFixedTEX0,
therefore we remove the reduce optimization,
and we don't handle bilinear filtering which might create wrong interpolation at the border.
Fixes FFX upscaling issues on Bilinear filter during cutscenes.
comment
lightningterror