- Don't create several persistent screen-sized images and buffers for screenshots, only create one temporary buffer during a frame that requests a screenshot.
- Attempt to handle more out-of-memory situations.
- Make sure the final screenshot is opaque.
- Swap R and B channels in the screenshot ImageData when the backbuffer is BGRA.
Add love.graphics.newTextureView(basetexture, viewsettings). Requires GLSL4 support.
Add 'viewformats' boolean setting to texture setting tables.
Add Texture:hasViewFormats.
The viewsettings table has the following fields:
format = nil, -- set this to a pixel format to override the base format. It must have the same number of bytes per pixel as the original, and the 'viewformats' setting on the original texture must be true for this to work. Cannot be used for compressed or depth/stencil textures.
type = nil, -- set this to a texture type to override the base texture type. 2d base textures can make [2d, array] views. [array, cube] base textures can make [2d, array, cube] views.
mipmapstart = nil, -- set to a number to use a less detailed mipmap level as a base.
mipmapcount = nil, -- set to a number to override the number of mipmaps in the texture view.
layerstart = nil, -- set to a number to use a specific layer or cube face as a base.
layers = nil, -- set to a number to override the number of layers in an array texture view.
- Setting the depth buffer for the backbuffer is now done as a boolean instead of a bit depth integer, in love.conf or love.window.setMode.
- Error if depth writes are enabled when the active canvas setup or backbuffer does not have a depth buffer (now it matches stencil behaviour).
- Don't allocate a backbuffer depth-stencil buffer if they're not requested. Metal also determines the format for that buffer based on which combination of depth and stencil is requested. OpenGL still has to allocate the depth-stencil buffer for the backbuffer all the time, because changing it requires the context to be recreated.
Add new higher level love.graphics.setStencilMode function.
- Add love.graphics.setStencilMode(mode [, value = 1])
- Add love.graphics.setStencilMode().
- Add mode, value = love.graphics.getStencilMode().
The possible modes are "off" (same as no parameters), "draw", and "test".
The "draw" mode disables color writes and sets the stencil state to ("replace", "always") using the given value.
The "test" mode enables color writes and sets the stencil state to ("keep", "equal") using the given value to compare to.
Sometimes no buffer for tex coords are given, however the shader
still expects some. We can fix this by using a default buffer, which
contains the constant tex coords (0, 0), which is then bound instead.
There is little to no performance gain, with a lot of burden
to maintenance of the extra path. It's better to unify the code
with the other render pass logic.
Before this patch some common patterns would lead to very suboptimal
render passes.
Consider the following code:
```lua
local canvas = love.graphics.newCanvas(...)
function love.draw()
love.graphics.setCanvas(canvas)
love.graphics.clear(...)
love.graphics.draw(...)
love.graphics.setCanvas()
love.graphics.draw(canvas)
end
```
This would lead to the following rendering:
1) render pass on main window with loadOp=load, followed by an immediate
call to vkCmdClearAttachments
2) render pass on canvas with loadOp=load, followed by an immediate call
to vkCmdClearAttachments
3) render pass on main window with loadOp=load
This patch changes the behaviour to the more performant
(and equivalent) version:
1) render pass on canvas with loadOp=clear
2) render pass on main window with loadOp=clear
This is especially helpful on mobile devices, where creating render
passes is an expensive operation.
Resolves#1879.
- Add an 'indirectdraw' boolean field to the graphics feature support table returned by love.graphics.getSupported. This is almost always supported when compute shaders are supported, except on a few older phones.
- Add an 'indirectarguments' boolean field to the settings table in love.graphics.newBuffer.
- Add love.graphics.dispatchIndirect(shader, argumentsbuffer [, argumentsindex = 1]).
The compute dispatch's threadgroup width, height, and depth values are fetched from the buffer (as 3 uints) instead of coming from function parameters.
- Add love.graphics.drawIndirect(mesh, argumentsbuffer, argumentsindex, x, y, ....).
Vertex or index count, instance count, and other related parameters for drawing the mesh are fetched from the buffer (as 4 or 5 uints depending on whether the mesh has an index buffer) instead of coming from function parameters. It's usually a good idea to keep parameters other than the instance count in sync with what the mesh should be using.
- Add love.graphics.drawFromShaderIndirect(drawmode, argumentsbuffer [, argumentsindex = 1] [, maintexture = nil]) and drawFromShaderIndirect(indexbuffer, argumentsbuffer [, argumentsindex] [, maintexture = nil]).
Vertex or index count, instance count, and other related parameters are fetched from the buffer as 4 or 5 uints, as above.
For the dispatch indirect arguments buffer, it has to have 3 uint32 elements: { uint threadgroupsX, uint threadgroupsY, uint threadgroupsZ }.
For non-indexed draws, the arguments buffer has to have 4 uint32 elements: { uint vertexCount, uint instanceCount, uint baseVertex, uint baseInstance }. Note that baseInstance should always be set to 0 as many drivers don't support non-zero values.
For draws which use an index buffer, the arguments buffer has to have 5 uint32 elements: { uint indexCount, uint instanceCount, uint firstIndex, uint baseVertex, uint baseInstance }. As above, the baseInstance value should always be 0.
A buffer can be created to have an array of those structures, which can be used with the argumentsindex parameter of the Indirect dispatch/draw functions.
This is not actually needed. The api version specified when creating
the vulkan instance should be the highest api version that might
be used by the application, which can be higher than the supported
instance version. VMA can make use of vulkan 1.3 features so this
can be hardcoded to VK_API_VERSION_1_3.