summaryrefslogtreecommitdiffstats
path: root/tests/manual/shadow-map-qml/doc/src/shadow-map-qml.qdoc
diff options
context:
space:
mode:
Diffstat (limited to 'tests/manual/shadow-map-qml/doc/src/shadow-map-qml.qdoc')
-rw-r--r--tests/manual/shadow-map-qml/doc/src/shadow-map-qml.qdoc252
1 files changed, 252 insertions, 0 deletions
diff --git a/tests/manual/shadow-map-qml/doc/src/shadow-map-qml.qdoc b/tests/manual/shadow-map-qml/doc/src/shadow-map-qml.qdoc
new file mode 100644
index 000000000..29cb4d391
--- /dev/null
+++ b/tests/manual/shadow-map-qml/doc/src/shadow-map-qml.qdoc
@@ -0,0 +1,252 @@
+// Copyright (C) 2015 Klaralvdalens Datakonsult AB (KDAB).
+// SPDX-License-Identifier: LicenseRef-Qt-Commercial OR GFDL-1.3-no-invariants-only
+
+/*!
+ \example shadow-map-qml
+ \title Qt 3D: Shadow Map QML Example
+ \ingroup qt3d-examples-qml
+
+ \brief A Qt 3D QML application that illustrates how to render a scene in Qt 3D
+ with shadows.
+
+ \image shadowmapping-qt3d.png
+
+ \e {Qt 3D Shadow Map} illustrates how to configure the renderer in order to
+ accommodate custom rendering techniques. The example application displays a
+ self-shadowed plane and trefoil knot.
+
+ We implement \l{Shadow Mapping}{shadow mapping} using a two pass rendering.
+ In the first pass, we generate the shadow information. In the second pass,
+ we generate the scene using the forward rendering technique with Phong
+ shading, while at the same time using the information gathered in the first
+ pass to draw the shadows.
+
+ The entire rendering is configured using QML, but it is possible to use C++
+ to achieve the very same result.
+
+ \include examples-run.qdocinc
+
+ \section1 Setting Up the Scene
+
+ We set up the entire scene in the \e main.qml file.
+
+ To be able to use the types in the Q3D and Q3D Renderer modules, we must
+ import the modules:
+
+ \quotefromfile shadow-map-qml/main.qml
+ \skipto import Qt3D.Core
+ \printuntil Render 2.0
+
+ The first entities we create are a \l Camera, which represents the camera
+ used for the final rendering, and a \l Configuration, which allows us to
+ control this camera using the keyboard or the mouse:
+
+ \printuntil }
+ \printuntil }
+
+ We then create a Light custom entity, which represents our light. It is a
+ directional spotlight, placed somewhere above the plane and looking down at
+ the scene’s origin:
+
+ \printuntil }
+
+ This light entity is used by our custom frame graph, ShadowMapFrameGraph,
+ and our rendering effect, AdsEffect, whose instances are created just after
+ the light:
+
+ \printuntil ]
+ \printuntil }
+
+ Last, we create three entities for the meshes in the scene: a trefoil knot,
+ a toy plane, and a ground plane. They aggregate a mesh, a transformation,
+ and a material that uses the AdsEffect. The toy plane and the trefoil knot
+ transformations are animated:
+
+ \printuntil /^\}/
+
+ \section1 Specifying the Light
+
+ We specify the Light custom entity in \e ShadowMapLight.qml.
+
+ Again, we import the necessary modules:
+
+ \quotefromfile shadow-map-qml/ShadowMapLight.qml
+ \skipto import Qt3D
+ \printuntil Qt3D.Render
+
+ We then use an \l Entity type as the root element of the custom QML type.
+ The light is a directional spotlight that exposes as properties a position,
+ intensity, and a 4×4 transformation matrix:
+
+ \printuntil matrix4x4
+
+ In the first rendering pass, we use the light as a camera, and therefore we
+ use a \l Camera entity within the light and expose it as a property:
+
+ \printuntil /^\}/
+
+ \section1 Configuring the Framegraph
+
+ In Qt 3D, the frame graph is the data-driven configuration for the rendering.
+ We implement the frame graph in the \e ShadowMapFrameGraph.qml file.
+
+ In addition to the Qt 3D and Qt 3D Render modules, we also import the
+ Qt Quick module:
+
+ \quotefromfile shadow-map-qml/ShadowMapFrameGraph.qml
+ \skipto import QtQuick
+ \printuntil Render 2.0
+
+ The code defines a \l RenderSettings node that has a tree of nodes as the
+ active frame graph:
+
+ \badcode
+ RenderSettings {
+ activeFrameGraph: Viewport {...}
+ }
+ \endcode
+
+ Any path from the leaves of this tree to the root is a viable frame graph
+ configuration. Filter entities can enable or disable such paths, and
+ selector entities can alter the configuration.
+
+ In our case, the tree looks like this:
+
+ \badcode
+ Viewport
+ RenderSurfaceSelector
+ RenderPassFilter
+ RenderTargetSelector
+ ClearBuffers
+ CameraSelector
+ RenderPassFilter
+ ClearBuffers
+ CameraSelector
+ \endcode
+
+ So we have two paths from the topmost \l Viewport entity. Each path
+ corresponds to a pass, or phase, of the shadow map technique. The paths are
+ enabled and disabled using a RenderPassFilter, a node that can filter
+ depending on arbitrary values defined in a given render pass. In this
+ example, it is a string:
+
+ \skipto RenderPassFilter
+ \printuntil ]
+
+ The actual passes are not defined within the frame graph. Instead the
+ available passes are declared in the Materials used in the scene graph. The
+ frame graph is only used to select which passes are used when rendering.
+
+ \section1 Generating the Shadow Map
+
+ In the shadow map generation pass, we must render to an offscreen surface
+ (Framebuffer Object) which has a depth texture attachment. In Qt 3D, it is
+ represented by the RenderTarget entity, which has a number of attachments.
+
+ In this example, we need only a depth attachment. We define it as a
+ RenderAttachment entity using the RenderAttachment.DepthAttachment \c type
+ that stores the depth and a Texture2D entity that actually configures the
+ exture storage used to store the depth information:
+
+ \printuntil ]
+ \printuntil }
+
+ Moreover, in this first pass, we must render using the light’s camera.
+ Therefore, we have a CameraSelector entity that sets the camera to the one
+ exported by the Light:
+
+ \skipto CameraSelector
+ \printuntil }
+
+ The second pass is more straightforward, because we simply render to the
+ screen using the main camera:
+
+ \skipto RenderPassFilter
+ \printuntil }
+ \printuntil }
+ \printuntil }
+ \printuntil }
+
+ \section1 Using Effects
+
+ The bulk of the magic happens in the \e AdsEffect.qml file, where our main
+ \l Effect is defined. It implements the Ambient, Diffuse and Specular
+ (ADS) Lighting Model using Phong shading with the addition of shadow mapping.
+
+ An effect contains the implementation of a particular rendering strategy. In
+ this example, shadow mapping using two passes:
+
+ \quotefromfile shadow-map-qml/AdsEffect.qml
+ \skipto Effect
+ \printuntil Light
+
+ The \c parameters list defines some default values for the effect. The
+ values will get mapped to shader program uniform variables, so that in the
+ shaders we can access their values. In this example, we expose some information from
+ the Light entity (position, intensity, view or projection matrix defined by
+ the internal camera) and the shadow map texture exposed by the frame graph:
+
+ \skipto parameters:
+ \printuntil ]
+
+ It is possible to put such parameters all the way down, from a \l Material,
+ to its \l Effect, to one of the effect’s \l {Technique}{Techniques} and a
+ \l RenderPass within a \l Technique. This allows a \l Material instance to
+ override defaults in an \l Effect, \l Technique or \l RenderPass.
+
+ To adapt the implementation to different hardware or OpenGL versions, we
+ could use one or more \l Technique elements. In this example, only one
+ technique is provided, targeting OpenGL 3.2 Core, or later:
+
+ \quotefromfile shadow-map-qml/AdsEffect.qml
+ \skipto techniques:
+ \printuntil }
+
+ Inside the technique, we finally have the definition of our two rendering
+ passes. We \e tag each pass with a \l FilterKey object, matching the ones
+ we specified in the frame graph configuration, so that each pass will have
+ different rendering settings:
+
+ \printuntil ]
+
+ The first pass is the shadow map generation. We load a suitable set of GLSL
+ shaders, which are actually extremely simple. They do only MVP (Model, View,
+ Projection) to bring meshes from their model space into clip space (and,
+ remember, in this first pass, the light is the camera). The fragment shader
+ is totally empty, because there is no color to be generated, and the depth
+ will be automatically captured for us by OpenGL:
+
+ \printuntil }
+
+ In this first pass, we also set some custom OpenGL state in the form of a
+ polygon offset and depth testing mode:
+
+ \printuntil ]
+
+ \section1 Rendering Using Phong Shading
+
+ The second pass is a normal forward rendering using Phong shading. The code
+ in the effect entity is extremely simple. We simply configure some
+ parameters and load a pair of shaders which will be used when drawing.
+
+ The first part of the shadow mapping happens in the vertex shader defined in
+ \e ads.vert file, where we output towards the fragment shader the
+ coordinates of each vertex in light space:
+
+ \quotefromfile shadow-map-qml/shaders/ads.vert
+ \skipto mat4(
+ \skipto positionInLightSpace
+ \printuntil ;
+
+ Actually, the coordinates get adjusted a little to allow us to easily sample
+ the shadow map texture.
+
+ The second part happens in the fragment shader defined in the \e ads.frag
+ file, where we sample the shadow map. If the currently processed fragment is
+ behind the one closest to the light, then the current fragment is in shadow
+ (and only gets ambient contribution). Otherwise, it gets full Phong shading:
+
+ \quotefromfile shadow-map-qml/shaders/ads.frag
+ \skipto main
+ \printuntil }
+*/