Graphics Pipeline

Goals

Understand the difference between inverse-mapping and forward-mapping approaches to computer graphics rendering

Be familiar with the graphics pipeline From transformation perspective From operation perspective

Approaches to computer graphics rendering

Ray-tracing approach Inverse-mapping approach: starts from pixels A ray is traced from the camera through each pixel Takes into account reflection, refraction, and diffraction in a

multi-resolution fashion High quality graphics but computationally expensive

Not for real-time applications Pipeline approach

Forward-mapping approach Used by OpenGL and DirectX State-based approach:

Input is 2D or 3D data Output is frame buffer Modify state to modify functionality

For real-time and interactive applications, especially games

Ray-tracing – Inverse mapping

For every pixel, construct a ray from the eye for every object in the scene intersect ray with objectfind closest intersection with the raycompute normal at point of intersectioncompute color for pixel shoot secondary rays

Pipeline – Forward mapping

Start from the geometric primitives to find the values of the pixels

The general view (Transformations)

Modeling Transformation

Lighting

Viewing Transformation

Projection Transformation

3D scene, Camera 3D scene, Camera Parameters, and Parameters, and light sourceslight sources

graphics Pipelinegraphics Pipeline FramebuffeFramebufferr

DisplayDisplay

Clipping

Rasterization

ViewportTransformation

Input and output of the graphics pipeline Input:

Geometric model Objects Light sources geometry and transformations

Lighting model Description of light and object properties

Camera model Eye position, viewing volume

Viewport model Pixel grid onto which the view window is mapped

Output: Colors suitable for framebuffer display

Graphics pipeline

What is it?The nature of the processing steps to display a computer

graphic and the order in which they must occur. Primitives are processed in a series of stages Each stage forwards its result on to the next stage The pipeline can be drawn and implemented in different

ways Some stages may be in hardware, others in software Optimizations and additional programmability are

available at some stages Two ways of viewing the pipeline:

Transformation perspective Operation perspective

Modeling transformation

3D models defined in their own coordinate system (object space)

Modeling transforms orient the models within a common coordinate frame (world space)

Modeling Transformation

Lighting

Viewing Transformation

Projection Transformation

Object space World space

Clipping

Rasterization

ViewportTransformation

Lighting (shading)

Vertices lit (shaded) according to material properties, surface properties (normal) and light sources

Local lighting model (Diffuse, Ambient, Phong, etc.)

Modeling Transformation

Lighting

Viewing Transformation

Projection Transformation

Clipping

Rasterization

ViewportTransformation

Viewing transformation

It maps world space to eye (camera) space

Viewing position is transformed to origin and viewing direction is oriented along some axis (usually z)

Modeling Transformation

Lighting

Viewing Transformation

Projection Transformation

Clipping

Rasterization

ViewportTransformation

Projection transformation(Perspective/Orthogonal)

Specify the view volume that will ultimately be visible to the camera

Two clipping planes are used: near plane and far plane

Usually perspective or orthogonal

Modeling Transformation

Lighting

Viewing Transformation

Clipping

Rasterization

Projection Transformation

ViewportTransformation

Clipping

The view volume is transformed into standard cube that extends from -1 to 1 to produce Normalized Device Coordinates.

Portions of the object outside the NDC cube are removed (clipped)

Modeling Transformation

Lighting

Viewing Transformation

Projection Transformation

Rasterization

Clipping

ViewportTransformation

Viewport transformation (to screen space)

Maps NDC to 3D viewport: xy gives the screen window z gives the depth of each point

Modeling Transformation

Lighting

Viewing Transformation

Projection Transformation

Rasterization

Clipping

ViewportTransformation

Rasterization (scan conversion)

Rasterizes objects into pixels Interpolate values as we go

(color, depth, etc.)

Modeling Transformation

Lighting

Viewing Transformation

Clipping

Projection Transformation

Rasterization

ViewportTransformation

Summary of transformations

glMatrixMode(GL_MODELVIEW);

//glTranslate; glRotate; glScale

//gluLookAt

glMatrixMode(GL_PROJECTION);

//glTranslate; glRotate; glScale

//gluPerspective

//glFrustrum

glViewPort(0, 0, w, h);

DirectX transformations

World transformation Device.Transform.World

= Matrix.RotationZ(…) OpenGL does not have one

View transformation: Device.Tranform.View

= Matrix.LookAtLH(…)

Projection transformation: device.Transform.Projection

= Matrix.PerspectiveFovLH

OpenGL pipeline (operations)

OpenGL pipeline Display list A group of OpenGL commands that have been stored (compiled) for later

execution. Vertex and pixel data can be stored/cached in a display list. (Why?) Vertex Operation

Each vertex and its normal coordinates are transformed by GL_MODELVIEW matrix from object coordinates to eye coordinates. When lighting is enabled, the lighting calculation of a vertex is performed using the transformed vertex and normal data; thus producing new color for the vertex.

Primitive AssemblyThe geometrical primitives are transformed by projection matrix then clipped by viewing volume clipping planes. After that, viewport transform is applied in order to map 3D scene to screen space coordinates. Lastly, if culling is enabled, the culling test is performed.

Pixel Transfer OperationUnpacked pixels may undergo transfer operations such as scaling, wrapping, and clamping. The transferred data are either stored in texture memory or rasterized directly to fragments.

OpenGL pipeline Texture Memory

Texture images are loaded into texture memory to be applied onto geometric objects. Rasterization

The conversion of both geometric and pixel data into fragment. Fragments obtained form a rectangular array containing color, depth, line width, point size, and anti-aliasing calculations (GL_POLYGON_SMOOTH). When requested, the interior pixels of a polygon will be filled. A fragment corresponds to a pixel in the frame buffer.

Fragment OperationFragments are converted to pixels onto frame buffer. The first step in this stage is to generate a texture element, texel, from texture memory and apply it to each fragment. Fog calculations are then performed. When enabled, several fragment tests are performed in the order: Scissor Test , Alpha Test, Stencil Test, and Depth Test. Finally, blending, dithering, logical operation, and masking by bitmasks are performed and actual pixel data are stored in frame buffer.

FeedbackUsed to get OpenGL’s current states and information (glGet*() and glIsEnabled() commands are just for that). A rectangular area of pixel data from frame buffer can be read using glReadPixels(). Fully transformed vertex data can be obtained using the feedback rendering mode and the feedback buffer.

Zoom into OpenGL pipeline (see the OpenGL bluebook)

Summary of operations

Fixed Pipeline

Programmable PipelineHigh-level Shading Language