Computer Graphics

Software/hardware
Visualisation
Windowing and clipping
Lighting and shading
Hidden surface and solid removal
Color models
CAPABILITIES FOR MODELING IN A TYPICAL GRAPHICS
WORKSTATION ENVIRONMENT
a. 2D and 3D modeling and drafting facilities
b. Display of text
c. Animation
d. Color selection (usually a palette of 256 colors)
e. 3D viewing
f. Z-buffering
g. Gouraud shading
h. Depth cueing
i. Lighting models
j. Anti-aliasing
k. Texture mapping
ARCHITECTURE OF A TYPICAL GRAPHICS WORKSTATION
CPU board with a RISC (Reduced
Instruction Set Computing)
Microprocessor with Cache Memory
and 4/8 GB RAM. communication facilities
HDD 1 TB

display generation and

manipulation.

A color (RGB) monitors with

19" screen size and resolution
of 1280 x1024 pixels is the
standard. Some workstations
will have monitors with
resolution of 1920 x 1200
pixels
Workstations consist of three basic
components
A primary processor
Associated memory
Graphics display system and software

Leading workstations like SUN, IBM, Silicon

Graphics, Hewlett-Packard etc., use special
architecture based on RISC (Reduced
Instruction Set Computing) chips and
specialized graphic boards.
The graphic features like display manipulation are realized in
hardware for fast response. Workstations typically place emphasis
on graphics display and manipulation.
The features of a display system are directly related to Functions such
as color fill; entity generation and transformation of displays are
controlled by the display processor and associated hardware.
CAD systems are no more than computer programs. The software comprises a
number of different elements or functions that process the data stored in the
database in different ways.
Main Functions of CAD Systems
1. Model definition: for example to add geometric elements to a model
2. Model manipulation: to move, copy, delete, edit or otherwise modify elements in
the design model.
3. Picture generation: to generate images of the design model on a computer screen
or on some hard-copy device.
4. User interaction: to handle commands input by the user and to present output to
the user about the operation of the system.
5. Database management: for the management of the files that make up the DB.
6. Applications: these elements of the software do not modify the design model, but
use it to generate information for evaluation, analysis or manufacture.
7. Utilities: a “catch-all” term for parts of the software that do not directly affect the
design model, but modify the operation of the system in some way (for example, to
select the color to be used for display, or the units to be used for construction of a
drawing).

These features may be provided by multiple programs operating on a common

database, or by a single program encompassing all of the elements.
What is GPU Computing?
GPU computing is the use of a GPU (graphics processing unit) as a co-processor to
accelerate CPUs for general-purpose scientific and engineering computing.

The GPU accelerates applications running on the CPU by offloading some of the
compute-intensive and time consuming portions of the code.
The rest of the application still runs on the CPU.

From a user’s perspective, the application runs faster because it’s using the massively
parallel processing power of the GPU to boost performance. This is known as
“heterogeneous” or “hybrid” computing.

A CPU consists of four to eight CPU cores, while the GPU consists of hundreds of smaller
cores. Together, they operate to crunch through the data in the application. This
massively parallel architecture is what gives the GPU its high compute performance.

There are a number of GPU-accelerated applications that provide an easy way to access
high-performance computing (HPC).
Application developers harness the performance of the parallel GPU architecture
using a parallel programming model invented by NVIDIA called “CUDA.” All NVIDIA
GPUs—GeForce®, Quadro®, and Tesla®--support the NVIDIA® CUDA® parallel-
programming model.

WHAT IS CUDA?

CUDA™ is a parallel computing platform and programming model invented by NVIDIA.

It enables dramatic increases in computing performance by harnessing the power of
the graphics processing unit (GPU).

Identify hidden plaque in arteries: Heart attacks are the leading cause of death worldwide.
Harvard Engineering, Harvard Medical School and Brigham & Women's Hospital have teamed up
to use GPUs to simulate blood flow and identify hidden arterial plaque without invasive imaging
techniques or exploratory surgery.

The latest Tesla 20-series GPUs are based on the latest implementation of the CUDA platform
called the “Fermi architecture ”. Fermi has key computing features such as 500+ gigaflops of IEEE
standard double-precision floating-point hardware support, L1 and L2 caches, ECC memory error
protection, local user-managed data caches in the form of shared memory dispersed throughout
the GPU, coalesced memory accesses, and more.
What are the difference between Cloud, Cluster and Grid Computing?
All three systems are distributed and share similar characteristics.

The similarities relate to resource pooling and broad network access – two criteria that are
fulfilled by all systems.

Network access to cluster and grid computing systems usually takes place within a corporate
network, while the services of a cloud computing system can also be accessed through public
network, i.e. the Internet.

Resources in grid and cluster environments are generally pre-reserved, while cloud computing
systems are demand driven, i.e. operation of these systems is geared to consumers’ actual needs.

Another difference concerns the “rapid elasticity” criterion, which forms an integral part of cloud
computing systems but is not normally supported by cluster or grid systems.

Compared to other distributed systems such as grids or clusters, cloud computing solutions give
enterprises significantly more flexibility. They can dispense with IT infrastructures of their own
and only have to pay for the resources and services they actually use (“pay-per-use”/ “pay as you
go)
Two interactive operations that carried out in CAD:

* Redrawing of the display to “clean-up” unwanted clutter on the screen

or to restore parts of the image that have been corrupted.

*Selection of entities from the screen.

In each of these cases, work is not just done with the base entity data, but with a
display file that stores the displayable vectors for an entity.

For example, in the selection of a surface without use of a display file it would be
necessary to recompute the surface display in order to identify which path is nearest
to the cursor. With a display file it is only necessary to find the nearest displayed
vector and to cross-reference back to the entity.

Display files also lend themselves to fast image manipulation or zoom facilities.
Instead of recomputing the entire image for a zoom within existing window boundary,
the display file vectors are used- leading to some loss in display resolution for curves,
but generally faster display control.

The display file is related to the rest of the data by cross-referencing between the
display file and the entity table.
Graphics system consists of four subsystems:
a. Geometry engine subsystem
b. Scan conversion subsystem
c. Raster subsystem
d. Display subsystem

GEOMETRY ENGINE
• The geometry engine accepts 3-D world co-ordinate data
and converts them into X, Y screen co-ordinates.
• Depth information is manipulated using Z-buffer.
• Colors are also processed.
• The geometry pipeline facilitates among other functions
lighting, clipping,and 3D to 2D projection, viewing
transformations, rotation, scaling and translation.
SCAN CONVERSION
The scan conversion subsystem carries out
polygon decomposition
edge slope calculations
span slope calculations and span interpolation

The output. of the scan

conversion is the
pixel information to the raster
subsystem
RASTER SUBSYSTEM
• The raster subsystem will have usually 24 bit planes.
This will provide eight bit planes for each primary
color (RGB)
• In a typical raster engine five 256K X 4D RAM provide
4 bits of Z-depth.
• The raster information is stored in the frame buffer.
• Twenty 64 K X 4 video RAM provide 4 bits for each
pixel of 1280 X 1024 resolution.
• Entry level systems will have 12 bit planes and high
end systems will have 32 bit planes for the frame
buffer.
DISPLAY SUBSYSTEM
The display subsystem has multi-mode graphics
processors which manage the display, send the Red,
Green, Blue color (RGB) data to the respective
digital to analog converters as well as provide a
video output.
 The display devices are classified into two groups:
i. Display devices based on CRT principle
ii. Flat screens

In a color CRT, there are three electron guns, one each for red, green
and blue.
The phosphor dots for red, green and blue are arranged in a triangular
pattern. The individual beams intersect at a shadow mask which directs
a red beam to a red phosphor dot and so on.
Raster Scan technique
• A frame buffer is a common method of implementing a raster CRT
graphics device.
• It can be considered to be a large contiguous piece of computer
memory. In the simplest , there can be one memory bit for each
pixel in the raster. This amount of memory is called a bit plane.
• The picture is built in the frame buffer one bit at a time. The
memory bits can be either in 0 or 1 state.
• If a particular pixel is to be addressed, the corresponding bit in the
frame buffer is changed from 0 to 1.
• Since the raster CRT is an analog device, it requires an electrical
voltage and the digital data of the frame buffer has to be converted
to an analog voltage through a digital to analog converter (DAC).

A four bit plane frame buffer

A Single bit frame buffer
FLAT SCREENS
• CRT has the disadvantage that it is extremely
bulky. Moreover, CRT consumes considerable
power with increased heat dissipation
requirements.
• Portability is reduced because of the size and the
material of CRT.
• These disadvantages have prompted the
manufacturers try different types of flat screens
as output devices for computers.
• Flat screens are necessary for laptop, notebook
and palm top computers. Flat screens operate on
the principle of liquid crystal display/LED
We can separate flat-panel displays into two categories: emissive
displays and nonemissive displays.

The emissive displays (or emitters) are devices that displays, and light-
emitting diodes are examples of emissive displays.

Nonemissive displays( or nonemitters) use optical effects to convert

sunlight or light from some other source into graphics patterns. The
most important example of a nonemissive flat-panel display is a
liquid- crystal device.
emissive device is the light-emitting diode (LED). A matrix of diodes is
arranged to form the pixel positions in the display, and picture
defination is stored in refresh buffer. As in scan- line refreshing of a
CRT, information is read from the refresh buffer and converted to
voltage levels that are applied to the diodes to produce the light
patterns in the display.
Liquid- crystal displays (LCDs) are commonly used in systems, such
as calculators (fig.below) and portable, laptop computers (fig.below).
These nonemissive devices produce a picture by passing polarized
light from the surrounding or from an internal light source through a
liquid- crystal material that can be aligned to either block or transmit
the light.