8/8/2019 gpuppt2003
1/24
GRAPHICS
PROCESSINGUNIT
Submitted by:--BRIJESH KUMAR PATEL
BT CS
0709510012
8/8/2019 gpuppt2003
2/24
A Graphics Processing Unit (GPU) is a
microprocessor that has been designed
specifically for the processing of 3D graphics.
Main purpose of GPU is to simulate the 3D
images as realistic as possible on the computer
screen
8/8/2019 gpuppt2003
3/24
GPU is mainly needed to relieve the CPUfrom graphical computations so that CPU
can be used for other processes
8/8/2019 gpuppt2003
4/24
CPU is where all the
program instructions are
executed . CPU performs less
computation than
GPU(150G flops)(floating
point operations per sec)
GPU is a dedicated piece of
hardware that processes graphic.
GPU performs high intensive
computing & massive parallel
processing (500G flops) (floating
point operations per sec)
8/8/2019 gpuppt2003
5/24
The first GPU, introduced in 1981 by IBM, weremonochrome cards designated as Monochrome DisplayAdapters (MDAs).
Then came Colour Graphics Adapter(CGA) & then theEnhanced Graphics Adapter(EGA).
IBM introduced the Video Graphics Array (VGA) in1987. It could support up to 256 colors at resolutions up
to 720x400
Then came the Super Video Graphics Array (SVGA)that supports upto 16.8 million colors and 1280x1024resolution
8/8/2019 gpuppt2003
6/24
Graphics HardwareInterface
PERIPHERAL COMPONENT INTERCONNECT(PCI)
ACCELERATED GRAPHICS PORT (AGP)
PERIPHERAL COMPONENT INTERCONNNECT-EXPRESS(PCI-E)
8/8/2019 gpuppt2003
7/24
8/8/2019 gpuppt2003
8/24
Graphics Processor
The processor is designed specifically toperform floating point calculations, which arefundamental to 3D graphics processing.
Graphics Accelerator
A graphics accelerator assists graphicsprocessing by executing instructionsconcurrently.
Frame Buffer
A frame buffer is a video output device thatdrives a video display from a memory buffercontaining a complete frame of data.
8/8/2019 gpuppt2003
9/24
Memory
Video memory may be used for storing screenimage as well as Z-Buffer which manages thedepth coordinates in 3D graphics.
Graphics BIOS
This contains the basic program, which isusually hidden, that governs the video card's
operations and provides the instructions thatallow the computer and software to interactwith the card.
8/8/2019 gpuppt2003
10/24
Digital-to-Analog Converter (DAC)
The RAMDAC or Random Access Memory Digital-to-Analog Converter, converts digital signals toanalog signals for use by a computer display thatuses analog inputs such as CRT displays.
Display Connector
This include connection system b/w the video cardand the display devices such as monitor or atelevision.
8/8/2019 gpuppt2003
11/24
Everything that is displayed on the
computer screen is 2D
The push for more realism, more finely-
detailed graphics and faster speeds in such
programs as means that more 3D work must be done in a shorter period of time has
resulted in need for 3D acceleration
8/8/2019 gpuppt2003
12/24
REAL IMAGE GRAPHICS
8/8/2019 gpuppt2003
13/24
8/8/2019 gpuppt2003
14/24
PERFORMANCE FACTORS OF GPU
Fillrate
The fillrate usually refers to the number of pixelsa video card can render and write to video
memory in a second.
Memory Bandwidth
Memory bandwidth is the rate at which data canbe read from or stored into a memory by aprocessor.
8/8/2019 gpuppt2003
15/24
Memory Clock
This tells us the amount of memory bandwidtha graphics card has.
Memory Interface (Memory Bus)
The larger the Memory Interface width, fasterthe speed of data traveling in it.
Core ClockThe actual speed at which the graphicsprocessor on a video card operates.
8/8/2019 gpuppt2003
16/24
STREAM PROCESSING
A stream is simply a set of records that require similarcomputation.
A technique used to accelerate the processing of manytypes of video and image computations is called streamprocessing.
Streams provide data parallelism
GPUs are stream processors processors that can operate inparallel by running a single kernel on many records in a stream
at once.
Kernels are the functions that are applied to each element in thestream.
8/8/2019 gpuppt2003
17/24
It is a set of routines, protocols, and tools forbuilding software applications
GLide by 3dfx
OpenGL by Silicon Graphics
Direct3D by Microsoft
API (Application Program
Interface)
8/8/2019 gpuppt2003
18/24
DEDICATED GPU
INTEGRATED GRAPHICS SOLUTIONS
HYBRID SOLUTIONS
TYPES OF GPU
8/8/2019 gpuppt2003
19/24
MANUFACTURERS OF GPU AMD
Matrox Graphics
Nvidia
S3 Graphics
Intel
SiS
PowerV
8/8/2019 gpuppt2003
20/24
ADVANTAGE OF GPU
Scattered reads code can read from arbitrary addresses inmemory.
Shared memory CUDA exposes a fast shared
memory region (16KB in size) that can be shared amongstthreads. This can be used as a user-managed cache,enabling higher bandwidth than is possible using texturelookups.
Faster downloads and read backs to and from the GPU
Full support for integer and bitwise operations, includinginteger texture lookups.
8/8/2019 gpuppt2003
21/24
1. System Design
The processing capacities and capabilities of the
GPU. The GPUs internal banbwidth.
GPU-CPU communication bandwidth.
If the system is part of a processing network, theexternal network bandwidth.
DISADVANTAGE OF GPU
8/8/2019 gpuppt2003
22/24
2.Techinal limitations:- Lack of fundamental computing constructs.
lack of High Quality Random Numbers.
Lack of Double Precision. Reduced Capability of Scatter Operations.
Expensive Branching
8/8/2019 gpuppt2003
23/24
CONCLUSION
From the introduction of the first GPUs in the1970s to the most recent ones manufacturedtoday, the present world of graphics has changed
enormously and would have never been the samewithout it.
Today lot of applications have become fasterand efficient by using GPU technology thussaving lot of time in many scenarios.
8/8/2019 gpuppt2003
24/24