SlideShare une entreprise Scribd logo
1  sur  22
GRAPHIC PROCESSING UNIT
Presented by KamranAshraf
13-NTU-4009
TOPICS
Advancements
Latest Models
Challenges
Impact on
computing
CONTRIBUTORS
Three main
contributors to
advancement:
Intel AMD/ATI Nvidia
WHAT IS RENDERING
 Rendering is the process of generating an image from a 2D or 3D model by
means of computer programs
 It is the process of add shading, color and lamination in order to create life-like
images on a screen
 A GPU assists a CPU in performing complex rendering calculations
 If a scene is to look relatively realistic and predictable under virtual lighting,
the rendering software should do it
 Rendering is also used to describe the process of calculating effects in a video
editing program to produce final video output
A variety of rendering
techniques applied to a
single 3D scene
WHAT IS GPU RENDERING?
 GPU rendering means, that there is used the power of GPU (Graphics card) for
rendering, instead of CPU (processor).
 Conventional CPUs (Intel, AMD) didn't make such dramatic progress, like GPUs in
last 10 years.
GPU RENDERING
 GPU rendering is definitely the future of rendering as one single GPU has the
same power as the huge and expensive computer clusters with many CPUs
 GPU rendering is incredibly fast and less power hungry
 You will need less computers for the same performance
 Much CHEAPER
THE WORLD'S FASTEST GPU
THEWORLD'S FASTEST GPU
The GTXTitan Z The Radeon R9 295X2
AMD’S R9 295X2
 It joins two of its R9 290X GPUs onto a single board
 Max Resolution: 4096 x 2160, Support 5x Display Monitors
 Memory: 8GB dedicated
 Cores: Powered by 5,632 cores meaning 2,816 cores for each GPU
 The most advanced dual GPU liquid cooling system ever designed
 Ups the power consumption to a scarily large 500 watts
 Cost around $1500
NVIDIA GTXTITAN Z
 5k resolutions supported:This GPU supports multi-monitors and a resolution upto 5k
 Dual GPU architecture:Two Kepler GK110 GPU’s inside,Titan Z generates the same power
as if twoTitan’s were running
 Memory:The card will support a total of 12 GB dedicated memory i.e 6 GB for each GPU.
 CUDA cores:The GTXTitan Z is powered by 5,760 CUDA cores
meaning 2,880 cores for each GPU.
 It won’t make much noise as compared to other powerful GPU’s
 Cost around $3000
CALCULATE IMAGE QUALITY
USING GPU
IMAGE QUALITY CALCULATOR
 The University of Manitoba developed an Image Quality Calculator.
 This utility, which is a plugin for ImageJ, does a great job analyzing the quality
(amount of noise)
COMPARING RESULTS
The GPU (2688 CUDA cores) was 6.2 times faster than the CPU.
IMPACT OF GPU IN
COMPUTING
GPU ACCELERATED COMPUTING
 GPU-accelerated computing is the use of aGPU together with a CPU to accelerate
scientific, analytics, engineering, consumer, and enterprise applications.
 Pioneered by NVIDIA, GPU accelerators now power energy-efficient datacenters
in government labs, universities, enterprises, and small-and-medium businesses
around the world.
 GPUs are accelerating applications in platforms ranging from cars, to mobile
phones and tablets, to drones and robots.
HOW ITWORKS
 Compute-intensive portions of the application are executed by the GPU
 The remainder of the code still runs on the CPU
 From a user's perspective, applications simply run significantly faster
GPUs have thousands of cores to process parallel workloads efficiently
CHALLENGES
THE LIGHTNING CHALLENGE
 One of the most important aspects of creating convincing 3D visuals is
implementing realistic lighting
 The challenge is not having an artist manually animate realistic-looking lighting,
but to simulate how a light reacts with, and bounces off, multiple surfaces in real-
time, otherwise known as path tracing
 In the film industry, path tracing is done offline, because the calculations required
are simply too complex
 Unity global illumination worked on advanced lighting model which supported
indirect lighting but it was very expensive so It's a technology that's not quite
ready.
POWER CHALLENGES
 Biggest, most complex processors on the planet are being made
 The biggest challenge is no longer being able to build a fast GPU
 The biggest challenge is power as Graphics can consume a near infinite amount of
horsepower
 We don't want to have to watercool everything, or have 19 fans and make things
really loud.
ADVANCEMENTS
LATEST ADVANCEMENTS
Nvidia's GameWorks technology
 Nvidia's GamesWorks is a set of visual
effects, physics effects, and tools
designed to give developers more options
when it comes to a game's visuals.
 Effects like HBAO+ and FaceWorks are
part of GameWorks.
 AO can account for the occlusion of light,
creating non-uniform shadows that add
depth to the scene.
 HBAO+ doubles the detail level of the AO
effect, runs three times faster, and uses the
latest DirectX 11 technologies.
 Nvidia's GameWorks is more copyrighted
in nature.
AMD's Mantle technology
 Mantle, allows console-like low-level
access to the CPU and GPU.
 By removing some of the software layers
of the operating system between the
hardware and the software
 With the right programming developers
are able to see an increase in
performance
 AMD claims its Mantle API is open, and
free to use
Thanks

Contenu connexe

Tendances

Graphic Processing Unit (GPU)
Graphic Processing Unit (GPU)Graphic Processing Unit (GPU)
Graphic Processing Unit (GPU)Jafar Khan
 
Graphics Processing Unit - GPU
Graphics Processing Unit - GPUGraphics Processing Unit - GPU
Graphics Processing Unit - GPUChetan Gole
 
graphics processing unit ppt
graphics processing unit pptgraphics processing unit ppt
graphics processing unit pptNitesh Dubey
 
Graphics processing unit ppt
Graphics processing unit pptGraphics processing unit ppt
Graphics processing unit pptSandeep Singh
 
Graphics processing unit (GPU)
Graphics processing unit (GPU)Graphics processing unit (GPU)
Graphics processing unit (GPU)Amal R
 
What are graphics cards
What are graphics cardsWhat are graphics cards
What are graphics cardsUsman Hashmi
 
Advanced Micro Devices - AMD
Advanced Micro Devices - AMDAdvanced Micro Devices - AMD
Advanced Micro Devices - AMDVedavyas Sheoy
 
Graphic card information search pp
Graphic card information search ppGraphic card information search pp
Graphic card information search ppPoornima Shetagar
 
Parallel computing with Gpu
Parallel computing with GpuParallel computing with Gpu
Parallel computing with GpuRohit Khatana
 
CPU vs. GPU presentation
CPU vs. GPU presentationCPU vs. GPU presentation
CPU vs. GPU presentationVishal Singh
 
GPU Architecture NVIDIA (GTX GeForce 480)
GPU Architecture NVIDIA (GTX GeForce 480)GPU Architecture NVIDIA (GTX GeForce 480)
GPU Architecture NVIDIA (GTX GeForce 480)Fatima Qayyum
 
Presentation Graphic cards - History
Presentation Graphic cards - HistoryPresentation Graphic cards - History
Presentation Graphic cards - HistoryDipen Vasoya
 
10. GPU - Video Card (Display, Graphics, VGA)
10. GPU - Video Card (Display, Graphics, VGA)10. GPU - Video Card (Display, Graphics, VGA)
10. GPU - Video Card (Display, Graphics, VGA)Akhila Dakshina
 
GRAPHIC CARD
GRAPHIC CARDGRAPHIC CARD
GRAPHIC CARDVPKV
 
Video/ Graphics cards
Video/ Graphics  cardsVideo/ Graphics  cards
Video/ Graphics cardsAmandeep Kaur
 

Tendances (20)

Graphic Processing Unit (GPU)
Graphic Processing Unit (GPU)Graphic Processing Unit (GPU)
Graphic Processing Unit (GPU)
 
Graphics Processing Unit - GPU
Graphics Processing Unit - GPUGraphics Processing Unit - GPU
Graphics Processing Unit - GPU
 
graphics processing unit ppt
graphics processing unit pptgraphics processing unit ppt
graphics processing unit ppt
 
Graphics processing unit ppt
Graphics processing unit pptGraphics processing unit ppt
Graphics processing unit ppt
 
Graphics processing unit
Graphics processing unitGraphics processing unit
Graphics processing unit
 
Graphics card
Graphics cardGraphics card
Graphics card
 
Graphics processing unit (GPU)
Graphics processing unit (GPU)Graphics processing unit (GPU)
Graphics processing unit (GPU)
 
Introduction to GPU Programming
Introduction to GPU ProgrammingIntroduction to GPU Programming
Introduction to GPU Programming
 
What are graphics cards
What are graphics cardsWhat are graphics cards
What are graphics cards
 
Advanced Micro Devices - AMD
Advanced Micro Devices - AMDAdvanced Micro Devices - AMD
Advanced Micro Devices - AMD
 
Graphic card information search pp
Graphic card information search ppGraphic card information search pp
Graphic card information search pp
 
Parallel computing with Gpu
Parallel computing with GpuParallel computing with Gpu
Parallel computing with Gpu
 
CPU vs. GPU presentation
CPU vs. GPU presentationCPU vs. GPU presentation
CPU vs. GPU presentation
 
GPU Architecture NVIDIA (GTX GeForce 480)
GPU Architecture NVIDIA (GTX GeForce 480)GPU Architecture NVIDIA (GTX GeForce 480)
GPU Architecture NVIDIA (GTX GeForce 480)
 
Graphic card
Graphic cardGraphic card
Graphic card
 
Presentation Graphic cards - History
Presentation Graphic cards - HistoryPresentation Graphic cards - History
Presentation Graphic cards - History
 
10. GPU - Video Card (Display, Graphics, VGA)
10. GPU - Video Card (Display, Graphics, VGA)10. GPU - Video Card (Display, Graphics, VGA)
10. GPU - Video Card (Display, Graphics, VGA)
 
GRAPHIC CARD
GRAPHIC CARDGRAPHIC CARD
GRAPHIC CARD
 
GPU Programming
GPU ProgrammingGPU Programming
GPU Programming
 
Video/ Graphics cards
Video/ Graphics  cardsVideo/ Graphics  cards
Video/ Graphics cards
 

Similaire à Graphic Processing Unit

Graphics Processing Unit: An Introduction
Graphics Processing Unit: An IntroductionGraphics Processing Unit: An Introduction
Graphics Processing Unit: An Introductionijtsrd
 
VisionizeBeforeVisulaize_IEVC_Final
VisionizeBeforeVisulaize_IEVC_FinalVisionizeBeforeVisulaize_IEVC_Final
VisionizeBeforeVisulaize_IEVC_FinalMasatsugu HASHIMOTO
 
GPU accelerated path rendering fastforward
GPU accelerated path rendering fastforwardGPU accelerated path rendering fastforward
GPU accelerated path rendering fastforwardMark Kilgard
 
Amd future of gp us - campus party
Amd   future of gp us - campus partyAmd   future of gp us - campus party
Amd future of gp us - campus partyCampus Party Brasil
 
Graphics Processing Units
Graphics Processing UnitsGraphics Processing Units
Graphics Processing UnitsAmrik Sadhra
 
VMworld 2013: Graphics and Users in VDI
VMworld 2013: Graphics and Users in VDI VMworld 2013: Graphics and Users in VDI
VMworld 2013: Graphics and Users in VDI VMworld
 
What is graphic card.docx
What is graphic card.docxWhat is graphic card.docx
What is graphic card.docxSurendra Gusain
 
Introduction to GPU
Introduction to GPUIntroduction to GPU
Introduction to GPUbabuece
 
TECHNICAL PAPER PRESENTATION copy
TECHNICAL PAPER PRESENTATION copyTECHNICAL PAPER PRESENTATION copy
TECHNICAL PAPER PRESENTATION copyBhargav Ramesh
 
Enabling Artificial Intelligence - Alison B. Lowndes
Enabling Artificial Intelligence - Alison B. LowndesEnabling Artificial Intelligence - Alison B. Lowndes
Enabling Artificial Intelligence - Alison B. LowndesWithTheBest
 
GPU power consumption and performance trends
GPU power consumption and performance trendsGPU power consumption and performance trends
GPU power consumption and performance trendsAlessio Villardita
 
Accelerating Data Science With GPUs
Accelerating Data Science With GPUsAccelerating Data Science With GPUs
Accelerating Data Science With GPUsiguazio
 
Fuel your imagination with a Microsoft Surface Laptop Studio
Fuel your imagination with a Microsoft Surface Laptop StudioFuel your imagination with a Microsoft Surface Laptop Studio
Fuel your imagination with a Microsoft Surface Laptop StudioPrincipled Technologies
 

Similaire à Graphic Processing Unit (20)

Apu fc & s project
Apu fc & s projectApu fc & s project
Apu fc & s project
 
Graphics Processing Unit: An Introduction
Graphics Processing Unit: An IntroductionGraphics Processing Unit: An Introduction
Graphics Processing Unit: An Introduction
 
VisionizeBeforeVisulaize_IEVC_Final
VisionizeBeforeVisulaize_IEVC_FinalVisionizeBeforeVisulaize_IEVC_Final
VisionizeBeforeVisulaize_IEVC_Final
 
APU in nepal 2
APU in nepal 2APU in nepal 2
APU in nepal 2
 
GPU accelerated path rendering fastforward
GPU accelerated path rendering fastforwardGPU accelerated path rendering fastforward
GPU accelerated path rendering fastforward
 
Amd future of gp us - campus party
Amd   future of gp us - campus partyAmd   future of gp us - campus party
Amd future of gp us - campus party
 
Graphics Processing Units
Graphics Processing UnitsGraphics Processing Units
Graphics Processing Units
 
VMworld 2013: Graphics and Users in VDI
VMworld 2013: Graphics and Users in VDI VMworld 2013: Graphics and Users in VDI
VMworld 2013: Graphics and Users in VDI
 
What is graphic card.docx
What is graphic card.docxWhat is graphic card.docx
What is graphic card.docx
 
Introduction to GPU
Introduction to GPUIntroduction to GPU
Introduction to GPU
 
APUs in Nepal
APUs in NepalAPUs in Nepal
APUs in Nepal
 
Gpu
GpuGpu
Gpu
 
TECHNICAL PAPER PRESENTATION copy
TECHNICAL PAPER PRESENTATION copyTECHNICAL PAPER PRESENTATION copy
TECHNICAL PAPER PRESENTATION copy
 
Enabling Artificial Intelligence - Alison B. Lowndes
Enabling Artificial Intelligence - Alison B. LowndesEnabling Artificial Intelligence - Alison B. Lowndes
Enabling Artificial Intelligence - Alison B. Lowndes
 
GPU power consumption and performance trends
GPU power consumption and performance trendsGPU power consumption and performance trends
GPU power consumption and performance trends
 
Accelerating Data Science With GPUs
Accelerating Data Science With GPUsAccelerating Data Science With GPUs
Accelerating Data Science With GPUs
 
Fuel your imagination with a Microsoft Surface Laptop Studio
Fuel your imagination with a Microsoft Surface Laptop StudioFuel your imagination with a Microsoft Surface Laptop Studio
Fuel your imagination with a Microsoft Surface Laptop Studio
 
GPU
GPUGPU
GPU
 
Nvidia
Nvidia Nvidia
Nvidia
 
Gpu
GpuGpu
Gpu
 

Plus de Kamran Ashraf

The Maximum Subarray Problem
The Maximum Subarray ProblemThe Maximum Subarray Problem
The Maximum Subarray ProblemKamran Ashraf
 
Ubiquitous Computing
Ubiquitous ComputingUbiquitous Computing
Ubiquitous ComputingKamran Ashraf
 
Application programming interface sockets
Application programming interface socketsApplication programming interface sockets
Application programming interface socketsKamran Ashraf
 
Error Detection types
Error Detection typesError Detection types
Error Detection typesKamran Ashraf
 
INSTRUCTION LEVEL PARALLALISM
INSTRUCTION LEVEL PARALLALISMINSTRUCTION LEVEL PARALLALISM
INSTRUCTION LEVEL PARALLALISMKamran Ashraf
 

Plus de Kamran Ashraf (6)

The Maximum Subarray Problem
The Maximum Subarray ProblemThe Maximum Subarray Problem
The Maximum Subarray Problem
 
Ubiquitous Computing
Ubiquitous ComputingUbiquitous Computing
Ubiquitous Computing
 
Application programming interface sockets
Application programming interface socketsApplication programming interface sockets
Application programming interface sockets
 
Error Detection types
Error Detection typesError Detection types
Error Detection types
 
VIRTUAL MEMORY
VIRTUAL MEMORYVIRTUAL MEMORY
VIRTUAL MEMORY
 
INSTRUCTION LEVEL PARALLALISM
INSTRUCTION LEVEL PARALLALISMINSTRUCTION LEVEL PARALLALISM
INSTRUCTION LEVEL PARALLALISM
 

Graphic Processing Unit

  • 1. GRAPHIC PROCESSING UNIT Presented by KamranAshraf 13-NTU-4009
  • 4. WHAT IS RENDERING  Rendering is the process of generating an image from a 2D or 3D model by means of computer programs  It is the process of add shading, color and lamination in order to create life-like images on a screen  A GPU assists a CPU in performing complex rendering calculations  If a scene is to look relatively realistic and predictable under virtual lighting, the rendering software should do it  Rendering is also used to describe the process of calculating effects in a video editing program to produce final video output A variety of rendering techniques applied to a single 3D scene
  • 5. WHAT IS GPU RENDERING?  GPU rendering means, that there is used the power of GPU (Graphics card) for rendering, instead of CPU (processor).  Conventional CPUs (Intel, AMD) didn't make such dramatic progress, like GPUs in last 10 years.
  • 6. GPU RENDERING  GPU rendering is definitely the future of rendering as one single GPU has the same power as the huge and expensive computer clusters with many CPUs  GPU rendering is incredibly fast and less power hungry  You will need less computers for the same performance  Much CHEAPER
  • 8. THEWORLD'S FASTEST GPU The GTXTitan Z The Radeon R9 295X2
  • 9. AMD’S R9 295X2  It joins two of its R9 290X GPUs onto a single board  Max Resolution: 4096 x 2160, Support 5x Display Monitors  Memory: 8GB dedicated  Cores: Powered by 5,632 cores meaning 2,816 cores for each GPU  The most advanced dual GPU liquid cooling system ever designed  Ups the power consumption to a scarily large 500 watts  Cost around $1500
  • 10. NVIDIA GTXTITAN Z  5k resolutions supported:This GPU supports multi-monitors and a resolution upto 5k  Dual GPU architecture:Two Kepler GK110 GPU’s inside,Titan Z generates the same power as if twoTitan’s were running  Memory:The card will support a total of 12 GB dedicated memory i.e 6 GB for each GPU.  CUDA cores:The GTXTitan Z is powered by 5,760 CUDA cores meaning 2,880 cores for each GPU.  It won’t make much noise as compared to other powerful GPU’s  Cost around $3000
  • 12. IMAGE QUALITY CALCULATOR  The University of Manitoba developed an Image Quality Calculator.  This utility, which is a plugin for ImageJ, does a great job analyzing the quality (amount of noise)
  • 13. COMPARING RESULTS The GPU (2688 CUDA cores) was 6.2 times faster than the CPU.
  • 14. IMPACT OF GPU IN COMPUTING
  • 15. GPU ACCELERATED COMPUTING  GPU-accelerated computing is the use of aGPU together with a CPU to accelerate scientific, analytics, engineering, consumer, and enterprise applications.  Pioneered by NVIDIA, GPU accelerators now power energy-efficient datacenters in government labs, universities, enterprises, and small-and-medium businesses around the world.  GPUs are accelerating applications in platforms ranging from cars, to mobile phones and tablets, to drones and robots.
  • 16. HOW ITWORKS  Compute-intensive portions of the application are executed by the GPU  The remainder of the code still runs on the CPU  From a user's perspective, applications simply run significantly faster GPUs have thousands of cores to process parallel workloads efficiently
  • 18. THE LIGHTNING CHALLENGE  One of the most important aspects of creating convincing 3D visuals is implementing realistic lighting  The challenge is not having an artist manually animate realistic-looking lighting, but to simulate how a light reacts with, and bounces off, multiple surfaces in real- time, otherwise known as path tracing  In the film industry, path tracing is done offline, because the calculations required are simply too complex  Unity global illumination worked on advanced lighting model which supported indirect lighting but it was very expensive so It's a technology that's not quite ready.
  • 19. POWER CHALLENGES  Biggest, most complex processors on the planet are being made  The biggest challenge is no longer being able to build a fast GPU  The biggest challenge is power as Graphics can consume a near infinite amount of horsepower  We don't want to have to watercool everything, or have 19 fans and make things really loud.
  • 21. LATEST ADVANCEMENTS Nvidia's GameWorks technology  Nvidia's GamesWorks is a set of visual effects, physics effects, and tools designed to give developers more options when it comes to a game's visuals.  Effects like HBAO+ and FaceWorks are part of GameWorks.  AO can account for the occlusion of light, creating non-uniform shadows that add depth to the scene.  HBAO+ doubles the detail level of the AO effect, runs three times faster, and uses the latest DirectX 11 technologies.  Nvidia's GameWorks is more copyrighted in nature. AMD's Mantle technology  Mantle, allows console-like low-level access to the CPU and GPU.  By removing some of the software layers of the operating system between the hardware and the software  With the right programming developers are able to see an increase in performance  AMD claims its Mantle API is open, and free to use

Notes de l'éditeur

  1. Why brute force rendering? The way the brute force rendering works is by looking at the scene which needs to be rendered, then essentially hammering away at the scene by sampling different areas of the shot; calculation information, calculation ray information etc. This was once a very heavy task for older cumputers to deal with. However in the past 10 years computing power has increased substantially and as a result Brute force rendering has become a much more viable option.
  2. Why brute force rendering? The way the brute force rendering works is by looking at the scene which needs to be rendered, then essentially hammering away at the scene by sampling different areas of the shot; calculation information, calculation ray information etc. This was once a very heavy task for older cumputers to deal with. However in the past 10 years computing power has increased substantially and as a result Brute force rendering has become a much more viable option.
  3. Path tracing is a computer graphics Monte Carlo method of rendering images of three-dimensional scenes such that the global illumination is faithful to reality. Fundamentally, the algorithm is integrating over all the illuminance arriving to a single point on the surface of an object.