The History of the GPU - Steps to Invention

This document was uploaded by one of our users. The uploader already confirmed that they had the permission to publish it. If you are author/publisher or own the copyright of this documents, please report to us by using this DMCA report form.

Simply click on the Download Book button.

Yes, Book downloads on Ebookily are 100% Free.

Sometimes the book is free on Amazon As well, so go ahead and hit "Search on Amazon"

This is the first book in a three-part series that traces the development of the GPU. Initially developed for games the GPU can now be found in cars, supercomputers, watches, game consoles and more. GPU concepts go back to the 1970s when computer graphics was developed for computer-aided design of automobiles and airplanes. Early computer graphics systems were adopted by the film industry and simulators for airplanes and high energy physics―exploding nuclear bombs in computers instead of the atmosphere.

A GPU has an integrated transform and lighting engine, but these were not available until the end of the 1990s. Heroic and historic companies expanded the development and capabilities of the graphics controller in pursuit of the ultimate device, a fully integrated self-contained GPU.

Fifteen companies worked on building the first fully integrated GPU, some succeeded in the console, and Northbridge segments, and Nvidia was the first to offer a fully integrated GPU for the PC. Today the GPU can be found in every platform that involves a computer and a user interface.


Author(s): Jon Peddie
Edition: 1
Publisher: Springer
Year: 2023

Language: English
Pages: 430
City: Cham
Tags: Computer Graphics; Central Processing Unit; CPU; Graphics Processing Unit; GPU; Computer-Aided Design; GPU Development; Graphics Controllers; Moore's Law

Foreword
Preface
What Is In and Not In These Books
Significant Things
The Author
A Lifetime of Chasing Pixels
Acknowledgments and Contributors
Contents
List of Figures
List of Tables
1 Introduction
1.1 Introduction
1.2 First Computer Graphics System (1949)
1.3 The Graphics Processor Unit (1999)
1.3.1 The Evolution of Graphics Controllers to GPUs
1.4 Performance (2000–2026)
1.5 The GPU’s Changing Role
1.6 The GPU’s Application
1.6.1 AI and Machine Learning
1.6.2 Accelerated Computing and Supercomputers
1.6.3 Content Creation
1.6.4 Gaming
1.6.5 Molecular Modeling
1.6.6 Video and Photo Editing
1.6.7 Vehicle Navigation and Robots
1.6.8 Crypto Mining
1.6.9 Summary
1.7 The Many Roles of the GPU Require Additional Names
1.8 Types of GPUs
1.9 Conclusion
References
2 1980–1989, Graphics Controllers on Other Platforms
2.1 Ikonas Graphics Systems (1978–1982)
2.2 Pixel Planes—The Foundation of the GPU (1980–2000)
2.2.1 HP Acquires Division (1996)
2.3 Processor per Polygon—The Demetrescu Caltech Architecture (1980)
2.4 The Geometry Engine (1981)
2.5 Matrox SM640 with Geometry Engine (1987)
2.6 SGI’s Personal Integrated Raster Imaging System (IRIS) Workstation (1988)
2.7 SGI’s IrisVision AIB (1988)
2.8 NEC’s µPD7220 Pioneering Graphics Display Controller (1982)
2.9 Hitachi ACRTC HD63484 (1984)
2.10 Truevision (1984–1987)
2.11 Tl 34010 (1986)
2.11.1 TI Epilogue
2.12 MAGIC—Multiple Application Graphics Integrated Circuit (1987)
2.13 Raster Technologies Vertex Processor (1987)
2.14 Amiga (1988)
2.15 Sun’s GX Graphics Accelerator Board (1989)
2.15.1 Summary
2.16 Conclusion
References
3 1980–1989, Graphics Controllers on PCs
3.1 1980–1989, Graphics Controllers on the PC Platform
3.2 CRT Control (1975–1987)
3.2.1 The Video Output—LUT-DAC (~1981–1987)
3.2.2 Brooktree (1983–1996)
3.2.3 Edsun Labs (1989–1991)
3.2.4 Summary of Video Output
3.3 IBM Graphics History (1981–1990)
3.3.1 IBM CGA (1981)
3.3.2 IBM EGA (1984)
3.3.3 EGA Begets VGA to XGA
3.3.4 The IBM Professional Graphics Controller—PGC (1984)
3.3.5 The IBM 8514/A (1987)
3.3.6 IBM VGA (1987–1991)
3.3.7 Those Clones
3.3.8 IBM Summary
3.4 The Market Expands (1986–1987)
3.5 Intel’s Pre-GPU History (1983–2003)
3.5.1 82720 (1983)
3.5.2 82786 (1986)
3.5.3 i860 (1989)
3.5.4 i740 (1998)
3.5.5 i810 (1999)
3.5.6 Extreme Graphics (2001)
3.5.7 Intel Summary
3.6 Tseng Labs (1983–1997)
3.6.1 Winning Awards Was not Enough
3.6.2 It Could Have Been the First GPU
3.6.3 The End
3.7 SGI’s IrisVision (1988–1994)
3.7.1 The Legacy of IrisVision—Pellucid, Media Vision, and 3dfx (1991–1994)
3.7.2 Media Vision (1990–1994)
3.7.3 Benchmarking
3.8 Conclusion
References
4 1980–1995 the Progenitors: Graphics Controller on PCs
4.1 1990–1995, Graphics Controllers on PCs
4.1.1 IBM XGA (1990)
4.1.2 Summary 1990 to 1995
4.2 The IGC to IGP (1991)
4.2.1 The First Workstation IGC
4.2.2 The First PC IGC
4.3 Bitboys (1991–1999)
4.3.1 Pyramid3D 25202
4.3.2 Pyramid3D 25201
4.3.3 The Eight P’s
4.3.4 Summary
4.4 Artist Graphics (1979–2098)
4.4.1 Artist Graphics Shows 3GA Graphics Accelerator
4.4.2 Summary
4.5 Number Nine Imagine 128 (1992–1999)
4.5.1 Summary
4.6 Rendition (1992–1998)
4.6.1 Summary
4.7 Stellar—RSSI (1993–2000)
4.7.1 Reality Simulations Systems PixelSquirt
4.7.2 Stellar is Born (1997)
4.7.3 VelaTX (1998)
4.7.4 Broadcom Acquires Stellar (2000)
4.7.5 Summary
4.8 Matrox Millennium (1994–2014)
4.8.1 Summary
4.9 VideoLogic/Imagination Technologies Tiling (1994–)
4.9.1 NEC-Imagination Technologies PCX (1994–1999)
4.9.2 Summary
4.10 Conclusion
References
5 1990 to 1999 Graphics Controllers on Other Platform
5.1 Workstations
5.1.1 Workstation Graphics
5.1.2 HP Artist (1993)
5.1.3 Silicon Reality (1994–1998)
5.1.4 The Saga of Evans & Sutherland’s Pre-GPU Effort (1995–2001)
5.1.5 3Dlabs Permedia (1997)
5.1.6 Intergraph Wildcat (1998–2000)
5.2 Game Consoles
5.2.1 Sega
5.2.2 Sega Genesis (1988)
5.2.3 Sony PlayStation (1994)
5.2.4 Atari Jaguar (1993)
5.2.5 Nintendo 64 (1996)—The First T&L in a Console
5.2.6 ArtX and the Nintendo GameCube (1998)
5.2.7 NEC Electronics’ PowerVR (1996)
5.3 Conclusion
References
6 1996–1999, Graphics Controllers on PCs
6.1 The ATI 3D Rage (1995)
6.1.1 Approaching the GPU
6.1.2 The Saga of ATI (1985–2006)
6.2 Nvidia’s Quadratic Processor, the NV1 (1993–)
6.2.1 Nvidia Epilogue
6.3 3dfx Voodoo (1994–2000)
6.3.1 SLI Was Not a New Concept
6.3.2 The SST-1
6.4 Yamaha YGV612 RPA (1995–1996)
6.5 Real3D (1995–1999)
6.5.1 A Stand-Alone Company
6.5.2 Real3D and Silicon Graphics Settle Out of Court
6.5.3 Intel Acquires Real3D (October 25, 1999)
6.5.4 3dfx and Intel Patent Cross-License Agreement
6.6 Microsoft Talisman—The Chip That Never Was (1996)
6.7 Nvidia Riva 128 (1996)
6.8 S3 Virge 86C385 (1996)
6.8.1 S3 Epilogue
6.9 Conclusion
References
7 What is a GPU?
7.1 What is a GPU?
7.2 The GPU
7.2.1 Vendors
7.2.2 Shaders, Processors, Units, and Cores
7.2.3 Getting to a Model
7.3 The Six Eras of GPUs
7.3.1 Pre-GPU Graphics Controllers (1960–1998)
7.3.2 First-Era GPUs (1999–2000) Fixed Function
7.3.3 Second-Era GPUs (2000–2006) Programmable Shaders
7.3.4 Third-Era GPUs (2006–2009) Unified Shaders
7.3.5 Fourth-Era GPUs (2009–2015) Compute Shaders
7.3.6 Fifth-Era GPUs (2015–2020) Ray Tracing and AI
7.3.7 Sixth-Era GPUs (2020–) Mesh Shaders
7.3.8 The Range of the GPU and This Book
7.4 Conclusion
7.5 Epilog
References
Appendix A Acronyms
Appendix B Definitions
Index