Organización De Computadoras

Total Page:16

File Type:pdf, Size:1020Kb

Organización De Computadoras Organización de Computadoras Principios de Conjuntos de Intrucciones Clasifición de las ISA Secuencia de Código para C = A + B Stack Acumulador Registro Registro (Reg-Mem) (load-store) Push A Load A Load R1, A Load R1,A Push B Add B Add R3,R1,B Load R2,B Add Store C Store R3, C Add R3,R1,R2 Pop C Store R3,C Combinaciones de operandos en memoria y operandos totales en instrucciones de ALU Nro. de Máximo nro. Tipo de Ejemplos direcciones de de operandos arquitectura memoria permitidos 0 3 Registro-registro Alpha, ARM, MIPS, PowerPC, SPARC. 1 2 Registro- IBM 360/370, memoria Intel 80x86, Motorola 68000 2 2 Memoria- VAX Memoria 3 3 Memoria- VAX Memoria Ventajas y desventajas de los 3 tipos más comunes de computadoras del tipo registro de propósito general Tipo Ventajas Desventajas R-R (0,3) Simple, codificación con instrucciones de Conteo de instrucciones tamaño fijo, modelo de generación de más alto, programas más código simple, instrucciones toman grandes similares ciclos de reloj para su ejecución R-M (1,2) Se pueden acceder datos sin un load Los operandos no son separado primero. Tendencia a ser equivalentes (uno es facilmente codificado, destuido). Pocos bits disponibles para codificar el registro M-M (2,2) Código compacto. No desperdicia Gran variación en el o (3,3) registros como temporarios tamaño de instrucción. CPI muy variable. Cuello de botella en la memoria Direccionamiento de Memoria Direccionamiento por byte , acceso a: • Byte • Half word • Word • Double word Tópicos • Alineación • Orden • Litttle endian • Big endian Alineación de objetos en memoria Direccionamiento de Memoria Interpretación de las direcciones de memoria. Word • “Little Endian ”: pone el byte cuya dirección es XX000 en la parte más baja de la palabra. Word • “Big Endian ”: pone el byte cuya dirección es XX000 en la parte más alta de la palabra. Modos de Direccionamiento MIPS 32 1. Immediate addressing o p rs r t Im m e d ia te 2. Register addressing o p rs r t rd . funct R e g is te rs R e g is te r 3. Base addressing op rs rt Address M e m o r y R e g iste r + B y te H a lfw o r d W o rd 4. PC-relative addressing o p rs r t A d d r e s s M e m o r y PC + W o rd 5. Pseudodirect addressing o p A d d re s s M e m o r y PC W o rd Modos de Direccionamiento en MIPS 32, ejemplos Registro Inmediato Desplazamiento add $1, $2, $3 regs[1] regs[2] + regs[3] addi $1, $2, 35 regs[1] regs[2] + 35 lw $1, 35($2) regs[1] mem[regs[2] + 35] Modos de Direccionamiento en MIPS C code: g = h + A[8]; g in $s1, h in $s2, base address of A in $s3 Compiled MIPS code: Index 8 requires offset of 32 4 bytes per word lw $t0, 32($s3) #load word add $s1, $s2, $t0 Frecuencia de modos de direccionamiento (VAX, SPEC89) Número de bits de desplazamiento (Alpha, SPEC CPU2000) Uso de inmediatos (Alpha, SPEC CPU2000) Tipo y tamaño de operandos Generalmente lo implica el OpCode • Carácter • ASCII • Unicode • Enteros • Half word • Word • Coma flotante, precisión simple y doble • IEEE 754 • Decimal • BCD Operaciones en el conjunto de instrucciones Top 10 instrucciones en 80x86 Soporte multimedia para desktop RISC Instrucciones de control del flujo del programa Salto condicional (branch) Salto incondicional (jump) Llamado a procedimiento (call) Retorno de procedimiento (return) Modos de direccionamiento para instrucciones de control Relativas al PC Registro indirecto (dirección del salto no conocida en tiempo de compilación) Evaluación de las condiciones de los Branchs Frecuencia de los diferentes tipos de comparaciones Programando a nivel de la máquina Programa en MIPS 32 que calcula la suma de los cuadrados de los numeros de 0 a 100. Programando en lenguaje ensamblador (crudo) Da una sintaxis con mnemónicos para las instrucciones, registros y modos de direccionamiento. Programando en lenguaje ensamblador (con abstracciones) Agrega: directivas, labels, pseudoinstrucciones, modos de direccionamiento no soportados por el procesador . Proceso de producir un ejecutable Ensamblador Programa que traduce el código assembly a binario. Genera como salida un archivo objeto . Archivo objeto generado por un assembler en Unix Linker Enlaza los archivos objeto y genera el ejecutable. Resuelve las referencias a bibliotecas externas. Asigna las posiciones finales en memoria. Necesidad de mayores de abstracciones: lenguajes de alto nivel El compilador Generación de un ejecutable El Roll de los Compiladores Expectativas para aplicaciones del tipo de escritorio Arquitectura carga/almacenamiento Registros de propósito general Modos de direccionamiento • Desplazamiento • Inmediato Tipos de datos • Enteros, 8, 16, 32 y 64 bits • Coma flotante, IEEE 754 Instrucciones simples • load, store, add, subtract, move reg-reg, shift • Compare, branch, jump, call, return Codificación de instrucción fija La Arquitectura MIPS MIPS 32 La Arquitectura MIPS Conjunto de instrucciones simple L/S Diseñada para eficiencia del pipeline Eficiencia para el compilador La Arquitectura MIPS Registros • 32 registros de 32 bits: $0, $1, $2, $31 (int) • 32 registros de 32 bits: $f0, $f1, $f2, $f31 (fp) Tipos de datos • Bytes (8 bits), Half Word (16 bits), Words (32 bits) • Double words (fp) Modos de direccionamiento • Inmediato • Desplazamiento CPU, FPU y el coprocesador 0 Causa de excepciones El registro CAUSE Convención de uso de registros Registers Name Number Use Callee must preserve? $zero $0 constant 0 N/A $at $1 assembler temporary No values for function $v0–$v1 $2–$3 returns and expression No evaluation $a0–$a3 $4–$7 function arguments No $t0–$t7 $8–$15 temporaries No $s0–$s7 $16–$23 saved temporaries Yes $t8–$t9 $24–$25 temporaries No $k0–$k1 $26–$27 reserved for OS kernel No $gp $28 global pointer Yes $sp $29 stack pointer Yes $fp $30 frame pointer Yes $ra $31 return address N/A Ejemplo de un loop # Computa la suma de los elementos de un vector. Precondiciones en zona de datos: length es la longitud del vector de enteros array. Postcondiciones: el resultado queda en donde estaba la longitud del vector. .text .globl main main: addi $v0, $zero, 0 # v0 = resultado = 0 la $t0, length # t0 = address de length lw $t1, 0($t0) # t1 = length la $t4, array # t4 = address de array[0] addi $t1, $t1, -1 # t1 = indice del ultimo elemento del vector sll $t1, $t1, 2 # multiplico length por 4, t1 = address de array[length-1] loop: add $t3, $t4, $t1 lw $t2, 0($t3) # t2 = array[i] add $v0, $v0, $t2 # resultado += array[i] addi $t1, $t1, -4 # i-- test: slti $t9, $t1, 0 # if !(i<0) ... beq $t9, $zero, loop # ... go to loop nop sw $v0, 0($t0) # reemplaza length con resultado halt: beq $zero, $zero, halt nop .data length: .word 8 array: .word 9, 8, 7, 6, 5, 4, 3, 2 .text .globl main main: addi $v0, $zero, 0 # v0 = resultado = 0 #la $t0, length # t0 = address de length lui ori lw $t1, 0($t0) # t1 = length #la $t4, array # t4 = address de array[0] lui ori addi $t1, $t1, -1 # t1 = indice del ultimo elemento del vector sll $t1, $t1, 2 # multiplico length por 4, t1 = address de array[length - loop: add $t3, $t4, $t1 lw $t2, 0($t3) # t2 = array [i] add $v0, $v0, $t2 # resultado += array[i] addi $t1, $t1, -4 # i-- test: slti $t9, $t1, 0 # if !(i<0) ... beq $t9, $zero, loop # ... go to loop nop sw $v0, 0($t0) # reemplaza length con resultado halt: beq $zero, $zero, halt nop .data length: .word 8 array: .word 9, 8, 7, 6, 5, 4, 3, 2 Código sin pseudoinstrucciones Para el código anterior calcular: Cantidad de instrucciones del programa. Tamaño del código en bytes. Cantidad de instrucciones ejecutadas. Cantidad de bytes de instrucciones transferidos entre CPU y memoria. Cantidad de accesos a datos en memoria realizados. Cantidad de bytes de datos transferidos entre CPU y memoria . Cantidad total de bytes transferidos entre CPU y memoria. Para el código anterior calcular: Porcentaje de accesos a datos que son lectura. Porcentaje de accesos a memoria que son lectura. Clasificar a las instrucciones en los grupos: ALU, L/S, BRANCH, JUMP, FP. Calcular la frecuencia de ejecución de cada grupo. Calcular el porcentaje de saltos tomados. Calcular el CPI del código (usar CPI típicos de grupos) Calcular el tiempo de ejecución si la CPU corre a 100MHZ. Calcular el porcentaje de tiempo que la CPU ejecuta intrucciones de ALU. Para el código anterior calcular: Se propone llevar el CPI de las instrucciones de BRANCH a 1 (branchs tomados y no tomados). • Calcular el SPEEDUP total usando la ecuación de desempeño de CPU. • Calcular el SPEEDUP total usando la ley de Amdahl. Formatos de Instrucciones MIPS Instrucción del Tipo I: addi (Texto) addi rt, rs, imm regs[rt] ← regs[rs] + imm Ensamblado 31 0 (Hardware) 8 rs rt imm Op Code fuente destino inmediato (6 bits) (5 bits) (5 bits) (16 bits) Ejemplo de addi (Texto) addi $1, $2, 15 regs[1] ← regs[2] + 15 Ensamblado (Hardware) 001000 00010 00001 0000000000001111 Op Code fuente destino inmediato Instrucción del Tipo I: lw (Texto) lw rt, offset(rs) Regs[rt] ← Mem[Regs[rs] + imm] Ensamblado 31 0 (Hardware) 0x23 rs rt Offset Op Code fuente destino inmediato (6 bits) (5 bits) (5 bits) (16 bits) Ejemplo de lw (Texto) lw $3, 33($4) Regs[3] ← Mem[Regs[4] + 33] Ensamblado 31 0 (Hardware) 010111 00100 00011 0000000000100001 Op Code fuente destino inmediato (6 bits) (5 bits) (5 bits) (16 bits) Instrucción del Tipo I: bne (Texto) bne rs, rt, label if (Regs[rs] ≠ Regs[rt]) then PC ← PC+4+(Offsetx4) Ensamblado else PC ← PC+4 31 0 (Hardware) 5 rs rt Offset Op Code fuente destino inmediato (6 bits) (5 bits) (5 bits) (16 bits) Ejemplo de bne bne $5, $6, loop if (Regs[5] ≠ Regs[6]) then PC ← PC+4+(-5x4)
Recommended publications
  • Computer Architectures an Overview
    Computer Architectures An Overview PDF generated using the open source mwlib toolkit. See http://code.pediapress.com/ for more information. PDF generated at: Sat, 25 Feb 2012 22:35:32 UTC Contents Articles Microarchitecture 1 x86 7 PowerPC 23 IBM POWER 33 MIPS architecture 39 SPARC 57 ARM architecture 65 DEC Alpha 80 AlphaStation 92 AlphaServer 95 Very long instruction word 103 Instruction-level parallelism 107 Explicitly parallel instruction computing 108 References Article Sources and Contributors 111 Image Sources, Licenses and Contributors 113 Article Licenses License 114 Microarchitecture 1 Microarchitecture In computer engineering, microarchitecture (sometimes abbreviated to µarch or uarch), also called computer organization, is the way a given instruction set architecture (ISA) is implemented on a processor. A given ISA may be implemented with different microarchitectures.[1] Implementations might vary due to different goals of a given design or due to shifts in technology.[2] Computer architecture is the combination of microarchitecture and instruction set design. Relation to instruction set architecture The ISA is roughly the same as the programming model of a processor as seen by an assembly language programmer or compiler writer. The ISA includes the execution model, processor registers, address and data formats among other things. The Intel Core microarchitecture microarchitecture includes the constituent parts of the processor and how these interconnect and interoperate to implement the ISA. The microarchitecture of a machine is usually represented as (more or less detailed) diagrams that describe the interconnections of the various microarchitectural elements of the machine, which may be everything from single gates and registers, to complete arithmetic logic units (ALU)s and even larger elements.
    [Show full text]
  • “LCD TV Matters” Volume 2, Issue 2
    “LCD TV Matters” Volume 2, Issue 2 "A Great TV in Every Room" LCD TV Association LCD TV Matters Pre-CES 2009 Contents Chairman’s Corner: A crazy few months... by Bruce Berkoff 3 LCD-TV News compiled by Veritas et Visus 6 Sharp unveils AQUOS Experience for holiday season 6 Samsung introduces a 2000-nit 70-inch LCD panel 6 Panasonic Viera TH-PZR900 HDTV sports terabyte drive 6 LG Display shows off 480Hz LCD TV panel 7 Viewsonic launches 24-inch 1080p LCD TV 7 Samsung introduces 52-inch “Touch of Color” LCD TV 7 HDMI adopted by 700+ manufacturers as new CE and PC products hit market 7 Sharp releases world’s first LCD TV with built-in Blu-ray Disc recorder 8 Sony unveils world's thinnest LCD TV 8 LifeSize introduces the next-generation telepresence and high definition video communications 8 Sharp adds enhanced inputs and network connectivity to HD LCD line-up 9 Panasonic launches PZ850 Web-enabled TVs – features H.264 and four HDMI ports 9 Schaub Lorenz introduces $130,000 40-inch LCD TV 10 Philips Research reveals ultra-thin backlight technology for TVs 10 LG and Samsung join forces to develop mobile digital TV standard via ATSC 10 eyevis introduces 56-inch LCD TV at 3840x2160 pixels 11 AUO and Qisda form JV to manufacture LCD TVs 11 ATSC approves AVC within DTV transmissions 11 NEC introduces 82-inch professional-grade LCD for digital signage 11 Fairchild Semiconductor’s LCD TV design achieves 90% efficiency 12 ZeeVee now shipping ZvBox for PC content streaming 12 GAO reports DTV transition making progress 12 Broadcom to acquire digital
    [Show full text]
  • Embedded Linux Primer: a Practical Real-World Approach
    Embedded Linux Primer: A Practical, Real-World Approach By Christopher Hallinan ............................................... Publisher: Prentice Hall Pub Date: September 18, 2006 Print ISBN-10: 0-13-167984-8 Print ISBN-13: 978-0-13-167984-9 Pages: 576 Table of Contents | Index Comprehensive Real-World Guidance for Every Embedded Developer and Engineer This book brings together indispensable knowledge for building efficient, high-value, Linux-based embedded products: information that has never been assembled in one place before. Drawing on years of experience as an embedded Linux consultant and field application engineer, Christopher Hallinan offers solutions for the specific technical issues you're most likely to face, demonstrates how to build an effective embedded Linux environment, and shows how to use it as productively as possible. Hallinan begins by touring a typical Linux-based embedded system, introducing key concepts and components, and calling attention to differences between Linux and traditional embedded environments. Writing from the embedded developer's viewpoint, he thoroughly addresses issues ranging from kernel building and initialization to bootloaders, device drivers to file systems. Hallinan thoroughly covers the increasingly popular BusyBox utilities; presents a step-by-step walkthrough of porting Linux to custom boards; and introduces real-time configuration via CONFIG_RT--one of today's most exciting developments in embedded Linux. You'll find especially detailed coverage of using development tools to analyze
    [Show full text]
  • Radeon® X1550 Graphics Technology - GPU Specifications
    Radeon® X1550 Graphics Technology - GPU Specifications Advanced Image Quality Features • 64-bit floating point HDR rendering supported throughout the pipeline Features • Includes support for blending and multi-sample anti-aliasing • 105 million transistors on 90nm fabrication process • 32-bit integer HDR (10:10:10:2) format supported throughout the pipeline • Ultra-threaded architecture with fast dynamic branching • Includes support for blending and multi-sample anti-aliasing • 4 pixel shader processors • 2x/4x/6x Anti-Aliasing modes • 2 vertex shader processors • Multi-sample algorithm with gamma correction, programmable sparse sample • 128-bit 4-channel DDR2 memory interface, 64-bit/2-channel patterns, and centroid sampling configurations • New Adaptive Anti-Aliasing feature with Performance and Quality modes • Native PCI Express x16 bus interface, AGP 8x/4x, 3.3v PCI • Temporal Anti-Aliasing mode Note: AGP and PCI supported through bridge ASIC • Lossless Color Compression (up to 6:1) at all resolutions, including widescreen HDTV resolutions High Performance Memory Controller • 2x/4x/8x/16x Anisotropic Filtering modes • Fully associative texture, color, and Z/stencil cache designs • Up to 128-tap texture filtering • Hierarchical Z-buffer with Early Z test • Adaptive algorithm with Performance and Quality options • Lossless Z Compression (up to 48:1) • High resolution texture support (up to 4k x 4k) • Fast Z-Buffer Clear • Z/stencil xache optimized for real-time shadow rendering Avivo™ Video and Display Platform • High performance programmable
    [Show full text]
  • Video Streaming
    Video streaming a research on free video streaming technologies Jaromil dyne.org / rastasoft.org Table of Contents Introduction................................................................................................................................................1 We need CPU for compression .................................................................................................................2 We need bandwidth for distribution ........................................................................................................2 We need harddisk space for storage.........................................................................................................3 Distribution of video archives...................................................................................................................4 Embedded solutions...................................................................................................................................6 Streaming video software..........................................................................................................................8 About the author of this document ........................................................................................................10 Video streaming terms glossary..............................................................................................................11 Introduction Given the vast panorama of video technologies available nowadays, this research could fill up way too much paper in the
    [Show full text]
  • Download X100 Graphics Overview In
    x100 - graphics overview Bringing a unique way of seeing things. The Shuttle X100 features ATI’s Mobility™ Radeon® X1400 - providing best-in-class graph- ics performance, delivering enhanced 3D gaming, high quality video playback, and high energy efficency. Boost graphics performance with the dedicated 90-nanometer process Graphics Processing Unit (GPU) with an ultra-threaded 3D architecture, ring-bus memory controller, and a native 16x PCI Express® bus interface. Render lifelike 3D characters and scenes with any OpenGL® and DirectX® 9 game and enjoy astonishing effects with full DirectX® 9 Shader Model 3.0 support. With ATI’s Avivo™ High-Definition video and display enhancements and PowerPlay™ 6.0 advanced power management, Mobility Radeon X1400 creates a new performance threshold for mobile productivity and entertainment. Best-in-class mobile graphics performance. Go for the industry’s best-in-class graphics performance for your small-form-factor—ATI’s Mobile on desktop. Mobility™ Radeon® X1400. A discrete 90-nanometer process GPU with a new ultra-threaded The X100 breaks the rules of the 3D architecture delivers visual power far beyond integrated UMA graphics, performance- computing world, by bring mobile enhanced with a new ring-bus memory controller and high-speed 128-bit 4-channel technology in a desktop form-fac- tor. The Intel Core Duo, Core Solo, DDR/DDR2/GDDR3 RAM support. Accelerate through tasks and entertainment with a native and Celeron M bring powerful and x16-lane PCI Express® bus interface, pumping data through 4 pixel shader pipelines and 2 energy efficent processors to time vertex processors. Mobility Radeon X1400 packs all the mobile graphics power and visual when environmental consideration velocity you need with your X100.
    [Show full text]
  • Development of an Immersive Challenge Racing Game for 3D-TV
    SBC - Proceedings of SBGames 2013 Computing Track – Full Papers Development of an Immersive Challenge Racing Game for 3D-TV Ana Grasielle Dionísio Corrêa, Alexandre Antonino Gonçalves Martinazzo, Erich Panzenboeck Lotto, Marcelo Archanjo José, Irene Karaguilla Ficheman, Roseli de Deus Lopes Escola Politécnica da Universidade de São Paulo (EP-USP) Laboratório de Sistemas Integráveis (LSI), São Paulo, Brasil {anagrasi, am, irene, roseli} @lsi.usp.br; [email protected]; [email protected] Abstract—Television is most people favorite pastime. [7]. For this reason, the Advanced Television Systems The way we watch TV has not changed since its invention Committee (ATSC) in North America has created a planning 75 years ago. However, research in computer graphics and team to examine the possible advantages and disadvantages, display technology favored the advances of TV platforms, requirements and steps required to manufacture and deliver such as interactivity and stereoscopy. Moreover device 3D-TV content [7]. Similar investigations are also being connectivity allows new ways of distributing content and carried out by international standardization bodies such as the applications, including films and games. Combining those European Broadcasting Union (EBU), the International innovations, 3D-TV is a new type of home entertainment Telecommunication Union (ITU) and 3D Home Consortium [8]. that provides a rich and fun experience, innovating not only the way we watch movies and television series but also Great interest in 3D-TV is the recognition that compared the way we interact. In this context, this paper presents the with conventional television (standard 2D), 3D technology development of an immersive racing game for 3D-TV, improves significantly television entertainment value [9], [10].
    [Show full text]
  • HP Xw4400 Workstation Overview
    QuickSpecs HP xw4400 Workstation Overview HP recommends Windows Vista™ Business 1. Monitor (sold separately) 7. 3 PCI slots, 1 PCI Express x1 slot, 1 PCI Express x16 slot 2. 2-Button Scroll Mouse (with x4 functionality) 3. Standard Keyboard 8. 1 PCI Express x16 Graphics slot 4. Front IO: 2 USB 2.0, IEEE-1394 (requires optional PCI 9. Rear IO: 5 USB 2.0, 1 standard serial port, 1 optional card to enable), headphone and microphone serial port, 1 parallel port, PS/2 keyboard, PS/2 mouse, 5. One 3.5" external bay for optional diskette drive or other RJ-45, audio in, audio out, mic in 3.5" device 10. 1 USB 2.0 internal port 6. Three 5.25" external bays (3rd external is not full depth), 11. Intel® Core™ 2 Duo processor or Intel Pentium® D and two 3.5" internal bays processor; both processors are EM64T capable 12. 460 watt (continuous) power supply At A Glance DA - 12568 North America — Version 21 — August 16, 2007 Page 1 QuickSpecs HP xw4400 Workstation Overview Choice of Operating Systems: Microsoft Vista™ Business 32 or 64* Microsoft Windows Vista™ Business 32-bit downgrade to Microsoft Windows XP Professional Microsoft Windows Vista™ Business 64-bit downgrade to Microsoft Windows XP Professional x64 Microsoft Windows XP Professional (32-Bit) Microsoft Windows XP Professional x64 Edition Preloaded: Red Hat Enterprise Linux® WS 4 (Update 4 or later) (32- or 64-bit version) HP Linux Installer Kit (see http://www.hp.com/workstations/software/linux) Red Hat Enterprise Linux WS 4 (Update 4 or later) (32- or 64-bit version) Red Hat Enterprise Linux
    [Show full text]
  • AMD M690T Databook
    AMD M690T Databook Technical Reference Manual Rev. 3.04 P/N: 42437_m690t_ds_nda_3.04 © 2007 Advanced Micro Devices, Inc Please note that in this databook, references to "DVI" and "HDMI" refer to the capability of the TMDS interface, multiplexed on the PCI Express® external graphics interface, to enable DVI or HDMI through passive enabling circuitries. Any statement in this databook on any DVI or HDMI-related functionality must be understood in that context. Advanced Micro Devices, Inc., will not provide any indemnity, pay any royalty, nor provide any license/sublicense to any: (a) Intellectual property rights relating to any of the following: (i) Macrovision for its Analog Protection System ("APS") technologies; (ii) Advanced Television Systems Committee (ATSC) standard and related technologies; or (iii) the High Definition Multimedia Interface (HDMI) standard and related technologies; or (b) Audio and/or video codecs or any industry standard technology (e.g., technology or specifications promulgated by any standards development organization, consortium, trade association, special interest group or like entity). This device is protected by U.S. patent numbers 4,631,603, 4,577,216 and 4,819,098 and other intellectual property rights. The use of Macrovision's copy protection technology in the device must be authorized by Macrovision and is intended for home and other limited pay-per-view uses only, unless otherwise authorized in writing by Macrovision. Reverse engineering or disassembly is prohibited. This device may only be sold or distributed to: (i) a Macrovision Authorized Buyer, (ii) a customer (PMA Customer) who has executed a Proprietary Materials Agreement (PMA) with Macrovision that is still in effect, (iii) a contract manufacturer approved by Macrovision to purchase this device on behalf of a Macrovision Authorized Buyer or a PMA Customer, or (iv) a distributor who has executed a Macrovision-specified distribution agreement with ATI.
    [Show full text]
  • A Platform for Building Programmable Multicore Video Processors
    mediaDSP™: A Platform for Building Programmable Multicore Video Processors Richard Selvaggi and Larry Pearlstein AMD Hot Chips 20, August 2008 Recent Trends Big-screen TV sets have been selling in ever-increasing numbers in the past 5 to 10 years. LCDs have recently overtaken CRTs in unit shipments. LCD display technology has advanced in many ways over these years. – Wider viewing angle – Higher contrast ratio – Wider color gamut – Higher resolution But...motion blur is one of the last remaining challenges for LCD display technology. – This is the problem our chip is addressing. 2 August 2008 Hot Chips 20: mediaDSP Perception of Motion Blur on LCD Panels 60 frames/sec 120 frames/sec 16ms 16ms 8ms Reduced blur Reduced Interpolated frames Eye moves continuously, smearing Eye moves a shorter arc in 8ms, thus the still image across the retina reducing amount of blur across retina LCD is a hold-type display (not impulsive, like CRT). Projected object is smeared on the retina by the amount that the eye moves in one frame-time (16 vs. 8ms). 3 August 2008 Hot Chips 20: mediaDSP TV System Block Diagram TV System (FullHD, 1080p) TV Main Board Panel Board 120Hz 60Hz AMD DTV SoC Xilleon™ 420 120Hz LCD Panel DRAM DRAM AMD Xilleon 420 (X420) performs motion compensated frame rate conversion (MC-FRC) and doubles the frame rate. Samsung Timing Controller (TCON) for LCD panel is integrated inside X420. 4 August 2008 Hot Chips 20: mediaDSP AMD Xilleon 420 Block Diagram AMD Xilleon™ 420 Timing mini-LVDS LVDS LVDS mini-LVDS Controller Pixel Pixel Input
    [Show full text]
  • ATI Radeon™ HD 4300 Series GPU Specifications
    ATI Radeon™ HD 4300 Series GPU Specifications Products & Technologies Support & Drivers at Work at Home at Play Worldwide | About AMD | News | Where to buy Find a Driver Products & Technologies ATI Radeon HD 4000 ATI Radeon HD 4870 X2 ATI Radeon HD 4890 ATI Radeon HD 4870 ATI Radeon HD 4850 ATI Radeon HD 4830 ATI Radeon™ HD 4770 ATI Radeon HD 4600 ATI Radeon HD 4550 ATI Radeon HD 4300 ATI Radeon HD 4200 Series ATI Radeon™ HD 4300 Series GPU Specifications ● 242 million transistors on 55nm fabrication process ● PCI Express 2.0 x16 bus interface ● GDDR3/DDR3/DDR2 memory interface (depending on model) ● Microsoft® DirectX® 10.1 support ❍ Shader Model 4.1 ❍ 32-bit floating point texture filtering ❍ Indexed cube map arrays ❍ Independent blend modes per render target ❍ Pixel coverage sample masking ❍ Read/write multi-sample surfaces with shaders ❍ Gather4 texture fetching ● Unified Superscalar Shader Architecture ❍ 80 stream processing units ■ Dynamic load balancing and resource allocation for vertex, geometry, and pixel shaders ■ Common instruction set and texture unit access supported for all types of shaders ■ Dedicated branch execution units and texture address processors ❍ 128-bit floating point precision for all operations ❍ Command processor for reduced CPU overhead ❍ Shader instruction and constant caches ❍ Up to 32 texture fetches per clock cycle ❍ Up to 128 textures per pixel ❍ Fully associative multi-level texture cache design ❍ DXTC and 3Dc+ texture compression ❍ High resolution texture support (up to 8192 x 8192) ❍ Fully associative
    [Show full text]
  • Club 3D Radeon X1300 Pcie New Standard for Mainstream VGA
    Club 3D Radeon X1300 PCIe New Standard For Mainstream VGA Introduction: The Club 3D Radeon® X1300Pro is a completely new generation of graphics based on latest 3D Technolo- gies. In fact it is a derivative of the ATI™ high end X1800 family to bring high performance graphics tech- nology to the discrete desktop market on extremely PowerPack sharp pricing levels. The Club 3D Radeon® X1300Pro is targeted to the more 3D games orientated PC user. It combines the excellent desktop capabilities for Home and Office applications of the X1300 with more rendering power for 3D games by having higher clock speeds. Order Information: • Radeon X1300 256MB : CGAX-136TVD Product Positioning: Colin McRae 2005 • Mainstream market • SOHO and Gaming Features: System requirements ® ® ® Specifications: • DirectX 9.x compliant • Intel Pentium or AMD™ Athlon™ or • Shader Model 3 support compatible Item code: CGAX-136TVD • High Dimension Floating-Point Textures • 128MB of system memory Format: PCIe • Full Screen Anti Aliasing Gamma Correction • Mainboard with free PCIe (x16) slot Engine Clock: 400MHz • Avivo™ • CD-ROM drive for software installation Memory Clock: 500MHz • Video Gamma Correction • 350Watt or greater Power Supply Driver/E-manual Memory: 256MB DDR2 • 3Dc™ Normal Map Compression • Enhanced HyperMemory Technology Memory Bus: 128 bit Pixel Pipelines: 4 Accessories: Operating System Support RAMDAC: 2x 400MHz • S-video/Composite cable • Microsoft® Windows® XP/x64, MCE ® ® DVI-I: √ • Panda Jr WWF, Colin McRae 2005 Game • Microsoft Windows 2000 • DVI/CRT converter
    [Show full text]