Taqi Raza Khan :: Blog

Processor Overview

The Opteron is AMD's x86 server processor line, and was the first processor to implement the AMD64 instruction set architecture (known generically as x86-64). It was released on April 22, 2003 with the SledgeHammer core (K8) and was intended to compete in the server market, particularly in the same segment as the Intel Xeon processor. Processors based on the AMD K10 microarchitecture (codenamed Barcelona) were announced on September 10, 2007 featuring a new quad-core configuration.

• AMD Opteron
• AMD Opteron 64
• AMD Athlon 64 FX
• AMD Turion
• AMD Sempron

Today, i am going to tell you about the AMD 64 Architecture overview. We are going to divide our session in few Blog Entries over a couple of days. I will cover the following topics :

+ Processor Overview

I. AMD64 Architecture Overview

   + Operating Modes

   + Register Set

   + Segmentation

   + Task Management

   + Interrupts and Exceptions

   + Demand Mode Paging

   + Instruction Set Extensions

   + x86 Virtualization Overview 

 II. CPU Microarchitecture

   + Processor Core Introduction and Terminology

   + Integer Pipeline

   + FPU Pipeline

   + Load/Store Unit

   + Caches

III. System Architecture

   + System Chip Components

   + Hypertransport

   + Configuration Space

   + PC Memory Technologies

   + System Power Up Process

   + APIC

   + Debug Registers

   + Performance Monitoring Registers

   + Machine Check Architecture

   + Power Management

 Keep looking this blog for further explanations and details of CPU.

Storage Components

Memory:
–Volatile: retains data only when it is powered
–Non-volatile: retains data even when it is not on
–Primary: hold programs when they are running
–Secondary: Store data and programs

DRAM vsMagnetic disks
–Access time for DRAM:40-80 nanoseconds
–Access time for Disks: 5-15 milliseconds
–However cost per megabyte of disk is 100 times less expensive

In DIRECT CODING we use 3bytes per pixel(as an accepted standard) with one byte for each primary color. Hence, each primary can have 256 different intesntiy level so we have a total of 256*256*256 poosible intensity level.

this is the true color representation. but the problem is a 1000*1000 true color image would take up 3 million bytes and also if each pixel in this image had a different color, there would only be 1 million colors so this representaion of 16.7 million diff colors is somewhat not practical.

 Hence we have LOOKUP TABLE . in this the pixel values do not code colors directly instead the have the addresses or indices into a table of color values.

there is a table having 256 entries each of 24 bits color value(8 bits per primary). Pixel values are now 1 byte telling number from 0 to 255 in the table where a color value is stored.

so a 1000*1000 image now requires 1 million+768 bytes( for the color values in lookup table). hence it allows 256 simultaneous colors that are chosen from 16.7 million possible colors.