i

Intel I3 Processor

Arpit

Section- E2E43, roll no.-A34, Reg. no.-11106840

School of Electronics and communication

Lovely professional university, phagwara, Punjab

Abstract-This term paper on Intel i3 processor is

to define the role played by the Intel processor in

the field of data manipulation and graphic display.

This term paper present report on the architecture

of i3 processor made by Intel and improvement in

it from predecessor.

1-Introduction

The Intel i3 processor with Intel HD Graphics

offers an unparalleled computing Experience. This

revolutionary new architecture allows for new

levels of intelligent performance and advanced

media and graphics features—all while being

energy efficient. The processor include an

Integrated Memory Controller (IMC) making

them monolithic processors. The IMC and the

multiple processor cores are connected by the new

Quick Path Interconnect (QPI).

2-History

Intel is the world’s biggest company which is

famous for manufacturing the best processors ever

created in processor history. Every time they have

introduced something interesting and new in their

processors and devices. They introduced the series

of processors in 1940s and till now the advanced

version of their processors are still considered the

best. Some of the famous processor are from the

latest group of the family ‘core’. Intel core is the

processors i3 family which is famous for its latest

revolutionary structure and integrated architecture

which also provide the advantage of the parallel

computing. It’s also wonderful in providing the

users with the excellent graphical user interfaces.

Intel was founded in 1968 and its first product was

Intel 3101 produced in 1969, Intel first product

was world’s first solid state memory device with

16 x 4-bit SRAM. Intel 1103 came in 1970 was

world first DRAM product with 1K-bit PMOS

and it was used in HP 9800 series computers. By

1972, it became world bestselling memory chip,

defeating Magnetic memory

Intel MCS Family

MCS Family

Intel CPU

MCS-4 4004

MCS-40 4040

MCS-8 8008

MCS-80 8080

MCS-85 8085

MCS-86

8086, 8088, 80186, 80188,

80286, 80386, 80486,

Pentiums

Intel 4004 produced in 1971 was world first

“general purpose” micro-processor and its Lead

ii

designers were– Ted Hoff, Federico Faggin, Stan

Mazor, and Masatoshi Shim. Intel 4004 have

Word width: 4-bit, 2300 transistors, Clock

frequency: 108KHz/500/740.It have 46

instructions, Registers: 16 x 4-bit, Stack: 12 x 4-

bit with Address space of 1Kb for program and

4Kb for data. In1972 came the Intel 8008 - world

first 8-bit microprocessor whose designers were–

Ted Hoff, Stan Mazor, Hal Feeney, and Federico

Faggin. Intel 8008 have word width: 8-bit, Clock

frequency: 800 KHz with 3500 transistors, 48

instructions, Registers: 6 x 8-bit, Stack: 17 x 7-bit

and Address space: 16KB. In 1974 came the Intel

8080 whose Lead designers were – Feder1ico

Faggin, Masatoshi Shima and Stan Mazor. "The

8080 really created the microprocessor market”. It

was used in MITS Altair 8800 in 1975 and termed

as “Microcomputer”. It have Word width: 8-bit

with 4500 transistors, Clock frequency: 2M-

3MHz and Address space: 64KB, Registers: 6 x 8-

bit, IO ports and Stack pointer.

Intel 16-bit Microprocessors came in 1978 when

Intel launched Intel 8086, first x86 family

microprocessor with Source compatibility with

80xx lines and its Followers were: 8088 (1979),

80186 (1982). It have 16-bit: all registers, internal

and external buses with 29,000 transistors, 5MHz

of clock frequency, 20-bit address bus, 4MB

address space and 16-bit register - segmentation

programming. IBM PC in 1981 used 8088

processor. Then came the Intel 80286 in 1982

with 134,000 transistors with clock frequency of

6M-8MHz and 1.5 MIPS it was used by IBM

PC/AT in 1984 it was Designed for multi-tasking

with MMU “protection mode”. Intel i432, Intel

first 32-bit microprocessor design it was “intel

Advanced Processor architecture”. Started in 1975

as the 8800, follow-on to the existing 8008 and

8080 CPUs, intended purely 32-bit, to be Intel

backbone in the 1980s, to support Ada, LISP,

advanced computations, the HW supports to all

the good terms with Object Oriented

programming and capability-based addressing,

multi-tasking and IPC, Multiprocessing, Fault

tolerance. But the Problems with it was two-chip

implementation, lack of cache, bit-aligned

variable length instructions. It Failed: ¼

performance of 286 as of 1982. Then in 1980 Intel

8087 came into picture with First floating-point

coprocessor for 8086 lines, its Performance was:

+20% ~ 5x; it have Floating registers form 8-level

stack: st0~st7 work in two mode: 8-bit/16-bit

follow IEEE 754 standard. Then follows: Intel

80287 – 16-bit and Intel 80387, 80487 – 32-bit.

Starting from Intel 80486DX, Pentium and later

model has on chip floating point unit and “DX”

was used for on-chip FP capability.

Intel introduced 80386 processor in 1985 it was

Intel first X86 32-bit flat memory model with

4GB space. 80386 instruction set, programming

model, and binary encodings were the common

denominator for all IA-32, i386, x86 series. It has

Paging to support VM, hardware debugging, first

use of pipeline it wasn’t necessarily a big

performance improvement over 80286,it contain

275,000 transistors with clock frequency of

12MHz initially, later 33MHz and 11.4MIPS.

Compaq: first PC using 80386, legitimize PC

“clone” industry.

In 1985 Intel produced i960. Intel 80960, Intel

was first RISC (Reduced instruction set

computing) microprocessor it was the Best-selling

embedded microcontroller at the time. It was

intended to replace 80286/i386, and for UNIX

systems, it used Berkeley RISC, flat memory

model, superscalar structure but Dropped after

acquiring Strong ARM in late 90’s when its

Price/performance/power remain no longer

competitive and team went to design another i386

processor.

Intel 80486 was another processor introduced by

Intel in 1989 with Improvements in Atomic

instructions, On-die 8KB SRAM cache, tightly

coupled pipelining: 1 IPC with clock frequency of

50MHz and 40MIPS on average and 50MIPS at

peak it has Integrated FPU (no longer need x87).

It was first chip which exceeds 1M transistors.

Now the competitor were more manufacturers,

AMD Am5x86, Cyrix Cx5x86, and Motorola

68040 in Macintosh Quadra.

iii

In 1989 Intel presented new masterpiece Intel i860

with entirely new RISC microprocessor, high-

performance FP operations it have 32-bit ALU

core, and 64-bit FPU (adder, multiplier, GPU)

along with Register sets: 32 x 32-bit integer, 16 x

64-bit FP. It GPU uses FP registers as 8 x 128-bit,

with iSIMD (Influenced MMX), 64/128-bit buses,

fetch 2 x 32-bit instructions. It was dropped in

mid-90 because Compiler support was mission

impossible and Context switch took 62 - 2000

cycles which was Unacceptable for GP CPU, it

was Incompatible with X86, Confusing the market

with Intel 486 CISC. It was being used in some

parallel computers and graphic workstations.

Intel introduced Pentium in

1993. Pentium means “5”, because court

disallowed number based trademark, later

Pentium was used in many Intel processors. P5

micro-architecture first used X86 superscalar

micro-architecture with dual integer pipelines,

separate D/I caches, 64-bit external data-bus and

60M-300MHz (at 75 MHz -126.5 MIPS) .It’s

Competitors were X86: AMD K5/K6, Cyrix 6x86,

etc. from Pentium processor Intel started to use a

cooler.

In 1996 Intel launched MMX which has SIMD

instruction set, introduced with P5 it has “Matrix

Math Extensions”, mainly for graphics and 8 x 64-

bit integer registers MM0 ~ MM7, alias of FPU

ST0 ~ ST7. Integer was not enough soon due to

gfx cards. Intel introduced SSE in 1999 and

started with Pentium-III it have new XMM

register set with 70 new instructions. It has "Intel

Wireless MMX Technology" and then Intel

Pentium Pro in 1995, P6 (or i686) was completely

new apart from Pentium (P5) it had no. of

transistors: Pentium 3.1M, Pentium MMX 4.5M,

Pentium Pro 5.5M with Speculative execution,

RISC-like micro-ops and three pipelines, 2

integer, 1 for floating point. Innovative on-

package level-2 cache but manufacturing did

allow on-die L2 cache it had Same CPU clock

rate, non-blocking, SMP advantage. Dies had to

be bonded early, it had Low yield rate and high

price it had36-bit address bus (PAE), low 16-bit

performance but Performance was better than best

RISC with SPECint95.

Pentium II by Intel in 1997 had 7.5M transistors

and Slot replaced Socket with a daughterboard,

solved the issues of off-package L2 cache in

Pentium Pro with half CPU clock. It implemented

MMX, improved 16-bit performance. Celeron and

Xeon was launched in 1998, Celeron: no on-die

L2-cache. And Pentium II Xeon: L2-cache,

100MT/s, SMP. Intel launched Pentium III in

1999 it introduced SSE for FP and vector

processing it had on-die L2 cache with .18um

Coppermine. Intel then Streaming SIMD

Extensions in 1999 in which MMX uses FP

registers for SIMD data, and has only integer

SIMD, SSE introduces separate XMM registers.

Intel Xscale came in light in 1997 when Intel

acquired Strong ARM from DEC, 1997 to replace

the RISC processors i860 and i960. It had Strong

ARM implemented ARMv4 ISA. It’s Successor,

Xscale implemented ARMv5 with Seven-stage

integer and an eight-stage memory super pipelined

microarchitecture, 32KB data cache and 32KB

instruction cache.• Xscale processor family had

Application Processors (with the prefix PXA), I/O

Processors (with the prefix IOP), Network

Processors (with the prefix IXP), Control Plane

Processors (with the prefix IXC).• Intel sold

Xscale PXA business to Marvell in 2006.

Intel Itanium was in limelight in 2001 which was

originated from HP, EPIC: explicitly parallel

instruction computing. All believed EPIC would

supplant RISC and CISC. Compaq and SGI gave

up Alpha and MIPS, Microsoft and SUN etc.

developed Operating system for it and in 1999,

Intel named it Itanium it had Speculation,

prediction, predication, and renaming with 128

integer registers, 128 FP registers, 64 one-bit

predicates, and eight branch registers 128-bit

instruction word has 3 instruction, dual-issue, max

iv

6 IPC, X86 support in HW initially and then

purely in SW.

Intel introduced Pentium 4 in 2000 with Net Burst

microarchitecture (P68, successor to P6), it Pursue

higher frequency, smaller IPC with Hyper

Pipelined: 20-stage Willamette. It had Rapid

Execution Engine: Two ALUs in the core are

double-pumped and execution Trace Cache,

SSE2, L3-cache (Extreme Edition). It was

confined with Hyper-Threading Technology. But

its performance worse than Northwood with

similar clock as designed to be 10GHz, but

achieved 3.8GHz. Core-based: 27W,

Pentium4:115W, Pentium4M:88W. Then comes

Pentium D – Dual-core which was abandoned due

to High power consumption and heat intensity and

Inability to increase clock speed, and inefficient

pipeline. Intel introduced Pentium M in 2003

derived From Pentium III, based on P6

microarchitecture. FSB interface of Pentium 4,

SSE2, much larger cache, improved

decoding/issuing FE, L2 cache only switches on

the portion being accessed. It has dynamically

variable clock frequency and core voltage it has

1.6 GHz Pentium M performance > 2.4 GHz

Pentium 4-M. Next generation of it released as

Intel Core brand on Jan 2006. Core 2: Intel-64

Core microarchitecture came in light in July 2006.

It has larger cache in size. Then comes the Intel

Tick-Tock Model which was introduced since

2007 to describe progress cadence where “Tick“:

shrinking of process technology – same

microarchitecture. And “Tock“: new

microarchitecture – same process. Tick-Tock is

expected alternating every year.

Intel

Nehalem came in 2008 it was successor of Core

micro-architecture and was planned as Net burst

evolution, but then a completely different design

of microarchitecture with size of 45nm. It was

Multi-core with on-package GPU and integrated

memory controller, Integrated PCI-E and DMI

replacing Northbridge. It had level branch

predictor and 20% gain performance/clock, 30%

cut power/performance and on this architecture

Intel introduced Core i3, i5, i7, Celeron, Pentium,

Xeon. Intel introduced Atom Processors in 2008

which was based on Bonnell microarchitecture,

45nm in size with dual-issue in order, 16-stage

pipeline and only around 4% of instructions

produce multiple micro-operations. It can contain

both a load and a store with an ALU operation

with Partial revival of old principle in P5 and 486

for performance/watt.

Intel Sandy Bridge came into light in 2011, it was

new microarchitecture after Nehalem, 32nm in

size and it shared L3 cache for cores, including

GPU, it has two load/store ops/cycle for memory

channel.in this there is Ring bus interconnect

between Cores, Graphics, Cache and System

Agent Domain. As compared to Nehalem, 17%

gain in performance/clock over Lynnfield, 2x

graphics over Clarkdale and then in 2012 came

the Ivy Bridge architecture with, size of 22nm and

3D gates.

3-Architecture

Intel i3 is a 64 bit microprocessor and is available

in different versions depending on different

architecture, and there are three main architecture

on which Intel i3 processor is designed.

a) Nehalem Computer Architecture

The predecessor to Nehalem, Intel’s Core

architecture, made use of multiple cores on a

single die to improve performance over traditional

single-core architectures. But as more cores and

processors were added to a high-performance

system, some serious weaknesses and bandwidth

bottlenecks began to appear. After the initial

generation of dual-core Core processors, Intel

began a Core 2 series processor which was not

much more than using two or more pairs of dual-

v

core dies. The cores communicated via system

memory which caused large delays due to limited

bandwidth on the processor bus. Adding more

cores increased the burden on the processor and

memory buses, which diminished the performance

gains that could be possible with more cores. The

new Nehalem architecture sought to improve

core-to-core communication by establishing a

point-to-point topology in which microprocessor

cores can communicate directly with one another

and have more direct access to system memory.

The approach to the Nehalem architecture is more

modular than the Core architecture which makes it

much more flexible and customizable to the

application. The architecture really only consists

of a few basic building blocks. The main blocks

are a microprocessor core (with its own L2 cache),

a shared L3 cache, a Quick Path Interconnect

(QPI) bus controller, an integrated memory

controller (IMC), and graphics core. With this

flexible architecture, the blocks can be configured

to meet what the market demands. For example,

the Bloomfield model, which is intended for a

performance desktop application, has four cores,

an L3 cache, one memory controller, and one QPI

bus controller. Server microprocessors like the

Beckton model can have eight cores, and four QPI

bus controllers. The architecture allows the cores

to communicate very effectively in either case.

In Nehalem architecture I3 has 32 nm process

technology with following specifications: 2

physical cores/4 threads, 64 Kb L1 cache, 512 Kb

L2 cache, 4 MB L3 cache, Introduced January,

2012, Socket 1156 LGA, 2-channel DDR3,

Integrated HD GPU with different variants.

b) Sandy bridge/ Ivy bridge architecture

Sandy Bridge is the codename for

a microarchitecture developed by Intel beginning

in 2005 for central processing units in computers

to replace the Nehalem microarchitecture. Intel

demonstrated a Sandy Bridge processor in 2009,

and released first products based on the

architecture in January 2011 under

the Core brand. Sandy Bridge implementations

targeted a 32 nanometer manufacturing process

based on planar double-gate transistors. Intel's

subsequent product, codenamed Ivy Bridge, uses

a 22 nanometer process.

The Ivy Bridge die shrink, known in the Intel

Tick-Tock model as the "tick", is based on Fin

FET (non-planar, "3D") tri-gate transistors. Intel

demonstrated the Ivy Bridge processors in 2011.

Developed primarily by the Israel branch of Intel,

vi

the codename was originally "Gesher" (meaning

"bridge” in Hebrew). The name was changed to

avoid being associated with the defunct Gesher

political party; the decision was led by Ron

Friedman, vice president of Intel managing the

group at the time. Intel demonstrated a Sandy

Bridge processor with A1 stepping at

2 GHz during the Intel Developer Forum in

September 2009. Upgraded features from

Nehalem include:

32 KB data + 32 KB instruction L1

cache (4 clocks) and 256 KB L2

cache (11 clocks) per core.

Shared L3 cache includes the processor

graphics (LGA 1155).

64-byte cache line size.

Two load/store operations per CPU cycle for

each memory channel.

Decoded micro-operation cache and enlarged,

optimized branch predictor.

Improved performance for transcendental

mathematics. With 256-bit/cycle ring bus

interconnect between cores, graphics, cache

and System Agent Domain.

Intel Quick Sync Video, hardware support for

video encoding and decoding. And it have up

to 8 physical cores or 16 logical cores

through Hyper-threading.

Integration of the GMCH (integrated graphics

and memory controller) and processor into a

single die inside the processor package. In

contrast, Sandy Bridge's predecessor,

Clarkdale, has two separate dies (one for

GMCH, one for processor) within the

processor package. This tighter integration

reduces memory latency even more.

A 14- to 19-stage instruction pipeline,

depending on the micro-operation cache hit or

miss.

Intel continues to drive platform enhancements

that increase the overall user experience. Some of

these enhancements include areas such as

connectivity, manageability, security, and

reliability, as well as compute capability. One of

the means of significantly increasing compute

capability is with Intel multi-core processors

delivering greater levels of performance and

performance-per-watt capabilities. The move to

multi-core processing has also opened the door to

many other micro-architectural innovations to

continue to even further improve performance.

Intel Core microarchitecture is one such state-of-

the-art micro architectural update that was

designed to deliver increased performance

combined with superior power efficiency. As

such, Intel Core microarchitecture is focused on

enhancing existing and emerging application and

usage models across each platform segment,

including desktop, server, and mobile.

4-Feature

Intel i3 is based on Hyper-Threading Technology

and the improvements to Intel Smart Cache

combine to create dynamic and adaptive

performance. Add the integration of the memory

controller and the graphics to the processor and the

Intel Core i3 processor gets things done faster

and more efficiently. Intel Hyper-Threading

Technology offers more computer muscle while

reducing wait time. Intel Smart Cache improves

responsiveness by providing faster access to

data. Intel HD Graphics is the ideal graphics

solution for your everyday visual computing needs.

vii

Performance has an immediate, impact on

what you do on your PC. Accelerate your

productivity, inspire your digital creations, and

enjoy video smoothness and music quality on

a system with the Intel Core i3 processor—the

smart choice for home and office. The basic

feature of the i3 are highly improved as compared

to the previous version of the processor by Intel,i3

offers the perfect accuracy and high performance

and response rate which in result provide the users

with the high throughput rates, also reduced time

in executing the programs by the processor. The

Intel i3 processor is fully equipped by the latest HD

graphics with powerful and video engine that

provide smooth high quality display along with 3d

graphics capabilities. On the whole i3 processor

can be considered as the high graphical and

multimedia display processors for daily computing.

Intel i3 processor also provide hyper threading

technology to its users which enable the

multitasking capability of both user and system.

The systems with i3 processor can perform

execution and compilation of two tasks

simultaneously without causing the executing

delays and debuggers errors. They are also so

responsive that output of the program can be

generated at the same time too. We can easily say

that Intel i3 are the best choice for homes and

offices. More than seven applications can run

simultaneously on the system with i3 processor

built on the motherboards. I3 processor are the

smarter, faster and more adaptive in all kinds of

networking scheme. They can be used with any of

the hard disk configurations. They are also famous

in the market with the name of desktop processor

because of the great quality resolution they have.

Integrated components on the motherboards also

makes i3 processors unique in their architecture

and circuit installations.

I3 processors have 3.06 GHz and 2.93 GHz core

speed which is very high as compared to the

previous configurations of the Intel processors.

They have 4 processing threads that enables

multithreading and multitasking. 4 megabyte

additional cache memory is also provided inside

the processor. Double channeled DDR with 1333

MHz memory sequence.

5-Advantage

I3 processors have remarkable advantages that are

of great use in the field of computers and

technology. Some of them are: dual processor have

the capability to run two independent program with

one hardware. I3 processor have improved Pentium

base, they have totally new architecture with more

integrations and high speed performance structure.

Hyper threading technology also enables the user

to enjoy the high speed and better performance

with more reliable outputs. It has 4 tasking threads

that allows user to easily execute 3-4 programs at a

time. Smart memory and cache sequence allows

user to enjoy the optimized and efficient data

access both direct and sequentially. Effect ive

shortcuts have reduced the access time of the file

and system. HD graphical features also make these

processors distinguished from the others because

they are considered as best in their resolution. I3

have advantage because it have different power

management and thermal management unit. Intel

Core processors include an Integrated Graphics

Device (IGD) providing excellent graphica l

capabilities. For the Intel Core processor models

that do not have an IGD, a PCI Express (PCIe)

interconnect is integrated into all processors to

support up to PCIe x16 video cards.

6) Difference from predecessor

Ivy Bridge CPUs (3rd gen) are basically on

average 6% more processing power than Sandy

Bridge CPUs (2nd gen). They also use a little less

power. The main difference is the new Intel HD

4000 which is on average around 35% - 40% more

powerful than the older Intel HD 3000. Sounds

impressive, but it is still relatively weak compared

to desktop graphic cards. The Intel HD 4000 is a

bit slower than AMD's most powerful graphic

core found in the older Llano A8 APUs.

Compared to desktop graphic cards, the mobile

version of the Intel HD 3000 is a little slower than

the Radeon HD 5450, while the Intel HD 4000

would be a little slower than the Radeon HD

5550.

viii

I3 processor is different from the predecessor

because it have: Integrated Memory Controller

(IMC) where, IMC offers high bandwidth and low

latency for memory I/O leading to faster memory

read and write cycle’s along with Hyper-

Threading Technology (HT); Hyper-Threading

technology allows one physical processor core to

be seen as two logical processors by firmware and

software. Each logical processor can execute a

thread allowing for two concurrent threads to be

executed. And Turbo Boost Technology - A

feature that automatically allows processor cores

to run faster than its base operating frequency

when other cores are not being utilized. Automatic

performance boost. It also provide service of

Quick Path Interconnect (QPI) which provide a

new high bandwidth, low latency bus that

connects processor cores and memory.

7) List of 64 bit i3 processor

Sandy Bridge – 32 nm process technology

2 physical cores/4 threads

32+32 Kb (per core) L1 cache

256 Kb (per core) L2 cache

3 MB L3 cache

624 million transistors

Introduced January, 2012

Socket 1155 LGA

2-channel DDR3-1333

Variants ending in 'T' have a peak TDP of 35 W, others 65 W

Integrated GPU

All variants have peak GPU turbo frequencies of 1.1 GHz

Variants ending in 'T' have GPUs running at a base frequency of 650 MHz; others at 850 MHz

Variants ending in '5' have Intel HD Graphics 3000 (12 execution units);

others have Intel HD Graphics 2000 (6 execution units)

Variants

i3-2100T – 2.5 GHz

i3-2120T – 2.6 GHz

i3-2100 – 3.1 GHz

i3-2102 – 3.1 GHz

i3-2105 – 3.1 GHz

i3-2120 – 3.3 GHz

i3-2125 – 3.3 GHz

i3-2130 – 3.4 GHz

Ivy Bridge – 22 nm Tri-gate transistor process technology

2 physical cores/4 threads

32+32 Kb (per core) L1 cache

256 Kb (per core) L2 cache

3 MB L3 cache

Introduced September, 2012

Socket 1155 LGA

2-channel DDR3-1600

Variants ending in '5' have Intel HD Graphics 4000; others have Intel HD Graphics 2500

All variants have GPU base frequencies of

650 MHz and peak GPU turbo frequencies of 1.05 GHz

TDP 55 W

Variants

i3-3220T – 2.8 GHz

i3-3240T – 2.9 GHz

i3-3220 – 3.3 GHz

i3-3225 – 3.3 GHz

i3-3240 – 3.4 GHz

References

1-A Brief History of Intel CPU Microarchitectures

by Xiao-Feng Li-2/10/13.

2- White paper on inside Intel Core

Microarchitecture by Intel.

3-Intel Nehalem Computer Architecture by Trent

Rolf/University of Utah Computer

Engineering/CS 6810 /Final Project/December

2009

4-Intel processor 4th generation/datasheet by Intel

5-http://www.wifinotes.com/computer-hardware-

components/I3- intel-processors.html

6-

http://en.wikipedia.org/wiki/List_of_Intel_microp

rocessors#Core_i3