Microprocessor

A HISTORICAL BACKGROUND

I. Funa – K. Gacos – W. Valentin – A.R. Herrera – K.A. Haspela – EE-

4A
Microprocessor Age

 The worlds first microprocessor is the Intel 4004

 4 bit microprocessor
 It addressed a mere 4096, 4-bit-wide memory locations
 The instruction set contained only 45 instructions
 Fabricated with P-channel MOSFET technology that only allowed it to
execute instructions at 50 KIPs
 Started designing in April 1970 and completed in January 1971
Microprocessor Age

 The 4-bit microprocessor debuted in early video

game systems and small microprocessor – based
control systems
 Main problem with this microprocessor is speed, word
width and memory size
Microprocessor Age

 Intel Released an an updated version of Intel 4004,

the Intel 4040. (1974)
 4040 operated at a higher speed .
 Texas instruments also produced 4-bit
microprocessors, the TMS-1000
Microprocessor AGE

 On April 1972, Intel introduced the 8008, 8-bit

version of the 4004 microprocessor
 Intel 8008
 8-bit
 Expanded memory size of 16k bytes
 Instruction set of 48 instructions
Microprocessor Age

 In 1973, Intel introduced the 8080,the first of the

modern 8 bit processors
 6 months after, Motorola Corporation released the
MC6800.
 After the MC6800, other companies produced their
own versions of 8-bit microprocessors.
What was special about 8080?

 The Intel 8080

 Addresses more memory
 10 times faster than 8008
 An addition that took 20µson 8008 based system required only 2.0µs on 8080
based system
 Compatible with transistor-transistor logic, whereas 8008 was not directly
compatible
 Made interfacing easier and less expensive
 4 times more memory than 8008
These improvements are
responsible for ushering in the
era of the 8080 and the
continuing saga of the
microprocessor.
Intel 8085

 IN 1977, Intel corporation introduced an updated version of the 8080

--- the 8085
 Intel 8085
 Slightly advanced than 8080
 An addition will require only 1.3µs
 Internal clock generator
 Internal system controller
 Higher clock frequency
Intel managed to sell well over 100million copies of 8085
 The Modern
Microprocessor
 1995

80386EX Microprocessor
32-bit Microprocessors
1986

80386 Microprocessor
1983

1978 - 1981
80826 Microprocessor

8086 & 8088 Microprocessors 16-bit Microprocessors

 8086 & 8088 Microprocessors
• 16-bit microprocessors, which executed
instructions in as little as 400 ns (2.5 MIPs or
millions of instructions per second)

• addressed 1M byte of memory

which was 16 times more memory than the
8085

1M byte of memory
contains 1024K byte-sized memory locations or 1,048,576 bytes
8086 & 8088
Microprocessors
• features a small 4- or 6-byte instruction cache or
queue that prefetched a few instructions before
they were executed and helped sped the
operation of many sequences of instructions

• increased memory size, number of instructions

and additional improved instruction set (like
multiply and divide instructions)
that’s why it’s also called CISC (complex instruction set computers)

• also provided more internal register storage space

than the 8-bit microprocessor. The additional
registers allowed software to be written more
efficiently
 80826 Microprocessor
• addressed a 16M-byte memory system instead
of a 1M-byte system

• instruction set have few additional instructions

• increased clock speed

execution of instructions: 250 ns (4.0 MIPs)

• some changes in the internal execution of the

instructions
80386 Microprocessor
• Intel’s first practical 32-bit microprocessor that
contained a 32-bit data bus and a 32-bit memory
address

• addressed up to 4G bytes of memory

• a 4G-byte memory can store an astounding 1,000,000

typewritten, double-spaced pages of ASCII text data

1G byte of memory
contains 1024M or 1, 073, 741, 824 locations
32-bit Microprocessors

• provides higher clocking speeds; significantly increases the speed of any

program that manipulates real (single-precision floating-point) numbers

• also includes a memory management unit that allowed memory resources

to be allocated and managed by the operating system

• included hardware circuitry for memory management and memory

assignment, which improved its efficiency and reduced software overhead

• instruction set is upward-compatible with the earlier 8086, 8088, and

80286 microprocessors
80386 Microprocessors

80386 Modified Versions

80386SX
addressed 16M bytes of memory through a 16-bit data and 24-bit
address bus

80386SL/80386SLC
addressed 32M bytes of memory through a 16-bit data and 25-bit
address bus. 80386SLC version contained an internal cache memory that
allowed it to process data at even higher rates.
80386EX Microprocessor
• called an embedded PC because it
contains all the components of the AT
class personal computer on a single
integrated circuit.

• The 80386EX also contains 24 lines for

input/output data, a 26-bit address bus, a
16-bit data bus, a DRAM refresh
controller, and programmable chip
selection logic.
 THE 80486
MICROPROCESS
OR
(Main Topic)
• Released by Intel in 1989, which incorporated an 80386-
like microprocessor, am 80387-like numeric
coprocessor, and an 8K-byte cache memory system
into one integrated package.
• Modified from the 80386 so that about of its
instructions executed in one clock instead of two clocks
 because the 80486 was available in a 50 MHz
version, about half of the instructions executed in 25
ns (50 MIPS)
Versions
(Sub-Topic)
 80486DX2 or Double-clocked Version
• Executed instructions at even higher speed at the rate 66 MHz, with memory
transfers executing at the rate of 33 MHz

80486DX4 or Triple-clocked Version

• Improved internal execution speed to 100 MHz with memory
transfers at 33 MHz
Other Versions
(Sub-topic)
 OverDrive Processor
• Registered trademark of Intel Corporation
• Actually a double-clocked version of the
80486DX that replaced an 80486SX or
slower-speed 80486DX
 *Triple-Clocked Version

•Produced by Advanced Micro Devices

that runs with a bus speed of 40 MHz and
a clock speed of 120 MHz
 Table 1-2 Many modern Intel and Motorola Microprocessors
Manufacturer Part Number Data Bus Width Memory Bus
Intel 8048 8 2K Internal
8051 8 8K Internal
8085A 8 64K
8086 `6 1M
8088 8 1M
8096 16 8K Internal
80186 16 1M
80188 8 1M
80251 8 16K Internal
80286 16 16M
80386EX 16 64M
80386DX 32 4G
80386SL 16 32M
80386SLC 16 32M + 8K cache
80386SX 16 16M
80486DX/DX2 32 4G + 8K cache
80486SX 32 4G + 8K cache
80486DX4 32 4G + 16K cache
Pentium 64 4G + 16K cache
Manufacturer Part Number Data Bus Width Memory Size
Pentium OverDrive 32 4G + 16K cache
Pentium Pro 64 64G + 16K L1 cache + 256K L2 cache
Pentium II 64 64G + 32K L1 cache + 256K L2 cache
Pentium III 64 64G + 32K L1 cache + 256K L2 cache
Pentium 4 64 64G + 32K L1 cache + 512K L2 cache (or
larger)
Pentium 4D (Dual Core) 64 1T + 32K L1 cache + 2 or 4M L2 cache
Core2 64 1T + 32K L1 cache + a shared 2 or 4M L2
cache
Itanium (Dual Core) 64 1T + 2.5M L1 and L2 cache + 24M L3
cache
Motorola 6800 8 64K
6805 8 2K
6809 8 64K
68000 16 16M
68008D 8 4M
68008Q 8 1M
68010 16 16M
68020 32 4G
68030 32 4G + 256 cache
68040 32 4G + 8K cache
68050 32 Proposed, but never released
68060 64 4G + 16K cache
PowerPC 64 4G + 32K cache
 THE PENTIUM
MICROPROCESSOR
• Introduced in 1993, was similar to the 80386 and 80486
microprocessors
• Originally labeled the P5 or 80586, but Intel decided not to use
• Most ingenious feature is integer processors
• Executes two instructions, which are not dependent on each other,
simultaneously because it contains two independent internal integer
processors called superscalar technology
• With another feature, jump prediction technology, speeds the
execution of program loops
• With 80486, also employs an internal floating-point coprocessor to
handle floating-point data, albeit at a five time speed improvement
• Replacesome of the RISC (reduced instruction set computer)
machines that currently execute one per clock
 Introductory Versions
• Operated with a clocking frequency of 60 MHz and 66 MHz, and a speed of 110
MIPs, with a higher-frequency 100 MHz one and one-half clocked version that
operated at 150 MIPs

Fastest Version
• Operated at 233 MHz, which is a three and one-half clocked version

Recent Version
• Included additional instructions, called multimedia extensions, or
MMX instructions
 Pentium OverDrive (P24T)

• For older 80486 systems that operate at either 63 MHz or 83 MHz

clock.
63 MHz version upgrades older 80486DX2 50 MHz
83 MHz version upgrades the 80486DX2 66 MHz; performs at a rate somewhere
between a 66 MHz Pentium and a 75 MHz Pentium

Note:
If older VESA local bus video and disk-caching controllers seem too expensive to toss out,
the Pentium OverDrive represents an ideal upgrade path from the 80486 to the Pentium
 *PowerPC
•Produced by Motorola, Apple and IBM
•A RISC microprocessor that has two
integer units and a floating-point unit.
PENTIUM PRO PROCESSOR

• A recent entry from Intel, formerly named the P6

microprocessor.
• Contains 21 million transistor, integer units, as well
as floating-point unit to increase the performance
of most software.
• Basic clock frequency was 150 Mhz and 166 MHz in
the initial offering made available in late 1995
• Internal 16K level-one (L1) cache (8K for data and
8K for instructions)
• In addition to the L1 cache, Has 256 KB L2 cache
• Has been optimized to efficiently execute 32-bit
code
• Can address either a 4G-byte memory system or
64G memory system.
• 36-bit address bus ; 64 GB address space
INTEL PENTIUM PRO
PROCESSOR
PENTIUM II AND PENTIUM XEON
MICROPROCESSORS
• Pentium II was introduced in 1997
• Introduced multimedia (MMX) instructions
• Doubled on-chip L1 cache
• 16 KB data
• 16 KB instruction
• Introduced comprehensive power management features
• Sleep
• Deep sleep
• In addition to the L1 cache

- Has 256 KB L2 cache

• Represents a new direction for Intel
• L2 cache operates at the system bus speed of 60
Mhz or 66 Mhz.
• The L2 cache and microprocessor on a circuit board
called “Pentium II module” operates at speed of 133
MHz and stores 512K bytes of information.
• In 1998, Intel change the bus speed of Pentium II.
The newer Pentium II microprocessor use a 100 MHz
bus speed
• Pentium II rated at 350 MHz, 400 MHz, and 450 MHz
all use this higher 100 MHz memory bus speed.
PENTIUM II
PENTIUM XEON MICROPROCESSOR

• In mid-1998 Intel announced a new version of the Pentium II called Xeon

• - Designed for high-end workstation snd server applications.
• - Main difference between Pentium II and Pentium II Xeon is that Xeon is
available with a L1 cache size of 32K bytes and L2 cache size of either
512K, 1M or 2M bytes.
• - this newer product represents a change in Intel’s strategy: Intelnow
produces a professional version and a home/ business version of
Pentium II microprocessor.
PENTIUM XEON
MICROPROCESSOR
PENTIUM III
MICROPROCESSOR
 Uses a faster core than
Pentium II, but it is still a
P6 or Pentium Pro
processor. (Speed is up to
1 GHz)
 Available in two
versions
 Slot
I version mounted
on a plastic cartridge
 Socket 370 or Flip chip
version
Pentium 4 and Core2
Microprocessors
  It was made in the late 2000
 The most recent version is Core2
by Intel
 It still uses a P6 architecture, but
faster than Pentium III (Pentium 4
is 3.2 GHz and faster and Core2 is
3 GHz)
 Uses RAMBUS and DDR memory
technologies
 Uses Copper instead of aluminum
 Pentium 4 and
Core2, 64bit and
Multicore
Microprocessors
 64 bit modification
 Inclusion of multiple
cores.
 Dual and quad cores
 THINGS TO
REMEMBER
 Pentium III
 Available with clock frequencies of up to
1 GHz
 Available in two versions
 Slot I version – contains 512K cache and
runs at one half clock speed
 Flip Chip version - contains 256K cache and
runs at the clock speed
 Both of versions use memory speed of 100
MHz (Celeron uses 66 MHz)
 Pentium 4 and Core2
 Pentium 4 is available at 3.2 GHZ and higher
 Core2 speeds up to 3GHz
 Chipsets use RAMBUS or DDR memory tech
 These higher microprocessor speeds are available
by an improvement in the size of the internal
integration (0.045 micron to 45 nm)
 Level 1 cache size is changed form 32K to 8K to
64K
 Level 2 cache size is 256K (Pentium Coppermine),
and latest version is 512K cache.
 Pentium 4 Extreme edition contains 2M 2L cache,
Pentium 4e contains a 1M Level 2 cache Core2
contains either 2M or 4M L2 cache.
 Copper interconnections inside the microprocessor
instead of aluminum
 Pentium 4 and Core2, 64 bit
and Multiple Core
Microprocessors
 64 bit microprocessor allows to
address more than 4G bytes of
memory
 64 bit machine allows 64 bit
integer arithmetic
 In a multiple core microprocessors,
each core execute separate task
so that it increases the speed of
execution. Programs that do this
are called MULTITHREAD apps.
 FUTURE OF
MICROPROCESSORS
FUTURE OF
MICROPROCESSORS
We will be expecting a change in RISC technology, but more likely
are improvements to a new technology jointly by Intel and Hewlett-
Packard called hyper threading technology. Even this new
technology embodies the CISC instruction set of the 80X86 family
of microprocessors, so that software for the system will survive.
The Basic premise behind this technology is that many
microprocessors communicate directly with each other, allowing
parallel processing without any change to the instruction set or
program. Currently the superscalar technology uses many
microprocessors, but they all share the same register set. This new
technology contains many microprocessors, each containing its
own register set that is linked with the other microprocessors’
registers. This technology offers true parallel processing without
writing any special program.
FUTURE OF
MICROPROCESSORS
In 2002, Intel released a new microprocessor architecture that is 64
bits in width and has 128-bit data bus. This new architecture,
named the Itanium, is a joint venture called EPIC(Explicitly Parallel
Instruction Computing) of Intel and HP.
Itanium :
Allows greater parallelism than traditional architectures
Include 128 general purpose integer registers, 128 floating-point
registers, 64 predicate registers, and many execution units.
Conceptual
views of
Intel
Microprocess
ors
FUTURE OF
MICROPROCESSORS
The major changes of we will be
expecting are:
Wider memory path
Faster memory speed
Faster mass storage
Intel Core
Intel Core

 Intel Core is a line of mid-to-high end consumer, workstation,

and enthusiast central processing units (CPU) marketed by Intel
Corporation. These processors displaced the existing mid-to-high
end Pentium processors of the time, moving the Pentium to the
entry level, and bumping the Celeron series of processors to low
end. Identical or more capable versions of Core processors are
also sold as Xeon processors for the server and workstation
markets.
 As of June 2017, the lineup of Core processors included the Intel
Core i9, Intel Core i7, Intel Core i5, and Intel Core i3, along with
the Y - Series Intel Core CPUs
Enhanced Pentium M based

 Core Duo
 Intel Core Duo (product code 80539) consists of two cores on one die, a
2 MB L2 cache shared by both cores, and an arbiter bus that controls both
L2 cache and FSB (front-side bus) access.
 Core Solo
 Intel Core Solo (product code 80538) uses the same two-core die as the
Core Duo, but features only one active core. Depending on demand, Intel
may also simply disable one of the cores to sell the chip at the Core Solo
price—this requires less effort than launching and maintaining a separate
line of CPUs that physically only have one core. Intel used the same
strategy previously with the 486 CPU in which early 486SX CPUs were in
fact manufactured as 486DX CPUs but with the FPU disabled.
64-bit Core microarchitecture-based

 Core 2 Solo
 the Core 2 Solo, introduced in September 2007, is the
successor to the Core Solo and is available only as an ultra-
low-power mobile processor with 5.5 Watt thermal design
power. The original U2xxx series "Merom-L" used a special
version of the Merom chip with CPUID number 10661 (model
22, stepping A1) that only had a single core and was also used
in some Celeron processors. The later SU3xxx are part of
Intel's CULV range of processors in a smaller µFC-BGA 956
package but contain the same Penryn chip as the dual-core
variants, with one of the cores disabled during manufacturing.
64-bit Core microarchitecture-based

 Core 2 Duo
 The majority of the desktop and mobile Core 2 processor variants
are Core 2 Duo with two processor cores on a
single Merom, Conroe, Allendale, Penryn, or Wolfdale chip. These
come in a wide range of performance and power consumption,
starting with the relatively slow ultra-low-power Uxxxx (10 W) and
low-power Lxxxx (17 W) versions, to the more performance
oriented Pxxxx (25 W) and Txxxx (35 W) mobile versions and the
Exxxx (65 W) desktop models. The mobile Core 2 Duo processors
with an 'S' prefix in the name are produced in a smaller µFC-BGA
956 package, which allows building more compact laptops.
64-bit Core microarchitecture-based

 Core 2 Quad
 Core 2 Quad processors are multi-chip modules consisting of two dies similar to
those used in Core 2 Duo, forming a quad-core processor. This allows twice the
performance of a dual-core processors at the same clock frequency in ideal
conditions.
 Initially, all Core 2 Quad models were versions of Core 2 Duo desktop
processors, Kentsfield derived from Conroe and Yorkfield from Wolfdale, but
later Penryn-QC was added as a high-end version of the mobile dual-core Penryn.
 The Xeon 32xx and 33xx processors are mostly identical versions of the desktop
Core 2 Quad processors and can be used interchangeably.
64-bit Core microarchitecture-based

 Core 2 Extreme
 Core 2 Extreme processors are enthusiast versions of
Core 2 Duo and Core 2 Quad processors, usually with a
higher clock frequency and an unlocked clock
multiplier, which makes them especially attractive
for overclocking. This is similar to
earlier Pentium processors labeled as Extreme Edition.
Core 2 Extreme processors were released at a much
higher price than their regular version, often $999 or
more.
Nehalem microarchitecture-based

 Core i3
 intel intended the Core i3 as the new low end of the performance processor line
from  Intel, following the retirement of the Core 2 brand.
 The first Core i3 processors were launched on January 7, 2010.
 The first Nehalem based Core i3 was Clarkdale-based, with an integrated GPU and
two cores. The same processor is also available as Core i5 and Pentium, with
slightly different configurations.
 Core i3
 The Core i3-3xxM processors are based on Arrandale, the mobile version of the
Clarkdale desktop processor. They are similar to the Core i5-4xx series but running
at lower clock speeds and without Turbo Boost. According to an Intel FAQ they do
not support Error Correction Code (ECC) memory. According to motherboard
manufacturer Supermicro, if a Core i3 processor is used with a server chipset
platform such as Intel 3400/3420/3450, the CPU supports ECC with UDIMM.When
asked, Intel confirmed that, although the Intel 5 series chipset supports non-ECC
memory only with the Core i5 or i3 processors, using those processors on a
motherboard with 3400 series chipsets it supports the ECC function of ECC
memory. A limited number of motherboards by other companies also support ECC
with Intel Core ix processors; the Asus P8B WS is an example, but it does not
support ECC memory under Windows non-server operating systems.
Nehalem microarchitecture-based

 Core i5
 The first Core i5 using the Nehalem microarchitecture was introduced on
September 8, 2009, as a mainstream variant of the earlier Core i7, the Lynnfield
core. Lynnfield Core i5 processors have an 8 MB L3 cache, a DMI bus running at 2.5
GT/s and support for dual-channel DDR3-800/1066/1333 memory and have Hyper-
threading disabled. The same processors with different sets of features (Hyper-
Threading and other clock frequencies) enabled are sold as Core i7-8xx and Xeon
3400-series processors, which should not be confused with high-end Core i7-9xx and
Xeon 3500-series processors based on Bloomfield. A new feature called Turbo Boost
Technology was introduced which maximizes speed for demanding applications,
dynamically accelerating performance to match the workload.
Nehalem microarchitecture-based

 Core I7
 Intel Core i7 as an Intel brand name applies to several families of desktop and
laptop 64-bit x86-64 processors using the Nehalem, Westmere, Sandy Bridge, Ivy
Bridge, Haswell, Broadwell, Skylake, and Kaby Lake microarchitectures.
 Intel introduced the Core i7 name with the Nehalem-based Bloomfield Quad-core
processor in late 2008. In 2009 new Core i7 models based on
the Lynnfield (Nehalem-based) desktop quad-core processor and
the Clarksfield (Nehalem-based) quad-core mobile were added, and models based
on the Arrandale dual-core mobile processor (also Nehalem-based) were added in
January 2010. The first six-core processor in the Core lineup is the Nehalem-
based Gulftown, which was launched on March 16, 2010. Both the regular Core i7
and the Extreme Edition are advertised as five stars in the Intel Processor Rating.
Next Gen Microarchitecture

 Sandy Bridge
 Ivy Bridge
 Haswell
 Broadwell
 Skylake
 Kaby Lake
 Coffee Lake
Core I9

 Core i9 Extreme Edition:

 Core i9-7980XE: (2.6GHz, 4.4GHz burst) 18 cores/36 threads, $1,999 
 Core i9:
 Core i9-7960X: (2.8GHz, 4.4GHz burst) 16 cores/32 threads, $1,699
 Core i9-7940X: (3.1GHz, 4.4GHz burst) 14 cores/28 threads, $1,399
 Core i9-7920X: (3.1GHz, 4.4GHz burst) 12 cores/24 threads, $1,199
 Core i9-7900X: (3.3GHz, 4.5GHz burst) 10 cores/20 threads, $999