RAM

RAM is an abbreviation for Random Access Memory. It usually refers to "temporary"

memory, as when the system is shut down, the memory is lost. This is why the memory is
referred to as being "random," as any piece of information can be circulated through the
memory regardless of its location and its relation to any other information within the
RAM. The first of the many RAM’s to come in the future was created in 1951 and was
used until the late 1960s and early 1970s.

Types of RAM’s
The most widely used RAM’s today are SRAM’s (static RAM’s), which stores data in a
state of a flip-flop, or DRAM’s (dynamic RAM’s), Flash, and EPROM, which stores data
as a charge in a capacitor. Another well known type of RAM, a ROM, is a type of a RAM
that has permanently enabled/disabled selected transistors by using a metal mask. ROMs
thus cannot store any further charges.

Flash memory is also widely used. Because SRAM’s and DRAM’s are known to be
highly volatile, many new products adopt the flash memory technology. Some examples
of devices using flash memory include portable music players, scientific calculators,
mobile phones, and even certain types of personal computers such as the One Laptop per
Child (OLPC) devices. Flash memory is divided into two types: the NOR type and the
NAND type. The NOR type is the one able to conduct a random access, therefore being
widely used as a ROM in today's market. However, as the NOR is not able to, it is often
used in flash USB drives.

In today's computers, RAM’s are shipped in a form of module known as DRAM modules
or memory modules. It is about the size of a chewing gum and with the Plug n Play
technology; these can easily be replaced by taking it out of the port and replacing it with
a new one. Not to forget, there is also an extremely small amount of RAM (known as
SRAM’s) within CPUs, motherboards, hard drives, and other parts of the system.

RAM Swapping
Sometimes people will need more memory than what the RAM installed can offer during
intensive application cycles. During these sessions, the computer will conduct a task
known as swapping. This is when the computer temporarily uses the hard drive space in
place of the memory. Thrashing is the term for constantly swapping, but as the name's
connotation suggest, constant thrashing can decrease the overall system performance.
More RAM being added can reduce the swapping rate.

Security Concerns
SDRAM’s are known to not fully shut off when the power is cut. Instead, the stored
memory is faded away. While this fading away only takes seconds in a normal room

1
temperature, which is not the case in lower temperatures, which then the fading away can
take up to minutes. Therefore, encrypted information could potentially be leaked as it is
still stored in the memory.

Recent Developments
Currently non-volatile RAM’s are under development, which mean that memory can still
will be stored when the power is shut off. The more advanced step yet to have been taken
was in 2004 when Nantero developed a fully functioning carbon nanotube memory type
holding 10 GB. Recently in 2006, "Solid State Drives" with sizes of 150 GB and a faster
speed than the normal hard drives have been created.

SRAM is Static RAM.

SRAM is used in small amounts in computers where very fast RAM is required, such as
in the L2 cache of many CPU's. SRAM is often contrasted with DRAM (Dynamic
RAM). Dynamic is much less expensive than SRAM, but is usually slower and must
constantly be refreshed in order to preserve its contents.

Types of SRAM include:

• Asynchronous Static RAM

• Synchronous Burst Static RAM
• Pipeline Burst Static RAM

What is NVRAM?
NVRAM is an acronym for Non-Volatile Random Access Memory. NVRAM is a type of
Random Access Memory (RAM) that retains its information when power is turned off.
The NVRAM is a small 24 pin DIP (Dual Inline Package) integrated circuit chip and is
thus able to obtain the power needed to keep it running from the CMOS battery installed
in your motherboard. It keeps track of various system parameters such as serial number,
Ethernet MAC (Media Access Control) address, HOSTID, date of manufacture, etc.
NVRAM is therefore a type of non-volatile memory that offers random access.

Types of NVRAM
One type of NVRAM is SRAM that is made non-volatile by connecting it to a constant
power source such as a battery. Since SRAM requires continual power supply in order to
maintain its data, an NVRAM that is made from an SRAM will need to use an available
power supply to make sure it continues working.

Another type of NVRAM uses EEPROM (Electrically Erasable Programmable Read-

Only Memory) circuit chips to save its information when power is turned off. In this case,
NVRAM is composed of a combination of SRAM and EEPROM chips incorporated into
a single semi-conductor die.

2
Benefits of NVRAM
 NVRAM chips work like static RAM
 NVRAMs provide superior performance over other NVM products
 NVRAM's serve applications that require high-speed read/write operations with
non-volatile memories such as parallel processing controllers for LANs and
antilock braking systems.
 NVRAM chips don't require much power and backup can be guaranteed for up to
ten years.

Bad NVRAM
When NVRAM is failing, it generally means that your computer hardware is not
retaining the necessary specialized settings that it ought to though the default BIOS
settings remain. Since the BIOS rely on the settings stored in NVRAM in order to handle
the particular hardware you have, performance may lack in stability. The contents of the
NVRAM chip can become corrupted for a variety of reasons:

 A failure of the embedded battery. If the battery embedded in the NVRAM chip
fails, then this means that your system clock will stop running and important
system configuration information may not be maintained.
 A failure of the CMOS (BIOS) chip on your motherboard. If the CMOS chip is
going bad or is not making proper contact with the motherboard's contacts, then
the NVRAM will fail.

When you get an error message about your NVRAM:

 You may need to purchase a new CMOS battery at your computer store to replace
your current one. It is advisable to have a technician observe the battery first and
determine whether you really need a new one.
 If the BIOS chip was the problem, then you will need to contact your hardware
manufacturer who may give you a replacement chip depending on your warranty.
If not, then you will need to replace your motherboard.
 You could also try to reprogram the NVRAM chip with a hosted and Ethernet
address. You should only attempt to do this if you know exactly what you are
doing; otherwise you should seek a technician's guidance.

What is SDRAM?
SDRAM is Synchronous Dynamic RAM. SDRAM is a variant of DRAM in which the
memory speed is synchronized with the clock pulse from the CPU.

This synchronization enables the SDRAM to pipeline read and write requests. Pipelining
enables the SDRAM to accept commands at the same time as it is processing other
commands.

3
There are three variants of SDRAM:

• Single Data Rate SDRAM - SDR SDRAM

What is SDR SDRAM?

SDR SDRAM is Single Data Rate SDRAM. SDR SDRAM is the original SDRAM
standard, which has since been replaced by DDR SDRAM. DDR SDRAM doubles the
bandwidth of SDR DRAM by transferring data twice per cycle on both edges of the clock
signal, implementing burst mode data transfer.

Standard SDR SDRAM DIMMs

SDR SDRAM is normally packaged in DIMM modules.

SDRAM DIMM modules are sold according to clock speed (MHz), bus speed (mega
transfers per second), and transfer rate (megabytes per second.

DIMM Module Chip Type Clock Speed Bus Speed Transfer Rate
PC66 10ns 66 66 533
PC100 8ns 100 100 800
PC133 7.5/7ns 133 133 1,066

What is DDR SDRAM?

DDR SDRAM is Double Data Rate SDRAM.

DDR SDRAM is an improvement over regular SDRAM, also known as SDR SDRAM
(Single Data Rate SDRAM).

DDR SDRAM doubles the bandwidth of SDR DRAM by transferring data twice per
cycle on both edges of the clock signal, implementing burst mode data transfer.

DDR SDRAM is being supplanted by DDR2 SDRAM.

Standard DDR SDRAM DIMM's

DDR SDRAM is normally packaged in DIMM modules.

DIMM Module Chip Type Clock Speed Data Rate Transfer Rate
PC1600 DDR200 100 200 1,600
PC2100 DDR266 133 266 2,133
PC2400 DDR300 150 300 2,400

4
PC2700 DDR333 166 333 2,667
PC3000 DDR366 183 366 2,933
PC3200 DDR400 200 400 3,200
PC3500 DDR433 216 433 3,466
PC3700 DDR466 233 466 3,733
PC4000 DDR500 250 500 4,000
PC4300 DDR533 266 533 4,266

Standards for DDR SDRAM

DDR SDRAM standards are still being developed and improved.

DDR SDRAM Standard Frequency (MHz) Voltage

DDR 400-533 2.5
DDR2 667-800 1.8
DDR3 1066 to ... 1.5

Higher frequencies enable higher rates of data transfer.

Lower voltages result in less heat radiation and longer batter life for portable computing
devices. They also allow greater component density, which allows higher capacity in the
same package size.

What is DDR2 SDRAM?

DDR2 SDRAM, an abbreviation for double data rate two synchronous dynamic random
access memory, is a type of a random access memory (RAM) commonly used in personal
computers and various digital electronics today. This offspring from the SDRAM
(synchronous dynamic random access memory), a part of DRAM (dynamic random
access memory), is like an evolution from the DDR SDRAM; it can operate the external
data bus twice as fast as its predecessor. This was possible by abandoning the original
clock rate of the DDR, and operating the memory cells at half the rate. If the DDR2 was
clocked at the same rate, the performance would be worse.

History
The DDR2 was first introduced in 2003 at 200 MHz (PC2-3200) and 266 MHz (PC2-
4200), but it failed to outperform the original DDR due to a latency problem.
Nonetheless, a combination of the original DDR technology being at a speed of about
266 MHz (but 533 MHz effective), the JEDEC (Joint Electron Device Engineering
Council) Solid State Technology Association not standardizing the faster DDR chips, and
a release of newer DDR2 modules with lower latencies, the DDR2 eventually began the

5
better solution. By the end of 2004, the DDR2 really began to compete against the
standard DDR.

Behind the DDR2 Technology

As DDR2 is a type of an SDRAM, it stores its memory in memory cells. These cells are
then activated at according times to correctly operate with an external data bus. Although
the DDR2 shares many similarities with the original DDR, as in the case with data
transfer happening with the rise and fall at the edge of the clock, it is different in that the
DDR2's bus is clocked at twice the speed. This basically allows the DDR2 to be twice as
more efficient than the original DDR.

The key behind the high bus frequency with DDR2s (although the latency is augmented
as a trade off) is the coalition of electrical interface improvements, prefetch buffers, and
various off-chip drivers. This is because DDR's prefetch buffer is twice of the DDR,
which is 2bits deep. Also, its read latencies are between four and six cycles; again this is
twice that of the DDR. Thus, twice as much bus speed is necessary to achieve an equal
latency.

Also, the DDR2 is packaged relatively more expensively and more complicatedly than
the DDR SDRAM or the SDR SDRAM, in exchange for a more blazing speed.

Backward Compatibility
Backward compatibility with DDR2 DIMMs are quite limited. In fact, they are not
designed to be compatible with DDRM DIMMs at all. This is because firstly, the notch
positions in the DDR2 and the DDR are different along with the pin density slightly
higher on the DDR DIMMs. Although not much of a "backward compatibility," but faster
DDR2 DIMMs can be used with slower DDR2 DIMMs. However, in such a situation, the
faster one would run at the speed of the slower DDR2.

Standard DDR2 SDRAM DIMM's

DDR2 SDRAM is normally packaged in DIMM modules.

DIMM Module Chip Type Clock Speed Data Rate Transfer Rate
PC2-3200 DDR2-400 200 400 3,200
PC2-4200 DDR2-533 266 533 4,266
PC2-5300 DDR2-667 333 667 5,333
PC2-6400 DDR2-800 400 800 6,400
PC2-7400 DDR2-933 466 933 7,460
PC2-8500 DDR2-1066 533 1066 8,530
PC2-9600 DDR2-1200 600 1200 9,600

6
PC2-10600 DDR2-1333 667 1333 10,660
PC2-11700 DDR2-1466 733 1466 11,730
PC2-12800 DDR2-1600 800 1600 12,800

What is DDR3 SDRAM?

DDR3 SDRAM (Double Data Rate Three Synchronous Dynamic Random Access
Memory) is the third generation of DDR SDRAM.

DDR3 SDRAM improves on DDR2 SDRAM in several significant ways:

• Higher bandwidth due to increased clock rate

• Reduced power consumption due to 90mm fabrication technology
• Pre-fetch buffer is doubled to 8 bits to further increase performance

The voltage of DDR3 SDRAM DIMM's was lowered from 1.8V to 1.5V. This reduces
power consumption and heat generation, as well as enabling more dense memory
configurations for higher capacities.

Standard DDR3 SDRAM DIMM's

DDR3 SDRAM is normally packaged in DIMM modules.

DIMM Module Chip Type Clock Speed Data Rate Transfer Rate
PC3-6400 DDR3-800 400 800 6,400
PC3-8500 DDR3-1066 533 1066 8,530
PC3-10667 DDR3-1333 667 1333 10,660
PC3-12800 DDR3-1600 800 1600 12,800
PC3-14900 DDR3-1866 933 1866 14,930

What is RDRAM?
RDRAM is Rambus DRAM.

RDRAM is a proprietary variant of DRAM which was developed by Rambus, Inc.

RDRAM incorporates technical advantages such as:

• Packet-based command protocol

• Command pipelining
• Data pipelining
• Low-voltage signaling
• Precise clocking

7
Manufacturers who wish to utilize RDRAM technology must pay royalties to Rambus
Inc.

What is MRAM Magnetic Memory?

MRAM (Magneto resistive Random Access Memory) is being considered as a viable
replacement for current DRAM technology. Its key difference to the latter is its use of
magnetic fields to save memory bits which makes the memory "permanent." As such,
MRAM technology allows for faster access and data security.

Current RAM Technology

Current RAM technology uses an electrical charge to store memory bits.
Microminiaturized capacitors and transistors in an integrated circuit are 'stacked up' to
create high-density, large-memory but physically small chips. A typical memory chip
stores up to 1 Gb and allows for more programs and data to be cached for immediate use
by the CPU.

The major problem with current RAM technology is that capacitors leak their electrical
charge. This results in the loss of information. As such, the capacitor's electrical charge
must be constantly refreshed to keep the information active and readily available. This
implies the need for a constant power source - not an issue with computers plugged into
electrical outlets (unless the power is inadvertently cut off) but a problem with laptops or
portable computers that use batteries; constant refreshing of memory drains the batteries.

This also leads to another issue - turning off electrical power (deliberate or accidental
shut down of the computer) means that all information cached in RAM disappears.
Restarting requires some time for all the programs, systems and data to be reloaded and
this results in 'idle time' for the user.

The Emergence of MRAM

MRAM uses magnetic elements to store information; usually in the form of two magnetic
plates - separated by a thin insulating layer - which form a single cell. One plate has a
specific magnetic polarity, while the other varies with an external field. An MRAM unit
is composed of a grid of such cells.

The primary advantage of an MRAM chip is that there is no need to constantly refresh
the information through the periodic application of an electrical charge; even shutting
down the computer does not erase the information. As such, start-up routines go faster;
just turn on the computer and the last session is immediately available. At the same time,
there is reduced risk of data loss from unexpected power outages.

There are two downsides to MRAM’s, however, which have prevented it from joining the
mainstream. Changing data in magnetic fields requires a larger power input than
conventional RAM chips; in effect, the need for a constant power supply is replaced by a

8
larger power requirement. Moreover, the nature of magnetic fields implies physically
larger memory 'cells' - an MRAM chip stores less memory than a RAM chip of the same
size.

Ongoing research aims to resolve both these problems using a variety of approaches.
Although some companies report readiness to mass manufacturer MRAM chips, it
remains to be seen whether these will successfully solve the power input and size issues.
If these are resolved and prices of MRAM chips match that of current RAM chips, we
would be seeing a revolution in how computers start up and operate.

What is Video RAM?

Video RAM is specialized RAM which is used on video cards. Video RAM is dual-
ported, which means it can be accessed by two different devices simultaneously. This
enables data to be read from video RAM (i.e. sent to the computer monitor) at the same
time data is written to video RAM.

What is Flash Memory?

Flash memory is memory which retains its contents even after power is removed. Flash
memory is a form of EAPROM (Electrically Alterable Programmable Read-Only
Memory).

Each bit of data in a flash memory device is stored in a transistor called a floating gate.
The floating gate can only be accessed though another transistor, the control gate. The
process the control gate uses to access the floating gate is a field emission phenomenon
known as Fowler-Nordheim tunneling. Tunneling allows voltage to flow from the control
gate to the floating gate through the dielectric layer of oxide which separates them.

Popular flash memory devices include:

• Sony's Memory Stick

• Compact Flash
• SDCard
• MultiMediaCard (MMC)

What is a SD Card?

9
A SD Card (Secure Digital Card) is an IC (Integrated Circuit) which is stored in a
compact and rugged plastic enclosure. SD Cards are designed to store data and to enable
the transfer of data between devices equipped with SD Card slots. Current SD Card
capacities range up to 1GB. A SD Card is 32mm long, 24mm wide and 2.1mm thick. An
even more compact format, the miniSD Card, is 20mm long, 21.5mm wide, and 1.4mm
thick. The theoretical transfer speed of a SD 1.0 Card is 12.5MB/s. SD 1.1 is expected to
raise this to 50MB/s. The SD Card standard was introduced by Toshiba, Matsushita
Electric, and Scan Disk in 1999.

SDIO
SDIO extends the SD Card standard to include 802.11b Wi-Fi cards, Bluetooth cards,
modems, GPS receivers, TV tuners, cameras, digital recorders, scanners, fingerprint
scanners and more.

What is a Multimedia Card (MMC)?

A Multimedia Card (MMC) is an IC (Integrated Circuit) which is stored in a compact and

rugged plastic enclosure. Multimedia Card (MMC) is designed to store data and to enable
the transfer of data between devices equipped with Multimedia Card (MMC) slots.

The Multimedia Card (MMC) standard was introduced in November of 1997 by Scandisk
and Siemens AG/Infeneon Technologies AG. Current Multimedia Card (MMC)
capacities range up to 2GB. A Multimedia Card (MMC) is 32mm long, 24mm wide and

10
1.4mm thick. Multimedia Cards can be used in SD Card readers and writers. The
theoretical transfer speed of a Multimedia Card is 2.5MB/s. For more information on
Multimedia Card, visit the Multimedia Card Association.

What is a SIMM?
A SIMM is a Single In-line Memory Module. A SIMM is a small circuit board designed
to hold a set of RAM chips. Two types of SIMM's have been in general use. 30-pin
SIMM's and 72-pin SIMM's. 30-bit SIMM's have 8-bit data buses; 72-pin SIMM's have
32-bit data buses.

The pin outs for all SIMMs is not completely standard. Some manufacturers, such as HP,
Compaq, and IBM used non-standard pin outs in their proprietary SIMMs. If you are
upgrading SIMMs, be sure to use compatible SIMMs.

SIMM's have largely been replaced with DIMM's.

What is a DIMM?
Dual Inline Memory Module or DIMM is a series of Random Access Memory (RAM)
chips mounted on a small printed circuit board. The entire circuit collectively forms a
memory module. DIMMs are commonly used in personal computers, servers and high-
end workstations. The DIMM makes physical contact with the data bus of the computer
through teeth like connectors that fit into a socket on the mother board.

The earlier memory modules were known as SIMMs or Single Inline Memory Modules
and had a 32-bit data path. DIMMs on the other hand use a 64-bit data path, since
processors used in personal computers including the Intel Pentium have a 64-bit data

11
width. Since SIMMs can handle only 32-bits at a time, they were always used in matched
pairs to fully utilize the processing power of the CPU. DIMMs were developed to rectify
this inefficient method of installing memory modules.

SIMMs have an identical pair of electrical contacts one on each side of the module and
the processor can access the SIMM through either side. Hence the connectors used in a
SIMM are redundant contacts, whereas the DIMM has unique contacts on either side of
the module and it makes much better use of the connectors.

DIMMs that implement error detection and correction are known as Error Correction
Codes enabled DIMMs or ECC DIMMs. Apart from the data bits, these DIMMs use extra
bits for ECC. While there are many types of ECC schemes, the SECDED or Single Error
Correct - Double Error Detect scheme is the most common and uses an extra 9th bit for
every byte of data.

DIMMs come in various standard sizes known as form factors. Earlier DIMMs came in
sizes of 1.5 inches and 1.7 inches. Later on when rack-mounted servers became common,
these DIMMs had to be squeezed into a narrow space and hence the DIMM sockets were
tilted to an angle to accommodate the memory modules. To address this issue, the next
standard of DIMMs had a low profile height of 1.2 inches eliminating the need for angled
sockets. When servers became even smaller, the sockets were again angled to
accommodate the LP form factor. This further led to the development of the VLP or Very
Low Profile form factor with a height of a mere 0.72 inches. The Mini-DIMM, the SO-
DIMM and the VLP Mini-DIMM are the other popular form factors.

What is a SODIMM?
SODIMM is Small Outline DIMM. SODIMM is a smaller DIMM which is designed for
use in notebook computers and other space-restricted environments.

SODIMM Types
Thee types of SODIMM's are manufactured.

SODIMM Types

12
Contact Count Data Bus Width
72 32
144 64
200 72

SODIMM is sometimes written as SO-DIMM.

What is Registered Memory?

Registered memory modules have built-in registers on their address and control lines. A
register is a very small temporary holding area (usually 64 bits) for data. These registers
act as buffers between the CPU and the memory. The use of registered memory increases
system reliability, but also slows the system down a very slight bit as data must be moved
through the registers.

Some systems do not support registered memory, others require registered memory, and
many more give you the option to use registered or unregistered memory. The use of
registered memory is recommended for server-class systems. It is not useful on gaming
systems. Registered memory is also known as buffered memory. Unregistered memory is
also known as unbuffered memory.

13
What is Unbuffered Memory?
Unbuffered memory is memory where the memory controller module drives the memory
directly, instead of using a store-and-forward system like registered memory. Some
systems do not support unbuffered memory, others require unbuffered memory, and
many more give you the option to use unbuffered or registered memory.

The use of unbuffered memory is reasonable for gaming systems. It is not recommended
for server-class systems. Unbuffered memory is also known as unregistered memory.
Buffered memory is also known as registered memory.

How do I Test Memory?

When something begins to go wrong with a computer and things start acting differently
than they should, the first thing to test should be the computer's memory. Problems, even
small ones, in the computer's memory can lead to very erratic behavior from the
computer, even causing it to crash.

Memory Testing Software

Memtest86 is our recommended memory testing utility. Memtest86 is a reliable program
that can be a great help to you if you think your memory may be causing problems in
your computer.

Strategies for Testing Memory

If you do find an error when running Memtest86 or any other memory testing program,
the next task is to find out which memory stick has the problem.

If you only have one RAM stick plugged in to the motherboard, replacing that single
stick is a very simple task.

If you are running two, three, or even more RAM sticks than it is advised that you pull
out the sticks one by one and then run the test again to find out which stick has the
problem. There is no sense in replacing all of the memory when only one might have an
error.

What is ROM?
Read-Only Memory or ROM is an integrated-circuit memory chip that contains
configuration data. ROM is commonly called firmware because its programming is fully
embedded into the ROM chip. As such, ROM is a hardware and software in one.

14
Because data is fully incorporated at the ROM chip's manufacture, data stored can neither
be erased nor replaced. This means permanent and secure data storage. However, if a
mistake is made in manufacture, a ROM chip becomes unusable. The most expensive
stage of ROM manufacture, therefore, is creating the template. If a template is readily
available, duplicating the ROM chip is very easy and affordable. A ROM chip is also non
volatile so data stored in it is not lost when power is turned off.

RAM versus ROM

Both RAM and ROM provide the user random access to stored data. However, RAM
provides only short-term memory since data stored in RAM is lost when power is turned
off. ROM, on the other hand, provides long-term storage since data is permanently etched
into the ROM chip.

One other difference between the two is that RAM's data can frequently and speedily be
altered and changed at will. ROM cannot be reconfigured at all.

ROM Chip Manufacture

The ROM chip has a matrix of columns and rows. The point of intersection between the
column and the row is called a cell. Each cell will have a value of either 1 or 0. The
manufacturer of the ROM chip has to determine each cell's value before the ROM
manufacture ensues.

After the program has been designed, the ROM chip can now be made. At each cell
where the value should be 1, a diode is placed to connect column to row. At each cell
where the value should be 0, no diode is placed.

The ROM Chip's Diode

The diode is the cell's electronic check valve. Its main function in the ROM chip is to
control and direct the flow of electric current in cells. Specifically, it transfers the electric
current from column to row.

The diode has a minimum voltage requirement, what is commonly called the forward
breakover, which needs to be surpassed before the diode will transfer the charge from the
column to the row of the cell. The usual ROM diode forward breakover voltage is around
0.6 volts.

The ROM activates cells by sending through voltage that is more than the cells' diode
forward breakover. The charge that is directed through the column gets passed on by the
diode to the appropriate grounded row. Since charge is successfully transmitted from
column to row, the cell where a diode is present has a value of 1.

In the ROM circuit, passing a charge through the column will not affect the cells whose
values are supposed to be 0. In any cell, this charge cannot be passed from column to row

15
if there's no diode to facilitate the transfer. Therefore, once a cell's value has been
determined to be 0, it will permanently be 0.

What is a PROM?
PROM stands for Programmable Read-Only Memory. It was invented by Wen Tsing
Chow in 1956. Unlike a ROM chip, a PROM chip comes from the manufacturers devoid
of any programming. Programming is done by the end user or the makers of electronic
devices which require a permanent data storage device.

Since it is cheaper than a ROM chip, it is also useful for making experimental
programming test runs. Successful programming can later on be applied to a ROM chip
at manufacture. A PROM chip can only be programmed with the desired data once. Data
or programming information stored in PROM chips is permanent. A PROM chip is also
non-volatile so data is not lost when power is turned off.

Physical Structure
The PROM chip is physically similar to a ROM chip as it also has an array of columns
and rows. Unlike the ROM chip, however, all columns and rows are connected by a fuse
at their points of intersection. Every point of connection determines one cell or one bit.
Since connection between the columns and rows is intact and therefore charges can pass
unhindered from column to rows, a newly manufactured PROM chip's cells all have a
value of 1.

PROM Programmer
A PROM Programmer, also known as a PROM blower, is used in programming a PROM
chip. The PROM programmer is a device that can hold and program more than one
PROM chip at the same time. Basically, what it does is melt the fuses where the truth
table data indicates the value of a cell or bit should be 0.

Programming a PROM
The process of programming a PROM chip is called PROM "burning" or "blowing"
because it involves burning or blowing out the fuses of some cells whose value needs to
be altered from 1 to 0.

Before programming a PROM chip, the PROM programmer is first configured so it

contains the desired PROM burning instructions. After this is done, the PROM chip is
inserted into one of the PROM programmer slots.

The PROM programmer sends a predetermined amount of voltage pulse ranging from 12
to 21 volts. This high voltage is directed only to the fuses belonging to cells whose value
has to be changed from 1 to 0.

16
When the charge reaches the fuse, the high voltage causes this fuse to blow out or burn
out. The burned out fuse no longer connects a column to a row in a cell and the row is
rendered incapable of receiving any charge from the column. The cell with a burned out
fuse has a value of 0.

PROM programming is irreversible because burning a PROM involves permanent

damage to some of the fuses. The cells whose fuses have been burned out will have a
permanent value of 0. After PROM programming has been accomplished, there's no
returning the altered cells' value back to 1. Thus, the PROM chip's programming cannot
be modified and no new data can be written to an already programmed PROM chip.

What is an EPROM?
EPROM or Erasable Programmable Read Only Memory was invented by Engr. Dov
Frohman. It is a ROM-type chip that can hold data from 10-20 years. It is different from
PROM because it can be programmed more than once. An EPROM programming is
erased only through exposure to ultra violet light. The EPROM is configured or
reconfigured using an EPROM programmer.

Configuring an EPROM
An EPROM is an array of columns and rows. Each intersection is called a cell and each
cell has a floating gate transistor and a control gate transistor. These two transistors are
separated by a fine oxide layer.

The floating gate is connected to the grid row (word line) through the control gate. The
cell's value is 1 if the floating gate's connection to the word line through the control gate
exists. All cells of a blank EPROM have a value of 1.

To configure an EPROM chip, you need to change certain cells from 1 to 0. The cell
value 0 is attained if the connection between the floating gate transistor and the word line
is removed. This is done through a process known as Fowler-Nordheim tunneling.

To accomplish tunneling, a charge of electricity, typically 10-13 volts, is passed through

the column (bit line). This electric charge proceeds to the floating gate. In reaction, the
floating gate releases electrons which are in turn trapped on the oxide layer. The electric
charge, in effect, rearranges the floating gate's electron configuration. The excited
electrons become a barrier between the floating gate and the control gate. The link
between the control gate and the word line is subsequently eliminated and the cell's value
changes from 1 to 0.

Erasing EPROM Programming

To erase EPROM programming, sufficiently strong energy should be passed through the
EPROM to break down the wall of negative electrons blocking the floating gate's link to

17
the word line. Most EPROM chip has a quartz window on top and this is where this
energy is directed.

Erasure of EPROM programming is accomplished by using ultraviolet light that belongs

to the UVC range and has a frequency of 253.7. Remove the EPROM chip from your
computer or electronic device then put it under the EPROM eraser for a number of
minutes. The EPROM Eraser will emit UV light which should be directed towards the
quartz window from where it proceeds to flow through the cells and unblock the floating
gates.

This energy will unblock all cells so that all cell values will return to 1. EPROM
programming erasure is non-selective. This means that a specific part of the EPROM chip
cannot be selected for reconfiguration. When UV light is directed towards the quartz
window of the EPROM, all data will be lost.

It is inadvisable to leave the EPROM chip under the EPROM eraser for too long. This
may lead to over erasure. An over-erased EPROM chip can no longer be used for storing
data. Due to overcharging of UV light, the EPROM's floating gates become unstable,
rendering them incapable of holding any electrons.

What is an EEPROM?
EEPROM stands for Electrically Erasable Programmable Read-Only Memory. An
EEPROM is like an EPROM chip since it can be written in or programmed more than
once. Unlike the EPROM chip, however, an EEPROM chip need not be taken out of the
computer or electronic device of which it is part when a new program or data needs to be
written on it.

Selective programming can be done to an EEPROM chip. The user can alter the value of
certain cells without needing to erase the programming on other cells. Thus, sections of
data can be erased and replaced without needing to alter the rest of the chip's
programming.

Data stored in an EEPROM chip is permanent, at least until the user decides to erase and
replace the information it contains. Furthermore, the data stored in an EEPROM chip is
not lost even when power is turned off.

History of the EEPROM

The EEPROM is a modification of the EPROM and was designed by George Perlegos .
Its development began in 1978 while Perlegos was still employed by Intel. However the
archetypal EEPROM still had to be taken out of the computer or electronic device of
which it was part if any reprogramming was necessary.

When Perlegos left Intel to form Seeq Technology, he designed the first fully functional
EEPROM. To eliminate the necessity of external programming, Perlegos and company
made the insulating layer thinner and integrated an oscillator and capacitor circuit into the

18
memory chip itself. This charge pump can produce the necessary programming voltage.
Since it is fully integrated in every EEPROM chip, there is no need to take out the
EEPROM chip for erasing and programming. To configure an EEPROM chip, an electric
field produced by the charge pump is applied locally to cells marked for modification.

EEPROM Structure
The EEPROM chip is physically similar to the EPROM chip. It is also composed of cells
with two transistors. The floating gate is separated from the control gate by a thin oxide
layer. Unlike the EPROM chip, however, the EEPROM chip's oxide layer is much
thinner. In EEPROM chips, the insulating layer is only around 1 nanometre thick whereas
in EPROM chips, the oxide layer is around 3 nanometres thick. The thinner oxide layer
means lower voltage requirements for initiating changes in cell value.

Tunneling the electrons of the floating gate towards the oxide layer separating the
floating gate and the control gate is still the method of changing a bit's value from 1 to 0.
To erase EEPROM programming, the electron barrier still has to be overcome by the
application of enough programming voltage.

EEPROM Limitations
While the EEPROM can be reprogrammed, the number of times it can be altered is
limited. This is the main reason why EEPROM chips are popular for storing only
configuration data such as the computer's BIOS code which doesn't require frequent
reprogramming. The oxide insulating layer can be damaged by frequent rewrite. Modern-
day EEPROMs can be rewritten up to a million times.

What is Firmware?
We all use electronic devices, whether they are mobile phones, computer routers, MP3
players, and cable boxes, etc. These electronic devices include hardware, the physical
electronic components and software; programs that help these components run
effectively. Firmware is a mix of both.

Firmware
Firmware is usually defined as a type of program that runs within an electronic device.
Firmware is not categorized either as hardware or software, but a mix of both. The reason
firmware is considered a combination of both categories is that firmware employs an
executable program (.exe) and includes an integrated piece of the electronic device.

Firmware is stored on ROM chips (read only memory). Each time your computer or
electronic device boots up, firmware is booted as well. Firmware can store their
instructions permanently and do not require a power source. One of the most common
types of firmware is the BIOS chip on your computer's motherboard. The BIOS chip
holds extremely important data (instructions). For instance, when your computer is

19
started, this activates the BIOS chip where important hardware is initialized and
components are checked by the BIOS chip to ensure that devices are working properly.
Once the BIOS chip has completed its duties, it sends a message to your computer's
operating system which then starts up.

While firmware is permanent, it has the advantage of being rewritten. It should be noted
that older firmware did not have the ability to rewrite its instructions. In fact, in the past,
when new instructions were required for a BIOS chip, the entire motherboard usually
needed replacing since the BIOS chip could not be rewritten. Now, firmware can easily
be rewritten enabling it to be upgraded with new features or to remove bugs and issues.
Since firmware is found in practically all electronic devices, it has extended the life of
these products, easily adding more functionality to them.

How Does Firmware Upgrade

For the most part, in order to upgrade firmware you will need to contact the device's
manufacturer. Whether your device is having issues or you would like to add new
functionality to it, your manufacturer should have the latest firmware update (if
available). There are two simple methods of upgrading your firmware. The first is to
download the new firmware directly into your device; the second is to request the new
firmware from the manufacturer (either on CD ROM or other media). With both
methods, upgrading your firmware is relatively quick and easy and your new
functionality or the disappearance of bugs will be immediate once installed

Computer devices that connect directly to the internet are by far the easiest firmware to
upgrade. Just log onto the device's manufacturer's homepage and search for a new
firmware update. If a new firmware update exists, you can follow the prompts to
download it. Sometimes you can download it directly to your device, other times you will
need to save it to your hard drive and then transfer it to your device with a special cable.

In certain cases, firmware can not be downloaded from the internet, instead new
instruction need to be inserted directly into the device, usually with the help of a memory
card and card reader. One type of device that can include direct installation is the cable
box or satellite receiver.

20