What's New Departments

The Ultimate Memory Guide

How Memory Works

[Back to the beginning...]

How memory works

Up to now, we've alluded to some of the technical attributes of memory and how it functions in a system. Now we're down to the technical aspect -- the bits and the bytes. In this section we discuss the binary numbering system that forms the basis of computing and how memory modules are designed to work within the binary system.

[ Table of Contents ]

Bits and bytes

Your computer speaks a language made up of only two numerals: 0 and 1. This two-numeral form of communication is called machine language; the numerals combine to form binary numbers. Machine language uses binary numbers to form instructions for the chips and microprocessing devices that drive computing devices -- such as computers, printers, hard disk drives, and so on.

You may have heard the terms bit (short for binary digit) and byte. A bit is the smallest unit used by your computer and can be either a 1 or a 0. A byte consists of 8 bits (more about bytes later). Because binary numbers consist of only 1's and 0's, binary number values appear different from the decimal values we use in everyday life. For example, in the decimal system when you see the number 1 followed by two 0's (100), you know it represents a value of one hundred. In binary, however, the same number combination -- 100 -- represents a value of four.



The process of counting in binary isn't all that different from the decimal system. As you count in decimals, when you reach 9, the count resets to 0 and a 1 shifts into the tens column. When you count in binary, the process is very similar. However, because there are only two numerals, the resetting and shifting of digits happens more quickly.


Take a look at the table above. On the top are the decimal numbers 0 through 15; on the bottom are the binary equivalents.

Going back to how machine language uses 1's and 0's, remember that each digit in a binary number represents 1 bit. Machine language thinks of each bit as either `on' or `off.' A bit with a value of 1 is considered `on,' and a bit with a value of 0 is `off.' Therefore, determining the value of a binary number is simply a matter of adding up the columns that are `turned on,' in other words, wherever a 1 appears. (This `on/off' idea comes in handy later.)

In the decimal numbering system, each column (ones, tens, hundreds, and so on) has ten times the value of the previous column. In the binary numbering system, however, each column has a value twice that of the previous column (ones, twos, fours, eights, sixteens, and so on).





This example compares the same number combination (111) in each of the two systems. In the decimal numbering system, the number 111 represents the addition of 100+10+1. In binary, 111 represents a binary value of 7 because it's the result of adding 4+2+1.







Because your computer understands binary values consisting of 1's and 0's, a binary value exists for every possible keyboard character. The most widely accepted standardized system for numbering keyboard characters is called the ASCII system. (ASCII is pronounced `askee' and stands for American Standard Code for Information Interchange.)

EIGHT BINARY DIGITS


It takes up to 256 different binary number combinations (0 through 255) to describe every possible keystroke you might enter from your keyboard. Representing the decimal numbers 0 through 255 requires 8 binary digits. As you can see here, the highest number value (255) is that in which all 8 columns are "turned on", or contain the number 1.





We mentioned earlier that 8 bits -- represented by 8 binary digits -- together are referred to as a byte. Almost all specifications of your computer's capabilities are represented in bytes. For example, memory capacity, data-transfer rates, and data-storage capacity are all measured in bytes or multiples thereof (such as kilobytes or megabytes). The following table summarizes how to refer to quantities of bits and bytes.

When calculating multiple bits and bytes, something potentially confusing is how the letter K (kilo) is used to express quantities of bytes or bits.

Outside the computer environment, a kilo represents 1,000 units. However, in computer science, the kilo prefix represents exactly 1,024 units, or 2 to the tenth power.

[ Table of Contents ]

CPU and memory requirements

A computer's CPU (central processing unit) processes data in 8-bit chunks. Those chunks, as we learned in the previous section, are commonly referred to as bytes. Because a byte is the fundamental unit of processing, the CPU's processing power is often described by the maximum number of bytes it can process at any given time. For example, the most powerful Pentium and PowerPC microprocessors currently are 64-bit CPUs, which means they can simultaneously process 64 bits, or 8 bytes, at a time.

Each transaction between the CPU and memory is called a bus cycle. The number of data bits a CPU can transfer during a single bus cycle affects a computer's performance and dictates what type of memory the computer requires. Most desktop computers use either 72- or 30-pin SIMMs. A 30-pin SIMM supports 8 data bits; a 72-pin SIMM supports 32 data bits.

[ Table of Contents ]

30-pin SIMMs

Let's look at the example of a CPU that supports 32 data bits. If the computer's system board has 30-pin SIMM sockets, each of which supports 8 data bits, you'll need four 30-pin SIMMs to supply 32 bits. (This is a common configuration for systems that use 30-pin SIMMs.) The memory configuration on such a system is typically divided into two memory banks -- bank zero and bank one. Each memory bank consists of four 30-pin SIMM sockets. The CPU addresses memory one bank at a time.

NOTE:
With most computer models, mixing different-capacity SIMMs within the same bank prevents the computer from accurately detecting the amount of available memory. This causes one of two things to occur:

  1. The computer will not boot up.
  2. The computer will boot up but will not recognize or use some of the memory in the bank. For example, if a bank had three 1-megabyte SIMMs and one 4-megabyte SIMM, it would recognize them all as 1-megabyte SIMMs.

    [ Table of Contents ]

    72-pin SIMMs

    The 72-pin SIMM was developed to satisfy the ever-expanding memory requirements of desktop computers. One 72-pin SIMM supports 32 data bits, which is four times the number of data bits supported by a single 30-pin SIMM. If you have a 32-bit CPU -- such as a 486 from Intel or Motorola's 68040 -- you need only one 72-pin SIMM per bank to provide the CPU with 32 data bits. As we saw in the previous section, that same CPU would require four 30-pin SIMMs per bank to get its 32 data bits.

    [ Table of Contents ]

    Credit card memory

    `Credit card memory' is designed for use in laptop and notebook computers. Because of its compact form factor, credit card memory is ideal for applications where space is limited. (Credit card memory got its name because the form factor is the approximate size of a credit card.)

    
    
    
    On the surface, there are few similarities between credit card memory and the SIMM-style memory described earlier. However, the same familiar components are used in the internal construction of SIMMs.
    
    
    NOTE: Although their appearances are similar, a credit card memory module is not to be confused with a PCMCIA (Personal Computer Memory Card International Association) card. Credit card memory uses a non-PCMCIA slot on the computer and is designed only to provide memory expansion. Kingston, however, does offer memory products that conform to the PCMCIA standard, which is designed for attaching input/output devices to laptop and portable computers.

    [ Table of Contents ]

    DIMM memory

    Dual In-line Memory Modules, or DIMMs, closely resemble SIMM-type memory. Like SIMMs, most DIMMs install vertically into expansion sockets. The principal difference between the two is that on a SIMM, opposing pins on either side of the board are `tied together' to form one electrical contact; on a DIMM, opposing pins remain electrically isolated to form two separate contacts.

    DIMMs are often used in computer configurations that support a 64-bit or wider memory bus. In many cases, these computer configurations are based on powerful 64-bit processors like Intel's Pentium or IBM's PowerPC processors.

    For example, Kingston's KTM40P/8 DIMM module used in IBM's PowerPC 40P RISC 000 computer is a 168-pin DIMM.

    [ Table of Contents ]

    Small Outline DIMMs

    Another type of memory commonly used in both notebook and laptop computers is called Small Outline DIMM, or SO DIMM. A small outline DIMM is like a 72-pin SIMM in a reduced-size package, but with some important technical differences. The small outline DIMM and the SIMM shown opposite both have 72 pins. However, it's the way the pins are arranged that differentiates these two types of memory.

    
    
    
    The three examples illustrate the differences among SIMM, DIMM, and SODIMM products. The full size, 168-pin DIMM supports 64-bits transfers without being twice the size of the 72-pin SIMM which support only 32-bit transfers.

    The SODIMM also supports 32-bit transfers and was designed for us in notebook computers.

    
    
    
    
    
    
    [ Table of Contents ]

    More about proprietary memory

    By definition, proprietary memory is memory designed specifically for a particular manufacturer or computer model. For example, the KCN-IB150/16 proprietary module is used in Canon INNOVA notebook computers.

    
    
    The KCN-IB150/16 is only one in many examples of proprietary memory. Kingston manufactures several other proprietary modules, including many credit card and DIMM components. As mentioned earlier, the term proprietary does not apply to a particular type of memory, it simply means the memory upgrade is applicable to only one type of computer. Keep in mind that a memory module doesn't require a unique appearance to be proprietary. Kingston also sells proprietary versions of SIMMs, DIMMs, and SO DIMMs. In fact, the DIMM memory example highlighted in the `DIMM memory' section is a proprietary memory module.

    [ Table of Contents ]


    | Product Information | Technical Support |

    | About Kingston | Distributor & Reseller Links | Main Menu |


    Copyright © 1996 Kingston Technology Corporation. All Rights Reserved.