'; window.popUpWin.document.write(zhtm); window.popUpWin.document.close(); // Johnny Jackson 4/28/98 } //--> Upgrading & Repairing PCs Eighth Edition -- Ch 20 -- T


Upgrading & Repairing PCs Eighth Edition

Previous chapterNext chapterContents

- 20 -

Portable Systems

Portable computers, like their desktop counterparts, have evolved enormously since the days when the word portable could mean as little as a case with a handle on it. Today, portable systems can rival the performance of their desktop counterparts in nearly every way, to the point at which many systems are now being marketed as "desktop replacements" that companies are providing to traveling employees as primary systems. This chapter examines the types of portable computers available and the technologies designed specifically for use in mobile systems.

Portables started out as suitcase-sized systems that differed from desktops mainly in that all of the components were installed into a single case. Compaq was among the first to market portable computers like these in the 1980s, and although their size, weight, and appearance were almost laughable when compared to today's portables, they were cutting-edge technology for the time. The components themselves were not very different from those in standard computers. Most portable systems are now built using the clamshell design that has become an industry standard (see Figure 20.1), with nearly every component developed specifically for use in mobile systems.

The computers have settled into a number of distinct roles that now determine the size and capabilities of the systems available. Traveling users have specific requirements of portable computers, and the added weight and expense incurred by additional features makes it less likely for a user to carry a system more powerful than is needed.

Form Factors

There are three basic form factors that describe most of the portable computers on the market today: laptops, notebooks, and subnotebooks. The definitions of the three types are fluid, however, with the options available on some systems causing particular models to ride the cusp of two categories. The categories are based primarily on size and weight, but these factors have a natural relationship to the capabilities of the system, because a larger case obviously can fit more into it. The three form factors are described in the following sections.

Laptops

As the original name coined for the clamshell-type portable computer, the laptop is the largest of the three basic form factors (see Figure 20.1). Typically, a laptop system weighs seven pounds or more, and is approximately 9x12x2 inches in size, although the larger screens now arriving on the market are causing all portable system sizes to increase. Originally the smallest possible size for a computer, laptops have today become the high-end machines, offering features and performance that are comparable to a desktop system.

FIG. 20.1  A laptop.

Indeed, many laptops are being positioned in the market either as desktop replacement or as multimedia systems suitable for delivering presentations on the road. Because of their weight, they are typically used by sales people and other travelers who require the features they provide, although many laptops are now being issued to users as their sole computer, even if they only travel from the office to the home. Large displays, 16M or more of RAM, and hard drives up to 2G in size are all but ubiquitous, with many systems now carrying CD-ROM drives, on-board speakers, and connectivity options that enable the use of external display, storage, and sound systems.

As a desktop replacement, many laptops can be equipped with a docking station that functions as the user's "home base," allowing connection to a network and the use of a full-size monitor and keyboard. For a person that travels frequently, this arrangement often works better than separate desktop and portable systems, on which data must continually be kept in sync. Naturally, you pay a premium for all of this functionality. Cutting-edge laptop systems can now cost as much as $6,000 to $7,000, more than three times the price of a comparable desktop.

Notebooks

A notebook system is designed to be somewhat smaller than a laptop in nearly every way: size, weight, features, and price. Weighing 5 to 7 pounds, notebooks typically have smaller and less capable displays and lack the high-end multimedia functions of laptops, but they need not be stripped-down machines. Many notebooks have hard drive and memory configurations comparable to laptops, and some are equipped with CD-ROM drives or sound capabilities.

Designed to function as an adjunct to a desktop system, rather than a replacement, a notebook probably won't impress your clients but can be a completely serviceable road machine. Notebooks typically have a wide array of options, as they are targeted at a wider audience, from the power user that can't quite afford the top-of-the-line laptop, to the bargain hunter that requires only basic services. Prices can range from less than $2,000 to more than $4,000.

Subnotebooks

Subnotebooks are substantially smaller than both notebooks and laptops, and are intended for users who must enter and work with data on the road, as well as connect to the office network. Weighing four to five pounds, and often less than an inch thick, the subnotebook is intended for the traveler that feels overburdened by the larger machines and doesn't need their high-end capabilities.

Usually, the first component omitted in a subnotebook design is the internal floppy drive, although some include external units. You also will not find CD-ROM drives and other bulky hardware components. However, many subnotebooks do include large high-quality displays, plenty of hard drive space, and a full-size keyboard (by portable standards).

As is common in the electronics world, devices become more inexpensive as they get smaller, but only up to a certain point, at which small size becomes a commodity and prices go up. Some subnotebooks are intended (and priced) for the high-end market, such as the executive that uses the system for little else but e-mail and scheduling, but wants a lightweight, elegant, and impressive-looking system. Subnotebooks can cost as much as $4,000. Others are much cheaper.

Portable System Designs

Obviously, portable systems are designed to be smaller and lighter than desktops, and much of the development work that has been done on desktop components has certainly contributed to this end. The 2 1/2-inch hard drives typically used in portables today are a direct extension of the size reductions that have occurred in all hard drives over the past few years. However, the other two issues that have created a need for the development of new technologies specifically for portables are power and heat.

Obviously, operating a computer from a battery imposes system limitations that designers of desktop systems have never had to consider. What's more, the demand for additional features like CD-ROM drives and ever faster processors has increased the power drain on the typical system enormously. The problem of conserving power and increasing the system's battery life is typically addressed in three ways:

Perhaps a more serious problem than battery life is heat. The moving parts in a computer, such as disk drives, generate heat through friction, which must somehow be dissipated. In desktop systems, this is usually accomplished by using fans to continuously ventilate the empty spaces inside the system.

The worst culprit by far, however, as far as heat is concerned, is the system processor. When they were first released, the amount of heat generated by Intel's 80486 and Pentium processors was a problem even in desktop systems. Heat sinks and tiny fans mounted on top of the chip became standard components in most systems.

Because many portable systems are now being designed as replacements for desktops, users are not willing to compromise on processing power, so the chips being manufactured for use in portables have all of the speed and capabilities of the desktop models. However, for reasons of power consumption, noise, and space, there are often no fans in portable computers and very little empty space within the case for ventilation.

To address this problem, Intel has created a special method for packaging its mobile processors that is designed to keep heat output to a minimum. Other components are also designed to withstand the heat within a portable computer, which is usually greater than that of a desktop, in any case.

Upgrading and Repairing Portables

The portable systems manufactured today are generally as upgradeable and repairable as traditional desktop systems. In fact, the process of replacing a device is often simpler than on a desktop, because portable systems use modular components with snap-in connectors that eliminate the need for ribbon cables, mounting rails, and separate electrical connections. Thus, common upgrades like adding memory and swapping out a hard drive can usually be accomplished in seconds.

The problem with replacing components in portables is that the hardware tends to be much less generic in portable systems than it is in desktops. Except for PC Cards, which are interchangeable by definition, and in some cases hard drives, purchasing a component that is not specifically intended for use in your exact system model can be risky.

In some cases, these compatibility problems are a matter of simple logistics. Portable system manufacturers jam a great deal of machinery into a very small case, and sometimes a new device just will not fit in the space left by the old one. This is particularly true of devices that must be accessible from the outside of the case, like CD-ROM and floppy drives. Keyboards and monitors, the most easily replaceable of desktop components, are so completely integrated into the case of a portable system that they may not be practically removed at all.

In other cases, your upgrade path may be deliberately limited by the options available in the system BIOS. For example, a manufacturer may limit the hard drive types supported by the system in order to force you to buy a replacement drive from them, and not a third party. This is something that you should check when shopping for a system by asking whether the BIOS is upgradeable and finding out how much the vendor charges for replacement components.

Most of the time, components for portable systems are sold by referencing the system model numbers, even when third parties are involved. If you look through a catalog for desktop memory, for example, you see parts listed generically by attributes like chip speed, form factor, and parity or non-parity. The memory listings for portable systems, on the other hand, will very likely consist of a series of systems manufacturers' names and model numbers, plus the amount of memory in the module.

There are always exceptions to the rule, of course. It is even possible to purchase a basic laptop case and populate it with individual components from various manufacturers. However, purchasing compatible components that fit together properly is certainly more of a challenge than it is for a desktop system.

NOTE: Generally speaking, purchasing a pre-assembled system from a reputable manufacturer is strongly recommended, as is purchasing only replacement components that are advertised as being specifically designed for your system.

Portable System Hardware

From a technical standpoint, some of the components used in portable systems are very similar to those in desktop computers, while others are completely different. The following sections examine the various subsystems found in portable computers and how they differ from their counterparts, discussed in the rest of this book.

Displays

Perhaps the most obvious difference between a portable system and a desktop is the display screen. Gone is the box with an emitter bombarding a concave glass tube with electrons, In its place is a flat screen, the whole assembly for which is more than half an inch thick. This is called an LCD, or liquid crystal diode display. Virtually all of the screens in portable systems today are color, although monochromes were at one point the industry standard, just as in standard monitors.

The display is usually the single most expensive component in a portable system, often costing the manufacturer $1,000 or more. In fact, it is sometimes more economical to replace the entire computer rather than have a broken display replaced. In the first laptops with color screens, the display was a poor replacement for a standard VGA monitor. Today's portable screens are much improved, and while not quite up to the standards of a good monitor, provide excellent performance, suitable even for graphic-intensive applications like bitmap editing and videoconferencing.

The LCD display in a portable system is designed to operate at a specific resolution. This is because the size of the pixels on an LCD panel cannot be changed. On a desktop system, the signal output from the video adapter can change the resolution on the monitor, thus changing the number of pixels created on the screen. Obviously, as you switch from a resolution of 640x480 to 800x600, the pixels must be smaller in order to fit into the same space.

An LCD panel, on the other hand, should be thought of as a grid ruled off to a specified resolution, with transistors controlling the color that is displayed by each individual pixel. The arrangement of the transistors defines the two major types of LCD displays used in portable systems today: dual scan and active matrix.

Dual Scan Displays

The dual scan display, sometimes called a passive matrix display, has an array of transistors running down the x- and y-axes of two sides of the screen. The number of transistors determines the screen's resolution. For example, a dual scan display with 640 transistors along the x-axis and 480 along the y creates a grid like that shown in Figure 20.2. Each pixel on the screen is controlled by the two transistors representing its coordinates on the x- and y-axes.

FIG. 20.2  Dual scan LCD displays use a combination of two transistors on intersecting axes to control the color of each pixel.

If a transistor in a dual scan display should fail, a whole line of pixels is disabled, causing a black line across the screen. There is no solution for this problem other than to replace the display or just live with it. The term dual scan comes from the fact that the processor redraws half of the screen at a time, which speeds up the refresh process somewhat.

Dual scan displays are decidedly inferior to active matrix screens. Dual scan displays tend to be dimmer, because the pixels work by modifying the properties of reflected light (either room light or, more likely, a white light source behind the screen), rather than generating their own. Dual scan panels are also prone to ghost images, and are difficult to view from an angle, making it hard for two or more people to share the same screen.

Of course, they are also far less expensive than active matrix screens. These drawbacks are most noticeable during video-intensive applications, such as presentations, full-color graphics, video, or fast-moving games. For computing tasks that consist largely of reading words on the screen, like word processing and e-mail, the display is quite serviceable, even for long periods of time.

The standard size for a dual scan display is 10 1/2 inches (measured diagonally), running at 640x480, but there are now some systems with 12.1-inch displays that run at 800x600 resolution. If you are familiar with the dual scan display of an older portable, you will find that today's models are greatly improved.

Active Matrix Displays

An active matrix display, also known as a TFT (thin film transistor) display, differs from a dual scan in that it contains a transistor for every pixel on the screen, rather than just at the edges. The transistors are arranged on a grid of conductive material, with each connected to a horizontal and a vertical member (see Figure 20.3). Selected voltages are applied by electrodes at the perimeter of the grid in order to address each pixel individually.

FIG. 20.3  Active matrix LCD displays contain a transistor for each pixel on the screen.

The pixels generate their own light for a brighter display. Because every pixel is individually powered, each one generates its own light of the appropriate color, creating a display that is much brighter and more vivid than a dual scan panel. The viewing angle is also greater, allowing multiple viewers to gather around the screen; and refreshes are faster and crisper, without the fuzziness of the dual scans, even in the case of games or full-motion video.

On the downside, it should be no surprise that, with 480,000 transistors rather than 1,400 (on a 800x600 screen), an active matrix display requires a lot more power than a dual scan. It also drains batteries faster, and costs a great deal more as well.

With all of these transistors, it is not uncommon for failures to occur, resulting in displays with one or more "dead pixels," due to malfunctioning transistors. Unlike a dual scan display, in which the failure of a single transistor causes an immediate and obvious flaw, a single black pixel is far less noticeable. However, many buyers feel (and rightly so) that a computer costing thousands of dollars should be perfect and have attempted to return systems to the manufacturer solely for this reason.

This has occurred often enough that many portable computer manufacturers refuse to accept returns of systems for less than a set number of bad pixels. This is another part of the vendor's purchasing policy that you should check before ordering a system with an active matrix display.

The 12.1-inch active matrix screen has become a standard on high-end laptops, running at a resolution of 800x600 or even 1,024x768. Many portable systems now also include PCI bus video adapters with 2M of RAM, providing extra speed, even at 16- or 24-bit color depths. These displays come very close to rivaling that of a quality monitor and video adapter in a desktop system.

Not to be accused of dragging their feet, however, manufacturers are now bringing new systems to market with 13- and even 14-inch TFT screens.

NOTE: Another flat screen technology, called the gas plasma display, has been used in large display screens and a few portables. Plasma displays provide a CRT-quality picture on a thin flat screen using two glass panes filled with a neon/xenon gas mixture. Unfortunately, the displays require far more power than LCDs and have never become a practical alternative for the portable computer market.

Screen Resolution

The screen resolution of a portable system's display can be an important factor in your purchasing decision. If you are accustomed to working on desktop systems running at 800x600 or 1,024x768 pixels or more, you will find a 640x480 laptop screen to be very restrictive. Remember that an LCD screen's resolution is determined as much by the screen hardware as by the drivers and the amount of video memory installed in the system.

Some portables can use a virtual screen arrangement to provide an 800x600 (or larger) display on a 640x480 pixel screen. The larger display is maintained in video memory while the actual screen displays only the portion that fits into a 640x480 window (see Figure 20.4). When you move the cursor to the edge of the screen, the image pans, moving the 640x480 window around within the 800x600 display. The effect is difficult to get used to, rather like a "scan and pan" video tape of a wide-screen movie. The most serious problem with this arrangement is that some manufacturers have advertised it as an 800x600 display, without being more clear about the actual nature of the display.

Color depth, on the other hand, is affected by video memory, just as in a desktop system. To operate any LCD screen in 16- or 24-bit color mode, you must have sufficient video memory available. Portables typically have the video adapter hardware permanently installed on the motherboard, leaving little hope for a video memory upgrade. There are, however, a few PC Card video adapters that you can use to connect to an external monitor and increase the system's video capabilities.

FIG. 20.4  A virtual screen lets you use a small display to view a portion of a larger screen.

Processors

As with desktop systems, the majority of portables use Intel multiprocessors, and the creation of chips designed specifically for portable systems is a major part of the company's development effort. The heat generated by Pentium processors has been a concern since the first chips were released. In desktop systems, the heat problem is addressed, to a great extent, by computer case manufacturers. The use of multiple cooling fans and better internal layout designs can keep air flowing through the system to cool the processor, which is usually also equipped with its own fan and heat sink.

For developers of portable systems, however, not as much can be accomplished with the case arrangement. So it was up to Intel to address the problem in the packaging of the chip itself. At the same time, users became increasingly unwilling to compromise on the clock speed of the processors in portable systems. Running a Pentium at 133 or 166MHz requires more power and generates even more heat than the 75MHz Pentiums that were originally designed for mobile use.

Tape Carrier Packaging

Intel's solution to the size and heat problems is the tape carrier package (TCP), a method of packaging Pentium processors for use in portable systems that reduces the size, the power consumed, and the heat generated by the chip. A Pentium mounted onto a motherboard using TCP is much smaller and lighter than the standard PGA (pin grid array) processors used in desktop systems. The 49mm square of the PGA is reduced to 29mm in the TCP processor, the thickness to approximately 1mm, and the weight from 55 grams to under 1 gram.

Instead of metal pins inserted into a socket on the motherboard, a TCP processor is bonded to an oversized piece of polyimide film, not unlike photographic film, using a process called tape automated bonding (TAB), the same process used to connect electrical connections to LCD panels. The film, called the tape, is laminated with copper foil that is etched to form the leads that will connect the processor to the motherboard (see Figure 20.5). This is similar to the way that electrical connections are photographically etched onto a printed circuit board. Once the leads are formed, they are plated with gold to guard against corrosion, bonded to the processor chip itself, and then the whole assembly is coated with a protective resin.

FIG. 20.5  A processor that is mounted using the tape carrier package is attached to a piece of copper-laminated film that replaces the pins used in standard desktop processors.

After testing, the processor ships to the motherboard manufacturer in this form. To mount the processor on a motherboard, the tape is cut to the proper size and the ends are folded into a modified gull wing shape that allows the leads to be soldered to the motherboard while the processor itself is suspended slightly above it (see Figure 20.6). Before the actual soldering takes place, a thermally conductive paste is added between the actual processor chip and the motherboard. Heat can therefore be dissipated through a sink on the underside of the motherboard itself while it is kept away from the soldered connections. Because TCP processors are soldered to the motherboard, they are not usually upgradeable.

It should be noted that there are manufacturers of portable systems who use standard desktop PGA processors, sometimes accompanied by fans. Apart from a greatly diminished battery life, systems like these can sometimes be too hot to touch comfortably. For this reason, it is a good idea to determine the exact model processor used in a system that you intend to purchase, and not just the speed.

FIG. 20.6  The tape keeps the processor's connections to the motherboard away from the chip itself and allows the underside of the processor to be thermally bonded to the motherboard.

Voltage Reduction

Intel has also taken measures to reduce the amount of power used by its mobile processors, which extends battery life and helps to reduce the output of heat. The pinouts of mobile Pentiums have operated at 3.3v since the original 75MHz chip, but the newer and faster models now incorporate a voltage reduction technology that uses only 2.9v for the chip's internal operations while retaining the 3.3v interface with the motherboard.

This results in processors that use up to 40 percent less power than their desktop counterparts without forcing systems manufacturers to modify the electrical standards that they use to design their machines.

Memory

Adding memory is one of the most common upgrades performed on computers, and portables are no exception. However, most portable systems are exceptional in the design of their memory chips. Unlike desktop systems, in which there are only three basic types of slots for additional RAM, there are literally dozens of different memory chip configurations that have been designed to shoehorn memory upgrades into the tightly packed cases of portable systems.

Some portables use extender boards like the SIMMs and DIMMs in desktop systems, while others use memory cartridges that look very much like PC Cards, but which plug into a dedicated IC (integrated circuit) memory socket. In any case, appearance is not a reliable measure of compatibility. It is strongly recommended that you install only memory modules that have been designed for your system, in the configurations specified by the manufacturer.

This recommendation does not necessarily limit you to purchasing memory upgrades only from the manufacturer. There are now many companies that manufacture memory upgrade modules for dozens of different portable systems, by reverse-engineering the original vendor's products. This adds a measure of competition to the market, usually making third-party modules much cheaper than those of a manufacturer that is far more interested in selling new computers for several thousand dollars rather than memory chips for a few hundred.

NOTE: Some companies have developed memory modules that exceed the original specifications of the system manufacturer, allowing you to install more memory than you could with the maker's own modules. Some manufacturers, such as IBM, have certification programs to signify their approval of these products. Otherwise, there is an element of risk involved in extending the system's capabilities in this way.

Inside the memory modules, the components are not very different from those of desktop systems. Portables use the same types of DRAM and SRAM as desktops, including the new memory technologies like EDO (Enhanced Data Out). At one time, portable systems tended not to have memory caches because the SRAM chips typically used for this purpose generate a lot of heat. Advances in thermal management have now made this less of an issue, however, and you expect a high-end system to include SRAM cache memory.

Hard Disk Drives

Hard disk drive technology is also largely unchanged in portable systems, except for the size of the disks and their packaging. Enhanced IDE drives are all but universal in portable computers, with the exception of Macintosh, which uses SCSI. Internal hard drives typically use 2 1/2-inch platters, and are 12.5mm or 19mm tall, depending on the size of the system.

As with memory modules, systems manufacturers have different ways of mounting hard drives in the system, which can cause upgrade compatibility problems. Some systems use a caddy to hold the drive and make the electrical and data connections to the system. This makes the physical part of an upgrade as easy as inserting a new drive into the caddy and then mounting it in the system. In other cases, you might have to purchase a drive that has been specifically designed for your system, with the proper connectors built into it.

In many portables, replacing a hard drive is much simpler than in a desktop system. Multiple users can share a single machine by snapping in their own hard drives, or you can use the same technique to load different operating systems on the same system.

The most important aspect of hard drive upgrades that you must be aware of is the drive support provided by the system's BIOS. The BIOS in some systems, and particularly older ones, may offer limited hard drive size options. This is particularly true if your system was manufactured before 1995 or so, when EIDE hard drives came into standard use. BIOSes made before this time support a maximum drive size of 508M. In some cases, flash BIOS upgrades may be available for your system, which provide support for additional drives.

Another option for extending hard drive space is PC Card hard drives. These are devices that fit into a type III PC Card slot that can provide as much as 450M of disk space in a remarkably tiny package (usually with a remarkably high price). You can also connect external drives to a portable PC, using a PC Card SCSI host adapter or specialized parallel port drive interface. This frees you from the size limitations imposed by the system's case, and you can use any size SCSI drive you want, without concern for BIOS limitations.

Removable Media

Apart from hard disk drives, portable systems are now being equipped with other types of storage media that can provide access to large amounts of data. CD-ROM drives are now available in many laptop and notebook systems while a few include removable cartridge drives, such as Iomega's Zip drive. This has been made possible by the Enhanced IDE specifications that let other types of devices share the same interface as the hard drive.

Another important issue is that of the floppy drive. Small subnotebook systems usually omit the floppy drive to save space, sometimes including an external unit. For certain types of users, this may or may not be an acceptable inconvenience. Many users of portables, especially those that frequently connect to networks, have little use for a floppy drive. This is becoming increasingly true even for the installation of applications, as more and more software now ships on CD-ROMs.

One of the features that is becoming increasingly common in laptop and notebook systems is swappable drive bays that you can use to hold any one of several types of components. This arrangement lets you customize the configuration of your system to suit your immediate needs. For example, when traveling you might remove the floppy drive and replace it with an extra battery, or install a second hard drive when your storage needs increase.

PC Cards

In an effort to give laptop and notebook computers the kind of expandability that users have grown used to in desktop systems, the Personal Computer Memory Card International Association (PCMCIA) has established several standards for credit card-size expansion boards that fit into a small slot on laptops and notebooks. The development of the PC Card interface is one of the few successful feats of hardware standardization in a market full of proprietary designs.

The PC Card standards, which were developed by a consortium of more than 300 manufacturers (including IBM, Toshiba, and Apple), have been touted as being a revolutionary advancement in mobile computing. PC Card laptop and notebook slots enable you to add memory expansion cards, fax/modems, SCSI adapters, network interface adapters, and many other types of devices. If your computer has PC Card slots that conform to the standard developed by the PCMCIA, you can insert any type of PC Card (built to the same standard) into your machine and expect it to be recognized and usable.

Detailed online information about the standard is available at

http://www.pccard.com

The promise of PC Card technology is enormous. There are not only memory expansion cards, tiny hard drives, and wireless modems, but also ISDN adapters, MPEG decoders, network interface adapters, sound cards, CD-ROM controllers, and even GPS systems that use global positioning satellites to locate your exact position on the earth.

Originally designed as a standard interface for memory cards, the PCMCIA document defines both the PC Card hardware and the software support architecture used to run it. The PC Cards defined in version 1 of the standard, called Type I, are credit card size (3.4x2.1 inches) and 3.3mm thick. The standard has since been revised to support cards with many other functions. The third version, called PC Card Specification--February 1995, defines three types of cards; the only difference between each one is their thickness. This was done to support the hardware for different card functions.

Most of the PC Cards on the market today, such as modems and network interface adapters, are 5mm thick Type II devices. Type III cards are 10.5mm thick and are typically used for PC Card hard drives. All of the card types are backwards compatible; you can insert a Type I card into a Type II or III slot. The standard PC Card slot configuration for portable computers is two Type II slots, with one on top of the other. This way, a single Type III card can be inserted, taking up both slots but using only one of the connectors.

NOTE: There is also a Type IV PC Card, thicker still than the Type III, that was designed for higher-capacity hard drives. This card type is not recognized by the PCMCIA, however, and is not included in the standard document. There is, therefore, no guarantee of compatibility between Type IV slots and devices, and they should probably be avoided.

The latest version of the standard, published in March 1997, includes many features designed to increase the speed and efficiency of the interface, such as:

The PC Card itself usually has a sturdy metal case, and is sealed except for the interface to the PCMCIA adapter in the computer at one end, which consists of 68 tiny pinholes. The other end of the card may contain a connector for a cable leading to a telephone line, a network, or some other external device.

The pinouts for the PC Card interface are shown in Table 20.1.

Table 20.1  Pinouts for a PCMCIA Card

Pin Signal Name
1 Ground
2 Data 3
3 Data 4
4 Data 5
5 Data 6
6 Data 7
7 -Card Enable 1
8 Address 10
9 -Output Enable
10 Address 11
11 Address 9
12 Address 8
13 Address 13
14 Address 14
15 -Write Enable/-Program
16 Ready/-Busy (IREQ)
17 +5V
18 Vpp1
19 Address 16
20 Address 15
21 Address 12
22 Address 7
23 Address 6
24 Address 5
25 Address 4
26 Address 3
27 Address 2
28 Address 1
29 Address 0
30 Data 0
31 Data 1
32 Data 2
33 Write Protect (-IOIS16)
34 Ground
35 Ground
36 -Card Detect 1
37 Data 11
38 Data 12
39 Data 13
40 Data 14
41 Data 15
42 -Card Enable 2
43 Refresh
44 RFU (-IOR)
45 RFU (-IOW)
46 Address 17
47 Address 18
48 Address 19
49 Address 20
50 Address 21
51 +5V
52 Vpp2
53 Address 22
54 Address 23
55 Address 24
56 Address 25
57 RFU
58 RESET
59 -WAIT
60 RFU (-INPACK)
61 -Register Select
62 Battery Voltage Detect 2 (-SPKR)
63 Battery Voltage Detect 1 (-STSCHG)
64 Data 8
65 Data 9
66 Data 10
67 -Card Detect 2
68 Ground

PC Card Software Support

PC Cards are by definition hot-swappable, meaning that you can remove a card from a slot and replace it with a different one without having to reboot the system. If your PC Card devices and your operating system conform to the Plug and Play standard, simply inserting a new card into the slot causes the appropriate drivers for the device to be loaded and configured automatically.

To make this possible, two separate software layers are needed on the computer that provide the interface between the PCMCIA adapter (that controls the card slots) and the applications that use the services of the PC Card devices (see Figure 20.7). These two layers are called Socket Services and Card Services. A third module, called an enable, actually configures the settings of the PC Cards themselves.

FIG. 20.7  The Card and Socket Services allow an operating system to recognize the PC Card inserted into a slot and configure the appropriate system hardware resources for the device.

Socket Services

The PCMCIA adapter that provides the interface between the card slots and the rest of the computer is one of the only parts of the PCMCIA architecture that is not standardized. There are many different adapters available to portable systems manufacturers and, as a result, an application or operating system cannot address the slot hardware directly, as it can a parallel or serial port.

Instead, there is a software layer called Socket Services that is designed to address a specific make of PCMCIA adapter hardware. The Socket Services software layer isolates the proprietary aspects of the adapter from all of the software operating above it. The communications between the driver and the adapter may be unique, but the other interface, between the Socket Services driver and the Card Services software, is defined by the PCMCIA standard.

Socket Services can take the form of a device driver, a TSR program run from the DOS prompt (or the AUTOEXEC.BAT file), or a service running on an operating system like Windows 95 or Windows NT. It is possible for a computer to have PC Card slots with different adapters, as in the case of a docking station that provides extra slots in addition to those in the portable computer itself. In this case, the computer can run multiple Socket Services drivers, all of which communicate with the same Card Services program.

Card Services

The Card Services software communicates with Socket Services and is responsible for assigning the appropriate hardware resources to PC Cards. PC Cards are no different from other types of bus expansion cards, in that they require access to specific hardware resources in order to communicate with the computer's processor and memory subsystems. If you have ever inserted an ISA network interface card into a desktop system, you know that you must specify a hardware interrupt, and maybe an I/O port or memory address in order for the card to operate.

A PC Card network adapter requires the same hardware resources, but you do not manually configure the device using jumpers or a software utility as you would an ISA card. The problem is also complicated by the fact that the PCMCIA standard requires that the computer be able to assign hardware resources to different devices as they are inserted into a slot. Card Services addresses this problem by maintaining a collection of various hardware resources that it allots to devices as needed, and reclaims as the devices are removed.

If, for example, you have a system with two PC Card slots, the Card Services software might be configured to use two hardware interrupts, two I/O ports, and two memory addresses, whether any cards are in the slots at boot time or not. No other devices in the computer can use those interrupts. When cards are inserted, Card Services assigns configuration values for the settings requested by the devices, ensuring that the settings allotted to each card are unique.

Card Services is not the equivalent of Plug and Play, although the two may seem similar. In fact, in Windows 95, Card Services obtains the hardware resources that it assigns to PC Cards using Plug and Play. For other operating systems, the resources may be allotted to the Card Services program using a text file or command-line switches. In a non-Plug and Play system, you must configure the hardware resources assigned to Card Services with the same case that you would configure an ISA board. Although Card Services won't allow two PC Cards to be assigned the same interrupt, there is nothing in the PCMCIA architecture to prevent conflicts between the resources assigned to Card Services and those of other devices in the system.

You can have multiple Socket Services drivers loaded on one system, but there can be only one Card Services program. Socket Services must always be loaded before Card Services.

Enablers

One of the oldest rules of PC configuration is that the software configuration must match that of the hardware. For example, if you configure a network interface card to use interrupt 10, then you must also configure the network driver to use the same interrupt, so that it can address the device. Today, this can be confusing because most hardware is configured not by physically manipulating jumpers or DIP switches, but by running a hardware configuration utility.

In spite of their other capabilities, neither Socket Services nor Card Services are capable of actually configuring the hardware settings of PC Cards. This job is left to a software module called an enabler. The enabler receives the configuration settings assigned by Card Services and actually communicates with the PC Card hardware itself to set the appropriate values.

Like Socket Services, the enabler must be designed to address the specific PC Card that is present in the slot. In most cases, a PCMCIA software implementation includes a generic enabler, that is, one that can address many different types of PC Cards. This, in most cases, lets you insert a completely new card into a slot and have it be recognized and configured by the software.

The problem with a generic enabler, and with the PCMCIA software architecture in general, is that it requires a significant amount of memory to run. Because it must support many different cards, a generic enabler can require 50K of RAM or more, plus another 50K for the Card and Socket Services. For systems running DOS (with or without Windows 3.1), this is a great deal of conventional memory, just to activate one or two devices. Once installed and configured, the PC Card devices may also require additional memory, for network, SCSI, or other device drivers.

NOTE: Windows 95 is definitely the preferred operating system for running PC Card devices. The combination of its advanced memory management, its Plug and Play capabilities, and the integration of the Card and Socket Services into the operating system makes the process of installing a PC Card usually as easy as inserting it into the slot.

When memory is scarce, there can be an alternative to the generic enabler. A specific enabler is designed to address only a single specific PC Card, and uses much less memory for that reason. Some PC Cards ship with specific enablers that you can use in place of a generic enabler. However, it is also possible to use a specific enabler when you have a PC Card that is not recognized by your generic enabler. You can load the specific enabler to address the unrecognized card, along with the generic enabler to address any other cards that you may use. This practice, of course, increases the memory requirements of the software.

Finally, you can avoid the overhead of generic enablers and the Card and Socket Services entirely, by using what is known as a point enabler. A point enabler is a software module that is included with some PC Cards that addresses the hardware directly, eliminating the need for Card and Socket Services. As a result, the point enabler removes the cap-ability to hot swap PC Cards and have the system automatically recognize and configure them. If, however, you intend to use the same PC Cards all of the time, and have no need for hot swapping, point enablers can save you a tremendous amount of memory.

Keyboards

Unlike desktop systems, portables have keyboards that are integrated into the one-piece unit. This makes them difficult to repair or replace. The keyboard should also be an important element of your system purchasing decision, because manufacturers are forced to modify the standard 101-key desktop keyboard layout in order to fit in a smaller case.

The numeric keypad is always the first to go in a portable system's keyboard. Usually, its functionality is embedded into the alphanumeric keyboard and activated by a Function key combination. The Function (or Fn) key is an additional control key used on many systems to activate special features like the use of an alternate display or keyboard.

Most systems today have keyboards that approach the size and usability of desktop models. This is a vast improvement over some older designs in which keys were reduced to a point at which you could not comfortably type with both hands. Standard conventions like the "inverted T" cursor keys were modified, causing extreme user displeasure.

Some systems still have half-sized function keys, but one of the by-products of the larger screens found in many of today's portable systems is more room for the keyboard. Thus, many manufacturers are taking advantage of the extra space.

Pointing Devices

As with the layout and feel of the keyboard, pointing devices are a matter of personal taste. Most portable systems have pointing devices that conform to one of the three following types (see Figure 20.8):

FIG. 20.8a Portable systems are usually equipped with one (or two) of these three pointing devices.

FIG. 20.8b

FIG. 20.8c 

The trackpad is still a relatively new innovation to the portable system market (although the technology has been around for years), and some systems allow users a choice by providing both a trackpoint and a trackpad. Unfortunately, on most of the systems that do this, the two devices use the same interrupt, forcing you to select one device or the other in the system BIOS, rather than letting you use both at the same time.

Another important part of the pointer arrangement is the location of the primary and secondary buttons. Some systems locate the buttons in peculiar configurations that require unnatural contortions to perform a click-and-drag operation. Pointing devices are definitely a feature of a portable system that you should test drive before you make a purchase. As an alternative, remember that nearly all portables have a serial port that you can use to attach an external mouse to the system, when your workspace permits.

Batteries

Battery life is one of the biggest complaints that users have about portable systems. Although a great deal has been done to improve the power management capabilities of today's portable computers, the average hardware configuration has grown enormously at the same time. The efficiency of a computer's power utilization may have doubled in the last two years, but the power required by the system in order to run a faster processor and a CD-ROM drive has also doubled, leaving the battery life the same as it was before.

Battery Types

The battery technology also has a role in the issue, of course. Most portable systems today use one of three battery types:

NOTE: All of the battery types in use today function best when they are completely discharged before being recharged. Lithium Ion batteries are affected the least by incomplete discharging, but the effect on the life of future charges is still noticeable. When storing charged batteries, refrigerating them helps them to retain their charges for longer periods.

Unfortunately, buying a portable computer with a Li-Ion battery does not necessarily mean that you will realize a longer charge life. Power consumption depends on the components installed in the system, the power management capabilities provided by the system software, and the size of the battery itself. Some manufacturers, for example, when moving from NiMH to Li-Ion batteries, see this as an opportunity to save some space inside the computer. They decide that since they are using a more efficient power storage technology, they can make the battery smaller and still deliver the same performance.

Unfortunately, battery technology trails behind nearly all of the other subsystems found in a portable computer, as far as the development of new innovations is concerned. Power consumption in mobile systems has risen enormously in recent years, and power systems have barely been able to keep up. On a high-end laptop system, a battery life of two hours is very good, even with all of the system's power management features activated.

One other way that manufacturers are addressing the battery life problem is by designing systems that are capable of carrying two batteries. The inclusion of multipurpose bays within the system's case enables users to replace the floppy or CD-ROM drive with a second battery pack, effectively doubling the computer's power supply.

Power Management

There are various components in a computer that do not need to run continuously while the system is turned on. Mobile systems often conserve battery power by shutting down these components based on the activities of the user. If, for example, you open a text file in an editor, the entire file is read into memory and there is no need for the hard drive to spin while you are working on the file.

After a certain period of inactivity, a power management system can park the drive heads and stop the platters from spinning until you save the file to disk or issue any other call that requires its service. Other components, such as floppy and CD-ROM drives and PC Cards, can be powered down when not in use, resulting in a significant reduction of the power needed to run the system.

Most portables also have systemic power saver modes that suspend the operation of the entire system when it is not in use. The names assigned to these modes can differ, but there are usually two system states that differ in that one continues to power the sys-tem's RAM while one does not. Typically, a "suspend" mode shuts down virtually the entire system after a preset period of inactivity except for the memory. This re- quires only a small amount of power and allows the system to be re-awakened almost instantaneously.

Portable systems usually have a "hibernate" mode as well, which writes the current contents of the system's memory to a special file and then shuts down the system, erasing memory as well. When the computer is reactivated, the contents of the file are read back into memory and work can continue. The reactivation process takes a bit longer in this case, but the system conserves more power by shutting down the memory array.

NOTE: The memory swap file used for hibernate mode may, in some machines, be located in a special partition on the hard drive, dedicated to this purpose. If you inadvertently destroy this partition, you may need a special utility from the system manufacturer to re-create the file.

In most cases, these functions are defined by the APM (Advanced Power Management) standard, a document developed jointly by Intel and Microsoft that defines an interface between an operating system power management policy driver and the hardware-specific software that addresses devices with power management capabilities. This interface is usually implemented in the system BIOS.

However, as power management techniques continue to develop, it becomes increasingly difficult for the BIOS to maintain the complex information states needed to implement more advanced functions. There is, therefore, another standard under development by Intel, Microsoft, and Toshiba called ACPI (the Advanced Configuration and Power Interface), which is designed to implement power management functions in the operating system.

Placing power management under the control of the OS allows for a greater interaction with applications. For example, a program can indicate to the operating system which of its activities are crucial, forcing an immediate activation of the hard drive, and which can be delayed until the next time that the drive is activated for some other reason.

Peripherals

There are a great many add-on devices available for use with portable systems, which provide functions that cannot practically or economically be included inside the system itself. Many of the most common uses for portable systems may require additional hardware, either because of the computer's location or the requirements of the function itself. The following sections discuss some of the most common peripherals used with portable systems.

External Displays

High-end laptop systems are often used to host presentations to audiences that can range in size from the boardroom to the auditorium. For all but the smallest groups of viewers, some means of enlarging the computer display is needed. Most portable systems are equipped with a standard VGA jack that allows the connection of an external monitor.

Systems typically allow the user to choose whether the display should be sent to the internal, the external, or both displays, as controlled by a keystroke combination or the system BIOS. Depending on the capabilities of the video display adapter in the computer, you may be able to use a greater display resolution on an external monitor than you would on the LCD panel.

For environments where the display of a standard monitor is not large enough, there are several alternatives, as discussed in the following sections.

Overhead LCD Display Panels

An LCD display panel is something like the LCD screen in your computer except that there is no back on it, making the display transparent. The display technologies and screen resolution options are the same as those for the LCD displays in portable systems, although most of the products on the market use active matrix displays.

You use an LCD panel by placing it on an ordinary overhead projector, causing the image on the panel to be projected onto a screen (or wall). Because they are not intended solely for use with portable systems, these devices typically include a pass-through cable arrangement, allowing a connection to a standard external monitor, as well as the panel.

While they are serviceable for training and internal use, overhead display panels do not usually deliver the depth and vibrancy of color that can add to the excitement of a sales presentation. The quality of the image is also dependent on the brightness of the lamp used to display it. While the panel itself is fairly small and lightweight, overhead projectors frequently are not. If a projector is already available at the presentation site, the LCD panel is a convenient way of enlarging your display. If, however, you are in a position where you must travel to a remote location and supply the overhead projector yourself, you are probably better off with a self-contained LCD projector.

Overhead display panels are not cheap. You will probably have to pay more for one than you paid for your computer. However, on a few models of the IBM ThinkPad, you can remove the top cover that serves as the back of the display and use the LCD screen as an overhead panel.

LCD Projectors

An LCD projector is essentially a self-contained unit that is a combination of a transparent display panel and a projector. The unit connects to a VGA jack like a regular monitor and frequently includes speakers that connect with a separate cable. Not all LCD projectors are portable; some are intended for permanent installations. The portable model varies in weight, display technologies, and the brightness of the lamp, which is measured in lumens.

A 16-pound unit delivering 300-400 lumens is usually satisfactory for a conference room environment; a larger room may require a projector delivering 500 lumens or more, weighing up to 25 pounds. LCD projectors tend to deliver images that are far superior to those of overhead panels; and they offer a one-piece solution to the image-enlargement problem. However, you have to pay dearly for the convenience. Prices of LCD projectors can range from $4,000 to well over $10,000, but if your business relies on your presentations, the cost may be justified.

TV-Out

One of the simplest display solutions is a feature that is being incorporated into many of the high-end laptop systems on the market today. It allows you to connect the computer to a standard television set. Called TV-out, various systems provide support for either the North American NTSC television standard, the European PAL standard, or both. Once connected, a software program lets you calibrate the picture on the TV screen.

TV-out is becoming a popular feature on high-end video adapters for desktop systems, as well as portables. There are also some manufacturers that are producing external boxes that plug into any computer's VGA port and a television set, to provide an external TV display solution. The products convert the digital VGA signal to an analog output that typically can be set to use the NTSC or PAL standard.

Obviously, TV-out is an extremely convenient solution, as it provides an image size that is limited only by the type of television available, costs virtually nothing, and adds no extra weight to the presenter's load (unless you have to bring the TV yourself). You can also connect your computer to a VCR and record your presentation on standard videotape. However, the display resolution of a television set does not approach that of a computer monitor, and the picture quality suffers as a result. This is particularly noticeable when displaying images that contain a lot of text, such as presentation slides and Web sites. It is recommended that you test the output carefully with various size television screens before using TV-out in a presentation environment.

Docking Stations

Now that many portable systems are being sold as replacements for standard desktop computers, docking stations are becoming increasingly popular. A docking station is a desktop unit to which you attach (or dock) your portable system when you are at your home or office. At the very least, a docking station provides an AC power connection, a full-size keyboard, a mouse, a complete set of input and output ports, and a VGA jack for a standard external monitor.

Once docked, the keyboard and display in the portable system are deactivated, but the other components, particularly the processor, memory, and hard drive, remain active. You are essentially running the same computer, but using a standard full-size desktop interface. Docking stations can also contain a wide array of other features, such as a network interface adapter, external speakers, additional hard disk or CD-ROM drives, additional PC Card slots, and a spare battery charger.

An operating system like Windows 95 can maintain multiple hardware profiles for a single machine. A hardware profile is a collection of configuration settings for the devices accessible to the system. To use a docking station, you create one profile for the portable system in its undocked state, and another that adds support for the additional hardware available while docked.

The use of a docking station eliminates much of the tedium involved in maintaining separate desktop and portable systems. With two machines, you must install your applications twice, and continually keep the data between the two systems synchronized. This is traditionally done using a network connection or the venerable null modem cable (a crossover cable used to transfer files between systems by connecting their parallel or serial ports). With a docking station and a suitably equipped portable, you can achieve the best of both worlds.

Docking stations are highly proprietary items that are designed for use with specific computer models. Prices vary widely depending on the additional hardware provided, but since a docking station lacks a CPU, memory, and a display, the cost is usually not excessive.

Connectivity

One of the primary uses for portable computers is to keep in touch with the home office while traveling by using a modem. Because of this, many hotels and airports are starting to provide telephone jacks for use with modems, but there are still many places where finding a place to plug into a phone line can be difficult. There are products on the market, however, that can help you to overcome these problems, even if you are traveling overseas.

Line Testers

Many hotels use digital PBXs for their telephone systems, which typically carry more current than standard analog lines. This power is needed to operate additional features on the telephone itself, such as message lights and LCD displays. This additional current can permanently damage your modem without warning, and unfortunately, the jacks used by these systems are the same standard RJ-11 connectors used by traditional telephones.

To avoid this problem, you can purchase a line-testing device for about $50 that plugs into a wall jack and measures the amount of current on the circuit. It then informs you whether or not it is safe to plug in your modem.

Acoustic Couplers

On those occasions when you cannot plug your modem into the phone jack, or when there is no jack available, such as at a pay phone, the last resort is an acoustic coupler. The acoustic coupler is an ancient telecommunications device that predates the system of modular jacks used to connect telephones today. To connect to a telephone line, the coupler plugs into your modem's RJ-11 jack at one end and clamps to a telephone handset at the other end. A speaker at the mouthpiece and a microphone at the earpiece allow the audible signals generated by the modem and the phone system to interact.

The acoustic coupler may be an annoying bit of extra baggage to have to carry with you, but it is the one foolproof method for connecting to any telephone line without having to worry about international standards, line current, or wiring.

Previous chapterNext chapterContents


Macmillan Computer Publishing USA

© Copyright, Macmillan Computer Publishing. All rights reserved.