![]() | Functions of DOS, Windows 3.x and Windows 95 |
![]() | Operating System's Files and Structure |
![]() | File Allocation Tables, Formatting and Partitioning |
![]() | Viewing and Changing File Attributes |
![]() | Operating System Navigation |
![]() | Managing Files and Directories in DOS/Windows |
![]() | Viewing and Changing File Attributes |
![]() | Disk Management |
![]() | File Allocation Tables |
This chapter serves to introduce you to some of the inner workings of the PC. It provides an overview of some of the more popular operating systems and contrasts some of their differences. By the time you complete this chapter, you should be familiar with the basic structure and operation of both DOS and Windows.
By definition, an operating system (OS) is a set of computer instruction codes, usually compiled into executable files, whose purpose is to define input and output devices and connections, and provide instructions for the computers central processor to operate on to retrieve and display data. The coding is usually broken down into many small, modular component files rather than a few large and complex files.
While there have been several competent, if not commercially successful, operating systems developed for the PC (CP/M, QDOS, NDOS, Dr. DOS, Geoworks, NeXTSTEP,among others), at this time, the PC workstation market belongs primarily to Microsoft, with DOS, Windows, and Windows NT maintaining a healthy market advantage over the various flavors of UNIX and IBMs OS/2. Originally, there were several vendors competing for a share in the "DOS wars," but Microsoft has since dominated the market with its various versions of MS-DOS. MS-DOS reached version 6.22 before being wholly incorporated into Windows 95 as version 7.
DOS is a widely used acronym for Disk Operating System. Microsoft was asked by IBM to develop a disk-based OS for the Personal Computer, which was still in development. Microsoft accepted this challenge and purchased a precursor to DOS from a Seattle company for $50,000. Microsoft modified this OS and created DOS 1.0.
DOS consists of a set of instruction files that include basic commands and drivers to allow a user to load and execute a set of computer instructions, also known as software. These instructions are loaded into the computer through multiple types of hardware devices and storage media.
DOS was created to give PC users easy access to applications and data stored on the PCs storage components. Given the limitations of the then-current hardware, it fulfilled its goal fairly well. It went through several development iterations before reaching a level of adequate device coverage and true stability. Through the use of its internal and external commands placed in scripting or batch files by systems programmers to provide simple task-selection menus, the vagaries of DOS became mostly invisible to the majority of computer users.
At the time DOS was first developed, floppy disks and cassette tapes were the only storage media available for the PC. Hard drives, often known as "Winchesters," taken from the name of one of the first successful manufacturers, were soon to be supported. Modified from their eight-inch version size, used in the large system world, hard drives were changed to fit the smaller PC format.
As DOS has evolved over the years, it has grown to support a multitude of other components, such as sound devices, scanners, CD-ROMs, tape drives, and so forth. It has also been modified to make provisions for networking activity. While most of that function has been off-loaded by the various vendors whose subsistence is drawn from network development, Microsoft has been a networking player since its inception, and the networking capabilities of both Windows and Windows NT are unmatched in their support for most every available networking hardware item.
DOS provides a fairly complete set of utilities and commands to perform most computer operations. However, two of its inherent limitations were that it was text-based, requiring strings of commands to be entered to perform the simplest of tasks, necessitating in-depth knowledge of the command structure and syntax, and that it only allowed a single application to be loaded and executed.
Through research done by the XEROX Corporations Palo Alto Research Center (PARC) into computer usage, it was demonstrated that through the use of a graphical computer environment, users could achieve a much higher level of production due to the enhanced accessibility of an applications components and features. Developing graphical applications with what was to be known as a "Common User Interface," where the access to compatible features remained consistent across the various applications, provided further enhancement.
This idea was seized on by both Apple Computer with its Macintosh computer, unveiled in the mid-1980s and shipped complete with a graphical OS, and Microsoft, who teamed with IBM to develop their own version of a graphical environment. This co-development became Windows. Version 1.0 of Windows hit the streets in 1985, with a resounding thud. While the Macintosh was delivered from the start with hardware capable of supporting the new graphical environment, Microsoft had the baggage of an already-installed user base of millions of computers already geared to text-based computing. Unlike the Macintosh, whose OS was included and preinstalled since its origin, DOS and Windows are (or were) additional-cost options, licensed separately from the PC on which they resided. These days, most systems vendors preinstall a copy of DOS and/or Windows prior to shipment, and include the cost of the license in their "system package."
Unlike the Mac, with its Motorola 680XX series processors, neither the Intel 8088 microprocessor or the then-current CGA video standard were capable of better than meager support for high-level graphics. It took the development of the 80286 microprocessor and EGA graphics to get Windows off the ground, and the 80386 and VGA to ensure its success.
Continuous improvements in available system "horsepower" meant that the development of a next-generation OS was needed to take better advantage of the burgeoning hardware and applications that were becoming available for the PC. Microsoft and IBM teamed up together to create a new, more powerful version of DOS and Windows to leverage the new hardware.
DOS was developed as a 16-bit application, with its internal code segments written to be accessed 16-bits at a time. This fit the 808X processor and memory format. Once Intel moved on to its 80X86 architecture, the processor and system memory were then able to read and write 32-bits of data per clock cycle. The two giants were determined to develop a completely new 32-bit OS that would take full advantage of this feature, and yet would be backward-compatible with all existing DOS and Windows software. This new OS was to become OS/2.
The Microsoft/IBM alliance ended in 1989, when it was decided by both companies in a much-publicized battle to split their development of OS/2. IBM was to keep all rights to existing and future versions of OS/2, and Microsoft got exclusive rights to Windows and further Windows development, soon to be known as Windows NT (New Technology)
The second limitation of DOS, single-task usage, also was hammered down. While a microprocessor could only operate on a single set of instructions (not true now, because Pentium-class processors can now "pipeline" up to 4 concurrent operations ) at any given moment, given their speed, task-switching became possible at a rate that produced the effect of multi-tasking. This ability allowed several applications to be loaded into memory concurrently, and a user could then easily switch between them. Windows, OS/2, and the Macintosh, along with some DOS vendors (notably Quarterdeck with its Deskview software) could not only handle the access, but also provided an easy method to exchange data between the concurrently loaded applications. Known as cut and paste, a user could select data, make a copy in memory, and then place the data (usually with its formatting characteristics) into a different application. This made juggling such things as e-mail use, database access, document creation, and event scheduling a snap.
Unlike Windows NT, and IBMs OS/2, which are true and complete OSs, Windows (up to Windows 95) is a graphical shell that overlays on DOS, and relies on DOS for its low-level device access. It even requires a separate license from DOS. In order to install Windows (pre-95), a user first had to install a working copy of DOS to the PC. Even though Windows 95 was created to eliminate a requirement for DOS and its reliance on a 16-bit architecture, it still relies on several DOS components, mostly for compatibility reasons. It now also includes the needed DOS components under the same installation and license, and doesnt require a pre-existing copy of DOS. The majority of its internal operations are performed 32-bits wide.
Operating systems provide both system access and a set of utilities. The core files are also known as the system files.
System files are responsible for all device access and user interface presentation. While they may require the use of additional device drivers for specific hardware items, by themselves they can startup (boot, or IPL) a computer and give a user some access to direct some specific actions.
These core files are listed in the following sections, along with a brief explanation of their function and contents.
DOS system files include the following:
![]() | AUTOEXEC.BAT |
![]() | CONFIG.SYS |
![]() | IO.SYS |
![]() | MSDOS.SYS |
![]() | COMMAND.COM |
![]() | HIMEM.SYS |
![]() | EMM386.EXE |
![]() | ANSI.SYS |
AUTOEXEC.BAT is located and automatically executed in the ROOT directory at startup. This file contains commands to modify the PC environment (PATH, COMSPEC, other SET commands), and to execute applications. It can be used to create a menu system, prompt for user input, or call other batch files to maintain a modular structure. By default, it carries no attributes, and is NOT required for OS startup.
CONFIG.SYS is located in the ROOT directory and loaded by MSDOS.SYS. This file loads low-level device drivers for specific hardware, and adjusts several system parameters for performance tuning and memory usage. By default, it carries no attributes, and is not required for OS startup. Since version 6.x, it also may be used in conjunction with an internal menu system to select multiple startup configurations.
IO.SYS is located in the ROOT directory, and defines basic input/output routines for the processor. By default, it carries the hidden, system, and read-only attributes, and is required for OS startup.
MSDOS.SYS is located in the ROOT directory, defines system file locations. By default, it carries the hidden, system, and read-only attributes, and is required for OS startup.
COMMAND.COM is located and automatically executed in the ROOT directory at startup. This file contains the internal command set and error messages. By default, it carries no attributes, and is required for OS startup. It may be executed from a different location and/or renamed if a SHELL command statement is placed in the CONFIG.SYS file pointing to the correct location and name.
HIMEM.SYS and EMM386.EXE control memory management and are located in the \DOS directory (\WINDOWS directory in Windows 95). They are not required for system startup (pre-95) and are explained in detail in Chapter 10.
ANSI.SYS is located in the DOS directory and loaded by CONFIG.SYS if required. This file loads an extended character set for use by DOS and DOS applications that includes basic drawing and color capabilities. Normally used for drawing and filling different boxes for menu systems, it is seldom in use today. By default, it carries no attributes, and is not required for OS startup.
Windows 3.x system files include the following:
![]() | WIN.INI |
![]() | SYSTEM.INI |
![]() | USER.EXE |
![]() | GDI.EXE |
![]() | KRNLXXX.exe (KRNL286.EXE for 80286 processors and KRNL386 for 80386) |
![]() | WIN.COM |
The WIN.INI file contains configuration information for Windows applications. Errors made in this file seldom have global implications to Windows operation, but can cripple specific applications or features. Printing is also controlled by settings in this file. The WIN.INI file is dynamic, and records the way the user configures how the system looks and behaves. The spacing of the icons, the type of wall paper, screen colors, and other interface features can be customized in the WIN.INI file. This file, as well as the SYSTEM.INI file, are ASCII text files.
The SYSTEM.INI file configures Windows to address specific hardware devices and their associated settings. Errors in this file can and do cause Windows to fail to start, or crash unexpectedly.
The rest of Windows startup components are explained in detail in the "Components" section, later in this chapter.
The Windows 95 system files include the following:
![]() | IO.SYS |
![]() | MSDOS.SYS |
![]() | COMMAND.COM |
![]() | WIN.INI |
![]() | SYSTEM.INI |
![]() | SYSTEM.DAT |
![]() | USER.DAT |
Because Windows 95 no longer relies on a separate copy of DOS, the fundamentals have changed. While IO.SYS and COMMAND.COM remain for compatibility reasons, MSDOS.SYS no longer contains system code, but allows for special system settings to be user-defined. WIN.INI and SYSTEM.INI are mostly vestigial, allowing for 16-bit (pre-WIN95) applications to still be able to register system variables. The bulk of configuration has now shifted to the SYSTEM.DAT and USER.DAT files. Also, while not listed here, the kernel file, KRNL386.EXE, still teams up with USER.EXE and GDI.EXE to deliver the complete Windows OS.
Because DOS and Windows share a common development heritage and supported hardware, it shouldnt come as a surprise that they share much of the same structure.
DOS installs onto a formatted disk by placing its system files in the root directory. Its external command set and device drivers are then, by default, placed in a directory named DOS just off the root. During installation, an alternate path and name can be chosen, but straying from this default usually leads to trouble.
WIN.COM is a file created by the Setup program when first installing Windows. You wont find a WIN.COM file on any of the distribution media. During installation, Setup takes three files and combines them together to form the WIN.COM file.
When "WIN" is typed on the command line and the Enter key is pressed, WIN.COM performs three functions. The first function of the WIN.COM is to ascertain what type of processor is in the computer. The file WIN.CNF, also located in the Windows\System subdirectory, performs this function.
The second function that WIN.COM performs is to switch the computer into the appropriate graphics mode by loading a *.LGO file for the logo display. The *.LGO file is selected based on the type of video card Setup determined was in the computer during installation. Setup selects only one graphics mode and adds it to the WIN.COM file. The following are the available choices that Setup selects from and installs that *.LGO file into the WIN.COM file and into the Windows\System subdirectory. Windows 95 leaves the logo file out of WIN.COM, and moves it to the root directory as LOGO.SYS
![]() | VGALOGO.LGOfor VGA, Super VGA, or 8514/A displays |
![]() | EGALOGO.LGOfor EGA color displays |
![]() | EGAMONO.LGOfor EGA monochrome displays |
![]() | CGALOGO.LGOfor CGA, EGA B&W (64K) and Plasma displays |
![]() | HERCLOGO.LGO for Hercules Monochrome Graphics displays |
The third function WIN.COM performs is to load the bitmapped graphic advertisement that Windows displays. The file displayed is in a compressed format with a RLE, Run Length Encoded, extension. Setup selects only one of the following available choices and installs it into the WIN.COM file and into the Windows\System subdirectory:
![]() | VGALOGO.RLEfor VGA, Super VGA, or 8514/A displays |
![]() | EGALOGO.RLEfor EGA color displays |
![]() | EGAMONO.RLEfor EGA monochrome displays |
![]() | CGALOGO.RLEfor CGA, EGA B&W (64K) and Plasma displays |
![]() | HERCLOGO.RLE for Hercules Monochrome Graphics displays |
After the three functions of WIN.COM are completed, WIN.COM hands the control of Windows over to either DOSX.EXE (for Standard mode) or WIN386.EXE (for Enhanced mode). Because Windows 95 no longer supports Standard mode, DOSX.EXE has been removed. WIN386.EXE switches the CPU to protected mode, thereby enabling advanced features such as access to all memory in the computer. Windows then loads its kernal, either KRNL286 or KRNL386, and starts a configuration and initialization process. It reads the values stored in the SYSTEM.INI file, and loads the specified device drivers for the display, keyboard and mouse, and other devices. These files include SYSTEM.DRV, KEYBOARD.DRV, MOUSE.DRV, VGA.DRV, MMSOUND.DRV, and COMM.DRV. Basic screen fonts and fonts for DOS applications are loaded next. If VGA mode is selected, the required fonts loaded are VGASYS.FON and VGAOEM.FON for DOS and FONTS.FON, VGAFIX.FON, and OEMFONTS.FON for Windows. The graphics subsystem is then loaded along with the user interface. These files are GDI.EXE and USER.EXE. The display driver, DISPLAY.DRV (used as a label here; the actual driver name will vary) is then loaded. This allows the other drivers to display an error message on screen if they fail to initialize during the next step. Once the DISPLAY.DRV is in control, messages go through Windows. Windows then initializes all drivers and then all fonts.
In the last phase of initialization, Windows 3.x starts a desktop manager, usually Program Manager (PROGMAN.EXE), the default shell that Windows uses unless another desktop manager has been selected in the SYSTEM.INI file. PROGMAN.EXE finishes the Windows initialization by running any startup applications specified in the WIN.INI file, and by starting any applications whose icons are stored in the StartUp Group. The last action PROGMAN.EXE takes is to display its own menus and interfaces for use of the system.
Windows 95 performs much of the same rituals, however it runs only in protected mode, so there is only one kernel, and the WIN.COM file no longer supports switches for real or standard mode. It still reads the SYSTEM.INI and WIN.INI files, and runs whatever applications are listed in the StartUp folder. The big change, though, is the reliance on the system Registry. Windows 3.x began the move to a Registry-based startup with a REG.DAT file, mostly used to support individual application parameters for Dynamic Data Exchange (DDE). This concept was greatly expanded for Windows NT, and then moved to Windows 95 as well. The Registry is a complex database of settings pertaining to both applications and hardware.
Using the FDISK (for Fixed DISK) utility provided with DOS, a hard drive is prepared for use through the act of partitioning and formatting. A partition is created to separate a single physical drive into multiple logical components. Each partition receives a letter, beginning with C:. If a second drive is installed, it is assigned the letter D: automatically. The third becomes E:, and so forth. Additional partitions may be created as desired until the entire drive (or drives) has been allocated. So that the computer knows where to start, one of the partitions (typically C:, but there are sometimes reasons to select a different partition) must be set as "active" with FDISK in order to boot from the hard drive.
Various versions of DOS have been stretched to their limit by the ever-expanding capacities of hard drives. In the early days of DOS, circa 1981, a large hard drive was 10 megabytes (MB). Today, drives for home computers are approaching 10 Gigabytes(GB) at about the same cost (~$400). This represents an increase of 10,000 times in storage capacity.
Once partitioned, a drive must be formatted. This process divides the drive logically into groups of sectors and tracks. Sector size is usually set at 512 bytes, and physically varies in size depending on whether the sectors track is located in the inner or outer portion of the drive platter. Hard drives consist of a number of platters rotating on a common spindle, with read/write heads mounted on a pivoting assembly that spans all platters concurrently. The sectors are combined logically into groups called clusters. When the partition is formatted, an array of addresses that correspond to the physical starting locations of each cluster is created. Known as the File Allocation Table, or FAT, it becomes the pointer for DOS to locate information on the drive.
Due to a limitation in DOSs heritage ("whoda thunk anyoned need any more?") , the FAT on each hard drive partition can catalog 65,535 (roughly 64K) individual addresses. Depending on the size of the partition, the 64K limit dictates the number of sectors, and therefore the size of each cluster. The cluster is the smallest accessible unit to DOS from the FAT of a hard drive. The smallest cluster used by DOS contains 4 512-byte sectors, or 2048 bytes (2KB). Times 65,535, once you reach a partition size of >128MB, cluster size jumps to 4096 (4KB). At >256MB, the jump is to 8KB clusters, at 512MB it goes to 16KB, at 1 GB to 32KB, and so forth. Up until recently, DOS wouldnt support a partition greater than 2 GB.
These increasing cluster sizes have an inherent drawback, the concept of slack. Slack is the space left between the end of a file and the end of the cluster in which the file resides. Because the smallest addressable unit is the cluster, once a file is written to a cluster, the cluster is marked as used and cant have anything else written to it until the file in it is erased or overwritten by another copy of the same file. For large files that span several clusters, this is of little consequence. However, say you write a batch file that has 5 lines of code, perhaps 150 bytes in all. When you save the file on a 1GB partition, youve just wasted about 32,000 bytes of space. Multiply this by the literally thousands of small (<2KB) files found on a typical PC, and you can see why its a good thing large drives have gotten cheap.
When formatting, or afterwards if desired, the partition marked as active must have an operating system installed. In the case of DOS, the FORMAT C: /S command formats the C: (boot) drive and puts the DOS system files on the hard drive. At the same time, it puts a pointer in a special location, known as the boot sector, which is located in the first cluster on the boot partition. This pointer tells the processor exactly where to find the system files. If for some reason this cluster becomes unusable, the drive can no longer be accessed by DOS. If you choose to add the system files at a later time, the SYS command can be used to reinitialize the boot sector and place the DOS system files in the root directory. Lets try that now, okay? Go to the DOS prompt on your C: drive and type FORMAT C: /S. Just Kidding.
Each logical drive contains an initial, or root, directory that is created when it is formatted. Each root directory can hold up to 512 files or other (sub)directories. Also known as folders and subfolders, the files and folders combine into a hierarchical tree known as the directory tree. In DOS, both files and folders are limited to eight characters to the left of a period and three characters to the right, also known as 8 dot 3 naming. The total possible length of a file name, including its path, is 128 characters. This is known as the "fully qualified path." In practice, its best to limit the fully qualified path to less than 60 characters, else some older applications and utilities can display anomalous behavior.
From the DOS command line, there are several commands used to create and manage files and directories. Table 9-1 shows some of the DOS commands. You can get the information listed here at any time by issuing the command with a "/?" extension and pressing the Enter key. First, in order to change from one drive to another, type in the drive letter to change to, followed by a colon (i.e., To change to drive D, type D: and press the Enter key.)
Command | Command Type | Action Performed | Available Command-Line Switches | Syntax |
CD (or CHDIR) | Internal | Displays the name of or changes the current directory | Specifies that you want to change to the
parent directory. Type CD drive: to display the current directory in the specified drive. Type CD without parameters to display the current drive and directory. |
CHDIR [drive:][path] CHDIR[..] CD [drive:][path] CD[..] |
COPY | Internal | Copies one or more files to another location. Notes: DOES NOT COPY SUBDIRECTORIES. The switch /Y may be preset using the COPYCMD environment variable. This may be overridden with /-Y on the command line. To append files, specify a single file for destination, but multiple files for source (using wildcards or file1+file2+file3 format). (This is also known as concantenation.) |
source Specifies the file or files to
be copied. /A Indicates an ASCII text file. /B Indicates a binary file. Destination Specifies the directory and/or filename for the new file. /V Verifies that new files are written correctly. /Y Suppresses prompting to confirm you want to overwrite an existing destination file. /-Y Causes prompting to confirm you want to overwrite an existing destination file. |
COPY [/A | /B] source [/A | /B] [+ source
[/A | /B] [+ ...]] [destination [/A | /B]] [/V] [/Y | /-Y Also; related environment variable, COPYCMD |
DIR | Internal | Displays a list of files and subdirectories in
a directory. Notes: Switches may be preset using the DIRCMD environment variable. You may override preset switches by prefixing any switch with - (hyphen)--for example, /-W. |
[drive:][path][filename] Specifies
drive, directory, and/or files to list. (Could be enhanced file specification or multiple
filespecs.) /P Pauses after each screenful of information. /W Uses wide list format. /A Displays files with specified attributes. Attributes D Directories R Read-only files H Hidden files A Files ready for archiving S System files - Prefix meaning not /O List by files in sorted order. Sortorder N By name (alphabetic) S By size (smallest first) E By extension (alphabetic) D By date & time (earliest first) G Group directories first - Prefix to reverse order A By Last Access Date (earliest first) /S Displays files in specified directory and all subdirectories. /B Uses bare format (no heading information or summary). /L Uses lowercase. /V Verbose mode. /4 Displays year with 4 digits (ignored if /V also given). |
DIR [drive:][path]filename] [/P] [/W] [/A[[:]attributes]] [/O[[:]sortorder]] [/S] [/B] [/L] [/V] [/4] |
DEL (or ERASE) | Internal | Deletes one or more files. | [drive:][path]filename specifies the
file to delete. Specify multiple files by using wildcards. /P Prompts for confirmation before deleting each file. |
DEL [drive:][path]filename [/P] ERASE [drive:][path]filename [/P] |
MOVE | External | Moves files and renames files and directories. | [drive:][path]filename1 Specifies the
location and name of the file or files you want to move. Destination Specifies the new location of the file. Destination can consist of a drive letter and colon, a directory name, or a combination. If you are moving only one file, you can also include a filename if you want to rename the file when you move it. [drive:][path]dirname1 Specifies the directory you want to rename. dirname2 Specifies the new name of the directory. /Y Suppresses prompting to confirm creation of a directory or overwriting of the destination. /-Y Causes prompting to confirm creation of a directory or overwriting of the destination. |
To move one or more files: MOVE [/Y | /-Y] [drive:][path]filename1[,...] destination To rename a directory: MOVE [/Y | /-Y] [drive:][path]dirname1 dirname2 |
MD or MKDIR | Internal | Displays the name of or changes the current directory. | None available | MKDIR [drive:]path MD [drive:]path |
REN or RENAME | Internal | Renames a file/directory or files/directories. | None available | RENAME [drive:][path][directoryname1 |
filename1] [directoryname2 | filename2] REN [drive:][path][directoryname1 | filename1] [directoryname2 | filename2] |
XCOPY | External | Copies files and directory trees. | Source Specifies the file to copy. Destination Specifies the location and/or name of new files. /A Copies files with the archive attribute set, doesn't change the attribute. /M Copies files with the archive attribute set, turns off the archive attribute. /D:date Copies files changed on or after the specified date. If no date is given, copies only those files whose source time is newer than the destination time. /P Prompts you before creating each destination file. /S Copies directories and subdirectories except empty ones. /E Copies directories and subdirectories, including empty ones. Same as /S /E. May be used to modify /T. /W Prompts you to press a key before copying. /C Continues copying even if errors occur. /I If destination does not exist and copying more than one file, assumes that destination must be a directory. /Q Does not display file names while copying. /F Displays full source and destination file names while copying. /L Displays files that would be copied. /H Copies hidden and system files also. /R Overwrites read-only files. /T Creates directory structure, but does not copy files. Does not include empty directories or subdirectories. /T /E includes empty directories and subdirectories. /U Updates the files that already exist in destination. /K Copies attributes. Normal Xcopy will reset read-only attributes. /Y Overwrites existing files without prompting. /-Y Prompts you before overwriting existing files. /N Copy using the generated short names. |
XCOPY source [destination] [/A | /M] [/D[:date]] [/P] [/S [/E]] [/W] [/C] [/I] [/Q] [/F] [/L] [/H] [/R] [/T] [/U] [/K] [/N] |
Table 1: DOS Command Reference
As you can see from the previous examples, most commands include an array of possible options that are accessed by command line switches. These switches are usually preceded by a slash (/), although some use a dash (-). It was the reliance on slash-triggered switches that forced the DOS developers to use the backslash character to distinguish subdirectories. When executing any .BAT, .COM, or .EXE file, the extension may be dropped. Be aware though that there is a hierarchy to execution if multiple commands with the same extension are in the same directory (or if executed from a different directory and they are in the DOS path). For example, say you issue the NET command from the command line. There may be a NET.COM, NET.BAT, and NET.EXE file all in the path. DOS looks first for .COM files, then .EXE files, then .BAT files. If you ever execute a command and get strange results, be sure and check for the presence of another command with the same name, but a different extension.
If you notice, the commands shown in Table 9-1 are listed as either internal or external. Internal commands are present in the COMMAND.COM shell, and can be executed on the command line at any time from any directory location. External commands are those that reside in the DOS directory. External commands must be executed either from there, or they may be run from any subdirectory if the DOS directory is included in the DOS path (see PATH Command) or by preceding the command with the fully qualified path to the DOS directory (e.g., C:\DOS\XCOPY).
File management is crucial for the rapid retrieval of information. One method used to organize data is by using logical filenames within the boundaries set by the operating system.
DOS and Windows 3.1 were limited to what is termed an 8.3 (8 "dot" 3) file naming specification. This meant that no filename could exceed eight characters followed by a period and up to three more characters (i.e. FILENAME.EXT). There was also no provision by the operating system for mixed-case names. Everything was in uppercase.
When Windows95 and Windows NT entered the desktop realm, these limitations were eliminated. Windows95 and Windows NT provide long filename support on FAT. This essentially links directory entries within the FAT together to create a long filename (up to 250 characters) with mixed case and spaces. Windows NT also provides long filename support with its native file system, NTFS (NT File System).
You can save or rename any file to most any combination of eight letters or numbers, a period, and then add a three-letter (or number or combo) extension. DOS only performs an action on certain files based on their three-letter extension. .COM, .EXE, and .BAT files are known as executables. These are the only files that can be executed (or "run") by the computer. These executables can then include additional data or functions by loading additional files. Rather than put all available functions into a single huge executable file, most developers choose to modularize their applications by creating library files that include additional commands and functions. These additional executable enhancement files are usually referred to as overlays. These files usually have a .BIN or .OVL extension. Windows took this one step further and developed Dynamic Linked Libraries (.DLL files). These overlays have the additional benefit of being shareable by all of the applications loaded in Windows.
There are additional "conventions" to file naming, but none that must be adhered to at the OS level. Device drivers usually end in .DRV. You can usually count on .TXT and .DOC files to be documents, .HLP to be help files, and so forth.
On the other hand, software applications are usually sticklers at requiring a certain extension to be recognized as a valid file to be opened by the application. Some applications allow the importing of data from several different file types.
Each file accessible to DOS, regardless of media type, has four attributes that may be associated with it. These attributes are READ ONLY, HIDDEN, SYSTEM, and ARCHIVE.
The READ ONLY (RO) attribute prevents a user or application from inadvertently deleting or changing a file. If set, the attribute must be removed before the file can be deleted or overwritten. If the attribute is removed, the file then has READ/WRITE capabilities (RW)
The HIDDEN (H) attribute keeps a file from being displayed when a DIR command is issued. It also prevents the file from being acted upon by standard DOS commands, such as COPY or XCOPY. Some third-party applications ignore the HIDDEN attribute, and will operate on the files regardless, but this is rare.
The SYSTEM attribute is usually set by DOS or Windows, and cannot be modified using standard DOS or Windows commands, including the ATTRIB command or File Manager. This is to ensure that a user doesnt inadvertently erase or modify these crucial files. Again, third-party utilities can sometimes be used to override and change this attribute.
The ARCHIVE attribute is set automatically when a file is created or modified, and is automatically removed by back-up software when the file is backed up.
File attributes can be viewed from DOS using either the DIR or ATTRIB command. Windows users can either use File Manager (all versions) or Windows Explorer in Windows 95 to both view and/or modify the attributes of selected files.
The ATTRIB command displays or changes file attributes, such as the following:
ATTRIB [+R | -R] [+A | -A] [+S | -S] [+H | -H] [[drive:][path]filename] [/S]
![]() | + Sets an attribute. |
![]() | - Clears an attribute. |
![]() | R Read-only file attribute. |
![]() | A Archive file attribute. |
![]() | S System file attribute. |
![]() | H Hidden file attribute. |
![]() | /S Processes files in all directories in the specified path. |
Notice this last switch. It allows a user to perform a change on ALL files in and below the directory in which the command is issued. The S switch allows a user to reset the SYSTEM attribute only on those files that were previously set on by the user.
There is a specific hierarchy that must be followed to apply the switches, and this hierarchy varies slightly depending on which version of DOS is in use. For example, if a file is hidden, it must be unhidden before the READ ONLY attribute can be set, but if the file is already marked RO, it must be set to RW (using the R switch) before it can be hidden.
All operating systems provide the user with the ability to manage information. The capabilities vary from one to another, but the goal is a common one. Windows, MacOS, OS/2, and so forth give the user the ability to easily manage their files and programs. Because everything is handled graphically, most functions are present for the user regarding data manipulation, but the features that are present are merely a function of the OS developers vision of what should be present. In DOS, and other command-line based OSs, ease of use is discarded in favor of power and control. This gap has gradually closed and will continue to do so as the use of the DOS prompt becomes less of a necessity. Keep in mind that a solid grasp how to accomplish tasks at a DOS prompt will save many hours of frustration when a machine does not boot.
Earlier in this chapter, Table 9-1 references some of the basic commands used in the DOS environment. Of those commands, the most commonly used commands for navigating the waters of the DOS world are CD (Change Directory) and the DIR (Directory) commands.
The CD command is used in the following example shown in Figure 9-1.
Figure 9-1: Changing directories in DOS
The DIR command can be used to search for specific files. This can be accomplished by using the /S (search subdirectories) switch. In most cases, it is best to be in the root, or top, directory of the drive. Figure 9-2 shows an example of this technique.
Figure 9-2: Using the DIR command to perform a file search
Windows pre-95 relies on an application called File Manager to navigate through a users storage devices. Both local and networked drives can be accessed from this single user interface. File Manager helps to organize files and directories. File management can encompass several tasks, such as formatting floppy disks or creating directories, but the primary task is searching for files and placing them in the appropriate locations. The WINFILE.INI is the configuration file that stores the names of the directories that File Manager displays when starting.
File Manager also dynamically tracks all files on the computer. Messages are sent by Windows to File Manager of many operations including file creation and deletion.
File Manager can also be used to start applications. By double-clicking on any document file, Windows automatically tries to find the application that works with that document, and then starts that application. File Manager determines this by the three-character extension of the document file and then consults with the Windows registration database and the [EXTENSIONS] section of the WIN.INI. By selecting a document in the File Manager window and then the File | Run menu option, File Manager also starts an application. File Manager can also be used to print documents by selecting them in a File Manager window and then using the File | Print menu option.
To select a file in File Manager, click on the files name. To select multiple files in File Manger, click on one file and then hold down the Shift key when clicking on a second file. File Manager selects the two specified files and all the files in between them. If the Ctrl key is held down while selecting files, File Manager allows selection of multiple individual files.
I always derive pleasure from watching a professional technician navigate through an operating system. I can pick up a few shortcuts while Im watching. Each operating system has its own structure as well as sharing the structure of other operating systems.
I once watched an administrator fix a system problem through DOS commands. He immediately shelled out of the program manager and started typing very complex commands. After 26 different commands, the network was up and running. Another tech commented that this particular administrator had the ability to think in DOS.
Because much of the work that you will be doing involves navigating through the operating system, you might as well learn as many shortcuts as you can. Your work will not only be quicker, but those who watch you will be impressed. Following is a list of my favorite Windows 95 shortcuts.
Shortcut | Function |
Right click the start button | Launch explorer |
Hit ctrl-esc | Open the task bar |
Ctrl-Tab | Shuffle Excel windows |
F5 | Update window (good when looking at multiple floppies) |
Ctrl-F | Find |
F1 | Help |
Ctrl-ESC the ALT-M | Minimizes all windows |
Ctrl-P | |
Ctrl-X | Cut |
Ctrl-C | Copy |
Ctrl-V | Paste |
Alt-Tab | Rotate active windows |
Shift + | Using Shift plus many shortcuts works in reverse |
F10 | Activate Menus |
Alt-F4 | Close the active window (can be used multiple times to close all windows) |
by Ted Hamilton, MCP, A+ Certified
The Windows Explorer is merely a turbo-charged version of File Manager. It provides a more solid integration with the Windows 95 and NT environments than File Manager. Another benefit is the fact that Explorer is highly configurable and extensible by third-party software vendors.
Due to their capacity to store thousands of files coupled with the constant creation and modification of data contained therein, several utilities have been created to help manage hard drive "real estate." Early versions of DOS came with fairly crude utilities, and many companies made a pretty good living by creating better applications to perform disk management functions. Beginning with DOS version 6.0, a disk repair tool (SCANDISK) and a file defragmentation utility (DEFRAG) were included out of the box. Windows brought several new utilities into play. These include file management utilities, such as File Manager and Explorer, graphical versions of SCANDISK and DEFRAG, and also new file-compression utilities to make better use of disk space.
SCANDISK.EXE and DEFRAG.EXE provided with MS-DOS 6.0 and later provide many benefits, as mentioned earlier. Similar functionality is provided with Windows 95s graphical versions of the programs, which are located in the Programs | Accessories | System Tools program folder.
The Scandisk utility provides the ability to scan and correct some data anomalies on a hard drive before they result in data loss. Typically, it moves the data to a non-damaged portion of the disk whenever possible. Table 9-2 lists the valid command line switches for SCANDISK.EXE
SCANDISK.EXE Command Line Switch | Function |
/ALL | Scans all local drives for errors. |
/AUTOFIX | Fixes damage without prompting for user confirmation. |
/CHECKONLY | Checks drive for damage, but does not repair. |
/CUSTOM | Forces SCANDISK to read the SCANDISK.INI file for settings. |
/FRAGMENT | Checks file for fragmentation. |
/MONO | Causes SCANDISK to use a monochrome display. |
/NOSAVE | Used with /AUTOFIX. Deletes lost clusters rather than saving them as a file. |
/NOSUMMARY | Used with /CHECKONLY and /AUTOFIX. Prevents SCANDISK from displaying summary screens. |
/UNDO | Back out of previous repairs. |
Table 2: Command Line Switches for SCANDISK.EXE
The defragmentation utility included with DOS 6.0 and later provides the ability to rearrange the clusters of data on the hard drive in order to achieve greater performance by placing all of the clusters for a given file together in a contiguous fashion. One mistake that is commonly made is that of running DEFRAG to fix errors. It should be noted that DEFRAG does no repair and will not make any error disappear.
Typically, the only command line switch that is used is /FULL. This causes a full defragmentation to take place. The command line would appear as shown in Figure 9-3.
Figure 9-3: Common DEFRAG.EXE command line syntax
Hard drives are by far the most important part of a computer. Thats because they contain the most vital element: your data. As such, its imperative that the data be kept safe from any failure of the equipment or potential operator error. The best way of doing so is to create a separate copy of the data, also known as a backup. Youd think it would be easy to make copies, because copying files is a feature inherent in DOS, yet this feature has been probably the most problematic of all.
DOS has included backup utilities since version 2.0. Unfortunately, due to the multitude of problems that have arisen, each succeeding version of DOS came with a backup utility that was incompatible with the previous version. This was by far one of the most difficult limitations to overcome. When installing a newer version of DOS, if problems arose with the new install, the new version would be unable to restore the previous versions data, necessitating a reinstallation of the previous DOS version and then often a complete restore. This could take anywhere from hours to days.
Things have gotten better, but not by much. This version incompatibility still occurs, even using Windows and/or third-party software. Be aware that no backup can be assumed to be complete and restorable, unless you actually perform a restore operation and it works correctly. Its a good idea as part of a prudent disaster-recovery plan to perform occasional full restorations of a backup set to a different, similarly configured PC to insure that what you thought you saved really was saved.
Another of the problems inherent in the use of the DOS FAT table is file fragmentation. Because DOS writes files to the hard drive by breaking the file into cluster-sized pieces and then storing each piece in the next available cluster, as files are deleted and then rewritten, they can be written in discontiguous clusters scattered all over the disk. To access a file, the heads that actually read the data from the platters must traverse to each cluster in sequence. If the clusters are contiguously stored, the read operation is smooth and fast. However, if the file is fragmented, the heads fly wildly across the platters, picking up each scattered cluster one by one. This can slow down access dramatically.
Recognizing this, developers created utilities that reorganize the files back in a proper, contiguous fashion. They do so by moving several of them to an unused portion of the drive, erasing the previous locations in contiguous clusters, then rewriting the files back in proper sequence. Performed periodically, defragmentation is probably the single best operation a user can perform to maintain a high-performance system. There is both a DOS version and a Windows version of DEFRAG, as well as third-party versions.
Besides fragmentation, there are a few other things that can go wrong with a hard drive. Actually, there are two FATs, identical copies, so that in case one FAT gets messed up, there is a built-in backup. Sometimes, the two get "out of sync," usually when something weird happens. An example might be while saving a file to disk, the power switch is accidentally turned off, or Windows "crashes" while swapping something to disk. Once this happens, the clusters that are marked in disagreement between the two FATs are considered "lost." These lost clusters can be recovered by running the SCANDISK utility.
Also, even though hard drives have become basically bullet-proof, with MTBF (Mean Time Between Failure) readings up in the 100,000 hr + range, occasionally clusters "go bad." These clusters must be detected and "marked" as bad, so that DOS wont try to write data to them. If data exists there already, some of it may be recoverable. SCANDISK notes any bad clusters, attempts to save any data already present, and flags the cluster as unusable. There is both a DOS version and a Windows version of SCANDISK, as well as third-party versions.
Windows versions prior to 95 were all written as 16-bit extensions to DOS. Windows 95 was written as a full 32-bit OS, but still is stuck with some 16-bit calls to some of the PCs subsystems for legacy support, mostly of existing DOS applications. The 32bit capability virtually doubled throughput to almost all I/O devices and memory, accelerating access to just about everything. Windows 95 also changed virtually everything else, from its networking capabilities to the user interface itself.
Again for compatibility purposes, Windows 95 retained the WIN.INI and SYSTEM.INI files, but shifted configuration entries to the SYSTEM.DAT and USER.DAT files. The distinction is fairly clear. The SYSTEM.DAT file resides in the WINDOWS directory, and is loaded at startup to provide specific hardware support. It is common to all users of a particular PC. In contrast, the USER.DAT file contains application and environment preferences specific to each user who logs on to the PC. These two files in conjunction are known as the Windows Registry database. Windows 95 borrows the registry concept from its "big brother," Windows NT. Whereas the SYSTEM.INI and WIN.INI files are ASCII text files that are directly editable, the Registry is a compressed database that requires special tools and skills to manipulate.
Fortunately, for the most part, a user usually doesnt need to mess with the Registry. Its entries are handled automatically both by Windows when it comes to hardware, and also by the various applications that are installed. This automaticity is labeled as "Plug and Play." If needed, Windows 95 provides a tool, REGEDIT.EXE, to access and edit the registry, but be aware that any changes made happen and are saved immediately. Any changes or deletions must be carefully considered before they are made.
Older versions of Windows used a shell application known as Program Manager to perform its menuing chore. Program Manager organizes applications into groups, with each group displayed in a window that contains a set of application icons. Each group window with individual icons can be viewed, tiled, or cascaded, or the group itself can be minimized to an icon in the Program Manager window. Program Manager stores the information about each group in a .GRP file in the Windows directory.
Major groups (Main, Accessories, StartUp, Games) are formed during installation. The Program Manager File | New menu selection allows creation of new groups. Application installation can also add new groups.
For each of its program icons, Program Manager remembers which *.EXE file should be started when double-clicking on an icon, which directory the *.EXE file resides in, and what title should appear under an icon. The individual groups to be loaded by Program Manager are listed in the PROGMAN.INI file.
Windows 95 now defaults to a "desktop" metaphor, with a hierarchical menu structure accessed from the Start button on the Task bar. Each item in the menu and on the desktop are "link" (.LNK) files stored in subdirectories under Windows. If the PC is configured for use by multiple users and set to maintain individual settings for each, then a PROFILES subdirectory is created in the WINDOWS directory and individual user settings (USER.DAT and Program subdirectories that hold .LNK files) are created for each user. These link files contain either Windows-related information about the to-be-launched application if they pertain to a Windows file, or information about "DOS Box" configuration if it is to launch a DOS application. Again, due to their small size (<2KB), link files produce enormous waste of hard drive territory due to slack.
Another change in Windows 95 was to allow for file and directory names to have more than the standard 8.3 characters. File names can now be up to 255 characters long, although this also must include the fully qualified path in the count. Also, the ability to use spaces between words is now present. This allows a user to save files with a much more descriptive filename to avoid confusion.
Another big change between Windows 3.x and Windows 95 pertains to the difference in their multitasking capabilities. There are two different types of multitasking, cooperative and preemptive. Windows 3.x supported 286 and later processors. The 286 had several limitations not present in the 386 and later. The greatest difference is that the 286 supported real and protected modes, but the 386 added virtual x86 mode. This limited the 286 considerably because it could not use virtual memory due to a limitation of the Windows architecture. This was also exacerbated because the 286 only had 24 address lines for memory, which translates to a maximum of 16MB. In contrast, the 386 and later have 32 address lines for memory, which is a maximum of 4GB.
Also, due to the virtual x86 mode not being present, all DOS applications were task switched, meaning that only one could be executing at a time. This was determined by the selected DOS application in the foreground.
On PCs with 386 and later processors, Windows 3.x cooperatively multitasks all native Windows applications and preemptively multitasks all DOS applications. The older 80286 PCs could load and run only one DOS application at a time.
Cooperative multitasking means that applications must voluntarily relinquish control of the CPU. When an application relinquishes control of the CPU, Windows then decides which application will execute next. The most common way for an application to relinquish control is by asking Windows if any messages are available.
In preemptive multitasking, control is passed from one program to another automatically by the Windows process scheduler. This is accomplished by using specific CPU hardware features of the 386/486/586 processor and the silicon chip timer to tell an application how long the application will be allowed to run. When the preset time expires, the timer interrupts whatever program is running and automatically switches back to the process scheduler. The program is temporarily suspended at this point. Windows is then free to do something else. A very short time later, Windows gives the program another slice of time, then another, and so on. This process of dividing up the time is called time slicing. The user perceives all of the running applications to be operating together, when in fact they are actually executing one at a time. Windows uses preemptive multitasking to multitask DOS sessions because DOS programs, unlike Windows programs, were not written to relinquish control of the CPU.
The collection of all the running Windows programs is treated as a single task by the Windows preemptive scheduler. When the collection of Windows programs receives a time slice, the applications multitask cooperatively during that slice of time. This type of "round-robin" scheduling worked pretty well, as long as the running Windows applications were written properly and "behaved well," meaning that they would relinquish control correctly. Errant applications, however, would cause Windows to lock-up, often forcing a reboot.
For this reason, Windows 95 followed Windows NT into the realm of complete preemptive multitasking. The new task-management structure that allows Windows to take control when an application fails to "behave" makes Windows 95 much more stable. Poorly written applications sometimes still misbehave, but now a user can force Windows to take control by pressing the Ctrl-Alt-Del key combination, which brings up the running Task List and allows the user to shut down specific running tasks. Errant applications show up in the Task List with a "(not responding)" entry next to the applications name.
While Windows 3.x was fully network-aware, it was geared around a single user. In order to support multiple users on a single PC with independent settings, access to a network server and custom installation procedures were required. Windows 95 allows multiple users to retain their own custom settings on the local PC, much like Windows NT. By default, Windows 95 installs for all users to use the same settings. Accessing the Passwords option in the Control Panel lets an administrator configure the PC for multiple user access.
Windows 95 also added a new feature, System Policies, to allow a network administrator to more easily configure Windows clients on a network. The System Policy editor, POLEDIT.EXE, while not installed with Windows, is located on the Windows 95 CD-ROM. It is used to set common-denominator defaults for all network users, and add certain restrictions on a global basis if deemed necessary.
A new VFAT (VirtualFAT) driver is incorporated in Windows 95, and provides a 32-bit protected-mode code path for manipulating the file system stored on a disk. It is multithreaded, providing smoother multitasking performance. The 32-bit file access driver is improved over that provided originally with Windows for Workgroups 3.11 and is compatible with more DOS device drivers and hard drive controllers.
Benefits of the 32-bit file access driver over previous DOSbased driver solutions include:
![]() | Dramatically improved performance and real-mode disk caching |
![]() | No conventional memory used (replacement for real-mode SmartDrive) |
![]() | Better multitasking when accessing information on disk |
![]() | Dynamic cache support |
Using DOS and Windows 3.x, manipulation of the FAT and writing to or reading from the disk is handled by the Int 21h MS-DOS function and is 16-bit real-mode code. Being able to manipulate the disk file system from protected-mode removes or reduces the need to transition to real mode in order to write information to the disk through MS-DOS, resulting in a large performance gain while accessing files.
FAT32, new to Windows 95 (beginning with the OEM 2 release), provides several enhancements over previous implementations of the FAT file system. While the current FAT supports drives up to 2GB, FAT32 can now address drives up to 2TB (terabytes) in size. FAT32 uses smaller clusters (that is, 4K clusters for drives up to 8GB in size), resulting in 10- to 15-percent more efficient use of disk space relative to large standard FAT (FAT16) drives. FAT32 has the ability to relocate the root directory and use the backup copy of the FAT instead of the default copy. In addition, the boot record on FAT32 drives has been expanded to include a backup of critical data structures. This means that FAT32 drives are less susceptible to a single point of failure than existing FAT16 volumes.
FAT32 was implemented with as little change as possible to Windows 95s existing architecture, internal data structures, Application Programming Interfaces (APIs), and on-disk format. However, in some cases, existing APIs will not work on FAT32 drives. Most programs are unaffected by these changes, and existing tools and drivers should continue to work on FAT32 drives.
All of Microsofts bundled disk tools (Format, FDISK, Defrag, and MS-DOS-based and Windows-based ScanDisk) have been revised to work with FAT32. In addition, Microsoft is working with leading device driver and disk tool vendors to support them in revising their products to support FAT32.
For most users, FAT32 has a negligible performance impact. Some programs may see a slight performance gain from FAT32. In other programs, particularly those heavily dependent on large sequential read or write operations, FAT32 may result in a modest performance degradation.
There are several available OSs for the Intel-based PC. Of the various choices, the marketplace has been fairly dominated by Microsofts DOS and Windows in various versions. While complex, this OS combination is now well-understood, reasonably stable, and well-documented. Taking the time to learn the various available commands, their syntax, and correct usage can turn an otherwise daunting occupation into an easy and sometimes fun endeavor. By understanding the inner workings of the various components, you, too, can be the expert.
![]() | At the time DOS was first developed, floppy disks and cassette tapes were the only storage media available for the PC. |
![]() | DOS commands can be either internal or external. Internal commands are present in the COMMAND.COM shell, and can be executed on the command line at any time from any directory location. External commands are those that reside in the DOS directory. |
![]() | CP/M, QDOS, NDOS, Dr. DOS, Geoworks, NEXT,among others, are all examples of competent PC operating systems. However, the PC market belongs primarily to Microsoft, with DOS, Windows, and Windows NT. |
![]() | DOS was developed as a 16-bit application, with its internal code segments written to be accessed 16-bits at a time. |
![]() | Windows (up to Windows 95) is a graphical shell that overlays on DOS, and relies on DOS for its low-level device access. |
![]() | Because of their small size (<2KB), .LNK files produce enormous waste of hard drive territory due to slack. |
![]() | System files are responsible for all device access and user interface presentation. |
![]() | As DOS has evolved over the years, it has grown to support a multitude of components, such as sound devices, scanners, CD-ROMs, tape drives, and so forth. |
![]() | On the Windows desktop, the spacing of the icons, the type of wall paper, screen colors, and other interface features can be customized in the Windows WIN.INI file. |
![]() | The bulk of configuration for Windows 95 has shifted to the SYSTEM.DAT and USER.DAT files. |
![]() | DOS installs onto a formatted disk by placing its system files in the root directory. |
![]() | By default, the DOS system files AUTOEXEC.BAT, CONFIG.SYS, and ANSI.SYS carry no attributes, and are not required for OS startup. |
![]() | The *.LGO file that WIN.COM selects is based on the type of video card setup determined was in the computer during installation. |
![]() | Review Table 9-1 for a DOS command reference. |
![]() | In Windows 3.x, PROGMAN.EXE finishes the Windows initialization by running any startup applications specified in the WIN.INI file, and by starting any applications whose icons are stored in the StartUp Group. |
![]() | The Windows 95 Registry is a complex database of settings pertaining to both applications and hardware. |
![]() | DOS is a widely used acronym for Disk Operating System. |
![]() | Using the FDISK (for Fixed DISK) utility provided with DOS, a hard drive is prepared for use through the act of partitioning and formatting. |
![]() | Review Table 9-2 for command line switches for SCANDISK.EXE. |
![]() | .COM, .EXE, and .BAT files are known as executables. |
![]() | Cooperative multitasking means that applications must voluntarily relinquish control of the CPU so that other applications may operate. |
![]() | Hard drives consist of a number of platters rotating on a common spindle, with read/write heads mounted on a pivoting assembly that spans all platters concurrently. |
![]() | Up until recently, DOS wouldnt support a partition greater than 2GB. |
![]() | Because the smallest addressable unit is the cluster, once a file is written to a cluster, the cluster is marked as used and cant have anything else written to it until the file in it is erased or overwritten by another copy of the same file. |
![]() | When formatting, or afterwards if desired, the partition marked as active must have an operating system installed. |
![]() | In practice, its best to limit the fully qualified path to less than 60 characters, else some older applications and utilities can display anomalous behavior. |
![]() | Each file accessible to DOS, regardless of media type, has four attributes that may be associated with it: READ ONLY, HIDDEN, SYSTEM, and ARCHIVE. |
![]() | The /S switch allows a user to reset the SYSTEM attribute only on those files that were previously set on by the user. |
![]() | The best way to keep your data safe from any failure of the equipment or potential operator error is to create a separate copy of the data, also known as a backup. |
![]() | Performed periodically, defragmentation is probably the single best operation a user can perform to maintain a high-performance system. |
![]() | Windows Registry entries are handled automatically both by Windows when it comes to hardware, and also by the various applications that are installed. This automation is labeled as "Plug and Play." |
![]() | In Windows 95, each item in the menu and on the desktop are "link" (.LNK) files stored in subdirectories under Windows. |
![]() | Accessing the Passwords option in the Control Panel lets an administrator configure the PC for multiple user access. |