#1-TuffTEST-Pro

Professional Level PC Diagnostic Software

USER INSTRUCTIONS
03251
#1-PC Diagnostics Company
The ESD Division of Windsor Technologies, Inc.
25 Bellam Blvd, Suite 250, San Rafael, CA 94901
phn (415) 456-2200 ~ fax (415) 456-2244
http://www.tufftest.com

T A B L E O F C O N T E N T S

I. Introduction
II. Systems Supported
III. #1-TuffTEST-Pro -- Components and Getting Started
IV. User Handbook
V. Making Program Diskettes
VI. Making TEST Diskettes
VII. TEST Plugs - Make or Buy
VIII. Product Overview
A. The Configuration Menu Summary <F2>
B. The Certification Menu Summary <F3>
C. The Error Logging Menu Summary <F4>
D. The Diagnostic Menu Summary <F5>
E. The Help Menu Summary <F10>
IX. Loading #1-TuffTEST-Pro
X. Troubleshooting Loading Failures
XI. Navigating within #1-TuffTEST-Pro
XII. Printing from Displays
XIII. Exiting #1-TuffTEST-Pro
XIV. Protecting Against Viruses
XV. Technical Support
XVI. Upgrade Information
XVII. Documentation Organization
XVIII. Using #1-TuffTEST-Pro
A. Configuration Menu <Main Menu F2>
1. Current Configuration
2. Switch Positions
3. System Memory Size
4. Extended Memory Size
5. Expanded Memory Size
6. CMOS Configuration
a. ISA/EISA/VLB/PCI Computer Systems
b. Micro Channel Architecture (MCA) Computer Systems
B. Certification Menu <Main Menu F3>
1. Abbreviated System Test
2. Extensive System Test
3. Interpreting the Certification Menu Display Information
a. System Board
b. Math Coprocessor
c. Main Memory
d. Extended Memory
e. Expanded Memory
f. Diskette Drives
g. Fixed Disk Drive
h. Serial Input/Output (I/O)
i. Parallel Input/Output (I/O)
C. Error Logging Menu <Main Menu F4>
D. Diagnostic Menu <Main Menu F5>
1. Keyboard Menu <Diagnostic Menu F2>
a. PC-Style Keyboard (10 Function Keys)
b. Enhanced Keyboard (12 Function Keys)
c. AT-Style Keyboard (10 Function Keys)
d. 102-Key Enhanced Keyboard (European Distribution
e. NCR PC-8 Keyboard (30 Function Keys)
2. Display Menu <Diagnostic Menu F3>
a. Alignment Aids
i. Cross-Hatch Pattern
ii. Dot Patterns
iii. Vertical Bars
iv. Horizontal Bars
v. Text Color Chart
b. Video Adapter Tests
i. Verify Screen Memory Test
ii. Character Set
iii. Video Attributes
iv. Fill Display with a Character
v. Test Cursor Addressing
vi. EGA/MCGA/VGA Submenu
a. Test Video Memory
b. Display High-Resolution Color Charts
c. Test Text Resolution
3. Parallel Menu <Diagnostic Menu F4>
a. Overview
b. Parallel Interface Architecture
i. Standard Pin Values for IBM/Epson DB-25 Parallel Connector
c. Parallel Interface Testing
i. Interface (Signal Loopback) Test
ii. Send the ASCII Character Set to a Printer
iii. Echo the Keys Pressed to a Printer
iv. Monitor Handshake and Data Signals While Printing
v. Monitor Status Signals from a Printer
vi. Display Error Log
d. Parallel Interface Error Codes
i. Data Test
ii. Control Test
iii. Interface Status Test
e. Parallel Interface Error Bit Chart
4. Storage Menu <Diagnostic Menu F5>
a. Diskette Drive Test Menu
b. Diskette Drive Tests and Functions
i. Specify Diskette Drive to Test
ii. Rotational Timing Test
iii. Seek/Read Test
iv. Write-Protect Verification Test
v. Read-Only
vi. Format Test/Make TEST Diskette
vii. Write/Read/Verify Test
viii. Select Testing Range
ix. Display the Error Log
c. Diskette Drive Error Codes
d. Fixed Disk Drive Subsystem
i. Overview
ii. Supported Fixed Disk Drive Interfaces/Types
a. ST506/412
b. Enhanced Small Device Interface (ESDI)
c. Integrated Drive Electronics (IDE & EIDE)
d. Small Computer Systems Interface (SCSI)
iii. Hard-Card-Type Fixed Disk Drives
iv. Fixed Disk Drive Tests and Functions
a. Select Drive to be Tested (ST506/412, ESDI, IDE, EIDE,
SCSI)
b. Select Host Adapter/Drive to be Tested (SCSI)
c. Controller Test (ST506/412, ESDI)
d. Controller Test (IDE, EIDE)
e. Controller Test (SCSI)
f. Seek (Hysteresis) Test
g. Find Previously Marked Bad Tracks (ST506/412, ESDI)
h. Surface Analysis (Read and Verify) Test
i. Find Optimum Interleave (ST506/412, ESDI)
j. Low-Level Format - CONDITIONAL (Preserve Bad
Tracks) (ST506/412, ESDI, IDE)
k. Low-Level Format - UNCONDITIONAL (ST506/412,
ESDI, IDE)
l. Low-Level Format (SCSI)
m. Media Analysis (Write/Read) Test (ST506/412, ESDI)
n. Media Analysis & Repair (Write/Read/Verify/Re-Assign
Bad Blocks) Test (IDE, EIDE, SCSI)
o. Edit and/or Mark Bad Tracks (ST506/412, ESDI)
p. Re-Assign Blocks (SCSI)
q. Select Testing Range
r. Display Error Log
s. Park the Disk Heads (ST506/412, ESDI)
t. Host Adapter ROM-Based Utilities (SCSI)
e. SCSI Tape Drive Tests
i. Select Drive to be Tested
ii. Combination Test
iii. Erase Test
iv. Retension Test
v. Write Filemarks Test
vi. Start/Stop/Write Test
vii. Short-Streaming Write Test
viii. Long-Streaming Write Test
ix. Display Error Log
f. Fixed Disk Drive and Tape Drive Error Codes
i. Fixed Disk Drives Only
ii. SCSI (Fixed Disk and Tape Drives)
a. ADAPTEC/NCR
b. Western Digital 7000
5. Serial Interface Menu <Diagnostic Menu F6>
a. Overview
b. Serial Interface Architecture
c. Serial Interface Testing
i. Select I/O Port Addresses
ii. Internal Operations Test
iii. External Operation Test
iv. Display Error Log
d. Serial Interface Error Codes
i. Transmitter Test or Receiver Test
ii. Data Test
iii. External/Internal Control Line Tests
e. Serial Interface Error Bit Chart
6. Memory Menu <Diagnostic Menu F7>
a. Overview
b. Common Memory Tests
i. Abbreviated Memory Test
ii. Extensive Memory Test
iii. ALLZEROS Memory Test
iv. ALLONES Memory Test
v. CHECKERBOARD Memory Test
vi. ADDRESS Memory Test
vii. MARCHING ONES Memory Test
viii. WALKING ONES Memory Test
ix. Select Memory Testing Range
a. Default Range (Main Memory)
b. Alternate Range (Main Memory)
c. Test Range (Extended Memory)
x. Display Error Log
c. Testing Variations in the Extended Memory Menu
i. Test Compaq Reserved Memory
d. Testing Variations in the Expanded Memory Menu
i. Test Controller
ii. Select Base Memory Segment
iii. Select Expanded Memory Board
e. Disable Cache Memory
i. First-Level Cache
ii. Second-Level Cache
f. Finding Failing Memory Modules
g. Finding Bit 0, Bank 0, and Parity
i. Memory Module Orientation - 8-Bit Systems
ii. Memory Module Orientation -16-, 32- and 64-Bit Systems
iii. "Bit Failure Packet" Decoding Examples
h. Memory Error Codes
i. Main Memory -8088-and 8086-based Computers
ii. Main Memory -80286-based Computers (ISA)
iii. Main Memory -80386- and i486-based Computers
(ISA/EISA/VLB/PCI)
iv. Main Memory -Pentium through Pentium 4-based Computers
(ISA/EISA/VLB/PCI)
v. Main Memory -IBM PS/2 Models -General
vi. Main Memory -IBM PS/2 Models 25 and 30
vii. Main Memory -IBM PS/2 Models 30-286, 50, 50Z, and 60
viii. Main Memory -IBM PS/2 Models 55SX, 70, 80, and 90
ix. Extended Memory -80286-based Computers (ISA/MCA)
x. Extended Memory -80386- and i486-based Computers
(ISA/EISA/MCA/VLB/PCI)
xi. Extended Memory -Pentium through Pentium 4-based
Computers (ISA/EISA/VLB/PCI)
xii. Expanded Memory -IBM PC, XT, PS/2 25 and 30
xiii. Expanded Memory -80286-, 80386-and i486-based Computers
(ISA/EISA)
i. Memory Bit Position Failure Chart
E. Help Menu <Any Menu F10>
XVIV. Licensing Information

I. INTRODUCTION

#1-PC Diagnostics Company, the Electronic Software Distribution (ESD) division of Windsor
Technologies, Inc., welcomes you to the growing family of personal computer service
companies, manufacturers, and support personnel who use #1-TuffTEST-Pro Professional
Level PC Diagnostic Software as their primary and preferred troubleshooting and diagnostic
tool.
#1-TuffTEST-Pro is the downloadable version of PC-Technician. It was developed by
Windsor Technologies, a leader in personal computer diagnostic products since 1984, and with
the assistance of their customers, whose business is the support and/or service of personal
computers. #1-TuffTEST-Pro is specifically designed to assist professional field service
engineers in maintaining, troubleshooting, and repairing personal computers.
#1-TuffTEST-Pro contains extensive tests for main, extended, and expanded memory; fixed
disk drives; diskette drives; video displays and adapters; serial and parallel interfaces; and
keyboards; and is able to log errors both to the display and to a printer. #1-TuffTEST-Pro has
an extremely friendly, function-key-driven, user interface.
#1-TuffTEST-Pro uses a proprietary loader/operating system so that software routines are as
close to the hardware as possible and do not have results of various tests masked by operating
system (i.e., Windows, DOS, Unix, OS/2, etc.) error recovery routines. All the diagnostic
routines in #1-TuffTEST-Pro are written in assembly language (instead of a high-level language
such as C), so they execute more efficiently, run faster and are smaller in size. In addition, the
#1-TuffTEST-Pro program is able to relocate when executing memory tests; something that is
NOT POSSIBLE UNDER DOS or other operating systems.
#1-TuffTEST-Pro is a high-precision tool dealing with complex personal computer hardware
and configuration issues. Therefore, it is extremely important that you read this User Handbook
prior to using #1-TuffTEST-Pro.

II. Systems Supported

#1-TuffTEST-Pro is professional-level diagnostic software that supports personal computers
that use Intel Pentium 4, Pentium III, Pentium II, Pentium Pro, Pentium, i486, 80386, 80286,
8088, 8086, and compatible AMD and Cyrix processors with Industry Standard Architecture
(ISA), Enhanced Industry Standard Architecture (EISA), Micro Channel Architecture (MCA),
VESA Local Bus (VLB), and Peripheral Component Inter-connect (PCI) bus configurations.

III. #1-TuffTEST-Pro -- Components and Getting Started

User Handbook: First print AND READ.
Program Diskettes: Make each format (3" and/or 5") as required.
TEST Diskettes: Make as required for diskette drive testing.
TEST Plugs: Required for parallel and serial interface external tests.
Make your own or buy from #1-PC Diagnostics Company.
To order from #1-PC Diagnostics Company, go to:
http://www.tufftest.com/ttp01.htm#TEST_Plugs

IV. User Handbook

The #1-TuffTEST-Pro User Handbook is established as an icon after running the downloaded
program, ttpsetup.exe, under Windows. It is in the Windows Write format and was compiled
with the standard Windows format defaults (font style and sized for 8" x 11" paper).
As #1-TuffTEST-Pro is a "Stand-Alone" tool, the User Handbook is not, by definition,
available for use in an on-line help mode. Therefore, it is best to print the entire document prior
to using
#1-TuffTEST-Pro, so that the User Handbook is available to use as a reference guide while
troubleshooting a PC.
To better understand the operation, tests, and functions of #1-TuffTEST-Pro, it is best to do a
THOROUGH review of this User Handbook prior to using #1-TuffTEST-Pro. This will also
insure that you will understand the User Handbook layout and contents for quick reference while
involved in a #1-TuffTEST-Pro test session.

V. Making Program Diskettes

To use #1-TuffTEST-Pro, you first have to make a Program Diskette, as #1-TuffTEST-Pro is
ALWAYS loaded from Drive A from a power-on start (cold boot).
To make Program Diskettes, click on the Windows Start button, select the Make #1-TuffTEST-
Pro Diskette icon from the #1-TuffTEST-Pro program group, then follow the easy-to-use
instructions.After making a Program Diskette, remember to label it, and write-protect it as
another safegard in protecting against viruses. See Section XIV, Protecting Against Viruses.
#1-TuffTEST-Pro Program Diskettes contain their own boot loader module, proprietary
operating system (NOT Windows, DOS, Unix, OS/2, etc.) and multiple programs designed
specifically for the various CPU/bus configurations that exist in the systems supported (see
Section II above).
NOTE: To be able to use #1-TuffTEST-Pro on the full range of PC configurations, it is best to
make Program Diskettes in the most common format, i.e. the lowest capacity media type. While
their presence is diminishing, there are still PC's that have "single mode" standard (low) capacity
Drive A diskette drives that only can read 360KB DS/DD 5" diskettes or 720KB DS/DD 3"
diskettes. The standard (low) capacity Program Diskettes can also be used in the "dual-mode"
diskette drives that primarily use 1.2MB 5" diskettes or 1.44MB 3" diskettes.
NOTE: IT IS ILLEGAL TO HAVE MORE THAN ONE OPERATIONAL COPY OF #1-
TuffTEST-Pro AT ANY TIME.

VI. Making TEST Diskettes

#1-TuffTEST-Pro TEST Diskettes are required to perform the Diskette Drive Tests because
TEST Diskettes are formatted with a worst-case data pattern, B6D9h (h = hexadecimal),
designed to put maximum stress on the target diskette drive during testing.
To make TEST Diskettes:
1) Load #1-TuffTEST-Pro and then remove the Program Diskette.
2) Select [F5] Diagnostic Menu.
3) Select [F5] Storage Menu.
4) Select a) Diskette Drive Tests Menu.
5) Insert appropriate blank diskette (3" 1.44MB/720KB or 5" 1.2MB/360KB).
6) Specify appropriate diskette drive.
7) Press [Enter] to continue.
8) Specify appropriate format.
9) Press [Enter] and special formatting will commence.
NOTE: Only diskettes that format without error on both sides should be used as TEST
Diskettes.
An added benefit of #1-TuffTEST-Pro's maximum stress format pattern, B6D9h (h =
hexadecimal), is that it can also be used as a high-precision media certification test. Any errors
occurring during the formatting process (on a known good diskette drive), would indicate that
the diskette being tested has a marginal recording capability due to imperfections and/or wear on
the diskette surface.

VII. TEST Plugs: Make or Buy

#1-TuffTEST-Pro has very powerful tests for the external operation of the parallel and serial
interface ports. To use these tests, TEST Plugs (external signal loopback connectors) must be
used. TEST Plugs receive an outbound signal from the interface connector and loop it back
inbound to the appropriate interface connector pins, thus testing that external function of the
interface.
There are three TEST Plugs used by #1-TuffTEST-Pro:
1 DB-25 Serial (asynchronous) TEST Plug
1 DB-9 Serial (asynchronous) TEST Plug
1 DB-25 Parallel TEST Plug
TEST Plugs can be made by the #1-TuffTEST-Pro licensee or they can be purchased directly
from #1-PC Diagnostics Company (online from our Web site as indicated below).
#1-TuffTEST-Pro is designed to only use TEST Plugs with the signal loopback pin-outs
indicated below. Other loopback connectors (also called wrap-plugs) available in the
marketplace MAY NOT work correctly with #1-TuffTEST-Pro.
The #1-TuffTEST-Pro TEST Plug pin-outs are as follows:
Serial Port - DB 25 subminiature (female) connector:
PIN to PIN
2 to 3
4 to 5
4 to 6
8 to 20
9 to 18
11 to 25
20 to 22
Note that pin #4 and pin #20 are double-wire pins.
Serial Port - DB 9 subminiature (female) connector:
PIN to PIN
1 to 4
2 to 3
4 to 9
6 to 7
7 to 8
Note that pin #4 and pin #7 are double-wire pins.
Parallel Port - DB 25 subminiature (male) connector:
PIN to PIN
5 to 15
6 to 13
7 to 12
8 to 10
9 to 11
As a #1-TuffTEST-Pro licensee, you are authorized to make your own TEST Plugs using the
above pin designations and parts that are readily available from most electronic parts suppliers
and catalogs.
If you do not want to make your own TEST Plugs, you can buy TEST Plugs online from our web
site. They are sold ONLY in complete sets of three, one of each type. Price, shipping and
ordering information is available at:
http://www.tufftest.com/ttp01.htm#TEST_Plugs

VIII. Product Overview

#1-TuffTEST-Pro's Main Menu contains five submenus. They are discussed briefly in this
section and in detail in the sections specifically addressing each menu item.

VIII.A. The Configuration Menu Summary <F2>:

Options within this menu report the system's hardware configuration, verify the size of main,
extended, and expanded memory, and list information about the installed Read Only Memory
Basic Input/Output Services (ROM BIOS).
Additionally, where applicable, #1-TuffTEST-Pro provides the capability to view and/or edit
CMOS-based configuration information (where applicable) and view system board configuration
DIP switches.

VIII.B. The Certification Menu Summary <F3>:

Selecting this option performs the following tests: System Board; Math Coprocessor; Main,
Extended, and Expanded Memory; Diskette and Fixed Disk Drives; and Serial and Parallel
I/O. Any failing device reported by the Certification tests should be thoroughly tested with
options from the Diagnostic Menu.

VIII.C. The Error Logging Menu Summary <F4>:

Options within this menu allow the Error Log to be displayed or cleared. It also indicates if the
Error Log is full.
NOTE: The Error Log can only be cleared from this menu.

VIII.D. The Diagnostic Menu Summary <F5>:

Options within this menu perform extensive tests of main, extended, and expanded memory;
diskette drives, fixed disk drives, and tape drives; keyboard; video adapter and display unit; and
serial and parallel adapters.

VIII.E. The Help Menu Summary <F10>:

The purpose of the Help Menu is to explain how to choose the various menus and how to select
the various options within each menu.

IX. Loading #1-TuffTEST-Pro

The #1-TuffTEST-Pro Program Diskette contains a proprietary loader/operating system, and
therefore cannot be loaded from within an operating system environment, i.e., Windows, DOS,
Unix, OS/2, etc. #1-TuffTEST-Pro is written this way to provide greater reliability in loading
and operating in systems with suspected hardware malfunctions. This feature also provides
maximum protection against computer viruses.
Because #1-TuffTEST-Pro uses a proprietary loader/operating system, the following loading
process MUST be used:
1) Turn the target computer OFF.
2) Insert a write-protected Program Diskette into Drive A.
3) Turn the target computer ON.
4) Once the Main Menu appears, remove the Program Diskette.
WARNING: DO NOT use the <Ctrl> <Alt> <Del> reset to load #1-TuffTEST-Pro, and DO
NOT attempt to read the Program Diskette from an operating system prompt prior to loading
#1-TuffTEST-Pro. See Section XIV, Protecting Against Viruses.
The #1-TuffTEST-Pro Program Diskette contains multiple diagnostic programs in a compressed
format. Each program is written for a specific microprocessor/bus combination. As
#1-TuffTEST-Pro loads, the micro-processor/ bus combination of the system in use is analyzed,
and the appropriate program is expanded and loaded into main memory.
NOTE: When using #1-TuffTEST-Pro, some tests may not be available. Test availability is
determined by the microprocessor/bus combination of the system in use.

X. Troubleshooting Loading Failures

If the #1-TuffTEST-Pro Program Diskette will not load, follow these procedures:
1) Make sure the target PC's CMOS Setup is configured to boot from Drive A first. If it is, go to
number 2.
2) Remove the Program Diskette from Drive A. Turn the computer OFF, carefully reinsert the
Program Diskette, and turn the computer ON. Often the clamping mechanism does not quite
center the diskette. If unsuccessful, go to number 3.
3) Attempt to load an application program (utility, wordprocessing, spreadsheet, etc.) from a
known good diskette. If it loads, it would indicate that the Program Diskette has become
damaged. Make a new Program Diskette (see Section V, Making Program Diskettes). If the
known good diskette does not load, go to number 4.
4) The problem is probably with the diskette drive, the diskette controller board, or with the
inter-connecting cable. Substitute a known good interface cable and try loading the diskette. If
it will not load, go to number 5.
5) Substitute a known good diskette controller board and attempt to load the Program
Diskette. If it will not load, go to number 6.
6) Substitute a known good diskette drive and attempt to load the Program Diskette. If the
diskette drive is not being accessed, the problem may be with the system board or with the power
supply.

XI. Navigating within #1-TuffTEST-Pro

#1-TuffTEST-Pro is a menu-driven diagnostic system. All possible options and associated key-
strokes necessary to initiate functions appear on the video display at all times. This eliminates
the need for remembering commands or keystroke sequences.
The Main Menu graphically displays five submenus and their associated function keys. Anytime
a menu and function key are visible as an option, they are available for selection. To enter the
desired submenu, press the function key indicated on the display. When in a submenu, the <F1>
key is used to go back one level.
In addition to menu selection with function keys, there are three other selection processes used
within #1-TuffTEST-Pro:
1) Terminating a diagnostic routine is achieved by pressing the <Esc> key. Once the routine has
stopped, pressing the <ENTER> (or <Ret>) key returns to that routine's previous menu.
2) Multiple-choice menu selection is achieved by pressing the key associated with the desired
selection.
3) User-defined values may be entered where appropriate, such as specifying the memory
testing range or fixed disk drive testing range.
NOTE: Anytime four incorrect keys are used, #1-TuffTEST-Pro will cycle through the HELP
section to review how to use function keys to move from menu to menu, and how to select items
within a menu.

XII. Printing from Displays

Any #1-TuffTEST-Pro display can be printed by using the Print Screen key while no tests are in
operation.

XIII. Exiting #1-TuffTEST-Pro

To exit #1-TuffTEST-Pro, the Program Diskette and TEST Diskettes must be removed from the
diskette drives, and the power switch on the computer turned OFF.

XIV. Protecting Against Viruses

The #1-TuffTEST-Pro Program Diskette(s) will be used to troubleshoot many computers, any
one of which may contain a virus. To insure the virus-free status of the Program Diskette(s) and
to prevent the spread of viruses, please read the following information carefully.
#1-TuffTEST-Pro is highly resistant to virus infection when used according to the instructions
provided below because it does not operate under an operating system (i.e., Windows, DOS,
Unix, OS/2, etc.).
Since most viruses use the operating system environment to inflict their damage, keeping the
Program Diskette(s) from interacting with the operating system environment provides maximum
protection against viruses. NEVER do an operating system directory command (DIR) or in any
other way attempt to read a Program Diskette from within an operating system.
To protect the Program Diskette(s) from viruses:
1) Always turn the computer OFF before inserting a Program Diskette into Drive A.
2) Verify that the Program Diskette is write-protected.
3) Never use the <Ctrl> <Alt> <Del> reset to load or reload #1-TuffTEST-Pro.
4) Remove the Program Diskette once the Main Menu is displayed.
Carefully following the above guidelines will provide maximum protection against viruses
contaminating Program Diskette(s).
If virus contamination should occur, make a new Program Diskette on a virus-free computer (see
Section V, Making Program Diskettes).

XV. Technical Support

Technical support is available at no charge during normal business hours via e-mail at:
[email protected]

XVI. Upgrade Information

Once you license #1-TuffTEST-Pro you are eligible to buy upgrades for life. For your
convenience, upgrades are always totally backward compatible so you never have to worry about
using your old diskettes with your new diskettes.
"Any" older version of #1-TuffTEST-Pro can be upgraded to the current version. To upgrade,
all we need is your #1-TuffTEST-Pro Serial Number when you place your order through our
online 24-hour automated ordering system
The cost of a #1-TuffTEST-Pro upgrade varies depending upon the size and scope of the
upgrade. Minor upgrades and patches are usually available at no charge.
Information on the current version and how to order it is always available at our World Wide
Web site at:
http://www.tufftest.com/ttp01.htm#Upgrade_Information
NOTE: #1-TuffTEST-Pro licensees - to determine your version number, load
#1-TuffTEST-Pro. The complete version number is shown on every display.

XVII. Documentation Organization

In general, this User Handbook follows the process of using #1-TuffTEST-Pro and the order of
the various menus and submenus. Due to the differences in the microprocessor/bus
combinations, the sequence or number of some menu/submenu items may vary.
Primary menus and submenus are accessed via the keyboard function (F) keys with options
within those menus accessed by the keyboard's lower case alpha keys. The documentation
indicates the function key in < > following the test/function title, with the keys for options within
menus indicated in ( ).

XVIII. USING #1-TuffTEST-Pro


XVIII.A. Configuration Menu <Main Menu F2> :

The Configuration portion of #1-TuffTEST-Pro contains some or all of the following selections,
depending on the type of computer being tested:

XVIII.A.1. Current Configuration:

Page one of Current Configuration displays the PC's configuration. In addition, a CPU speed
test is performed and MMX extensions are tested. Page 2 of Current Configuration displays
BIOS information and also tests the BIOS.
CPU SPEED TEST NOTE: #1-TuffTEST-Pro determines the speed of the installed
microprocessor by using algorithms to test its internal clock. This is the actual speed of the
microprocessor (in MHz). This report should not be confused with a benchmark, as it does not
reflect the actual speed of the computer as affected by memory speed, wait states, fixed disk
drive speed, etc. Also note that due to CMOS and other system settings, the speed may vary
from the speed rating of the processor.
BIOS TEST NOTE: When a BIOS is manufactured a mathematical value is calculated for the
sum of the data it contains and placed in a register. This is called a BIOS Checksum. #1-
TuffTEST-Pro reads the BIOS contents, calculates a checksum for the data it finds, and then
does a compare to the BIOS manufacturer's checksum. The actual values of the checksums don't
matter. The only thing that matters is that the 2 checksums match.
If the two checksums don't match exactly it means that the BIOS must have dropped or gained at
least 1 bit since it was manufactured. When this happens #1-TuffTEST-Pro generates an error
message in the form of a red blinking word UNSTABLE.
If you see this error message AND the PC is experiencing symptoms, a failing BIOS chip or
CPU chip is the most likely cause of the PC's problems.
NOTE: The displayed BIOS Release Date often differs from the date displayed in the copyright
message during system initialization. The BIOS release date displayed by #1-TuffTEST-Pro is
the actual release date of the BIOS code, not the updated copyright information.

XVIII.A.2. Switch Positions:

This selection graphically displays the current system board switch settings on IBM PC, PC XT
& compatible PCs.
NOTE: The switch settings displayed are those of the IBM-branded PC and PC XT. If this
product is being used to test non-IBM-branded PC or PC XT compatible computers, the switch
settings may differ from the actual system board switch configuration.

XVIII.A.3. System Memory Size:

Main memory size (to 640K) is displayed in a bar graph format. The display contains both
memory size as reported by the ROM BIOS and memory size as detected by #1-TuffTEST-
Pro. If the two memory size reports are not the same, CMOS configuration should be checked.
NOTE: #1-TuffTEST-Pro reports the amount of main memory that is actually usable by an
operating system or an application program. Some computers reserve a small amount of main
memory (usually 1K to 2K) for internal use, making it inaccessible to operating systems or
application programs. In these instances, #1-TuffTEST-Pro will report a System Memory Size
that is smaller than expected (such as 638K or 639K, versus the expected 640K).

XVIII.A.4. Extended Memory Size:

Extended memory starts at 1024K. Its size is displayed in a bar graph format. The display
contains both memory size as reported by the ROM BIOS and memory size detected by
#1-TuffTEST-Pro. If the memory sizes reported are not the same, CMOS configuration should
be reviewed.

XVIII.A.5. Expanded Memory Size:

Expanded memory size is displayed in a bar graph format when installed on IBM AT, 386, 486
& compatible PCs. The display contains both memory size as reported by the ROM BIOS and
memory size detected by #1-TuffTEST-Pro. If the memory sizes reported are not the same,
CMOS configuration should be checked.

XVIII.A.6. CMOS Configuration:

The current status of the CMOS Configuration is displayed when this item is selected.
XVIII.A.6.a. ISA/EISA/VLB/PCI Computer Systems
On some PCs the CMOS Configuration can also be initialized or modified from this selection.
NOTE: If the CMOS Configuration is modified, changes will not become effective (except for
date and time) until the next time the computer is turned OFF and then ON again.
XVIII.A.6.b. Micro Channel Architecture (MCA) Computer Systems:
The actual configuration cannot be permanently modified using #1-TuffTEST-Pro because of
IBM's use of proprietary Adapter Definition Files (ADFs) in its setup program.
NOTE: Where #1-TuffTEST-Pro uses CMOS to identify devices (such as diskette drive B: and
fixed disk drives C: and D:), they will be displayed as "Not Available," since MCA computers do
not use CMOS in the traditional ISA/EISA fashion.

XVIII.B. Certification Menu <Main Menu F3>:

The Certification portion of #1-TuffTEST-Pro consists of a number of discrete diagnostic
routines. These single tests, when run as a group, allow the testing of an entire computer
system. TEST Diskettes and TEST Plugs must be inserted before running the certification tests.
The Certification Menu within #1-TuffTEST-Pro provides two testing options: the Abbreviated
System Test and the Extensive System Test. Prior to running the Abbreviated System Test or the
Extensive System Test, the Test Selection Menu is presented which allows individual tests to be
selected or deselected prior to executing either of these two testing options. When the asterisk
(*) is displayed next to the listed system resource, the individual test for that particular system
resource is part of the set of diagnostic routines to be performed. When the asterisk (*) is not
displayed, the diagnostic routine associated with the listed system resource is not
performed. The default setting is for all tests to be selected and performed. Choose the
appropriate letter to select or deselect a test.

XVIII.B.1. Abbreviated System Test:

This test performs the diagnostic routines once and then terminates. Its purpose is to perform a
single-pass verification of the entire system, including all installed devices.

XVIII.B.2. Extensive System Test:

The diagnostic routines performed under this option are exactly the same as the Abbreviated
System Test, except that this option adds the Marching Ones and Walking Ones memory tests
and repeats the tests continuously until the operator intervenes by pressing the <Esc> key. It
may be used to burn-in a system (running it under stress for an extended period of time to certify
operation), to detect intermittent failures, or to certify newly installed/repaired devices.

XVIII.B.3. Interpreting the Certification Menu Display Information:

The Abbreviated and Extensive System Tests are compilations of many system component
tests. These tests, along with information regarding the resulting error codes for each test
module, are described below.
XVIII.B.3.a. System Board
If the system passes, a report of "OK" will be displayed. A report of "FAILED" would indicate
that system board replacement or further chip-level troubleshooting using hardware-based
equipment might be appropriate.
XVIII.B.3.b. Math Coprocessor:
This diagnostic routine tests the math coprocessor by performing write/read/compare operations
to its internal registers. Its status is reported as "OK," "FAILED," or "NOT INSTALLED." If a
math coprocessor other than an Intel-branded math coprocessor is installed (i.e., Weitek), it will
not be detected.
XVIII.B.3.c. Main Memory:
The installed memory is subjected to four types of write/read/compare tests: ALLZEROS,
ALLONES, CHECKERBOARD, and ADDRESS. In the ALLZEROS test, zeros are written to
all memory locations followed by a read/compare operation. Any memory location not
containing a zero is reported as "FAILED," and an error code is displayed. The same procedures
are followed in the ALLONES test, except that ones are written to all memory locations. The
CHECKERBOARD test alternately writes zeros and ones to successive memory locations and
verifies through read/compare operations that the memory location contains the correct value. If
a discrepancy is detected, an error code is displayed.
The first three memory tests (ALLZEROS, ALLONES and CHECKERBOARD) generate data
bytes having an even number of ones, and therefore do not provide maximum stress testing of
the parity generators. The fourth memory test, ADDRESS, generates data having approximately
the same number of bytes with an even number of ones as those with an odd number of ones,
imposing greater stress conditions on the parity generators. If errors occur, the main memory
should be further tested using the routines available from the Diagnostic Menu <MM F5>.
NOTE: The Extensive System Test adds the more rigorous Marching Ones and Walking Ones
memory tests.
XVIII.B.3.d. Extended Memory:
#1-TuffTEST-Pro tests extended memory using the same procedures as for main memory. If
errors occur, the extended memory should be further tested using the routines available from the
Diagnostic Menu <MM F5>.
XVIII.B.3.e. Expanded Memory:
#1-TuffTEST-Pro tests expanded memory using the same procedures as for main memory. If
errors occur, the expanded memory should be further tested using the routines available from the
Diagnostic Menu <MM F5>.
XVIII.B.3.f. Diskette Drives:
This test is comprised of two parts: The Start test and the Write/Read test. In the Start test, the
diskette drive is accessed to verify that the diskette drive controller is functioning properly. The
Write/Read test writes a maximum stress test pattern, B6D9h (h = hexadecimal), and verifies
through a read/compare operation that the test pattern has not changed. Any discrepancies are
reported as "FAILED," and an error code is reported. If errors occur, the diskette drive should be
further tested using the routines available from the Diagnostic Menu <MM F5>.
If a second diskette drive is installed, it is also tested.
XVIII.B.3.g. Fixed Disk Drive:
Testing of the fixed disk drive in this portion of #1-TuffTEST-Pro is limited to non-destructive
tests. If errors occur, the fixed disk drive should be further tested using the routines available
from the Diagnostic Menu <MM F5>.
XVIII.B.3.h. Serial Input/Output (I/O):
Both the internal operation and interface of the serial (asynchronous) I/O adapter(s) are tested.
In the internal test, the Universal Asynchronous Receiver Transmitter (UART) or Universal
Synchronous/Asynchronous Receiver Transmitter (USART) is switched into the test mode,
enabling #1-TuffTEST-Pro to test its internal registers without transmitting data. While in this
mode, data is written in parallel format to the UART/USART which converts it to serial data,
internally loops it back within the UART/USART, then reconverts it to parallel data. The
reconverted data is then read and compared by #1-TuffTEST-Pro. If a discrepancy is detected,
an error code is displayed indicating that the Data Test failed and the failing bit(s) are displayed.
The interface test requires the DB-25 Serial TEST Plug (the yellow plug if purchased from #1-
PC Diagnostics Company) and/or the DB-9 Serial TEST Plug (the green plug if purchased from
#1-PC Diagnostics Company) to be inserted into the DB-25 and/or DB-9 connector on the back
of the serial adapter. During this test, #1-TuffTEST-Pro causes the UART/USART to send data
out via the interface connector and receive it back after it is looped through the TEST Plug. It
also sends and receives the handshake signals. #1-TuffTEST-Pro then reads the Modem Status
Register in the UART/USART and displays an error code if the expected values of the status bits
are not correct.
If errors occur, the serial interface should be further tested using the routines available from the
Diagnostic Menu <MM F5> followed by Serial <DM F6>.
XVIII.B.3.i. Parallel Input/Output (I/O):
The test routines for the parallel (I/O) adapter card(s) contain two parts: internal testing and
interface testing.
The internal testing involves writing, reading, and comparing data and control signals through
the
circuitry. If the data and control signals match, an OK status is reported. If they do not match, a
failure is reported along with an error code.
The interface testing requires the DB-25 Parallel TEST Plug (the blue plug if purchased from
#1-PC Diagnostics Company) to be inserted into the DB-25 connector on the back of the parallel
adapter. The test involves feeding back five of the eight parallel data lines through the five status
reporting pins, emulating the status signals that would normally originate from the printer. The
data is sent out and received back with the Data, Control, and Status words indicated as
"PASSED" or "FAILED." For each failing word, an error code is displayed indicating the
failing bit(s).
If errors occur, the parallel interface should be further tested using the routines available from the
Diagnostic Menu <MM F5> followed by Parallel <DM F4>.

XVIII.C. Error Logging Menu <Main Menu F4>:

The Error Logging facility can maintain up to 180 errors in four pages of 45 errors each. The
Error Log format of "Errors -Code" indicates the total number of errors for each separate and
unique error code. Consult the individual tests below for the error codes generated.
The Error Logging feature of #1-TuffTEST-Pro is automatically enabled when #1-TuffTEST-
Pro is loaded. When the Error Logging Menu option is selected, the following choices are
displayed for selection:
(a) Display Log: Displays the currently logged errors.
(b) Clear Error Log: Deletes all currently logged errors.
NOTE: Although the Error Log can be viewed from numerous menus, clearing the log can only
be done from the Error Logging Menu.
If there are enough errors to fill the log, the lower right corner of the current screen will display a
blinking "Error Log is Full" message.

XVIII.D. Diagnostic Menu <Main Menu F5>:

The Diagnostic Menu selections provide the majority of #1-TuffTEST-Pro's power and
precision. Each major system component/component group has a separate menu selection,
extensive tests, and appropriate displays which indicate the activity and errors found. The
precision and stressfullness of the tests represent the maximum amount possible using a RAM
resident program. Each is engineered to be easy to learn and use. It is strongly suggested that
this section's documentation be thoroughly reviewed as part of using #1-TuffTEST-Pro.

XVIII.D.1. Keyboard Menu <Diagnostic Menu F2>:

The Keyboard Test verifies the functionality of every key position on the keyboard, as well as
multiple-key combinations. For each key or keys pressed, the ASCII hex code and the keyboard
scan code are displayed along with a graphic representation of the key(s) pressed. The ASCII
hex code and keyboard scan code displayed should be compared with the standards established
by IBM.
NOTE: Keys that have specific system functions will still perform those functions.
When the <PrtSc> (print screen) key is pressed, the BIOS printing routine will attempt to send
the image on the video display to a printer at LPT1. If a printer is not attached, or the attached
printer is offline or otherwise disabled, the system will hang for an indeterminate period of
time. To bypass the printing function, hold down the <Ctrl> key while pressing the <PrtSc> key.
When the <Pause> key is pressed, the BIOS pause routine will take effect. The system will not
continue until the <Pause> key is pressed again. To bypass the pause function, hold down the
<Ctrl> key while pressing the <Pause> key.
Upon selection of the Keyboard Test, some of the following selections will be listed, which may
vary, depending on the target computer:
XVIII.D.1.a. PC-Style Keyboard (10 Function Keys):
This item should be selected if the keyboard has ten function keys along the left or across the
top.
XVIII.D.1.b. Enhanced Keyboard (12 Function Keys):
This item should be selected if the keyboard has twelve function keys across the top.
XVIII.D.1.c. AT-Style Keyboard (10 Function Keys):
This item should be selected if the keyboard has ten function keys along the left, or across the
top.
XVIII.D.1.d. 102-Key Enhanced Keyboard (European Distribution):
This item should be selected if the keyboard has 102 keys and supports the European character
set.
XVIII.D.1.e. NCR PC-8 Keyboard (30 Function Keys):
This item should be selected if the system being tested is an NCR PC-8.

XVIII.D.2. Display Menu <Diagnostic Menu F3>:

#1-TuffTEST-Pro's display tests are designed to visually verify correct operation of the display
via the alignment tests, and to test the operation and components of the video adapter and
memory via the video adapter tests.
NOTE: When the computer contains a Black on White VGA display, the menu headers will not
be visible. This can be corrected by pressing the <V> key at the Main Menu.
XVIII.D.2.a. Alignment Aids:
The alignment aids verify that the alignment circuitry is functioning properly. Convergence and
other alignment issues can be determined by observing changes in the various patterns in the
corners versus the center of the display. Any irregularities in the appearance of the display
indicate a need for servicing the video display unit. The following alignment patterns are
available:
XVIII.D.2.a.i. Cross-Hatch Pattern
XVIII.D.2.a.ii. Dot Patterns
XVIII.D.2.a.iii. Vertical Bars
XVIII.D.2.a.iv. Horizontal Bars
XVIII.D.2.a.v. Text Color Chart:
This pattern verifies proper focusing and operation of the red, green, and blue color guns.
Inconsistencies in color quality and the displayed color patterns indicate a need for service.
XVIII.D.2.b. Video Adapter Tests:
XVIII.D.2.b.i. Verify Screen Memory Test:
This test writes every ASCII character with all possible attributes to each screen location. A
successful screen memory test displays a sequential series of Xs. If any character other than an
X is displayed at any character location on the video display (including a space), a screen
memory failure is indicated.
Screen memory for monochrome displays consists of 4K of screen memory (one page). Screen
memory for color displays consists of 16K of screen memory (four pages). The screen memory
tests on systems using monochrome displays scan the screen one time, displaying the test results
of the one page of screen memory. The screen memory test with systems having color displays
scans the screen four times, displaying the test results of the four pages of memory.
XVIII.D.2.b.ii. Character Set:
This test verifies the proper operation of the video character generator ROM. The entire standard
and extended ASCII character sets are displayed. This display should be carefully compared to
the established standard. Incorrect or missing characters indicate a problem with the character
generator and/or the display adapter.
XVIII.D.2.b.iii. Video Attributes:
All possible combinations of video attributes are displayed on the screen: blinking, underlined,
reverse, normal and high intensity, and blank. The display should be carefully checked to verify
the attribute specified is correct. If it is not correct, a faulty video display adapter is indicated.
XVIII.D.2.b.iv. Fill Display with a Character:
This test fills the display with a character selected by the user. Any incorrectly displayed
character or a blank indicates a failure of screen address generation, screen memory, or the video
character generator ROM. In any of these cases, a failure indicates a problem with the video
display adapter.
XVIII.D.2.b.v. Test Cursor Addressing:
This test successively places the cursor in different screen locations, giving the appearance of
smaller and smaller rectangles on the screen. The display should be carefully monitored by the
user for breaks in the rectangles. Any break indicates a problem with cursor addressing and/or
screen memory, both of which may be caused by a problem with the video display adapter.
XVIII.D.2.b.vi. EGA/MCGA/VGA Submenu:
(EGA = enhanced graphics adapter; MCGA = multicolor graphics array; VGA = video graphics
array). If and only if #1-TuffTEST-Pro identifies the presence of an EGA, MCGA, or VGA
video display adapter, the displayed menu offers three options:
XVIII.D.2.b.vi.a. Test Video Memory:
Selecting this option tests each installed video memory bank using the WALKING ONES
Memory Test. The status of each bank is reported.
XVIII.D.2.b.vi.b. Display High-Resolution Color Charts:
This test verifies proper focusing and operation of the Red, Green, and Blue color guns in the
high-resolution mode. Inconsistencies in the color patterns displayed indicate a need for service.
XVIII.D.2.b.vi.c. Test Text Resolution:
This test displays the ASCII character set in three resolution modes: 35 x 80, 42 x 80, and
50 x 80.
NOTE: When the computer contains multiple video display adapters, all but the adapter to be
tested must be removed.

XVIII.D.3. Parallel Menu <Diagnostic Menu F4>

XVIII.D.3.a. Overview:
#1-TuffTEST-Pro provides extensive testing for all standard parallel (IBM/Epson-
type) communication interfaces (LPT1, LPT2, and LPT3 ports) at the standard addresses, both
internally and externally.
The parallel interface was designed to communicate with a printer, although other devices that
communicate via the standard can be used. It is also designed to predominately be a one-way
communicator, rapidly and efficiently transferring data to be printed.
The computer's BIOS controls the basically "dumb" parallel interface by directing the requests of
the operating system (example -the DOS Print Screen command) or application program's
requests to the interface. The interface receives the data, i.e., 8-bit characters in parallel, sends a
CONTROL word to initialize the printer, receives back a STATUS word indicating the printer is
initialized and ready to receive and handle (print) the data. Refer to the Error Code charts below
for the CONTROL and STATUS word possibilities.
The process as described above is based on a character-by-character communication view. The
interface can always operate faster than any of the currently available printers can print, thus the
interface is always waiting on the printer. To make the process of the interface more efficient,
memory buffers were introduced in the computer and then also in the printer thus allowing larger
amounts of characters to be transferred through the interface while not requiring the full and
continuous attention of the PC's BIOS and actual printing by the printer.
XVIII.D.3.b. Parallel Interface Architecture:
The Interface (Signal Loopback) Test validates that the interface circuitry is sending CONTROL
words (signals) and DATA and receiving, via the TEST Plug, a looped-back signal that emulates
the possible return STATUS words (signals). Once the interface is determined to be operating
correctly, the other tests are designed to operate with a printer connected to determine if the
computer is sending correctly and that the printer is receiving, printing, and responding
correctly. As described below, some of the tests operate in a delayed fashion based on the
computer/printer buffer configuration.
The parallel interface is comprised of electrical components and circuits either on a separate add-
in board that fits into a bus slot, as part of several functions on an add-in board, or mounted
directly on the system board.
In all configurations, the interface resides at a set "port" address from which it communicates
with the bus and to the outside world via a DB-25 (female is standard) 25-pin connector.
Within the interface, there are three 8-bit registers for DATA, CONTROL, and STATUS
"words." The DATA register is loaded with bits coming off the bus in parallel form. These 8
bits represent the standard American Standard Code for Information Interchange (ASCII)
character set (128 characters) and the extended set (standard 128 plus an additional 128
characters).
The CONTROL and STATUS registers' 8 bits are only partially used, as there are only
4 CONTROL word bit configurations and 5 STATUS word bit configurations. The location of
the "on" or "1" bits in each word are equated and physically mapped to a separate and unique pin
in the DB-25 connector.
XVIII.D.3.b.i. Standard Pin Values for IBM/Epson DB-25 Parallel Connector:
DATA/CONTROL/
Signal Name Pin STATUS Word Description
Strobe 1 CONTROL (Out) Output signal - the 8 data lines can be read
Data bit 0-7 2-9 DATA (Out) 5 volts = logical 1
0 volts = logical 0
Acknowledge 10 STATUS (In) Input signal - data received
Busy 11 STATUS (In) Input signal - Do Not Transmit
Paper Out 12 STATUS (In) Input signal - printer is out of paper
Select 13 STATUS (In) Input signal - printer is on line, ready to receive
Auto/Feed 14 CONTROL (Out) Output signal selecting the line feed option. If
the signal is low, the printer is commanded to
automatically feed one line when it detects a
carriage return; if the signal is high, an explicit
line-feed character is required.
Error 15 STATUS (In) Input signal - error
Initialize 16 CONTROL (Out) Output signal resetting the printer to its
default
operating parameters
Select 17 CONTROL (Out) Output signal to switch the printer on and off
line
Ground 18-25 Ground
NOTE: The displays for Echo the Keys Pressed to a Printer, Monitor Handshake and Data
Signals While Printing, and Monitor Status Signals from a Printer indicate either "active high"
(+) or "active low" (-) for each of the signal names.
XVIII.D.3.c. Parallel Interface Testing
XVIII.D.3.c.i. Interface (Signal Loopback) Test:
The DB-25 Parallel TEST Plug (the blue plug if purchased from #1-PC Diagnostics Company)
must be inserted into the DB-25 connector on the back of the parallel adapter for the interface
selected. The test involves feeding back five of the eight parallel data lines through the five
status reporting pins, emulating the status signals that would normally originate from the
printer. The data is sent out and received back with the Data, Control, and Status words
indicated as "PASSED" or "FAILED." For each failing word, an error code is displayed
indicating the failing bit(s).
XVIII.D.3.c.ii. Send the ASCII Character Set to a Printer:
This test sends a printer initialization control word and continuously sends the entire ASCII
character set in the classic "barber pole" format to the printer attached to the selected
interface. It also displays the characters that are being sent to the printer. The printer output can
be verified by comparison with the display output.
NOTE: Some printers with buffers do not print until the buffer is full or a carriage return has
been received. Accordingly, there may be a delay before a line is printed after the test begins. A
delay should not be more than a few seconds as parallel interfaces transfer data at an extremely
high rate of speed.
XVIII.D.3.c.iii. Echo the Keys Pressed to a Printer:
This test permits the user to select the character to be printed via the keyboard and echoes
(prints) it on the printer. As each character is selected, the active data and control lines are
highlighted on the display.
NOTE: Some printers with buffers do not print until the buffer is full or a carriage return has
been received. Accordingly, there may be nothing printed until many keys (not including the
ENTER key) have been pressed or until the ENTER key is pressed. Printer buffers of 80
characters or more must be completely filled before it will print a line. To confirm that the
printer is ready for a thorough test, start by pressing the ENTER key once or twice. The paper
should advance one line each time ENTER is pressed. If it does not, check to see that the printer
is on line, that the cable between it and the computer is securely connected, and that paper is
properly loaded in the printer.
XVIII.D.3.c.iv. Monitor Handshake and Data Signals While Printing:
This test sends the entire ASCII character set to the printer one character at a time in the same
sequence as the Send ASCII Character Set Test. As each character is sent, the active data and
control lines are highlighted on the display.
NOTE: Some printers with buffers do not print until the buffer is full or a carriage return has
been received. Accordingly, there may be a delay before a line is printed after the test begins. A
delay should not be more than a few seconds as parallel interfaces transfer data at an extremely
high rate of speed.
XVIII.D.3.c.v. Monitor Status Signals from a Printer:
This test verifies the functionality of the status word sent from the printer. The status word
signals are Acknowledge, Busy, Paper-out, Select, and Error. Verification of the Paper-out
signal is accomplished by inserting a single sheet of paper into the printer and then removing it.
Verification of the Error, Select, Acknowledge, and Busy signals is made by turning the printer
on and off.
XVIII.D.3.c.vi. Display Error Log:
This selection displays the Error Log. See the Error Logging Menu section in Section XVIII.C
for complete information.
XVIII.D.3.d. Parallel Interface Error Codes:
Parallel error code format is Pit-xx where:
P = Parallel error
i = Interface that failed (1 -3)
1 = Interface 1 (LPT1)
2 = Interface 2 (LPT2)
3 = Interface 3 (LPT3)
t = Test number where:
1 = Data test
2 = Control test
3 = Interface status test
xx = Bits that failed where:
x = Value of 1st x
x = Value of 2nd x
The "xx" portion indicates a status code or error code, depending on the test. Each "x" is
expanded into 4 bits via the Parallel Interface Error Bit Chart in Section XVIII.D.3.e. Equate the
"FAILED" bits to the status or error code explanation chart for each test below.
XVIII.D.3.d.i. Data Test:
Each bit in the "xx" error code represents a data bit in the internal circuitry. An error code where
the two xx's = 01 indicates that bit position 0 (oooo oooF) failed the test, and an error code where
the two xx's = FF indicates 2that ALL bits (FFFF FFFF) failed the test.
1st 'x' 2nd 'x'
Bits 7 6 5 4 3 2 1 0
For example, an "xx" error code of 42 (oFoo ooFo) indicates that bits 6 and 1 failed.
NOTE: Refer to the Parallel Interface Error Bit Chart in Section XVIII.D.3.e for the "xx" error
code values.
XVIII.D.3.d.ii. Control Test:
Each bit in the "xx" error code represents a control bit in the internal circuitry. Bits 4-7 are not
used in the error code. An error code where the two xx's = 01 indicates that bit position 0 (oooo
oooF) failed the test, and an error code where the two xx's = FF indicates that ALL bits (FFFF
FFFF) failed the test.
1st 'x' 2nd 'x'
Bits 7 6 5 4 3 2 1
0
|________| | | |
|
| | | |
STROBE signal
Not
Used | | AUTO FEED signal
| INITIA
LIZE PRINTER signal
SELECT
INPUT signal
For example, an "xx" error code of 08 (oooo Fooo) indicates that bit 3 (SELECT INPUT signal)
failed.
NOTE: Refer to the Parallel Interface Error Bit Chart in Section XVIII.D.3.e for the "xx" error
code values.
XVIII.D.3.d.iii. Interface Status Test:
Each bit in the "xx" error code represents a status bit in the external interface. The lower three
bits (0-2) are not used in the error code. An error code where the two xx's = 01 indicates that bit
position 0 (oooo oooF) failed the test, and an error code where the two xx's = FF indicates that
ALL bits (FFFF FFFF) failed the test.
1st 'x' 2nd 'x'
Bits 7 6 5 4 3 2 1 0
| | | | | |_____|
| | | | | |
| | | | | Not Used
| | | | ERROR Signal
| | | |
| | | SELECT Signal
| | PAPER OUT Signal
| ACKNOWLEDGE Signal
BUSY Signal
This test requires that the DB-25 Parallel TEST Plug be connected to the interface being
tested. If the TEST Plug is not installed, all status bits will indicate "FAILED" (an "xx" error
code of F8).
For example, an "xx" error code of 80 (Fooo oooo) indicates that bit 7 (BUSY signal) failed.
NOTE: Refer to the Parallel Interface Error Bit Chart in Section XVIII.D.3.e for the "xx" error
code values.
XVIII.D.3.e. Parallel Interface Error Bit Chart:
BIT POSITION BIT POSITION
7 6 5 4 3 2 1 0

Value of: 1st 2nd
"x" "x"
0 = o o o o 0 = o o o o
1 = o o o F 1 = o o o F
2 = o o F o 2 = o o F o
3 = o o F F 3 = o o F F
4 = o F o o 4 = o F o o
5 = o F o F 5 = o F o F
6 = o F F o 6 = o F F o
7 = o F F F 7 = o F F F
8 = F o o o 8 = F o o o
9 = F o o F 9 = F o o F
A = F o F o A = F o F o
B = F o F F B = F o F F
C = F F o o C = F F o o
D = F F o F D = F F o F
E = F F F o E = F F F o
F = F F F F F = F F F F

o = passing (ok), or OFF status
F = failing, or ON status
For example, an "xx" error code of 02 would have the value of oooo ooFo and would indicate
that bit 1 was failing. An "xx" error code of 4A would be oFoo FoFo, indicating bits 6, 3, and 1
were failing.

XVIII.D.4. Storage Menu <Diagnostic Menu F5>:

Depending upon the PC being tested, selection of the Storage Menu displays the following
potential device selection choices:
Diskette Drive Tests
Fixed Disk Drive Tests
ST506/412 and ESDI Fixed Disk Drive Tests
IDE Fixed Disk Drive Tests
SCSI Fixed Disk Drive Tests
SCSI Tape Drive Tests
XVIII.D.4.a. Diskette Drive Test Menu:
#1-TuffTEST-Pro performs tests and functions on high-density 3" (1.44MB), standard 3"
(720KB), high-capacity 5" (1.2MB) and standard 5" (360KB) diskette drives to verify the
controller/controller board and/or the drive's mechanical functions and electrical integrity.
Some of the following selections perform read and/or write operations. Therefore, the Program
Diskette must be removed and TEST Diskette(s) inserted into each diskette drive. If the Program
Diskette is detected in the selected drive, the message, "Program Diskette in Drive, Remove
Before Continuing" will be displayed.
The special #1-TuffTEST-Pro TEST Diskettes must be used. See Making TEST Diskettes in
Section VI. The TEST Diskettes have a very stressful data pattern to verify that the drive is
operating at specifications and that it can change "mode" appropriately on dual-mode drives
(where two storage sizes are available: 3" 1.44MB/720KB and 5" 1.2MB/360KB).
Using the correct media (diskette) type is important when working with dual-mode drives. High-
Density (3") and high-capacity (5") media will work for all standard-media
functions. However, the opposite is not reliable. Results are unpredictable if high-capacity/high-
density functions are attempted on standard media.
Some diskette drives may indicate soft ("S") errors on some tests due to the speed of the
select/initialization process of the drive versus the speed of the system in which it is
installed. Disregard soft errors on track zero, sector zero.
XVIII.D.4.b. Diskette Drive Tests and Functions
XVIII.D.4.b.i. Specify Diskette Drive to Test:
THIS OPTION MUST BE SELECTED FIRST. This is used to select the diskette drive to be
tested. Upon selection, #1-TuffTEST-Pro checks for the presence of the drive, and determines
whether a TEST Diskette is inserted. If no TEST Diskette is detected or no diskette to be made
into a TEST Diskette, #1-TuffTEST-Pro will return the message: "To test the diskette drive,
insert a TEST Diskette or, to make a TEST Diskette, insert a blank diskette."
XVIII.D.4.b.ii. Rotational Timing Test:
This test graphically displays the rated rotational speed of the diskette drive vs. the actual
rotational speed of the diskette. If the speed deviates more than plus or minus 3% from
the correct speed, the drive should be serviced or replaced.
XVIII.D.4.b.iii. Seek/Read Test:
This test seeks the first sector of every track and reads the data contained in that sector. The
purpose of this test is to verify the ability of the controller to recognize specific track and sector
identifications and of the write/read head assembly to move to and read all tracks and sectors.
Any errors are reported in the error portion of the display along with the track, sector, and head
identification.
XVIII.D.4.b.iv. Write-Protect Verification Test:
CAUTION!! SELECTION OF THIS TEST MAY DESTROY SOME DATA ON THE
DISKETTE.
The purpose of this test is to verify proper functioning of the write-protect switch/sensor and
circuitry. Prompts are given to first insert a TEST Diskette that is write-protected and then to
insert a TEST Diskette that is not write-protected. To write-protect a TEST Diskette, use a write-
protect tab to cover the write-protect notch on a 5" TEST Diskette or move the slide to the
open position on a 3" TEST Diskette. Upon completion, a message is displayed indicating
whether the test "PASSED" or "FAILED."
XVIII.D.4.b.v. Read-Only Test:
This test reads all sectors on the entire diskette surface. As the sectors are read, a graphic display
follows the progression of the write/read heads as they move from track to track and from sector
to sector. Errors are reported with an "S" for a soft error and an "H" for a hard error. A soft
error is an initial error found in any given sector. When the controller detects an initial error, it
performs five read operations of that sector in an attempt to achieve a non-error status. If these
attempts produce a successful read operation of that sector, the graphic display report continues
to show an "S" in the sector being read, and also logs on the display the fact that a soft error
occurred. If an error-free read operation is not achieved after five attempts, the "S" report
changes to "H," indicating a hard error. A hard error is defined as a failure of the controller to
achieve an error-free read operation after five attempts.
XVIII.D.4.b.vi. Format Test/Make TEST Diskette:
CAUTION!! SELECTION OF THIS TEST WILL DESTROY ALL DATA ON THE
DISKETTE.
This test formats the diskette using an extremely stressful data pattern, B6D9h (h =
hexadecimal). It then reads the data pattern, and compares it with the known written pattern. As
the diskette is formatted and written to, the sectors and tracks affected are graphically displayed
on the screen with a "-" indicating a format of that sector is in progress and a "+" indicating a
write operation is in progress. The error codes displayed are the same as the Read-Only Test. If
any errors are reported, this indicates a problem with the diskette drive controller, the media
itself, or the index sensor.
This function is also used to create #1-TuffTEST-Pro TEST Diskettes. See Making TEST
Diskettes in Section VI.
XVIII.D.4.b.vii. Write/Read/Verify Test:
CAUTION!! SELECTION OF THIS TEST WILL DESTROY ALL DATA ON THE
DISKETTE.
This test writes a worst-case data pattern, B6D9h (h = hexadecimal), to all sectors of the diskette
and then performs a read and compare operation. Progress of the write/read operations is
graphically displayed and errors are reported as described in the Read Only Test.
XVIII.D.4.b.viii. Select Testing Range:
This function allows user selection of a particular portion of the diskette for testing. This is
helpful because some diskette drive tests can take a considerable amount of time. If a particular
portion of the diskette, drive, or controller is suspect, this feature can reduce testing time.
XVIII.D.4.b.ix. Display the Error Log:
This selection displays the Error Log. See the Error Logging Menu section in Section XVIII.C
for complete information.
XVIII.D.4.c. Diskette Drive Error Codes:
The diskette drive error code format is Dd-xx during Certification test, where:
D = Diskette
d = Drive number where:
0 = A:
1 = B:
2 = External
xx = Diskette error code where:
01 = Invalid diskette parameter
02 = Address mark not found
03 = Write-protect error
04 = Requested sector not found
06 = Diskette change line active
08 = DMA overrun on operation
09 = Attempted to DMA across a 64K boundary
0C = Media type not found
10 = Cyclic Redundancy Check (CRC) error on diskette read
20 = General controller failure
40 = Seek operation failed
80 = Diskette drive not ready
9E = Data compare error
9F = Start test failed
FF = Sense operation failure
The diskette drive error code format for Error Logging for Certification is #c-Ddxx, where the
additional parameters are:
# = Number of passes
c = Test types -Certification
The diskette drive error code format is cc:hh:ss/Ddtxx, where the additional parameters used for
Diagnostic and Error Logging are:
t = Diskette drive specific test where:
0 = Certification
1 = Start test
3 = Seek/Read
4 = Read Only
5 = Format -Write
6 = Format -Read
7 = Write/Read - Write
8 = Write/Read - Read
cc = Track
hh = Side
ss = Sector
XVIII.D.4.d. Fixed Disk Drive Subsystem
XVIII.D.4.d.i. Overview:
#1-TuffTEST-Pro provides a wide variety of tests and functions designed for the specific
technology of each fixed disk interface/type, including ST506/412, ESDI, IDE, EIDE and SCSI.
#1-TuffTEST-Pro thoroughly tests and verifies the controller's circuitry, the drive's mechanical
and electrical functions, and the reliability of the fixed disk's surface to retain written data.
The fixed disk drive subsystem physical components include the controller, the drive, and the
connectors. One or several of these physical components may be malfunctioning at any one
time. These types of problems are referred to as "hard" damage and usually require repair or
replacement in a rework facility.
Other problems are referred to as "soft" failures and in many cases can be resolved with #1-
TuffTEST-Pro. Note that in some situations a problem may appear to be a physical one (hard
damage) when it is actually a soft failure. These may include a spot failure of the disk surface in
a previously good sector, data recording errors due to electrical surges/spikes or improper shut-
down activities, or a read/write head mistracking due to mechanical wear or temperature
conditions (when the drive is cold or very hot). Also, the drive may appear "dead" due to
corruption of the track 0 boot loader information.
XVIII.D.4.d.ii. Supported Fixed Disk Drive Interfaces/Types:
#1-TuffTEST-Pro supports the following fixed disk drive interfaces/types:
XVIII.D.4.d.ii.a. ST506/412:
All drives of this type are supported if their geometries are supported by the system's BIOS.
The first ST506/412 drives were introduced using the Modified Frequency Modulation (MFM)
data encoding methodology. Soon thereafter, the Run Length Limited (RLL) methodology was
introduced. RLL encoding increased data density by 50% over MFM. Over time, ST506/412
drives became referred to as either MFM or RLL, depending on the data encoding method being
used.
While all fixed disk drives, regardless of type, are capable of using either MFM or RLL
encoding, RLL has become the preferred method due to its greater storage capability.
The original PC interface, the ST506, was introduced by Seagate Technologies for the XT
computers. This was followed by modifications for the AT identified as the ST412, thus this
interface is called the ST506/412. This interface is referred to as a "device-level interface"
where the computer's BIOS routines direct the limited, or "dumb," controller to perform the head
positioning, reads/writes and other functions.
XVIII.D.4.d.ii.b. Enhanced Small Device Interface (ESDI):
Support for fixed disk drives of this type is limited by the make and model of host adapter being
used.
The ESDI interface standard, while an improvement over the ST506/412 type, also had its
shortcomings when compared to IDE and SCSI. ESDI saw its greatest success on the earlier
Compaq desktop systems and in the early IBM PS/2 products. Support for fixed disk drives of
this type is limited by the make and model of the host adapter being used. This is due to the lack
of standardization in alternate sector mapping, translation modes, controller ID procedures, and
device communication modes. The following host adapters (not drives) are currently supported:
- Western Digital 1007 (A and WAH versions)
- Adaptec ACB-232xB
- DTC 6x80
Any ESDI fixed disk drive attached to one of these host adapters can be thoroughly tested using
#1-TuffTEST-Pro.
NOTE: #1-TuffTEST-Pro will not support more than one ESDI fixed disk drive in the same
computer.
XVIII.D.4.d.ii.c. Integrated Drive Electronics (IDE & EIDE):
All drives of this type are supported if their geometries are supported by the system's BIOS.
IBM, with the introduction of the AT, also introduced the AT Attachment (ATA) standard that
would allow a wider variety of larger, more capable storage devices to be attached. IDE &
Enhanced IDE (EIDE) drives are based on the ATA standard.
The IDE (ATA) interface is referred to as a "logic-level interface." The IDE controller has its
own BIOS and localized intelligence and is located within the drive unit. A simple interface bus
card performs signal decoding and buffering, the computer's BIOS requests logical functions,
and the controller/drive (the "drive") has complete control of the actual drive activity.
It is important to note that the ATA standard defines how functions and data are requested from
and received by the computer with no set definition of how it is done within the drive. As such,
there is a wide range of storage and retrieval methodology implemented from manufacturer to
manufacturer, sometimes within a manufacturer's product line and from the older drives to the
newer units.
NOTE: Fixed disk drives in a PC whose BIOS does not support INT13h extensions will be
tested in Cylinder/Head/Sector (CHS) mode.
NOTE: Fixed disk drives in a PC whose BIOS supports INT13h extensions will be tested in
Logical Block Addressing (LBA) mode. LBA is a way of addressing fixed disk drives by
assigning numbers sequentially to each sector on the drive, starting from 0 for the first sector.
NOTE: Fixed disk drive drives over 8.4 gigabytes (GB) cannot be identified or tested using CHS
mode. They must always be tested in LBA mode.
XVIII.D.4.d.ii.d. Small Computer Systems Interface (SCSI):
All drives of this type are supported provided the SCSI host adapter's BIOS supports INT13h and
drive counting located at 40:75 in the BIOS data area.
The Small Computer System Interface (SCSI) standard is an intelligent subsystem (not a fixed
disk interface) that manages up to eight different devices (fixed disk drives, tape drives,
scanners, CD-ROM drives, etc.). The SCSI host adapter (an add-in bus card or components set
on the system board) depends on its own internal BIOS and bus, which are separate from that of
the host computer. During system power-on process, the SCSI controller queries each device it
locates on its bus for the device's operating parameters. These parameters are then used by the
controller to operate that device.
NOTE: With SCSI drives attached to older SCSI Host Adapters whose BIOS does not support
INT13h and drive counting located at 40:75 in the BIOS data area, #1-TuffTEST-Pro's SCSI
tests must bypass the system BIOS and work directly with the attached SCSI device using the
SCSI command set. This requires a special device driver for each supported host
adapter. Therefore, support for fixed disk drives of this type is limited by the make and model of
host adapter being used. The following host adapters are supported:
- Adaptec AHA-1540/1542 (A, B, C, CF) and AHA-1740/1742A (Standard Mode)
- Western Digital WD7000-ASC Host Bus Adapter
- NCR Intelligent SCSI Host Adapter (as found in NCR System 3000 Class 3433
computers)
- NCR SCSI Host Adapter (as found in NCR System 3000 Class 3421 computers)
Any SCSI fixed disk drive or tape drive attached to one of these host adapters can be thoroughly
tested using #1-TuffTEST-Pro. Selecting these tests for host adapters other than those listed
above will produce unpredictable results.
XVIII.D.4.d.iii. Hard-Card-Type Fixed Disk Drives:
#1-TuffTEST-Pro does not currently support hard-card-type fixed disk drives due to the
nonstandard format used. Consult the manufacturer of the hard card for information regarding
proprietary diagnostics for that particular product.
XVIII.D.4.d.iv. Fixed Disk Drive Tests and Functions
Depending upon the PC being tested, the following menu choices will be available:
XVIII.D.4.d.iv.a. Select Drive to be Tested (ST506/412, ESDI, IDE, EIDE, SCSI):
This option must be selected first. If there is only one fixed disk drive installed, it is selected
when <Enter> is pressed. If more than one fixed disk drive is installed, the proper drive
designator must be entered. The remaining tests will not run unless a fixed disk drive is selected.
ESDI NOTE: Newly installed fixed disk drives using Adaptec ESDI Controllers must first be
initialized using the Adaptec-supplied documentation/procedure and using the format programs
supplied on the controller's BIOS ROM. The Adaptec controller's BIOS ROM can also be
accessed via the DOS DEBUG.COM utility. Until this beginning initialization is done, #1-
TuffTEST-Pro will display the message, "This drive cannot be tested until it is initialized."
XVIII.D.4.d.iv.b. Select Host Adapter/Drive to be Tested (SCSI):
On SCSI drives attached to supported host adapters that do not support INT13h, this option must
be selected first. It displays all supported SCSI host adapters. If two or more host adapters are
installed and recognized by #1-TuffTEST-Pro, they must be tested individually by selecting the
proper designator. Upon selection, the identification and testing process described above applies
to each selection.
XVIII.D.4.d.iv.c. Controller Test (ST506/412, ESDI):
This test verifies operation of the fixed disk drive controller using the controller's ROM-based
internal diagnostics. #1-TuffTEST-Pro reports the title of each test that the controller is running
and displays the result of each test as "PASSED" or "FAILED."
XVIII.D.4.d.iv.d. Controller Test (IDE, EIDE):
This test verifies operation of the IDE-type fixed disk drive controller circuitry that is an integral
component of the fixed disk drive. Therefore, any reported failure indicates that the fixed disk
drive's circuitry is malfunctioning and the fixed disk drive needs to be replaced.
XVIII.D.4.d.iv.e. Controller Test (SCSI):
This test verifies operation of the SCSI host adapter independent of the fixed disk
drive. Therefore, any reported failure indicates a failing host adapter rather than a failing fixed
disk
drive.
XVIII.D.4.d.iv.f. Seek (Hysteresis) Test:
This test can be run in single pass or continuous mode. A graphic display shows the progress of
the test. This test examines the selected drive's ability to accurately position the heads over each
track, from both directions, and at maximum head positioning speed. The test moves the heads
from track 1 to the last track (farthest inside near the spindle), track 2 to last, track 3 to last,
etc. When reaching the last track, it reverses and goes last track to last track minus 1, then last to
last minus 2, etc., until it reaches last track to first track. Any failure is reported in that portion of
the display labeled "ERRORS." If errors are detected it indicates the head positioning
mechanism may need to be repaired or replaced. As this may also indicate potential mechanical
failure the data on the drive should be backed up as soon as possible using a suitable backup
utility or application program.
XVIII.D.4.d.iv.g. Find Previously Marked Bad Tracks (ST506/412, ESDI):
This function reads each track within the selected range to determine if it is currently marked
as "bad," either by the manufacturer or a previous low-level format. This is a simple read-and-
identify process, with no testing or verification. All tracks previously marked bad are noted in
#1-TuffTEST-Pro's bad tracks table in Memory. Select Edit and/or Mark Bad Tracks to review
the table.
XVIII.D.4.d.iv.h. Surface Analysis (Read and Verif) Test:
This test can be run in single pass or continuous mode. A graphic display shows the progress of
the test. Error Correction Code (ECC) and Cyclic Redundancy Check (CRC) are two types of
algorithmic data integrity testing routines used in fixed disk drives. Either technique, but not
both, will be used as part of that controller's architecture. When data is written to a sector on the
fixed disk drive, the controller automatically performs a mathematical operation on all the data
bytes within that sector and records the result of that operation on a portion of the sector
immediately following the 512 bytes of data. When data is read from a particular sector, the
mathematical operation is again performed and the results compared with the previous results.
Any discrepancy between the ECCs/CRCs is reported in that portion of the display labeled
"ERRORS."
If errors are detected it indicates either logical (soft) or physical (hard) failures of the sector. As
this may also indicate potential mechanical failure, the data on the drive should be backed up as
soon as possible using a suitable backup utility or application program. Then, depending on the
system, Media Analysis or Media Analysis & Repair should be run. It will repair all soft errors
and report all hard errors.
ESDI NOTE: In using the Surface Analysis Test, if the sector mapping process on the selected
fixed disk drive has been changed and a full format has not been performed, the message,
"Alternate sector mapping changed -format MUST be run," will be displayed. If a format is
required, select Low-Level Format - UNCONDITIONAL.
XVIII.D.4.d.iv.i. Find Optimum Interleave (ST506/412, ESDI):
The interleave factor determines the logical sequence of sector markings, and therefore
determines the number of media revolutions necessary for the fixed disk drive to read an entire
track. The interleave setting greatly affects the data throughput. If the interleave factor is set too
high (more revolutions than necessary), maximum efficiency and speed are not achieved. If the
interleave factor is set too low, one complete revolution is required to read each sector. On a
drive with 17 sectors per track, the result is the same as an interleave factor of 17 (one sector per
revolution). Therefore, it is important that interleave be set properly.
When this test is used, the service cylinder of the fixed disk drive is formatted with interleave
factors 8 through 1. The service cylinder is an inner track not used for the storage of data,
therefore write/read tests can be performed there. After each interleave factor is written, that
cylinder's data is read, the process timed in bytes per second, and the calculated read time is
written to the display. Upon completion of this process, #1-TuffTEST-Pro indicates the
optimum interleave factor on the display. This interleave factor should then be entered within
the format portion of #1-TuffTEST-Pro and the fixed disk drive should be reformatted with the
optimum interleave factor.
It is important to note that the interleave factor is determined based on the retrieval of sequential
data, meaning that the data to be read is available in adjacent sectors. If the primary application
of the drive being formatted is one that accesses fragmented data or not sequentially, i.e.,
randomly, optimum interleave performance may be achieved by increasing the interleave
factor by 1.
XVIII.D.4.d.iv.j. Low-Level Format - CONDITIONAL (Preserve Bad Tracks) (ST506/412,
ESDI):
CAUTION!! SELECTION OF THIS OPTION WILL DESTROY ALL DATA ON THE
SELECTED FIXED DISK DRIVE. Therefore, the drive should be backed up using a suitable
backup utility or application program before running this test.
This test overwrites all existing data and sector/cylinder address marks. New address marks
and data blocks are written. Bad track entries from previous formatting processes are left
intact. If it is desired to delete previous bad tracks entries, use the Low-Level Format -
UNCONDITIONAL selection.
With some fixed disk drive controllers, the routines used to preserve bad tracks markers cause
recalibration seeks to track 0 after each track is formatted. These recalibration seeks
significantly increase the time required to format. When this is the case, the following alternate
method is suggested:
1) Use the Find Previously Marked Bad Tracks function to locate and log all existing bad tracks
flags to #1-TuffTEST-Pro's bad tracks table in memory.
2) Select Low-Level Format -UNCONDITIONAL to perform a timely format of the fixed disk.
3) Select Media Analysis Test to locate and log newly identified bad tracks to the bad tracks
table in memory.
4) Select Edit and/or Mark Bad Tracks to review and mark all tracks indicated in the bad tracks
table in memory.
The fixed disk drive is now ready for an operating system partition and format.
ESDI NOTE: For ESDI-type drives, the Low-Level Format - CONDITIONAL function does
not support Alternate Sector Mapping. If the currently selected fixed disk drive is using alternate
sector mapping, the message, "This option is not available with this drive or controller," will be
displayed. If a format is required, select Low-Level Format - UNCONDITIONAL.
XVIII.D.4.d.iv.k. Low-Level Format - UNCONDITIONAL (ST506/412, ESDI):
CAUTION!! SELECTION OF THIS OPTION WILL DESTROY ALL DATA ON THE
SELECTED FIXED DISK DRIVE. Therefore, the drive should be backed up using a suitable
backup utility or application program before running this test.
This test overwrites all existing data and sector/cylinder address marks with new address marks
and data blocks. Bad track entries from previous formatting processes are overwritten.
As fixed disk drives wear, the relative track positions can change. When this happens, the
relative location of defects will also change. In these circumstances, if previous bad track entries
are not cleared, the old bad track locations will still be marked bad, even though the defect has
moved and the location now contains reliable media.
In addition, this option must be used if the drive will no longer boot because the track 0 boot
loader information has become corrupted.
ESDI NOTE: For ESDI-type drives, the Low-Level Format - UNCONDITIONAL function
determines whether alternate sector mapping is currently enabled or disabled on the selected
fixed disk drive. A message will be displayed indicating the fixed disk drive's current sector
mapping selection, followed by the message: "Do you want to use alternate sector
mapping (Y/N)?" Input the desired Y/N selection. If the sector mapping process on the selected
fixed disk drive is changed, a full format is required.
XVIII.D.4.d.iv.l. Low-Level Format (SCSI):
CAUTION!! SELECTION OF THIS OPTION WILL DESTROY ALL DATA ON THE
SELECTED FIXED DISK DRIVE. Therefore, the drive should be backed up using a suitable
backup utility or application program before running this test.
On SCSI drives attached to supported host adapters that do not support INT13h, this test rewrites
all existing data and sector/cylinder address marks. New address marks and data blocks are
written. Bad block entries from previous formatting processes are left unchanged. No defect
data is used for this process. If media defects are suspected, use Surface Analysis Test or Media
Analysis Test to locate defects, then perform the Re-Assign Blocks (SCSI) function.
XVIII.D.4.d.iv.m. Media Analysis (Write/Read) Test (ST506/412, ESDI, IDE, SCSI):
CAUTION!! SELECTION OF THIS OPTION WILL DESTROY ALL DATA ON THE
SELECTED FIXED DISK DRIVE. Therefore, the drive should be backed up using a suitable
backup utility or application program before running this test.
The Media Analysis Test is one of #1-TuffTEST-Pro's most thorough media tests. This test can
also be used to security erase a fixed disk drive. For best results, it should be run in continuous
mode for at least 2-5 passes.
This test writes seven different high-stress data patterns, calculates and enters the appropriate
CRC (the cyclic redundancy check character) for each pattern, reads the data back, recalculates
the CRC, and compares both CRC results in the selected test range. A graphic display is
presented as the test is running in addition to designating the head and cylinder being
tested. Any discrepancy between CRCs is reported in that portion of the display labeled
"ERRORS."
As this test runs, it logs each defect location into #1-TuffTEST-Pro's bad track table in
memory. Select Edit and/or Mark Bad Tracks to review the table. It is important to note that
defects can be affected by bad or improperly placed format information. A Media Analysis Test
should only be performed as a final step AFTER a Low-Level Format - CONDITIONAL or a
Low-Level Format - UNCONDITIONAL. Only this defect information, combined with
manufacturer-supplied defect maps, should be used to update the bad track table on the fixed
disk using the Edit and/or Mark Bad Track function.
NOTE: It is possible to run this test without destroying existing data if the service cylinder is
isolated using Select Testing Range.
ESDI NOTE: In using the Media Analysis Test, if the sector mapping process on the selected
fixed disk drive has been changed and a full format has not been performed, the message,
"Alternate sector mapping changed format MUST be run," will be displayed. If a format is
required, select Low-Level Format - UNCONDITIONAL.
XVIII.D.4.d.iv.n. Media Analysis & Repair (Write/Read/Verify/Re-Assign Bad Blocks) Test
(IDE,EIDE, SCSI):
CAUTION!! SELECTION OF THIS OPTION WILL DESTROY ALL DATA ON THE
SELECTED FIXED DISK DRIVE. Therefore, the drive should be backed up using a suitable
backup utility or application program before running this test.
The Media Analysis & Repair Test is #1-TuffTEST-Pro's most thorough media repair
test. This test can also be used to security erase a fixed disk drive. For best results, it should be
run in continuous mode for at least 2-5 passes.
This test writes seven different high-stress data patterns, calculates and enters the appropriate
CRC (the cyclic redundancy check character) for each pattern, reads the data back, recalculates
the CRC, and compares both CRC results in the selected test range. A graphic display is
presented as the test is running in addition to designating the head and cylinder being
tested. Any discrepancy between CRCs is reported in that portion of the display labeled
"ERRORS."
This test will automatically repair all logically corrupted sectors. For example, if a PC will not
boot because its boot sector is logically corrupted this test will repair it.
In addition, this test will automatically re-assign, ie. replace, all physically bad sectors with
known good sectors from the drive's pool of spare sectors. If no more spare sectors are available
appropriate error codes will be displayed if additional physically bad sectors are detected.
Important: Once the drive's pool of spare sectors is depleted, the drive is no longer reliable and
should be replaced.
If the Media Analysis & Repair Test produces no errors, it indicates the drive has been
successfully repaired and is now safe to use. To re-use the drive, use the operating system's
utilities to partition and format the drive. Then install your operating system.
XVIII.D.4.d.iv.o. Edit and/or Mark Bad Tracks (ST506/412, ESDI):
This function allows the user to view, add, delete, and/or edit entries in #1-TuffTEST-Pro's Bad
Track Table in memory. Bad track locations are automatically logged here by Find Previously
Marked Bad Tracks, Surface Analysis, and Media Analysis. Additionally, manufacturer-
supplied defect/bad track parameters, which may be found as an external label on the drive's
housing, can be entered here.
Finally, when requested by the user, this function writes bad track markers to all indicated
sectors/tracks on the fixed disk drive. The actual locking out of bad tracks from the
file allocation table is a function of the operating system (high-level) format.
XVIII.D.4.d.iv.p. Re-Assign Blocks (SCSI):
On SCSI drives attached to supported host adapters that do not support INT13h, this function
takes the blocks found to be bad as a result of using Surface Analysis or Media Analysis, and
deactivates and re-assigns them via the SCSI host adapter.
XVIII.D.4.d.iv.q. Select Testing Range:
This function allows user selection of a particular portion of the fixed disk drive media for
testing. This is often advantageous, as most fixed disk drive tests take a considerable amount of
time. If a particular portion of the fixed disk drive media is suspect, this feature dramatically
decreases testing time.
XVIII.D.4.d.iv.r. Display Error Log:
This selection displays the Error Log. See the Error Logging Menu section in Section XVIII.C
for complete information.
XVIII.D.4.d.iv.s. Park the Disk Heads (ST506/412, ESDI):
This utility function moves the write/read heads of the selected fixed disk drive to the service
cylinder. The purpose of this function is to move the heads away from any data. This function
should be invoked before removal and/or shipping of any fixed disk drive.
XVIII.D.4.d.iv.t. Host Adapter ROM-Based Utilities (SCSI):
Many newer SCSI host adapters, such as Adaptec's AHA-2940, have configuration and
diagnostic utilities in their ROM BIOS. Consult your host adapter's product documentation for
instructions on how to access these utilities.
XVIII.D.4.e. SCSI Tape Drive Tests:
On SCSI tape drives attached to supported host adapters that do not support INT13h, this
grouping of SCSI tests bypasses the system BIOS and works directly with the attached SCSI
device using the SCSI command set. A special device driver is required for each supported host
adapter. Therefore, support for tape drives of this type is limited only by the make and model of
host adapter being used. The following host adapters are directly supported:
- Adaptec AHA-1540/1542 (A, B, C, CF), and AHA-1740/1742A (Standard Mode)
- Western Digital WD7000-ASC
- NCR Intelligent SCSI Host Adapter (as found in NCR System 3000 Class 3433 computers)
- NCR SCSI Host Adapter (as found in NCR System 3000 Class 3421 computers)
Any SCSI tape drive attached to one of these host adapters can be thoroughly tested using
#1-TuffTEST-Pro. Selecting these tests for host adapters other than those listed above will
produce unpredictable results.
XVIII.D.4.e.i. Select Drive to be Tested:
This option must be selected first. If there is only one tape drive installed, it is selected when
<Enter> is pressed. If more than one tape drive is installed, the proper drive designator must be
entered. The remaining tests will not run unless a drive is selected.
NOTE: A blank tape cartridge must be properly inserted in the tape drive BEFORE the drive is
selected.
XVIII.D.4.e.ii. Combination Test:
This test executes all of the SCSI tests listed below.
XVIII.D.4.e.iii. Erase Test:
This test determines if the erase function of the tape drive is working properly. The tape is first
rewound and checked for errors. The entire tape is then erased and again checked for errors. All
errors are reported in the error portion of the display.
XVIII.D.4.e.iv. Retension Test:
This test determines if the retension function of the tape drive is working properly. The tape is
first rewound and checked for errors. The tape is then retensioned and checked for errors. All
errors are reported in the error portion of the display.
XVIII.D.4.e.v. Write Filemarks Test:
This test determines the ability of the tape drive to write and read filemarks. The tape is first
rewound and checked for errors. Then 512 consecutive filemarks are written to the tape and
checked for errors. The tape is again rewound and checked for errors. Finally, the tape is spaced
past 512 consecutive filemarks and again checked for errors. All errors are reported in the error
portion of the display.
XVIII.D.4.e.vi. Start/Stop/Write Test:
This is a high-stress test that exercises the tape drive's ability to start, stop, read, and write. The
tape is first rewound and checked for errors. Then 1024 blocks of data are written, forcing the
tape to stop, rewind, and verify after each block. The tape is again rewound and checked for
errors. All 1024 blocks of data are then sequentially verified and checked for errors. All errors
are reported in the error portion of the display.
XVIII.D.4.e.vii. Short-Streaming Write Test:
This is a quick check of the tape drive's ability to write data in a streaming mode. The tape is
first rewound and checked for errors. Then, while the status is continually monitored, 256 32-
block writes are performed. The tape is again rewound and checked for errors. All 8192 blocks
are then verified and checked for errors. All errors are reported in the error portion of the
display.
XVIII.D.4.e.viii. Long-Streaming Write Test:
This is a long test of the tape drive's ability to write data in a streaming mode. The tape is first
rewound and checked for errors. Then, while the status is continually monitored, 32-block
writes are performed over the entire tape. The tape is again rewound and checked for
errors. The entire tape is then verified and checked for errors. All errors are reported in the error
portion of the display.
XVIII.D.4.e.ix. Display Error Log:
This selection displays the Error Log. See the Error Logging Menu section in Section XVIII.C
for complete information.
XIV.D.4.f. Fixed Disk Drive and Tape Drive Error Codes
XIV.D.4.f.i. Fixed Disk Drives Only:
For all fixed disk drives except for SCSI drives attached to host adapters whose BIOS does not
support INT13h and drive counting located at 40:75 in the BIOS data area.
The fixed disk drive error code format for Certification and Diagnostics is Fd-xx where:
F = Fixed disk
d = Drive number
xx = BIOS Error code where:
01 = Invalid function request - bad command
02 = Address mark not found
03 = Write-protect error
04 = Sector not found
05 = Reset failed
07 = Drive parameter activity failed
08 = DMA overrun on operation
09 = Data boundary error -DMA boundary error
0A = Bad sector flag detected - bad sector (marked bad)
0B = Bad cylinder detected - bad track
0D = Invalid number of sectors in format
0E = Control data address mark detected
0F = DMA arbitration level out of range
10 = Uncorrectable Error Checking and Correction (ECC)
error.
11 = ECC corrected data error (ECC error that was corrected)
20 = General controller failure
40 = Seek operation failed
80 = Time-out, no response from device
9E = Data compare error
AA = Drive not ready
BB = Undefined error occurred
CC = Write fault on selected drive
E0 = Status error/error register=0 unknown status error
FF = Sense operation failed
The fixed disk drive error code format for Error Logging is ccc:hh:ss/Fdtxx when the drive is
tested in CHS mode, or ss/Fdtxx when the drive is tested in LBA mode where the additional
parameters are:
t = Test number where:
1 = Seek (Hysteresis)
2 = Media Analysis
3 = Surface Analysis
4 = Low-Level Format - Test CONDITIONAL
5 = Low-Level Format - Test UNCONDITIONAL
6 = Park Heads Test
xx = See above
hh = Head number
ccc = Cylinder number
ss = Sector number
XVIII.D.4.f.ii. SCSI (Fixed Disk and Tape Drives)
For SCSI drives attached to "supported" host adapters whose BIOS does not support INT13h and
drive counting located at 40:75 in the BIOS data area.
XVIII.D.4.f.ii.a. ADAPTEC/NCR:
The error code for Adaptec/NCR SCSI drive types is DTLt-ABS, where:
D = Device where:
A = Tape Drive
F = Fixed Disk
T = Target
L = LUN (Logical Unit Number)
t = Test Number where:
If Fixed Disk
0 = Controller
1 = Media Analysis
2 = Surface Analysis
3 = Re-Assign Blocks
4 = Format Disk
If Tape
1 = Combination
2 = Erase
3 = Retension
4 = Write Filemarks
5 = Start/Stop Write
6 = Short-Streaming Write
7 = Long-Streaming Write
A = Adapter Error Code where:
00 = No host adapter detected
0A = Linked command complete without error
0B = Linked command complete without error, interrupt
generated
11 = Selection time out
12 = Data overrun/underrun
13 = Unexpected bus free
14 = Target bus phase sequence failure
15 = MBO (Mail Box Out) command was not 00, 01, or 02
16 = First byte of the CCB (Command Control Block) was
not 00, 01, 02, or 81
17 = Linked CCB does not have the same status byte codes
18 = Invalid CCB parameter received from host in target
mode
19 = Duplicate CCB received in target mode
20 = Linking violation in target mode
B = Target Status:
00 = Good status
02 = Check status
08 = LUN busy
18 = Reservation conflict
S = Sense Key:
0 = No Sense
1 = Recovered Error
2 = Not Ready
3 = Media Error
4 = Hardware Error
5 = Illegal Request
6 = Unit Attention
7 = Data Protect
8 = Blank Check
9 = Vendor Unique
A = Copy Aborted
B = Aborted Command
C = Equal
D = Volume Overflow
E = Miscompare
F = Reserved
XVIII.D.4.f.ii.b. Western Digital 7000:
The error code for the Western Digital 7000 SCSI device types is DTLt-ABS where:
D = Device where:
A = Tape Drive
F = Fixed Disk
T = Target
L = LUN (Logical Unit Number)
t = Test Number where:
If Fixed Disk
0 = Controller
1 = Media Analysis
2 = Surface Analysis
3 = Re-Assign Blocks
4 = Format Disk
If Tape
1 = Combination
2 = Erase
3 = Retension
4 = Write Filemarks
5 = Start/Stop/Write
6 = Short-Streaming Write
7 = Long-Streaming Write
A = Adapter Error
00 = Power-up condition
01 = No errors occurred
02 = RAM failed
03 = FIFO (first-in first-out) buffer Read/Write failure
04 = SBIC (SCSI Bus Interface Controller) failure
05 = Initialization D-FF (D Flip-Flop Register) failure
06 = Host IRQ D-FF failure
07 = ROM checksum error
08 = Unexpected SBIC response during data out phase
09 = Unexpected SBIC response during data in phase
0A = Unexpected SBIC response during command
out phase
0B = Unexpected SBIC response during status in phase
0C = Unspecified SCSI out phase
0D = Unspecified SCSI in phase
0E = Unexpected SBIC response during message out phase
0F = Unexpected SBIC response during message in phase
10 = Unexpected SBIC interrupt status code
18 = DMA (Direct Memory Access) error during host system
memory diagnostics
20 = Command issued with OGMB (Outgoing Mail Box)
marked empty
21 = Illegal command parameter
22 = Command cannot be executed at the present time
23 = Target data direction contrary to the one specified
24 = Set/read execution parameters index out of range
25 = Set/read execution parameters count out of range
26 = Host has a previous command in progress for the same
SCSI target and LUN
40 = Target sent less data than the allocation length (bytes 16
-18)
41 = Target wants to send more data than the allocation length
42 = Target wants to transfer data but host allocation length =
0
43 = Reserved
44 = Specified tag (SCSI ID+LUN)command aborted
45 = Specified target received reset message - command
aborted
4D = Timeout occurred during target selection/reselection
4E = Target did not send ID message, or invalid source ID
4F = No record of the SCB (SCSI Command Execution Block)
to be executed
50 = Unrecoverable error or disconnect in ID or selection
phase
51 = Unrecoverable error or disconnect in command phase
52 = Unrecoverable error or disconnect in message in phase
53 = Unrecoverable error or disconnect in message out phase
54 = Unrecoverable error or disconnect in ID or reselection
phase
55 = Unrecoverable error or disconnect in data transfer phase
56 = Unrecoverable error or disconnect in status phase
57 = Unrecoverable error or disconnect in command phase
80 = SCSI hard reset during a solicited selection
81 = valid command bytes in scratchpad buffer
82 = No command scratchpad buffer opened
83 = Illegal CDB (Command Data Block) transfer length
84 = No inbound data buffer opened, wrong ID, or
unqualified LUN
85 = No sense data buffer opened for the specified LUN
86 = Buffer opened but next command was not request sense
87 = No inquiry buffer opened as qualified by LAV/IAV
(LUN Acceptance Vector/ID Acceptance Vector)
88 = No outbound data buffer opened, wrong ID, or
unqualified LUN
FF = Not applicable in current field
B = Target Status where:
00 = Reserved
01 = Command complete, no errors
02 = Command complete, error logged in SCB/ICB(SCSI
Command Execution Block/Interface Command Block) bytes 14 and 15
03 = Scan OGMB (Out-Going Mail Box) command has been
completed
04 = Command failed to complete without SCSI status
05 = Command terminated, bus reset by external device
during SCB
06 = SBIC or hardware failure, requires host reset
07 = SCSI soft reset command completed
08 = Link command complete with no errors when flag bit is
set
80 = Unexpected reselection
81 = Unexpected selection
82 = Abort command message
83 = SCSI reset message received
84 = SCSI hardware reset with no command in queue
S = Sense Key
0 = No Sense
1 = Recovered Error
2 = Not Ready
3 = Media Error
4 = Hardware Error
5 = Illegal Request
6 = Unit Attention
7 = Data Protect
8 = Blank Check
9 = Vendor Unique
A = Copy Aborted
B = Aborted Command
C = Equal
D = Volume Overflow
E = Miscompare
F = Reserved

XVIII.D.5. Serial Interface Menu <Diagnostic Menu F6>

XVIII.D.5.a. Overview:
#1-TuffTEST-Pro provides extensive testing for all standard serial (asynchronous/RS-
232) communication interfaces (COM1, COM2, COM3, and COM4 ports) at the standard
addresses, both internally and externally.
In addition to the standard interfaces (ports), #1-TuffTEST-Pro also tests up to 64 User-Defined
Addresses (UDA) for interfaces (ports) installed at addresses other than the standard
addresses. All multiport I/O boards and other nonstandard address interfaces can be tested if
they comply with the industry standard asynchronous (RS-232) communications protocol.
XVIII.D.5.b. Serial Interface Architecture:
The serial interface is designed for two-way communication. It can send data or reverse the
process and receive data in exactly the same fashion with its full range of capability.
Data, in the form of 8-bit bytes, is sent over the bus in parallel form to the port. The
asynchronous interface board at the port address takes the parallel data in and converts it to serial
format by "clocking" it out of its buffer at set intervals (baud rate) in a serial fashion. The
interface board conditions the signals for transmission and sends them out through the interface
connector either remotely (via modem) or locally (via cabling). This parallel-to-serial
conversion when transmitting and serial-to-parallel conversion when receiving is the primary
function of the serial interface.
The primary component of the serial interface is either a Universal Asynchronous
Receiver Transmitter (UART), or Universal Synchronous/Asynchronous Receiver Transmitter
(USART) integrated circuit module (chip).
All of the data flows through two lines (wires--as in the "twisted" pair telephone line), with one
line for signals going out and one line for signals coming in. This is known as "full duplex"
communication.
The lines carry data, control, and status signals in both directions. Data and control go out and
the status comes back in. This can be reversed where data and control comes in and status is sent
out. The interface can create each type of signal for transmission, or identify and verify them
when receiving. To do this the interface (printed circuit board that goes into a bus slot) has
buffers and registers for data, control, and status functions. As they are an integral part of the
UART/USART design, they can be tested both internally and externally. The external testing
also includes the line driver/signal conditioning components and circuits that work with the
external connector (either a male-type DB-25 or DB-9).
If all of the #1-TuffTEST-Pro tests pass and do not indicate a malfunction on the interface
adapter board, check the cabling or the serial device at the other end of the cable.
XVIII.D.5.c. Serial Interface Testing
XVIII.D.5.c.i. Select I/O Port Addresses:
To perform either the internal or external tests, the address(es) must first be specified. This
selection displays the four standard COM ports and their addresses and 64 additional User-
Defined Addresses (UDA).
Each standard BIOS-supported port that is found by #1-TuffTEST-Pro will be identified as
installed by a "+" mark after the address (e.g., 03F8+). It can be toggled ON/OFF for testing by
pressing the "T" key. The default setting is ON. Ports that have been toggled ON for testing will
be marked with an "*" before the address (e.g., *03F8+).
NOTE: A port may be installed at a standard address, but if it is not indicated as installed nd
BIOS-supported by #1-TuffTEST-Pro, it would appear to be malfunctioning. To verify the
port, the address can be toggled ON and the interface can be tested.
Each standard UDA can also be toggled ON/OFF for testing with the default setting as
OFF. The UDAs start at 0100h address (the accepted convention for additional interfaces) and
increment by 8 bytes up to 64 addresses. Note that the address range of 0200-03FF is not
included in
#1-TuffTEST-Pro's standard display. These addresses are often used for attaching devices
other than serial interfaces. If the system has a serial interface installed at an address in this
range, or any other address not listed, it can be entered as a UDA and tested.
Navigating via the arrow keys, place the cursor at any UDA listed on the display, press the space
bar, and enter the 4-digit hexadecimal address to be tested. The entered address will appear at
the location of the cursor (the prior UDA). All user-entered addresses can be toggled ON/OFF
for testing.
The following tests can be performed on all addresses selected (toggled ON):
XVIII.D.5.c.ii. Internal Operations Test:
The UART/USART is switched into the test mode, enabling #1-TuffTEST-Pro to test its
internal registers without transmitting data to the external interface. While in this mode, data is
written in parallel format to the UART/USART which converts it to serial data and adds the
control (handshake) status signals. It is transmitted internally within the UART/USART, which
loops it back internally, checks the control line (handshake) status signals, then reconverts it to
parallel data. The reconverted data is then read and compared by #1-TuffTEST-Pro. Any
failure is reported as "FAILED" and is displayed next to the failing transmission signal(s).
XVIII.D.5.c.iii. External Operation Test:
This test requires the DB-25 Serial TEST Plug (the yellow plug if purchased from #1-PC
Diagnostics Company) and/or the DB-9 Serial TEST Plug (the green plug if purchased from
#1-PC Diagnostics Company) to be inserted into the DB-25 and/or DB-9 connector on the back
of the serial adapter(s). During this test, #1-TuffTEST-Pro performs the transmission portion of
the internal test with the test mode turned off. This causes the UART/USART to send data out
via the interface connector and receive it back after it is looped through the external TEST Plug
which is installed in the interface connector. It also sends and receives the control line
(handshake) status signals. #1-TuffTEST-Pro then reads the Modem Status Register in the
UART/USART and displays an error code if the expected values of the status bits are not
correct. Any failure is reported as "FAILED" and displayed next to the failing transmission
signal(s).
XVIII.D.5.c.iv. Display Error Log:
This selection displays the Error Log. See the Error Logging Menu section in Section XVIII.C
for complete information.
XVIII.D.5.d. Serial Interface Error Codes:
Serial (Asynchronous) Interface error code format is Stxx-aaaa where:
S = Asynchronous (Serial) error
t = Test number where:
1 = Transmitter test
2 = Receiver test
3 = Data test
4 = External Control Line test
5 = Internal Control Line test
xx = Bits that failed where:
x = Value of 1st x
x = Value of 2nd x
aaaa = The address, in hex, of the port that failed
The "xx" portion indicates either a status code or an error code, depending on the test. Each "x"
is expanded into 4 bits via the Serial Interface Error Bit Chart in Section XVIII.D.5.e. Equate
the "FAILED" bits to the status or error code explanation chart for each test below.
The "xx" portion of the error codes generated by the various tests (the error code "t" value) are as
follows:
XVIII.D.5.d.i. Transmitter Test or Receiver Test:
(Error code "t" = 1 or 2) Reports the status of the UART/USART internal circuitry. Bit 7 is not
used. Where xx = 01 indicates that bit position 0 (oooo oooF) failed the test, and an error code
of FF (FFFF FFFF) would indicate that ALL bits failed the test.
1st 'x' 2nd 'x'
Bits 7 6 5 4 3 2 1 0
| | | | | | | |
| | | | | | | DATA READY flag
| | | | | | OVERRUN error
| | | | | PARITY error
| | | | FRAMING error
| | | |
| | | BREAK INTERRUPT detected
| | TRANSMITTER EMPTY
| TRANSMITTER SHIFT REGISTER EMPTY
not used
For example, an error code where the "xx" portion is 02 (oooo ooFo) indicates that bit 1
(OVERRUN error) failed.
XVIII.D.5.d.ii. Data Test:
(Error code "t" = 3) The error code indicates the data bit(s) in the internal circuitry that
failed. Where xx = 01 indicates that bit position 0 (oooo oooF) failed the test, and an error code
of FF (FFFF FFFF) would indicate that ALL bits failed the test.
1st 'x' 2nd 'x'
Bits 7 6 5 4 3 2 1 0
For example, an error code where the "xx" portion is 96 (FooF oFFo) indicates that bits 7, 4, 2,
and 1 failed.
NOTE: Refer to the Serial Interface Error Bit Chart in Section XVIII.D.5.e for the error code
values.
XVIII.D.5.d.iii. External/Internal Control Line Tests:
(Error code "t" = 4 or 5) Report the status of the looped-back signal by reading the modem
status register. Bits 4-7 are not used. Where xx = 01 indicates that bit position 0 (oooo oooF)
failed the test, and an error code of FF (FFFF FFFF) would indicate that ALL bits failed the test.
1st 'x' 2nd 'x'
Bits 7 6 5 4 3 2 1 0
|________| | | | |
| | | | CLEAR TO SEND (CTS) signal
Not Used | | DATA SET READY (DSR) signal
| RING INDICATOR (RI) signal
RECEIVER SIGNAL DETECTED (CD) signal
For example, an error code where the "xx" portion is 04 (oooo oFoo) indicates that bit 2 (RING
INDICATOR signal) failed.
NOTE: Refer to the Serial Interface Error Bit Chart in Section XVIII.D.5.e for the error code
values.
NOTE: This test requires that the appropriate Serial Interface TEST Plug be connected to the
interface being tested. If the TEST Plug is not installed, all of the tests will fail (xx = oF).
XVIII.D.5.e. Serial Interface Error Bit Chart:
BIT POSITION BIT POSITION
7 6 5 4 3 2 1 0
Value of: 1st 2nd
"x" "x"
0 = o o o o 0 = o o o o
1 = o o o F 1 = o o o F
2 = o o F o 2 = o o F o
3 = o o F F 3 = o o F F
4 = o F o o 4 = o F o o
5 = o F o F 5 = o F o F
6 = o F F o 6 = o F F o
7 = o F F F 7 = o F F F
8 = F o o o 8 = F o o o
9 = F o o F 9 = F o o F
A = F o F o A = F o F o
B = F o F F B = F o F F
C = F F o o C = F F o o
D = F F o F D = F F o F
E = F F F o E = F F F o
F = F F F F F = F F F F
o = passing (ok), or OFF status
F = failing, or ON status
For example, an "xx" error code of 02 would have the value of oooo ooFo and would indicate
that bit 1 was failing. An "xx" error code of 4A would be oFoo FoFo, indicating bits 6, 3, and 1
were failing.

XVIII.D.6. Memory Menu <Diagnostic Menu F7>

XVIII.D.6.a. Overview:
#1-TuffTEST-Pro's memory tests are divided into three areas: Main Memory, Extended
Memory, and Expanded Memory. The Main Memory tests appear on all computers and relate to
the first 640K of memory. The Extended Memory tests were introduced with the IBM PC AT-
level systems (80286) and exist on all subsequent systems. The Expanded Memory tests were
also introduced on the IBM PC AT and usually involve an add-in board with additional memory
that resides outside the standard memory map and is accessed through address-swapping
techniques. Expanded memory was phased out with the introduction of Intel i486-based PCs.
This memory testing documentation is arranged to show the common tests that appear in Main
Memory, Extended Memory, and Expanded Memory, with any unique memory testing
information indicated in the appropriate sections that follow for each type of memory.
There are many personal computer brands, makes, and models in the marketplace today, all
having their "unique" system board layout, and in some cases, BIOS modifications. This
#1-TuffTEST-Pro User Handbook is designed to identify the memory modules that are failing
based on industry standard systems and provide a general set of rules that assist in identifying the
failing module from the error code for systems that deviate from the standard systems.
As PC architecture evolved, so have memory modules. In the earlier PC, XT, AT types of PCs,
memory modules were the Dual Inline Package (DIP) type, the classic memory "chips." With
the introduction of PS/2, 80386, i486, Pentium and later PCs, newer memory designs were used,
including Single Inline Pin Package (SIPP), Single Inline Memory Module (SIMM), Dual Inline
Memory Module (DIMM), SDRAM and, more recently, Rambus RDRAM, and DDR SDRAM.
Since #1-TuffTEST-Pro's memory tests test memory directly by hexadecimal address through
the PC's CPU, it can be used to rigorously test ALL types and sizes of these memory
modules.
XVIII.D.6.b. Common Memory Tests:
The following memory tests verify memory and report the specific failing bits in addition to the
specific hexadecimal address at which the failure(s) occurred. All tests run continuously until
the user intervenes by pressing the <ESC> key.
NOTE: The test will not stop until it has completed the current test range. In a fast system with a
small amount of memory the test will stop almost instantaneously. However, in a slow system
with a large amount of memory it could take 5-20 minutes to stop.
Common memory menu selections that appear in Main, Extended, and Expanded Memory
subsections include:
XVIII.D.6.b.i. Abbreviated Memory Test:
All of the installed memory is subjected to four types of write/read/compare tests: ALLZEROS,
ALLONES, CHECKERBOARD, and ADDRESS.
XVIII.D.6.b.ii. Extensive Memory Test:
The diagnostic routines performed under this option are exactly the same as the Abbreviated
System Test, except for the addition of the MARCHING ONES test and the WALKING
ONES test. It may be used to burn-in a system (running it under stress for an extended period of
time to certify operation), to detect intermittent failures, or to certify newly installed/repaired
memory modules/boards.
XVIII.D.6.b.iii. ALLZEROS Memory Test:
In this test, zeros are written to all memory locations, followed by a read/compare
operation. When a memory location does not contain a zero, an error code is reported.
XVIII.D.6.b.iv. ALLONES Memory Test:
This test is functionally the same as the ALLZEROS Test except that ones are written to all
memory locations.
XVIII.D.6.b.v. CHECKERBOARD Memory Test:
This test alternately writes zeros and ones to successive memory locations and then performs
read/compare operations. If the memory location does not contain the correct value, an error
code is displayed.
NOTE: The ALLZEROS, ALLONES, and CHECKERBOARD tests all produce a data pattern
that naturally has an "even" parity, thus if the parity module is failing in a "high" state or is not
present, it would not be detected by these tests. If a parity module is suspect, use the ADDRESS
Memory Test to verify.
XVIII.D.6.b.vi. ADDRESS Memory Test:
This test generates data that produce approximately the same number of bytes with an even
number of ones as with an odd number of ones, generating greater stress conditions for
the parity generators.
XVIII.D.6.b.vii. MARCHING ONES Memory Test:
Using THREE different data patterns, this test writes to, reads from, and compares with all
specified memory locations. This test is more stringent than the above tests.
XVIII.D.6.b.viii. WALKING ONES Memory Test:
Using NINE different data patterns, this test writes to, reads from, and compares with all
specified memory locations. This is the most stringent memory test.
NOTE: The WALKING ONES Memory Test can take a considerable amount of time to
run. The length of the test depends upon the amount of memory being tested and the speed of
the computer. If a specific area of memory is suspect, testing time may be reduced by using the
Select Memory Testing Range option to limit the area of memory tested.
XVIII.D.6.b.ix. Select Memory Testing Range:
This function allows the user to select the memory address range to be tested.
One of the primary reasons for selecting an alternate memory testing range in Main Memory is
to permit the testing of the portion of memory in which #1-TuffTEST-Pro normally resides the
172K segment of memory between 24K (0602h) and 196K (3112h) and the reserved 21.5K
segment of memory between 2.5K (00A0h) and 24K (0602h).
NOTE: The 2.5K segment of memory between 0K (0000h) and 2.5K (00A0h) cannot be
tested. This area of memory contains the system's BIOS interrupt vectors and data tables.
NOTE: The 384K Upper Memory Area (UMA) between the top of Main Memory at 640K and
the start of Extended Memory at 1024K doesn't contain physical memory. Mapped into the 384K
UMA are the system BIOS, ROM chips and display adapter memory. When you install other
accessory cards, such as network adapters, video or disk controllers, they may also occupy space
within the 384K UMA. The 384K UMA is always located in the same area of the IBM
compatible computer's address space: from 640K to 1024K (A000 to FFFF hex). There are no
exceptions to this rule. This means that a standard IBM compatible PC with 640K of
conventional memory installed really has 1 MB (1024K) of address space. The system memory
occupies the first 640K, and the 384K UMA occupies the area from 640K to 1 MB (1024K).
This does not mean that the PC has 1 MB of memory. A PC with 1 MB of physical memory has
an address space of 1408K. This consists of the 640K of conventional memory, the 384K UMA,
and the 384K of extended memory starting at 1024K.
The primary reason for selecting an alternate memory testing range in Extended Memory is to
isolate a suspect subset of memory for more exhaustive testing. When Select Memory Testing
Range is chosen, a prompt asks for the beginning address. When the address has been entered, it
is automatically loaded into the Start portion of the test range. A prompt then asks for the ending
address. It is loaded into the End portion of the Test Range.
Addresses may be entered either in a hexadecimal ("h") paragraph format or in a decimal
kilobyte format.
XVIII.D.6.b.ix.a. Default Range (Main Memory):
This line indicates the default memory test range.
XVIII.D.6.b.ix.b. Alternate Range (Main Memory):
This line indicates the suggested alternate testing range [00A0h (2.5K) through 739Ch (462K)]
and the area of memory that #1-TuffTEST-Pro will relocate to if the suggested alternate range
is selected [739Dh (462K) through A000h (640K)].
NOTE: The actual address may vary slightly from system to system due to each
system's BIOS.
XVIII.D.6.b.ix.c. Test Range (Extended Memory):
This line indicates the actual testing range. When Select Memory Testing Range is first selected,
this line will be the same as the Default Range line. When an alternate range is selected, it will
be displayed on this line.
After selection of the memory test range, the type of memory test to be performed must be
selected (ALLZEROS, ALLONES, etc.).
XVIII.D.6.b.x. Display Error Log:
This selection displays the Error Log. See the Error Logging Menu section in Section XVIII.C
for complete information.
XVIII.D.6.c. Testing Variations in the Extended Memory Menu
XVIII.D.6.c.i. Test Compaq Reserved Memory:
The Compaq 386/20 series uses a memory mapping scheme that allows the memory area
between 640K and 1024K to be used as extended memory. This memory is mapped to the area
just prior to 16MB. Therefore, standard memory tests won't work. To test this memory,
select: K) Test Compaq Reserved Memory. The following tests are performed against this
reserved memory: ALLZEROS, ALLONES, CHECKERBOARD, ADDRESS, and
MARCHING ONES.
XVIII.D.6.d. Testing Variations in the Expanded Memory Menu
XVIII.D.6.d.i. Test Controller:
This test writes values to the page mapping register bits on the controller and then reads them
back. If discrepancies are detected between the write and read operations, an error is displayed.
XVIII.D.6.d.ii. Select Base Memory Segment:
Expanded memory uses one of the two available base memory segments between 640K and 1M
for data I/O between main and expanded memory. If unusual results occur while within
the expanded memory tests, the base memory segment that had not been tested previously
should be tested.
XVIII.D.6.d.iii. Select Expanded Memory Board:
The resulting error codes within individual expanded memory tests do not indicate which
memory board failed. Since #1-TuffTEST-Pro defaults to testing ALL installed memory
boards, a single memory board should be selected when individual expanded memory tests are
executed. This insures proper interpretation of any resulting error codes.
XVIII.D.6.e. Disable Cache Memory:
NOTE: Cache memory should not be disabled unless #1-TuffTEST-Pro is reporting memory
problems.
XVIII.D.6.e.i. First-Level Cache:
This is the high-speed cache imbedded in the microprocessor.
XVIII.D.6.e.ii. Second-Level Cache:
This is the high-speed cache imbedded in the motherboard.
Disable and Test Cache Memory:
The original PC design did not anticipate cache memory and therefore did not allocate memory
addresses for it. Thus, cache memory chip sets must must "borrow" memory addresses to
work. Due to the numerous "proprietary" memory address swapping schemes employed by the
cache memory chip set manufacturers, we know of no program that RELIABLY tests cache
memory DIRECTLY.
When cache memory is malfunctioning, symptoms may occur that appear to be failures in non-
cache memory. If memory failures disappear when cache memory is disabled, it is probable that
cache memory is the failing device.
The most RELIABLE way to test cache memory is to test it INDIRECTLY as follows:
1) Run the target PC with cache memory enabled and note the symptoms/error codes that you
think cache memory is causing.
2) Re-boot the target PC and go into its CMOS Setup program and disable Second-Level cache
memory.
3) Run the target PC and if the symptoms/error codes noted in step 1) go away, Second-Level
cache memory is bad. If they remain, something else is the cause.
4) Re-boot the target PC and go into its CMOS Setup program and enable Second-Level cache
memory. Then disable First-Level cache memory.
5) Run the target PC and if the symptoms/error codes noted in step 1) go away, First-Level
cache memory is bad. If they remain, something else is the cause.
Important: Whenever you disable or enable First- or Second-Level cache memory in CMOS
Setup, some systems will not incorporate the change until the next time the PC is started from a
cold boot. So make sure you save the change when you exit CMOS Setup and then shut down
the PC before it boots into the operating system. This will insure that the change takes effect the
next time you boot the PC.
XVIII.D.6.f. Finding Failing Memory Modules:
In order to find failing memory modules, it is important to first understand how various
microprocessor/bus combinations process data and use memory.
The following chart indicates the relationship between CPUs and the internal data path and bus
size.
CPU Overview:
CPU Internal Data Data Bus
or Compatible Path Size in Bits Size in Bits
8088 16 8
8086 16 16
80286 16 16
80386 SX 32 16
80386 SL 32 16
80386 DX 32 32
80486 SX 32 32
80486 DX 32 32
Pentium 32 64
Pentium Pro 32 64
Pentium II 32 64
Pentium III 32 64
Pentium 4 32 64
8-BIT SYSTEM:
In an 8-bit system memory is processed 1 byte (8 bits) at a time. A byte is equal to 8 data bits
(Bits 0-7) and 1 parity bit (P). Each of the 9 bits must be placed into a cell in a separate memory
module or DIP. Therefore, an 8-bit system with parity must have at least one Memory Bank
(Bank 0) composed of one physical row of 9 discrete memory modules or DIPs (see below). The
memory capacity of each module may vary so long as all 9 modules have the same capacity and
speed rating.
BANK 0 - MEMORY MODULES EXAMPLE - DIPs
XXX XXX XXX XXX XXX XXX XXX XXX XXX
0 1 2 3 4 5 6 7 P
BIT POSITIONS
16-BIT SYSTEM:
In a 16-bit system, memory is processed 2 bytes (16-bits) at a time. Therefore, a 16-bit system
must have at least one Memory Bank (Bank 0) composed of two 8-bit (30-pin) SIMMs or two
physical rows of nine DIPs each (see below). One byte is placed into each physical row. Thus
there is a LO BYTE (Row 0) and a HI BYTE (Row 1) within each bank.
BANK 0 - MEMORY MODULES EXAMPLE
ROW
0 XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX SIMMS
LO BYTE XXX XXX XXX XXX XXX XXX XXX XXX XXX DIPS
0 1 2 3 4 5 6 7 P
BIT POSITIONS
ROW
1 XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX SIMMS
HI BYTE XXX XXX XXX XXX XXX XXX XXX XXX XXX DIPS
8 9 10 11 12 13 14 15 P
BIT POSITIONS
32-BIT SYSTEM:
In a 32-bit system, memory is processed 4 bytes (32 bits) at a time. Therefore, a 32-bit system
must have at least one Memory Bank (Bank 0) composed of four 8-bit (30-pin) SIMMS or one
32-bit (72-pin) SIMM or four physical rows of nine DIPs each (see below). One byte is placed
into each physical row. Thus there is a LO WORD/LO BYTE (Row 0), a LO WORD/HI BYTE
(Row 1), a HI WORD/LO BYTE (Row 2), and a HI WORD/HI BYTE (Row 3) within each
bank.
NOTE: One 32-bit SIMM equates to the four 8-bit SIMMS shown in Rows 0, 1, 2 and 3.
BANK 0 - MEMORY MODULES EXAMPLE
ROW
0 XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX SIMMS
LO WORD, XXX XXX XXX XXX XXX XXX XXX XXX XXX DIPS
LO BYTE 0 1 2 3 4 5 6 7 P
BIT POSITIONS
ROW
1 XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX SIMMS
LO WORD, XXX XXX XXX XXX XXX XXX XXX XXX XXX DIPS
HI BYTE 8 9 10 11 12 13 14 15 P
BIT POSITIONS
ROW
2 XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX SIMMS
HI WORD, XXX XXX XXX XXX XXX XXX XXX XXX XXX DIPS
LO BYTE 16 17 18 19 20 21 22 23 P
BIT POSITIONS
ROW
3 XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX SIMMS
HI WORD, XXX XXX XXX XXX XXX XXX XXX XXX XXX DIPS
HI BYTE 24 25 26 27 28 29 30 31 P
BIT POSITIONS
64-BIT SYSTEM:
In a 64-bit system, memory is processed 8 bytes (64 bits) at a time. Therefore, a 64-bit system
must have at least one Memory Bank (Bank 0) composed of two 32-bit SIMMs or one 64-bit
DIMM. In 64-bit systems the Bit Failure Packet scheme (See Section XVIII.D.6.g.iii for Bit
Failure Packet explanation) is not used, and in its place the actual address of the failing bit(s) is
displayed.
XVIII.D.6.g. Finding Bit 0, Bank 0, and Parity:
Regardless of the size or configuration of the memory modules, it is always important to
locate Bit 0 of Bank 0 as a point of reference. While the concept now is somewhat inappropriate
due to the advances in technology, the original IBM PC, which had 16K memory modules,
reserved Bank 0 (the first 16K) to be used for the vector tables and other information that would
be loaded at power-on start-up time by the system's BIOS and Power-On Self Test. While in
reality the information only occupies a small portion of the 16K bank size, this was a soldered
versus socketed bank (physical row) to ensure its integrity. Today with 256K and 512K DIPs
and larger SIMMs/DIMMs, the same Bank 0 approach has been preserved. Regardless of the
current technical viability of the original approach, it is the norm today that if Bank 0 is disabled,
then the computer will not operate beyond the Power-On Self Test.
On a system board or an add-in board, some manufacturers have screen-printed orientation
information on the board. With those that have the bank and module position screen-printed, it is
fairly easy to find the first bit position (Bit 0) of Bank 0. For those that do not, a process of
elimination in conjunction with the #1-TuffTEST-Pro error codes will clearly identify Bit 0 of
Bank 0, as well as the location of the parity bit. Note in the following section that in some cases
the parity bit precedes Bit 0 and in others it follows Bit 7.
XVIII.D.6.g.i. Memory Module Orientation - 8-Bit Systems:
In most cases, 8-bit systems from different manufacturers do not have the same memory module
orientation. The Parity bit and the 0 or 7 bit may be on either end of the row.
To determine the physical layout of an 8-bit system's DIP memory modules:
1) Remove the first and second DIP modules in any bank except Bank 0.
MODULES XXX XXX XXX XXX XXX XXX XXX XXX XXX DIPS
| |
| |
Remove these 2 DIP modules.
NOTE: If there is more than one row of DIPs and you accidentally remove DIPs in Bank 0, the
computer will not operate.
2) Load #1-TuffTEST-Pro and select Diagnostic Menu <F5>. Within this menu select the
option Memory Test <F7>. Run one of the memory tests and an error code will be
generated. See Section XVIII.D.6.h for error code interpretation. Use the xxP Bit Failure
Packet portion of the error code and the Memory Bit Position Failure Chart in Section
XVIII.D.6.i to identify the failed bits. See Section XVIII.D.6.g.iii for Bit Failure Packet
explanation.
If the xxP error code is 01P (equal to oooo oooF on the Chart), the Bit-7 module and the Parity
Bit module failed. Therefore, the DIP memory module layout in the row would be the parity bit
module followed by the Bit-7 module.
MODULES XXX XXX XXX XXX XXX XXX XXX XXX XXX
P 7 6 5 4 3 2 1 0
BIT POSITION
If the xxP error code is 80P (equal to Fooo oooo on the Chart), the Bit-0 module and the Parity
Bit module failed. Therefore, the DIP memory module layout in the row would be the parity bit
module followed by the Bit-0 module.
MODULES XXX XXX XXX XXX XXX XXX XXX XXX XXX
P 0 1 2 3 4 5 6 7
BIT POSITION
If the xxP error code is C0* (equal to FFoo oooo on the Chart), the Bit-0 and Bit-1 modules
failed and the Parity Bit module passed. Therefore, the DIP memory module layout in the row
would be the Bit-0 module followed by the Bit-1 module with the parity bit module to the far
right of that physical row.
MODULES XXX XXX XXX XXX XXX XXX XXX XXX XXX
0 1 2 3 4 5 6 7 P
BIT POSITION
If the xxP error code is 03* (equal to oooo ooFF on the Chart), the Bit-7 and Bit-6 modules
failed and the Parity Bit module passed. Therefore, the DIP memory module layout in the row
would be the Bit-7 module followed by the Bit-6 module with the parity bit module to the far
right of that physical row.
MODULES XXX XXX XXX XXX XXX XXX XXX XXX XXX
7 6 5 4 3 2 1 0 P
BIT POSITION
NOTE: Some systems have 18 modules in a row versus the standard 9 (parity plus 8 bits or 8
bits plus parity). The same decoding technique as above can be used, but you must visually note
the end of the first bank and the beginning of the second bank within the physical row.
NOTE: Later IBM XT and compatible systems and later IBM AT and compatible systems,
contain larger and perhaps mixed size memory modules. In those cases you first need to use #1-
TuffTEST-Pro to determine the size of the main memory, which is usually 640K, then through
simple mathematics and visual inspection of the main memory area on the system board,
determine how many rows there are, and through simple division determine which rows are
combined into what banks. Then using the same approach of pulling two modules beside each
other, typically as a starting point in the upper left-hand corner, you can determine whether it's
Bank 0 (the computer will not operate). Then move on to the next row below it to determine
what bits are failing and with the same logic as above, determine the bank and row. Through the
process of elimination, the physical layout of the memory modules can be mapped.
XVIII.D.6.g.ii. Memory Module Orientation - 16-Bit, 32-Bit and 64-Bit Systems
A 16-bit system must have at least two 8-bit (30-pin) SIMMs or two physical rows of nine DIPs
each (LO BYTE and HI BYTE), but can have more rows, always in sets of two. The
module/SIMM size (64K, 256K, 512K, 1MEG, etc.) does not matter. If it only has two 8-bit
SIMMs or nine DIP rows, those will be Bank 0.
Earlier 32-bit systems had four 8-bit (30-pin) SIMMs. If the system has only four 8-bit SIMMs,
then those constitute Bank 0. If there are more than four, then the next set of four would
represent Bank 1, and so on. More recent systems have 32-bit (72-pin) SIMMs. If the system
has only one 32-bit SIMM, then it constitutes Bank 0. If there is more than one, then the next
one would represent Bank 1, and so on.
A 64-bit system must have at least two 32-bit SIMMs or one 64-bit DIMM, but can have more,
always in sets of two or one respectively. If it only has two 32-bit SIMMs, or one 64-bit DIMM,
those will be Bank 0.
NOTE: If the system has an interleaved memory board, the CMOS interleave value should be
set to OFF, or disabled. And if the system has cache memory, that also should be set to OFF, or
disabled, to effectively use the procedure to find the failing bit in the failing memory module,
thus identifying which physical module is to be replaced.
NOTE: It is important, through physical examination of the printing/coding on the module, to
determine the speed of the memory modules installed in the system. The modules' speed within
Memory Banks should always be identical otherwise problems could occur. When you must use
different speed memory modules in different Memory Banks, always put the slower modules in
Bank 0 because the CPU uses BANK 0 to calibrate the system's memory speed. A 120 NS speed
is relatively slow, where an 70 NS or faster speed is a more efficient module.
It is assumed that if a memory module is removed from Bank 0, the system will not
operate. Thus by using a module removal approach similar to what was described in the prior
section regarding DIP-type memory modules, Bank 0 can be located in systems with more than
one Memory Bank. However, it is possible to have a bad bit in Bank 0 (within the
SIMMs/DIMMs) and the system may continue to operate. In general this would indicate that the
bad bit is above the 2.5K that the BIOS/Power-On Self Test verifies for its reserved information
loading.
If more than two 8-bit SIMMs are observed in a 16-bit system, or more than four 8-bit SIMMs or
one 32-bit SIMM in a 32-bit system, or more than two 32-bit SIMMs or one 64-bit DIMM in a
64-bit system, then there is more than one Memory Bank and a procedure must be used to
determine which is Bank 0.
Use the motherboard's documentation to determine the location of Bank 0. If documentation is
not available check the motherboard for silk-screened legends depicting Bank 0, Bank 1, etc. If
there are no silk-screen identifiers, pull the SIMM(s) from the Memory Bank closest to the
front the system, as a point of reference, to determine which is Bank 0. When the SIMM(s) are
removed from Bank 0 the system will not boot. If the system still boots, pull the SIMM(s) from
the next Memory Bank. Repeat this procedure until the system stops booting.
Locating Failing Memory Module(s)
In systems with more than one Memory Bank, use the error code address to determine which
Bank(s) contains the failing memory module(s). Then use SIMM swapping techniques to
determine which memory module is bad.
Important: In a 16-bit system with two 8-bit SIMMs in a Memory Bank, a 32-bit system with
four 8-bit SIMMs in a Memory Bank or a 64-bit system with two 32-bit SIMMs in a Memory
Bank, if the system operates and an error is detected, reverse the SIMMs within the Bank. If the
error code address moves, the error code indicates a failing SIMM. If the error code address
does not move, the error code indicates a failing socket or system board.
XVIII.D.6.g.iii. "Bit Failure Packet" Decoding Examples:
The xxP portion of the error code is referred to as the "Bit Failure Packet." Each Bit Failure
Packet represents one byte composed of eight data bits, xx, and one parity bit, P. An 8-bit
computer would have one Bit Failure Packet, a 16-bit computer would have two Bit Failure
Packets, and a 32-bit computer would have four Bit Failure Packets.
NOTE: All #1-TuffTEST-Pro memory error code Bit Failure Packets are decoded via the
Memory Bit Position Failure Chart in Section XVIII.D.6.i. The Memory Bit Position Failure
Chart is a hexadecimal/binary translation chart depicted in such a way as to not burden the user
in trying to understand the binary relationship to the failing bit or in performing hexadecimal
math calculations.
An 8-bit system with the error code xxP = 28* (equal to ooFo Fooo on the chart) would indicate
that bits 2 and 4 failed and the parity bit passed (* equals parity passed and P equals parity
failed). The following graphic indicates the physical bit layout of the modules (DIPs) with their
bits indicated below and the "xx" error code status indicated above.
1st "x" 2nd "x" 1st P






xxP VALUE o o F o F o o o *
DIPs XXX XXX XXX XXX XXX XXX XXX XXX XXX
BIT POSITION 0 1 2 3 4 5 6 7 P
A 16-bit system with the error code xxPxxP = 40*22P (equal to oFoo oooo ooFo ooFo on the
chart) indicates that the LO BYTE bit 1 and HI BYTE bits 10 and 14 and the second parity bit
failed.
1st "x" 2nd "x" 1st P






LO xxP VALUE o F o o o o o o *
BYTE DIPs XXX XXX XXX XXX XXX XXX XXX XXX XXX
BIT POSITION 0 1 2 3 4 5 6 7 P
3rd "x" 4th "x" 2nd P






HI xxP VALUE o o F o o o F o P
BYTE DIPs XXX XXX XXX XXX XXX XXX XXX XXX XXX
BIT POSITION 8 9 10 11 12 13 14 15 P
LO SIMM XXXXXXXXXXXXX bit 1 failed 1st xxP(40*)
BYTE BITS 0 7
HI SIMM XXXXXXXXXXXXX bits 10, 14 and 2nd xxP(22P)
BYTE BITS 8 15 parity failed
A 32-bit system with the error code xxPxxPxxPxxP = 00*01P08*00* (equal to oooo oooo oooo
oooF oooo Fooo oooo oooo on the Chart) indicates that the LO WORD/HI BYTE bit 15 and the
parity bit failed and HI WORD/LO BYTE bit 20 failed.
LO SIMM XXXXXXXXXXXXX OK 1st xxP(00*)
LO BYTE BITS 0 7
WORD
HI SIMM XXXXXXXXXXXXX bit 15 and 2nd xxP(01P)
BYTE BITS 8 15 parity failed
LO SIMM XXXXXXXXXXXXX bit 20 failed 3rd xxP(08*)
HI BYTE BITS 16 23
WORD
HI SIMM XXXXXXXXXXXXX OK 4th xxP(00*)
BYTE BITS 24 31
A 64-bit system does not use the Bit Packet orientation as the error code contains the actual
address of the failure.
XVIII.D.6.h. Memory Error Codes:
Due to the architectural differences of personal computer types and #1-TuffTEST-Pro's testing
precision, there are slight variations of the basic memory error code format for the various
computer types as follows.
XVIII.D.6.h.i. Main Memory - 8088- and 8086-based Computers:
The Main Memory error code format is MTt-BbxxP (in Certification and Diagnostic
Abbreviated and Extended Tests) or MTt-BbaaaaxxP (in Error Logging), where:
M = Main Memory
T = Test type where:
A = Abbreviated
E = Extensive
S = Single
t = Test number of specific test where:
1 = ALLZEROS
2 = ALLONES
3 = CHECKERBOARD
4 = ADDRESS
5 = MARCHING ONES
6 = WALKING ONES
B = Bank number (64K)
b = Block number (16K) within bank
aaaa = Address within bank, 0000 -FFFF (Error Logging)
xxP = Bit failure packet where:
xx = Failed bits
P = Letter P for parity error (asterisk [*] if no parity error)
NOTE: See Memory Bit Position Failure Chart, Section XVIII.D.6.i, to decode xxP Bit Failure
Packets.
XVIII.D.6.h.ii. Main Memory - 80286-based Computers (ISA):
Main Memory error code format is MTt-BxxPxxP where:
M = Main Memory
T = Test type where:
A = Abbreviated
E = Extensive
S = Single
t = Test number of specific test where:
1 = ALLZEROS
2 = ALLONES
3 = CHECKERBOARD
4 = ADDRESS
5 = MARCHING ONES
6 = WALKING ONES
B = Bank number (256K banks)
xxP = Bit failure packet for each xxP, where:
xx = Failed bits
P = Letter P for parity error (asterisk [*] if no parity error)
NOTE: See Memory Bit Position Failure Chart, Section XVIII.D.6.i, to decode xxP Bit Failure
Packets.
XVIII.D.6.h.iii. Main Memory - 80386- and i486-based Computers (ISA/EISA/VLB/PCI):
Main Memory error code format is MTt-xxPxxPxxPxxP, where:
M = Main Memory
T = Test type where:
A = Abbreviated
E = Extensive
S = Single
t = Test number of specific test where:
1 = ALLZEROS
2 = ALLONES
3 = CHECKERBOARD
4 = ADDRESS
5 = MARCHING ONES
6 = WALKING ONES
xxP = Bit failure packet for each xxP, where:
xx = Failed bits
P = Letter P for parity error (asterisk [*] if no parity error)
NOTE: The Bit Failure Packets in the 80386SX error code are redundant. The first pair of Bit
Failure Packets display exactly the same information as the second pair.
NOTE: See Memory Bit Position Failure Chart, Section XVIII.D.6.i, to decode xxP Bit Failure
Packets.
XVIII.D.6.h.iv. Main Memory - Pentium through Pentium 4-based Computers
(ISA/EISA/VLB/PCI):
Main Memory error code format is MTt-f-aaaaaaaa where:
M = Main Memory
T = Test type where:
A = Abbreviated
E = Extensive
S = Single
t = Test number of specific test where:
1 = ALLZEROS
2 = ALLONES
3 = CHECKERBOARD
4 = ADDRESS
5 = MARCHING ONES
6 = WALKING ONES
f = Type of failure where:
Bit = only a data bit failed ("B" in Error Log)
Parity = only a parity bit failed ("P" in Error Log)
Bit&Parity = both a data bit and a parity bit failed
("B&P" in Error Log)
aaaaaaaa = address of failing data bit
XVIII.D.6.h.v. Main Memory - IBM PS/2 Models, General:
The error codes report significant quantities of data regarding locations of errors. Models of the
PS/2 family, however, provide no opportunity for specific chip replacement (except PS/2 Model
25) because of the surface-mounted technology used and the SIMM packages. #1-PC
Diagnostics Company, nevertheless, decided to retain the detailed error reporting in anticipation
of PS/2 compatibles entering the market using potentially different memory configurations.
XVIII.D.6.h.vi. Main Memory - IBM PS/2 Models 25 and 30:
The Main Memory error code format is MTt-BbxxP (in Certification and Diagnostic) or MTt-
BbaaaaxxP (in Error Logging), where:
M = Main Memory
T = Test type where:
A = Abbreviated
E = Extensive
S = Single
t = Test number of specific test where:
1 = ALLZEROS
2 = ALLONES
3 = CHECKERBOARD
4 = ADDRESS
5 = MARCHING ONES
6 = WALKING ONES
B = Bank number (64K)
b = Block number (16K) within bank
aaaa = Address within bank, 0000-FFFF (Error Logging)
xxP = Bit failure packet, where:
xx = Failed bits
P = Letter P for parity error (asterisk [*] if no parity error)
NOTE: See Memory Bit Position Failure Chart, Section XVIII.D.6.i, to decode xxP Bit Failure
Packets.
XVIII.D.6.h.vii. Main Memory - IBM PS/2 Models 30-286, 50, 50Z, and 60:
Main Memory error code format is MTt-BxxPxxP, where:
M = Main Memory
T = Test type where:
A = Abbreviated
E = Extensive
S = Single
t = Test number of specific test where:
1 = ALLZEROS
2 = ALLONES
3 = CHECKERBOARD
4 = ADDRESS
5 = MARCHING ONES
6 = WALKING ONES
xxP = Bit failure packet for each xxP, where:
xx = Failed bits
P = Letter P for parity error (asterisk [*] if no parity error)
NOTE: See Memory Bit Position Failure Chart, Section XVIII.D.6.i, to decode xxP Bit Failure
Packet.
XVIII.D.6.h.viii. Main Memory -IBM PS/2 Models 55SX, 70, 80, and 90:
Main Memory error code format is MTt-xxPxxPxxPxxP where:
M = Main Memory
T = Test type where:
A = Abbreviated
E = Extensive
S = Single
t = Test number of specific test where:
1 = ALLZEROS
2 = ALLONES
3 = CHECKERBOARD
4 = ADDRESS
5 = MARCHING ONES
6 = WALKING ONES
xxP = Bit failure packet for each xxP, where:
xx = Failed bits
P = Letter P for parity error (asterisk [*] if no parity error)
NOTE: See Memory Bit Position Failure Chart, Section XVIII.D.6.i, to decode xxP Bit Failure
Packets.
XVIII.D.6.h.ix. Extended Memory -80286-based Computers (ISA, MCA):
Extended Memory error code format is Tat-BbxxPxxP where:
T = Extended Memory error
a = Test type where:
A = Abbreviated
E = Extensive
S = Single
t = Test number of specific test where:
1 = ALLZEROS
2 = ALLONES
3 = CHECKERBOARD
4 = ADDRESS
5 = MARCHING ONES
6 = WALKING ONES
B = Which 1MB block where the error occurred, where:
1 = between 1 and 2MB
2 = between 2 and 3MB, etc.
b = Which 64K bank (0 -Eh) within the 1MB block where the error
occurred, where:
0 = 0 - 64K
1 = 65 - 128K
2 = 129 - 192K
3 = 193 - 256K
4 = 257 - 320K
5 = 321 - 384K
6 = 385 - 448K
7 = 449 - 512K
8 = 513 - 576K
9 = 577 - 640K
A = 641 - 704K
B = 705 - 768K
C = 769 - 832K
D = 833 - 896K
E = 897 - 960K
F = 961 - 1024K
xxP = Bit failure packet for each xxP, where:
xx = Failed bits
P = Letter P for parity error (asterisk [*] if no parity error)
NOTE: See Memory Bit Position Failure Chart, Section XVIII.D.6.i, to decode xxP Bit Failure
Packets.
The bank in which the error occurred is reported on the display. When combined with the bit
failure packet information, the exact failing memory module can be isolated.
XVIII.D.6.h.x. Extended Memory - 80386- and i486-based Computers
(ISA/EISA/MCA/VLB/PCI):
Extended Memory error code format is either Tat-xxPxxPxxPxxPaaaaaaaa (most computers)
or Tat-32xxPxxPxxPxxP (Acer system boards) where:
T = Extended Memory
a = Test type where:
A = Abbreviated
E = Extensive
S = Single
t = Test number of specific test where:
1 = ALLZEROS
2 = ALLONES
3 = CHECKERBOARD
4 = ADDRESS
5 = MARCHING ONES
6 = WALKING ONES
3 = State of A3 when error was detected (Acer only)
2 = State of A2 when error was detected (Acer only)
xxP = Bit failure packet for each xxP, where:
xx = Failed bits
P = Letter P for parity error (asterisk [*] if no parity error)
aaaaaaaa = Address
NOTE: The Bit Failure Packets in the 80386SX error code are redundant. The first pair of Bit
Failure Packets display exactly the same information as the second pair.
On an 80386 Acer system board, if A2 is 0, a SIMM in an ODD-numbered socket is bad. If A2
is 1, a SIMM in an EVEN-numbered socket is bad. If A2 is B, ALL SIMMs are bad. On an
80486 Acer system board, use the following truth table:
Failing SIMM
A3 A2 Module(s)
0 0 = 1 or 5
0 1 = 2 or 6
0 B = 1 and 2, or 5 and 6
1 0 = 3 or 7
1 1 = 4 or 8
1 B = 3 and 4, or 7 and 8
B 0 = 1 and 3, or 5 and 7
B 1 = 2 and 4, or 6 and 8
B B = ALL SIMMs BAD
NOTE: See Memory Bit Position Failure Chart, Section XVIII.D.6.i, decode xxP Bit Failure
Packets.
XVIII.D.6.h.xi. Extended Memory - Pentium through Pentium 4-based Computers
(ISA/EISA/VLB/PCI):
Extended Memory error code format is TAt-f-aaaaaaaa where:
T = Extended Memory
A = Test type where:
A = Abbreviated
E = Extensive
S = Single
t = Test number of specific test where:
1 = ALLZEROS
2 = ALLONES
3 = CHECKERBOARD
4 = ADDRESS
5 = MARCHING ONES
6 = WALKING ONES
f = Type of failure where:
Bit = only a data bit failed ("B" in Error Log)
Parity = only a parity bit failed ("P" in Error Log)
Bit&Parity = both a data bit and a parity bit failed
("B&P" in Error Log)
aaaaaaaa = address of failing data bit
XVIII.D.6.h.xii. Expanded Memory - IBM PC, XT, PS/2 25 and 30:
Expanded Memory error code format is either Xat-BbPxxpppp (in Certification and Diagnostic)
or Xat-BbaaaaPxx (in Error Logging), where:
X = Expanded Memory error
a = Test type where:
A = Abbreviated
E = Extensive
S = Single
C = Certification
t = Test number of specific test where:
1 = ALLZEROS
2 = ALLONES
3 = CHECKERBOARD
4 = ADDRESS
5 = MARCHING ONES
6 = WALKING ONES
B = Bank the error occurred in (0 -7)
b = Block (16K) of bank error occurred in (0 -F)
aaaa = Upper byte of address of error (Error Logging)
Pxx = Bit failure packet where:
xx = Failed bits
P = Letter P for parity error (asterisk [*] if no parity error)
pppp = Base address in hex (Certification and Diagnostic)
NOTE: See Memory Bit Position Failure Chart, Section XVIII.D.6.i, to decode xxP Bit Failure
Packets.
XVIII.D.6.h.xiii. Expanded Memory - 80286-, 80386- and i486-based Computers
(ISA/EISA):
Expanded Memory error code format is Xat-BxxPpppp where:
X = Expanded Memory error
a = Test type where:
A = Abbreviated
E = Extensive
S = Single
C = Certification
t = Test number of specific test where:
1 = ALLZEROS
2 = ALLONES
3 = CHECKERBOARD
4 = ADDRESS
5 = MARCHING ONES
6 = WALKING ONES
B = Bank in which the error occurred (0 -7)
xxP = Bit failure packet where:
xx = Failed bits
P = Letter P for parity error (asterisk [*] if no parity error)
ppp = Base address in hex (Certification and Diagnostic)
NOTE: See Memory Bit Position Failure Chart, Section XVIII.D.6.i, to decode xxP Bit Failure
Packets.
XVIII.D.6.i. Memory Bit Position Failure Chart:
Decode the "xxP" portion of the Memory Error Code by computer type to determine failing
bit(s), then bank and block decoding to locate the failing memory module(s).
Value of "xx": o = Passing Bit (OK)
F = Failing Bit
Value of "P": * = No Parity Error
P = Parity Error
SYSTEM
TYPE BIT POSITION BIT POSITION
8-BIT:
BYTE 0 1 2 3 4 5 6 7
16-BIT:
HI BYTE 8 9 10 11 12 13 14 15
LO BYTE 0 1 2 3 4 5 6 7
32-BIT:
HI WORD
HI BYTE 24 25 26 27 28 29 30 31
LO BYTE 16 17 18 19 20 21 22 23
LO WORD
HI BYTE 8 9 10 11 12 13 14 15
LO BYTE 0 1 2 3 4 5 6 7
_________________ _________________
Value of: 1st 2nd
"x" "x"
0 = o o o o 0 = o o o o
1 = o o o F 1 = o o o F
2 = o o F o 2 = o o F o
3 = o o F F 3 = o o F F
4 = o F o o 4 = o F o o
5 = o F o F 5 = o F o F
6 = o F F o 6 = o F F o
7 = o F F F 7 = o F F F
8 = F o o o 8 = F o o o
9 = F o o F 9 = F o o F
A = F o F o A = F o F o
B = F o F F B = F o F F
C = F F o o C = F F o o
D = F F o F D = F F o F
E = F F F o E = F F F o
F = F F F F F = F F F F

XVIII.E. Help Menu <Any Menu F10>

Help screens in #1-TuffTEST-Pro are available from most menus by pressing the <F10>
key. When Help is selected, the following choices are displayed for selection:
(a) How #1-TuffTEST-Pro Works
(b) When to Use #1-TuffTEST-Pro
(c) How to Move from Menu to Menu
(d) How to Select Functions within a Menu

XVIV. Licensing Information

#1-TuffTEST-Pro Professional Level PC Hardware Diagnostic Software product, copyright
1984-2002, #1-PC Diagnostics Company All rights reserved. The #1-TuffTEST-Pro computer
programs and User Handbook are the property of #1-PC Diagnostics Company and are protected
by copyright and intellectual property laws. #1-TuffTEST-Pro is a trademark of #1-PC
Diagnostics Company. All other brand and product names are trademarks or registered
trademarks of their respective companies.
The #1-TuffTEST-Pro computer program(s) and User Handbook are licensed to the registered
user and may not be copied (other than the one and only copy currently being used), reproduced,
disclosed, transferred, or reduced to any electronic medium or machine-readable form without
the express written approval of #1-PC Diagnostics Company.
#1-PC Diagnostics Company MAKES NO WARRANTIES, EXPRESSED OR IMPLIED,
RELATING TO THE #1-TuffTEST-Pro COMPUTER PROGRAMS, User Handbook, OR
TEST Plugs, INCLUDING BUT NOT LIMITED TO IMPLIED WARRANTIES OF
MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE, AND ALL SUCH
WARRANTIES ARE EXPRESSLY AND SPECIFICALLY DISCLAIMED. IN NO EVENT
SHALL #1-PC Diagnostics Company BE RESPONSIBLE FOR ANY INDIRECT OR
CONSEQUENTIAL DAMAGES OR LOST PROFITS, EVEN IF #1-PC Diagnostics Company
HAS BEEN ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
Unless otherwise specifically stated, #1-PC Diagnostics Company warrants the #1-TuffTEST-
Pro product (computer programs and User Handbook, and TEST Plugs if purchased from #1-PC
Diagnostics Company) to be free from any defects in materials or faulty workmanship, in normal
use and service, for a period of thirty (30) days from the date of its licensing. #1-PC Diagnostics
Company shall not be responsible for any damage caused to the product by the purchaser.