BurnInTest Users Guide

PassMark
Software
BurnInTest
Users Manual
for version 5.0 of BurnInTest
Document Edition: 5.0
Date:
17 /February / 2006
Feedback:
While every effort has been made to produce an accurate and comprehensive document,
nobody is perfect; please send corrections and suggestions to help@passmark.com
BIT users manual
Page 1
Table of Contents
OVERVIEW.......................................................................................................................................................... 4
GETTING STARTED .......................................................................................................................................... 6
TEST DESCRIPTIONS ....................................................................................................................................... 7
MATHS TESTS ...................................................................................................................................................... 7
2D GRAPHICS TESTS – VIDEO MEMORY ................................................................................................................ 7
2D GRAPHICS TESTS – GRAPHICS FUNCTIONS ...................................................................................................... 8
MONITOR EMC TESTING ..................................................................................................................................... 8
HARD DISK AND FLOPPY DISK TESTS ................................................................................................................... 8
CPU MMX & SSE CPU TESTS .......................................................................................................................... 10
3D GRAPHICS TEST ............................................................................................................................................ 11
PRINTER TEST .................................................................................................................................................... 11
CD ROM, CD-RW BURN AND DVD TEST ......................................................................................................... 12
DON’T TEST THE DRIVE SELECTED. .................................................................................................................... 12
NETWORK TEST.................................................................................................................................................. 14
ADVANCED NETWORK TEST .............................................................................................................................. 16
MEMORY (RAM) TEST....................................................................................................................................... 16
ADDRESSING WINDOWS EXTENSION (AWE) MEMORY TEST ............................................................................. 17
SOUND CARD TEST ............................................................................................................................................. 17
VIDEO PLAYBACK TEST ..................................................................................................................................... 18
SERIAL PORT TEST (BURNINTEST PRO ONLY) .................................................................................................... 19
PARALLEL PORT TEST (BURNINTEST PRO ONLY) ............................................................................................... 20
USB PORT TEST (BURNINTEST PRO ONLY) ........................................................................................................ 21
TAPE DRIVE TEST (BURNINTEST PRO ONLY)...................................................................................................... 23
PLUGIN TEST (BURNINTEST PRO ONLY) ............................................................................................................ 24
TEST PREFERENCES ...................................................................................................................................... 25
DISK PREFERENCES ............................................................................................................................................ 25
CD-RW/DVD PREFERENCES ............................................................................................................................. 26
LOGGING PREFERENCES ..................................................................................................................................... 28
ERROR HANDLING PREFERENCES ....................................................................................................................... 29
NETWORK PREFERENCES ................................................................................................................................... 29
ADVANCED NETWORK TEST OPTIONS ............................................................................................................... 30
TEMPERATURE AND BATTERY MONITORING PREFERENCES ............................................................................... 30
PARALLEL PORT PREFERENCES (BURNINTEST PRO ONLY) ................................................................................ 31
2D GRAPHICS TEST / EMC TESTING (BURNINTEST PRO ONLY) ........................................................................ 31
USB PORT PREFERENCES (PRO VERSION ONLY) ................................................................................................. 31
RAM PREFERENCES ........................................................................................................................................... 32
VIDEO PLAYBACK PREFERENCES........................................................................................................................ 33
TAPE PREFERENCES (BURNINTEST PRO ONLY) .................................................................................................. 33
CPU PREFERENCES (BURNINTEST PRO ONLY)................................................................................................... 34
SERIAL PORT PREFERENCES (BURNINTEST PRO ONLY) ...................................................................................... 34
SOUND PREFERENCES......................................................................................................................................... 35
MATHS PREFERENCES ........................................................................................................................................ 35
PRINT PREFERENCES .......................................................................................................................................... 35
PRE-TEST PREFERENCES ..................................................................................................................................... 36
POST-TEST PREFERENCES .................................................................................................................................. 36
TEST PREFERENCES – “PLUGIN” ........................................................................................................................ 38
AUTOMATIC TEST STOP, TEST SELECTION & DUTY CYCLES......................................................... 40
AUTO-STOP ....................................................................................................................................................... 40
TEST CHECK BOXES ........................................................................................................................................... 41
SLIDE BARS & DUTY CYCLE............................................................................................................................... 41
BUTTONS ........................................................................................................................................................... 41
SELECTING WHICH TESTS TO RUN ...................................................................................................................... 41
INTERPRETING THE RESULTS ................................................................................................................... 43
BIT users manual
Page 2
STATUS DISPLAY ................................................................................................................................................ 43
TEST WINDOW DISPLAY ..................................................................................................................................... 45
DETAILED ERROR LOG HISTORY ........................................................................................................................ 46
MENU OPTIONS ............................................................................................................................................... 48
FILE ................................................................................................................................................................... 48
EDIT ................................................................................................................................................................... 49
VIEW.................................................................................................................................................................. 49
CONFIGURATION ................................................................................................................................................ 49
TEST .................................................................................................................................................................. 49
HELP .................................................................................................................................................................. 50
ADVANCED OPTIONS..................................................................................................................................... 51
MAINTAINING MULTIPLE TEST CONFIGURATIONS .............................................................................................. 51
COMMAND LINE ARGUMENTS ............................................................................................................................ 51
EXAMPLE 1 ........................................................................................................................................................ 52
EXAMPLE 2 ........................................................................................................................................................ 52
TEMPERATURE AND BATTERY MONITORING ...................................................................................................... 52
REPORT INFORMATION (MACHINE IDENTITY AND ADDITIONAL NOTES)............................................................. 54
PREPARING A CUSTOMER TEST CERTIFICATE .................................................................................................... 54
REBOOT / RESTART CYCLING ............................................................................................................................ 55
SCRIPTING.......................................................................................................................................................... 56
OTHER INFORMATION ................................................................................................................................. 62
CONTACTING PASSMARK SOFTWARE ................................................................................................................ 62
SYSTEM REQUIREMENTS .................................................................................................................................... 62
PRECAUTIONS FOR THOROUGH AND CAREFUL TESTING ...................................................................................... 63
SETTING THE WINDOWS ‘LOCK PAGES IN MEMORY’ RIGHT .............................................................................. 64
APPENDIX A – EXAMPLE ASCII LOG FILE .............................................................................................. 66
APPENDIX B – FAULT FINDING................................................................................................................... 68
APPENDIX C – COMMON ERROR MESSAGES......................................................................................... 69
APPENDIX D – LOOPBACK PLUGS ............................................................................................................. 77
SERIAL PORT LOOPBACK PLUG ........................................................................................................................... 77
PARALLEL PORT LOOP BACK PLUG ..................................................................................................................... 78
AUDIO LOOPBACK CABLE .................................................................................................................................. 79
USB1 PORT LOOP BACK PLUG ............................................................................................................................ 79
USB 2.0 LOOPBACK AND BENCHMARKING PLUG................................................................................................ 79
APPENDIX E – WHAT IS S.M.A.R.T.?........................................................................................................... 81
BIT users manual
Page 3
Chapter 1 - Overview
Overview
BurnInTest is a software tool that allows all the major sub-systems of a computer to be simultaneously tested for
reliability and stability. Anyone who has used computers for any period of time knows the importance of having
a stable system. The cost of a hardware failure can be enormous. If you are lucky your data will only become
inaccessible while faulty components are replaced. If you’re unlucky you can lose your files completely or have
them permanently corrupted. Even if you have good backups, you can still lose days of work returning the PC to
its pre-failure state. In many cases the lost data can never be replaced and businesses can be brought to their
knees, with the loss of documentation, customer details and financial records.
BurnInTest thoroughly exercises the hardware in a PC in a short period of time, in the same way as normal
applications use a PC over a long period of time. This takes advantage of the fact that computer hardware tends
to fail either very early in its life or after many thousands of hours of use.
Typically BurnInTest should be used before a PC is put into service, as a failure of a computer that isn’t being
used for a critical application and doesn’t have any critical data is much less damaging, (especially if it is still
under warranty).
BurnInTest will bring to the surface intermittent or hidden problems so that after a successfully run the computer
can be used with a much higher level of confidence.
BurnInTest tests,
• The CPU via mathematical operations and MMX calculations
• Hard drives / Floppy drive
• CD ROM, CD-RW burn and DVD
• Sound cards
• 2D graphics
• 3D graphics
• Printers
• Memory & Network connection
• Video file playback
• Tape drives (Professional version only)
• Serial ports (Professional version only)
• Parallel ports with loop back connectors (Professional version only)
• USB 1.x and 2.0 Ports (Professional version only)
• If you have specialized hardware that BurnInTest does not test, you can even write your own test and
integrate it with BurnInTest (Professional version only). PassMark has produced plugin tests for
Keyboards, Dial up Modems and FireWire ports.
All these tests can be done simultaneously as the application is multithreaded.
You can use BurnInTest to,
• Avoid delivering D.O.A. (Dead on Arrival) hardware to your customers.
• Dramatically reduce your burn in times with multithreaded simultaneous testing of components.
• Build your image as a supplier of quality systems.
• Avoid costly downtime, system rebuilds and lost data.
• Test the stability of a system after configuration changes or hardware upgrades (critical for over clocking).
• Interfaces for logging temperature and laptop battery charge levels (additional software required).
• Help with Electromagnetic Compliance Testing (EMC) to ANSI C63.4-1992
BIT users manual
Page 4
Chapter 1 - Overview
This document is applicable to both the Standard and Professional Windows editions of BurnInTest. More
testing will be possible with the Professional edition however. This document does not cover the Linux version
of BurnInTest.
BIT users manual
Page 5
Chapter 2 – Quick Start
Getting started
Here’s a summary of the steps to go through to start testing.
1.
2.
3.
4.
Use the Test Duty Cycles window to select the type of tests you
wish to perform (see chapter 0, Automatic test stop, Test
selection & Duty cycles).
Use the Configuration, Test Preferences window to set any
parameters that you wish to use, e.g. which hard drive to use
(see chapter 0, Test preferences).
Put paper in the printer, a data CD in the CD-drive and a floppy
disk in the disk drive (if you selected these tests).
Click on the Go button
That’s it.
BIT users manual
Page 6
Chapter 3 - Test descriptions
Test descriptions
Several tests are built into the software. Each of these tests contains one or more sub-tests designed to exercise
different aspects of your computer system. These tests can be run concurrently, with each test having its own
window and using its own CPU thread or as individual tests. It is also possible to vary the load mix by increasing
or decreasing the duty cycle for a particular test.
Each test keeps an error counter, a cycle counter and an operations counter. The significance of each of these
counters is detailed with each test.
The following tests are currently available.
Maths tests
This test contains two sub categories, integer and floating point, which contain 8 different tests in total.
For each of the integer tests, a large array of random 32Bit integers are processed using the particular
mathematical operations list below. For the floating-point test single precision floating-point numbers are used.
The following tests make up the suite.
-
Addition
Subtraction
Multiplication
Division
Floating Point Addition
Floating Point Subtraction
Floating Point Multiplication
Floating Point Division
Each mathematical calculation is performed twice and the results compared to each other. If the results obtained
from the two calculations are different then this is flagged as an error.
The numbers displayed in the window for this test represent how many millions of mathematical operations have
been performed and verified. Each different math’s test is run for half a second. After all 8 tests have been run
the cycle count is incremented. The duty cycle and the CPU speed determine how many operations can be
processed during the half-second period.
2D Graphics tests – video memory
This test directly writes, reads and verifies data to the primary display device using the Microsoft DirectDraw
interface. As such, DirectX 8.0 or higher must be installed on the machine under test.
You may test…
•
•
Just the local video memory – that is the graphics adapters onboard memory.
All video memory – this includes the local video memory and any other video memory, such as RAM
reserved for graphics.
In each case, the test first determines how much of the appropriate memory is free and then fills this memory
with a series of test patterns. Each test pattern is written to video memory and then displayed in the test window.
After display, the test pattern data is read back from video memory and verified. Any errors are reported.
The screen color depth must be set to a minimum of 16 bits.
BIT users manual
Page 7
Chapter 3 - Test descriptions
The test patterns used are similar to the memory test, plus some color specific patterns.
• White
11111111
• Black
00000000
• Binary 1
10101010
• Binary 2
01010101
• Sequence
(1, 2, 3, …)
• Red
• Green
• Blue
2D Graphics tests – graphics functions
This suite contains a number of tests that exercise the standard Windows graphics functions and the video card.
The numbers displayed at the top of the window for this test represent how many millions of graphical
operations have been performed. Each different math’s test is run for half a second. After all 3 tests have been
run the cycle count is incremented. The duty cycle, the CPU speed and the video card speed, determine how
many operations can be processed during the half-second period.
The codes used at the top of the windows are:
LI = Millions of lines drawn
SH = Millions of shapes drawn
RE = Millions of rectangles (Bitmaps) drawn
The total of these three values is displayed in the main window.
Line Drawing
Lines are drawn into window. The color is changed every 500 lines to enable the lines to been seen in the test
window.
Painting Bitmaps
A bitmap is painted into a window as quickly as possible.
Outline shapes
Two different shapes are drawn into a window:
-
An ellipse
A square with rounded corners
The color is changed every 64 shapes to enable the shapes to be seen in the test window.
For each graphical operation the Windows return code is checked to verify that the operation was successful.
Monitor EMC Testing
BurnInTest can be used to help with Electromagnetic Compliance Testing (EMC) to ANSI standard C63.4-1992.
Section 11.1.3 of this standard details the requirements for testing of Visual Display Units.
In brief, this standard requires that screens of scrolling 'H' characters be displayed while EMC testing takes
place. The monitor and the image displayed need to be adjusted by the tester in order to maximize the
electromagnetic output of the monitor (E.g., by selecting maximum brightness and black text on a white
background).
BurnInTest provides the ability to display a screen of Scrolling 'H'. This option can be selecting instead of the
standard 2D graphics test in the Preferences window (see section 4). This option is only available in the Pro
version of the software.
When activated, the Scrolling ‘H’ pattern fills the entire screen but Alt-Tab can be used to switch back to the
main Window or other applications. Scrolling only takes place when the test window is the active window.
Hard Disk and floppy disk tests
This suite contains a number of tests that exercise the mass storage units (Hard disk or otherwise) connected to
the computer. By default drive C: is used for the hard disk and drive A: for the floppy disk but this can be
BIT users manual
Page 8
Chapter 3 - Test descriptions
changed from the Preferences Dialog. (See section, 4, Disk preferences) and up to 20 drives can be selected for
simultaneous testing. From this twenty it is possible to select any combination of available floppy disk and hard
disks. Drives can be connected via the IDE bus, SATA, SCSI, USB, Firewire or any other method supported by
Windows.
The user can select one of the following patterns or choose to automatically cycle between the patterns. This
selection is done in the preferences dialog (see section, 4, Disk preferences). Possible test mode patterns are,
• Default (Cyclic)
• Sequential data pattern (0,1,2...255)
• Random data with random seeking
• High Low freq data overwrite (10101 then 00001)
• Butterfly seeking
• Binary data pattern 1 (10101010)
• Binary data pattern 2 (01010101)
• Zeros data pattern (00000000)
• Ones data pattern (11111111)
• Random data pattern
During each test cycle (except butterfly seeking) a file is created and verified on the disk. The size of the file is
equal to a certain percentage of the disks capacity and all files are created in the root directory of the drives
selected. The default file size is 1.0% of the disk size. For small capacity drives, like floppy disks, there is a
minimum file size of 32KB. It is possible to change the default file size percentage from the preferences
window.
Each test file is filled with a coded number sequence (a Pattern) that is used to verify the correct operation of the
disk when the file is verified (read). Files are created in the free space on the disk until the disk is at least 94%*
full (we’ll refer to this as the ‘full’ level). When the disk has reached this limit, all the test files are deleted and
the test starts again.
(*) 94% is used on disks that have a swap file or a Windows directory. A limit of 94% is used to ensure that
some space is kept for use by other applications (including the windows swap file). On disks that do not have a
swap file or a Windows directory, between 99.5% and 100% of the disk will be used.
A number of test modes perform additional testing to that described above. These are described below:
The Default (Cyclic) setting is to cycle between the patterns. In this case a new pattern will be selected each time
the disk has reached the full level. Where the disk drive is a floppy/CD/DVD or the operating system is less than
Windows 2000, Butterfly seeking will be skipped in the Cyclic testing.
The Sequential data pattern (0,1,2...255) setting will use sequential data, 0, 1, 2…255, 0, 1, 2, etc.
Random data with Random seeking generates 7 blocks of random data for each test disk and sequentially writes a
file using the first block of random data, followed by the second, third then first, until the file size specified is
reached. This file is then verified with sequential reads of data blocks from the file just written. This is followed
by seeking to a block within the files created, selected at random. At random either a read and verify or a write
will occur at this position on the disk. This random seek and random read/verify or write will be repeated the
number of times specified by the user in the disk preferences “seek count” field. New files are written and
verified with this process until the disk full level is reached. At this point, all the test files written to disk will be
deleted and the cycle will be repeated.
High Low frequency data overwrite works by first writing a file with a high frequency pattern
(10101010101010010101…), then overwriting this with a low frequency pattern (00001000010000100001…)
and then verifying that the low frequency pattern has been fully and correctly written.
Butterfly seeking works by detecting the hard disk geometry (number of cylinders etc) and then seeking between
cylinder 0 and cylinder X and back to cylinder 0. This is repeated the user specified ‘seek count’ times, before X
is incremented by 1. This is repeated until the last cylinder on the Volume is reached. The number of operations
reported refers to the number of combined Seek/Read operations. Note (1) that this test is only supported in
Windows 2000/XP/2003 server and above on non-Floppy/CD/DVD drives. (2) Performing a seek in many cases
does not move the disk driver head, to ensure the disk drive head is moved to the seek position, a small read (of
the Sector size) is done at each seek position. Also, incremental offsets of the sector size are used for the ‘X’
BIT users manual
Page 9
Chapter 3 - Test descriptions
seek and the return to ‘0’ seek, to ensure that the disk drive cache is overflowed and physical disk seeks
consistently occur (once the disk cache has overflowed). (3) Disk drive geometry may be faked or translated by
hard disks/ device drivers and may not be the actual drive geometry. (4) The Cylinders, track and sector size
used are typically logical values provided by the hard disk. (5) Up to 8 different physical extents per Volume (or
Drive letter) are supported in the butterfly-seeking test. (6) As with other disk tests, if a physical disk is
partitioned into multiple volumes (drives) and testing is simultaneously carried out on these volumes, the nature
of the test will change. In the case of Butterfly testing, the movement of the physical disk drive head will not
reflect butterfly seeking, but something between butterfly seeking and random seeking, as the seeks for volume 1
go from its logical cylinder 0 to X and the seeks for volume 2 go from its logical cylinder 0 to Y are mixed in
time. (7) Some disk drive device drivers do not support the supply of disk geometry information that is needed
for this test. Error reporting may be configured not to report “Butterfly seeking test not supported for this disk”
errors. This can be achieved by editing the Error Classifications file (BITErrorClassifications.txt) for Error
Number 160 to NONE. For example, changing the line from:
160,"Butterfly seeking test not supported for this disk",INFORMATION,
to
160,"Butterfly seeking test not supported for this disk",NONE,
The user can chose to log SMART errors. To learn more about SMART, see section Appendix E, What is
S.M.A.R.T?
If an error is detected in the coded pattern then the error count is incremented. The number of bytes written and
read from the disk are displayed in the test window. The addition of these two values is displayed in the main
window.
Graphical progress bars indicate if the test is currently writing or verifying (reading) information, the percentage
complete for the particular file and the space remaining on the disk.
The speed of the hard disk and the duty cycle determine how quickly test files are created.
When multiple copies of BurnInTest are used to test shared or networked drives at the same time, care must be
taken to avoid file name collision on the drive being tested. File name collision will manifest itself as file
creation errors. As the name of the test files created in the root directory of the disk being tested contains the
drive letter, collisions can be avoided by assigning different drive letters to the network drive on each machine
running BurnInTest.
There are a few issues to aware of when interpreting the results of the disk test. These are covered in the
precautions section (see section 9, Precautions for thorough and careful testing).
Note (1) To test Firewire ports using BurnInTest it is recommended that an external hard disk via the Firewire
port is used in conjunction with the BurnInTest disk test. (2) Similarly, to test external memory card
readers/writers or other removal drives, it is recommended that the BurnInTest disk test be used.
CPU MMX & SSE CPU tests
Tests of the MMX (multimedia extensions) and SSE extensions within newer CPUs.
First some definitions:
MMX = Multimedia Extensions
SSE = Streaming SIMD extensions
SIMD = Single Instruction Multiple Data
3DNow! = The name that chipmaker AMD, gave to their new CPU instructions.
MMX, SSE & 3DNow! are all sets of CPU instructions that have been introduced to enable blocks of data to be
processed at higher speeds. At the moment only AMD chips support 3DNow! and only Intel chips support SSE.
This may change in the future.
MMX allows 64bit Integer mathematical and logical operations. SSE allows 128bit floating-point mathematical
and logical operations. 3DNow! allows 64bit floating-point mathematical and logical operations.
BIT users manual
Page 10
Chapter 3 - Test descriptions
For each test a large array of random numbers are processed using the particular mathematical operations list
below.
MMX Addition (64bits)
MMX Subtraction (64bits)
MMX Multiplication (64bits)
SSE Addition (128bits)
SSE Multiplication (128bits)
3DNow! Maths (64bits, floating point addition, multiplication and subtraction)
Each mathematical calculation is performed several times and the results compared to each other. If the results
obtained are different from each other then this is flagged as an error. The numbers displayed in the window for
this test represent how many millions of calculations have been performed and verified. Each different test is run
for half a second. After all tests have been run the cycle count is incremented. The duty cycle and the CPU speed
determine how many operations can be processed during the half-second period.
BurnInTest 64bit version specific:
MMX and 3DNow! have been superseded and are not supported in BurnInTest. The MMX and 3DNow! Tests
have been removed from the 64bit version of BurnInTest.
End 64bit specific
3D Graphics test
Tests the DirectX 3D graphics system.
This test creates and animates a 3D image to test the 3D functions of the video card and its associated software.
The test creates 20 texture-mapped spheres. Each sphere contains 20 ‘slices’ and 20 ‘segments’. Making a total
of 16,000 triangular polygons rendered for each scene.
For the 3D test a cycle is defined to be 200 frames of 3D animation. The ‘operations’ count represents the
number of polygons displayed.
The test makes use of the Windows DirectX 3D interface. If DirectX (version specified in the System
Requirements) is not installed on your PC you will not be able to run this test (in fact you may not even be able
to start BurnInTest). DirectX can be downloaded from Microsoft and comes standard with all new versions of
Windows.
Some video cards don’t support hardware acceleration at high color depths and you may have to reduce the color
depth to 16bit color to get the test to run.
Warning: A number of manufacturers produce video cards, which have very poor software support for 3D
graphics (DirectX). This is especially the case for the cheaper, bottom of the range video cards. Poor drivers can
result in strange visual artifacts appearing on the screen, poor 3D performance and system crashes.
Printer test
This test verifies the correct operation of the printer by
either sending a standard page of text, or a sending a
command file, to the printer.
Standard test
The standard test sends pages of text to the default
Windows printer. (See Start, Settings Printers in Windows to define a default printer). The pages printed by the
printer should be a page of solid text in a fixed-point font filling the printable area of the page. There should be a
small margin (usually under 1cm) around the block of text.
BIT users manual
Page 11
Chapter 3 - Test descriptions
The number of pages printed during a period of time depends on the print duty cycle. For each duty cycle
increment there will be an additional 20second delay between each page. This means that a Duty Cycle of 1
corresponds to a delay of 33 minutes between pages. A duty cycle of 50 corresponds to a delay of about
16minutes between pages.
The pages printed should be visually inspected to check if they are complete.
The cycle count for the printer test corresponds to the number of pages printed. The ‘ops’ count corresponds to
the number of characters printed.
Standard color printer test
The standard color test sends pages of color text to the default Windows printer. The pages printed by the printer
should be a page of solid color text in a fixed-point font filling the printable area of the page. There should be a
small margin (usually under 1cm) around the block of text. The colors are as follows:
Title: Black
Repeated blocks of:
Red
(RGB: 0xFF0000)
Red
(RGB: 0x7F0000)
Green (RGB: 0x00FF00)
Green (RGB: 0x007F00)
Blue
(RGB: 0x0000FF)
Blue
(RGB: 0x00007F)
Printer command file
The user can define a printer command file to be used in place of the standard test. A printer command file is a
file containing a series of printer commands that are used by the printer to carry out certain actions. For example
a printer command file could contain instructions on how to print an image or page of text. Specialized printer
command files can also be created by printer manufacturers in order to run their printers through a particular test
routine.
See Print Preferences, section 4, for more details about how to create and select printer command files.
CD ROM, CD-RW burn and DVD test
Tests the CD-ROM, CD-RW burner and DVD and optionally the sound card (when used with a music CD).
This test can be used with many different types of CD / DVD media, including the following types.
- Music CD
- Data CD-ROM
- Data CD-R
- Data CD-RW
- Specialized test CD (see below for details)
- Data DVD-ROM
- Data DVD-R
- Data DVD-RW
- Specialized test DVD (see below for details)
The type of disc being used must be suitable for the test mode set in the CD / DVD preferences window
however. If multiple CD / DVD players are installed in your system, you can select all or some of them from the
Preferences window (see section 4, Disk preferences). Up to twenty drives can be selected for simultaneous
testing.
Test modes
NO TEST
Don’t test the drive selected.
Music CD playback
For a music CD, the test verifies the operation of the CD ROM and the PC’s sound system by repeatedly playing
a music CD. The number of cycles corresponds to the total number of times the entire CD has been played. The
number of ‘ops’ corresponds to the number of bytes read from the CD in order to generate the sound. The duty
cycle affects the time spent waiting between tracks. Choose a band you like and turn the volume up loud.
BIT users manual
Page 12
Chapter 3 - Test descriptions
Data CD / DVD read and verify
For a data CD/DVD, the CD/DVD is scanned with each file being read in turn. A 32bit checksum is created for
each file as it is read from the CD/DVD. The CD/DVD is then re-read and the checksums verified to complete
the cycle. The best data CD/DVD’s to use are those, which are full and contain a good variety of files. Some
large ones and some small ones. If the CD has more than 10,000 files only the 1st 10,000 will be scanned.
PassMark Test CD / DVD
This test method is the most advanced CD-ROM test mode and provides a level of in depth testing not available
with the other two methods. When possible this method should be selected in preference to the other two. It
allows a complete read and verify of the data on the CD, and optionally random seek testing (see Additional
seeking and Seek count below).
There is the possibility to create CDs and DVDs especially for use with BurnInTest. These specialized CDs and
DVDs contain a set of files that have been specifically created to thoroughly check your drive. With these CDs
and DVDs, BurnInTest is able to know the layout and exact content for each file on the CD/DVD. This allows
BurnInTest to read back each file and test every bit for accuracy.
PassMark Software sells these specialized CDs and DVDs but you can also make your own file set using a utility
called CD-Maker. You will still need 3rd party CD or DVD burner software to actually create the disc however.
CD-Maker creates a set of files that can be burned to CD or DVD. More information about this tool can be found
on the PassMark Software web site. (http://www.passmark.com/products/cdmaker.htm )
No CD in Drive test option
This test method only attempts to detect the presence of a CD/DVD drive. No attempt is made to read from the
drive. This can be a useful option when there are no discs available for testing but a basic check is still required
to ensure that the drive was detected and enumerated by Windows. This test option is not as thorough as the
other options and, if possible, one of the other options should be used.
Burn CD-RW
This test method provides testing of a CD burner through continuously erasing, burning, verifying then randomly
seeking across a CD-RW.
PassMark 650MB or 700MB (user specified) Test CD data (see PassMark Test CD / DVD above) is created
temporarily on the harddisk. For the period of the test, the following testing is then repeated:
- Erase the CD-RW media (User specified Quick or Full format). The quick option erases content
type information in typically less than two minutes. The Full option also erases all user data on the
disc however this can take a considerable amount of time (over 1 hour) and the test cannot be
cancelled during this stage.
- Burn the PassMark Test CD data to the CD.
- A complete read and verify of the data on the CD.
- And optionally random seek testing (see Additional seeking and Seek count below).
Additional seeking and Seek count
When Additional seeking is selected (for PassMark CD/DVD’s and Burn CD-RW only), seeking to different
positions on the CD/DVD or CD-RW and verifying the data at this position will occur Seek Count number of
times for each cycle of a file read and verify.
Display
Depending on the test the following information is displayed in the test window.
Progress bar
Displays the progress graphically for the longer CD burn phases.
Phase
Data CD and specialized CDs :
• Starting
• Checksum creation
BIT users manual
Page 13
Chapter 3 - Test descriptions
•
•
•
•
•
Searching for files
Opening file
Reading Test CD
Verifying checksums
Seeking
CD burn:
• Starting
• Creating test data files
• Setting active CD burner
• Erasing CD-RW media
• Erasing CD-RW media retry
• Preparing to Burn CD
• Adding staged image data
• Burning image to CD
• Closing the CD
• Finished CD Burn
• Checksum creation
• Searching for files
• Opening file
• Reading Test CD
• Verifying checksums
• Seeking
Files scanned
This is the number of files that have been scanned during the current cycle.
Current file
This is the name of the file that is currently being scanned. If this is in the seek phase, this will display the file
and block within the file that the drive has seeked to.
Bytes read
This is the total number of bytes read from the CD.
CD burn:
This is the total number of bytes written or read from the CD (depending on the phase).
Throughput (Current and Average)
This is read speed from the CD in Megabytes per second and a conversion to standard CD Drive speed (e.g. 4x).
It should be noted that this measurement includes the seek and rotational latency time, so that the speed reported
will always be lower than the maximum drive speed. This is especially the case for a CD with many small files.
Also most drive manufacturers quote the maximum speed for their drives, the actual transfer rate is usually
significantly lower than the quoted maximum. E.g. It is not unusual for a 32x speed drive to only reach a speed
of 2MB/Sec, 13x.
CD burn:
This is the write or read speed (depending on the phase).
Errors
Data CD and specialized CDs :
The number of errors detected. See Common Errors (Appendix C) for a description of the errors that may be
encountered.
Network test
Tests the Network connection.
The Network test, tests your networking hardware and software. This is done by sending special ICMP_ECHO
message using the Internet Control Message Protocol (ICMP). The ICMP_ECHO message is more commonly
known as a ‘ping’. This message is echoed back to your computer by a remote host. This allows the reliability of
BIT users manual
Page 14
Chapter 3 - Test descriptions
the network connection to be determined. Up to 4 remote hosts can be selected by setting the Destination
Network address in the Test Preferences window (see section 4, Network preferences).
These values are used for the Network test. Each one must be a
URL or an IP address (or left blank).
• A URL is the name of a network host, e.g.
www.hostname.com
• An IP address is a sequence of 4 numbers that
correspond to a network host. e.g. 169.192.0.1
The host selected must be accessible from the computer and
capable of responding to the ‘ping’ command. PassMark
recommends the selection of a local host to minimize data link
problems, which are fairly common on the Internet. The IP
address 127.0.0.1 can also be used for local loop back testing.
It should be noted however that the loop back is done in the Windows TCP/IP software and not on the network
card.
The packet sent to the remote host contains a data payload and a checksum. Every time a packet is echoed from
the remote host the checksum is verified and the data payload is compared byte by byte with the data that was
sent. Any differences in the payload or an incorrect checksum will result in an error. The data payload is 64
bytes in length.
To use this test the TCP/IP Internet protocol must be installed on your computer. This can be determined from
the Windows Control Panel, Network Window. If you try and use this test without the Internet protocol being
installed, you will get a message like, “Could not create a Network Socket”.
The amount of time BurnInTest will spend waiting for a packet can also be set in the Test Preferences window
(see section 4, Network preferences).
The meaning of the information displayed in the network test window is given below.
Packets Sent
The number of packets sent to the remote host.
Packets Received
The number of packets received back from the remote host. This should remain at the same level as the Packets
Sent counter. If after 2 second a sent packet is not echoed, this causes a timeout error and Packets Sent will be
greater than Packets received.
Average Delay
The average round trip time in milliseconds for a packet.
Max Delay
The maximum round trip time in milliseconds for a packet. The maximum values often happen at the very start
of the test session. This is because Windows is still loading and caching the required networking software.
Current delay
The round trip time in milliseconds for the last packet sent.
Bytes sent
The total number of bytes transmitted to the network.
Flow rate
The average number of packets sent per second for the test period. The duty cycle set for the Network Test
determines how many packets are being sent per second.
Errors
The number of errors that have been detected and the percentage of packets that had an error are displayed. The
definition of what an “error” is can be changed from the Test Preferences window (see section 4, Network
BIT users manual
Page 15
Chapter 3 - Test descriptions
preferences). Depending on the settings in the preferences window, the error count can be incremented for every
detected error or alternatively it can be set to only increment when the specified error ratio is exceeded.
When the specified error ratio is exceeded, the ratio is reset to zero to avoid triggering a continuous stream of
errors. Thus the ratio value displayed is the ratio of bad errors to good packets since the last time the threshold
was crossed.
For the purposes of detecting a crossing of the threshold and signaling an error, the ratio is ignored until a
sufficient number of packets have been sent to make the ratio valid. For example, if the ratio is set to 2%, at least
50 packets must be sent before the ratio is deemed to be valid. If the ratio is set to 0.1%, at least 1000 packets
must be sent before the ratio is deemed to be valid.
Compatibility issues
If you are running Windows 2000, XP, 2003 server or above, you need to have administrator privileges to run
this test.
Advanced Network Test
(Applicable to BurnInTest Pro version only)
The advanced network test works in conjunction with an external program called EndPoint, which the network
cards being tested connect to and send and receive data from. To use the advanced network test there must be at
least 1 endpoint running on another machine on the network separate to the machine running BurnInTest. The
test will dynamically connect to the EndPoint applications as needed.
To successfully run the test, BurnInTest uses broadcasts to find all the available EndPoint applications on the
network, so anything that may block the broadcasts, such as a router, needs to be taken into account. Firewalls
could also interfere with the test and may need to be configured to allow BurnInTest and the EndPoint
applications to use the network.
It should be noted that when there is more than one network card in the machine being tested the FTP option
cannot lock to a specific network card in the same way that the UDP and TCP tests do.
Memory (RAM) test
Tests the Memory in the computer.
The Memory test, tests the reliability of the RAM installed in the computer. As BurnInTest runs within windows
some of the available RAM is being used by windows and any other applications running. Any memory that is
not already in active use will be grabbed by the Memory test.
The memory test works by writing a pattern numbers in the RAM, then verifying the numbers read from the
RAM match this sequence. The pattern used will change automatically from one cycle to the next. Possible
patterns are:
Sequence (0,1,2...255)
Binary 1 (10101010)
Binary 2 (01010101)
Zeros (00000000)
Ones (11111111)
The Test pattern may be selected to be one of the above, or to cycle through each of the above in the above
order.
There are 3 sequences that the test goes through:
1/ Memory allocation. The test will dynamically grab and release memory depending on how much is currently
available. The amount that has been grabbed and is under test is displayed in the Test Ram Field.
2/ Writing the number sequence.
3/ Verifying the number sequence.
The total amount of free RAM is displayed in the Memory Test Window. This number should always be bigger
than the Test Ram number. Some memory is always left available to avoid Out of Memory Errors, and disk
thrashing caused by Windows swapping to disk.
BIT users manual
Page 16
Chapter 3 - Test descriptions
For the Memory test a cycle is defined to be the number of times the above 3-step sequence is completed. The
‘operations’ count represents the number of bytes read or written.
It should be noted that not all RAM faults will be detected by this test. This is especially the case if Windows or
the Windows cache is using a large proportion of the available RAM. RAM faults may show up as system
crashes or disk errors however.
Standard memory test
The standard memory is the normal test for testing memory.
Torture test (Memory over-allocation & disk swapping test)
The torture test is a multi-process memory test. Multiple processes are started in their own virtual address space
and each process allocates and tests a block of RAM. This avoids the problem of virtual memory fragmentation,
which the standard test can experience trying to allocate a single large block of RAM. Each process runs
asynchronously, so writing and reading of various memory blocks will take place at the same time in different
processes. The other advantage over the standard test is the possibility to over-allocate the RAM. (The standard
test attempts to prevent this to avoid disk swapping). Over-allocation takes place when more RAM is used by the
torture test than is currently available in the system. This then results in Windows disk swapping memory blocks
into a paging file on the disk. This continual swapping to and from the disk places a very heavy load on the
system. The I/O activity on the disk will increase dramatically but CPU load can actually decrease as more and
more time is spent waiting for the paging activity to complete. Depending on the level of over-allocation
Windows may need to extend the paging file or may even fail as it runs out of available RAM.
Addressing Windows Extension (AWE) memory test
The Addressing Windows Extension (AWE) memory test allows a larger area of memory to be tested on 32-bit
versions of Windows. It requires some additional administrator user rights. It will also only work in Window
2000, XP and 2003 Server. See the description in the RAM test preferences window for more details. The
advanced memory test is only available in the Professional version of the software.
BurnInTest 64bit version specific:
The AWE memory test is not available under 64bit Windows. However, the standard test will test very large
memory sizes under 64bit Windows.
End 64bit specific
The AWE and Torture test are only available in the Professional edition of BurnInTest.
Sound card test
This test operates in two different modes. The standard mode verifies that the sound card can play back MP3,
waveform audio and MIDI (Musical Instrument Digital Interface) sound.
There is also a loopback mode, which requires a loopback cable. Loopback mode measures the distortion
between audio input and audio output. Loopback cables are available from http://www.passmark.com
Standard Test:
Three small sound clips are played by in a loop. The duration of play back is determined by the duty cycle
setting selected by the user. A larger duty cycle setting will result in a longer playback period. The sound test
window displays the amount of time remaining in the playback. If the sound clip is shorter than the test period,
then the sound will be played in a loop until the test period is complete.
You may select the files that are played during the test in the Sound preferences window. The default test files
are in the BurnInTest installation directory.
C:\Program Files\BurnInTest\Testsound.mid (Midi file)
C:\Program Files\BurnInTest\Testsound.wav (Wave file)
C:\Program Files\BurnInTest\Testsound.mp3 (MP3 file)
While the test is running the user should verify that the sounds produced are clear and without distortion. If no
sound is heard and BurnInTest detects no errors, check
- The mute and volume settings in the Windows mixer / volume control window
BIT users manual
Page 17
Chapter 3 - Test descriptions
-
The speakers are plugged into the correct plug on the sound card
-
The speakers volume control is turned up
Loopback Test:
Loopback testing involves connecting the audio out and microphone in plugs of your computers sound card
together with an audio loopback cable.
The test uses a single channel (mono), 8-bit sound format with telephone quality sample rate (11025Hz). It
outputs a sin wave tone with a frequency of 300Hz and measures the input values for distortion. It is possible to
set the maximum level of distortion before an error is logged in the Sound preferences window.
Note: If you use the Sound Loopback test at the same time as you use the Video Playback test, you must use a
Video file with no audio component as the audio will impact the distortion ratio of the sound test.
Before beginning a loopback test, there are a few important configurations, which must be carried out in
Windows in order for the test results to be meaningful.
Getting ready for the loopback test
1. Open the mixer.
This can be done from Control Panel =>Sounds and Audio devices. Alternatively, the mixer can be started by
double clicking on the small speaker icon on the task bar. Note that the mixer device has two properties sections,
Playback and Recording, which contain different volume controls depending on your sound card. It is possible to
toggle between the two sections through the Options =>Properties menu item of the mixer device.
2. Check that Wave output is enabled.
Look at the settings for the overall Volume Control and Wave volume control. The slide bars should be near the
centre of the range and the Mute check boxes should be unchecked. It is advisable to mute all other output
devices (e.g. Midi, CD Audio) in order to maintain a pure output signal. In particular, the Line In volume control
should be muted to prevent feedback loops in some sound cards. Initially leave the balance setting in the middle.
If these settings are not correct, there will be no sound output.
4. Switch to the mixer recording (input) window.
From the mixer window select the Options => Properties menu item,
and then select the Recording option and the volume control check
boxes. After clicking on OK the recording volume controls for the
mixer inputs are displayed.
5. Check that Line In input is enabled.
This volume control is the most important setting for the loopback
test. If the volume setting is too low, the input signals will be
correspondingly vague and therefore distortion will be high. If the
volume setting is too high, the input signal may become “capped” as
the sound waves become truncated. This will also lead to high
distortion and errors being logged. You may need to play with this
control while the test is in progress in order to find the “sweet spot”.
A useful tool for calibrating your input and output audio levels is
PassMark SoundCheck, available from http://www.passmark.com
Once the Volume Controls and Recording Controls have been
configured correctly, the test is ready to begin. BurnInTest measures
the level of distortion between the output signal and the input signal on
a wave-by-wave basis. If the distortion measured exceeds the value set
in the Sound Preferences window and an error is logged and the output
and input waves are displayed in the Sound Test window. Waves that
exceed 0.5 of the maximum distortion level are also displayed, but no
error is logged.
Video Playback test
This test allows video files to be played as a BurnInTest test. This allows for the detection of numerous error
conditions that a PC might exhibit, from lack of media codec support, playback errors and dropped samples
(reduced quality due to the PCs inability to process the video quickly enough) and Bit Error Rates. From the
Video Playback Preferences window, up to 3 video files can be selected for playback, with each video being
played in turn (1st, 2nd, 3rd, 1st etc).
BIT users manual
Page 18
Chapter 3 - Test descriptions
Video files come in a number of different file formats and within these file formats, video and audio are encoded
using different compression methods.
The file formats supported include:
• Audio-Video Interleaved (AVI)
• Motion Picture Experts Group (MPEG, MPG)
The Compression formats supported (where installed) include:
• AVI
• Cinepak
• Digital Video (DV)
• DivX
• MJPEG
• MPEG-1
• MPEG-4 video: ISO version 1.0 , Microsoft version 3
• Microsoft Windows Media Video codec version 7.0
MPEG-2 is not supported.
(Note: Each of the above formats are registered trademarks of their respective owners).
Sample test videos are available from the PassMark Web site.
A test window will be opened when the tests are started. This will be sized with width based on the BurnInTest
test window size and height such that the aspect ratio is maintained (i.e. so it is not squashed). The window may
moved, resized or made Full screen
The title bar will show the File format, Video compression method and native video size (Width x Height).
BitErrorRate will be logged to the detailed log, where available and greater than 0.
The number of operations will indicate the number of seconds of Video played. A cycle represents each video
file that is played, e.g. if Videos 1,2,3,1,2 were played, this would represent 5 cycles.
Duty cycle sets the amount of delay between each of the video files played.
Error reporting may be configured not to report reduced quality of Video playback (dropped samples) errors.
This can be achieved by editing the Error Classifications file (BITErrorClassifications.txt) for Error Number 182
to NONE. For example, changing the line from:
182,"Video Playback is dropping samples",WARNING,
to
182,"Video Playback is dropping samples",NONE,
Serial port test (BurnInTest Pro only)
Tests the serial ports connected to the computer.
Tests the serial communications ports connected to the PC. Up to 64 serial ports may be
tested simultaneously. The serial ports and test speed can be selected from the Test
Preferences window (see section 4, Serial port preferences).
A serial port loop back plug per port is required to run this test. These can be purchased
from the PassMark web site (www.passmark.com) or you can make them yourself (see Appendix D).
Each test cycle corresponds to about 10 seconds of data transmission followed by a signal pin test phase. The
signal pin test phase checks that the following pins on the serial port are functioning correctly.
RTS – Request to Send
CTS - Clear to Send
DTR – Data terminal ready
DSR – Data set ready
BIT users manual
Page 19
Chapter 3 - Test descriptions
The number of ‘ops’ corresponds to the number of bytes sent and received. The duty cycle affects the time spent
waiting between cycles.
The serial port selected must not already be in use by Windows (for example by the mouse or an active modem),
for the test to be carried out.
The speed that the serial port operates at is independent from the modem speeds. Even if you have a 56Kbit/s
modem your serial port may operate at a higher speed. The maximum serial port speed depends on the type of
chip installed on your motherboard. Most PCs will only do up to 115Kbit/s, so don’t be alarmed if the test fails
at 128Kbit/s or above.
If the “detect only” option was selected in the preferences window then the loopback test will not be performed.
The presence of the serial port in the system will still be checked for however.
The following information is displayed for each port being tested.
Serial Port
This is the Windows name for the serial port being tested. The port can be selected from the Test Preferences
window. Any port between COM1 and COM64 is supported.
Speed
This is speed that the serial port is configured for. The speed can be selected from the Test Preferences window.
Bytes Sent
This is the number of bytes that have been sent to the serial port.
Bytes Received
This is the number of bytes that have received from the serial port.
Errors
This is the number of errors detected (see Appendix C, p.48).
Throughput
This is the real measured throughput for the port. This will generally be less than the Speed (see above) as there
is some overhead in the code and in the data transmission itself (e.g. Stop bits).
Parallel port test (BurnInTest Pro only)
Tests the parallel port connected to the computer.
Tests the parallel communications port connected to the PC. The parallel port
to be tested can be selected from the Test Preferences window (see section 4,
Parallel port preference). A parallel port loop back plug is required to run this
test. These can be purchased from the PassMark web site (www.passmark.com)
or you can make them yourself (see Appendix D).
Each test cycle corresponds to 500,000 bytes of data transmission. The number
of ‘ops’ corresponds to the number of bytes sent and received. The duty cycle affects the time spent waiting
between cycles.
The parallel port selected must not already be in use by Windows (for example by the printer or other external
device), for the test to be carried out. The default on-board Parallel port settings are that Port1 is named LPT1
and corresponds to the physical IO memory address 0378. Depending on the version of Windows you are using
you may be able to change a number of Parallel Port settings, including the port name (e.g. You may have
renamed LPT1 to LPT2) and the IO address (e.g. You may have changes port1 to use 0x3BC instead of 0x378).
BurnInTest will automatically detect the changes for the on-board parallel port. For W2000 and later versions of
Windows, PCI and ISA parallel ports will also be automatically detected and available for testing. For W98 and
ME, PCI and ISA parallel ports will need to be manually configured by writing the required IO address (e.g.
0xB400) into a file named “ioports.dat” and placing this in the executable directory. Contact
help@passmark.com if you need help to do this.
BIT users manual
Page 20
Chapter 3 - Test descriptions
Note: That the Windows operating system may block exclusive access to the Parallel port. This is reported with
the error “Unable to lock Parallel Port”. When this occurs there is a retry pattern over a number of minutes to try
and lock the port. Regardless of the result, the parallel port test will then commence. If you wish to ignore the
“Unable to lock Parallel Port” message the BITErrorClassifications.txt file can be edited to change this error
message classification (Number 138) to NONE.
In Windows 2000, XP and 2003 Server you need to be the administrator to run this test.
Also note that the old ‘bi-directional’ BIOS mode is not supported. ECP or EPP mode is required is be set in
BIOS for the port being tested.
Parallel Port
This is the port name for the parallel port being tested. The port can be selected from the Test Preferences
window.
Bytes Sent
This is the number of bytes that have been sent to the parallel port.
Bytes Received
This is the number of bytes that have received from the parallel port.
Errors
This is the number of errors detected (see Appendix C, p.48).
Throughput
This is the measured throughput for the port.
USB port test (BurnInTest Pro only)
Tests the USB (Universal Serial Bus) communications ports connected to the PC. The USB devices that are
connected can be viewed from the Test Preferences window. A USB port loopback plug is required to run this
test. There are two versions available, USB1 and USB2 loopback plugs. The USB1 loopback plug supports a
range of testing for the USB1.x (USB 1.x supports rates up to 12 Mb/sec), while the USB2 loopback plug
provides a wider range of testing, as well as benchmarking, for USB 2.0 and USB 1.x (USB 2.0 supports rates up
to 480Mb/sec). These can be purchased from the PassMark web site (www.passmark.com).
Using a USB loopback plug (USB1 or USB2 loopback plugs) and BurnInTest, it is possible to,
•
•
•
•
•
•
•
Quickly check if a USB port is powered (the red LED)
Check that data can be sent and received from the port
Check USB errors rates and transmission speed.
Check that the system remains stable under long periods of load
Test that USB support in Windows has been correctly installed
Check USB cabling
Concurrently check multiple USB ports at the same time.
In addition, with the USB2 loopback plug and BurnInTest (V4.0 and above), it is possible to test:
• Check if your PC ports are high speed 480Mbits/sec (USB 2.0), or full speed 12Mbits/sec (USB 1.x) via
a LED or the BurnInTest USB test Window
• Check USB errors rates and transmission speed (using the USB2.0 Highspeed protocol)
It is worth noting that with the USB2 loopback plug and the PassMark USB2Test software it is possible to:
•
•
Benchmark the maximum speed of your PC's USB ports
Measure the level of recovered data (retransmitted frames)
USB1 loopback plug
BIT users manual
USB2 loopback plug
Page 21
Chapter 3 - Test descriptions
Using these plugs on a system that is functioning correctly, you can expect error free loopback transfer speeds of
around:
• USB1 loopback plug:
700 - 800 Kbits/sec per port
• USB2 loopback plug:
5 – 7 Mbits/sec per port. (Note: With benchmarking functionality provided in
the USB2Test software you can expect to see between 250Mb/sec and 380Mb/sec for a USB2 port
connected to the motherboard).
To reach this level you need to have the USB test duty cycle set to 100%. Up to 10 of these USB test plugs can
be simultaneously connected to a PC (providing that free USB ports are available on the PC or on a down stream
hub). Note: If required, both USB1 and USB2 loopback plugs can be used to test different USB ports
simultaneously.
To use the plugs they must be correctly installed with their device driver. See the installation and users guide that
is supplied with the plugs for more details. Additional information about the plugs is also available on the
PassMark web site.
Each test cycle corresponds to:
• USB1 loopback plug:
• USB2 loopback plug:
300 data blocks (Roughly 600KB)
8000 data blocks (4000KB)
The number of Operations (‘ops’) corresponds to the number of bytes sent and received. The duty cycle affects
the time spent waiting between cycles.
Each plug also has its own serial number stored in EPROM, on the plug, so it is possible to identify each plug
when multiple plugs are connected. USB2 loopback plugs are firmware controlled and upgradeable.
The USB test sends data to the USB loopback plug in:
• USB1 loopback plug:
2KB blocks. The USB loopback plug receives this data, checks the cyclic
redundancy checksum is correct, and then re-encodes the data before sending it back to the PC.
• USB2 loopback plug:
0.5KB blocks. The USB loopback plug receives this data, copies it to a new
buffer and transmits it back to the PC. The PC compares the data in the block for an exact match, and
then builds a new packet of random data bytes before sending it back to the PC.
Any differences between the data send and received is flagged as an error.
As the low level USB protocol has some ability to detect errors and re-transmit bad data it is important to pay
attention to the transfer speed. A low transfer speed may be an indicator of an intermittent or partial hardware
fault. With the USB2 loopback plug an error LED will be lit when more than a defined number of recoverable
bus errors occur (currently set to 4). (Note: The USB2Test software will report the number of recoverable bus
errors per block). A lack of high quality cabling within or external to the PC is a typical cause of retransmission
due to recoverable bus errors.
The USB2 loopback plug Error LED is reset at the start of each test if results are configured to be cleared:
• At the start of each test (Preferences, Logs, Log Clearing, Automatically clear logs at the start of each
run),
• On selecting “Edit, Clear All Results” or
• On a power reset (e.g. PC sleep mode).
USB test window
The following information is displayed in the USB test window.
BIT users manual
Page 22
Chapter 3 - Test descriptions
USB Identity
An indication of whether the USB port is connected to a USB1 or USB2 loopback plug is shown together with
the maximum speed supported by the protocol currently supported by the USB port (e.g. 12Mb/s or 480Mb/s).
The device number and the unique device serial number are then displayed. The device number is determined by
the order in which plugs are connected and disconnected and is not linked to a physical USB socket on the PC.
(Note that this is different from Serial and Parallel ports). The device serial number is collected during the USB
device enumeration process and comes from the memory (EEPROM) in the plug itself.
Bytes Sent
The number of Kilobytes sent to the USB device.
Bytes Rec.
The number of Kilobytes received from the USB device.
Throughput
The number of Kilobits or Megabits of data received and transmitted in the last second.
Ave. Throughput
The average number of Kilobits or Megabits of data received and transmitted per second since the start of the
test.
Errors
The number of errors that have occurred (i.e. Transmitted data does not match received data).
Tape drive test (BurnInTest Pro only)
Tests the tape drive connected to the PC. Drives can be connected via the SCSI bus, IDE bus or any other
method supported by Windows2000/XP/2003 Server. Note that Windows 2000 and later operating systems no
longer supports tape drives that use the floppy disk interface. The tape drive to be tested can be selected from the
Test Preferences window (see section 4).
A blank tape (media) is required to run this test. If the tape is not blank any existing data will be overwritten and
permanently lost. There are three optional steps that can be included as part of each test cycle. See the Test
Preferences window for more details (see section 4).
Test description
The test is carried out by writing a number of files onto the tape, rewinding the tape and then verifying the files
were correctly written by re-reading them from the tape. A progress bar shows the percentage complete for the
current test cycle. The contents of the files will correspond to a particular data pattern. The following patterns
will be used in a cyclic manner in the following order:
Sequence (0,1,2...255)
Binary 1 (10101010)
Binary 2 (01010101)
Zeros (00000000)
Ones (11111111)
The number of files created and the size of each file can be set in the Preferences window. The tape needs to be
large enough to contain the number of files selected plus a small amount of overhead for file marks.
A test cycle is completed each time all the files have been written and re-read. A cycle may end prematurely for
some errors conditions. The number of ‘Ops’ reported in the main window is the total of the bytes read and bytes
written.
Pattern
The current data pattern that is being used to fill the test files (see the list above).
Bytes Written
This is the total number of bytes that have written to the tape.
Bytes Read
This is the total number of bytes that have read from the tape.
Errors
BIT users manual
Page 23
Chapter 3 - Test descriptions
This is the number of errors detected (see Appendix C).
Plugin test (BurnInTest Pro only)
If you have specialized hardware that BurnInTest does not test, you can write your own test and integrate it with
BurnInTest.
Up to 3 plugin’s can be specified. PassMark has developed plugins for the following tests:
- Modem testing, using PassMark’s ModemTest software;
-
Keyboard testing, using PassMark’s KeyboardTest software,
-
Firewire port testing, using a Passmark developed plugin and a “Kanguru FireFlash” drive.
-
Sample software in C and C++ is also available to assist developing a plugin for your specific hardware.
The plug-in test will appear as another test, much like all of the tests. It may be switched on or off, and duty
cycle set. The Plug-in application can define the following, which may be displayed, logged and included in the
BurnInTest results:
- The test window title text.
-
The number of test cycles completed.
-
The number of plug-in test errors in the current test run. The plug-in may flag BurnInTest when a new error
occurs, and pass a plug-in defined error message (this will be displayed in the BurnInTest windows and may
be logged in the log file).
-
The plug-in may flag BurnInTest when it wants to display new status text (defined by the plug-in), such as
“Waiting for event”.
-
The Plug-in may define up to 3 “number of operations counts”. Typically this would be for something like,
the number of writes, reads and verifies. All three operation counters may be given plug-in defined labels,
which will be displayed in the BurnInTest test window. Any operation counter may be flagged as not in use,
hence you could use a single operations counter, such as just “Write”. BurnInTest will display each label
and number of operations counters separately in the test window. They will be summed for the main
window operation count. These values are 64 bit integers.
-
Two Plug-in defined string values and labels are provided for the display of other interesting data to the
user, e.g. the plug-in could provide throughput in Mb/s to BurnInTest. BurnInTest displays this information
in the test window. Both of these fields may be flagged as not in use.
Software may be written to interface an external plug-in test module with BurnInTest. An interface is defined to
pass test parameters between an external test application and BurnInTest. BurnInTest will attempt to start this
application when the Plug-in test is run and this application is specified as the Plug-in file. BurnInTest will pass
a flag that indicates that the test run is still underway (e.g. has/ has not been stopped by the user) and BurnInTest
will pass in the Duty cycle. Before starting the test, the plug-in application must initialize the test parameters
(such as display labels) and flag that they have been set. Once the test is started, the test result parameters and
most test window labels may be updated based on the test results. BurnInTest will pick up any changes. Specific
cases where the plugin must flag to BurnInTest that new parameters are available are for a new labels (as a group
flag), error, a new status message and the user defined strings 1 & 2. You may need to check that BIT has read
the old values before writing new values. See the Interface definition, PLUGININTERFACE, and the sample
source code for details.
BIT users manual
Page 24
Chapter 4 - Test Preferences
Test preferences
Disk preferences
Configuring hard disk, Floppy disk and tests for other
types of drives
These fields and buttons allow the user to select the drives
used in the disk test.
Disk Selection
To edit the test preferences for a disk, select the Drive in the
“Disk Selection” List. Any changes you make in the “Edit
details for drive: <drive name>” will be reflected in this list.
Test this drive
To specify that a drive should be tested, check the “Test this
drive”. This will copy the last used settings to the drive
specified. To test a shared network drive, the drives needs to
have a drive letter mapped to it. Only the drives detected by
the Windows operating system as displayed in the drop down
list. The following settings can be configured differently for
each drive: Slow drive warning threshold, File size, Test
Pattern and Log SMART errors. To use the general disk drive
Duty cycle for each disk just set the Duty Cycle override value
to blank (no value), otherwise set the required value per disk.
To remove a disk from the test, uncheck the “Test this drive”.
Up to twenty drive letters can be selected for simultaneous testing. Note: The simultaneous testing of two
partitions on the same physical hard drive will result in a lot of seeking between partitions and slow down the
test significantly.
Test Mode
The test patterns that can be selected are explained in the section, “Disk test (Floppy and hard).
Slow drive warning threshold
BurnInTest can be configured so that a warning message is generated when the transfer speed from a disk device
drops below a certain level. The transfer speed is measured in Megabytes / second and a different threshold can
be assigned for each drive.
Selecting a level of 0 will disable the feature.
When the measured I/O transfer speed is below the threshold, a warning message will appear in the detailed error
log. The entry will be similar to the following,
2002-08-30 15:01:26, Disk, Threshold level: 20.00 MB/Sec, Measured speed: 18.29 MB/Sec
The disk test periodically samples the transfer speed every second (600ms when the duty cycle > 90). Every
sample below the threshold generates a warning.
The duty cycle for the disk test should be set to 100% when using this feature. Using a lower value doesn’t make
a lot of sense. Setting a threshold can be useful in helping to determine if the drive is functioning correctly and
the correct device drivers have been loaded.
BIT users manual
Page 25
Chapter 4 - Test Preferences
File size
The use can select the test file size and test pattern that is used with the disk test for each disk. The size of the
file is equal to a certain percentage of the disks capacity. The default file size is 1.0% of the disk size. So if a
disk had a total capacity of 200GB then the default size of each file would be 2GB. Setting a smaller percentage
results in more files being created on the disk and the read / verify cycle occurring more quickly.
Note that the file system type (NTFS, FAT32 or FAT) and the version of the Windows operating system (e.g.
Windows 2000) may limit the size of the file that can be created on a disk.
Log SMART errors
You may choose to have BurnInTest monitor the disk drives SMART attributes. If a SMART attribute for any of
the hard drives being tested exceeds its threshold, an error is logged along with the attribute exceeded. To learn
more about SMART see Appendix E: What is S.M.A.R.T?
Note: This feature is only available in Windows 2000/XP/2003 Server.
Seek count
For test modes that perform seeking to different positions on the disk drive (e.g. Random data with random
seeking), the seek count specifies the number of seeks for a particular iteration (e.g. After Random data with
random seeking has created 100 test files, the Seek count specifies the number of times a seek should occur
between blocks within these 100 files).
Duty cycle override
To use the general disk drive Duty cycle for each disk just set the Duty Cycle override value to blank (no value),
otherwise set the required value per disk.
Automatically select all hard drives
If you would like to test all fixed disks (hard disks) over multiple systems that have different drives letters, you
can select “Automatically select all hard drives” and all of the hard drives on the system will be selected prior to
running the test.
This option will not alter Floppy, CD or DVD disk drive preferences.
CD-RW/DVD preferences
CD/DVD Selection and test method
To edit the test preferences for a CD or DVD, select the Drive in the “CD/DVD Selection” List. Any changes
you make in the “Edit details for drive: <drive name>” will be reflected in this list.
Adding a CD/DVD drive
To specify that an optical drive should be tested, select a “Test Mode” from the drop down list. Only those CD
and DVD drives detected by the Windows operating system as displayed in the drop down list.
To remove an optical drive from the test, select a “Test Mode” of “NO TEST” from the drop down list.
Up to twenty drive letters can be selected for simultaneous testing. The CD burn test supports a maximum of 1
drive simultaneously The type of CD used in the actual test should match the setting in this window.
Test mode
Specifies the testing mode for the selected optical drive: NO TEST, Music CD playback, Data CD / DVD read
and verify, PassMark Test CD / DVD, No CD in Drive and Burn CD-RW.
Additional seeking and Seek count
For PassMark CD/DVD’s and Burn CD-RW only. When Additional seeking is selected, seeking to different
positions on the CD/DVD or CD-RW will occur Seek Count number of times for each cycle of file read/verify.
Quick Erase and Full Erase
For Burn CD-RW only. Erase the CD-RW media. The quick option erases content type information in typically
less than two minutes. The Full option additionally erases all user data on the disc, however this can take a
considerable amount of time (over 1 hour) and the test cannot be cancelled during this stage.
650MB and 700MB
BIT users manual
Page 26
Chapter 4 - Test Preferences
For Burn CD-RW only. CD-RWs are available in a number of sizes. The most common are 650Megabytes (MB)
and 700MB of data. This option allow the user to specify whether the CD-RW to be tested is supports 650MB or
700MB.
Test data drive
For Burn CD-RW only. PassMark format Test CD data is temporarily created on a harddisk before burning this
test data to the CD-RW. This test data is very large. By default, it will be created in the BurnInTest installation
directory, and is deleted after the CD-RW burn test. The target drive for this test data may be changed from the
“default installation directory” to the root directory of a hard disk on the system. This hard disk drive letter is
specified from the drop down “Test data drive” list. The selected drive will be saved as user configuration after
selecting “OK” (you do not need to add a CD drive to alter this drive configuration).
This option is provided for when BurnInTest is run from read-only media (CD/DVD’s), Floppy disks or other
disks with insufficient space and where the hard disk that BurnInTest is installed is to be tested by the
BurnInTest hard disk test (writing and reading a large temporary test file to a disk that is being tested with
BurnInTest should be avoided).
Blocking auto play
When a CD or DVD is inserted into a drive, Windows will attempt to auto-play the disc. In Windows XP a
window prompting the user to take some action is displayed even if the CD is just a data CD and can not be
played. The behavior can be prevented by selecting the ‘block removable storage auto play’ option in the CD
preferences window.
For this feature to work the following conditions must be meet.
- The main window of BurnInTest must be the active foreground window.
- Internet explorer version 4.0 or above must be installed.
- The auto-play feature must not already be disabled by some other means (e.g. changes to the registry)
Note: In some cases it is also possible to block the auto-play function by holding down the shift key when a disc
is inserted. (This is a feature of Windows and not BurnInTest).
Using the Registry to Disable AutoRun / Autoplay (Experts only)
There are two registry values that can be used to persistently disable AutoRun: NoDriveAutoRun and
NoDriveTypeAutoRun. The first value disables AutoRun for specified drive letters and the second disables
AutoRun for a class of drives. If either of these values is set to disable AutoRun for a particular device, it will be
disabled.
Note The NoDriveAutoRun and NoDriveTypeAutoRun values should only be modified by system
administrators to change the value for the entire system for testing or administrative purposes. Applications
(such as BurnInTest) should not modify these values, as there is no way to reliably restore them to their original
values.
The NoDriveAutoRun value disables AutoRun for specified drive letters. It is a REG_DWORD data value,
found under the
HKEY_CURRENT_USER\Software\Microsoft\Windows\CurrentVersion\Policies\Explorer key. The first
bit of the value corresponds to A:, the second to B:, and so on. To disable AutoRun for one or more drive letters,
set the corresponding bits. For example, to disable the A: and C: drives, set NoDriveAutoRun to 0x00000005.
The NoDriveTypeAutoRun value disables AutoRun for a class of drives. It is a REG_DWORD or 4-byte
REG_BINARY data value, found under the
HKEY_CURRENT_USER\Software\Microsoft\Windows\CurrentVersion\Policies\Explorer key. By
setting the bits of this value's first byte, different drives can be excluded from working with AutoRun.
The following table gives the bits and bit mask constants that can be set in the first byte of
NoDriveTypeAutoRun to disable AutoRun for a particular drive type. For Microsoft Windows 2000, you must
restart Windows Explorer before the changes take effect.
Bit Number
Bit mask Constant
Description
0x04
DRIVE_REMOVEABLE
Disk can be removed from drive (such as
a floppy disk).
0x08
DRIVE_FIXED
Disk cannot be removed from drive (a
hard disk).
0x10
DRIVE_REMOTE
Network drive.
0x20
DRIVE_CDROM
CD-ROM drive.
0x40
DRIVE_RAMDISK
RAM disk.
BIT users manual
Page 27
Chapter 4 - Test Preferences
Logging preferences
Logging to the Detailed Error log history
Errors are automatically logged to the Detailed Error log history. This log is viewed using “View”, ”Error Log
File”. This type of logging is not affected by the “Logging on” option, which refers to automatic disk logging.
Note: Text and HTML reports can be manually created using “File”, "Save report”. This will save System
information, a result summary and the recent Detailed Error log history (up to 1000 entries). A customer style
test certificate is also available.
Automatic disk logging (to disk)
Logging to disk is switched on and off with the “Logging on” option. When disk logging is on, BurnInTest will
log System information, results, result summaries, and detailed error descriptions to files based on the following
user settings.
The log directory allows the directory and filename prefix to be specified. For the default of “Time stamped
files”, the file will automatically be appended with _YYMMDD_HHMMSS.log, .htm or .trace, specifying the
Year, Month, Day, Hour, Minute & second that the disk log file was created.
Specify “Log name prefix” if you wish to prefix the standard BurnInTest Text or HTML log filenames you can
type a prefix string into the “Prefix” edit box. For example, if you want to prefix the standard BurnInTest log
filenames with the text “SPECIAL_TEST_” then just enter,
SPECIAL_TEST_
If you want to prefix using an environment variable, %COMPUTERNAME%, %USERNAME%, %OS%,
%PROCESSOR_ARCHITECTURE% are supported. So, if for example you wanted to prefix log files with a test
run description, the computer’s name and the user’s name, just enter,
Test123_%COMPUTERNAME%_%USERNAME%_
A “Log detail level” may be selected for disk logging to specify the amount of information to be written to disk.
The options that may be selected are:
1. Result Summary only. Produces a .log or .htm file containing system information, a summary of the
tests run (e.g. Which tests, the number of operations and errors per test) and a summary of the serious
and critical errors during a test run (cleared at the end of a non-scripted test run or at the end of the
script for a scripted test run).
2.
Normal. Produces a .log or .htm file containing the “Result Summary only” information and detailed
errors (1 or 2 lines per error).
A “Trace level” may be selected for detailed disk logging to specify the amount of information to be written to
disk. The options that may be selected are:
1. No Trace log. No trace log is produced. This is the default setting.
2.
Activity Trace level 1. Produces a .trace with the “Normal” information and a summary of test activities
(e.g. The starting and stopping of tests).
3.
Activity trace level 2. Produces a .trace file with the “Activity Trace level 1” information and a detailed
description of the test activity.
Setting “Max file size” specifies the approximate maximum file size in terms of the number of lines in either the
log or trace files.
The format of the Result summary and Normal log files may be specified to be either ASCII text or HTML via
the “ACSII Format” or “HTML Format” buttons. Trace logs are always in ASCII text.
By default, time stamped log files are created, this is the “Time stamped files” option. A new set of log files will
be created when Logging On is switched on, any of the logging options (Log Directory, Detail Level,
ASCII/HTML Format) are changed, or when BurnInTest is started with Logging switched on. The only
exception is when BurnInTest is automatically stopped with a REBOOT or REBOOTEND scripting command,
and then the original log files will be re-opened and appended to.
The log files are closed and available for use by another application when Logging On is switched off, any of the
logging options (Log Directory, Detail Level, ASCII/HTML Format) are changed, or when BurnInTest is exited
or automatically stopped with a REBOOT or REBOOTEND scripting command.
By specifying “Single file”, no date or time stamp is appended to the filename and only a single log file (and if
specified, a single trace file) will be created and all results will be logged to this file. “Single file” may be a
useful option when the output is to be processed by an external program.
Periodically log result summary reports during a test
When logging is switched on, by default a result summary report will be logged at the end of a test run. Change
this value if you would like to log interim summary reports during the test (every X minutes). The main
requested use of this option is to help obtain information about the testing status of a system prior to a system
BIT users manual
Page 28
Chapter 4 - Test Preferences
crash on an unstable system. Note: All error events are written in real time, so the main use of this option will be
if there are few errors occurring.
Pre-test logging options for test result clearing and log file creation
Test results, including the detailed error and status log history, can either be cleared each time a new test run is
started or test results can be accumulated across several test runs. Accumulating test results is useful when doing
scripted testing, including, reboot testing. If test results are accumulated, the only way test results can be cleared
is with the ‘Edit / Clear All Results’ menu item.
Accumulating test results means that the error count, cycle count and operations count values are the total values
that have been accumulated across all test runs on this machine.
Accumulation will also occur across scripted reboots, sleeps and execute commands.
Log files may be appended to the existing log file or a new log file created at the start of each test run.
Three options are available for clearing/accumulating test results and creating/appending log files. These options
are:
• Clear test results and create a new log
•
Clear test results and append to the existing log
•
Accumulate test results and append to the existing log
Note: Regardless of the configuration settings, the detailed error and status log history is maintained at a
maximum of 1000 lines, with the oldest data cleared if this size is exceeded.
See also
Calling external programs via Auto stop actions (see § 4 “Timer and test duration preferences”).
Maintaining multiple test configurations (see § 8.1 “Maintaining multiple test configurations” p.35).
Error handling preferences
Action on Error
When BurnInTest detects an error, three options are available.
1. Continue. The error counter is incremented and testing continues.
2. Continue and beep. The error counter is incremented, the computer beeps and testing continues.
3. Stop all tests. All windows (except the main window) are closed and all the tests are stopped. The failed
window is then displayed.
Windows errors (Abort / Retry / Cancel)
You can specify whether BurnInTest should block or allow critical Windows errors.
Network preferences
(The Standard Network Test is available in the Standard and Professional editions of BurnInTest. The Advanced
Network Test is available only in the BurnInTest Professional edition.)
Standard Network Test and Advanced Network Test
Use these radio button to switch between using the standard and advanced network tests. This option is only
available in the BurnInTest Pro version.
Standard Network Test
Network addresses
Up to 4 network addresses can be entered for the simultaneous testing of the connection to 4 different locations.
These values are used for the Network test. These connections can pass via different network interface cards
(NIC) or the same one. The NIC used will depend on the TCP/IP configuration set-up in Windows. Each address
must be a URL or an IP address.
A URL is the name of a network host, e.g. www.hostname.com
An IP address is a sequence of 4 numbers that correspond to a network host. e.g. 169.192.0.1
If less than 4 addresses need to be tested uncheck the active check box.
The host selected must be accessible from the computer and capable of responding to the ‘ping’ command. See
the Network Test description (section 3) for more details. PassMark recommends the selection of a local host to
minimize data link problems, which are fairly common on the Internet.
Timeout
BIT users manual
Page 29
Chapter 4 - Test Preferences
This value determines how long BurnInTest will wait for a data packet to be sent or received before an error is
reported. The value is measured in Milliseconds. 2000ms (2 seconds) is the default value.
Error handling
There are two options for handling errors that occur during the execution of the network test. BurnInTest can
either log an error for every bad (or missing) packet or can be set to only log an error when the number of errors
exceed a threshold set by the user. This threshold is expressed as the percentage of bad packets compare to the
overall number of packets sent.
For more information see the description of the Network Test description (section 3).
Edit Options (Advanced Network Test)
This option is only available in the BurnInTest Pro version.
When the Advanced Network Test option is selected at the top of the Window, the “Edit Options” button will be
available. Clicking this button will open the Advanced Network Test Options Window and allow the options for
the advanced test to be set.
Advanced Network Test Options
(Applicable to BurnInTest Pro version only)
Select local network card for testing
This list will show the IP of up to 10 network cards that are installed on the system and plugged in. The values in
the fields below are saved for each network card and can be different for each card.
Test On/Off
Check this box the enable testing for the selected IP.
Protocol
Select the protocol to use for testing the card, TCP, UDP, FTP or Cycle. Cycle will test each protocol for 15
minutes before moving on to the next one.
Mode
Full duplex – will send and receive data during the test.
Half duplex – will only send data during the test.
Transfer
Max MB - when testing a file is created in memory (or on the disk for FTP) of this size to send over the network.
Random file size - if this is checked then the file created will be a random size up to size specified in Max MB.
Data Validation – if this is checked then the data sent during the test will be validated , this may slow down the
testing due to the CPU overhead of validation.
Log file update interval
A separate log is kept for each network card being tested and this will be updated approximately by the interval
entered here, minimum interval is 10 seconds.
Port Number (Data)
This is the starting port number that is used when sending data during the test, as the testing progresses more
then one port may be in use by each network card. Unless it conflicts with other services in use this should be
left at the default value.
Transfer Block Size
This value is buffer size used when breaking the file into packets and sending it across the network.
Max Load % on NIC
This is the required load placed on the network card during testing, the test will be throttled to maintain the
necessary amount of data being sent through the network card. For example, a load of 15% for a 1000 megabit (1
gigabit) network card will be a transfer rate around 17 MB/s.
UDP loss % warning
During a UDP test if the percentage of lost packets rises above this limit then an error is generated.
Temperature and Battery monitoring preferences
Battery monitoring (via BatteryMon)
BurnInTest can work with another product from PassMark Software called BatteryMon. BatteryMon allows the
charge level in laptop batteries and uninterruptible power supplies UPSes to be monitored. Click the check box
to turn on the feature.
See Temperature and battery monitoring (section 8) for more details.
BIT users manual
Page 30
Chapter 4 - Test Preferences
Temperature monitoring
PassMark software does not produce a temperature monitoring application, however we do provide support for
many third party system monitoring applications currently available. By using BurnInTest with these
applications, is it possible to Burn in your hardware while monitoring and logging the temperature of your
CPU(s) and/or your motherboard.
Make sure temperature monitoring is turned on in BurnInTest (from the Test preferences Window) and if
required, the temperature source value is selected. For Hmonitor, up to two sources can be selected for display
and logging in BurnInTest.
See Temperature and battery monitoring (section 8) for more details.
Parallel port preferences (BurnInTest Pro only)
Parallel port (BurnInTest Pro only)
The communications port that will be used by the parallel port test, e.g. LPT1, LPT2 or LPT3 (only ports
currently installed will be displayed). See the parallel port test for more details. See the parallel port test (section
3) for more details.
Only a single parallel port may be tested at one time.
Test type
It is possible t select either a “Detect and loopback test” or a “Detect only” test. The, detect and loopback test,
requires the use of parallel port loopback plug (see Appendix D). The, detect only, test does not require a
loopback plug. When possible use the loopback test as it is a more thorough test.
2D Graphics Test / EMC Testing (BurnInTest Pro only)
BurnInTest provides the ability to display a screen of Scrolling 'H' characters and the possibility to adjust,
The color of the text
The color of the background
The Scroll speed (via the duty cycle window, slide bar)
The character used, as a replacement for 'H'
The font face and font size
The 2D Graphics preferences window can be used to either select the standard 2D graphics video memory test,
the 2D graphics GDI functions load test, or the EMC testing scrolling ‘H’ pattern.
USB port preferences (Pro version only)
Maximum number of USB ports
This is the maximum number of USB devices that BurnInTest attempts to connect to. In general you will want to
set this value to be equal to the number of USB loopback devices that you would typically have connected to the
machine. A maximum of 10 plugs may be used at the same time. If there are more plugs connected than
configured in this section, USB ports will be tested based on the following: Ports with USB2 loopback plugs
attached are first allocated for test (lowest device number to highest), followed by ports with USB1 loopback
plugs attached (lowest device number to highest).
USB Loopback plugs detected
This is a list of the loopback plugs that have been detected as being connected to the PC. Examples are:
PMUSB2-0: 480Mb/s (PASSMARK USB2Test, Serial #:PMK5C0KV, V2)
PMUSB2-1: 12Mb/s (PASSMARK USB2Test, Serial #:PMK5C0SD, V2)
PMUSB1-0: 12Mb/s (USB Loopback plug, Serial #:PM29K8PH)
PMUSBn-m specifies whether the USB port has a Passmark USB loopback plug connected to it, where n
specifies whether the plug is a USB1 or USB2 loopback plug, and m specifies the device number. The device
number is just a sequential number (for USB1 and separately for USB2) that is incremented for each plug
connected.
BIT users manual
Page 31
Chapter 4 - Test Preferences
The maximum speed supported by the connected protocol for this Port is then displayed (Note (1) that the 2nd
port in the list has a USB2 loopback plug attached, but the port only supports USB1 and hence only the lower
maximum speed of 12Mb/s is available and displayed. (2) This is not a benchmarked speed, but based on the
protocol the plug has been able to connect to the PC with. (3) Benchmarking speed can be undertaken using the
USB2Test software.
The text after the Maximum Speed is the description of the device that was found in the memory of the plug.
This text will normally be, " USB Loopback plug " for a USB1 loopback device and “PASSMARK USB2Test”
for a USB2 loopback plug.
The serial number is a unique identification number that is different for each plug. A typical serial number would
be, " PMK5C0KV ". This unique serial number is stored in the plug.
For USB2 loopback plugs, the Version of the firmware is then displayed.
USB Transmission timeout
Only applies to the USB1 loopback plug.
The user may specify the transmission timeout (in milliseconds). If a reply hasn’t been received after this time,
an error will be logged.
RAM preferences
RAM Test mode
Standard memory test
The standard memory test allocates RAM from the Windows operating system. BurnInTest then uses the virtual
address range supply by Windows. Having RAM allocated in this way means that Windows can swap these
RAM pages out to disk and generally move them around in physical memory. As BurnInTest accesses each
memory page frequently during a test, this swapping does not happen very much in practice but it does create a
small amount of uncertainty about what is actually being tested. Because of the limitations of the Windows
virtual address space and fragmentation, only around 1GB of RAM can usually be tested using this method with
the 32bit version of BurnInTest. This limitation is removed with the 64bit version of BurnInTest.
Torture Test (Pro version only)
The torture test is a multi-process memory test. Multiple processes are started in their own virtual address space
and each process allocates and tests a block of RAM. Disk swapping and the over-allocation of RAM can be
tested using this method. The number of test processes to start and the percentage of total RAM that each process
will use can be specified by the user.
Start 32bit specific
Addressing Windows Extensions Test (Pro version only)
The Addressing Windows Extensions memory test directly allocates physical RAM and locks it down to avoid
any inference from Windows. The memory allocated is effectively removed from the Windows operating system
and can never be swapped or moved (until unlocked). It is still not possible to use all of the installed RAM, as
Windows still needs to run. However this is a much better state of affairs as compared to the standard test. There
is also a second advantage with the Advanced test. It can use, in theory, up to 64GB of RAM. The standard
version of Windows2000 supports 4GB of RAM. Windows Advanced Server and Windows Data Center Server
support 8GB and 64GB respectively.
The disadvantage of the Advanced memory test is two fold. Firstly it can only be used with Windows2000 and
XP. Secondly you must manually assign an additional administrator right to lock pages in RAM.
End 32bit specific
Test pattern
The Test pattern may be selected to be one of the following, or to cycle through each of the above in the above
order:
Sequence (0,1,2...255)
Binary 1 (10101010)
Binary 2 (01010101)
Zeros (00000000)
Ones (11111111)
BIT users manual
Page 32
Chapter 4 - Test Preferences
Log memory allocations
Turning this option on will cause BurnInTest to log memory allocations (and de-allocations) into the detailed log
file. This is mainly useful for debugging purposes.
Video playback preferences
The Video Playback preferences window allows you to specify up to 3 video files. These may be typed directly
into each of the 3 edit boxes (full directory path and file name required), or you can browse to the required video
file by selecting the browse “…” button.
1.Video File Path
2.Video File Path
3.Video File Path
The order the videos are specified, is the order they will be played.
Video’s in the file formats .avi, .mpg & .mpeg are supported.
Sample test videos with different file formats and video compression are available from the PassMark Web site.
To automatically register three samples video files within BurnInTest (rather than typing them in manually)
please follow the following steps:
1. Download the videos from the PassMark website.
2.
Copy the files into the BurnInTest directory.
3.
If you have set up other test video files in preferences, delete these from within Preferences, Video
Playback, by simply highlighting the text in each edit box and pressing delete on your keyboard.
4.
Exit BurnInTest
5.
Start BurnInTest
If these PassMark Video sample files exist in the BurnInTest directory, you will now see these are configured in
Preferences, Video Playback.
Steps 3 & 4 can be skipped if no video files have been previously configured in BurnInTest, Preferences, Video
Playback.
If the PassMark video samples are not installed in the BurnInTest directory. BurnInTest will attempt to find the
Microsoft Windows “Clock.avi” video, and configure this as Video file number 1.
Tape preferences (BurnInTest Pro only)
Tape drive
The tape drive ID can be entered into this field. Windows normally assigns the name of the tape drive. Typical
names are TAPE0, TAPE1, etc..
File size and Number of files
The number of test files and the size of each file can be selected here. The size of the tape must be large enough
to hold the files selected.
Option: Formatting
The formatting step tries to create a new partition on the tape. The size of the partition will depend on the tape
drive in use. Some tape drives may not support this option as not all media types require this formatting step to
be executed before the tape media is used.
Option: Locking
The locking step attempts to physically lock the tape in the drive, so that it cannot be manually ejected during the
test. Some tape drives may not support this option.
Option: Writing file marks
File marks are used to separate files on a tape. Normally a file mark is a short section of blank tape. Different
tape drives support file marks that are different in length and content. File marks are not required to run this test
as BurnInTest knows the exact length of each file written to the tape and thus has no need to be notified when
the end of a file is reached. If you get file mark errors when running the tape test, turn this option off as your tape
drive probably doesn’t support the type of file marks used by BurnInTest.
BIT users manual
Page 33
Chapter 4 - Test Preferences
Drive info button
This window display information that has been extracted from the tape drive. It allows the user to see which
features are supported by a particular tape drive. This window is really only intended for tape drive experts.
Media info button
This window display information that has been extracted from the tape drive and its current media. It allows the
user to see some information about the media currently in the drive. This window is really only intended for tape
drive experts.
CPU preferences (BurnInTest Pro only)
Only systems with multiple CPUs support this option.
CPU affinity
Normal task scheduling or CPU targeted testing can be selected here. Normal task scheduling will result in
threads being dynamically scheduled on the CPU with the least load. In this way the load is spread between all
available CPUs. CPU targeted testing schedules all the threads onto a particular CPU.
Note that CPU selection is only possible if more than one CPU is installed and the user has administrator access
rights.
Even if CPU targeted testing is used some load can “spill” onto other CPUs. This is because system processes
will still be allocated to other CPUs.
Serial port preferences (BurnInTest Pro only)
Adding a serial port
To select a serial port for testing, select the COM port number from the right hand drop down list, select the test
type, then click on the add button. Up to 64 ports can be selected for simultaneous testing.
Removing a serial port
To remove one or more a serial ports, select the COM port(s) from the left hand list then click on the remove
button.
Serial port drop down list
The communications ports that will be used by the serial port test, e.g. COM1.
Test type
For each serial port it is possible to select either a “Detect and loopback test” or a “Detect only” test. The, Detect
and loopback test, requires the use of serial port loopback plugs (see Appendix D). The detect only test does not
require a loopback plug. When possible use the loopback test as it is a more thorough test.
Send and Receive timeouts
This is the amount of time in milliseconds that BurnInTest will wait for the successful completion of a Serial I/O
operation.
Disable RTS/ CTS and DSR / DTR
The checkbox will disable the part of the serial port that checks that the signal pins are functioning correction.
RTS = Request to send pin
CTS = Request to send pin
DSR = Data set ready pin
DTR = Data Terminal ready pin
This can be useful to avoid errors when testing non-standard serial ports.
Port speed
The communications speed can be selected. Note that most serial ports do not support speeds above
115200bits/sec.
See the serial port test (section 3) for more details.
BIT users manual
Page 34
Chapter 4 - Test Preferences
Sound preferences
Sound test type
Select the type of test you wish to carry out.
Wave File; Midi and MP3 File
Browse to select the files which will be used during the standard test, or enter the path to the audio file.
Max Distortion
Select the maximum distortion allowable for the loopback test. The values available are between 1 and 20.
For more information see the description of the Sound Test.
Maths preferences
Select maths tests
Using the check boxes it is possible to select which maths tests will be performed and which ones will be
skipped. See the maths test for more details.
Select SIMD tests
Using the check boxes it is possible to select which Single Instruction Multiple Data (SIMD) tests are performed
and which ones will be skipped. See the MMX /SSE test for more details.
Print preferences
Standard test
The standard test sends pages of text to the default Windows printer. The pages printed by the printer should be a
page of solid text in a fixed-point font filling the printable area of the page. There should be a small margin
(usually under 1cm) around the block of text.
Standard color printer test
The standard color test sends pages of color text to the default Windows printer. The pages printed by the printer
should be a page of solid color text in a fixed-point font filling the printable area of the page. There should be a
small margin (usually under 1cm) around the block of text. The colors are as follows:
Title: Black
Repeated blocks of:
Red
(RGB: 0xFF0000)
Red
(RGB: 0x7F0000)
Green (RGB: 0x00FF00)
Green (RGB: 0x007F00)
Blue
(RGB: 0x0000FF)
Blue
(RGB: 0x00007F)
Printer command file
The user can define a printer command file to be used in place of the standard test. A printer command file is a
file containing a series of printer commands which is used by the printer to carry out certain actions.
How to create a printer command file
Print cover page before sending job to printer
The user may choose to print a cover page before sending the command file to the printer. This may help
troubleshooting possible command file, as opposed to, printer errors.
There is an example printer command file included with BurnInTest. The example file is called “HP Printer
file.pcl” and can be found in the BurnInTest installation directory. This file is a list of printer commands used
with HP inkjet printers. It moves the print head without actually printing any pages.
The language used for each make of printer is different, but you can still make your own printer command file.
You need to create a document, then print this document to a file using the correct printer device driver.
For example in Microsoft word, when printing your document select the “Print to file” checkbox in the Print
window. Once a file is created it can then be used with BurnInTest.
Normally printer command files have a ‘.prn’ file name extension but this may vary depending on the printer
language used.
BIT users manual
Page 35
Chapter 4 - Test Preferences
PCL = Printer command language (from Hewlett Packard)
PS = Postscript.
Pre-test preferences
Action before running tests
You may specify the default behavior of BurnInTest before it begins any test cycle.
None (start immediately)
Immediately starts test(s) without any warning message or reminders.
Display warning message (default)
Displays the default warning message reminding the user of some basic precautions, such as possible port
conflicts and the presence of loopback plugs and cables. Note: Pre-test warnings are not provided for Plugin
tests.
Run external application
Run another application prior to starting the test run. For instance the user may wish to launch the Windows
Performance Monitor to examine the effects of the test on the system.
Run external application &wait for it to exit (Passing configuration information from an external program to
BurnInTest).
Run another application prior to starting the test run. BurnInTest will wait for this application to exit before it
continues. On continuing, BurnInTest will check for a subscript file (see below). The executable file may
produce a file of scripting commands that are to be run by BurnInTest once the executable file has closed. As an
example, this may be used to set the Machine type, serial number, test notes etc at the start of each individual test
run.
The only script commands that will be processed by BurnInTest are the SET… commands. All other commands
will be ignored. If this file, called a subscript file, is created is must:
- be placed in the same directory as the application executable file;
-
must be called “bit-script-input.txt”; and
-
must conform to the scripting file format.
After processing the script commands in this file, BurnInTest will delete the file.
See also, EXECUTEWAIT
External Application
The path to the external application above.
Post-Test preferences
There are different actions that may be taken based on whether a test Passes or Fails, and whether the test is
automatically stopped (i.e. the test has run for the set duration) or manually stopped by the user pressing the Stop
button.
Auto Stop
Actions specific to automatically stopped tests are described below.
Option1: Stop Tests (Normal behavior)
This option will stop all the running tests and close all test windows. The Result window indicating a PASS or
FAIL will be displayed. This is the normal default setting.
Option2: Stop Tests and print results
This option will stop all the running tests and close all test windows. The results will be sent to the default
printer.
Option3: Run external application and exit
To help with the process of test automation it is possible to have BurnInTest run an external application, and exit
automatically, after the tests have been completed. This could be used, for example, to call an application or
batch file that would undertake a different type of testing (e.g. PassMark PerformanceTest) or customer specific
reporting.
BIT users manual
Page 36
Chapter 4 - Test Preferences
Option4: Exit and Reboot PC
This option is useful for reboot cycling testing. Once configured BurnInTest can be setup so that the PC will
reboot itself in a cycle, with test runs occurring between each reboot.
Manual Stop
Actions specific to manually stopped tests are described below.
Option1: Stop Tests (Normal behavior)
This option will stop all the running tests and close all test windows. This is the normal default setting.
Option2: Display option to run application
This option allows tests to be stopped prior to reaching the test duration, while maintaining a post test process
that is consistent with the automatic stop actions (such as customer specific reporting). BurnInTest can display
an option to the user asking whether they want to run the Post test application after a test run is manually
stopped. If the user selects to run the external application, BurnInTest will exit after running the application. If
the user selects not to run the external application, then BurnInTest will not exit.
Other Options
Always display result Window
By default, this option is unchecked, and the result window indicating PASS/FAIL is only displayed after the
Auto Stop, Stop Tests option. The user may specify to display this result window for the other options (i.e.
manually stopped tests, and running an external application). If this option is checked the result window will be
displayed after any external application is run, and before exiting BurnInTest.
External Application
This specifies the application to be run if the appropriate option above is selected.
Syntax:
< Filename> <Parameter List>
Parameters:
< Filename >
The name of the executable file. The file name must be enclosed in double quotes
(“”). The file must be a batch or executable, and the “.bat” or “.exe” must be
included.
< Parameter List > An optional list an parameters to be passed to the application. Each fixed parameter
must be enclosed in quotations.
A number of variable parameters are supported. These parameters will be substituted
with fixed strings enclosed in double quotes (“”). These are:
$RESULT - “PASS” or “FAIL” will be substituted if the test is stopped
automatically. If the test is manually stopped by the user, “PASS (manual abort)” or
“FAIL (manual abort)” will be substituted.
$SERIAL – The serial number will be substituted.
$MACHINETYPE – The machine type will be substituted.
$NOTES – The notes will be substituted.
Examples:
#Set plug-in test application file with full path
c:\temp\cleanup.bat
c:\temp\cleanup.exe $RESULT $SERIAL $MACHINETYPE $NOTES
#Set plug-in test application file with path relative to BurnInTest drive
\temp\cleanup.exe
#Set plug-in test application file with path relative to BurnInTest application path
cleanup.exe
Reboot options
Selecting this button will open the “Passmark Rebooter” application to set the reboot configuration.
BIT users manual
Page 37
Chapter 4 - Test Preferences
Test Preferences – “Plugin”
You can specify up to 3 external Plugin test applications.
Application
Specifies the plugin filename and command line arguments. See Available Plugin’s below for examples.
Run as pre-test plugin
Some Plugin tests may require manual user testing, for example, testing a keyboard. You can specify that a
plugin be run before the main set of tests, such that the Plugin test will run, exit on completion of that test, and
then the main set of tests will be run.
Plugin’s available from Passmark
PassMark has developed plugins for Modem, Keyboard and Firewire port testing.
Modem testing
When PassMark’s ModemTest is used as a BurnInTest Plugin, it tests in the Local Diagnostics mode. This mode
requires only a single modem and no telephone line. ModemTest requests Identification information from the
Modem (via a series of AT commands) and then if supported by the modem starts a local loop back on the
analog side of the modem, ModemTest then sends data packets which are echoed back by the modem.
Version of Plugin required: PassMark ModemTest V1.3 1010 (or higher)
Test name within BurnInTest displays and reports: Modem.
Test information displayed in BurnInTest:.Status text, number of operations, errors and cycles, error text and
severity.
Application field should be set to:
ModemTest.exe /p <COM port number> <Results file name> <Config file>
Where:
/p specifies ModemTest to run as a plugin,in local diagnostic mode;
<COM port number> is the serial port the Modem is asociated with;
<Results file name> is a ModemTest specific result file;
<Config file> is a ModemTest specific configuration file.
e.g.
C:\Program Files\ModemTest\ModemTest.exe /p 1 results.log "C:\Program
Files\ModemTest\ModemTest.cfg"
Note: any arguments that contain a directory path, must be enclosed in
quotation marks.
Run as pre-test plugin, should be unchecked.
Notes:
1) You should install ModemTest before starting.
2) You should license ModemTest so that the “Welcome” window is not displayed when the BurnInTest
tests are started.
3) You should set up a configuration file and save it in the ModemTest directory before starting.
Keyboard testing
When PassMark’s KeyboardTest is used as a BurnInTest Plugin, it tests in the Batch mode without collecting the
manually entered information such as Keyboard Serial number. This mode allows automatic detection of all keys
being successfully pressed, of User failure with User selected reasons. The keyboard can be failed at any time
during the test or on requested user verification. A reason from a list of failure reasons will also be logged in
BurnInTest.
It should be noted that the KeyboardTest Keyboard Layout file (.kbl) needs to describe the keyboard you plan to
test. This file describes all of the keys on the keyboard and is used (among other things) to determine when all
the keys have been pressed and hence the trigger for automatically passing the keyboard. You may be able to use
one of the default keyboards shipped with KeyboardTest, you may find a matching keyboard on our website
Keyboard Layout Download Page or you may have to create your own .kbl file as described in the KeyboardTest
help file.
Version of Plugin required: PassMark KeyboardTest V2.2 1011 (or higher)
Test name within BurnInTest displays and reports: Keyboard.
Test information displayed in BurnInTest:.Status text, number of operations, errors and cycles, error text and
severity.
BIT users manual
Page 38
Chapter 4 - Test Preferences
Application field should be set to:
KeyboardTest.exe /p
Where:
/p specifies KeyboardTest to run as a plugin,in a modifed batch mode
e.g.
C:\KeyboardTest\KeyboardTest.exe /p
Run as pre-test plugin, should be checked.
Notes:
1) You should install KeyboardTest before starting.
2) You should license KeyboardTest so that the “Welcome” window is not displayed when the BurnInTest
tests are started.
3) You should set up a Keyboard configuration file and bitmap of your test keyboard before starting.
Please refer to the KeyboardTest help file.
Firewire testing
Firewire port testing, using a Passmark developed plugin and a “Kanguru FireFlash” drive.
Version of Plugin required: PassMark Plugin V1.0 1000 (or higher)
Test name within BurnInTest displays and reports: Firewire.
Test information displayed in BurnInTest:.Status text, number of operations, errors and cycles, error text and
severity.
Application field should be set to:
Firewire.exe [/a|/k <drive letter>]
Where:
The parameter specifies the Firewire port to test based on the drive
letter of the “Kanguru FireFlash” drive
/a select the first drive letter with a “Kanguru” drive;
/k <drive letter> select the drive specifed in <drive letter>;
<drive letter> is a single character, C to Z.
e.g.
C:\FireWire\Firewire.exe /k D
C:\FireWire\Firewire.exe /a
Run as pre-test plugin, should be unchecked.
Notes:
1) You should install the Firewire plugin before starting.
2) You should use a “Kanguru Fire Flash” drive with this plugin. This should be plugged in before starting
the BurnInTest test.
BIT users manual
Page 39
Chapter 5 - Test section & duty cycle
Automatic test stop, Test selection & Duty cycles
From this window it is possible to configure the automatic stopping of tests, select the tests to be performed and
the level of load placed on the system.
The options for Tape, Serial, Parallel &
USB port testing will only be displayed in
the professional version.
Auto-Stop
BurnInTest can be set to automatically
stop a test run by either configuring a test
duration or a number of test cycles.
Test duration
This option allows the user to select the
period of time that tests are to run for.
After the time has expired all windows
(except the main window) are closed and
all the tests are stopped.
The duration, in minutes should be
entered in the “Auto Stop after n
Minutes” field. A value of 0 in this field
means that the tests will run until a manual Stop command is issued via the button bar/menu or the number of
test cycles has been reached. The maximum test time in the unregistered shareware version is 15minutes per test.
After the software is registered the maximum auto stop time is 10 days. (the registered version can also run
forever if the manual stop option is selected).
Number of test cycles
This option allows the user to select the number of cycles that tests are to run for. After all current tests have
undergone the configured number of cycles in testing, all windows (except the main window) are closed and all
the tests are stopped.
The number of cycles should be entered in the “or n Cycles” field. A value of 0 in this field means that the tests
will run until a manual Stop command is issued via the button bar/menu or the test duration has been reached.
NOTES:
1) Both options are based on the current test run and not accumulated results.
2) The test duration and number of cycles are both used to automatically stop a test run (with a logical OR). For
example, if duration is set to 15 minutes and number of cycles is set to 3. The test run will be stopped when the
first of either 15 minutes OR 3 cycles is reached. If you want to automatically stop a test run based only on
duration, set the number of cycles to 0 (i.e. Run for the set duration OR forever). If you want to automatically
stop the test run only based on the number of test cycles, set the test duration to 0 (i.e. Run forever OR for the set
number of cycles).
3) If you need to run each test a single time, one after the other (i.e. In series, rather than in parallel), a script
should be created to: Set the number of cycles, run test1, run test2, …, run test n. See SETCYCLES.
BIT users manual
Page 40
Chapter 5 - Test section & duty cycle
Test Check boxes
Each test has an associated check box that can be used to turn the test on or off. Your CPU must support the
specific CPU instructions sets (e.g. MMX), for the check box to be activated.
Slide bars & Duty cycle
Each test has a slide bar and an associated edit box. The slide bar allows the user to determine the “duty cycle”
for each test. A low duty cycle means that a delay will be inserted during the execution of the test, reducing the
load on the system and reducing the number of operations performed during any particular period of time. A
high duty cycle corresponds to higher load.
A value can also be directly entered into the edit box. The background color of the edit box will vary from dark
red to white to bright green depending on the duty cycle selected.
Buttons
The buttons at the bottom of the window allow all tests to be activated or deactivated with a single click. The
default values can also be restored. The settings selected are saved when the OK button is clicked.
Selecting which tests to run
In order to help select which tests to run, here are some general guidelines.
Only select tests that match your hardware
The tests you select should correspond to the hardware installed in the PC. For example, if your computer
doesn’t have a tape drive installed, there is no point having the tape drive test enabled. This will only result in a
lot of errors being generated.
Selecting tests for specific testing
If you suspect a problem with a particular device, (e.g. the RAM), leave the other tests turned off and just run
this particular test at 100%. This will maximize the load on this element.
Selecting tests for general burn in testing
For a general burn in, select a variety of different tests. The RAM and Disk tests are the most important. Select
the other tests based on how you plan to use the computer. For example if the machine is to be used as a server,
then the Network test, Tape drive test, CPU tests and CD test should also be enabled. In general it is better to test
those elements that will receive the most usage once the machine is put into active use. For example the floppy
drive test could be left off, (or tested at a low duty cycle), if the floppy drive is not critical for the machines
intended use.
Optimize the load
As all the tests run at the same time in different threads1 . Some care should be taken to ensure that important
tests are not starved of the CPU and run too slowly. Thus it can be advantages to initially leave the CPU tests off
and run with the other tests. Then add in the CPU tests but adjust the duty cycle down until the CPU load just
hits 100%.
For example, assuming that the CD and disk drive are critical parts of your system, run just these tests at 100%
duty cycle, then note the load on the CPU 2 . Then add in the RAM and CPU tests at a lower duty cycle in order to
fully load the CPU.
Note that there is no point trying load up the CPU to more than 100%. Adding more load once the CPU is
running at 100% doesn’t result in any more processing being done. The CPUs available processing time is just
redistributed and all the running processes run more slowly.
Experiment
1
A thread is a section of an application that can run in parallel with other sections of the same application. Each
thread can run simultaneously on multiple CPU’s or share the same CPU. For example the disk test thread can be
reading the disk will the RAM test continues to read data from memory.
2
The CPU load can be seen from the Windows Performance Monitoring tool. (which comes with Windows).
BIT users manual
Page 41
Chapter 5 - Test section & duty cycle
As just about everyone has different requirements don’t hesitate to experiment with the settings to obtain the best
result in your environment.
BIT users manual
Page 42
Chapter 6 - Interpreting the results
Interpreting the results
Status display
In the main window there is a
display that shows a summary of
all the results of all the tests that
are currently running.
Test Name
This column shows a picture
depicting the test type and the
name of the test. Only those tests
actually running are displayed.
It should be noted that:
(i) The USB test picture will be
of the PassMark USB2 loopback
plug if a USB2 loopback plug is
connected, otherwise it will be of
a picture of the PassMark USB1
loopback plug. The USB pictures
also have an “F” (FullSpeed) or
“H” (HighSpeed) indication to
the bottom right of the picture
displayed. These indicate the
speed that the USB port has
connected at, either FullSpeed
(up to 12Mb/s) or HighSpeed (up
to 480Mb/s).
(ii) The CD/DVD picture has a
“B” indicator if the CD test is a
CD Burn test.
Cycle
The number of test cycles that have been executed for a particular test. The meaning of a ‘test cycle’ varies from
test to test. For example for the Printer test it is the number of full pages printed, for the Hard disk test it is the
number of file write / verify cycles that have occurred. See the test description (chapter 0) for more details about
the significance of this field.
Ops (Operations)
The number of test operations that have been executed for a particular test. The meaning of a ‘Operation’ varies
from test to test. For example, for the Printer test it is the number of characters printed, for the Hard disk test it is
the number of bytes that have been written or verified. See the test description (chapter 0) for more details about
the significance of this field. Note: The values are expressed in Units, Millions, Billions, Trillions and
Quadrillions.
BIT users manual
Page 43
Chapter 6 - Interpreting the results
DEFAULT ERROR VIEW
Errors
The number of errors that have been encountered while the test has been executing. This value should normally
stay at zero. A value of greater than zero indicates there has been an error in the hardware or the software
controlling the hardware. In some cases it is possible for the computer to self-detect an error. (such as the math’s
and disk tests). In other cases the user must check themselves that no error has occurred (e.g. Is there sound
coming from the speakers? Are printouts complete, clear and legible? ).
Last error description
This is the description of the last error that occurred. This will give some indication as to the cause of the error.
See Appendix C for a description of the errors that may be encountered.
ERROR BY CATEGORY VIEW
Errors are tabulated based on their severity as defined in the file BITErrorClassification.txt. The number of
errors for each test, for each level of severity: Critical, Serious, Warning and Informational will be displayed.
BIT users manual
Page 44
Chapter 6 - Interpreting the results
Test Window display
Each test type has its own display window. The contents of the windows are described in the various tests. An
example of some of these test windows is included below.
BIT users manual
Page 45
Chapter 6 - Interpreting the results
Detailed Error log history
The detailed error log history can be opened from the ‘View / Error log file’ menu. It contains a record of recent
errors and status messages that have occurred during a test run. It is possible to save a copy of this log of recent
history using the ‘File / Save test log’ menu item.
The log can contain up to 1000 messages. After 1000, the oldest messages are overwritten by the newer ones.
The log window is updated in real time and can left open while a test run is in progress.
Each log entry can be divided up into 3 fields.
Error Classification
Where logging records an error, the severity of this error (CRITICAL, SERIOUS, WARNING or
INFORMATION) is logged. Where an informational line (related to this error) follows the error line, the error
classification is not repeated (to aid determination of the number of unique errors per classification).
Time stamp
The time stamp records the time when the event occurred. The format is, YYYY-MM-DD HH:MM:SS
Test type
The test type refers t the name of the test which originated the event. Examples are, “RAM”, Floppy Disk”. The
word “Status”, indicated that the event is not an error but just a status message.
Details
This is a text field that contains additional information about the error (or status message). See Appendix C for a
description of the errors that may be encountered. In some cases a log entry may be spread across two lines. The
1st line will contain some general information about the error and the 2nd line will contain more detail.
Some error messages may contain internal program variables (and thus be fair cryptic). These are not always
intended to be interpreted by the user can but help us track down faults in our software.
Configuring Error Classifications
Each of the standard errors detected by BurnInTest can configured to include a customer specific string and error
classification. The file BITErrorClassification.txt (in the software directory) contains the error text and
classification used by BurnInTest.
The file is formatted as a Comma Separated File with four fields:
1.
2.
3.
Error number – this must not be changed by the customer;
Error description text – this may be changed. This must be a sting enclosed by quotes;
Error classification – this may be changed. This must be one of the following text values (not in
quotes):
CRITICAL, SERIOUS, WARNING, or INFORMATION
A number of errors may be configured to be ignored (not reported as errors), as they may be
considered ‘below the radar’. These are configured by setting the Error classification to NONE.
These are:
• 138,"Warning: Could not lock parallel port",WARNING,PP
• 146,"Frame could not be displayed and was skipped",INFORMATION,
• 182,"Video Playback is dropping samples",WARNING,
Other errors may not be configured to be ignored in this way.
4.
Customer comment – this is optional and may be changed – it is ignored by BurnInTest.
Comment lines may be inserted by using a “#” at the start of the line. These lines will be ignored by BurnInTest.
An example is shown below:
# PassMark BurnInTest,,,
# Error message text and classification,,,
0,"No errors",NONE,
1,"Incorrect mathematical addition",SERIOUS,
BIT users manual
Page 46
Chapter 6 - Interpreting the results
2,"Incorrect mathematical subtraction",SERIOUS,
3,"Incorrect mathematical division",SERIOUS,
4,"Incorrect mathematical multiply",SERIOUS,
5,"Incorrect MMX addition",SERIOUS,
6,"Incorrect MMX subtraction",SERIOUS,
7,"Incorrect MMX multiplication",SERIOUS,
8,"Failed Windows call - Line Drawing",SERIOUS,
BIT users manual
Page 47
Chapter 7 - Menu options
Menu options
File
Save report
Save the recent results as a text report, HTML report or a customer targeted Test Certificate. See Preparing a
Customer Test Certificate for more information about configuring a test certificate for your company.
Save as Image
Allows the main window (or its contents) to be saved in one of 3
different graphics file formats. Bitmap (BMP), GIF and Enhanced
Meta file format (EMF).
This dialog allows the main window (or it’s contents) to be saved as
an image file, which can then be imported into other application or
placed on a Web page.
Image format:
One of 3 different graphics file formats can be selected.
•
BMP. Stores the image in Bitmap format. Many applications
support Bitmaps but as no compression is used the file size is very large.
•
GIF. Stores the image in GIF87a format. Fewer applications support GIFs and only 256 colors can be used.
Powerful compression gives small file sizes and it is one of the standard image formats used on the Internet.
•
EMF (Enhanced Meta file format). Saves a scalable vector image that can be resized within other
applications. File size is very small but only limited support is available in other applications.
Option – Visible section only (or Save entire window as it appears on the screen)
If Visible section only is not selected (the default and typical selection), the full results section of the window
will be saved, including any part of the results that is currently not displayed in the window (i.e. Result
information that you need to scroll down to).
If Visible section only is selected, the full visible Window as it appears on the screen (including menu bar, button
bar, border, etc) will be saved. If result data is now visible (i.e. Result information that you need to scroll down
to) then this will not be saved.
Print
Allows the contents of the main window to be printed. The standard Print dialog window will be displayed
allowing the user to select a printer, etc..
Exit
Exits from the application.
BIT users manual
Page 48
Chapter 7 - Menu options
Edit
Copy
Allows the contents of the main window to be copied to the clipboard in an Enhanced MetaFile. Format (EMF).
The results can then be pasted as a graphical image into other documents that support this format, e.g. Microsoft
Word
Some applications may not accept a pasted EMF format image. In this case it is better to use the Save as Image
menu option (above) to create an image file, which can then be imported into another application.
Clear
Clear the results produced so far, resetting all values to zero. This is done automatically each time a new test run
is started (unless the accumulated logging option is set in the preferences window).
Machine ID and Notes
Displays the machine identity and notes window.(see section 8)
View
Error log file
Opens the log file window. The log file window contains status & error messages.
Error by categories
Toggles the main display between the following two error table views:
1. Number of error and last error description per Test Name, and
2. Number of errors for each category per Test Name (that is, CRITICAL, SERIOUS, WARNING AND
INFORMATION).
Configuration
Test duty cycles
Displays the test setup dialog which allows automatic stopping of tests to be configured, a subset of tests to be
selected for execution and a duty cycle to be set for each test.
Test Preferences
This option allows the user to customize some aspects of the way BurnInTest works. See the description of the
preferences dialog box (Chapter 0) for more details.
Load Config
A test configuration that has been previously saved to disk can be loaded using this menu option. The loaded
configuration will set all the parameters available in the ‘test duty cycles’ and ‘test preferences’ windows (see
above) to the values contained within the file. Using this menu option in conjunction with the ‘Save Config as’
function it is possible to create a number of standard test configurations. See section 8, Maintaining multiple test
configurations, for more information.
Save Config as
This option allows the current configuration settings to be stored in a file for later use. All the values available in
the ‘test duty cycles’ and ‘test preferences’ windows (see above) will be saved into a file, whose name is selected
by the user. Using this menu option in conjunction with the ‘Load config’ function it is possible to create a
number of standard test configurations. See section 8, Maintaining multiple test configurations, for more
information. Saved configuration files can also be loaded automatically by including them on the command line.
See section 8, Command line arguments for more details.
Test
Start test run
Runs the tests selected in the ‘Test duty cycles’ window. Each test will be run in it’s own window using it’s own
execution thread. This allows multiple tests to be executed simultaneously. The results from each test are
summarized in the main window. Depending on what tests are being started, the speed of the machine and the
selected duty cycles; the tests may take several seconds to start.
BIT users manual
Page 49
Chapter 7 - Menu options
After tests have been started the menu’s and buttons are ‘Grayed’ out, and no another actions can take place until
the tests are stopped manually or automatically.
Stop test run
Stops all the tests that are currently running. It may take several seconds to stop all the tests, remove test files
from the disk and de-allocate system resources. If one of the test windows is the active window, Alt-F4 can also
be used to stop all the tests. Individual tests cannot be stopped after they have been started, as all tests will be
stopped at the same time.
Execute script
Runs the tests selected in the specified BurnInTest script file (.bits).
Help
A series of options that allow access to this Online help file and to the PassMark home page. Context related
help is available from most windows using the “F1” key.
BIT users manual
Page 50
Chapter 8 - Advanced options
Advanced options
Maintaining multiple test configurations
If you are in the business to testing many different computers then you may benefit from maintaining different
test configuration. By saving a particular configuration to a file, you can avoid resetting all the parameters each
time BurnInTest is used with a different computer set-up.
A configuration file contains all the parameters available in the test duty cycles and test preferences windows.
The saving and loading of configuration files can be done from the menu options, ‘Load Config’ and ‘Save
Config As’. Configuration files can also be used from the command line. You can drag and drop a configuration
file into BurnInTest.
Configuration files normally end in the extension ‘bitcfg. For example the current configuration is stored
automatically in the file ‘LastUsed.bitcfg’
Using this feature it is possible to make your own configuration for each computer set-up that your business
encounters / builds. e.g. The heavy load situation, the Disk test only configuration.
After the configuration files have been created they can be transferred between computers where BurnInTest is
installed
Command line arguments
The following are the BurnInTest command line parameters…
-C [configfilename]
Loads the configuration file specified by [configfilename]
-D [minutes]
Sets the test duration to the value specified by minutes. Decimal values can be used.
-K
Keep disk test files. Specifies not to delete the disk drive test files when an error (e.g. Verification error) occurs.
This is intended to assist investigating disk errors. It is recommended that this option is used in conjunction with
the Auto Stop Tests on Error feature within Preferences. Once the test files have been investigated, they should
be deleted manually.
-L [x,y,wt,ht]
Starts BurnInTest with the main window located at top-left co-ordinates x, y and with width wt and height ht. It
is important that there are no white space characters in [x,y,wt,ht] specifications.
-R
Executes the tests immediately without needing to press the go button. It also skips the pre-test warning
message.
-S [scriptfilename]
On startup, BurnInTest will automatically run the script file specified by [scriptfilename]. [scriptfilename] can be
an absolute or relative path to the script file, but if the path and/or filename contain any space characters, you
should enclose the entire string in double quotes (“”).
BIT users manual
Page 51
Chapter 8 - Advanced options
Example 1
Here is an example of a Windows batch file that could be used to start BurnInTest,
c:
cd "\program files\BurnInTest"
bit.exe –c “heavyload.bitcfg”
Example 2
Here is an example that could be placed in the Windows Startup folder to run BurnInTest automatically with the
current default settings, (after Windows had booted up).
bit.exe /r
Temperature and Battery monitoring
Battery monitoring (via BatteryMon)
BurnInTest can work with another product from PassMark Software called BatteryMon. BatteryMon
allows the charge level in laptop batteries and uninterruptible power supplies UPSes to be monitored.
Turning on this feature will cause BurnInTest to log the battery statistics collected by BatteryMon.
Both programs need to be active and running for this feature to work. Statistics collected will appear
in the HTML, Text and binary log files, where they will be broken down on a per battery basis. There is also a
summary charge level displayed in the main window.
For more information about BatteryMon see the PassMark Software web site:
http://www.passmark.com/products/batmon.htm
Temperature monitoring
PassMark software does not produce a temperature monitoring application, however we do provide support for
many third party system monitoring applications currently available. By using BurnInTest with these
applications, is it possible to Burn in your hardware while monitoring and logging the temperature of your
CPU(s) and/or your motherboard.
In the case of each temperature monitoring application…
- Check that your main board supports temperature monitoring. This can be done from the documentation that
comes with the board / computer or from the manufacturers web page.
-
Ensure that the third party monitoring application is correctly installed. See below for a list of supported
applications.
-
Check that third party monitoring application supports your motherboard. Each application comes with a list
of supported boards. See below for more details on the supported applications.
-
Check that the third party monitoring application is a compatible version. See below for a list of compatible
versions.
-
Run the third party monitoring application, then Run BurnInTest.
-
Make sure temperature monitoring is turned on in BurnInTest (from the Test preferences Window) and if
required, the temperature source value is selected. For Hmonitor, up to two sources can be selected for
display and logging in BurnInTest.
-
Start testing.
BurnInTest reads the CPU temperature values from the third party monitoring application, displaying them on
the screen and writing them to disk if logging is turned on. Please bear in mind that the accuracy of the
temperature values in BurnInTest depend entirely on the third party application in question.
About Hmonitor
Versions tested with BurnInTest: 3.1.2.5 and 4.0.1.2
Hmonitor Pro monitors and displays hardware parameters collected by the several sensor chips, installed on
certain motherboards. The program can monitor voltages, CPU temperature, Motherboard temperature and
cooler fans RPMs. It is fully customizable with alarms that can be set to go off when the temperature goes too
high. In addition to this there is a built in "thermo control" function, that kicks in when the CPU is idle.
Hmonitor can be used under Windows 98 or 2000/XP operating systems on Intel-based personal computers.
Hmonitor is currently available at http://www.hmonitor.com/. PassMark also maintains a Web page that can tell
you where to get Hmonitor.
BIT users manual
Page 52
Chapter 8 - Advanced options
http://www.passmark.com/products/temperature.htm
About Intel Active Monitor
Versions tested with BurnInTest: 1.19
Intel Active Monitor monitors CPU temperatures, motherboard temperatures, voltages and fan speeds. It
contains some basic system information and supports programmable alerts. It is compatible with most new Intel
motherboards.
Intel Active Monitor is available from http://www.intel.com/design/motherbd/active.htm
About MBM (Motherboard Monitor)
Versions tested with BurnInTest: 5.3.3.0
MBM monitors CPU temperatures, motherboard temperatures, voltages and fan speeds. It contains some basic
system information and supports programmable alerts (also email alerts or application launch on alert), and
logging.
MBM is available from http://mbm.livewiredev.com/
About SpeedFan
Versions tested with BurnInTest: 4.08
SpeedFan monitors CPU temperatures, motherboard temperatures, voltages and fan speeds. It allows the user to
alter fan speeds to aid the over clocking enthusiast. It reports system information - in particular SM Bus info and
SMART attributes.
SpeedFan is available from http://www.almico.com/speedfan.php
BIT users manual
Page 53
Chapter 8 - Advanced options
Report information (Machine Identity and additional notes)
General report information
The fields in this section allow the entry of information related to the identity of the machine being tested. This
can be an aid to users that plan to test numerous machines and need to associate the results obtained with a
particular machine at a later date. Users would typically do this in order to maintain an audit trail for quality
assurance reasons.
Three free form text fields are available. The information entered into this window will be displayed in saved
result files, in the main window and on printed results.
Machine Type
Typically the make and model of the machine would be entered in this field, e.g. Dell Dimension T800 XPS
Serial Number
Typically the serial number would be entered here, e.g. D345-789-YT99
Notes
This is a spare text field that users may make use of to save any additional information that needs to be
associated with the machine being tested or the organization doing the testing. No more than 300 characters may
be entered into this field.
Certificate report information
The fields in this section allow the entry of customer specific information for the system under test. This
information is used in the Customer Test Certificate report. See Preparing a Customer Test Certificate for more
information about configuring a test certificate for your company.
Customer Name
The name of the Customer you are testing the system for.
Technician’s Name
The name of the tester.
Save As Defaults button
This button will save the contents of these three fields to disk. Each time BurnInTest starts the fields will
automatically be reloaded. This can save retyping the same information over and over again.
Preparing a Customer Test Certificate
A Test Certificate is a report targeted at your customer. It includes your company details, Technician
information, System summary, Result summary and a Certification statement. The report would typically be
created after a test run. It is produced by selecting File -> Save report menu and the Test Certificate option.
The Test Certificate is a HTML template. This HTML template will need to be modified with your company’s
details.
The items in the report that are specific to your customer, and likely to change frequently, can be modified from
Edit -> Machine ID & Notes. This is the Customer’s Name and Technician’s name.
The items in the report that are specific to your company, and unlikely to change frequently, can be modified in
the HTML template.
Test certificate HTML template
The test certificate HTML template must be in the BurnInTest directory and be named
BITCertificateTemplate.html. A sample BITCertificateTemplate.html is provided in the installation package.
Style
The HTML file defines the Styles for the Report in the <style> block. You can change the style, such as font size
and colors by modifying the style block.
Company Logo
The first item in the <body> of the HTML file is a <table> containing your company name and logo as a gif file.
For example:
<p>Put your company logo here:</p>
<img src="./Passmark_logo3.gif">
You should modify the text in the first line to be your Company name, and change the logo gif filename to a file
containing your company logo.
BIT users manual
Page 54
Chapter 8 - Advanced options
BurnInTest generated report components
BurnInTest will automatically insert 3 sections where the following marker is placed in the HTML file,
<!--BITCERTIFICATE--><!--Do not edit this line as it will be replaced by the results report-->
The sections are:
BurnInTest Certificate: Includes Customer name, date, technician name and BurnInTest version.
System summary: Includes summary information about the CPU,OS, RAM, video card and drives.
Result summary: Includes test start and stop times, duration, and a list of tests run with a PASS or FAIL
indication.
Certification statement
The certification heading and title can be modified in the HTML file by editing the following sections:
<h2>Certification</h2>
<p>This document certifies that the Tests described above have been carried out by a suitably qualified
technician on the System described </p>
Reboot / Restart Cycling
BurnInTest can be set-up so that the PC will reboot itself in a cycle, with test runs occurring between each
reboot. BurnInTest uses another software utility called Rebooter to reboot / restart a PC. Rebooter can be used
from within BurnInTest or it can be run by itself (look for the executable called, rebooter.exe in the BurnInTest
installation directory).
To use Rebooter from within BurnInTest, go to the BurnInTest Preferences window and then select the timer tab.
Clicking on the ‘Reboot options’ button will open the Rebooter configuration window. You can get additional
help about Rebooter options by clicking on the help button in the Rebooter configuration window.
Here is a brief description of how a BurnInTest can be set-up to reboot itself in a cycle.
Step1 – Select and save Rebooter settings
Open the Rebooter configuration window and use the “Clear” button to return the reboot count to 0.
Set the ‘Maximum Reboots’ value to the number of cycles required.
Set the other parameters in Rebooter. (Reboot type, Delay, etc). Don’t set the delay value to be too short, as you
want BurnInTest to fully stop before rebooting. 20 Seconds is a good value.
Save the settings, with the ‘Save Options’ button. Then close the Window.
Step2 – Configure the settings you want in BurnInTest
From the Preferences and Duty Cycle windows enter all the settings that you want for your test runs. When you
close these windows, the settings are saved to disk and will become the new default values. Alternatively a
separate configuration file could be created (using the ‘Save As Config’ menu option) and used on the command
line.
Make sure you
- Set a test period with the Auto-Stop option in the preferences window.
- Select ‘Reboot PC’ in the Action after Auto-Stop check boxes.
Step3 – Create an auto run shortcut
Create an auto run short cut that points to the BurnInTest executable. You need to do this manually in Windows.
The command line in the shortcut should use the ‘/a’ option. This will start tests executing in BurnInTest
automatically. The command line for the shortcut should be something like
C:\Program Files\BurnInTest\bit.exe
/a
The Rebooter help file (Rebooter.hlp ) also contains more details about how to start programs automatically with
Windows.
Step4 – Start the cycle
Start the 1st test run from within BurnInTest, with the Go button. At the end of the test period you have entered
the PC will reboot according to the settings in Rebooter and after the reboot, BurnInTest will automatically restart and do another test run, then Reboot again.
More about Rebooter
Rebooter is a small utility program developed by PassMark Software to help automate the PC hardware testing
process. It has been designed to work with PassMark BurnInTest but will also work with 3rd party application.
Rebooter allows you to,
- Shutdown, Reboot or Logout of a PC.
BIT users manual
Page 55
Chapter 8 - Advanced options
-
Reboot a PC from the command line
Set a timer so that the PC will reboot after a certain amount of time
Set-up a reboot loop, to reboot a PC over an over again in a cycle.
Force a shutdown or request a shutdown.
Enable and disable the Windows auto-login feature. (2000/XP/2003 Server only)
Include reboots into your hardware stress testing plan, (when used with BurnInTest).
Scripting
BurnInTest includes a simple scripting language that allows tests to be executed in a sequence.
A PASS/FAIL indication for all of the tests included in the script can be displayed at the end of the scripted test
run. To achieve this you should select “Accumulate logs until manually cleared” from the Logging Preferences
and results should be cleared before or at the start of the scripted test run.
Script files are ASCII text files that you can create with a text editor (e.g. notepad). The file name must end with
the extension ‘.bits’. To start a script use the ‘Test / Execute script’ menu item.
Each script command must appear on its own line in the text file and the entire command must appear on a single
line. (i.e. a single command can not be split across multiple lines).
The command and its parameters must be separated by one or more spaces.
Comments can be included by starting the line with the ‘#’ character.
The following commands are available in the current version of the software
RUN COMMAND
Syntax:
RUN <Test Name>
Parameters:
<Test Name>
See below for a list of all test names.
Examples:
#Run the CD test with the current settings
RUN CD
#Run all the tests in the current configuration simultaneously
RUN CONFIG
LOAD COMMAND
Syntax:
LOAD <File name>
Parameters:
<File name>
The full path name to a configuration file. This configuration file must have been
previously created from within BurnInTest. A partial path name can be used to load
a configuration file from the current directory. A loaded configuration file will over
write all preferences currently selected.
Examples:
#Load the low load configuration file that we created earlier
LOAD “C:\ConfigurationFiles\LowLoad.bitcfg”
MESSAGE COMMAND
Syntax:
MESSAGE <Message Text>
Parameters:
< Message Text > A single line of text that will be displayed in a window with an OK button. The user
must click on the OK button to continue with the script.
Examples:
MESSAGE “Insert the test disc into the DVD drive then click on OK to proceed with the test”
SLEEP COMMAND
Syntax:
SLEEP <Delay period>
BIT users manual
Page 56
Chapter 8 - Advanced options
Parameters:
< Delay period >
An integer that represents the number of milliseconds to pause before continuing
with the next command in the script..
Examples:
#Pause 2 seconds
SLEEP 2000
SETDURATION COMMAND
Syntax:
SETDURATION <Duration>
Parameters:
<Duration>
Sets the test duration in minutes. Using this command is the same as changing the
auto-stop period from the preferences window.
Examples:
#Set the test duration to 90 seconds
SETDURATION 1.5
SETCYCLES COMMAND
Syntax:
SETCYCLES <Number of test cycles>
Parameters:
< Number of test cycles > Sets the number of test cycles that will lead to an automatic stopping of the
test runs after all selected tests have reached or exceeded this number of test cycles.
Using this command is the same as changing the auto-stop number of cycles from
the preferences window.
Example 1:
#Set the number of test cycles to 1
SETCYCLES 1
Example : Run each test one cycle in series.
SETCYCLES 1
SETDURATION 0
LOG "Run Maths Test”
RUN MATHS
LOG "Run MMX Test”
RUN MMX
…
NOTE: Automatic stopping after a set number of test cycles is only supported in the licensed version of
BurnInTest.
SETDUTYCYCLE COMMAND
Syntax:
SETDUTYCYCLE <Test Name> <Duty setting>
Parameters:
<Test Name>
See below for a list of all test names.
<Duty setting>
Sets the duty cycle for the specified test to the value specified. Values must be
between 1 and 100. Using this command is the same as changing the duty cycle
value from the Test duty cycle window.
Examples:
#Set the disk test to maximum load
SETDUTYCYCLE DISK 100
#Set the CPU test to medium load
SETDUTYCYCLE DISK 65
SETSERIAL COMMAND
Syntax:
SETSERIAL < Serial Number >
BIT users manual
Page 57
Chapter 8 - Advanced options
Parameters:
<Serial Number>
The serial number string. Must be enclosed in double quotes (“”).
Examples:
#Set the serial number
SETSERIAL "1234-shdfgdhs-GHGHG"
SETMACHINETYPE COMMAND
Syntax:
SETMACHINETYPE < Machine Name>
Parameters:
< Machine Name > The Machine Name. Must be enclosed in double quotes (“”).
Examples:
#Set Machine type
SETMACHINETYPE "Dell XPS800"
SETNOTES COMMAND
Syntax:
SETNOTES <Notes >
Parameters:
< Notes >
The Notes. Must be enclosed in double quotes (“”).
Examples:
#Set Notes
SETNOTES "Test notes defined by the external application."
SETLOG COMMAND
Syntax:
SETLOG <Filename>
Parameters:
<Filename>
The name of the log file. The file name must be enclosed in double quotes (“”). The
file must be an executable.
Examples:
#Set log file with full path
SETLOG "C:\Program Files\Plugin\plugin_log"
#Set log file with path relative to BurnInTest drive
SETLOG "\Program Files\Plugin\plugin_log"
#Set log file with path relative to BurnInTest application path
SETLOG "plugin_log"
SETPLUGIN COMMAND
Sets up the first external test plug-in application executable file name.
Syntax:
SETPLUGIN <Filename>
Parameters:
<Filename>
The name of the plugin file. The file name must be enclosed in double quotes (“”).
The file must be an executable, and the “.exe” must be included.
Examples:
#Set plug-in test application file with full path
SETPLUGIN "C:\Program Files\Plugin\plugin.exe”
#Set plug-in test application file with path relative to BurnInTest drive
SETPLUGIN "\Program Files\Plugin\plugin.exe”
#Set plug-in test application file with path relative to BurnInTest application path
SETPLUGIN "plugin.exe”
<Test Name> Parameter
The test name parameter can takes the following values. The first value “CONFIG” is special because it does not
refer to the name of an individual test. When used with the RUN command it causes all tests in the current
configuration file to be started simultaneously.
CONFIG
MATHS
BIT users manual
Page 58
Chapter 8 - Advanced options
MMX
CD
DISK
MEMORY
NETWORK
PARALLEL
PRINTER
SERIAL
SOUND
TAPE
USB
2D
3D
VIDEO
PLUGIN
BurnInTest 64bit version specific:
MMX and 3Dnow! have been superseded and are not supported in BurnInTest. The MMX and 3Dnow!
Tests have been removed from the 64bit version of BurnInTest.
End 64bit specific
Example script
#Load my preferred test configuration
LOAD "MyConfiguration1.bitcfg"
#Override the test duration for all tests
SETDURATION 60
MESSAGE "Click on OK to start test run"
RUN MATHS
MESSAGE "Insert test discs into both the CD and DVD drive"
RUN CD
#Load my preferred test configuration for disk testing
LOAD "MyDiskConfig.bitcfg"
RUN CONFIG
EXECUTE COMMAND
Overview:
Executes an external file and continues processing the script.
Syntax:
EXECUTE <Filename> <Parameters>
Parameters:
<Filename>
The name of the file to execute. The file name must be enclosed in double quotes
(“”). The file must be an executable.
<Parameters>
Any command line parameters that you wish to pass to your executable. If any of
these parameters are filenames, you should enclose them in double quotes (“”)
Examples:
#Start up Notepad with some results before starting tests.
EXECUTE "c:\winnt\system32\notepad.exe" "c:\MyResults\Results.txt"
RUN CONFIG
EXECUTEWAIT COMMAND
Overview:
Runs an executable file and waits for that process to finish before continuing to process the script.
Passing configuration information from an external program to BurnInTest.
The executable file may produce a file of scripting commands that are to be run by BurnInTest once the
executable file has closed. As an example, this may be used to set the Machine type, serial number, test notes etc
at the start of a script file..
The only script commands that will be processed by BurnInTest are the SET… commands. All other commands
will be ignored. If this file, called a Sub-Script file, is created is must:
BIT users manual
Page 59
Chapter 8 - Advanced options
-
be placed in the same directory as the EXECUTEWAIT executable file;
-
must be called “bit-script-input.txt”; and
-
must conform to the scripting file format.
After processing the script commands in this file, BurnInTest will delete the file.
Syntax:
EXECUTE <Filename> <Parameters>
Parameters:
Filename
The name of the file to execute. The file name must be enclosed in double quotes
(“”). The file must be an executable.
Parameters
Any command line parameters which you wish to pass to your executable. If any of
these parameters are filenames, you should enclose them in double quotes (“”)
Examples:
#Start PassMark’s Sleeper application with user defined parameters.
EXECUTEWAIT "sleeper.exe" -S1000 -R 30 -N 1 –E
#Start up Notepad with some results before starting tests.
EXECUTEWAIT "c:\winnt\system32\notepad.exe" "c:\MyResults\Results.txt"
RUN CONFIG
LOG COMMAND
Overview:
Writes text to the detailed error and status log history.
Syntax:
LOG <Text>
Parameters:
Text
The text to be added to the detailed error and status log history.
Examples:
#Start an S1 sleep and log the start and stop times
LOG "Sleep S1 Duration 60 seconds starting"
EXECUTEWAIT SLEEPER -S1 -D 60
LOG "Sleep S1 Duration 60seconds complete"
REBOOT and REBOOTEND COMMAND
Note: These commands should only be used where multiple reboots within one script are required. The
commands are designed in such a way that they ONLY make sense in the following context…
BurnInTest must be launched automatically at start up using a shortcut to bit.exe in the StartUp directory. The
shortcut must use the /s command line parameter to automatically run the script, which contains the REBOOT
command. So if, for example the script file containing the REBOOT command was called ‘Reboot.bits’, then the
command line ‘Target’ of the shortcut would look something like
“C:\Program Files\BurnInTest\bit.exe” /s Reboot.bits
These commands require that the Rebooter application is present in the BurnInTest application directory. Any
reboots occurring as a result of these commands will use the current Rebooter settings.
Overview:
REBOOT reboots the computer. After the computer boots up, and BurnInTest restarts, the script will continue to
execute at the line following the REBOOT command.
REBOOTEND reboots the computer. After the computer boots up, BurnInTest will restart, but the script will no
longer continue to execute.
Example:
MESSAGE “Run some 3D tests”
RUN 3D
MESSAGE “Reboot for the first time”
REBOOT
MESSAGE “Run some 2D tests”
BIT users manual
Page 60
Chapter 8 - Advanced options
RUN 2D
MESSAGE “Reboot for the second time”
REBOOT
MESSAGE “And now one final reboot”
REBOOTEND
MESSAGE “This message will never be displayed”
Note: It is recommended to use “Accumulate logs” when using REBOOT and REBOOTEND.
BIT users manual
Page 61
Chapter 9 – Other information
Other information
Contacting PassMark Software
On the Web
You can contact PassMark on the web at:
http://www.passmark.com
E-Mail
For technical support questions, suggestions:
help@passmark.com
For sales & commercial issues:
sales@passmark.com
To order online visit this page,
http://www.passmark.com/sales
For any other issues:
info@passmark.com
FAQ
You may also want to check out the list of Frequently Asked Questions.
http://www.passmrk.com/support/
System Requirements
Operating System:
BurnInTest 32bit version specific:
32bit versions (x86) of Windows 98, 2000, ME, XP, 2003 server (*)
BurnInTest 64bit version specific:
64-bit versions (x64) of Operating System: Windows XP, 2003 server
RAM: 32 Meg
Disk space: 5 Meg of free hard disk space (plus an additional 10Meg to run the Disk test)
DirectX software for 3D graphics and video tests (plus working DirectX drivers for your video card).
DirectX 9.0c (or above) software
CPU:
SSE compatible CPU for SSE tests
Other:
A printer to run the printer test, setup as the default printer in Windows.
A CD ROM + 1 Music CD or Data CD to run the CD test.
A CD-RW to run the CD burn test.
A network connection and the TCP/IP networking software installed for the Network Tests
A serial port loop back plug for the serial port test. (Pro version only)
A parallel port loop back plug for the parallel port test. (Pro version only)
A USB port loop back plug for the USB port test. (Pro version only)
BIT users manual
Page 62
Chapter 9 – Other information
A USB 2.0 port loop back plug for the USB 2.0 port test. (Pro version only)
(*) – Windows98 and Windows Me do not support the Tape drive, Video playback , CD-RW burn, USB2 or the
Disk test mode of Butterfly seeking (all other test modes supported) tests. Windows 2000 does not support the
CD-RW burn test. The advanced RAM test is only available under Windows 2000, XP professional and 2003
Server (the other RAM tests are supported under the other OS’s).Users must have administrator privileges in
2000, XP and 2003 Server.
Precautions for thorough and careful testing
For a hardware test to be useful several precautions need to be taken. Failure to take into account these factors
may result in tests being misleading or other unwanted results.
-
Stop all other applications before running BurnInTest. BurnInTest can be run in the background but it just
doesn’t seem prudent to do any important work when you’re testing a computer to see if it will fail. In any
case BurnInTest will place such a load on the system that any other applications will run at a snails pace.
Not having other applications running also frees up more RAM that can be used by the Memory test.
-
Back-up any important files before you start. BurnInTest can simulate many days of typical office PC use in
a few hours, this increases the risk of hardware failure. Note that the testing process itself doesn’t touch any
existing files on the hard disk or floppy disk. It does however overwrite the contents of tapes used in the
tape drive test (Pro version only).
-
When testing multiple disk drives at the same time you may not want to test multiple partitions that are on
the same physical drive at the same time. Doing this can result in an enormous amount of seeking between
partitions and not as much reading and writing.
-
PassMark recommends running BurnInTest just after you install a PC for the first time, as this is the ideal
time to find a problem. The PC will be still under warranty and you can’t lose any of your data (because you
haven’t loaded any). Any disruption caused by a failure will be minimal.
-
Remember that BurnInTest does not create problems in your hardware, it just helps you find them in a
controlled manner. BurnInTest doesn’t use any nasty programming tricks to try and make your hardware
fail. It uses the same functions and procedures that standard Windows applications and file servers use. If
your computer fails when running BurnInTest, it was going to fail in the near future anyway !
-
If you only want to test a particular component of the computer, turn the other tests off. There’s no point
using the CD-Drive when you only wanted to test the new hard drive.
-
Doing a successful test run doesn’t mean that the computer will never fail. Software problems, viruses, and
the fact that no computer component has an unlimited life span means that precautions need to be taken.
Having good BurnInTest results is NOT a substitute for making good file backups in the future.
-
Because BurnInTest doesn’t delete any of the existing files from a disk, this occupied portion of the disk
will not be tested. Thus the more free space that the disk has before the test, the larger the portion of the disk
that will be tested.
-
For the Printer test, inspect printed pages by hand. Check the text is complete and legible.
-
When using the CD test with a music CD verify that the music is being played clearly though the PC’s
sound system.
-
On some new computers, the warranty may be voided if you open up the case. Check your warranty before
you start poking around in the case.
-
Old computers tend to fill up with dust over the period of many years. This dust layer can cause heat build
up and even short circuits. Check for dust build-up in old computers before you start.
-
Check the computer has adequate ventilation & check all the fans are in good working order.
BIT users manual
Page 63
Chapter 9 – Other information
-
Check that the computer isn’t full of bugs. (i.e. the insect type). Depending on where you live, insects can be
a real problem. The term ‘Computer Bug’ was coined after a dead moth was found to have shorted out one
of the first computers built. In Australia, cockroaches are the most common cause of failure in microwave
ovens. Recent studies have even suggested that some insects are attracted to electro-magnetic fields. So
watch out !
-
There are many cases where a software bug may appear to be a problem with the hardware. Knowing who
or what to blame isn’t easy. Check with your hardware manufacturer(s) from time to time in case they have
released new software that fixes some problems they may have found. The hardware components that in
general have the most problems with their ‘Driver’ software are, Video cards (especially the 3D functions),
Sound cards and CDs.
-
By using the Network test you can test both your computer and the network it is connected to. If an error
does occur it may be difficult to determine the location of the error. If you are using an Internet address then
it is very likely that any transient errors are the result of problems on the Internet. The best compromise is
probably to set the test address to the address of a machine on your local area network, (if you have one).
-
(Pro version only) When selecting a Serial or Parallel port to use for loop back testing, ensure that the port
selected is not already in use by the system. (e.g. by a mouse or printer).
-
Because of limitations in the memory test, faulty RAM errors may not be picked up by the test and faulty
RAM can often manifest itself in different ways. These include disk I/O errors, system crashes and lockups.
Setting the Windows ‘Lock Pages in Memory’ right
To use the advanced RAM test, the
user needs to be logged in as the
administrator and have the “Lock
pages in memory” right. This right is
turned off by default when Windows
is installed. To turn it on you need to
be logged in as the administrator and
complete the following steps.
1/ Open the Windows control
panel.
2/ Open the “Administrative Tools”
window from the control panel
and then open the “Local
Security Policy” utility. This
should open a new window
called. “Local Security Settings”.
3/ In the left hand section of the window, click on the “User Rights Assignment”, folder. This folder can be
found under the “Local Policies” folder.
4/ In the right hand section of the window find the “Lock pages in memory” policy. Then select “Action /
Security” from the menu to display the “Local security Policy Setting” window. (In Windows XP, the menu
option is now called “Properties” instead of “Security”).
5/ Use the add button to assign the right to the administrator. The “Local Policy Setting” check box should now
be checked. Click on OK to close the window. (In Windows XP, this is slightly more complex and you will
probably need the “Advanced” button to search for users to add).
6/ You now need to log off and log back on (or reboot your machine) for this new setting to take effect.
If you check back in the “Local Security Policy” utility after you login you should now see that the
Administrator has “Effective rights” to lock pages in memory.
Note1: This function is only available in Windows 2000, XP Professional, (but not XP home) and 2003 Server.
Note2: Windows XP security can be a complex thing. If you have a problem after following this procedure and
rebooting, set the “Lock pages in memory” right to “Everyone” and try again.
BIT users manual
Page 64
Chapter 9 – Other information
BIT users manual
Page 65
Appendix A
Appendix A – Example ASCII Log file
ASCII (Text) log files can be created using the logging option in the Test preferences dialog. The ASCII text log
file contains all the information that is in the main test window plus some additional information such as the
amount of RAM in the system, the current color depth (in bits), and the level 2-cache size for the CPU(s).
PassMark BurnInTest Log file - http://www.passmark.com
========================================================
BurnInTest V5.0 Pro 1000
Logging detail level: Normal
******************
SYSTEM INFORMATION
******************
Network Name: PASSMARKXPP12
Date: 12/21/05 14:02:35
Operating system: Windows XP Service Pack 2 build 2600 (32-bit)
Number of CPUs: 1
CPU manufacturer: GenuineIntel
CPU type: Genuine Intel(R) CPU 3.00GHz
CPU features: MMX SSE SSE2 SSE3 DEP PAE
CPU1 speed: 2999.6 MHz
CPU L2 Cache: 2 MB (L3 Cache: 0 KB)
RAM: 510 MB
Video card: ASUS Extreme AX300 (Resolution: 1920x1200x32)
Disk drive: Model ST3160827AS (Size: 149.0GB)
Disk drive: Model ST380013AS (Size: 74.5GB)
Optical drive: TSSTcorp DVD-ROM TS-H352A
Optical drive: PIODATA DVD-RW DVR-108DX
******************
DETAILED EVENT LOG
******************
LOG NOTE: 2005-12-21 14:03:28, Status, Main Tests started
CRITICAL
: 2005-12-21 14:03:29, USB, Could not detect USB loopback plug
INFORMATION: 2005-12-21 14:03:31, Disk, A: Not enough free disk space
CRITICAL
: 2005-12-21 14:03:35, Serial, Error while receiving from the COM port
LOG NOTE: 2005-12-21 14:03:35, Serial, \\.\COM1 Speed: 9600baud BlockSize: 100 BlockNum: 0 Transferred: 0
Handle:0000030C Total: 0, ErrorCode: 0
CRITICAL
: 2005-12-21 14:03:36, USB, Could not detect USB loopback plug
CRITICAL
: 2005-12-21 14:03:41, Serial, Error while receiving from the COM port
LOG NOTE: 2005-12-21 14:03:41, Serial, \\.\COM1 Speed: 9600baud BlockSize: 100 BlockNum: 0 Transferred: 0
Handle:0000030C Total: 0, ErrorCode: 0
CRITICAL
: 2005-12-21 14:03:42, USB, Could not detect USB loopback plug
LOG NOTE: 2005-12-21 14:03:44, Status, Test run stopped
WARNING
: 2005-12-21 14:03:44, Parallel, Warning: Could not lock parallel port
**************
RESULT SUMMARY
**************
Test Start time: Wed Dec 21 14:03:27 2005
Test Stop time: Wed Dec 21 14:03:44 2005
Test Duration: 000h 00m 17s
Test Name
CPU - Maths
CPU - MMX / SSE
Memory (RAM)
2D Graphics
Disk (A: )
Disk (C: )
Network 1
Sound
Parallel Port
USB Plug 1
Serial Port 1
Advanced Net
TEST RUN FAILED
BIT users manual
Cycles
3
2
0
0
0
0
0
0
0
0
2
0
Operations
419 Million
534 Million
72.321 Million
90
0
8.962 Million
1840
286650
0
0
900
0
Result
PASS
PASS
PASS
PASS
FAIL
PASS
PASS
PASS
FAIL
FAIL
FAIL
PASS
Errors
0
0
0
0
1
0
0
0
1
3
2
0
Last Error
No errors
No errors
No errors
No errors
Not enough free disk space
No errors
No errors
No errors
Warning: Could not lock parallel port
Could not detect USB loopback plug
Error while receiving from the COM port
No errors
Page 66
Appendix A
*********************
SERIOUS ERROR SUMMARY
*********************
CRITICAL
: 2005-12-21 14:03:35, Serial, CRITICAL: Error while receiving from the COM port (x 2)
CRITICAL
: 2005-12-21 14:03:29, USB, CRITICAL: Could not detect USB loopback plug (x 3)
-----------------------------------------------------------------------------------------------------
BIT users manual
Page 67
Appendix B
Appendix B – Fault finding
What follows are some hints on how go about finding the cause of particular system instability. (i.e. The system
locks up, you get the windows blue screen, etc..). We don’t want to try and explain the steps involved in each of
these processes, they are just points that may warrant future investigation.
-
-
Check you don’t have any viruses.
Check the drive for errors using the built in Window error checking tool.
Check that space is available on the disk for the windows swap file
Have a look through the issues in the section, 9 ‘Precautions for thorough and careful testing’.
Don’t run all the BurnInTest tests at once. Run just the 2D graphics, then run just the 3D graphics, then just
the disk, etc.. This will allow the problem to be isolated to one area.
Boot Windows up in Safe Mode and see if the system is more stable. The Windows98, ‘msconfig’ tool may
also help here.
If you suspect hardware, and you know what you’re doing, pull out all the “optional hardware”, e.g. LAN
cards, I/O cards and see if the system is more stable.
Once again, if you know what you’re doing, start swapping out components of the system to see if the fault
can be localized. Obviously you’ll need some spare hardware to do this.
If you suspect DirectX 3D problems consider using the dxdiag.exe tool that comes with DirectX.
Faulty RAM may not always be detected by the memory test. It may manifest itself as a disk fault of system
crash.
If you’re really stuck you may want to try a reinstallation of Windows on a reformatted disk. Think carefully
about this option before you attempt it, there are lots of good reasons why you don’t want to reformat your
hard disk.
Make sure you’ve got the most up to date software drivers for your hardware. Drivers are a never-ending
source of problems.
Check that you haven’t ended up with an over clocked CPU and don’t know about it.
Check that you haven’t purchased the cheapest and nastiest hardware in the hope of saving a couple of
dollars (or pounds, francs, etc). Often it may not be the cheap hardware that causes problems but the quality
and support of the software drivers that comes with the hardware that are a problem. Don’t shop on price
alone, check out the support and product reviews.
BIT users manual
Page 68
Appendix C
Appendix C – Common Error messages
When BurnInTest encounters an error during a test run a short description of the error is displayed in the main
window. What follows is an explanation of the common errors that may be encountered.
Incorrect mathematical addition / subtraction / division / multiply
The execution of a mathematical operation came up with the wrong result (e.g. 1 + 1 = 3). This is a very
serious error for a computer as it means the computer is incapable of execution the same sequence of
instructions to get the same result. Possible reasons are faulty RAM, System bus, CPU or overheating.
There is a strong chance that your computer will crash or lock up just after an error like this because if
the computer can’t add two numbers correctly, there is a good chance that it can’t continue to run a
program either.
Incorrect MMX addition / subtraction / multiplication
Incorrect 128bit (SIMD-SSE) floating point addition
Incorrect 128bit (SIMD-SSE) floating point multiply
Incorrect 64bit 3DNow! floating point result
Similar to the above, mathematical faults except that the error occurred while using the extended CPU
instructions. (Rather than the standard 32bit Maths)
Failed Windows call - Line Drawing / Bitmaps / Shapes
The Windows operating system has failed to correctly complete the graphical operation that was
requested by BurnInTest. This could indicate that Windows is running low on internal resources or that
there is a problem with the software drivers for the video card.
No free memory for buffer
Windows does not have enough free memory to allow the allocation of a buffer.
Not enough free disk space
There is not enough free space on the disk to create a test file for the Disk tests.
Test file could not be created
BurnInTest was unable to create a file on the disk in order for it to be able to run the disk tests. Check
that you have access to the disk with Windows explorer, Check file write and create permissions in the
root directory of the disk under test.
Error while writing to the disk
The test file could not be completely written to the disk. This could indicate a hardware error or a
sudden lack of free disk space
Error writing to disk - Disk is full
The test file could not be completely written to disk because the selected hard drive is full.
Test file could not be re-opened
The test file was created successfully, but could not be reopened for verification. This could indicate a
hardware error or a resource problem within Windows.
Error while reading from the disk
The test file created on the disk can no longer be read. This could indicate a hardware error or a
problem within windows.
Data Verify failure
This is a serious error that indicates that the data read from the disk is NOT the same as the data that
was written to the disk. This could indicate a hardware error.
BIT users manual
Page 69
Appendix C
Disk is full or FAT root file system limit reached
A test file could not be created on the disk being tested because the disk is full or the capacity of the file
system has been exhausted. Normally this error only happens when the disk is formatted using the old
FAT files system. Floppy disks still use the old FAT file system and there is a limited number of files
that can be created in the root directory of a floppy disk. The number of files depends on the length of
each file name. (and does NOT depend on available disk space). Thus this not really an error, but a
limitation of the FAT file system. (Which has now been replaced by FAT32 and NTFS).
Warning - Disk low speed threshold crossed
The disk speed has fallen below the low speed threshold as specified in the Disk test preferences
window.
Could not set CD Time format
Could not get number of audio tracks
Could not play the audio track
CD-ROM failed to return status information
Can not detect audio information on CD
No audio media in CD-ROM
These errors relate to one of the following problems.
There is not a CD in the first CD-ROM connected to the computer.
The CD in the CD-ROM is not a Music CD.
The CD is already in use by another application.
The CD has an error on it and is unreadable.
The CD is dirty, dusty or scratched and can not be read
The CD Software drivers are not correctly installed.
The CD-ROM has a hardware error.
Checksum failure for CD/DVD file
This indicates that the file read from the data CD/DVD being tested failed the checksum verification.
This means that the CD/DVD Drive is unable to accurately read data from the CD/DVD being tested.
Could not open file on CD/DVD for reading
The CD/DVD drive selected could not be opened for reading as a data CD/DVD.
Error while reading file from CD/DVD
A full block of data could not be read from the data CD/DVD.
Error while searching for files on data CD/DVD
An error occurred while searching for files on a data CD/DVD. This can be the result of a corrupted
(Error code 1117) or blank CD/DVD (Error code 21).
Data read from CD/DVD was incorrect
A block read from a specialized PassMark Test CD or DVD was incorrect. There was at least one byte
in the block that was not the value expected. In the detailed error log there is additional information that
give the number of bytes in error and displays the expected value and the value actual read from the
disc. It may be that the drive is faulty but you should check that the disc is not scratched, dusty or
damaged before assuming a hardware fault.
Could not determine type of Test CD/DVD
BurnInTest was expecting to find a specialized test CD/DVD in the drive selected. A specialized test
CD/DVD has a specific set of files, which all have a specific file size. These disc are normally
purchased from PassMark Software or made using the PassMark CD-Maker utility. BurnInTest was not
able to find the correct files or the files appear to have the wrong file size. If you don’t have a
specialized test CD/DVD available, select one of the other two test options in the preferences window.
Error loading DirectX .dll
DirectX (version specified in the System Requirements section) is not installed correctly.
Error initializing DirectX
Error initializing DirectX device
These errors relate to one of the following problems.
DirectX (version specified in the System Requirements section) or above is not installed correctly.
You need to install the most recent driver for your graphics card.
You’ve selected an incompatible screen resolution and/or color depth for you desktop.
BIT users manual
Page 70
Appendix C
Error creating DirectX textures
Out of video memory
This error means that you have exhausted all of your graphic adapters onboard memory.
Warning: Switching to reference rasterizer
BurnInTest was unable to initialize a hardware accelerated device and is using software emulation to
render the 3D graphics test. This will result in a much lower frame rate.
Video memory corruption
One of the pixel values read from video memory is not the same as that which was written to video
memory. This may indicate that your video memory is damaged.
Error initializing DirectDraw
DirectDraw could not be initialized. Check that DirectX (version specified in the System Requirements
section) is installed. Check that the most recent version of your video adapter driver is present.
Frame could not be displayed and was skipped
DirectDraw reported an error when displaying one of the test patterns to the test window. If you receive
many of these errors, reboot and check that the most recent version of your video adapter is present.
There are several instances when this message is expected. One example is when the video display
mode is changed or power control settings are altered.
Not enough video memory available for test
The minimum video memory required for the test (usually 100KB) could not be allocated. Check your
display adapter is not carrying out any other resource hungry processes, such as games. Attempt a
reboot and ensure that the most recent version of your video adapter is present.
Minimum color depth of 16 bits required.
Your screen mode is set to a color depth of less than 16 bits. Set the screen mode to 16 bit or 32 bit
color and run the test again.
Could not find the default printer
Could not open the default printer
Could not send page to printer
Could not open printer spooler
Error writing data to spooler
These errors relate to one of the following problems.
There is no printer defined in the Windows setup.
There is no default printer selected.
The default printer is not connected to the printer.
There is no paper / ink / toner in the printer.
There was a communications problem to the printer. e.g. The parallel port may not be functioning
correctly.
Could not open printer command file
The printer command file specified by the user was not found on the hard disk. Check the path name
and file name are correct in the preferences window.
Error verifying data in RAM
The data written to memory is not the same as the information read from memory. This is a very serious
error, much like the “Incorrect mathematical…”, error above. It’s highly likely that your computer is
about to crash or lock up.
Error allocating RAM from Windows
The Windows operating system was not able to allocate the amount of RAM requested by BurnInTest
for the memory test. As the RAM must be allocated in a continuous block, this error can sometimes be
seen as a result of free memory fragmentation.
Error connecting to network
Could not create a network socket
Could not set socket timeouts
During the establishment of a network connection an error was encountered. The Network address
selected has no effect on if these errors occur. They can be the result of the following problems.
– There is no network connection configured for the computer
– The computer is physically disconnected from the network
– The Internet TCP/IP protocol is not installed on this computer
BIT users manual
Page 71
Appendix C
Networking functions aren't available (ws2_32.dll)
During the establishment of a network connection an error was encountered. This error indicates that
the Windows Network Socket functions supplied by the library ws2_32.dll are not available on this
system. This main mean that
- The Internet TCP/IP protocol is not correctly installed on this computer
Could not allocate memory for packets
Windows is low on resources and cannot allocate any more memory.
Could not resolve host name, check settings
The Network address selected does not seem to be correct. Try another address, or using a TCP/IP
address directly. The address 127.0.0.1 is good for testing as it is an internal loop back.
No permission to open RAW network socket
If you are a non-administrator logged onto a Windows 2000, XP or 2003 Server system you will get this
error with the Network test. In Windows 2000, there is no way to disable this security check. Access to
Raw Sockets is granted on a per-transport basis. For the address family AF_INET, only administrators
have the access necessary to create Raw Sockets. The test should be run while logged in as the
‘administrator’.
Timeout sending packet
Timeout waiting for packet
Error sending packet
Error receiving packet
Bad packet
These errors relate to one of the following problems.
The Host using the network address selected doesn’t reply to ‘ping’ messages. Try a different host.
There is a configuration problem in your network connection.
Your network card may be faulty.
The network itself is not reliably. The information send was not the same as the information echoed by
the remote host.
The network is congested or faulty and the packets are not being echoed within the timeout period
specified in the preferences window.
Network test alarm. Error ratio exceeded
The Bad Packet ratio specified in the Network test preferences has been exceeded.
Got someone else’s packet
This is not really an error. It’s more of an information message. Don’t worry about this message.
Bad packet. Checksum incorrect
The checksum in the echoed, incoming data packet is not correct. This indicates data corruption or a
fault on the remote machine. Note that this checksum is calculated by the remote machine.
Bad packet. Corrupt data
The contents of the incoming data packet are not correct. Normally this error would not been seen as the
checksum should detect the incorrect data before this error occurs.
COM port is already in use by another program
The serial port selected for the test is already is use by Windows. This may be for the mouse, a modem
or another serial device.
The requested COM port could not be found
The serial port selected for the test does not exist in this computer. This could happen if COM4 is
selected but the computer has only 2 serial ports, COM1 and COM2.
Error while opening COM port
Windows has reported an error while trying to open the serial port selected for the test. This could be a
configuration problem in Windows. This error should not normally been seen.
Error getting current COM port configuration
Error while setting new COM port configuration
Windows has reported an error while trying to configure the selected serial port. The most common
cause for this error would be the selection of a speed that is not supported by the serial port chips
installed in the computer (the UART). Most chips only support speeds up to 115Kbit/s.
Corruption. Data received didn't match data sent
BIT users manual
Page 72
Appendix C
BurnInTest has detected that the data received from the serial port doesn’t match the data sent. This
could indicate that there is a hardware problem. This type of data corruption would however be a fairly
rare type of event. The more common result of a hardware failure would be the total inability to send or
receive data. (see below).
Error while setting current COM port timeouts
Windows has reported an error while trying to set the timeout periods for the data transmission and
reception. This could be a configuration problem in Windows. This error should not normally been
seen.
Error while sending data to the COM port
Error while receiving from the COM port
Windows has reported an error while trying to send / receive data through the serial port. If a device
(such as a loop back plug) is not connected to the serial port then no data can be sent. If the loop back
plug is connected and is not faulty, then this error may indicate a hardware fault.
COM port Clear To Send (CTS) line stuck high
COM port Clear To Send (CTS) line stuck low
COM port Data Set Ready (DSR) line stuck high
COM port Data Set Ready (DSR) line stuck low
The signal pin test phase of the serial test has failed. This might be because of an incorrectly wired up
loop back plug, a cabling problem between the serial socket and the motherboard, a non standard COM
port or a problem with the Windows configuration for the COM port. A failure of the CTS pin may be
caused by the associated RTS pin, to which it is looped. A failure of the DSR pin may be caused by the
associated DTR pin, to which it is looped.
Parallel device driver not initialized
To access the Parallel port under Windows 2000, XP and 2003 Server BurnInTest needs to load a
special device driver that allows direct access to the parallel port. If this driver, “DIRECTIO.sys”, can
not be found or loaded then this is the error message. You need to be logged in as the administrator in
order to load this driver. This error may also occur if the file is missing.
Parallel device driver could not detect port
This error usually occurs if the parallel port test attempts to access a parallel port, which doesn’t exist
(such as perhaps LPT3 or LPT4). It can also happen if the BIOS settings for the port are not correct.
Note that the old ‘bi-directional’ BIOS mode is not supported. ECP or EPP mode is required.
Could not open parallel device driver
This error usually occurs if the parallel port test fails to access a parallel port, which does exist.
Warning: Could not lock parallel port
When the parallel port test initializes, it attempts to lock the port for exclusive access. This is essentially
to prevent Windows XP “warm polling” the port, which can produce invalid test results. This error
results in a failure to lock the port and does not represent a failure in testing the parallel port.
“Unable to lock Parallel Port” message the BITErrorClassifications.txt file can be edited to change this
error message classification (Number 138) to NONE.
Corruption. Data received didn't match data sent
The data sent to the Parallel port was not the same as the data received. This may indicate a hardware
problem or a missing or faulty loop back connector.
Error while sending data to the parallel port
Error while receiving data to the parallel port
Windows has reported an error while trying send or receive data. This could be a configuration problem
in Windows. This error should not normally been seen.
Could not detect the parallel port selected
BurnInTest was unable to find a parallel port at the location selected. Try picking another port and see
the Parallel port test description for more details about port selection.
Could not access the tape drive selected
The tape drive selected from the preferences window (e.g. TAPE1) could not be found connected to the
computer.
BIT users manual
Page 73
Appendix C
Error while formatting tape partition
The tape drive reported an error while trying to format the tape in the tape drive.
Error while loading (rewinding) tape
The tape drive reported an error as a result of a request to load and position the tape for writing.
Error while locking tape in drive
The tape drive reported that it was unable to lock the tape in the drive. Some tape drives may not
support this option and this art of the test can be turned off in the tape preferences window.
Tape device is not ready to start test
The tape drive is not in a state where it can accept a new command (but it should be).
No media in tape drive
The tape drive reported that there is no media in the drive.
No free memory for write buffer
There is not enough free RAM to allocate a write buffer.
Error while seeking to new position on tape
The tape drive reported an error while seeking to a new position on the tape.
Error while writing to tape
The tape drive failed to fully write the current data block to the tape.
Error while writing file marker between files
The tape drive failed to write a file marker. (A flag to indicate a new file)
Error while reading data block from tape
The tape drive failed to fully read the current data block from the tape.
Corruption. Data read didn't match data written
The data read from the tape didn’t match the data written to the tape.
Unable to get/set tape drive parameters
A request to get/set the tape drive capabilities resulted in an error.
Unable to get/set tape media parameters
A request to get/set the capabilities of the media current in the tape drive resulted in an error.
Tape is write protected
The tape media is write protected and the test cannot continue.
Tape drive has reported that it requires cleaning
The tape drive and/or the media in the drive are dirty. Clean the tape drive before reattempting the test.
Tape partition is too small for test files
The media is too small or the tape drive does not have the capability to fit the requested files on the
tape. Select fewer files or smaller files or use a bigger tape.
Could not open MIDI Sequencer or MIDI file
The sound test was not able to open the sound card device or the midi sound file on the disk for
playback. This can be because the test file has been moved or deleted from the disk or because the
sound card has not been installed correctly.
Failed to play back MIDI via sequencer
The MIDI sequencer was not able to play back the file. This error is normally accompanied with a 2nd
error message in the detailed log that gives additional information.
Could not play back wave file
The sound test was not able to open the sound card device or the wave sound file on the disk for
playback. This can be because the test file has been moved or deleted from the disk or because the
sound card has not been installed correctly.
Audio Input doesn't match Audio Output
The data received through the audio line in is in a wave type form, but differs from the output
waveform by a greater degree than the Max Distortion value set in the Sound test preferences. It is not
rare for occasional high distortion values to occur on certain sound cards, especially if the system load
is particularly high.
Corrupt Audio input
BIT users manual
Page 74
Appendix C
This is a more serious error than the previous one. The audio line in data is corrupt and in no way
resembles the output wave. Check that the audio loopback cable is connected to the correct ports and
the Windows mixer settings are properly configured (see section 3 Sound card test).
Could not detect USB loopback plug
The USB enumeration process failed to find a USB loopback plug device. This might be because, (1)
The USB plug is not physically connected. (2) The USB loopback device driver was not successfully
installed. (3) The USB plug was connected while a test was already in progress and thus was not
recognized. (4) The Windows O/S is not installed correctly. (5) The USB socket on the PC is faulty. (6)
The USB cable being used is faulty or is prone to errors. (7) The plug itself has become faulty. (8) The
USB cable is too long or of poor quality (shielded cables under 5m are recommended). As there are
many possible causes for this error it can take some investigation to find the origin of the fault. The first
step however should always be to remove the plug, wait a few seconds then reconnect the plug.
USB loopback plug identified itself incorrectly
The USB enumeration process succeeded, but the unique identification information stored in the USB
plugs EEPROM memory is incorrect. Under normal conditions this should never happen. The plug can
not be used unless it is programmed correctly. Contact PassMark Software for assistance.
Could not allocate RAM for USB I/O buffers
The system is low on memory
USB device open failed
The initial attempt to connect to the device failed. The device might already be open and locked by
another process. With all the tests stopped, try removing the plug, waiting a few seconds then
reconnecting the plug.
USB device reset failed
Setting USB data rate failed
Setting USB data characteristics failed
Setting USB flow control failed
Setting USB I/O timeouts failed
Purging of USB I/O data buffers failed
Set USB loopback DTR failed
Set USB loopback RTS failed
These errors are all very rare. Normally errors will occur earlier in the initialization process and not at
these steps. In addition to the possible causes for errors mentioned above, it is also possible that there is
a software configuration problem. If one of theses errors persists, contact PassMark Software for
assistance.
Data packet transmission failed
USB driver reported send error
Data could not be sent to the USB device. It may be that the device has an intermittent connection to the
host and has disconnected itself. (i.e. a faulty cable). It may also indicate that there is a problem with the
device drivers or USB hardware.
Data packet reception failed
USB driver reported receive error
Data could not received from the USB device. It may also indicate that there is a problem with the
device drivers or USB hardware.
USB data packet verify failed
A block of data was sent to, and received from, the USB plug but the data was corrupted. What was
received did not match what was sent. This could indicate a poor quality cable that is subject to external
electromagnetic interference or faulty USB hardware.
Warning: Too many USB errors attempted reset
After 10 errors BurnInTest automatically attempts clear the error condition start by provoking a reset on
the device and device drivers. If the error condition does clear after this action the problem was almost
definitely a software problem.
BIT users manual
Page 75
Appendix C
Insufficient rights to lock pages
Either you are not running Windows 2000, XP or 2003 Server or the administrator user rights to lock
pages in memory is not set. See the page in the help file called, Setting the Lock Pages in Memory right
for more information. This right is required for the advanced memory test and is turned off be default. If
you have set this right, make sure you are logged on as the administrator and have rebooted the machine
so that the change takes effect.
Could not get extended RAM info. W2000 or XP Req.
The user has selected the advanced RAM test but there was a problem detecting while trying to obtain
information about extended RAM. This maybe because the system is not running Windows 2000, XP or
2003 Server.
SMART Threshold exceeding condition detected.
One of the drives S.M.A.R.T Attributes has exceeded its threshold. This signals imminent drive failure and
indicates the tests should be stopped and the drive replaced. To learn more about S.M.A.R.T, see What is
S.M.A.R.T?
BIT users manual
Page 76
Appendix D
Appendix D – Loopback plugs
(Applicable to BurnInTest Pro version only)
You can purchase high quality molded loop back plugs from PassMark or, for the serial and parallel versions,
you can make your own. A loop back plug is a small connector that plugs into the serial, parallel or USB port on
your computer in order to loop the data output line into the data input line. In the case of the serial and parallel
plugs, it also sets a few control lines to trick the computer into thinking that an external device is connected to
the port. In the case of the USB plug, an intelligent USB controller chip is required to loop the data.
You'll find below the instructions for making your own serial and parallel loop back plugs. These plugs are
designed to be plugged into the small 9 pin serial port and 25 pin parallel port on your computer. In conjunction
with software such as BurnInTest they allow the testing of the port.
Warning: These are the pin outs of the plug that we have made and successfully tested with our own software on
our own PC's. You use this information at your own risk! It's easy to make a mistake and we take no
responsibility for the potentially serious consequences.
Serial port loopback plug
The serial port on IBM compatible PCs conform to the RS232 electrical standard and use a male DB-9
connector. (older PCs may use a DB-25 in which case a converter will be required). The DB-9 connector as the
following 9 pins.
DB9 Pin
EIA Description
Description
1
DCD
Data carrier detect
2
RXD
Receive data
3
TXD
Transmit data
4
DTR
Data terminal ready
5
GND
Signal ground
6
DSR
Data set ready
7
RTS
Request to send
8
CTS
Clear to send
9
RI
Ring Indicator
To make the loop back plug the following pins need to be connected together:
ƒ Receive and transmit (2 & 3)
ƒ Request to send and Clear to send (7 & 8)
ƒ Data carrier detect, Data set ready and Data terminal ready (1, 6 & 4)
This diagram shows the connections that need to be made. It's the rear view of the female DB-9 connector that's
required for the plug. The red lines and gray dots show the connections that need to be made on the back of the
plug.
BIT users manual
Page 77
Appendix E
Female DB-9 connector - Rear view
Parallel port loop back plug
The parallel port on IBM compatible PCs have always used a DB-25 connector. However, over time several
changes have taken place to the electrical interface with the introduction of enhanced and bi-directional parallel
ports. We believe that this plug will work with all styles of parallel ports.
DB-25 Pin
2
Direction
3
4
5
6
10
11
12
13
15
Out
Out
Out
Out
In
In
In
In
In
Out
Description
Data bit 0 (Least significant bit)
Data bit 1
Data bit 2
Data bit 3
Data bit 4
Acknowledge status
Busy status
Paper out status
Select status
Error status
To make the loop back plug the following pins need to be connected together:
ƒ
ƒ
ƒ
ƒ
ƒ
Data 0 and Error status (Pin 2 & 15)
Data 1 and Select status (3 & 13)
Data 2 and Paper out status (4 & 12)
Data 3 and Acknowledge status (5 & 10)
Data 4 and Busy status (6 & 11)
This diagram shows the connections that need to be made. It's the rear view of the male DB-25 connector that's
required for the plug. The red lines and grey dots show the connections that need to be made.
Male DB-25 connector - Rear view
BIT users manual
Page 78
Appendix E
Audio loopback cable
PassMark audio loopback cables contain two 3.5mm gold plated stereo plugs which connect the Audio out and
line in ports of your sound card during sound card loopback testing.
The following are the cable specifications.
Dimensions:
Total length: 330mm ± 0.5mm
Plug length: 14mm ± 0.5mm
Plug diameter: 3.5mm
Conductors:
American wiring gauge: (26(10/0.12)x2F, 0.43mm diameter
Insulation: PBC, 1.0mm diameter
Shielding:
Spiral shield: Bare copper wire, 30/0.10
Sheath:
Material: PVC
Color: Black
Audio loopback cables can be purchased from http://www.passmark.com
USB1 port loop back plug
Active USB loopback plugs are the quickest, most effective way to verify that a PC's USB ports are
working correctly. The Red, Green and Yellow LED's on the front of this USB adapter plug
provide a quick visual indication of I/O activity and if the USB bus is powered.
Using these plugs, error free transfer speeds of around 700 - 800 Kbits/sec per port
can be expected on a system that is functioning correctly.
Technical specifications
USB standard:
1.1 (but is forwardly compatible with USB 2.0’s FullSpeed)
Plug and play:
Compliant
Physical connector:
4 wire. (2 x Balanced Signal, 1 Voltage [VBUS], 1 Ground [GND])
USB Transfer modes: Bulk
Voltage:
4.4V – 5.25V
Current:
90mA (typical operation)
Clock speed:
6.00Mhz input to x8 multiplier
Device buffer:
384 bytes Rx, 128 byte Tx
Case:
High impact MABS plastic
Size:
65mm x 50mm x 20mm
(2.5 x 2.0 x 0.8 inches)
Weight:
30.5g (1.1oz)
EMC standards:
AS/NZS 3548:1995, EC
Storage Temperature: -20 ºC to + 70 ºC
Usage Temperature: 0 ºC to + 50 ºC
USB 2.0 loopback and benchmarking plug
USB 2 Loopback plugs can be used to help diagnose, troubleshoot and load test
your PC’s USB 1.x and 2.0 ports. Five status LEDs on the USB 2 Loopback plug
will dramatically improve the troubleshooting and diagnostic ability of computer
technicians and system integrators, who have previously been working blind.The
Red, Green and Yellow LED's on the front of this USB adapter plug provide a quick
visual indication of I/O activity and if the USB bus is powered. The Orange LED
will provide a quick indication whether the PC port is currently supporting the
higher USB 2.0 speed (up to 480Mb/s) or the lower speed found with USB 1.x (of
up to 12 Mb/s). The leftmost red LED indicates whether recoverable bus error has
occurred.
BIT users manual
Page 79
Appendix E
These plugs are unique! They are the only true USB 2.0 loop back adapters on the market and have been
especially designed for use with our BurnInTest Professional software package. They do not require an external
power supply and do not require a connection to any other port on a PC. Each plug also has its own serial
number stored in EPROM, on the plug, so it is possible to identify each plug when multiple plugs are connected.
Each plug also contains its own CPU and executes its own firmware (which can be upgraded in the future by
download)
Technical Specifications
USB standard:
Plug and play:
Physical connector:
USB Transfer modes:
Voltage:
Current:
Clock speed:
Device data buffers:
Device memory:
Case:
Size:
Weight:
Indicators:
EMC standards:
Storage Temperature:
Usage Temperature:
Max speed:
2.0 HighSpeed and FullSpeed
(and is backwardly compatible with USB 1.1 and 1.0)
Compliant
Standard 4 wire USB
(2 x Balanced Signal, 1 Voltage [VBUS], 1 Ground [GND])
Bulk
4.4V - 5.25V. (No external power supply is required)
500mA (less than 100mA during initial enumeration)
24Mhz
3 x 64-byte control. 4Kbyte data available
8KB Firmware code and data available
High impact MABS plastic
65mm x 50mm x 20mm (2.5 x 2.0 x 0.8 inches)
35g (1.3oz)
Orange LED = USB2.0 (HighSpeed ) or USB1.0 (FullSpeed )
indicator
Green LED = Transmit data
Red LED 1 = Power from USB bus
Yellow LED = Receive data
Red LED 2 = USB bus transmission error indicator
AS/NZS 3548:1995, EC
-20 ºC to + 70 ºC
0 ºC to + 45ºC
480 Mbits/sec
Compatibility
Windows 2000, XP and 2003 Server are supported.
Window 95, 98, ME, NT4, Mac & Linux are not supported.
In order to use the plugs BurnInTest Professional version 4.0 or above is required or the stand alone test software
USB 2.0 and 1.1 ports are supported. (when connected to a USB 1.1 port only the lower speed, USB 1.1, mode
will be used)
For further details regarding any of the products in this section please see the PassMark website:
www.passmark.com
BIT users manual
Page 80
Appendix E
Appendix E – What is S.M.A.R.T.?
S.M.A.R.T. (Self-Monitoring Analysis and Reporting Technology) is an interface between the BIOS and a
computers hard disk. It is a feature of the Enhanced Integrated Drive Electronics (EIDE) technology that
controls access to the hard drive. If S.M.A.R.T is enabled when a computer is set up, the BIOS can receive
analytical information from the hard drive and determine whether to send the user a warning message about
possible future failure of the hard drive.
S.M.A.R.T monitors a set of drive attributes, such as temperature, throughput performance and the number of
hours the drive has been powered on. It compares these attributes to a corresponding set of thresholds. If an
attribute drops below its threshold, the drive is no longer considered fit for use and should be replaced
immediately. Not all attributes are available on all drives, however - it is the responsibility of the drive
manufacturer to include S.M.A.R.T support.
Some of the more common S.M.A.R.T attributes are listed here.
Raw Read Error Rate
Represents the rate of uncorrected read errors. An error condition indicates that there is a problem with either
disk surface or read/write heads.
Throughput Performance
Represents the throughput performance of the drive. I.e. The speed at which the drive is reading and writing
data.
Spin Up Time
Represents the average amount of time required to spin up the drive spindle to operational speed from a stopped
state.
Start/Stop Count
Represents the number of start/stop cycles for the drive. The drive being powered on/off or suspended/woken up
are considered as start/stop cycles.
Reallocated Sector Count
Represents the amount of spare sector pool available. Spare sectors are used to replace sectors that became bad
for some reason (for instance, if a read error occurs). Therefore the more sectors reallocated, the worse the
condition of the drive.
Seek Error Rate
Represents the number of seek errors. Each time the drive attempts a seek operation, but fails to position its head
correctly, the seek error rate increases.
Seek Time Performance
Represents how efficiently the drive is performing seek operations. An error indicates problems with the drive
subsystem, for instance the servo responsible for positioning the head.
Spin Retry Count
Represents the number of times a drive fails to spin its spindle up to operation speed on the first attempt.
BIT users manual
Page 81
Appendix E
Calibration Retry Count
Calibration is the act of repositioning the drive read/write head to cylinder 0. This value represents the number of
times a calibration has failed on the first attempt.
Reallocation Event Count
Represents the number of reallocation events, which have taken place. Sometimes multiple sectors are
reallocated together – this corresponds to one reallocation event. (See also: Reallocated Sector Count and
Current Pending Sector Count).
Current Pending Sector Count
Represents the number of sectors currently pending reallocation.
Offline Scan Incorrect Sector Count
Represents the amount of errors detected during the last offline scan.
Ultra ATA CRC Error Count
Represents the number of CRC error found in the Ultra DMA high-speed transfer mode. (CRC stands for Cyclic
Redundancy Check and is data verification algorithm which uses polynomial checksums).
Write Error Count
Represents the rate of uncorrected write errors. An error indicates that there is a problem with either disk surface
or read/write heads.
BIT users manual
Page 82