PerformanceTest V4 FAQ

Updated: 5/Dec/2003

Q. My License key doesn’t seem to work

Both the User Name and Registration Key must be correctly entered before the software turns itself into the registered version. See this step by step guide for help.

Q. I've lost my license key and I need to re-install the software

Email us at , giving us all the details you can about when & how you purchased the software. We'll then search our records and resend you your key once we have found it.

Q. My system doesn't meet the minimum requirements for version 4.0 of PerformanceTest. Can I still use the software?

PerformanceTest version 4.0 no longer supports Windows 95 and Windows NT4. You also require DirectX 8.0 or above for version 4.0. If you are using Windows 95 and Windows NT4 you can download the previous version of the software, version 3.5. This old version does support 95 and NT4 but we no longer provide updates for this version. If you have purchased version 4.0 and need a license key for a previous version please email . If you are using Windows 2000 or Windows 98 but don't have Direct X 8.0. It can be downloaded from the Microsoft web site.

Q. I don't understand the results. What do all these numbers mean?

If you don't have a lot of computer knowledge interpreting the results can be confusing. There are however a few simple concepts that can help you.

• The bigger the number the faster the computer.
• The results are not a percentage figure. They are relative figures.
• A computer with a result of 40 can process roughly twice as much data as a computer with a result of 20.

Q. The amount of level 2 cache is not reported correctly for my new CPU.

On a few of the new Intel CPUs, like the Celeron 2.2Ghz CPU, the amount of level (L2) cache is reported incorrectly in PerformanceTest V4.0. In this case it should report 128KB of L2 cache but PerformanceTest is reporting 0KB. This problem occurred becuase Intel just recently created a few new codes in their CPUs to represent new cache configurations. The problem has been corrected and will be available in the next public release of PerformanceTest.

Q. PerformanceTest fails to start and displays an error message "Not responding"  OR  takes several minutes to start in Windows 2000 or XP.

See this page for details on our to help us trace this problem.

Q. PerformanceTest locks up during the 2D video tests and I have a ATI Radeon video card, with Catalyst video drivers and I am running Windows XP with a theme active.

These conditions expose a bug in the ATI video drivers. This problem was seen in driver version 7.74 (6118), but was corrected in later versions. A work around was to disable Windows themes.

Q. I have dual monitors and my 2D tests are performing slower than expected.

Using extended desktops with dual monitors affect the 2D test performances significantly. Try disabling extended desktops and running the 2D tests again.

Q. I am running Windows98 and the CPU load % seems to be too high in the advanced disk, CD and network tests.

PerforrmanceTest gets the CPU utilization percentage from the Windows operating system. On some systems Windows 98 and Windows 98 second edition have a known bug that causes incorrect information to be returned by the operating system. This problem also affects the built in Windows system monitor. See the Microsoft web site, knowledge base question Q227131, "System monitor displays incorrect CPU usage", for more details. Microsofts workaround for this problem is, "you can safely ignore it." !

Q. I have a hyperthreading capable Xeon (2Ghz or greater) or P4 (2.8Ghz or greater). Why is the MegaFLOPS result so low.

Testing has shown that having hyperthreading capability enabled actualy reduces the performance of the CPU for certain tasks. Better results can be obtained by turning the hyperthreading CPU feature off. The MegaFLOPS test is very CPU and L2 cache intensive and we suspect the poor performance is a result of bottle necks when accessing L2 cache. More information and additional results for Hyperthreading can be found in for this forum article.

Q. I am getting strange results for the floating point division test.

This test may produce inconsistent values from one run to the next depending on the current state of the floating point control register. For details refer to this forum article. We have a solution for this problem which will be available in the next release of PerformanceTest.

Q. The results I obtained in version 4 are not exactly the same as in previous releases

There were many changes made to the PerformanceTest software in Version 4. So results created with previous versions of the software are not directly comparable with version 4 results. However old baseline files will have their values scaled up and down when loaded into PerformanceTest version 4 to approximate the results that might have been obtained with version 4.

Q. The results for my hard disk aren't what I expected.

There are many factors that affect the disk test results in a Windows environment. Here are the main ones

• Cluster size. Larger clusters in general means better performance for large files. See Appendix 1 for typical default values. Reformatting your drive with larger clusters will give better results in our benchmark, but will also waste disk space. See Appendix 2 for an easy way to check your current cluster size.
• If the disk is fragmented and almost full, this can badly affect performance. Windows includes a utility for defragmenting the disk. Look up “defrag” in the Windows online help for more details.
• The position of the test file on the disk (inner cylinder or outer cylinder) can also affect the performance. The only way to avoid this problem is to only test newly formatted disks.
• The file system being used, FAT, FAT32 or NTFS.
• The operating system, Window9x, NT, 2000, XP.
• The disk controller (IDE or SCSI) and what mode it is running in, ATA-33,66,100, RAID0, 1,etc.
• The amount of memory that Windows are currently allocated to the disk cache. This can (and will) vary from one run to the next. The more you use the disk the more free memory Windows will allocate to the cache (see Appendix 3). Thus on systems with lots of free RAM, results can increase slightly on successive runs.

Q. OK, So there are a lot of factors that effect disk performance but my Seek R/W results still seem too low

This test creates a large test file on the disk under test. The size of this test file is proportional to the amount of RAM available, this stops the entire file being cached in RAM. The file is then read randomly; a seek is performed to move the file pointer to a random position in the file, a 16KB block is read or written then another seek is performed. The amount of data actually transferred is highly dependant on the disk seek time. 

Here are some Seek R/W results for the same machine with various amounts of RAM:
Pentium II 266 FAT32, 16KB Clusters, W98
32MB = 2.6
64MB = 2.9
256MB = 3.9.

Q. When running Windows 2000 Server the disk results are very different from Windows 2000 Professional, but the hardware is the same. Why?

The Server edition of Windows 2000 has a different caching policy compared to the standard edition of Windows 2000. Windows 2000 Server allocates a lot more free RAM to the disk cache, thus making the results look artificially better. This change in behaviour can also be forced by altering the value of the LargeSystemCache registry key.

Q. How can I setup PerformanceTest to run from a floppy disk ?

1. Install PerformanceTest on any PC using the normal setup program.
2. Run PerformanceTest and enter the License Key and User name associated with the license (if you don't have a license then the floppy version will only work for 30days).
3. Exit PerformanceTest.
4. Copy all the files from the PerformanceTest installation directory to an empty floppy disk. Depending on the version of PerformanceTest, normally the installation directory is C:\Program Files\PerformanceTest3.
5. That's it, you're finished. Using the Windows Start / Run menu option you can for example enter the following 'a:\pt.exe' to start PerformanceTest from the floppy disk.

You can now uninstall PerformanceTest from the PC harddisk and use the floppy in the future on other machines. When running PerformanceTest this way, there will be no files left on the PC after PerformanceTest has finished. The floppy disk must be left writeable however. IMPORTANT NOTE: If you are using the unlicensed valuation version you need to be careful that the date and time is set correctly on all the PCs.

Q. How often are the Baseline results updated?

Every few weeks or when we have a significant number to upload.

Q. Why aren't the MMX / SSE tests available on my computer?

MMX instructions only became available in later model Pentium Processors. If you don't have one of these CPUs then the test can not be run.

Q. Why is the network test so slow on my laptop?

Laptop network speeds are usually limited by the PCMCIA bus speed. You may have a 100 MBps network card, but the PCMCIA bus will not allow translmission rates much above 10 MBps.

Q. How can I get the network test to run through a firewall?

You may have to set up a rule allowing access on the port used by the network test. If you don't have direct control over the firewall, you'll have to contact your network administrator. The port number used by the test can be found in the network test dialog.

Q. The calculated CPU clock speed is not correct. Why?

1. CPU's made for portable computers can decrease their clock speed to lower power usage. For example, a Pentium III 1000 Mobile CPU may be measured as having a clock speed of 730Mhz at the time the measurement is done.
2. The speed rating values that AMD assign to their Athlon CPU's do not correspond to their clock speed. For example a Athlon 2000+ runs at a clock speed of 1660Mhz. This is also the case for the older Cyrix 150+ & 200+ CPU's

Q. I don't need to do any more tests, how do I uninstall PerformanceTest?

Use the "add / remove programs" icon in the Windows control panel.

Symantec Users

Customers who have purchased PerformanceTest as a part of Symantec SystemWorks 2003, should refer to the Symantec website www.symantec.com for support.

Appendix 1 - Default hard disk cluster sizes

Default Cluster Sizes for Volumes with Windows 2000 File Systems

Appendix 2 - Checking your cluster size

Versions of PerformanceTest later than 3.2 display the cluster size of the disk. In earlier versions of the software you'll need to check it yourself. Find or make a file on your hard disk with just a few bytes in it. Then while in Explorer right click on the file and select properties. Compare the values for 'Size' and 'Size on disk'. You can see that in the smallfile.txt example (left)  that the disk is used extremely inefficiently (0.015% efficiency in fact). This is because this disk has been formatted with 32KB clusters. If the disk was formatted with 4KB clusters (top right) the results would have been very different. (0.12% efficiency, a 8 fold improvement).

Now compare the PerformanceTest 3.1 and 3.2 disk results for these two partitions on the same ATA-66 7200 RPM 16GB hard drive.

Drive E: NTFS (4KB clusters: 512 byte sectors, 8 sectors per cluster)
Drive C: FAT (32KB clusters: 512 byte sectors, 64 sectors per cluster).
Note that PerformanceTest 3.2 provides more balanced results regardless of the cluster size.

Appendix 3 - Windows Caching

The graphs (left) show the response of Windows 98 to the PerformanceTest disk tests. The heavy use of the disk has resulted in Windows allocating more RAM to the disk cache. (18Mbytes to 36Mbytes in this case). There are several ways to avoid this type of effect.

1. Reboot the machine between each test run
2. Limit the amount of free RAM available for the cache (not as easy as it sounds)
3. Run the disk tests until the cache has reached it maximum size.