README PC Magazine Labs Benchmarks Release 5.1 ---------------------------------------------------------------------- TABLE OF CONTENTS ---------------------------------------- Introduction..........................2 Installation..........................2 Operation.............................3 Command Line Options...............3 File Menu..........................3 View Menu..........................3 Performance Menu...................4 Compatibility Menu.................5 Quality Menu.......................5 Set Menu...........................5 Help Menu..........................6 Descriptions of Tests Processor Performance Tests........6 Instruction Mix.................6 128K NOP Loop...................6 Do-nothing Loop.................7 Integer Add.....................7 Integer Multiply................7 String Sort and Move............7 Prime Number Sieve..............7 Floating Point Mix..............7 Coprocessor Performance Tests......8 Coprocessor Speed Test..........8 Disk Performance Tests.............8 BIOS Disk Seek..................8 DOS Disk Access.................8 DOS File Access.................8 Video Performance Tests............9 Direct Screen Access............9 Teletype Without Scrolling......9 Teletype With Scrolling.........9 Memory Performance Tests...........9 Printer Speed Test................10 Additional Information...............10 README PC Magazine Labs Benchmarks Release 5.1 ---------------------------------------------------------------------- INTRODUCTION ------------ Welcome to Release 5.1 of the PC LABS BENCHMARK SERIES. Since the release of 5.0 last fall, we have received hundreds of registrations and have incorporated many of the changes and improvements you have asked for in our questionnaire. The latest features are outlined in the file called WHATS.NEW. This file will help pinpoint the new features for those who are already familiar with the benchmark programs. INSTALLATION ------------ The PC Labs Benchmark Series Release 5.1 is distributed on one low density floppy or as an archived file on PC MagNet. The archive contains the same files as the disk. The files are: BENCH5-1.EXE - The benchmark programs and data README - This file WHATS.NEW - New features description The BENCH5-1.EXE file is a self-extracting archive. To install the benchmarks just copy BENCH5-1.EXE to your preferred drive and subdirectory and type "BENCH5-1". Alternatively, you can make the destination directory your current directory and put the disk in drive A: and type "A:BENCH5-1". The quotes (") do not get typed. The following additional files should then be created (requiring approximately 750k disk space): BENCH.EXE - The benchmarks programs VGAHDW.EXE " EGAVGA.EXE " PRSPEED.EXE " PRSPEEDN.ASC " PRSPEEDW.ASC " FLOATMIX.EXE " BENCH.KEY " REGISTER - Form to send in to register for future updates Page 2 README PC Magazine Labs Benchmarks Release 5.1 ---------------------------------------------------------------------- OPERATION --------- To run the main benchmarks program, type "BENCH". Command line parameters: GO bypasses the initial logo screens. AUTO starts the benchmarks in automatic mode. The processor tests will cycle through one at a time. This option uses the BENCH.KEY file. NOTE: A reboot is required to end AUTO mode. LCD Uses black and white attributes. Use for forcing better display legibility on laptop computers. FILE MENU --------- The "Save results" selection allows you to save the test results in a file on your disk. You will have the ability to change the machine description (See SET MENU) and the name of the file. The file will be saved in the current directory and will overwrite an old file with the same name. The "About" selection tells about the benchmark program. The people involved with the project over the years as well as the languages and tools used to create the programs are listed. The "Quit" selection terminates the benchmark program. VIEW MENU --------- The "Results" selection allows you to browse through the test results on the screen. The format is the same as the results file. You can scroll up and down using the arrow, PgUp, PgDn, Home, and End keys. Use the Esc key to return to the main menu. Page 3 README PC Magazine Labs Benchmarks Release 5.1 ---------------------------------------------------------------------- PERFORMANCE MENU ---------------- The selections on this menu measure the performance capabilities of your machine. These are the indicators of the raw power of your machine. The "Processor..." selection activates the Processor Performance Tests submenu. The following tests are available: Instruction Mix 128K NOP Loop Do-nothing Loop Integer Add Integer Multiply String Sort and Move Prime Number Sieve Floating Point Mix Use the space bar to "check-off" the tests you wish to run and hit the Enter key. To abort hit the Esc key. Each of your chosen tests will be run and the results displayed. The "Coprocessor" selection runs the math coprocessor test. The results are displayed at the completion of the test. The "Disk..." selection activates the Disk Performance Tests submenu. The following tests are available: BIOS Disk Seek DOS Disk Access DOS File Access The "Video..." selection activates the Video Performance Tests submenu. The following tests are available: Direct Screen Access Teletype Without Scrolling Teletype With Scrolling Page 4 README PC Magazine Labs Benchmarks Release 5.1 ---------------------------------------------------------------------- The "Memory..." selection activates the Memory Performance Tests submenu. The following tests are available: Conventional Read Conventional Write Expanded Read Expanded Write Extended Read Extended write COMPATIBILITY MENU ------------------ The selections on this menu test the compatibility of a piece of hardware against a published specification. At this time there is only one choice on this menu. The "VGA Hardware" selection is only valid if you have a VGA graphics device. It tests your VGA adapter against the IBM standard. Any deviations from the standard do not necessarily indicate a defective or unusable board. Many application programs do not use all the available features and may also contain code to correct for deficiencies in your particular device. QUALITY MENU ------------ The selections on this menu represent tests which must be interpreted by the individual user. At this time there is only one choice on this menu. The "EGA/VGA Monitor" selection allows you to display various test patterns on your EGA or VGA Monitor using different display modes. The results will depend on your tastes with respect to size of characters, text readability, brightness, and contract as well as color perception. SET MENU -------- The "Enter info..." selection allows you to enter a name for the machine being tested. Up to 25 characters are available. The name you enter will appear on results screens as well as in the results file. Page 5 README PC Magazine Labs Benchmarks Release 5.1 ---------------------------------------------------------------------- HELP MENU --------- The "Help" selection presents a menu of help topics. Select the topic to view and hit the Enter key. Hit the Enter key again to return to the menu. The "Timer accuracy" selection allows you to test the accuracy of the internal system timer. By comparing the internal timer to an external time base, such as a stop watch, you can see if the results derived by the tests are valid for your machine. A correction factor is displayed to allow you to adjust for any inaccuracies in your timer. DESCRIPTIONS OF TESTS --------------------- PROCESSOR PERFORMANCE TESTS PLEASE NOTE: Since the execution speed of microcomputers is incredibly fast, the functions described in the Processor Performance Tests are performed many times in a loop to produce the published results. While the number of iterations varies from test to test, they are consistant within each test when testing different machines. INSTRUCTION MIX --------------- The Instruction Mix benchmark test times a series of tasks specific to the 8088, 80286, or 80386 chip. Since this test shows how the CPU operates in the context of the bus, processor, system memory, and motherboard architecture, a faster time means better overall computer performance. 128K NOP LOOP ------------- The 128K NOP Loop benchmark test times the execution of a loop containing 128K of NOP instructions. Page 6 README PC Magazine Labs Benchmarks Release 5.1 ---------------------------------------------------------------------- DO-NOTHING LOOP --------------- The Do-nothing Loop benchmark test times the execution of a loop containing one NOP instruction. INTEGER ADD ----------- The Integer Add benchmark test times the execution of a loop containing an ADD instruction. INTEGER MULTIPLY ---------------- The Integer Multiply benchmark test times the execution of a loop containing a IMUL instruction. STRING SORT AND MOVE -------------------- The String Sort and Move benchmark test times the execution of a bubble sort performed on 200 random strings containing 16 characters each. PRIME NUMBER SIEVE ------------------ The Prime Number Sieve benchmark test times the execution of a routine to find the prime numbers between 0 and 8190. FLOATING POINT MIX ------------------ The Floating Point Mix (also known as the Floating Point Calculation Without Coprocessor) benchmark test sets up a floating point emulation program in RAM and then exercises the processor and tests RAM access speeds during floating point calculations. The floating point processes performed are identical to those used in the Coprocessor Test. Page 7 README PC Magazine Labs Benchmarks Release 5.1 ---------------------------------------------------------------------- COPROCESSOR PERFORMANCE TESTS COPROCESSOR SPEED TEST ---------------------- The Coprocessor Speed benchmark test exercises the coprocessor using the same floating point calculations used in the Floating Point Mix Test. In addition to testing the speed differences of coprocessors in different machines, the speed gained by using a coprocessor can be seen by comparing the two floating point tests on the same machine. DISK PERFORMANCE TESTS BIOS DISK SEEK -------------- The BIOS Disk Seek benchmark test measures mechanical track-to-track disk drive access times. Fast times are helpful with programs such as databases, which often store and must later find data in many separate places on a drive. Both sequential and random access is tested. DOS DISK ACCESS --------------- The DOS Disk Access benchmark test measures the time necessary to perform 1000 read requests at random locations on the disk via DOS calls. DOS FILE ACCESS --------------- The DOS File Access (Small and Large Records) benchmark test times disk throughput as a result of mechanical disk drive speed, hard disk controller function, and bus speed. The test may be used to show the effects of disk caching by running both with and without a cache. Fast times for a particular record length are advantageous when working with an application that uses a similar record length. Page 8 README PC Magazine Labs Benchmarks Release 5.1 ---------------------------------------------------------------------- VIDEO PERFORMANCE TESTS DIRECT SCREEN ACCESS -------------------- The Direct Screen Access benchmark test times the writing of data directly into screen memory. Fast times are advantageous for application programs that bypass slower BIOS and DOS calls and write directly to the screen. TELETYPE WITHOUT SCROLLING -------------------------- The Teletype Without Scrolling benchmark test times the writing of data to the screen using BIOS calls. Fast times are advantageous for programs that display large amounts of data in a non-scrolling fashion. TELETYPE WITH SCROLLING ----------------------- The Teletype With Scrolling benchmark test times the writing of data to the screen using BIOS calls but adds the time necessary to scroll the entire screen up one line at a time once the screen is initially filled. This test writes the same amount of data as the non-scrolling test. The difference between the two times indicates the overhead attributable to scrolling the screen. Fast times are advantageous for application programs that frequently cause the screen to scroll. MEMORY PERFORMANCE TESTS ------------------------ The Memory Performance benchmark tests indicate the time needed to read and write data to conventional, expanded, and extended memory areas. Due to advances in the 80386 processor chip, the results obtained using the Extended Memory tests with this processor are not valid. The test uses the BIOS function for switching in and out of Protected Mode. The BIOS call is not commonly used to perform this function on a 386 machine. Page 9 README PC Magazine Labs Benchmarks Release 5.1 ---------------------------------------------------------------------- PRINTER SPEED TEST ------------------ The Printer Speed benchmark test times the throughput of your printer by printing a typical two-page business letter. Because this is a throughput test, our results are almost always slower than the speeds claimed by manufacturers, who usually clock a one line "burst" speed. ADDITIONAL INFORMATION ---------------------- Additional information may be found accompanying reviews in PC Magazine. If you have any specific questions or suggestions, please send them along with your registration form or in a separate letter addressed to: PC Labs Benchmark Registration, 1 Park Avenue, New York, NY, 10016. PC MagNet is also an excellent source of information as well as a place to post questions. Thank you for using the PC Labs Benchmark Series Release 5.1. Sincerely, Frederic E. Davis Director, PC Labs Page 10