Stories By Fabian Pascal32-Bit ComputingPerformance
Tweaks Performance
Tweaks The
Communications Subsystem Best
of Breed System -- 2nd Generation: Intel "Compatibility" Under Windows NT Disk SubsystemPC Advisor Column (www.dsi2000.com) Selecting a Slot 1 System board
Also of Interest: Transferring Files With Flipdisk By Ed Kahn By Woody Liswood Linux: By Kevin Savetz and Neil Randall Open Source Software: More Than An Emerging Alternative By Michael Robin |
|Feedback To Author | Feedback To Webmaster Copyright © 1999 by Fabian Pascal. All Rights Reserved 32 Bit ComputingBENCHMARKS AND TESTING TOOLS Applications (System) Level Winstone 98, 99 (www1.zdnet.com/zdbop/) Benchmark suites running leading business and high-end Windows applications scripts of common activities and produces a weighted system performance score from the individual application scores, using their market-shares as weights. The 98 version runs applications individually and produces scores for each; the 99 version multitasks applications and provides only one overall score. The overall score is scaled to a base system's score of 10.0. where >10 score indicates higher performance than the base system's. Video Subsystem DisplayMate (www.displaymate.com) A comprehensive set of video utilities for setting up, tuning-up, testing and evaluating video displays and complete video systems directly or indirectly to a computer. Can be used to compare and evaluate displays and thoroughly test entire video systems for performance and compatibility, including the monitor, video board and video BIOS and printer setup and calibration.. Disk Subsystem HD Tach (www.tcdlabs.simplenet.com/) Bypasses the file system to measure performance of fixed, removable and floppy drives, namely: random access time; read/write transfer rates from/to areas all overthe drive (graphs and reports maximum, minimum and average rates);read burst rate (a measure of the drive and controller interface performance). ThreadMark (www.adaptec.com) Uses a mixture of single and multithreaded read and write operations across a range of block sizes, to measure how a multidrive disksubsystem handles periods of high I/O activity. Operation results are weighted according to their occurrence in a typical environment,then averaged to produce single scores for data transfer rate and CPU utilization. CDTach 98 (www.tcdlabs.simplenet.com) Measures inner, center and outer track transfer rates, full stroke and random access time, CPU usage at any speed, and drive buffer/interface burst performance. Benchmark Caveats Benchmark Illusions ( www.x86.org/ddj/Mar98/Benchmarks.html) Benchmark Limitations ( www.intel.com/procs/perf/limits.htm)
|