In the Forums...
Posted: July 30, 2001
Written by: Tuan Huynh
Benchmarks can be divided into two kinds, component and system. Component benchmarks measure the performance of specific parts of a computer system, such as a microprocessor or hard disk, while system benchmarks typically measure the performance of the entire computer system. In either case, the performance you see in day-to-day use will almost certainly vary from benchmark performance, for a number of reasons. First, individual components must usually be tested in a complete computer system, and it is not always possible to eliminate the considerable effects that differences in a system design and configuration will have on a benchmark result. For instance, system vendors sell systems with a wide variety of disk capabilities and speeds, system memory, system bus features and video and graphics capabilities, all of which influence how the system components (such as the processor) and the computer system perform in actual use and can dramatically affect benchmark results. Also, you may not actually purchase the exact components we use in your benchmark system. This is just a reference you can base your purchase decisions on. Also, differences in software, including operating systems and compilers, will affect component and system performance. Finally, benchmark tests are typically written to be exemplary of only a certain type of computer application, which may or may not be similar to your applications.
Benchmarks are, at most, only one kind of information that you may use during the purchasing process. To get a true picture of the performance of a component or system you are considering purchasing, you must consult other sources of information (such as performance information on the exact system you are considering purchasing). You may also want to try actually sitting down and using the system if possible. There’s no better way to get a feel of a system than actually using it.
Let’s take a look at how the Pentium 3 Tualatin performs……
I’m using SiSoft Sandra CPU test because end users use it for benchmarking. The Pentium 3 at equal clock has always fallen behind the Athlon in the CPU tests, lets see if the Tualatin can keep up.
Here we see the Athlon still scoring higher then the Pentium 3 1.2 GHz. The Pentium 3 at 1.2 GHz is still able to keep up with the Athlon falling only approximately 4% behind the Athlon 1.2 GHz in the ALU test. Even at a speed of 1.2 GHz, the P3 1.2 GHz Tualatin is starting to catch up with Intel’s own P4 1.5 GHz. We all know the Athlon’s has a strong FPU unit, but here we see the P3 1.2 GHZ narrowing in on the Athlon falling only 3% behind.