Setup can also have an affect on overall system performance. One thing I did not cover previously is that by changing from the integrated SiS IDE ports to the High Point Technology ATA/133 port (using the Ultra ATA mode instead of RAID), I was able to get a significant gain in the disk scores under both Winbench and Winstone.
Did you notice that even though the HPT IDE port showed a larger percentage gain in the Business Winbench test results it showed a smaller percentage gain in the Business Winstone test? Kind of interesting since the Winbench disk tests are using the same data and applications as the Winstone tests. I would guess the reason is that disk access has less influence with the type of applications run and data used. Along with the fact that Winbench measures only the actions that result in I/O operations, while Winstone measures the operations that do other things – thus diluting the impact of I/O, as is the case in the ‘real world’.
And what about the upcoming 533MHz FSB iP4 (133MHz SysClk) due later this half of 2002? How much of an impact will it have? Well I was able to do some testing using the Willamette core 2.0GHz CPU:
As you can see, it makes a difference, but not much. I also tried setting the memory to 166MHz (DDR333 / PC2700) when using a 100MHz SysClk, but there was no gain in performance. However, when using a 133MHz SysClk and 166MHz memory there was a gain in performance.
Be the first to discuss this article!