|
Advice Beginners BIOS Guide CPUs Links Mainboards Memory Network Storage Video/Sound Cards Contact Forum SiteMap Sponsors WebNews Home |
. | . |
|
Prices: CPU Intel P4 2.4C-800 P4 2.6C-800 P4 2.8C-800 P4 3.0-800 P4 3.2-800 AMD AthlonXP XP 1700+ XP 2000+ XP 2400+ XP 2500+ XP 2700+ XP 3000+ XP 3200+ Athlon64 Athlon64 3200+ Athlon64 FX-51 Opteron Opteron 240 Opteron 242 Opteron 244 Opteron 246 Memory Corsair Crucial Kingston Mushkin OCZ |
LOSTCIRCUITS |
|
|
| ABit BH7 The Legend Lives On | ||
| (Review by MS, March 31, 2003) |
I/O Performance
We hinted earlier at the fact that the use of the Marvell bridge chip to interface one SATA connector with the ICH4 may not be the greatest idea for the main reason that the use of the SATA connector actually kills the entire secondary IDE channel. Another issue we did not go into yet is the performance of the bridge solution in terms of raw burst transfers. Keep in mind that the UATA 100 interface natively can only digest 100 MB/sec which means that putting a high spee interface in front of it can have all kinds of adverse effects, starting from a simple backlog to dropped packets and retry that effectively could cut the performance in half or less.
HDTach
It is certainly not the latest and greatest HDD benchmark but it still has some merits, especially for single drive configurations. In this case, it turned out that the SATA interface is running quite a bit more slowly than the parallel interface which was somewhat along the lines of what we thought might happen.
SiSoft Sandra 2003 FileSystem
Left: PATA Drive, first partition (outer tracks), center: same drive, last partition (inner tracks), right: SATA drive outer tracks. Note that the burst transfer rate (buffered reads and writes do not change as one goes from the outer to the inner tracks. However, on the SATA drive, the burst performance is the same as shown by HDTach. Click thumbnails for larger images.
Leaving the SATA issues aside, it is time to move on to application performance.
next page: => Winstones =>