Navigate: 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 Search Prices:

IDE Performance

The main reason for the delay of this review has been that we encountered some issues with the IDE controller of the new SouthBridge and it is still not quite clear what exactly happened. The first "casualties", that is a CDROM and one IBM drive could possibly be attributed to the IDE driver bug in Windows2000. After that, we switched to WinXP, got ourselves a new board, new IDE cables and new drives and, to make a long story short, we still got corrupted drives. And we admit that the IBM drives are very delicate but three drives in 2 days of test time is somewhat beyond a random failure rate.

HDTach 2.61, Onboard IDE

We did notice that HDTach returned some rather low values for the burst rate and often we got some serious depressions rather than inverted spikes.

A "good" run of HDTach using the onboard IDE controller. Note that the burst rate is limited to some 81 MB/s. In some runs, the two inverted spikes in the middle of the trace were fused to a flat-line at about 1300 kB/sec before going back to normal transfer rates. We saw these drop-outs with all 60 GXPs used at different positions and independent of whether drives had been freshly low level formated.

HDTach 2.61, Maxtor ATA133 PCI Card

Using the Maxtor ATA133 controller card, the same drive as above performs flawlessly and with a much higher burst transfer rate every single time except for the partition boundaries that give drops. We continued to use the Maxtor "Promise PDC20269" controller and had no further issues.

One of the new features of the SIS963 is the added support for ATA133. The IBM drives are only ATA100, however, the controllers are backwards compatible and autoselect the correct speed. Most likely, we are looking at a driver issue here, the fact that the IDE drivers did not work at all in Win2K would argue in favor of this idea. As long as it is software, it should be easy to fix the problem. Keep in mind also that ATA133 is at the very edge of what is technically possible with the parallel 16 bit technology used in current controllers and that at this speed, drive corruption is one of the major challenges the HDD industry faces.

next page:    => Gaming Performance =>