I'm still struggling, trying to find clues that'll lead to a solution to the instability that I constantly witness on my newly-built US3L, namely the constant rhythmic access to the HDD, random accesses to the FDD, gross sound distortion, and occasional BSODs. Yesterday, I had another look at the device allocations for IDE in the BIOS, and once again I confess that I find the way that the channels have been named and handled by the designers quite baffling. As to the available ports, this is what I see in Standard CMOS Features:
IDE Channel 0 Master
IDE Channel 0 Slave
IDE Channel 1 Master
IDE Channel 1 Slave
IDE Channel 2 Master
IDE Channel 3 Master
IDE Channel 4 Master
IDE Channel 4 Slave
IDE Channel 5 Master
IDE Channel 5 Slave
Do appreciate that I don't use RAID on this new PC. I use the Intel SATA ports, and only in IDE mode.
Okay, one can see that the nomenclature used by the BIOS in this regard is really of the old PATA arrangement, but forget that for the moment. Clearly, by looking at the board itself, there are 10 physical ports available. We know that six of these are Intel SATA ports, two are Gigabyte SATA ports, and two are a primary/secondary PATA port. But it's not at all obvious from this BIOS listing which channel in the listing corresponds to each port. I've already done a certain amount of experimenting and it seemed to merely put the three SATA drives I use - a HDD and two optical drives - at the head of the list, irrespective of the ports on the board into which I plugged them. This seems odd in the extreme.
Notice that Channels 2 and 3 only have Master allocations. Why this is so seems to be a complete mystery. The text on this in the BIOS separates out Channels 2 and 3 from 0 and 1 but doesn't explain why. Could it be that Channels 2 and 3 are meant to be reserved for the two Gigabyte ports only? Or is this entire nomenclature just randomly chosen and meaningless?
If Gigabyte has kept to the old PATA convention, then at least one of these master/slave pairs must be for the onboard PATA connector. But which one? It can only be determined, it seems, by trial and error. (I've not done tried it, as I've no reason to use a PATA drive in the PC).
The information that the INF driver puts into Device Manager in Windows further confuses the issue. This is what I see there, with the Gigabyte SATA Controller disabled in the BIOS so as to simplify things:
Intel 5/3400 Series chipset 2 port Serial ATA storage controller
Intel 5/3400 Series chipset 4 port Serial ATA storage controller
Primary IDE channel
Primary IDE channel
Secondary IDE channel
Secondary IDE channel
To which onboard Controllers do the 2 port and 4 port storage controllers refer? And looking at the Properties of the four IDE channels listed there give only a fleeting clue, through the transfer mode settings, as to which physical drives they refer to. Indeed, one of the channels is permanently greyed out in PIO mode. This, I believe, is the DVD/RW drive, but there seems to be no way of confirming that.
The HDD is currently connected into the first SATA port (SATA2_0). What I'm wondering is whether somehow I'm unknowingly bucking Gigabyte's intention of using certain of my drives in particular ports. Perhaps the onboard Controller goes haywire if you don't use the ports in a certain way? Perhaps it's mistaken to plug the HDD into SATA2_0?
Anyone got any thoughts on this? It's all about as clear as mud to me, at the moment. Whatever is hammering away at my HDD all the time is dominating the PCI bus. It's as though the Intel Controller (whatever one of the two referred to in Device Manager it is) is constantly struggling to sort the matter out but never succeeding.
