Yeh, I kinda got the feeling, from reading the user manual, that Native IDE was not the same as IDE. So, what IS the difference, then?
As for the driver, well, I find it all a bit confusing. What you say makes sense but, on the hand, no driver is required for the Intel SATA controller (the H55 chip), so one would have thought no driver would also be needed for the Gigabyte controller, when operating in IDE mode.
I prefer to install all chipset drivers that I'm ever likely to want just after installing the OS and its service pack, and to keep that to the minimum. I've no particular plan at present for using the Gigabyte chip for SATA but in the coming weeks and months I might still want to experiment a little with its usage. From my experience with mobos from other manufacturers, it's always best to install chipset drivers in that slot between effectively the OS being installed and the rest of one's software being loaded. So, in respect of this Gigabyte SATA driver, I've had it in mind to initially enable the Gigabyte SATA chip in the BIOS, load its driver, then disable the chip. That'd mean that, later on, if I want or need to use that chip, all I'll then need do is to re-enable it via the BIOS. Does that all make sense?
Addendum: The handling of interrupt requests has always seemed a bit of a black art to me, but my understanding is that Legacy IDE is for OSs like Windows ME and Windows 98, where the IDE ports would be required to use specific, non-shared IRQs. Native IDE is for newer OSs like Windows XP, Vista and 7, where IRQs are shareable. So, given that I'll be using Windows XP and I don't wish to use RAID or AHCI, my assumption has been to set up the BIOS in this regard as specifically Native IDE. In particular, I've presumed that, for the Intel controller, I should choose "IDE" and then enable "Native IDE" mode, and for the Gigabyte controller, I should choose "IDE", rather than "RAID/IDE" or "AHCI". Right?
