chipset identification

[rev1.0]

I am having trouble positively identifying the chipset on my kt266a motherboard. I am aware that Viatech renamed the northbridge chip from vt8366A to KT266A but 'am confused as to the stepping code. As I understood, the original northbridge chip from the kt266 chipset (vt8366) had a stepping code of (XXXXCD). With the advent of the enhanced replacement chipset, originally named kt266A to denote the (vt8366A northbridge and vt8233C southbridge combo), the stepping code of the vt8366A was made (XXXXCE). Although Viatech could have chosen a less confusing name change, I can follow it well enough. However, the KT266A northbridge chip that I have is sporting a stepping code previously used by the original vt8366. In addition, every monitoring and benchmarking program I have used, and the bios itself, reports the chipset as being the combo of vt8366 and vt8233. There are a number of other factors that have left me a little suspicuous. Has anyone else with a kt266A motherboard had any similar experiences? I would greatly appreciate hearing from anyone able to explain why this condition exists and would like to know if anybody has also noticed this on any of the kt266a motherboards they may have. Thanks.

rev1.0

[Peter M]

This is because a PCI chip revision never alters the device ID. Likewise, from KT266 to KT266A, the software readable device ID remained unchanged, only the revision byte was upped. This is how it's meant to be - as long as the chip's programming interface essentially remains the same and is compatible to existing drivers, BIOSes and software, you don't use a new device ID.

So software that knows the KT266's PCI device ID but doesn't know that a KT266 from revision xx onward is to be called KT266A will still report and handle it as a KT266. Existing software won't use the revised chip's new features (if any), but at least it'll work as good as it did on the old one. That's the idea behind all this, so your software acts perfectly fine.

regards, Peter