But the other side of it is that the pre-mitigation microcode that was originally shipped did not actually fix the problem, hence the roll back on it. Furthermore, the Intel compliance PDF shows that my CPUID and yours
both contain the Core i7 920s - so which is which?
Either way, I don't see why the 965 EEs have a different CPU ID from the 975 and 980, which are grouped with the majority of the rest of the original generation Core i7s. So, I really don't understand what is going on here.
I know it is not the architecture, because the motherboard I have , which has the LGA1366 socket, can also accommodate that 975 and 980 - in fact, IIRC, that entire family of Bloomfield CPUs were all LGA1366 CPUs, and set up for triple channel memory (again, if memory serves me). So, I don't see why the 965 was partitioned off with a different CPUID, but then when the 975 (and then 980) were developed, that Intel would dump them back onto the older CPUID.
That, in fact, is what made me begin to doubt the veracity of the April compliance PDF in the first place.
Here is a comparison of the three:
Intel Product Specification Comparison