Core for core is the performance the same across the entire family? For instance, does an i7 2600 with hyperthreading disabled perform the same as an i5 2500 at the same clocks? Are there differences in the uncore? Would a single threaded app perform the same on and i3 vs an i5 & i7 at the same clocks?
Also, comparison of chipsets only lists feature differences. overclocking and sata3 aside, does a H61 have the same IO performance as P67, Z68?
*edit* I'm aware there are differences in cache size.
just wondering, Thanks
In short, yes the i3, i5 and i7 will perform ROUGHLY the same per core disregarding frequency, cache and hyper threading.
As far as IO on motherboards, it depends on the motherboard. In general, the memory speed is the key IO metric considering that almost no peripheral will saturate a PCIE 2.0 x16 lane. Performance wise, I would assume H61 performs roughly similar to P67 and Z68, though I do not claim to be an expert on motherboards. The biggest performance factor here is the features, bios, drivers and other software provided by the motherboard's manufacturer.
The overall explanation about your question is as follows. Ignoring performance metrics such as frequency, cache size etc. is basically removing the key variables that cause a CPU to be defined as an i3, i5 or i7 and thus more/less expensive. All of these CPUs have the same underlying microarchitecture, Sandy Bridge, and will therefore perform identically if you removing everything that makes them different.