Gigabyte Mini ITX overclocking

[cbn]

http://www.anandtech.com/show/3769/reviewed-gigabyte-h55nusb3-miniitx-done-the-gigabyte-way/2

3.66 Ghz for a Core i7-K sounds very strong.

Does anyone believe the unlocked multiplier is factoring into this overclock. Am I right in assuming increasing BCLK would stress mainboard power circuits more than increasing the multi?

Lastly, how would a process shrink on future ITX mainboard chipsets (P65/H55, P75/H75, etc) affect overclocking? At the moment I believe Intel is using 65nm process for the p55 chipset.

 

[Zap]

It depends on the base multiplier of the CPU. If it is high enough, then it wouldn't stress the board too much. I think these days power circuitry often is the limiting factor, not bus clocks except for lowest end CPUs.

Die shrinks probably won't affect overclocking much, since these chipsets are basically just Southbridges. All the important stuff that is affected by clocks (PCIe lanes, memory controller) is in the CPU.

 

[cbn]

I think these days power circuitry often is the limiting factor, not bus clocks except for lowest end CPUs.

Thanks for your answer.

How about process shrinks in the power circuitry itself?

 

[Zap]

I don't have an answer for that question.

 

[PCTC2]

Thanks for your answer.

How about process shrinks in the power circuitry itself?

It's more physics than size that is limiting power circuitry. Between the current needed for an overclocked processor, the corresponding power draw and heat produced, and the physical limitation of the materials used for power regulation, without a major innovation, I don't see enthusiast power regulation on a mini-ITX board any time soon without sacrificing features for the real estate on the board.