Performance and Power Consumption during Overclocking

We checked out the overclocking potential of Asus Rampage II Gene mainboard according to our standard methodology, but taking into account all the typical peculiarities of Asus mainboards that we are aware of. At first, as usual, we found the maximum base frequency when the board could work stably with the lowered CPU clock frequency multiplier. 215 MHz is a very good result, which is typical of many good mainboards. then we checked of the board could raise the frequency of our Kingston HyperX DDR3-1866 KHX14900D3T1K3/3GX memory modules to the nominal value of 1866 MHz. No, we had no issues or complications. We only had to increase the memory voltage to 1.65 V and QPI/DRAM Core Voltage – only to 1.325 V. The CPU wasn’t overclocked in this case, but further tests showed that even when the CPU is overclocked the board experiences no problems if the memory is working at higher frequencies.

These were all standard parts of our overclocking routine, now we should proceed with the special tricks. You may already know that Asus mainboards for Intel processors have one typical drawback that the company cannot eliminate for many years already. At some point during CPU overclocking the mainboard suddenly decides that it must increase processor core voltage to ensure its stability. In fact, it is a theoretically correct step and we increase the CPU core voltage ourselves during overclocking, but only when it is really necessary. The mainboard cannot know the potential of a specific CPU sample that is why it plays it safe and increases the voltage sooner than needed. Much sooner. As a result, Intel processor power-saving technologies partially stop working. In idle mode the multiplier will be lowered, but the voltages will remain increased. For example, Asus P6T mainboard allowed to overclock our Intel Core i7-920 processor “for free” only until 152 MHz base frequency. As soon as the frequency increased to 153 MHz or higher, the CPU core voltage was pushed up and we had to forget about energy-efficient overclocking for good.

Knowing these peculiarities of Asus mainboards, we decided to try and raise the base frequency to 160 MHz. The result was quite predictable: CPU core voltage was increased and remained increased when idle. Everything is just the way it was before, isn’t it? No it isn’t! It turned out that when “CPU voltage Control” parameter was changed from Absolute (VID) to Relative (Offset), Asus Rampage II Gene mainboard starts acting completely differently. If the voltage is increased, you can barely notice it, and all Intel processor power-saving technologies continue working as they should. Finally!

We know from our previous reviews that the optimal overclocking for our Intel Core i7-920 processor is by raising the base frequency to 181 MHz. In this case you should increase the memory frequency and lower the timings to 8-8-8-22-1T. The tests confirmed that Asus Rampage II Gene can work in this overclocking mode just fine. Only when we enabled protection against voltage drop under load, the board slightly increased it. If we disable “CPU Load-Line Calibration” parameter, the voltage won’t be sufficient to ensure CPU stability in overclocked mode. It is no big deal, because now when we set “CPU Voltage Control” parameter to Relative (Offset), the CPU core voltage may be increased freely. To ensure stability we had to increase it just by 0.0125 V, and all Intel power-saving technologies remained up and running in idle mode lowering not only the clock frequency multiplier, but also the CPU Vcore.

When the load increased, the processor clock multiplier rose to 21x according to the static implementation of Intel Turbo Boost technology raising the resulting CPU clock to 3.8 GHz.

As a result, we can sum up the main peculiarities of Asus Rampage II Gene behavior during overclocking in the following table:

Great! Now we should only find out if Asus Rampage II Gene is faster than Gigabyte GA-EX58-Extreme working in the same mode, or slower:

And again we don’t see any serious difference in performance results between these two completely different mainboards. In identical testing conditions they both demonstrate the same performance level. And what will our power consumption tests show? The results here turned out much better than in nominal mode:

In most cases power consumption of Asus Rampage II Gene mainboard is still higher than that of Gigabyte GA-EX58-Extreme, but this time the difference is no higher than 10 W, and under maximum CPU load the results of both boards level out completely.