Overclocking Performance

Overclocking performance was somewhat less spectacular than the competition, but still pretty good. We managed to squeeze 3.72GHz out of our Core 2 Quad Q9550 (8.5 x 438MHz FSB) at 1.4325V Vcore, 2.3V memory, 1.38V VTT, 1.75V CPU PLL voltage, 1.25V southbridge and 1.4V P45 MCH. However it was eclipsed by the cheaper Gigabyte GA-EP45-UD3R, clocking the same Q9550 to just a touch higher at 3.91GHz (8.5 x 460MHz) at a very similar set of voltages, but just with slightly more CPU PLL at 1.77V, with a 1.4V MCH and 1.2V ICH10.

In general the overclocking options are very easy to work with for the Asus boards, with helpful auto-overclocking tools from the CPU Level Up feature, to more hardcore alternatives for seasoned professionals already explained in the BIOS section. Thankfully, the BIOS has not been tamed for "gamers" - there is still everything there ready to use, but getting that full overclocking headroom out of the Gene proved more difficult.

The board flat out resisted to overclock with manual memory timings and frequency set. Despite the stock clock of 1,066MHz ran successfully automatically by the board (we were using 1,150MHz memory) it would not allow anything other than auto when cranking the FSB. In comparison, the Gigabyte board was happy to reach into the clouds with an elevated memory performance, while still stretching the front side bus option even further.

Despite the options available, we didn't adjust the clock skews and GTLs of either board because, quite frankly, it's a dark art that takes too long and we were interested to see what the board could handle left to "Auto". We did however play with the OC Charger, Clock Twister, Static Read Control and Read Training to aid memory overclocking, but nothing helped the memory specifically, only the Clock Twister aided in getting a few more MHz on the FSB.

Not that we'd complain at a 3.7GHz quad core overclock mind and clearly no overzealous use of power phases are needed: eight is simply enough. Under overclocked load the heatpipes naturally get pretty hot, but don't feel like they will spontaneously combust, although we'd still recommend some extra airflow - particularly for the Southbridge. If you are going multi-GPU with two graphics cards over it, it's then highly recommended that there's a fan parallel to the board over that whole area for long term stability, especially in a small case.

Power Consumption

Resetting the BIOS to default settings we used the exact same set of hardware in all our tests below (listed in the test setup page), only changing the motherboard in between. The Vista power setting was set to Balanced and all CPU power saving states were restored to their BIOS default values. The Asus board does no feature any advanced Energy Saving (EPU) soft/hardware onboard, however we did enable the Gigabyte's Dynamic Energy Saver software to test against it, since it was available. By testing at the wall power consumption, we know exactly what kind of typical power consumption we'll be paying for.

Idle Power Consumption

Power at wall socket.

  • Asus Maximus II Gene
  • Gigabyte GA-EP45-UD3R
  • 102
  • 99
0
25
50
75
100
Watts (lower is better)

Load Power Consumption

Power at wall socket.

  • Asus Maximus II Gene
  • Gigabyte GA-EP45-UD3R
  • 164
  • 157
0
25
50
75
100
125
150
175
Watts (lower is better)

As we'd expect the Asus is a bit more power hungry despite its size because it doesn't feature specific power saving bits on board. That's not that we'd want them on an overclocking gaming board mind - we're not shedding a tear over a few watt here, however it is worth noting that without its DES enabled the Gigabyte board is still more energy efficient, sipping just two watts more without the software running and still undercutting the Gene.
Discuss this in the forums
YouTube logo
MSI MPG Velox 100R Chassis Review

October 14 2021 | 15:04