Gigabyte GA-EP45T-Extreme Motherboard Review

Benchmark Methodology

Today's benchmarking sections are going to be a bit different than those done in the past. Normally we run the benchmarks of our overall overclock against the XMP defaults but since our overall overclock isn't much different from that of the XMP profile, we are going to change things up a bit. Using the same Corsair Dominator TWIN3X2048-1800C7DFIN memory that we used for overclocking, we will be comparing the XMP profile against the system as it powers up as if there were no XMP profile. This will provide a very unique look at just how much performance an XMP enabled kit of memory can bring to the table, without the end user touching all but one setting in the BIOS. On X38/X48 based motherboards, it is even easier as the XMP profile tends to load all by itself.

The blue results will depict the system as it would be with system defaults loaded. We will make absolutely no adjustments to the BIOS meaning memory will be running at the default DDR3-1333 with whatever timings the Dominator memory has programmed for SPD (9-9-9). The FSB will be set to 333x9 equating to a 3.00GHz CPU frequency. The red results will be the default XMP profile for the Dominator memory and thus upping the FSB to 450 with a CPU multiplier of 7 for a 3.15GHz CPU frequency and of course, the memory at DDR3-1800 with 7-7-7 timings. The following will outline the rest of the testing methodology:

1. Windows is installed using a full format.
2. Intel Chipset drivers and accessory hardware drivers (audio, network, GPU) are installed followed by a defragment and a reboot.
3. At time of benchmarks the latest drivers were downloaded from their official web sites as the latest drivers, most notable, Intel Chipset 9.0.0.1008 and ATI Catalyst 8.7.
4. Programs and games are then installed followed by another defragment.
5. Windows updates are then completed installing all available updates followed by a defragment.
6. Benchmarks are each ran three times after a clean reboot for every iteration of the benchmark unless otherwise stated, the results are then averaged.