Treading the ‘boards

Justin Robinson weighs up the pros and cons of your next p45-based motherboard.

Ever since the original P35 Bearlake chipset came out, enthusiasts have enjoyed a great chipset with a solid feature set. With the successor, the name has changed to ‘Eaglelake’ – definitely indicative of the speed of the new series.

With the combined effort of Intel’s massive blue piggybank, and an R&D team that would make a small country look miniscule, this new chipset seems strangely familiar in both name and function. More and more manufacturers are jumping on the Intel team, pumping out a torrent of boards, washing away most people’s hopes for an easy choice – but that’s where we step in.

This month we place ten motherboards in a circle with a sharpened stick, and only one will survive our testing and come out on top. Which board will it be? Is the P45 a worthwhile upgrade from the P35? Well, you’re going to have to keep reading to find out.

How We Tested
A motherboard roundup involves very precise measurements of the performance of each board, especially in cases where each mobo is based on the same chipset, offering a similar amount of expansion options. We chose some of our more powerful hardware, and used this in each mobo at stock settings, 3.4GHz and 3.8GHz where possible.

This is what we used with each board:

P45-based motherboard
Intel E8400 @ 333x 9 = 3000MHz
TeamXtreem DDR2 @ 1066MHz, Patriot Viper DDR3 @ 1066
Antec TruePower Quattro 1000W
WD Velociraptor 300GB
Thermalright Ultra 120 Extreme

Each system used the same components to ensure that results are easily compared, and the parts were chosen to reflect a common enthusiast’s system. The operating system was Microsoft Windows Vista Home Premium 64-bit SP1, with no audio or network drivers installed.

We used three different programs to benchmark the boards: SuperPi 4M to test the memory bandwidth, wPrime to test the CPU speed and Cinebench R10 to test how well the motherboards can handle scaling with multiple cores present.
As well as theoretical benchmarks at stock speeds, we pushed each board to the same levels (though a few could not make it) and recorded how well they were able to handle the speed. The physical layout of the boards was also looked at, and the features of each board are compared in the table over the page. Any irregularities such as heat, inability to overclock and potential limitations with the BIOS’ were also noted.

Now that you know how we went about testing, and before we get into our actual reviews, here are the raw comparison results:



Want to test your own rig?
There’s no better way to get an idea of what our figures mean than by benching your own gear. With that in mind, here’s where you can go to find the programs that we use in this Head2Head:

SuperPi
wPrime
Cinebench R10

Happy testing, and on with the show!