Our testing methods
As ever, we did our best to deliver clean benchmark numbers. Tests were run three times, and the results were averaged.
Given the target market of the chips tested, we used Shuttle's SV24 barebones system to benchmark the processors. The system's specs are as follows:


Gigabyte was kind enough to send over its GA-6VEML motherboard to test the processors, but things didn't work out all that well. It seems that Gigabyte requires you to unplug the PSU and set a jumper before you add or remove memory. I, in my infinite wisdom, fried the board halfway through testing when I failed to follow these simple instructions. In my defense, however, these instructions are buried away in the manual without the blinking red lights and neon lettering that should cover something like this. So, while I'm an idiot for frying the board, it shouldn't have been that easy to do.
Shuttle's FV24 motherboard, found in the SV24, shares the same chipset and basic components as the GA-6VEML. I retested everything on the FV24, and predictably, the results closely matched what I had been able to collect on Gigabyte's motherboard before my RAM-swapping adventure.
The test systems' Windows desktops were set at 1024x768 in 32-bit color at a 75Hz screen refresh rate. Vertical refresh sync (vsync) was disabled for all tests.
We used the following versions of our test applications:
All the tests and methods we employed are publicly available and reproducible. If you have questions about our methods, hit our forums to talk with us about them.
| Windows 8 frightens me, and here's why | 237 |