OverclockingSince FX-8150 is advertised as an overclockable processor, and AMD kindly provided a great ASUS Crosshair V Formula motherboard together with AMD liquid cooling solution, I decided to give overclocking a more decent try than usual.
The idea was to try to find weak spots in FX-8150 factory defaults, or performance bottlenecks if you will, and try to improve on that to squeeze out every bit of extra performance while keeping power consumption at acceptable levels.
In the following charts you will find a total of five separate FX-8150 results and to understand what they mean here's an quick legend on them:
- FX-8150 - factory settings, no overclock, Turbo Core enabled
- FX-8150 northbridge overclock - NB speed set to 2.60 GHz, everything else untouched
- FX-8150 memory overclock - DDR3 2125 MHz, NB speed 2.27 GHz, HT 228 MHz, HT Link 2.50 GHz and CPU 3.63 GHz
- FX-8150 CPU overclock - overclocked via multiplier to 4.70 GHz
- FX-8150 combined overclock - CPU 4.77 GHz, HT 236 MHz, NB 2.60 GHz, DDR3 2201 MHz, HT Link 2.60 GHz
(click on image to see more details)
As you'll see from the charts, increasing memory speed from already fast 1866 MHz to ~2200 MHz can boost performance noticeably, while overclocking just the northbridge gives marginal improvements. Overclocking the cores of course gives you the biggest performance boost, but as shown you can squeeze out even more with balanced overclocking of all key components.
The achieved results are more than satisfying, although I was excepting a bit more from pure core overclock speeds, but it would seem Global Foundries needs to iron out that new 32 nm fabrication process. Speaking of fab process, there seems to be a distinct threshold for core voltage. If you step over it, power consumption and thermal output goes through the roof and is hard to keep in check even with water cooling.
For maximum stable clock, core voltage was set to 1.415 V, any lower and the system would crash almost instantly. With that voltage, an end speed of 4.77 GHz was achieved.
Also note that results below are without Windows 7 core scheduler and core parking patches applied. Overclocking was done before Microsoft published them, but as seen in regular tests, that shouldn't make much of a difference to these results.