AMD FX

Lore & Background

In the years prior to the AMD FX range, the AMD Phenom II and Athlon II lineup were generally competitive when price was taken into account, but by the end of Phenom's lifespan, Intel's Sandy Bridge-based Core processors provided performance that Phenom II could not match. Rumors suggested that the FX line would change that, based on leaked information about the upcoming Bulldozer architecture.

The FX series launched on October 12, 2011, with the Bulldozer architecture. The launch lineup included the 4-core FX-4100 at $115, the 6-core FX-6100 at $165, and the 8-core FX-8120 at $205 and FX-8150 at $185. The FX refresh on the Piledriver architecture launched on October 23, 2012, with the 4-core FX-4300 at $122, 6-core FX-6300 at $132, and 8-core FX-8320 at $169 and FX-8350 at $195.

Notable features included that all FX microprocessors were unlocked and overclockable, a feature usually reserved for Intel's high-end K suffix SKUs. In 2012, the personal world record for highest overclock was achieved on an FX-8350 at 8794.33 MHz; this record was broken in 2024 using an Intel Core i9 14900KF, but the FX-8350 continues to hold the AMD overclocking record. The chips used a module design: 4× dual-core modules in FX-8 series, 3× in FX-6 series, and 2× in FX-4 series, with two integer clusters (seen as logical cores from the OS) and a shared floating-point unit in each module. All models were manufactured from 8 logical cores with a simple Orochi die production, in a 938-pin μPGA package AM3+ socket. Unlike most Intel counterparts, FX chips offered no integrated graphics, a feature reserved for AMD's APU line.

Reader's Guide

Upon launch, the FX series was met with criticism from reviewers. Due to multiple cores sharing common resources, most tasks were substantially slower on the FX lineup than the Intel Sandy Bridge equivalent. In many single-threaded applications, it was worse than the previous generation of Phenom II microprocessors. Power consumption, while not as poor as the Phenom II generation, was still worse than what Intel provided at the time. The Piledriver-based FX refresh in 2012 generally improved performance across the board by increasing clock speeds at similar power consumption levels, but Intel's Ivy Bridge architecture was available and provided much better performance per watt and total performance. With AMD only responsible for 20% of consumer CPU sales in 2016, Intel continued to gain market share during the FX series' lifespan.

In 2015, AMD was accused of falsely advertising the core count of its FX lineup, claiming that every two cores shared one of multiple resources such as the FPU. The company settled out-of-court and voluntarily paid $12.1 million to California residents who bought a high-end FX chip.

The AMD FX line-up generally performed worse than its Intel competitors during its lifespan. Floating-point performance was relatively poor due to a single shared FPU per module. Most games could not take advantage of the high core counts. In applications that benefited from more threads, AMD SKUs typically pulled ahead, but at a great cost in thermal efficiency, often worse than the previous generation. Updates with Piledriver allowed higher clock speeds and better performance, but with even higher thermal output on the high end, as seen with the FX-9590's 220-watt TDP. The FX series was succeeded by the Ryzen brand of CPUs, based on the Zen architecture, which initially launched in 2017 to compete with Intel's later generation processors such as Skylake.

Did You Know?

• The FX-8350 set a personal world record overclock of 8794.33 MHz in 2012, which remains the AMD overclocking record.
• AMD settled a false advertising lawsuit in 2015 over the core count of its FX lineup, paying $12.1 million to California residents.
• The FX-9590 had a TDP rating of 220 watts.