AMD A10 5800K Launch Review



Product: AMD A10 5800K / ASUS F2 A85-M Pro
Company: AMD
Author: James Prior
Editor: Charles Oliver
Date: October 1st, 2012

Introduction

Following our preview of the 2nd Generation A-series APU last week, the full performance embargo has lifted and we can now bring you our full review of the new Virgo platform. Trinity is launching at the price points listed below, which to be honest quite surprised yours truly; I had anticipated higher. The pricing, $122 for both A10 models, takes aim at the second and third generation Core i3 series of which there are four similarly priced models - the Core i3 2100, 2120, 2130 (Sandy Bridge) and 3220 (Ivy Bridge).

AMD A10 5800K APU
APU Pricing

Our first question was why there were two A10 models at the same price point - what's different? There are three key differences between the A10 5800K we're testing and the A10 5700. The biggest is thermal design power; the A10 5800K is rated at 100W while the A10 5700 is 65W. This difference is made up from two sources: CPU clocks and the GPU clocks. The A10 5800K has 200MHz higher Turbo clock and 400Mhz higher base clock, running 3.8GHz / 4.2GHz vs. the A10 5700's 3.4/4.0. The GPU in the 5700 is still rated the same, both A10s being marketed as featuring the Radeon HD 7660D despite the 40Mhz decrease in GPU engine clock from A10 5800K to A10 5700. The final difference is the overclocking ability, the K suffix denotes that the A10 5800K is unlocked for upward CPU multiplier adjustment whereas the 5700 is not. The large difference in power rating indicates that the 5700's are likely binned for lower voltages to run the designated speeds, whereas the 5800's are 'leakier' and thus able to clock up a bit more right out of the box.

Specification Overview - A10 5800K with Radeon HD 7660D graphics

  • Quad core CPU [Piledriver architecture, two modules]
  • 3.8GHz CPU core clock / 4.2GHz CPU all-core Turbo
  • 1.3Bn transistors / 246mm2 die size
  • Manufactured on 32nm SOI from GlobalFoundries
  • Socket FM2
  • Dual Channel 64-bit Memory Controller
    • Two DIMMS - 1866MHz
    • Four DIMMS - 1600MHz
    • Low Voltage DIMM support (1.35v)
    • Simultaneous Read/Write in unganged mode w/four DIMMs
    • Dynamic Speed Control for power savings
    • Supports AMD Memory Profiles (AMP) & Intel XMP
  • 384 Radeon Core 2.0 [VLIW4 architecture, 96 ALU as 6 SIMD]
  • 800MHz GPU core clock
  • 24 Texture Units / 8 ROPs
  • BIOS controlled dedicated GPU system RAM
  • 256-bit Radeon Memory Bus
    • GPU access to CPU coherent memory space
    • CPU access to GPU dedicated buffer
  • Hardware Accelerated Media Decode - UVD3
  • Hardware Accelerated X264 Encode - VCE
  • DisplayPort 1.2 / DVI / HDMI supporting 3 Display Eyefinity
  • DirectX 11 / OpenCL 1.2 / OpenGL 4.1 / C++ AMP
  • 100W TDP with AMD Turbo Core 3.0 bidirectional power monitoring
  • 736 Single Precision GLFOPS [614.4GFLOPS GPU, 121.6GFLOPS CPU]
  • 184 Double Precision GFLOPS [153.6GFLOPS GPU, 30.4 GFLOPS CPU]

A few more abilities are added, depending on the chipset used in your selected mainboard. The platform in use today is based on the new A85 chipset, and we'll go into more details on that next.