Articles: Mainboards

Bookmark and Share

Pages: [ 1 | 2 | 3 | 4 | 5 | 6 | 7 ]

Testbed Configuration

We have already tested the performance of AMD E-350 based systems several times before and compared it against direct competitors as well as against fully-functional desktop CPUs. Therefore this time we will only check out a small number of test apps where Brazos-based Sapphire mainboards will compete against similar Mini-ITX solutions from Gigabyte built on Intel Atom D525 and AMD E-350.

As a result, the complete list of hardware and software components for our testbeds included the following:

  • Mainboards:
    • Gigabyte GA-D525-TUD (Intel Atom D525 + Intel NM10 Express);
    • Gigabyte GA-E350N-USB3 (AMD E-350 + AMD Hudson M1);
    • Sapphire PURE WHITE FUSION E350 (AMD E-350 + AMD Hudson M1);
    • Sapphire PURE FUSION MINI E350 (AMD E-350 + AMD Hudson M1).
  • Memory:
    • 2 x 2 GB DDR3-1333 SDRAM DIMM (Kingston KHX1600C8D3K2/4GX) 9-9-9-27;
    • 2 x 2 GB DDR3-1333 SDRAM SODIMM (Apogee AS2G733-13G) 9-9-9-27;
  • Hard drive: OCZ Vertex2 (OCZSSD2-2VTXE120G).
  • Power supply unit: Tagan TG880-U33II (880 W).
  • Operating system: Microsoft Windows 7 SP1 Ultimate x64.
  • Drivers:
    • ATI Catalyst 11.5 Display Driver;
    • Intel Chipset Driver;
    • Intel GMA 3150 Driver


To estimate the average platform performance PCMark 7 measures the speed of typical real-life algorithms that are very popular in every-day tasks. In this case we are particularly interested in two scores: PCMark test that characterizes performance under typical desktop load, and Lightweight test estimating the performance of relatively weak systems, which normally do not run resource-hungry and multi-threaded applications.


The performance of Brazos-mainboards from Sapphire is a little lower than that of a “reference” Gigabyte platform. The problem lies with relatively weak timings that Sapphire mainboards set by default, which cannot be corrected, because the BIOS doesn’t have the corresponding options for that.

We used the Entry profile in 3DMark Vantage suite to test the graphics sub-system performance.

Here we have no problems with Sapphire boards. Their performance is practically the same as that of the competing solution. As for the Atom based board, it didn’t pass this test, because GMA 3150 integrated graphics core in the processor doesn’t support DirectX 10.

The computational SuperPi test is a great way of checking the performance in single-threaded mode:

As we see, AMD E-350 in single-threaded mode is 60-70% faster than Atom D525. So, the only way energy-efficient Intel processor may reach parity with the AMD solution is under multi-threaded load, when the advantages of Intel Hyper-Threading technology will truly shine.

A great example of a test like that is the Fritz chess benchmark.

Here Atom D525 even slightly outperforms AMD E-350 based mainboards, which also perform very closely to one another.

To test the performance during data archiving we took WinRAR archiving utility. Using maximum compression rate we archived a folder with multiple files 550 MB in total size.

We use the Apple iTunes utility for our audio transcoding test. We convert an audio CD into AAC format. Take note that Apple iTunes can only use two CPU cores.

We used Peacekeeper test to check the performance in Internet Explorer 9.

Brazos platform shows its potential real well in applications, leaving the Pine Trail platform behind in most cases. But our review is about specific Sapphire products and therefore we must say that like any other contemporary platform, Brazos implementations by different manufacturers still perform practically identically.

In conclusion to our Performance section we would like to offer you the HD video transcoding test performed using Cyberlink MediaEspresso 6.5.1704 37777 that allows using the graphics accelerator as well. We measured the time it took to transcode a small H.264 1080p video clip into an iPhone 4 friendly format (H.264, 1280x720, 4 Mbps).

Again both Sapphire products performed impeccably.

Pages: [ 1 | 2 | 3 | 4 | 5 | 6 | 7 ]


Comments currently: 12
Discussion started: 06/20/11 08:01:01 AM
Latest comment: 03/08/16 03:55:19 AM

View comments

Add your Comment