Socket 939 Roundup: Battle at the Top

Performance Test Configuration

If you are interested in more information comparing the Athlon 64, Athlon 64 FX, Prescott, P4, and P4EE, please see our in-depth comparisons in the recent 775/939/Prescott/3400+/3000+ launch reviews:

Intel's 925X & LGA-775: Are Prescott 3.6 and PCI Express Graphics any Faster?
Intel 925X/915: Chipset Performance & DDR2
Socket 939 Chipsets: Motherboard Performance & PCI/AGP Locks
AMD Athlon 64 3800+ and FX-53: The First 939 CPUs
The Athlon 64 FX-53: AMD's Next Enthusiast Part
Intel's Pentium 4 E: Prescott Arrives with Luggage
Athlon64 3400+: Part 2
AMD's Athlon 64 3400+: Death of the FX-51
Athlon64 3000+: 64-bit at Half the Price

Performance Test Configuration
Processor(s):	AMD Athlon64 FX53 (2.4GHz) Socket 939
RAM:	2 x 512MB Mushkin PC3500 Level II OR 2 x 512MB OCZ PC3500 Platinum Ltd
Hard Drive(s):	Seagate 120GB 7200 RPM IDE (8MB Buffer)
Video AGP & IDE Bus Master Drivers:	NVIDIA nForce Platform Driver 4.24 (5-10-2004) VIA 4in1 Hyperion 4.51 (12-02-2003)
Video Card(s):	nVidia 6800 Ultra (AGP 8X)
Video Drivers:	nVidia nForce 61.80 Beta
Operating System(s):	Windows XP Professional SP1
Motherboards:	Abit AV8 PRO (VIA K8T800 PRO) Asus A8V Deluxe Revision 2 (VIA K8T800 PRO) ECS KV2 Extreme (VIA K8T800 PRO) Gigabyte K8NSNXP-939 (nVidia nForce3 Ultra) MSI K8N Neo2 Platinum (nVidia nForce3 Ultra) MSI K8T Neo2-FIR (VIA K8T800 PRO)

Tests used either Mushkin PC3500 Level II or OCZ PC3500 Platinum Ltd memory modules. Both memories use Winbond BH5 chips and perform virtually the same in our benchmarks. The BH5 memory modules that we used for testing are no longer available for purchase, but we have not yet established our standard memory for future testing. We will be using a new standard memory in future motherboard tests.

Since the nVidia 6800 Ultra PCI Express card was used for testing in the 925X/915 chipset launch, we used the AGP 8X version of the 6800 Ultra for all testing in this roundup. This allows for better comparison with Intel Socket 775 results, since our standard ATI 9800 PRO is not available in a PCIe configuration. We are still evaluating video cards to determine our new Video standard for future benchmarking. All performance tests were run with AGP Aperture set to 128MB with Fast Write enabled. Resolution in all benchmarks is 1024x768x32 unless otherwise noted.

Additions to Performance Tests

AutoGK (Auto Gordian Knot) has been selected as the new standard for Encoding benchmarking. This benchmark is partially based on the DVD2AVI engine and we use DivX 5.1.1 as the encoding codec. Test results are compiled with a 2-pass encoding using "Sum of All Fears", Chapter 9 as the video source.

Game Benchmarks

We have added four new Gaming tests to our standard suite of gaming benchmarks. Far Cry is a popular DirectX 9.0b game, which is run with a custom benchmark called Airstrip. We have also added Unreal Tournament 2004 with a custom benchmark called UTBench. Return to Castle Wolfenstein - Enemy Territory is a recent OpenGL game based on the Quake engine, and we run the Radar benchmark. We are finding RCW-ET to be an excellent benchmark for measuring system performance, and it will likely replace Quake3 as our standard OpenGL benchmark down the road. We have also included the Final Fantasy XI benchmark, which is DX9. With the addition of the 4 new games, we have dropped Splinter Cell from our motherboard game benchmarks.

Other standard game benchmarks include Halo, Microsoft's Direct X 9.0b game; X2 Benchmark, a DX 8.1 game, which includes Transform and Lighting effects; the DX9 Aquamark 3; and Unreal Tournament 2003, which is included for comparison with the new UT2004 test results. The DX 8.1 Comanche 4 benchmark is also used with the 4X anti-aliasing setting a 1280x1024 resolution to differentiate system performance better using the nVidia 6800 Ultra.