Testbed and Methods
Our major task was to compare the performance of systems built with dual-core processors of different architectures. Therefore, we assembled three test platforms around the top-of-the-line Athlon 64 X2, Core Duo and Pentium D CPUs. All these processors belong to the same price group and cost $600.
The testbeds we used were built using the following hardware components:
- AMD Athlon 64 X2 4800+ (Socket 939, 2.4GHz, 2x1MB L2);
- Intel Core Duo T2600 (Socket 479, 2.16GHz, 2MB L2);
- Intel Pentium D 960 (LGA775, 3.6GHz, 2x2MB L2).
- ASUS P5WD2-E Premium (LGA775, Intel 975X Express);
- ASUS N4L-VM DH (Socket 479, Intel 945GM);
- DFI LANParty UT NF4 SLI-DR Expert (Socket 939, NVIDIA nForce4 SLI).
- 2048MB DDR400 SDRAM (Corsair CMX1024-3500LLPRO, 2 x 1024 MB, 2-3-2-10);
- 2048MB DDR2-667 SDRAM (Corsair CM2X1024-6400PRO, 2 x 1024 MB, 4-4-4-12).
- Graphics card: PowerColor X1900 XTX 512MB (PCI-E x16).
- Storage subsystem: Maxtor MaXLine III 250GB HDD (SATA150).
- Operating system: Microsoft Windows XP SP2 with DirectX 9.0c.
The tests were conducted with the mainboards BIOS set to the maximum performance.
Here I would like to add two things. Firstly, when we tested the dual-core processor built on the NetBurst architecture, we used the Pentium D 960 working at 3.6GHz that hasn’t been officially announced yet at the time of testing. We selected this particular CPU, because it was due to be released any day and the previous model, the Pentium D 950, dropped down in price and was selling for $350-$400, so it couldn’t be regarded as a fair competitor to Core Duo T2600.
Secondly, we decided that we wouldn’t use any Nvidia based graphics cards in our test platforms, because ForceWare drivers still have quite a few unsolved issues in systems with dual-core processors. Unfortunately, the results from using ForceWare drivers in these systems may be quite unpredictable. Therefore, from now on we equip our testbeds with ATI Radeon X1900 XTX graphics card from PowerColor.