Test Systems And Benchmark Setup
We'll test all of these systems with a single Radeon HD 4890 graphics card. While this seems a little overkill for a $100 CPU, consider that the Radeon HD 4870 512MB is now as low as $130 online and would make a fine companion to any of these CPUs. The Radeon HD 4890 is merely an overclocked Radeon HD 4870, so the performance numbers we see here today aren't out of the realm of possibility for a budget machine.
In sync with the budget theme is our choice of motherboards. Our upcoming 785G-based board roundup has shown us that the difference between DDR2 and DDR3 memory on the 785G chipset is absolutely insignificant. Therefore, we'll use an AM2+ board with the new 785G chipset to represent the AMD corner because we can use the same DDR2 RAM on the Intel chipset to keep things more fair.
The Intel board sports a G41 chipset. We realize that neither of these platforms is the best-of-the-best. But we're evaluating $100 CPUs here, so we're keeping things realistic for a low-cost build. Budget-friendly boards based on either of these chipsets can be had for as low as $45 online.
The motherboard we're using for the AMD CPUs is the Biostar TA785GE 128M, which you'll see prove itself capable in the aforementioned upcoming 785G story.
In the Intel corner, we'll use Asus' P5QPL-VM EPU based on the Intel G41 chipset, and also featuring support for 1,066 MT/s dual-channel DDR2.
We'll test at two resolutions: 1280x1024 to demonstrate the CPU bottleneck inherent on smaller screens and 1920x1200, to show how these CPUs can handle higher-resolution gaming. We won't use anti-aliasing (AA) or anisotropic filtering (AA) in an effort to keep the focus on CPU, and not graphics card, performance.
One final note before we get started: a lack of available Phenom II X3 710 processors at the time of writing forced us to simulate the performance of this CPU by underclocking a Phenom II X4 945 and disabling one of its four cores. We've proven the validity of this test method in our How Many CPU Cores Do You Need? article.
Header Cell - Column 0 | AMD Test System | Intel Test System |
---|---|---|
CPU | Athlon II X4 640 (2.6 GHz, No Shared Cache)Phenom II X2 550 BE (3.1 GHz, 6MB L3 Cache)Phenom II X3 710* (2.6 GHz, 6MB L3 Cache)*Phenom II X3 710 simulated with a Phenom II 945by lowering the multiplier and disabling the fourth CPU core | Dual-Core Intel Pentium E6500 (2.93 GHz, 2MB L2 Cache) |
Motherboard | Biostar TA785GE 128M AMD 785G, BIOS 080015 | ASUS P5QPL-VM EPU Intel G41, BIOS 0407 |
Networking | Onboard Realtek Gigabit LAN controller | Onboard Realtek Gigabit LAN controller |
Memory | A-Data Extreme DDR2 800+ 2 x 2,048MB, DDR2-800, CL 5-5-5-18-24 2T at 1.8 V | |
Graphics | ATI Radeon HD 4890 Reference900 MHz Core, 1GB GDDR5 @ 993 MHz | |
Hard Drive | Western Digital Caviar Black 640GB 7,200 RPM, 32MB Cache SATA 3.0 Gb/s | |
Power | ePower EP-1200P10-T21,200 W, ATX 12V, EPS 12v | |
Software and Drivers | ||
Operating System | Microsoft Windows Vista Ultimate 64-bit 6.0.6001, SP1 | |
DirectX version | DirectX 10 | |
Graphics Drivers | ATI Catalyst 9.8 |
Benchmark Configuration | |
---|---|
3D Games | |
Crysis | Patch 1.2.1, DirectX 10, 64-bit executable, benchmark tool, High Quality, No AA |
Far Cry 2 | Patch 1.02, DirectX 10, in-game benchmark, Ultra High Quality, No AA |
World In Conflict | Patch 1009, DirectX 10, timedemo, Very High Details, No AA/No AF |
Tom Clancy's H.A.W.X. | Patch 1.02, DirectX 10.1, in-game benchmark, Highest Settings, No AA |
Left 4 Dead | Version 1.0.1.4., Custom THG Benchmark, Highest Settings, no AA, no AF |
Grand Theft Auto IV | Patch 1040, in-game Benchmark, Medium Settings, High Textures |
Resident Evil 5 | Resident Evil 5 Benchmark Version, Highest Settings, Motion Blur Enabled, no AA, no AF |
Fallout 3 | Patch 1.4.0.6., Custom THG Benchmark, Ultra High Quality, No AA, No AF |
Prototype | Version 1.0.0.1., Custom THG Benchmark, High Quality, No AA, No AF |