Testing methodology
CPU comparisons |
|||||||||
---|---|---|---|---|---|---|---|---|---|
CPU | Approx. pricing |
Default/Turbo frequency | Cores threads | Motherboard/BIOS | Chipset | Chipset driver | IGP | ||
AMD FX-8350 | £150 | 4.0GHz/4.2GHz | 8/8 | ASUS Crosshair V Formula (1605) | 990FX | Cat 12.8 SB | NA | ||
AMD FX-8150 | £135 | 3.6GHz/4.2GHz | 8/8 | ASUS Crosshair V Formula | 990FX | Cat 12.8 SB | NA | ||
AMD Phenom II X6 1100T | NLA | 3.3GHz/3.7GHz | 6/6 | ASUS Crosshair V Formula | 990FX | Cat 12.8 SB | NA | ||
AMD A10-5800K | £90 | 3.8GHz/4.2GHz | 4/4 | Gigabyte F2-A85X-UP4 (F3G) | A85 | Cat 12.8 SB | HD 7660D | ||
AMD A10-5700 | £90 | 3.4GHz/4.0GHz | 4/4 | Gigabyte F2-A85X-UP4 | A85 | Cat 12.8 SB | HD 7660D | ||
AMD A8-5600K | £75 | 3.6GHz/4.0GHz | 4/4 | Gigabyte F2-A85X-UP4 | A85 | Cat 12.8 SB | HD 7560D | ||
Intel Core i7-3770K | £250 | 3.5GHz/3.9GHz | 4/8 | Gigabyte Z77X-UD5H (F3C) | Z77 | Intel Inf 9.3.0.1021 | HD 4000 | ||
Intel Core i5-3570K | £170 | 3.4GHz/3.8GHz | 4/4 | Gigabyte Z77X-UD5H | Z77 | Intel Inf 9.3.0.1021 | HD 4000 | ||
Intel Core i3-3225 | £110 | 3.3GHz/NA | 2/4 | Gigabyte Z77X-UD5H | Z77 | Intel Inf 9.3.0.1021 | HD 4000 | ||
Intel Core i5-2500K | NLA | 3.3GHz/3.7GHz | 4/4 | Gigabyte Z77X-UD5H | Z77 | Intel Inf 9.3.0.1021 | HD 3000 | ||
*NLA - no longer available |
Common components |
||||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|
Discrete GPU | NVIDIA GeForce GTX 680 (306.97) | |||||||||||
Memory | Patriot IEM 8GB (2x4GB) | |||||||||||
Memory speed | 9-9-9-24-1T @ 1,600MHz | |||||||||||
Power Supply | Corsair AX750 | |||||||||||
Monitor | BenQ EW2430 | |||||||||||
Disk drive(s) | Samsung Series 830 256GB | |||||||||||
Chassis | Corsair Graphite 600T | |||||||||||
Operating system | Windows 7 Ultimate, SP1, 64-bit |
Benchmarks |
||||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|
CPU | ||||||||||||
AIDA64 | Useful application to test memory bandwidth and latency | |||||||||||
HEXUS.PiFast | Using CPU power to crunch the constant to 10m places - light load | |||||||||||
CINEBENCH R11.5 | Multi-CPU render - heavy load | |||||||||||
TrueCrypt 7.0a | Encryption benchmark - heavy load | |||||||||||
X264 HD 5.01 | Two-pass media encoding - heavy load | |||||||||||
7-zip 9.20 | Multi-threaded compression tool - heavy load | |||||||||||
GPU - GTX 680 | ||||||||||||
3DMark 11 | Run at the performance preset | |||||||||||
Batman: Arkham City | 1,920x1,080, DX11, FXAA High, Extreme Quality | |||||||||||
DiRT: Showdown | 1,920x1,080, DX11, 4x MSAA, High Quality | |||||||||||
Aliens vs. Predator | 1,920x1,080, DX11, 4x MSAA, Very High Quality | |||||||||||
GPU - IGP | ||||||||||||
3DMark 11 | Run at the entry preset | |||||||||||
3DMark Vantage | Run at the performance preset | |||||||||||
Batman: Arkham City | 1,280x720, 2x MSAA, High-quality settings (DX9) and Medium-quality settings (DX11) | |||||||||||
DiRT: Showdown | 1,280x720, 4x MSAA, Medium settings, DX9 | |||||||||||
LuxMark | Uses LuxRender and OpenCL for GPGPU performance evaluation | |||||||||||
Power-draw | Run at idle, 2D load (X264 HD) and 3D load (Batman: Arkham City) |
Notes
This review provides an opportunity to bring together every chip we've recently tested for the AMD Trinity APU and AMD Piledriver CPU launches. From AMD's side we've got three APUs and three CPU-only processors that need a discrete graphics card in the motherboard. From Intel's side the quartet of chips all have integrated graphics.
We're going to start with a slew of CPU-intensive tests, follow that up with an examination of high-resolution gaming performance with a GeForce GTX 680 card in situ, and then remove select CPUs and investigate IGP performance. Power-draw for both discrete and IGP systems is noted. Don't worry if this doesn't make a whole heap of sense right here; it should be self-explanatory as we go through the graphs.