Testing methodology
Comparison Processor Configurations |
|||||
|---|---|---|---|---|---|
| CPU | AMD |
Intel |
|||
A8-7600 |
A10-6800K |
A8-6500T |
Core i3-4330 |
||
| Chip TDP | 65W/45W |
100W |
65W |
54W |
|
| Current price | £90 |
£105 |
£80 |
£95 |
|
| Motherboard | ASRock FM2A88X-ITX+ |
Gigabyte Z87-D3HP |
|||
| BIOS | 1.9.0 |
F5 |
|||
| Chipset Driver | AMD Catalyst 13.12 |
Intel Inf 9.4.0.1027 and IMEI 9 |
|||
| DDR3 Memory | AMD Gamer Series 16GB (2x8GB) |
||||
| Memory Timings | 10-11-11-28-2T @ 2,133MHz |
9-10-9-27-2T @ 1,600MHz |
9-10-9-27-2T @ 1,866MHz |
||
| Integrated Graphics | Radeon R7 |
HD 8670D |
HD 8550D |
HD 4600 |
|
| IGP driver | AMD Catalyst 13.30 RC3 |
Intel 15.33.8.64.3379 |
|||
| Disk Drive | Samsung 840 Pro 250GB |
||||
| Chassis | Xigamtek Nebula |
Corsair Graphite 600T |
|||
| Power Supply | Antec High Current Pro 750W |
Corsair AX760i |
|||
| Operating System | Windows 8.1 Pro 64-bit |
||||
CPU and Memory Benchmarks |
|||||
| HEXUS PiFast |
Our number-crunching benchmark stresses a single core by calculating Pi to 10m places | ||||
| CineBench R15 | Using Cinebench's multi-CPU render, this cross-platform benchmark stresses all cores | ||||
| wPrime 2.1.0 | Another number-crunching benchmark that stresses all available CPU cores/threads | ||||
| AIDA64 v4.00.2746 |
Benchmark that analyses memory bandwidth and latency | ||||
Multimedia Benchmarks |
|||||
| LuxMark 2.0 | An OpenCL rendering benchmark | ||||
| MuseMage 1.9.6 | An OpenCL image-manipulation benchmark (64-bit) | ||||
| HandBrake 0.9.9.1 | Free-to-use video encoder that stresses all CPU cores (64-bit) | ||||
System Benchmarks |
|||||
| PCMark 8 v2.0 | System-wide examination that uses the Home preset, run with OpenCL acceleration | ||||
| 3DMark | DX11, run at the Firestrike default test | ||||
| SiSoft Sandra 2014 | Aggregate score that takes a composite of 12 system-wide benchmarks | ||||
Gaming Benchmarks |
|||||
| BioShock Infinite | DX9, 1,280x720 and 1,920x1,080 medium quality | ||||
| GRID 2 | DX9, 1,280x720 and 1,920x1,080 high quality | ||||
| Total War: Rome II | DX9, 1,280x720 and 1,920x1,080 medium quality | ||||
Miscellaneous Benchmarks |
|||||
| Power Consumption | While idling and when running wPrime and GRID 2 | ||||
Notes
We're benchmarking the A8-7600 in both 65W and 45W modes, to see how much performance is sacrificed as the TDP is lowered. Common sense dictates that while CPU-centric performance will drop, graphics oomph is likely to be less restricted.
Tested from the grounds-up, we're also comparing the third-rung Kaveri chip to AMD's A10-6800K and A8-6500T, representing the very best and 45W models from the previous Richland generation. Any instance where the Kaveri A8-7600 is faster/better than the older A10-6800K shows genuine improvements from one APU architecture to another.
We can also compare like-for-like with reference to TDPs - the A8-7600 and A8-6500T share a common 45W rating - and, looking across to Intel, the Core i3-4330 is the obvious comparison. AMD's been cute in sending us the A8-6500T - the non-T version, rated at 65W, is clocked in much higher on the CPU and is over 10 per cent faster on the GPU side of things.
