When we see one-second performance differences in timed applications that should yield identical results, we generally chalk that up to rounding. After all, 49.49 seconds rounds down, while 49.50 seconds rounds up.
We expect that an indiscernible frequency difference is responsible for the top results in Apple iTunes, where core count doesn’t matter.
Lame also produces substantially similar results from dramatically different hardware with similar CPU frequencies, indicating that it’s not optimized for multi-threading. The fact that Intel's desktop Sandy Bridge- and enthusiast Sandy Bridge-E-based chips share a common core architecture help explain the photo-finish.
HandBrake and MainConcept take excellent advantage of the new system’s extra cores.
- Au Contraire: More CPU, Less GPU
- CPU, CPU Cooler, And Memory
- Motherboard, Graphics, And Power
- SSD, Hard Drive, And Optical Drive
- Case And Case Cooling
- Hardware Installation
- Overclocking A Truly Overclockable Platform
- Test Settings And Benchmarks
- Benchmark Results: 3DMark And PCMark
- Benchmark Results: SiSoftware Sandra
- Benchmark Results: Battlefield 3 And DiRT 3
- Benchmark Results: Skyrim And StarCraft II
- Benchmark Results: Audio And Video Encoding
- Benchmark Results: Productivity
- Power, Heat, And Efficiency
- Does The CPU Gamble Pay Off?