4 ms·
That depends heavily on the workload. For optimized floating-point code, Haswell is a beast (16 double precision flops per core per cycle, vs 8 for Ivybridge).
by stephencanon 12y ago
That depends heavily on the workload. For optimized floating-point code, Haswell is a beast (16 double precision flops per core per cycle, vs 8 for Ivybridge).
- judk 12y agoFor 95% of consumers, all Apple CPUs are phenomenally overpowered.