Their methodology is rubbish. The fact that they can't possibly be blind invalidates the approach to begin with. But they also didn't remain within one class o

Their methodology is rubbish. The fact that they can't possibly be blind invalidates the approach to begin with. But they also didn't remain within one class of device when the physical coupling method does a lot to affect the result. They also didn't do any work whatsoever to research models which are often recommended as being good. If Engadget didn't have such high journalistic integrity, I'd suspect it of being a sideways ad for Bose and Monster.