There are ways around Amdahl's Law. The main cause of it, AFAIR, is the memory architecture of traditional computer systems (the bottleneck is the memory bus of designs like the Von Neumann architecture).
Amdahl's law doesn't have anything to do with memory bandwidth. It's very simple, and trust me, there is no way around it. Some algorithms, or parts of algorithms, cannot be parallelized; they are inherently serial. CABAC in H.264 video compression is a good example. Your overall software performance will be bounded by the execution time of the most complex piece of the algorithm which cannot be parallized. Once that piece of your algorithm is taking up 100% of an execution core, your software cannot go any faster, regardless of how many more CPUs you throw at it.
I read a paper a few days ago claiming that it's better to have a few beefy cores than a bunch of wimpy cores. This is exactly what I've also seen in my experience with high-performance computing, and the reason why serious people run HPC compute loads on badass x86 chips and not ARM or Atom.
