I would extend this to say that when you get down to the level of synapses you're now talking about memory and not CPU. (If you actually want to implement this)

I would extend this to say that when you get down to the level of synapses you're now talking about memory and not CPU. (If you actually want to implement this) 100 trillion synapses is better mapped to 100TB of RAM. It's also very important to remember how incredibly slow the human brain is. Were talking 100s of ms slow to go from sensory input to motor output. A single neuron might take between 1 and 10 ms to fire, and a single dendrite might take 1/10th that time, so at best your doing computation at 10khz. CPU has 5 orders of magnitude over biology. The problem is we have only a vague idea of how the network is connected and don't really know the algorithm thats being implemented by that network. So we fall back on things like simulating ion channels which take way more compute resources than necessary. There is a lot of cargo culting going on right now, but of course it's also insanely exciting and fun to find out what does and doesn't work.