A few of these reasons boil down to "it's faster to not NAT". That makes some intuitive sense, but does anyone know of any studies/tests so we can get numbers? Are we talking higher time to first connect? Slight increase in hops/latency on every packet?
The "faster than NAT" argument made very little sense to me. A NAT adds like a few microseconds of latency, nothing to be perceptible for virtually all applications.
Of course these are generalised stats, there will be some users with tunnels which will impair IPv6 performance, while other users might be behind multiple layers of NAT.
