Yes, but where will the devs hardfork if asic is capable to change N as well ? Don't mean to spread fud, I'm all in on vtc, but looks like things are moving a bit too fast ...
Just read a little bit about scrypt functions, e.g.
http://en.wikipedia.org/wiki/Scrypt and the pages referenced there. You'll see that these algorithms have lots of parameters. Just change one/some of them and you're rendering any perhaps-at-some-time-in-Q3-existing ASIC useless.
So, basically, nothing to see here. Just move on and ignore the FUD and panic kids.