Also, keep in mind when you're looking at WinISD or Unibox port velocity graphy, you're usually looking at a worst case scenario. Most of us use the max sustained amp power as the input power for the sub. This is great for making sure you won't destroy the driver if someone cranks the master volume to '11'. However, in the real world, you're unlikely to drive your sub with that much steady state power. Drop the power by half, -3dB in SPL, and you'll see air velocity drop also.
I once read a white paper published by Peerless, the driver manufacturer, that recommended 34 m/s as a max....I think that's 0.1 Mach. And at one time, that was the recommendation included with WinISD Pro. Later versions have dropped the recommendation to 17 m/s, which seems to coincide with some of the research that
Collo did. But note, that even his research suggested there are variables that can't be completely accounted for in actual usage, such as masking effect of real content vs. sine waves and port orientation (toward listener vs. firing into a carpet pad).
I like to point to a real world example. The SVS PB10 has a high excursion 10" woofer and only a 3" dual flared port. If you model something similar, say the Dayton Titanic 10", you'll see some really high port velocity numbers with "max" power. However, none of the reviews, paid or owner testimonials, I've read have noted port noise as a problem.
For me personally, I tend to stick with Peerless' 34 m/s number for now, but I generally never push my subs near max excursion. I've got a small ported sub planned with a RS315-HO that has a worst case scenario that gets into the low 40s near 20hz...I don't expect a problem with my real world volume requirements for that sub, but it'll be an interesting experiment to see how easily I can induce noise.
-Brent