Tavis,

Check out http://www.brynosaurus.com/pub/net/p2pnat/, it's not quite what you're asking for but it does include a table of data gathered from volunteers, detailing the percentage of NATs that support udp/tcp hole-punching and hairpin (loopback) translation. I'm concerned about the latter, it appears that not even ICE can resolve that issue.

Mike

Date: Fri, 18 Nov 2005 15:36:35 -0800 From: Tavis P tavis.lists@galaxytelecom.net Subject: [Users] How Effective is STUN? To: users@openser.org, serusers@iptel.org Message-ID: 437E6583.5050905@galaxytelecom.net Content-Type: text/plain; charset=ISO-8859-1

I'm trying to find some statistics as to what the ratio of Cone vs Symmetric NAT solutions deployed in the world are, has anyone done some research into this?

I'm curious what percentage of users in certain demographics (broadband clients, for example) i can expect to be serviced using STUN alone, so i can come up with some figure to help me build out my network

Even just some anecdotal information of peoples experiences would be very useful

Tavis