On Tue, Mar 1, 2011 at 06:58, Karl Auer <[email protected]> wrote: > > The only plausible theory I have is that after 8 bits of "ff" and 16 > bits of ethernet multicast prefix (0x3333) are factored in, there is > only room at layer 2 for 24 more bits, so there is no point having more > bits in layer 3. This is not a satisfying theory because it appears to > tie layer 3 multicast to a specific layer 2 technology (Ethernet, with > its 48-bit MAC addresses). >
"This is not a satisfying theory because it appears to tie layer 3 multicast to a specific layer 2 technology (Ethernet, with its 48-bit MAC addresses)." Factoring reality in to any decision typically yields better results, yes? More importantly - what problem are you really trying to solve? Do you expect to have lots of collisions in this 24 bit space on a given link? /TJ
-------------------------------------------------------------------- IETF IPv6 working group mailing list [email protected] Administrative Requests: https://www.ietf.org/mailman/listinfo/ipv6 --------------------------------------------------------------------
