On Fri, Feb 19, 2010 at 12:09 AM, Sa?l Ibarra Corretg? <saul at ag-projects.com> wrote: > Hi Benny, > >> You're right, it's violating the spec. I just fixed this in r3101 >> (basically swapped the SRFLX and PRFLX priority values). >> > > Ok, I can see the fix, but could you please explain me why do we want to > swap the priorities when using STUN and not keep the host candidate as the > highest priority? I'm a bit confused about it :) > > It's so that SRFLX candidate will be checked first, otherwise it will be queued behind all those host candidates, potentially would delay it by few tens/hundreds of msecs Cheers Benny
