Thread Links Date Links
Thread Prev Thread Next Thread Index Date Prev Date Next Date Index

Re: [EFM] FEC needs a good discussion home!


I have long supported the work on FEC, and I continue to encourage the
pursuit of this topic.  It will stay within the optics subtaskforce
for two reasons: 

1) The FEC that you are discussing is slated for use with the optical PMD,
though the implementation of the FEC mechanism is a logic problem.  
For now, we are considering the need for and the efficacy of FEC, and 
this is best done amongst the body of experts who understand the 
behavior of the optical links.

2) The FEC that you are discussing is not a global solution for EFM.
The Copperheads have their own FEC mechanisms, and I don't want to
create another overarching group that will reach its fingers into both
optics and copper.  OAM has a hard enough time coordinating between
the different groups as it is.  We don't need to create another
cross-PMD topic.

What we really need to do is to crank up the level of effort on FEC,
and I mean this as a word of encouragement, not criticism. The interest
is there, and the support is there, but time is running out.  We
have employed FEC before in 802.3ab, and we will probably employ it
again in the Copper PHY for 802.3ah.  FEC is used on optical links
in commercial applications today, at rates much faster than those we
are specifying in EFM.  It would be a real shame if EFM dropped the ball
on FEC, but we need to be convinced of 1) the need, 2) the cost and
3) the efficacy.

I think that the fans of FEC should prepare a motion for the
consideration of the Task Force at the Edinburgh meeting. It's
time to make up our minds on this subject.