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

RE: [EFM] 100 Mbps Proposals

thanks for correcting me the two cases. I was actually thinking the two proposals earlier at EFM. Specifically there were some proposals at EFM regarding 100 meg, one was on single pair (TDD-combined up/down) and one on 4-pairs of cat-5/3.
What I am suggesting is regarding any telephone copper pair and spectrally compatible with services in the bundle which is different from what was discussed on those proposals.
I guess I was hoping to hear  "is this something of interest to ieee?" since we brought up the 100 meg fiber discussion.
that is all
-----Original Message-----
From: Geoff Thompson [mailto:gthompso@xxxxxxxxxxxxxxxxxx]
Sent: Thursday, February 14, 2002 5:06 PM
To: Behrooz Rezvani
Cc: Geoff Thompson; O'Mahony, Barry; 'larry rennie'; Behrooz Rezvani; Bruce Tolley;; Roy Bynum
Subject: RE: [EFM] 100 Mbps Proposals


At 01:39 PM 2/14/02 -0800, Behrooz Rezvani wrote:
Sorry just got of a meeting did not realize that I have created some

You are right Geoff. 100BASE-T4 was 100 Meg combined on a single pair.

No, it was 100 Mb/s on 3 pair. There was one pair dedicated to each direction and 2 pair that were used in the direction of data transmission. It only takes one pair to easily handle the carrier sense information. So the CSMA/CD was done on 1 pair each direction. The data transmission was done on 3 pairs with 8B/6T encoding. See Figs 23-3, 23-4, 23-23. That is why you can't do full duplex on 100BASE-T4.

So this is not that. I am suggesting just like 100BASE-T put 25Meg full duplex on each pair

100BASE-T does not put 25Meg full duplex on each pair.
I assume that you mean that 1000BASE-T puts 25Meg full duplex on each pair. That is not true either. There is no pair specific coding in 1000BASE-T. The coding is spread over all 4 pair at once. The decoder looks at 5 levels on 4 pairs (which yields a code space of 625 points) and derives 1 byte from it (256 code points plus control codes). The excess code space is used to provide more than one possible code target for the subsequent code. Choosing the one that provides the greatest voltage swing generates extra noise immunity.


you are correct Spectrum compatibility should be dealt with I think it can
be dealt with on 400-500 meter reaches

If the team thinks that there is interest for 100 Meg I'll be happy to send
a short presentation off line to people who are interested


-----Original Message-----
From: Geoff Thompson [mailto:gthompso@xxxxxxxxxxxxxxxxxx]
Sent: Thursday, February 14, 2002 1:09 PM
To: O'Mahony, Barry
Cc: 'larry rennie'; Behrooz Rezvani; Bruce Tolley;; Roy Bynum
Subject: RE: [EFM] 100 Mbps Proposals

In addition, 100BASE-T4 does not support full duplex


At 12:04 PM 2/14/02 -0800, O'Mahony, Barry wrote:

>As was brought up in Raleigh, it has not been demonstrated that 100BASE-T4
>meets the spectrum compatibility objective (and it is unlikely that it
>does).  Leaving aside the discussion as to how close installed POTS wiring
>is to CAT3.
>There may be MxU instances where spectrum compatibility is not required,
>instances where it would be.  It is an adopted EFM objective, however.
>Barry O'Mahony
>Intel Labs
>Hillsboro, OR, USA
>tel: +1 (503) 264-8579
>-----Original Message-----
>From: larry rennie [mailto:Larry.Rennie@xxxxxxx]
>Sent: Thursday, February 14, 2002 11:28 AM
>To: Behrooz Rezvani
>Cc: Bruce Tolley;; Roy Bynum
>Subject: Re: [EFM] 100 Mbps Proposals
>Is not 100BASE-T4 100Mbits/sec over 4, Cu pairs?
>Behrooz Rezvani wrote:
> > Bruce,
> >
> > if there is a success in starting such an effort, I would very much to
> > encourage you and other people to consider 100 mbps over 4 copper pairs
> > reach <xyz> meters. That has a lot more practical applications in MxU.
> >
> > Thanks
> > Behrooz
> > ----- Original Message -----
> > From: "Roy Bynum" <rabynum@xxxxxxxxxxxxxx>
> > To: "Bruce Tolley" <>; <>
> > Sent: Wednesday, February 13, 2002 7:59 PM
> > Subject: Re: [EFM] 100 Mbps Proposals
> >
> > >
> > > Bruce,
> > >
> > > I am concerned about putting effort into developing a standard for
> > > technology that already exists for a market that is current, not
> > in
> > > the future.
> > >
> > > Thank you,
> > > Roy Bynum
> > >
> > > At 07:48 PM 2/13/2002 -0800, Bruce Tolley wrote:
> > >
> > > >Colleagues:
> > > >
> > > >Those of you who are also on the 802.3 reflector saw that there is
> > > >for interest on the agenda of the March meeting in St Louis to
> > > >starting a 100 Mbps dual fiber SM fiber project outside of 802.3ah
> > force.
> > > >
> > > >While I have not yet decided where I stand on 100 Mbps solutions for
> > > >I wanted to communicate that I think this call for interest is
> > > >I would strongly encourage the proponents of 100 Mbps on SM fiber to
> > > >converge on one strong proposal for the March IEEE 802.3ah meeting.
> > > >
> > > >We are already facing the challenge of perhaps too many EFM PHYs.  To
> > > >specify an additional PHY for EFM outside of the 802.3ah TF only
> > > >life more difficult.
> > > >
> > > >Bruce
> > >