Subject: Re: Netgear gigabit GA621 fibre interface problems
To: David Laight <david@l8s.co.uk>
From: David Maxwell <david@vex.net>
List: port-i386
Date: 12/11/2002 14:01:38
On Wed, Dec 11, 2002 at 12:35:24AM +0000, David Laight wrote:
> On Wed, Dec 11, 2002 at 11:10:06AM +1100, Daniel Carosone wrote:
> > On Tue, Dec 10, 2002 at 10:07:54AM -0800, Michael Graff wrote:
> > > If you get right down to it, so is copper.  There is a transmit pair
> > > and a receive pair.  It's the capabilities on each end that matter. 
> > 
> > I'm given to understand that this is not true for Gig-E on copper;
> > there's something smarter going on in the PHY's and they use all
> > wires for all signals.
> 
> IIRC it is the same as 100Base4, you transmit on 3 pairs
> and listen for collision on the 4th.

Nope.

http://www.mitech.co.uk/whitepapers/Insert-Guidance(Gigabit%20Eth).doc

"The 802.3ab task force selected Enhanced TX/T2 line code developed by
Level One communications for the implementation of Gigabit Ethernet,
this inherits the 100Base-T symbol rate and spectrum allowing auto
sensing 100/1000 products. 

The 1000Base-T achieves full duplex throughput by transmitting over each
of 4 pairs from both ends simultaneously. Each pair carries a dual
duplex 250Mbps data signal encoded as a 5 level pulse amplitude
modulation (PAM-5). "


1000Base-T does also support a half-duplex mode, but I don't know how
the signalling of collisions is handled.
 
-- 
David Maxwell, david@vex.net|david@maxwell.net --> Mastery of UNIX, like
mastery of language, offers real freedom. The price of freedom is always dear,
but there's no substitute. Personally, I'd rather pay for my freedom than live
in a bitmapped, pop-up-happy dungeon like NT. - Thomas Scoville