RE: From serial PMD call, 3 April

[Jonathan Thatcher <Jonathan.Thatcher@xxxxxxxxxxxxxxxxxxxx>]

Title:

Regarding:

Other issues raised by the shift to a box level spec
----------------------------------------------------
Perhaps without noticing it, we have moved from a component level standard
(Gigabit Ethernet) to a box level standard (10 Gigabit Ethernet).  We need
to check that we have made the transition correctly.  One known issue is
that of some components which try to appropriate the whole box's allowance
to themselves (e.g. jitter).

This is an interesting discussion point. Strictly speaking, if you look carefully at the Gigabit Ethernet specification, you will see nothing about a component specification. The specification is for operation at TP2 (Tx) and TP3 (Rx). The specification requires connection to a system implementing 1000BASE-X.

Now, it could be argued that the specification is not complete because it includes no details for how some of the optical tests can be accomplished in a system environment. It can also be argued that there is no case where a company tests the optical parameters in a system configuration. Even if this was strictly true (it  isn't), it would be beside the point. The standard does not specify components. It specifies operation of a piece of equipment implementing the standard. A transceiver does not do this. Neither does a SERDES or any other component.

We, therefore, have not shifted the specification to the box level. We have simply done a better job describing how to test the optics at the box level. Congratulations! Good work!

This is an important distinction....

jonathan