Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Weaver]

I am trying to understand what the order of things was historically. This document talks about the deeply wonderful goodness that is PhyR and Broadcom people give themselves a pat on the back. Since G.992.3 uses G.998.4 did the authors of G.992.3 even have an L2 retx mechanism in mind at all when they were putting together G.992.3, or what was the sequence of events? I am sure someone will set me straight with the chronology anyway.

That document talks about ‘DSL standardisation’ making it sound as if Broadcom were willing to donate their work to be the basis of either just some nameless standard, or to actually become the G.998.4 that we have today. That would mean them giving up their unique advantage.  Since BT has afaik no G.992.3/G.992.5 DSLAMs that speak G.998.4 anyway, in the ADSL2/2+ space, Broadcom ended up with a unique competitive advantage. A big reason for users to buy Broadcom-based ADSL2/2+ modems and for telcos to buy Broadcom DSLAMs to have happier ADSL2+ users and links that are more reliable.

I wonder about the extent of Broadcom patent coverage here. Is there any reason why other manufacturers could not have put together their own PhyR implementations?

[PhilipD]

Hi

This is basically G.INP as used on VDSL. PhyR was/is Broadcoms brand name.

Regards

Phil

[ejs]

Is there any reason why other manufacturers could not have put together their own PhyR implementations?

They may have made their own retransmission scheme, I doubt they will have made their own implementation of Broadcom's PhyR.

Perhaps have a read of this Ikanos whitepaper.

Ultimately, both schemes were combined into a final unified scheme, where management of
retransmission queues can be implemented transparently in either the PMS-TC or in the TPS-TC
layers of the DSL reference model—see Figure 4. Modems are unaware of the layer that the
other modem has implemented the retransmission buffers.