spacer 2Telebyte Gfast Testing Guide 320

gfast map nov

Dark Green: Firm commitments from incumbent: BT (10M), Belgacom, Australian NBN, Swisscom,  Austria, Bezeq Israel, Chunghwa Taiwan, Telus Canada, Telekom South Africa, SK Korea, (U.S.) AT&T, Century, Frontier, Windstream, Belgium, Omantel

Mid Pink: Smaller carriers in Germany, Norway, Finland, Japan

Light Green: Incumbent likely:  France, Germany, Italy

Jochen Maes slide 9Brilliant engineers disagree. As g.fast extends to 212 MHz, noise on the line becomes increasingly difficult to tame. There is more noise than signal. Dong, of Huawei, had presented the idea of non-linear pre-coding previously. A respected engineer told me he was on target. May 20, his Paris presentation was convincing. 

Jochen Maes of Nokia Alcatel came next to the podium and contended that non-linear coding was unnecessary. Maes works on the five gigabit Alcatel systems that go to 500MHz. He’s close to the data and believes what he calls “optimized linear” coding is actually best. He also believes that optimized linear requires far fewer calculations. The chips would require less power and run cooler, important factors.

Hard to characterize “bad” lines are a prime area of disagreement. Maes introduced some new tools to model the impact of channel perturbations. His results supported linear. Maes writes,

"All analysis on NLP so far assume theoretically ideal conditions. My presentation is the first that analyses the impact of channel fluctuations, e.g. due to mechanical vibrations or temperature fluctuations or imperfect channel estimates. It shows that NLP is more sensitive to such fluctuations and thus that under practical conditions NLP loses its advantage."

Because of the complexity, it is difficult to determine which factors are crucial. I am not an engineer. Even if I were it would be hard to decide who has this right. Definitive answers will not be available until there are extensive field trials. Even then, the question of the robustness of each implementation is likely to muddle the argument.

I remember the "QAM" versus "DMT" debate in VDSL a decade ago. That was an intense worldwide battle, with Texas Instruments and Alcatel employing a high-profile pr agency to protect their royalties. There were $hundreds of millions at stake. This one was bitter.

A prominent TI engineer accused me of being in the pay of the QAM team. That was surprising. I had written that it seemed to me DMT was technically superior. The only money I took from any of the parties was very visible advertising - from companies on the DMT side. (She never apologized.) 

The QAM guys, Metalink and Infineon, had inexpensive chips ready to go, proven interoperability and support from China Telecom and others. They simply asked to be included as an option. Off the record, they contended the tests were unfair; the DMT folks had brought in John Cioffi at the last minute to tweak the system. (That may not have been true. I never asked John.) As I recall, after the QAM guys did their own tweaks, the result was very close. 

I wouldn't be surprised if this debate goes on for a long time. Meanwhile, the chipmakers are in a chicken and egg situation. Until they incorporate new algorithms, it will be hard to test which coding offers better performance. Without a clear choice, it's hard for them to invest in adding the functionality to the chips. 

I welcome on or off the record comments from all and have asked the principals for more details.

 

The Site for gfast 230
 

G.fast News

I’m still working through remarkable presentations from the Broadband Forum events. Michael Weissman, Bernd Hesse and team did a remarkable job choosing the speakers. http://bit.ly/BBFBASE

Deutsche Telecom: 35b Supervectoring Delayed to 2019 http://bit.ly/35blater
Broadcom is now over 3 years late. DT briefed German reporters after their financial call and revealed 35b was now delayed until 2019. 35b should deliver 200+ meg downloads 500-600 meters, a crucial tool for DT, which is losing share to cable. Cable now covers about 70% of Germany and is expanding. DT now only offers 50-100 megabit DSL while cable is often 400 megabits, going to a gigabit. 

The problem is software; the hardware is shipping and supposedly will work. DT says 35b is not ready to turn on. Broadcom in 2015 said 35b was in "production" in the press release below. Alcatel in early 2016 said to expect complete systems very soon. "35g is very similar to 17a so there should be little delay."

Broadcom's problems are leading major telcos and vendors to have a plan B, using Sckipio G.fast. DT itself is planning extensive G.fast deployments in 2019, mostly in apartment buildings. http://bit.ly/35blater

Gigabit 100 Meters - Unless the Wires are Lousy http://bit.ly/gflousy
Speeds are fine, "Unless there's a line problem." I've been reporting for three years that ~10% of lines have problems. In the chart by Rami Verbin of Sckipio, he finds G.fast goes ~130 meters on good lines. Poor lines have about half the reach. 

His chart roughly matches the reports from Swisscom, Belgacom, and England for both G.fast & vectored DSL. The 10% with problems can cause the majority of the line-related complaints to support. The angry customers drive up cost.

Rami's solution to reach the gigabit is bonding, supported on the Sckipio chips. Verbin made some additional points:

  • 4 gigabits is possible by bonding two decent 2 gigabit lines.
  • Even in a service from remote cabinets, ~25% are close enough to get a full gigabit."
  • cDTA and iDTA are practical ways to deliver much higher upstream by switching some bandwidth from downstream to upstream only when needed.
  • 35B will probably be similar but Deutsche Telecom doesn't expect to deploy until 2019. http://bit.ly/gflousy

AT&T Wants Coax 2-5 Gigabit G.fast. Very Soon. http://bit.ly/ATTCoax
AT&T faces intense competition from cable, talking about 10 gigabits in both directions. (Cable will only be 1 gig down, ~100 meg up, until ~2021.) AT&T wants something to brag about as well.

AT&T gained millions of lines of coax as part of the DirecTV deal. Alcatel and Huawei are leading the development of G.mgfast. That uses 424 MHz, full duplex, to achieve ~2.5 gigabits in both directions. The reach on telco twisted pair is only about 30 meters. On coax, those speeds can probably extend far enough to service most apartment buildings. Using 848 MHz, speeds can reach 5 gigabits. The ITU standards group has been aiming for 2019-2020 for G.mgfast, too slow for AT&T's marketers. David Titus wants a high-speed standard for coax "early in 2018." He believes that is "doable."http://bit.ly/ATTCoax

Read more ...