spacer 2Telebyte Gfast Testing Guide 320

Gfast map July 2017

Dark blue: 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

Light blue: Smaller carriers in Germany, Norway, Finland, Japan

Green: Incumbent likely:  France, Germany, & Poland  

Stonehenge Wikipedia 320Pioneers get arrows in their back and schedules do slip. G.fast technology is rapidly improving, with Amendment 2 and Amendment 3 chips delayed until the second half of 2017. BT had originally promised to start their 10M home rollout late this year or early next. In particular, BT wants 48 or 96 ports while today's tech only supports 16-24 ports. They also want the higher power and longer reach of the Amendment 2/3 chips. That's particularly important. Their last goal (350 megabits down at 350 meters) is beyond the current state of the art. It may be impossible, but that's unclear.

The original plan, promised by CEO Gavin Patterson, would go to local distribution points for speeds generally 500+ megabits. The 500 meg promise was part of Gavin's attempt to persuade OFCOM not to break up BT because they would deliver great Internet to "nearly all of Britain by 2025." Then the bean counters jumped in and said they wouldn't pay for the high speed. Using the existing cabinets brings the cost so low it fits into the existing capex budget but also reduces the speed.

Some very good BT engineers redesigned the project for 300 meters rather than 100 meters.

They brought the changes (more power, more bits/signal) to the ITU standards committee. Broadcom and Sckipio, the main chipmakers, went to work. They tell me it's been challenging, especially with the heat and power budget. In addition, vectoring 48-96 ports requires massive processing power, adding heat. Adtran's solution, which works well for VDSL, is to move the vectoring engine into an external module, the SLV. That's inspired a debate about whether vectoring need to be onboard or can be external. The Adtran SLV isn't ready for G.fast in any case.  

I'm going to call the downstream speed 200+ megabits, rather than the "up to 330" in the BT pr. It's time to use actual speeds that most users can achieve. Anything else is false advertising. BT's regulator, OFCOM, has promised to clamp down on false claims. So have the FCC in the U.S. and BnetzA in Germany. So far, they've all been afraid to take on the telcos. Deutsche Telekom has been particularly egregious, building a vectored VDSL network designed for 30-50 megabits to many as "up to 100." Frontier isn't even vectoring, but also claiming "up to 100." 

They are expanding the trials to 140,000 homes. Light Reading expects that by March. If BT concentrated on apartment buildings, the 500 meg is easily achieved already. But Brits love their gardens, and only about 11% live in MDUs.

BT's speeds can be increased by an important 100 megabits by eliminating the VDSL carveout.

 

The Site for gfast 230
 

G.fast News
A remarkable 400 people attended the very strong Broadband Forum BASE events in Berlin and Las Vegas. Trevor confirmed BT would pass the million this year. Cioffi projected “Waveguide DSL” could carry 10 gigabits a kilometer as well as a terabit 100 meters. Werner sees a 4X improvement in upstream with cDTA. Much more in next issue.

Deutsche Wants a Gigabit, Finally Realizes 50 Meg Isn't Enough http://bit.ly/2zeZ5oZ
Deutsche Telekom is finally realizing that 50 megabit DSL won't make it against gigabit cable. VP Franz Seiser is blunt. "We must change radically, become disruptive and, above all, throw away things," he proclaims at BBWF. After years of DT insisting 50 megabits is plenty, we now hear "it is about Gigabit products" from DT's Robert Soukup.  
    A lucky building in Frankfurt will receive 500+ megabit service as ultra-conservative Deutsche Telekom experiments with G.fast. Soukup told BBWF, "We're going to have a field test in Frankfurt with G.fast and Fiber To The Building (FTTB.) We will know by the end of the year if this is the right way to go." Hint to Soukup: Yes it is. G.fast is working well at a dozen telcos I;ve talked to.
     The details are surprising. DT is going for CORD, Open Source, Calix, and Radisys. http://bit.ly/2zeZ5oZ

*** The new Telebyte Guide to Testing Gfast follows the Broadband Forum IR-337 Gfast test specification, the same used by the University of New Hampshire (UNH-IOL) for Gfast certification testing. Free download http://bit.ly/telebyte (ad) It is the best technical guide to G.fast  I have seen. Grab it. Dave

1.6 Gig in Sckipio-Calix Test http://bit.ly/Calix16
A telco tells me they are getting impressive early results from the Calix 48 port DSLAM with the new Sckipio 212 MHz chips. There still is work to do but this is encouraging. 
    Carriers want DSLAMs with more than 16 ports to reduce the deployment costs from the basement or larger field cabinets. Speed matters to the marketing side of the company; AT&T's CEO believes he must offer a true gigabit to match cable. (They've been getting ~750 megabits with first generation chips. http://bit.ly/Calix16

*** Self-Healing Wi-Fi With ASSIA Real-Q 
Beyond-the-Box visibility and control extends quality-of-experience (QoE) beyond the gateway to the end-user device for every device in the home. Based on ASSIA technology, proven across 80 million subscribers http://bit.ly/2dj7FJk (ad)

Reverse Power 4 Port DSLAM for Australia http://bit.ly/NetcommRP
Australia is connecting 1M homes to G.fast, some with a Netcomm distribution point mini-DSLAM. It's a small unit designed for pole or pit mounting. It's waterproof, pressure proof, and temperature resistant. Their matching home modem is bittorrent friendly, with two USB ports for a hard drive dedicated to sharing.
     A reverse power unit at the customer, the NDD-0100-01, can save the cost of bringing power to the DSLAM. They don't expect many orders until the second half of 2018, as nbn is waiting for the second generation chips. Netcomm demonstrated RP with BT Openreach in August. http://bit.ly/NetcommRP

*** Sckipio's Three advances are taking G.fast to the next level.http://bit.ly/Sckipio (ad)

Australia Makes it Official: G.fast to Million Plus http://bit.ly/GFAussie
No news here. In September, 2015, I reported Australia's nbn Going G.fast. This June. I reported the million home fiber to the curb (kerb?) was beginning. Unfortunately, they are no closer to figuring out where to find the needed $10B to $20B to cover the cost overruns. Instead, the parties are battling in Parliament about who is to blame. http://bit.ly/GFAussie

2 Bonded 212 Lines = 3 Gigabitshttp://bit.ly/twobonded
Sckipio at BBWF is demonstrating 3 gigabits down, nearly a gigabit up, over two phone lines, bonded. Twice the bandwidth (212 MHz instead of 106 MHz) times two lines is fast. Sckipio does great demos; at CES, they showed G.fast first generation chips delivering almost 1 gig upstream.
    “Sckipio is pushing Gfast to astonishing speeds with production silicon,” CEO David Baum proclaims. Calix is using the SCK23000 chipset in their 48 port gig+ DSLAM at the show. http://bit.ly/twobonded

Read more ...