In 2016, G.fast looked very promising.
Thousands worked at developing and deploying.
It wasn't enough.
Most carriers are investing
in fiber or 5G instead.
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
Mid Blue: Smaller carriers in Germany, Norway, Finland, Japan
- Published: 03 April 2016 03 April 2016
Better cable doubles reach. Calix used a 16 port fiber-fed distribution point to deliver 500+ megabits 2,000 feet. For many years, American builders have wired homes with coax, most of which are served by the local cable provider. But 8% of the U.S. does not get cable broadband. If coax is in place, Calix has just demonstrated an alternative.
In addition, some developers prefer to maintain exclusive control of broadband. G.fast over coax could be a tool for a non-cable company to compete for the business. Within a building, there's a pretty stark performance difference between G.fast and VDSL.
If you have fiber to the basement, anything but G.fast is obsolete by the end of 2016 and possibly sooner. BT is proving G.fast works in the field
; they and others are solving the remaining bugs and other issues. G.fast gear is only modestly more expensive that VDSL to manufacture. Hard bargaining with the vendor should bring the price close.
Today's networks easily deliver 300 meg wired, 100 meg wireless, to most of the people most of the time. Anything less is obsolete and will suffer where competition is effective.
I hadn't reported that Calix, like Adtran, had publicly demonstrated bonding G.fast. I missed the press release from last year's BBWF. When I reported Adtran's similar demonstration, I should have included Calix as well. Cable is now upgrading to a gigabit in much of the world; Comcast & Cox are committed to delivering that gigabit to over half the United States. Without bonding of other improvements, calling G.fast a gigabit is simply false advertising.
AT&T has been vocal they need to be able to offer a gigabit in turn. The original G.fast pr talked of a gigabit but that was never realistic. It only applied under lab conditions and implied that no bandwidth was used for upstream. I've reported G.fast speeds as 300-800 megabits, depending on several variables. The coming "long-range" deployments of G.fast will leave many with much lower speeds. G.fast was geared for performance at 50-100 meters and trails off very badly after about 300 meters. Carriers want to save money by building fewer field terminals. Several intend to go 400-500 meters and maybe more.
Governments who allow false advertising of speeds are key enablers.
Calix Announces Innovations in G.fast and Vectoring at Broadband World Forum
LONDON – October 20, 2015 – Calix, Inc. (NYSE: CALX), the world leader in gigabit fiber deployments, today announced the expansion of its copper solutions portfolio to be highlighted at the Broadband World Forum (BBWF) this week in London. Among the new innovations are:
- Live demonstrations of G.fast bonding technology delivering broadband speeds in excess of 1 gigabit per second (1 Gbps) over existing copper infrastructure up to approximately 250 meters
- Live demonstrations of interoperability between Calix G.fast DPUs and modems and those of other vendors at the Broadband Forum Interoperability Pavilion
- The debut of breakthrough VDSL2 system level vectoring (SLV) solutions on the Calix E7-2 modular access systems that deliver up to 96 vectored ports without the economic and operational burden of a dedicated vectoring control processor card (VCP)
- The introduction of new VCPs on the E7-2 modular chassis that can scale from 97 to 384 vectored VDSL2 ports to flexibly serve more dense deployment scenarios
With these innovations, Calix has set new standards for performance over existing service provider copper infrastructure, and is leading the way in bringing to market the breadth of solutions necessary to meet the emerging demands of copper-fed broadband consumers globally.
"It is critical for service providers like Windstream to be on the forefront of technologies that will allow us to leverage our existing infrastructure to meet ever increasing subscriber service demands," said Randy Nicklas, executive vice president of engineering and CTO at Windstream. "Optimally leveraging our copper infrastructure is essential to our strategy going forward, and we are excited by the innovations that are allowing us to leverage our strategic copper assets to both stay ahead of demand and our competition. We've seen these technologies in action, and these innovations are leading the way in helping us to deliver a world-class subscriber experience effectively and efficiently, independent of the media type over which they are served."
Coax in the lab test setup
100 ft - 2,000 feet RG6. RG59 16 port unit
Star topology - lots of loop problems
16 port fiber DP
2,000 feet 500 meg + over coax
cpe - 801s g.fast twisted
bonded g.fast at BBWF and user group
get into labs in international big incumbents
southeast Asia - downtown
48 & 96 on the roadmaps
Bill in Santa Barbara
Kicking tires a lot of folks
larger folks want to go fast
G.fast is superior to vectored VFSL2 in MDU pretty stark performance difference
video from satellite in MDU with two infrastructures - the satellite guys may move further
Want to know about zero touch provisioning. - management issues