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: 14 November 2016 14 November 2016
Not a typographical error, unfortunately. To reach the projected "USD 4,216.3 Million" in G.fast chip sales, chip price per home served would have to be $200 or so. That's ten times the price today in some markets, with every expectation the prices will go down over six years. (Moore's Law may be slowing but it isn't dead.) Save your $5650; an outfit called Markets and Markets has produced a work of low fantasy in the guise of a research "study." Buy these folks a new abacus. They include estimates for Broadcom, Sckipio, and Metanoia, who recently demonstrated interoperability, as well as Qualcomm/Ikanos, shuttered months before the November 2016 report date.
They include estimates for Broadcom, Sckipio, and Metanoia, who recently demonstrated interoperability, as well as Qualcomm/Ikanos, shuttered months before the November 2016 report date. If G.fast chips were even a quarter the price implied, most companies would use either fiber home or supervectoring.
At a more realistic price of $20/port, it would require deploying 200M lines that year. Ovum estimates 11.5M G.fast connections in 2021, in a report paid for by British Telecom and Australia's nbn, both strong advocates.
That's higher than many estimates, although I can envision somewhat higher scenarios. It would be remarkable if 20M lines were connected in 2022.
Predictions about the future are hard - see Trump, Donald. Precision five years out on a technology barely out of the labs is impossible. Hitting the target within 10% would be remarkable here. The $4,216.3M estimated sales requires precision to less than 1/10 of 1%. My eight grade math teacher would have flunked me.
A natural paper for The Journal of Irreproducible Results.
G.Fast Chipset Market by Deployment Type (DPU and CPE), End User (Residential, and Enterprise/Commercial), Copper-Line Length (Lines Shorter Than 100 Meters, Lines Of 100 Meters–150 Meters, Lines Longer Than 250 Meters) - Global Forecast to 2022
Publishing Date: November 2016
Report Code: SE 4709