Silicon to stop next-gen powerline standards war
Will the mains networking biz put its squabbles behind it?
Battle lines drawn - where are the mediators?
With big names on both sides, the stage is set for the inevitable 'standards war', a conflict companies backing either technology appear to be keen to promote.
Ironically, the market for powerline products remains well short of its potential, leading some commentators, such as David Waks of long running powerline news site Broadband Home Central, to throw their hands up and wonder why the likes of the HPA and HGF aren't working together to expand the market to the benefit of both.
That may yet happen, albeit indirectly. According to Jean-Philippe Fauré, head of the IEEE team charged with turning 1901 from a notion into a specification and then formalising it as a standard, interoperability provisions built into both 1901 and G.hn will ensure that the two standards can happily co-exist.
Co-existence is not compatibility, of course, but it ensures that if you put in place a 1901-based powerline network, the G.hn-based adaptors bundled with your broadband box by your service provider can be used too, without killing both LANs.
Fauré told Reg Hardware that both standards have agreed measures by which they will equally share the available mains bandwidth in the 2-30MHz band.
All very well for new 1901- and G.hn-based adaptors, but older HomePlug AV and UPA kit, all of which lack the co-existence tech, will likely kill them. Moral: if you upgrade one box to 1901 or G.hn, upgrade them all. This is not likely to be expensive.
And it may not be necessary to choose one new standard or the other, says Fauré. Mobile phone chips already support numerous networking technologies, each with distinct physical links - the so-called PHYs - and do so happily.
Savvy powerline chip makers can do the same, building support for both G.hn and 1901 into their products, allowing adaptor A to talk to adaptor B no matter which of the two standards the latter is dedicated to.
The G.hn specification contains one PHY, but 1901 has two. A three-PHY chip is not hard to do.
None of the chip makers eying either standard have yet gone public on plans to support both rather than one, but it's a logical step for them to make. The first powerline chip makers were relatively small operators, now long since acquired by big names - Intellon by Atheros, DS2 by Marvell, Gigle by Broadcom - who have the size to take on multi-standard developement efforts and the competitive motivation to do so.
The powerline standards war may yet be called off before the troops march into battle. ®