G.hn-ing for gigabit It's surely no coincidence that shortly after one technology for high-speed home networking achieved a key step toward becoming a world standard, the minds behind an alternative approach said they too had finally got the ball rolling. It won't do it much good - the second would-be standard, IEEE P1901, is arguably doomed, a …
Remember that USB has predominated over Firewire, even though high-speed USB needs two PHYs (one for 480 Mbit signalling and another for 12 and 1.5 Mbit). It only needs one MAC though. It's an example of a technology that has prevailed despite needing more stuff to implement it. Why? Two possibilities are the cost of licensing and the influence of early adopters. If one of these home networking proposals is cheaper to license than the other, or if someone like Intel gets behind it, then that will probably have more effect on the outcome than the number of PHYs.
I get around 10 megabits/s from my "200 megabit/s" powerline adaptors, I wonder what real world throughput over real world mains cabling will be for these new "gigabit" products? If the RLF (reality lag factor) is similar at 5%, I suppose 50 megabits/s would still be a welcome improvement. But it is my observation that typical RLF worsens with each new technology, so I'm not holding my breath.
In a sense, who cares what the reality lag is? No real-world wireless connection has ever come close to the headline speed, and neither has a powerline link. Heck, even ye olde analogue modems never actually reached 56Kb/s.
What matters is that it's faster than what we have now. I just hope it's 5x faster - '200Mb/s -> 1000Mb/s' as promised by the headline speed.
We shall see.
I recently wired my whole house with good quality CAT6 Ethernet cabling and jacks for LESS than AU$300. I get FULL gigabit speeds, I don't have the local HAM radio operators building a bonfire with a large pole sticking out the top on my front lawn, and It's all based on a tried, proven, trusted and extremely well supported standard!
If I want a Network printer installed, plug a patch cable into the wall.
If junior wants XBOX-Live in his room, plug a patch cable into the wall.
If I get a new Media PC for the lounge room, plug a patch cable into the wall.
No need to run out, spend AU$100 on a converter box to be able to plug the patch cable into the power point, only to get inferior connection speeds and a bunch of pissed off HAMs.
@Phil Endecott
> if someone like Intel gets behind it, then that will probably
> have more effect on the outcome than the number of PHYs.
Intel is already a big supporter of the G.hn approach. Intel is a board member of HomeGrid Forum, an organization set up to promote G.hn technology and ensure interoperability between vendors.
http://www.homegridforum.org
On the face of it, notches sound like a good idea but it's really just a convenient word. In their current form they only cover ham radio and even that is not perfect.
The shortwave band is used for many other things, including broadcast radio, and if all the required notches were in place (http://www.mikeandsniffy.co.uk/UKQRM/if.htm) then there's not sufficient bandwidth remaining!
The current thinking seems to be that because PLT equipment is not classed as a radio transmitter, it can get around all sorts of regulations. IMHO it's only a matter of time before it is realised that these devices are in fact simply radio transceivers which use mains wiring for their transmission medium.
In any case, it has been demonstrated time and time again (see articles in RadCom, RSGB for example) that these devices all fail against current standards regarding EMC so I don't understand why they are on the market.
The regulator should do their job and enforce the standards, even if that means admitting failure to do so in the first instance. It's not their job to decide what's in the public greater interest.
The less the IEEE have to do with networking standards the better it is for all of us -- they invariably make a pigs' ear of everything they touch. (Wireless, anyone?) This is no exception, especially as its tied the rather quaint notion that everything is bolted to the wall and plugged into something. While it is true that we're unlikely to buy cordless refrigerators any time soon the majority of our information consuming devices are portable these days and are only likely to get more so over time.
The problem with wireless and bandwidth is just technical, an iffy standard and data transmission crammed into a tiny sliver that nobody else wanted because it was deemed useless. Open the R/F window just a little more and we'll have as much bandwidth as we need, especially if we're only talking about limited range.
One of the reasons why I still use wired networking - and would have used coaxial cable if was still viable -- is that I care about standing power consumption. Unlike an office a home network isn't being used that much so leaving equipment on 24/7 is wasteful. Intelligent wireless can be made low powered. Powerline networking cannot -- its a crock, and an antisocial one at that.
That would be a bridge, right?
Seriously, put your CAT5e cabling above the ceiling and drop it down into cupboards.
Then let me know when the ADSL router manufacturers start putting 8 or 12 port gig-ethernet switches into their devices.
Icon: Ethernet: physical star, logical bus
Good article. For the record, HomePNA is a 2 wire technology -- the same PHY operates over both coax and phone wires (most North American use like AT&T's U-verse IPTV are over coax. Phone is used in the rest of the world).
This type of wired home networking is widely used for commercial triple play IPTV and VoIP so Service provider support is important. G.hn is heavily supported by telcos but IEEE P1901 isn't. There is a lot of information comparing G.hn with IEEE P1901 and home networking in general on the HomePNA blog. http://homepnablog.typepad.com/
Also, I was working at one of the major 802.11 companies during the PHY wars. There were three 801.11 PHY technologies and direct-sequence spread spectrum won for technical reasons - frequency hoppers couldn't deliver the higher data rates in the same bandwidth. The market window opened and volume drove the price down. G.hn has put technology first in its standard development.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2025
