Comments Locked

48 Comments

Back to Article

  • Murloc - Tuesday, May 3, 2016 - link

    powerline has always been a pain for me, it's less hassle to call up the electrician and get a cable pulled. It costs money and it needs space in the wall but at least it works with decent speed.
  • Samus - Tuesday, May 3, 2016 - link

    I've never had it work well unless you use it on circuits of the same phase. It seems as soon as you cross phases the reliability and speed fall off significantly.
  • magreen - Wednesday, May 4, 2016 - link

    Agreed. I don't think powerline has ever lived up to its billing.

    I think coax is the way for people who don't want to pull cable. Old cheap DECA units handle 100Mbps flawlessly. Newer bonded MOCA units are pushing actual throughput of 1 gigabit.
  • kmmatney - Wednesday, May 4, 2016 - link

    I wasn't able to get my powerline adapters to work with my desired sockets, so had to use another plug across the room. And then one of the adapters failed after about 2 years. I switched to a cheap programmable wireless adapter. You just connected it to a computer to input your wireless settings, and then it worked on it's own after that. The wireless solution was much cheaper and overall less hassle.
  • barleyguy - Friday, May 6, 2016 - link

    I've had pretty good luck with speeds with some TP-Link powerline adapters, but I'm getting rid of them anyway and moving to 802.11AC. The problem is that they create noise on the electrical line, and I can hear it in my speakers when there's not music playing. I don't feel like the utility is worth the hassle either.
  • cswilly - Tuesday, May 3, 2016 - link

    Informative review. Looking forward to your testing with BRCM's AV2 part. As it uses 2-87MHz it should outperform Marvell's G.hn on good links.
  • spawnbsd - Tuesday, May 3, 2016 - link

    Marvell's 2nd gen 88LX5153 are starting to sample now, and support 2-100MHz in powerline MIMO. In the limited testing we've done in our labs, we're seeing near ~1Gbps throughput in common powerline topologies. Hopefully some hit retail sooner rather than later.
  • morgi - Thursday, May 5, 2016 - link

    that is awesome! what lab are you working in?
    Is it real TCP traffic?
  • nathanddrews - Tuesday, May 3, 2016 - link

    It's a real shame that powerline networking hasn't received more interest and R&D. It's the perfect backbone for IoT, home automation, and retrofitting older homes that don't have ethernet or homes that struggle with radio interference.
  • hojnikb - Tuesday, May 3, 2016 - link

    I also wonder why its not adopted as a wide area network ie for internet. Instead of using phone lines use existing power network.
  • FATCamaro - Tuesday, May 3, 2016 - link

    There are curbside transformers that would completely mangle any data on the line for internet. I'm sure you've seen these buzzing boxes around you?
  • Manch - Wednesday, May 4, 2016 - link

    A long time ago there was research into using power lines for data transmission. The idea was to use the em field around the lines to carry the signal. Spurious signals can be picked up like this so why not intentionally use it to send/receive data.That research laid the ground work which turned into this tech.
  • GTVic - Wednesday, May 4, 2016 - link

    That is called native encryption.
  • willis936 - Tuesday, May 3, 2016 - link

    Phone lines have much better properties for signals. Also every house in America already has a phone line thanks to the Communications Act.
  • cigar3tte - Tuesday, May 3, 2016 - link

    But older homes are also that ones that don't have the proper wirings to support powerline networking.
  • Ninhalem - Tuesday, May 3, 2016 - link

    I used power line networking to great effect in a home built in 1892. Wiring is dependent on how good the builders were at the time. You can have modern home builders and electricians who shouldn't have been allowed to touch power tools. I find that most construction from the turn of the 20th century and in the 1930's was some of the best construction ever.
  • JoeyJoJo123 - Tuesday, May 3, 2016 - link

    What is "Anectdotal Evidence" for $5000, Jerry.
  • Samus - Tuesday, May 3, 2016 - link

    LOL, are you denying his claim that pre-WWII construction was superior to modern construction quality?

    Having a major in architecture and having worked in construction, most modern residential homes are built to last 30-50 years. Materials and quality of construction. The only advantage is the homes are more efficient and have more creature comforts.

    In the 20's and 30's, that mentality didn't exist. Things were just built. Well.

    Prior to the 70's, the concept of "flipping" also didn't exist. Homes were built and bought to live in forever, much of the time planned to be passed down through the family. These days, almost half of homes are purchased for short term occupancy. aka, 'flipping.'

    Buying modern construction is scary, especially if you don't know what to look out for. I'd rather have a 100 year old used and abused home than a 10 year old cookie cutter that was built cheap and has already been through 3 clueless owners.
  • sor - Wednesday, May 4, 2016 - link

    I don't know. I've seen my fair share of knob and tube, insulation that has settled and compacted, moisture barrier issues, rot, insufficient foundations. Most homes that are that old have already had major work done or are piles of crap that either need to be stripped to the studs or they're just run down homes that people put up with because of cost. You can find the occasional well built home, but since building code and inspections were sparse it's a total crap shoot. Looking at the past thinking everything was well built is looking with Rose colored glasses.
  • eek2121 - Wednesday, May 4, 2016 - link

    Took the words right out of my mouth.
  • kamm2 - Wednesday, May 4, 2016 - link

    Yeah, 99.99% chance a home built in 1892 with good wiring does not have 1892 wiring. If it was even built with any wiring.
  • extide - Wednesday, May 4, 2016 - link

    While I agree with most of what you say, the words 'Occupancy' and 'Flipping' should never be used in the same sentence when talking about homes...
  • kmmatney - Wednesday, May 4, 2016 - link

    I would think it would just be easier to use wireless in an older home. With most people having laptops , phones, ipads, etc.. your going to be on wireless most of the time anyways. You can get repeaters easily enough. My home was built in 1999, with ethernet cables wired across the whole house, but it turns out a lot of the connections are not convenient, so I still use mostly wireless.

    I did use a powerline adapter for one connection, and found it was a bit finicky, and wouldn't work between all plugs (and I'm sure my wiring is fine).
  • mgrier - Tuesday, May 3, 2016 - link

    Definitely. It would rock for a streaming box but I guess those tend to be small USB-powered and use wireless. I'd prefer to have a real outlet needed and avoid cluttering my wifi with video.
  • Murloc - Thursday, May 5, 2016 - link

    because the power network in houses is a mess or unknown to the owner and something always ends up not working.
    Wireless is just a better solution for most people and applications and even that isn't hassle-free.
  • fazalmajid - Tuesday, May 3, 2016 - link

    I am getting spotty performance on a pair of PLA-5405 I use to bridge upper and lower floors in my 1936 house with electrical wiring upgraded in 2006 (odd layout makes Cat5/6unpractical, plaster and lath construction is highly effective at blocking WiFi). I ordered a pair of these and a pair of HomePlug AV2000 Extollo Lansocket 1500 that outperform them in other reviews. G.hn proponents claim it has better noise resistance, I'd like to put that to the test.
  • Guspaz - Tuesday, May 3, 2016 - link

    You seem to use the terms "HomeGrid" and "G.hn" interchangeably without ever explaining that they're the same thing. That was very confusing, as for most of the first page of the article, I thought you were talking about three different standards.
  • Valantar - Tuesday, May 3, 2016 - link

    I agree. It took me all of the first page to understand that G.hn and HomeGrid were the same thing. No clear link between the two names, and no mention in the introduction. What do the two names mean? Is one a standards organization, while the other is the name of their tech? If so, which is which? And why have two names at all?
  • Guspaz - Tuesday, May 3, 2016 - link

    I think adding a single word would solve the problem. Change this:

    However, despite silicon getting demonstrated at various trade shows, G.hn was unable to get a retail product out for a long time.

    To this:

    However, despite silicon getting demonstrated at various trade shows, HomeGrid's G.hn was unable to get a retail product out for a long time.
  • grazapin - Tuesday, May 3, 2016 - link

    Yeah that first paragraph drops the names HomePlug, HomeGrid, G.hn and Comtrend with no explanation of how they relate to each other. I picked up most of it from context, but how G.hn fit in was not clear for most of the article. In fact, the only explanation that jumps out at me now is in the very last paragraph: "HomeGrid / G.hn"
  • bigboxes - Tuesday, May 3, 2016 - link

    Agreed. I was able to infer that they were the same, but yes it was confusing.
  • geniekid - Tuesday, May 3, 2016 - link

    +1.
  • spawnbsd - Tuesday, May 3, 2016 - link

    HomeGrid forum is just an organization that promotes G.hn, that's all. Many of the equipment and chipset vendors building G.hn gear belong to HomeGrid forum.
  • sor - Wednesday, May 4, 2016 - link

    +1, sounds like three different competing standards in the article.
  • Icehawk - Tuesday, May 3, 2016 - link

    What is idle power after turning power saving on?
  • mgrier - Tuesday, May 3, 2016 - link

    Any chance I could get you to test latency also? This is important for competitive gamers...
  • gobaers - Tuesday, May 3, 2016 - link

    I just did a ping sweep on my Homeplug connection:

    Reply to request 4861 (1 ms) (size 997)
    Reply to request 4862 (10 ms) (size 998)
    Reply to request 4863 (10 ms) (size 999)
    Reply to request 4864 (1 ms) (size 1000)
    Success rate is 100 percent (4865/4865), round-trip min/avg/max = 1/5/30 ms

    Mostly 1ms, but can sometimes spike. I wonder if G.hn handles this better.
  • spawnbsd - Tuesday, May 3, 2016 - link

    Marvell's G.hn chipsets use G.hn's TDMA MAC mode, and generally always given lower (more determinstic) latency and throughput compared to CSMA based HomePlug AV/AV2 gear. Latency should always be in the ~1ms range, unless impulse noise or other interference in the home forces link layer retransmissions (which will add jitter).
  • qlum - Tuesday, May 3, 2016 - link

    I wonder how the performance of these devices differs under the European 230v/50hz compared to the us.
  • Murloc - Thursday, May 5, 2016 - link

    they're just as much finicky and dependent on where you put them in although there is no weird old wiring in most european houses.
  • gobaers - Tuesday, May 3, 2016 - link

    My current setup is two Ubiquiti Unifi access points, one connected directly into the firewall and the other connected via TP-Link HomePlug adapters, as well as an IP phone via HomePlug adapter as well. They're somewhat reliable, but I see one of the powerline adapters drop connectivity every few days. Really, I should wire the house with Cat6, but it's going to be a messy job.

    Replacing the consumer Netgear WNDR3400 with a Cisco ASA5505, and adding Unifi WAPs has been an awesome. Running the Unifi controller software on a desktop, live handoffs and great wifi through the house, it has been rock solid at least until the powerline adapter disconnects.
  • Acarney - Tuesday, May 3, 2016 - link

    Any word on ping times with just a single brand/point to point network setup? I'm mainly looking to hardware my TV/AppleTV/PlayStation 4 with my network. I have an old G wifi hub connecting via bridge mode to my router which is about two rooms over. I need to hard connect these devices because if I don't they decide to connect to the wifi access point upstairs... Which is ok, except it randomly drops occasionally to where the one downstairs a couple rooms over is stronger. It's just a pain if gaming or streaming and a cut out happens. My internet pings are about 25 to 35ms at the best. Would a power line network drastically extend those? (Say into 100ms+) if it was just router -> power line plugin -> power line plugin -> network hub to break it out into 3 plugs for TV, AppleTV, PS4.
  • paranoised - Thursday, May 5, 2016 - link

    I have a Homeplug AV2 device and the latency I see is typically in the 2-4 milliseconds. Using powerline networking will probably increase your throughput and reduce latency tangibly.
  • rtho782 - Tuesday, May 3, 2016 - link

    I wish I could find testing like this in a UK setup.

    As our electrical wiring is significantly different (ring mains rather than radial circuits, leading to signal reflections etc, much more common use of RCBOs for individual rings, meaning that to go upstairs to downstairs, or either to the kitchen, you travel through two RCBOs) you can't directly compare the results, and UK/Europe seems to be an afterthought to the US!
  • Lepton87 - Wednesday, May 4, 2016 - link

    I concur, I actually looked up what G.hn means once the author used it because he didn't explain what he meant by that. I don't like reading something and not be sure what I'm reading about so even though it would be clear after a while what G.hn means just from the article it still makes the first page more confusing that it should be. Please change it so the others don't have to look it up or not be sure at first what they are reading about.
  • Lepton87 - Wednesday, May 4, 2016 - link

    I concur, I actually looked up what G.hn means once the author used it because he didn't explain what he meant by that. I don't like reading something and not be sure what I'm reading about so even though it would be clear after a while what G.hn means just from the article it still makes the first page more confusing that it should be. Please change it so the others don't have to look it up or not be sure at first what they are reading about.
  • jardows2 - Wednesday, May 4, 2016 - link

    **Disclaimer: I work for ARRIS consumer support**

    We received samples of the G.hn products to test and learn for support purposes. We were unable to do any detailed tests, but were able to take some units home for pre-release trials. So far, everything seems to work well for everyone that tested, even those who had old wiring and fuse boxes! No word on speed, but connectivity seemed to be fine, except certain surge protectors were problematic. I am looking forward to AnandTech's full evaluation of the products!
  • sepecat - Wednesday, May 11, 2016 - link

    The most problematic weakness of all RF-based tech (including powerline, MoCA, and wireless) for me is high latency. I've tried multiple iterations of each of these and always have a latency of at least 2ms and often over 10ms. This is a showstopper for my primary application, streaming from my PC to a SHIELD console. I don't see how that could be resolved without a dedicated, noise-free medium, though.

Log in

Don't have an account? Sign up now