In depth reviews of ethernet over power?

[ViviTheMage]

I ask because in my new house, my modem is pretty far away from where I have my computers, so I have to run EoP...I use netgear 500mbps devices, which connect at around 200mbps ... my speedtests are great, no issues there, but I am wondering if there will be latency, or heavy packet loss using it?

Has anyone done any indepth reviews on the technology?

 

[JoeMcJoe]

Nice speeds. I get just over 100 Mbps, works great, no latencies really.

This site has reviewed them many times.
http://www.smallnetbuilder.com/lanwan/powerline-charts/view

 

[ViviTheMage]

Thanks, i'll read through those ... I was actually looking on that site hoping they had a review!

So far the speeds are rock solid, but I just wanted to make sure I am losing out on some latency/packet loss because I am using them as my internet drop.

 

[JackMDS]

CAT5e, CAT6 follow Ethernet standard, power lines do not.

CAT cables are twisted to provide level of protection from noise. Power wires are actually creator of Noise.
If your electrical lines are identical in any way possible to the one that are used by others you will get the same performance.

Otherwise, YMMV.

 

[ViviTheMage]

yes, huge YMMV, but I was more curious about drawbacks when comparing to cat cable ... is there added latency/packet loss at all?

 

[mammador]

I don't understand. Do you mean PoE? or do you mean powerline (using mains electricity as a Layer 1 medium)?

Depending on the size of your house, it shouldn't be an issue generally. One thing is that available bandwidth via powerline is not as high as via Ethernet.

 

[ViviTheMage]

I don't understand. Do you mean PoE? or do you mean powerline (using mains electricity as a Layer 1 medium)?

Depending on the size of your house, it shouldn't be an issue generally. One thing is that available bandwidth via powerline is not as high as via Ethernet.

EoP = ethernet over power...that is just what I call it, figured everyone else did too. I already have the XAV5001 kits, and so far work great, I just wanted to know if I was losing out on latency/packet loss ... I don't really know a good way to test the connection and see what bottlenecks it is creating.

 

[frowertr]

I ask because in my new house, my modem is pretty far away from where I have my computers, so I have to run EoP...I use netgear 500mbps devices, which connect at around 200mbps ... my speedtests are great, no issues there, but I am wondering if there will be latency, or heavy packet loss using it?

Has anyone done any indepth reviews on the technology?

Jack basically hit the nail on the head. Your milage WILL vary. Some people love powerline setups. Others, not so much...

Only way to really know is to buy it and try it out. You can also use MoCA if you have coax jacks at the appropriate areas around your house.

 

[ViviTheMage]

I have had powerline for a few years, I moved into a new house and am wondering more about latency/packet loss though ... since my coax is in another room, I had to put my modem there, which connects to my router, so the only traffic that passes on it is the internet.

 

[JackMDS]

It was 13 years ago when I saw in the Computer show (Jabez Center NYC) the big presentation of the future. Power Line Networking, and PowerLine Internet connection.

13 years and the marketing share is in low single digit.

Why? Because it is grossly iffy in its consistency.

The consensus right now that in the future when Frig. Washing machine Toasters and similar appliances would made Auto connect to our Home Network it will be done through the power lines and we will be able to control them from our Computer/Tablets.

The above will be nice but Bandwidth, Latency, Packets losing, etc., is for No consequence in controlling appliances.

For activities that do depend on Bandwidth, Latency, and any other Networking variables, PowerLine Networking will stay in the YMMV category for long time.

P.S. The terms are.

General term Home Plug.

Power Line Networking (PLN)

Broadband Over Power Lines (BPL) for Internet connection.

 

[ViviTheMage]

So you don't like EoP? Ethernet over Power

 

[Phantomaniac]

Long ago I tried using 65Mbps powerline networking adapter to get an internet connection on the top floor of my house. I got about 3-4Mbps of actual bandwidth. Needless to say I switched back to wireless G quickly. I have a feeling that even the new 500Mbps powerline adapters are probably less reliable than typical wireless N. This isn't the fault of the technology though, its the fault of power wiring in a house (typically daisy chained over long distances). Latency is probably similar, though the edge could go to powerline networking if your router and client adapter have several walls in between them.

As far as testing, run a long running ping test between computers on either end and see how much variance there is in latency. For bandwidth testing I use iperf. Its a linux utility but you can get windows builds here: http://linhost.info/2010/02/iperf-on-windows/

 

[ViviTheMage]

thanks, i'll give that a shot!

 

[imagoon]

"EoP" is the same concept as wireless over a power line.

Single collision domain in a lot of cases and collision detection. It is pretty much a 100% that there will losses and collisions of some sort. Kinda like wireless.

Also when your neighbor fires up the microwave (if you share a transformer) you might lose a connection etc.

Mileage varies based on the condition of the medium, IE the copper in the walls.

 

[xperan.rich]

I try to avoid them where possible, but I've never had any problems using them either (eg, my home is rented, so I won't stick a load of structured cabling in for the landlords benefit), never done any huge testing as I've never had to, net connection is fine, can stream and skype, don't really need any more, and still get ~78mbps to speedtest.net - best bet, give them a go, if they work, they work, if they don't, buy some cat5

 

[ViviTheMage]

odd, I fired up my microwave last night, no issues with connection ... I still have to find time to test the throughput of it.

 

[Eug]

Long ago I tried using 65Mbps powerline networking adapter to get an internet connection on the top floor of my house. I got about 3-4Mbps of actual bandwidth. Needless to say I switched back to wireless G quickly. I have a feeling that even the new 500Mbps powerline adapters are probably less reliable than typical wireless N.

It depends. I used both in my house. I could use powerline for areas where 802.11g/n would give me no signal at all.

The Achilles' heel of WiFi is distance and walls. Powerline isn't as susceptible since it's actually wired, through those walls. The problem of course though is the serious noise on those wires, etc.

As for reliability, all the consumer webcams I tried would flake out on WiFi. Unreliable. Only wired would work 24/7. I have mine on Ethernet right now, but they would also work in some spots with powerline. Even in spots with good WiFi signal, the cameras would occasionally lockup. Not so with powerline, since the camera just sees an Ethernet signal. Even if the video stream was lost for a few seconds at a time when there was more noise via powerline, that's still better than the camera crashing on WiFi. Yeah, that's not WiFi's problem per se, but nonetheless it's a real world issue with a lot of consumer-class equipment. 24/7 streaming over WiFi seems to stress out a lot of equipment.

 

[ViviTheMage]

------------------------------------------------------------
Server listening on TCP port 5001
TCP window size: 8.00 KByte (default)
------------------------------------------------------------
[260] local 10.10.5.20 port 5001 connected with 10.10.5.117 port 60946
[ ID] Interval Transfer Bandwidth
[260] 0.0-10.0 sec 70.0 MBytes 58.7 Mbits/sec
[268] local 10.10.5.20 port 5001 connected with 10.10.5.117 port 60949
[ ID] Interval Transfer Bandwidth
[268] 0.0-10.0 sec 70.4 MBytes 59.0 Mbits/sec

tests look pretty good ... only gripe is sometimes my pings are weird... and I am plugged directly into my router, not into the powerline.

Pinging 10.10.5.1 with 32 bytes of data:
Reply from 10.10.5.1: bytes=32 time<1ms TTL=64
Reply from 10.10.5.1: bytes=32 time=7ms TTL=64
Reply from 10.10.5.1: bytes=32 time<1ms TTL=64
Reply from 10.10.5.1: bytes=32 time<1ms TTL=64
Reply from 10.10.5.1: bytes=32 time<1ms TTL=64
Reply from 10.10.5.1: bytes=32 time<1ms TTL=64
Reply from 10.10.5.1: bytes=32 time<1ms TTL=64
Reply from 10.10.5.1: bytes=32 time<1ms TTL=64
Reply from 10.10.5.1: bytes=32 time=3ms TTL=64
Reply from 10.10.5.1: bytes=32 time=10ms TTL=64
Reply from 10.10.5.1: bytes=32 time<1ms TTL=64
Reply from 10.10.5.1: bytes=32 time<1ms TTL=64
Reply from 10.10.5.1: bytes=32 time<1ms TTL=64
Reply from 10.10.5.1: bytes=32 time=9ms TTL=64
Reply from 10.10.5.1: bytes=32 time<1ms TTL=64
Reply from 10.10.5.1: bytes=32 time<1ms TTL=64
Reply from 10.10.5.1: bytes=32 time<1ms TTL=64
Reply from 10.10.5.1: bytes=32 time=20ms TTL=64
Reply from 10.10.5.1: bytes=32 time<1ms TTL=64

any other tests I can run to see if it is using the powerline when not needed?