NETGEAR XAVB5001 Powerline AV 500 Adapter Kit Reviewed

[JoeJoe]

Depends on a lot of things. Both apartments need to be within 300 ft of each other and on the same side of a distribution transformer. The only way to know is to try.

Is 300ft the realistic range of powerline networking, or is this for the Netgear units?

The Zyxel unit's specs read 300 metres.

 

[thiggins]

I erred in quoting the spec, sorry. But 300M is a pretty optimistic maximum.

 

[JoeJoe]

No problem. I got the Belkin kit today, seems to work ok, haven't done any speed tests yet.

The Zyxel software says that the speeds are Local to Remote of 120 Mbps Tx and 25 Mbps Rx, the speed LED is amber.

The Netgear software says that the speeds are Local to Remote of 157 Mbps Tx and 32 Mbps Rx.

Not sure of why the software gives different speed readings or why the Rx is so low.

I think a great design would be if they sold a 4 port switch model that included PoE for cameras.

 

[rhombus]

I got the Belkin kit today, seems to work ok, haven't done any speed tests yet. The Zyxel software says that the speeds are Local to Remote of 120 Mbps Tx and 25 Mbps Rx, the speed LED is amber. The Netgear software says that the speeds are Local to Remote of 157 Mbps Tx and 32 Mbps Rx.

Not sure of why the software gives different speed readings or why the Rx is so low.

There is nothing unusual here.

The link rates constantly fluctuate, minute-to-minute and even second-to-second, so the software reading will depend at the exact moment it was reported. In the first couple of minutes after switch-on and with initial data passing, rates can especially fluctuate a lot and then tend to settle. Note that long-term variation day-to-day might be far worse than what you quote here.

It is also not unusual for the Tx and Rx rates to be asymmetrical, especially at range, and often with a big difference like here. The Rx rate is indeed very low, but not surprisingly or alarmingly so. In the real-world, powerlines at some location pairs can just be really slow (as well as unpredictable and variable). This is the nature of the beast. They have their advantages too though, of course.

You can see my 500Mbps Homeplug Shootout here on this forum (scroll down for test results and charts).

 

[JoeJoe]

Thanks for more good information.

I'm pretty happy with the powerline network performance so far. Its going about 400 ft right now inside the house. The transformer in the yard only feeds our house. AWG#12 throughout the house.

I got them originally to extend my network to a couple of detached buildings.

If I can get at least 10 Mbps I'd be very happy, as my wireless drops out most days, because of metal roofs I think. I don't fancy digging trenches either.

I wonder if any performance issues are the difference in house wiring standards between the UK and US. In the UK they have circular circuits with hardly any breakers (in older houses) with thinner wire (higher voltage) and in the US they have straight daisy chained circuits with a few dozen breakers with thicker wire.

 

[rhombus]

I wonder if any performance issues are the difference in house wiring standards between the UK and US. In the UK they have circular circuits with hardly any breakers (in older houses) with thinner wire (higher voltage) and in the US they have straight daisy chained circuits with a few dozen breakers with thicker wire.

I'm not sure exactly what you mean. What performance issues? The big story is the difference in the ranking of the plugs compared to the Smallnetbuilder Roundup. A complete reversal for the Netgear - it aced it every place, every time in my tests.

In the UK, yes we use 230V ring mains. But we do use 2.5mm2 mains cable, and I can't remember the last time I saw a property without a circuit breaker unit.

I did get slightly lower best-case results compared to SNB even at 490Mbps+ link rates on the same double socket. If due to voltage and device operation, then yes it will be a universal US/UK thing. If due to my local mains condition, then not so.

I also captured a much greater range of performance than the SNB roundup, by doing a whole house survey: 40Mbps application throughput minimum by SNB is a big overestimation. You have just found that out for yourself I think.

On the rather large long-term day-to-day variability, SNB has not done any explicit longitudinal tests. However, it has in effect performed some: look at the Netgear results from its standalone review on one day, then look at its results from the Roundup on another day. And do the same for the Trendnet. You can draw your own conclusion! (I can tell you now, I recorded all the firmwares before and after updates, and my shootout blows any theories of firmware being responsible out of the water).

 

[scharf]

Distance

The only part that didn't make sense was, what was the distance between the points? Were they 10 ft,100ft, 1000ft? There is no correlation shown.

 

[lazna]

About HomePlug distance...

I test few powerline devices (netgear, wodaplug, tp-link, and some other cheepest chinese "noname" brands) based on Quallcom/Atheros 64xx and 74xx chipsets, but all of them with comparable results. After this testing I can say, there is no major differences between "brands" utilize same chipset. FW can play minor role only: e.g. Qualcomm 5.3.1 FW increase rates in my link by something up to 10%, but cause more rates fluctuating too, so I expect there is only some game with thresholds instead of real improovement.

Always keep few things on mind: Electrical wiring is not suitable to transfering RF signals. Electromagnetic circumstances differ any moment, as home appliances switched on/off. Wires isnt shielded and thus not durable for interference. And we can go on and on... This all practically means the larger distance than about 60-100m causes rates fluctuating and outages, because larger distance means unwanted interferences increase but level of signal itself decrease.

And one more issue should be mented here: As by Atheros, there is one chipset family for both powerline and TV coax line devices. As powerline spectrum (1.8-30Mhz Homeplug AV, 1.8-67Mhz for IEEE1901 and 1,8-86Mhz for HomePlugAV2) is wide, there is output signal limiting to about 50% (compare to chip HW maximum) by regulatory, to limit interference to other services (for example Short wave AM radio). Some specific subcarriers in the HPAV spectrum are switched-off definitely. Both of this limits not exist in the coax HomePlug, because coax is shielded and interferences with environment is negligible. So only influence in coax cables is attenuation of cable itself, which allows to use about 1000m lenght (and even more!!).

L.

BTW: I wrote HomePlug monitoring program which display more detailed informations about line quality, you can find it here http://forum.ethernetovercoax.eu/index.php?board=34.0