What am I doing wrong?

[rhombus]

D-Links sometimes drop speed way down, and everything is back to normal if you just shift one of them out and in again. No such problems with Netgear AV500?

Hi, thanks. I'm afraid I have seen this a couple of times, not a total slowdown but quite a slowdown which is also seen on the link rates, but then suddenly fixes itself on replugging. Which it shouldn't do as link rate is constantly, and I mean constantly, being revised automatically by the plug. One other 500Mbps plug I am currently testing for the shootout has twice got stuck at 5Mbps both ways despite a good link rate, and only replugging in many times fixed it.

My results will take some more time in coming, but will be 100% reliable and representative. The reason is that if you do one run on these products and try to compare, your results are null and void and could easily be inside out due to chance. These powerlines are very inconsistent moment-to-moment and also over the course of time, and by a significant amount (+/- 30% sometimes). You therefore need to do tens if not hundreds of tests in each location to gain an accurate and reliable picture over different periods of time, including fast swapping and monitoring of subsequent link negotiation. If you did a test at a different time of day, forget trying to compare. And some plugs start off fast for one minute, then slow right down as they try to settle on a link rate. Whereas others build up slowly over 5 minutes. I have seen the SNB roundup newly published, and will now try to get my hands on the Trendnet too.

 

[rhombus]

Are you a player?

I dabble

Good news for your sleep: all the 500Mbps plugs I have so far are "silent", except the Zyxel in standby.

Where abouts are you in the world?

 

[rhombus]

I can confirm my 500Mbps shootout will be Netgear vs Zyxel vs Trendnet vs Billion vs Devolo vs Solwise. I will post a new thread in the coming days.

In price, that is 1 cheap adapter, 2 mid-range and 3 high-end, and most of the 500Mbps plugs available in the UK! 3 of the adapters are included in Tim's roundup.

I have my preferred testing methods for useful and meaningul results, but will try to replicate the SNB protocol at one or two representative locations for direct comparison. I also now have gigabit LAN endpoint-to-endpoint for accurate close-range results.

You may be wondering (and myself too at this point!) why am I investing my own time and money in this? Well, simply for personal research, knowledge, a bit of geeky fun and because the winner gets a place to stay in my home! Let's see what happens

 

[StratmanX]

I dabble

Where abouts are you in the world?

Like my Strat - born, raised and reside in the USA.

Bought the Netgear XAVB5001 plug kit yesterday but not opened yet after I read thiggins review of home plugs. Was surprised to see thiggins results with the Netgear 500 Mbps plug performing opposite of your results. Now I'm confused.

I see a Belkin 1000 Mbps plug is now available! So many choices, so little money.

 

[rhombus]

The Belkin 1000Mbps is a couple of years old and based on the Gigle chipset like the Solwise 1000Mbps adapter I tested. This is the one that made the bad noise, and was much slower and unsteady than the 500Mbps adapters at range. Tim reviewed the Belkin with similar performance results. You can help your decision by ruling it out I think...

The HomePlug AV500 roundup claims homeplugs are consistent, but this is far from my experience. SNB's own tests show the Netgear plugs had results of 82, 78 and 68 first time out vs 96, 54 and 47 in the roundup. If you put the original results into the roundup, the Netgear would come top in all locations except the same room - and would beat its nearest competitor at the furthest location by 36%! Why does one set of results go into the powerline chart and not the other? I think this kind of variation needs to be controlled as best as possible - and if datasets are throwing up this kind of uncertainty, it really should be mentioned in the results and the procedure modified till you have enough data to be reliable and comparable. Please see my earlier post as to how my approach will try best to overcome this.

My shootout of 500Mbps adapters is not done yet of course, I haven't even opened all of them so do not know how it will go. But I can definitely confirm the Netgear beats the Zyxel every time in thorough like-for-like comparisons. Remember though I am at 230V in the UK, in my house, with UK versions of the plugs (and maybe different innards, as the Netgear says 230V only)...

 

[thiggins]

The Belkin 1000Mbps is a couple of years old and based on the Gigle chipset like the Solwise 1000Mbps adapter I tested. This is the one that made the bad noise, and was much slower and unsteady than the 500Mbps adapters at range. Tim reviewed the Belkin with similar performance results. You can help your decision by ruling it out I think...

Broadcom acquired Gigle last December. Perhaps Broadcom is preparing to enter the powerline market. But no sign of it right now. I would leave "Gigabit" powerline products on the shelf for now.

The HomePlug AV500 roundup claims homeplugs are consistent, but this is far from my experience. SNB's own tests show the Netgear plugs had results of 82, 78 and 68 first time out vs 96, 54 and 47 in the roundup. If you put the original results into the roundup, the Netgear would come top in all locations except the same room - and would beat its nearest competitor at the furthest location by 36%! Why does one set of results go into the powerline chart and not the other?

The charts can show only one set of data per product. I chose the more recent set of data to ensure that I was testing all products the same way.

When I call powerline throughput consistent, I mean its short term variation. Compare the IxChariot plots for powerline products vs. any wireless networking plots and powerline delivers much more consistent throughput.

I think this kind of variation needs to be controlled as best as possible - and if datasets are throwing up this kind of uncertainty, it really should be mentioned in the results and the procedure modified till you have enough data to be reliable and comparable. Please see my earlier post as to how my approach will try best to overcome this.

There is no attempt to hide data or mislead in these results. I have neither the time nor resources to perform long term performance tests on products. Even if I did, non-Ethernet networking technologies produce different results in different environments. The results in the charts are useful for relative performance comparison. They are not meant to represent the highest performance that a product will produce.

 

[rhombus]

When I call powerline throughput consistent, I mean its short term variation. Compare the IxChariot plots for powerline products vs. any wireless networking plots and powerline delivers much more consistent throughput.

Not in my testing so far and not in a lot of your own IxChariot plots for powerlines - they both look as bad as each other really. Powerline drifts considerably over time as well, which wireless doesn't do nearly as much.

You also do have very poor wireless dropout performance in all your tests, which I never see in my environment and equipment. Of course wireless is not perfect, but I have used it every day for many years and monitor live video streaming as part of my job - never once have I had a dropout in the stream.

If you pass any traffic whatsoever other than the testing stream using IxChariot, you get those dropout spikes on the charts.

There is no attempt to hide data or mislead in these results. The results in the charts are useful for relative performance comparison. They are not meant to represent the highest performance that a product will produce.

Woah, of course I wasn't saying you were intentionally hiding or misleading anything!

I am not capturing highest performance, but precisely what you say - relative performance. I don't see how your tests with single-runs capture that.

It's a lot of effort to write up results on a site with advertising, I think I'll pass on it now.

 

[StratmanX]

Guys:

I wasn't trying to stir up a shirtestorm here. I saw two different results and I got to wondering. Never did I think anyone was being misleading.

I think I understand better why the results are different. Correct me if I am wrong. Re-reading thiggins review it appears he was using 1 or 2 test runs per plug per location, while rhombus was using multiple test runs to determine performance. Thiggins results are a snapshot in time while rhombus' was a whole movie, so to speak. Also, there is the usual issues of different setups and testing conditions (electrical wiring) that must be traversed which can result in different data.

I do not want to see anyone not contribute because of a misunderstanding on my part. I appreciate what you two have done and provided freely for my, and others, understanding. I truly hope to read more from both of you on this subject.

 

[rhombus]

Hatchets are buried, and exhaustive testing has finally been completed.

I have a busy week ahead, but expect a writeup in the next few days

 

[StratmanX]

Excellent!!!

 

[rhombus]

My Test Outcome

My little test turned into a big group test, and it's just impossible to condense that into a forum post. I did begin to write up a review though, here's a tiny summary in the form of excerpts:

Netgear XAVB5001
Zyxel PLA4205
Trendnet TPL-401E2K
Billion BiPac P108
Devolo dLAN 500 AVmini
Solwise NET-PL-500AV-PIGGY

Performance

At typical locations, the Netgears consistently negotiated a higher link rate, with a corresponding throughput advantage in the order of 20-35% over their nearest competitor. A clear pattern and order of performance emerged for the rest too, and it was remarkable how this remained consistent across most sockets. There was a second best group of the Zyxel and Trendnet, closely followed by a third best group of the Solwise and Devolo. The Billion was a distant last - always a slow, erratic and dreadful performer.

At the best location, where the going was good, the plugs all performed identically (and I mean identically) with single stream, simulatenous up/down and four stream tests. Single stream was 100Mbps measured throughput. The exception was again the Billion which was a little behind even here.

At the worst location, where the going was bad, the plugs all performed just as badly but with the Netgear even here eeking out a small advantage, and the Billion still much worse than the rest. Single stream was 24Mbps here.

Short-term throughput was steady without dropouts, just like wireless can achieve with a steady signal.

Vs Wireless

In terms of throughput around the house, there is no clear winner. At the Typical Location (what I named one specific location for testing and with recorded IxChariot plots), the powerlines would easily beat any wireless because the path has many walls and a ceiling to penetrate. Yet conversely, at the Worst Location, wireless would easily beat the powerlines because the path it is just a short distance across the hallway. In general, the spread of typical real-world speeds probably overlap quite closely.

Powerlines have the big advantages of no deadspots and decent throughput even in the furthest reaches of a house, and no interference in the airspace.

Long-term performance however was remarkably inconsistent. On some days, at some times, performance would just be mysteriously poor. There was nothing I could do to control it. At my intended location, the performance on a good day averaged 60Mbps, but on some days was dipping below 30Mbps. In the end, this led me to abandon my planned use of powerlines.

Audible Noise

Someone specifically asked about noise. For all intents and purposes, all the plugs were silent with two notable exceptions:

The Zyxel made a high-frequency fizzing and whining noise in standby mode; the nature and spectrum of the noise might make it distracting if you need total silence.

The Trendnet made a louder high-pitched whistle noise in all modes; although still quiet, it could be heard 3m away over a laptop fan and therefore might easily be distracting. I noted that one of the plugs was always significantly louder than the other.

Software

Trendnet, Billion and Solwise all provide the Atheros OEM utility for Windows. They are identical, with just simple branding replacement. You can use the software to: check link rates to and from each plug, display and update firmware verions, create separate networks on the same mains, see MAC addresses of the plugs, name the devices and enter a manual encryption key.

Netgear, Devolo and Zyxel all provide their own custom utility. You can’t mix and match these utilities with plugs from other manufacturers : the Devolo utility won’t work at all, and with the others you will find features missing as well as incorrect or blank firmware reported. The Netgear and Zyxel plugs also have extra, unique features only accessible from their respective utilities and so you should stick with these if you have those plugs.

The Netgear utility was my favourite of all and unlike the others, it offers a Mac version and doesn’t annoyingly install WinPCap in the background. The Atheros utility was good and did its job fine. I found the Devolo a real pain to use however, even for such a piece of simple software.

 

[StratmanX]

Long-term performance however was remarkably inconsistent. On some days, at some times, performance would just be mysteriously poor. There was nothing I could do to control it. At my intended location, the performance on a good day averaged 60Mbps, but on some days was dipping below 30Mbps. In the end, this led me to abandon my planned use of powerlines.

There lies the money shot. Sounds like you wanted enough throughput for Bluray-sized data amounts and the Powerline products didn't consistently deliver that quantity of throughputs. Is this due something inherent in the Powerline technology no matter how clean the electrical wiring is or because of external interference from devices on the circuit or the vagaries of your electrical wiring itself? In other words, no matter who uses the Powerline products in whatever location, the throughput will mysteriously significantly drop temporarily?

Great review, rhombus. Thank you very much for all your work. Fantastic information.

 

[rhombus]

The Contenders




Typical Location - Downlink Results

 

[rhombus]

Results Table
Mbps, IxChariot, Throughput Script

 

[rhombus]

Is this due something inherent in the Powerline technology no matter how clean the electrical wiring is or because of external interference from devices on the circuit or the vagaries of your electrical wiring itself? In other words, no matter who uses the Powerline products in whatever location, the throughput will mysteriously significantly drop temporarily

All the plugs fluctuated with each other at the same time, and I tried unplugging all appliances on the mains with no change. I dont believe in voodoo, and I'm also quite sure on a clean artificial mains in an Atheros test lab they would all remain steady day in, day out. My house is 10 years old, with a professional, certified job done of the mains wiring and consumer units, so they should have no issue.

In short, I determine it is the vagaries of the mains to blame.

How prevalent or isolated this issue is I have no idea. I am far from a mains expert. Multi-dwelling, multi-location tests would be nice, but were beyond my means here!

 

[RogerSC]

You might consider MoCA, if you have coax in your walls for cable TV. This is working well for me, and seems to have fairly consistent throughput. Aside from the power glitch we had the other day, which caused me to have to reboot several pieces of network hardware. I do have my cable modem and main router on a UPS, so they didn't see the power glitch. But everything's back to normal now.

I was seeing about 85Mb/s when I measured the throughput using the MoCA. Now I'm using it as input to an AP, and the Roku box comes off that. Has really helped the wireless strength in the part of my house that didn't have strong wireless, as well as providing good streaming speed for the Roku box.

 

[VAT]

Thanks for the test! That, I guess, was as good as powerline test gets
Funny thing, if you take into account all thiggins's tests too, it seems that daily variations completely negate the vendor difference.
Meanwhile I got my XAVB5501 kit. I can't do a direct comparison with dlink kit, since I returned it because of speed dropping glitch (it was really a glitch, since the speed dropped 10 times and could be restored instantly by resetting the adapter).
However, if the daily difference is omitted, I can say that Netgears in best case get around 100 Mbps, 40 more than older 200Mbps Dlinks. That said, in my "real world" scenario, which puts the adapters where I actually need them, both Dlinks and Netgears achieve 45-50 Mbps. Netgears are only a little ahead, 5-7 Mbps.
In my tests I did find that performance varies day to day, although not as greatly as in rhombus case - around 10 Mbps difference for me, on Dlinks. I didn't test the Netgears on different days yet.
I guess, glitch-free HD streaming is still under a big question mark.

 

[rhombus]

it seems that daily variations completely negate the vendor difference.

Absolutely not! Not in my tests. The long-term variation over time affected all plugs equally, with the vendor differences intact.

Meanwhile I got my XAVB5501 kit... I returned it because of speed dropping glitch (it was really a glitch, since the speed dropped 10 times and could be restored instantly by resetting the adapter).

You described in an earlier post this also happens with your 200Mbps Dlinks. I did see something akin to this with all the plugs - they tend to be sticky towards lower rates as performance drifts over time. So when a good day came back around, a reset or replug was needed and instantly restored the best performance.

I can say that Netgears in best case get around 100 Mbps, 40 more than older 200Mbps Dlinks. That said, in my "real world" scenario, which puts the adapters where I actually need them, both Dlinks and Netgears achieve 45-50 Mbps. Netgears are only a little ahead, 5-7 Mbps.

This was also SNB's result of 500Mbps vs 200Mbps at range, but it definitely was not my result. My tests capture an even bigger spread of performance over an exhaustive number of trialled sockets. I tried out 3 different 200Mbps plugs during the course of my main 500Mbps shootout, using two different manufacturer chipsets. Throughput for all of them was about half that of the 500Mbps plugs, and this was maintained at all locations and sockets I tried (and I really do mean all) almost exactly. In a lot of socket pairings in my house, this meant throughput was limited to just 11Mbps.

 

[VAT]

rhombus

You described in an earlier post this also happens with your 200Mbps Dlinks. I did see something akin to this with all the plugs - they tend to be sticky towards lower rates as performance drifts over time. So when a good day came back around, a reset or replug was needed and instantly restored the best performance.

I returned the DLinks, not Netgears. The behaviour was different to what you described - they just sometimes suddenly dropped to about 0.5 MBps, not over time. Sometimes even right after the power-up.

I guess, all the experience just confirms the well-known fact: homeplugs depend a lot on the electrical network parameters. So try before you buy. Or buy from a place with dependable return policy
They're easy to try too, since there are no advanced configuration options for sleepless testing sprees
In the end, Netgears work for me, since I'm satisfied with 15-20Mbps FHD video.

 

[rhombus]

Sounds like you wanted enough throughput for Bluray-sized data amounts

I'm satisfied with 15-20Mbps FHD video.

Hey guys, an important point.

When studios do a high-quality online trailer encode, or your friend re-encodes a video of theirs to send across the internet, they usually preserve the peak bitrate for quality. This can be many multiples of the average bitrate, and is typically around 30-40-50Mbps. It can even exceed the max bitrate and be a legal encode, I have seen such real material topping 70Mbps.

Also, you cant translate the numbers directly from a download test or specific IxChariot script to a different real-world application or for different devices. And to further complicate, Powerlines do not behave as pseudo-cables. I found that chances are, to a device expecting a real cable connection such as a blu-ray player or console, they will stream a bit slower in practice.

For this reason, my 30-60Mbps on an IxChariot throughput script, although it sounds like thoroughly good throughput, was not sufficient for my personal quest to stream the video content I wanted to a PS3.

Also, my secondary quest to add a 3rd homeplug and use it between router and access point was scuppered by the really quite slow and limiting throughput and high running costs. This made my mind up to ditch the powerlines.