Mar 13, 2013

TP-Link TL-PA511 KIT Review

Today I’m doing a review of the TP-Link AV500 Gigabit Powerline Adapter starter kit. For those of you that aren’t aware of the powerline/homeplug standard it basically allows for internet (and LAN) access through your electrical circuitry within your home.


I’m also going to see how it stands up to a Gigabit connection and an older set of 85mbps powerline/homeplug adapters. I have been using Solwise 85mbps homeplugs for a few years now (they were one of the early models on the market) and one of the things that really bothered me about them was the very slow transfer rate of files from my NAS to my PC, download speed from the NAS was around 700kb/s and upload roughly the same. Considering I Get around 9mb/s on WiFi when I am back in the other side of the house (where WiFi reaches) this was quite a poor result for me. I have recently purchased a couple of TP-Link TL-WA901ND access points in order to upgrade the WiFi in both sides of the house and the homeplugs were the only thing holding me back in this half of the house where I have my ‘Zone 2′ AP running.


TP-Link offer 200mbps, 500mbps and wireless versions of their powerlines. I am reviewing their top of the range 500mbps gigabit adapter today. The product comes in their familiar green-coloured packaging with TP-Link’s 3 year warranty clearly stamped on the top of the packaging.

The TP-Link TL-PA511 KIT comes with the following:
  • 2x AV500 Gigabit Powerline Adapters
  • 2x CAT5 cables (ethernet)
  • Resource CD
  • 2x Purchasing Guides
  • Quick Installation Guide

One thing I did notice was that CAT5 cables were supplied instead of CAT5e, this does mean that despite the adapters being equipped with Gigabit ports the max transfer rate we could see with the provided cables would be 100mbps (0.1gbps). I assume the reason is that, as with every other homeplug/powerline on the market, you will never see a full 500mbps because of varying quality in the electrical circuitry in your home, distance, intereference etc etc. All of that should be revealed with testing!


Now that we have the box contents out of the way, let’s get into the adapters themselves and testing.

The adapters themselvs come in white, with ventilation holes on either side of the units. When in use they get warm, but nothing to be concerned about. Each adapter has three LEDs, starting from the top and working our way down to the bottom we have the power LED, the powerline LED (green = data rate is more than or equal to 80mbps, orange = data rate is between 48mbps and 80mbps, red = data rate is less than or equal to 48mbps) and the ethernet LED which flashes when data is being transferred.

Pair button located next to the Gigabit ethernet port
TP-Link states that the maximum range for their homeplugs is 300 metres within your home. As previously mentioned, each unit comes equipped with a Gigabit ethernet port. The link between the two units is encrypted with 128-bit AES encryption. Simply plug one unit near your modem/router/switch and make sure the power is switched on, then press the pair button (located next to the Gigabit port) for 1 second. Next head over to wherever you intend to plug in your second homeplug unit, plug it in, ensure it is powered, connect an ethernet cable up to the device(s) that require internet and press the pair button on the second unit (for 1 second).

Pair button located next to the Gigabit ethernet port
Allow 60 seconds for the devices to pair together. Note: You do not need to connect ethernet cables up at this point, it may just be easier to do so incase you intend on placing one or both of the devices behind furniture etc. The units are rather energy efficient, they conserve up to 85% power when inactive (no data transfer) – this is particularly useful at night when very little data is likely to be travelling through your powerline/homeplug adapters. In my case, no data whatsoever will travel down the units at night as my phone (and any other device that receives e-mail, updates, notifications etc) are always in the WiFi Zone 1 during the night and not in the extended Zone 2.

Methodology:

All runs were repeated 3 times with the average result used (providing no erroneous data was collected)

All runs were recorded using LAN Speed Test by Totusoft (http://www.totusoft.com)

Both powerline/homeplug adapters were reading/writing data to a Gigabit NAS (Synology DS210J) however in order to avoid a bottleneck when performing the Gigabit CAT6 cable test I ALSO opted to run a LAN Speed Test between two computers as my NAS will not read/write much faster than 30mb/s and 20mb/s respectively. (240mbps and 160mbps) but a Gigabit CAT5 cable should be able to read/write much faster than this.

For the 500mbps TP-Link adapters: One test (3 runs) with both adapters side by side, one test with both adapters approx. 25 metres away from each other, one test with both adapters approx. 40 metres away from each other.

For the 85mbps Solwise adapters: One test (3 runs)with both adapters side by side, one test with both adapters approx. 25 metres away from each other, one test with both adapters approx. 40 metres away from each other.

For the direct Gigabit connection: One test (3 runs) with one 15m CAT6 cable running from one computer to the Gigabit switch and then to the NAS (both the computer and the NAS have a Gigabit NIC).

For the direct Gigabit connection between two computers: One test (3 runs) with one 15m CAT6 cable running from one computer to the other (both computers have a Gigabit NIC).


TP-Link adapter located in loft for 40m testing

Testing:

Hardware used: Belkin CAT6 cable (15m), 1x Synology DS210j, Linksys Gigabit Switch, 3x PC’s (i7 930, ASUS R3E // Xeon E5-2630, EVGA SR-X // i3 2120, MSI P67A-GD53).

Immediate Observations: It appears that the Gigabit CAT6 cable was able to read/write faster than what most are reporting with their Synology DS210j NAS, still the figures aren’t too far out and I reckon that most of the examples of max read/write that I found online were from users installing green/eco 5400rpm drives whereas I am using a 7200rpm drive in my NAS. There could still be a SATA controller limitation with regards to read/write speeds on my NAS though.

Without further ado, here’s the results:

Please bear in mind that all results are averages of 3 runs, however each LAN Speed Test run comprised of a MINIMUM of 10 packets (for the large byte count runs) and a MAXIMUM of 1000 packets (for the small byte count runs), this means that the actual average is the number of runs multiplied by the number of packets. I decided to do it this way to ensure that discrepencies/erroneous data was kept to a bare minimum as powerline/homeplug networking can be adversely affected by as little as appliances being switched on/off around the home.

Results:


Graphical Representations (Click to Enlarge):


First of all we can see that there is next to no difference between the two adapters when it comes to reading and writing very small files (1000 byte test performed). All of the results obtained are well within the margins of error. Things get more interesting when we start to test read/write speeds of 10,000,000, 100,000,000 and 300,000,000 byte file sizes. We can see that in each test the TP-Link adapters outperformed the older 85mbps Solwise adapters even when the TP-Link adapters were positioned 40 metres apart and teh Solwise adapters were positioned next to each other.


It is only at this point that we start to realise just how far the technology has progressed and I expect to see further improvements in longer distances with the release of the AV2 standard back in January 2012. It shouldn’t be long before we see some AV2 devices out there on the market! Writing speeds (to the NAS) on the TP-Link was consistently slower, which is to be expected as that is upstream. Interesting to note was that the Solwise adapters had a faster upstream than downstream when positioned next to each other, although by only 2-4mbps with all 4 packet sizes testes and this is still well within the margins of error.


In this test, I wanted to try and show the limitations of the NAS. We can see that writing a test packet from one computer to another computer (both running standard 7200rpm drives) the results strongly outperform those obtained from writing a test packet to the NAS. This is a limitation of the controller within the NAS.


One interesting thing to note was that the NAS caught up with the computer to computer transfer speed when performing the read test, which is rather odd because the computer it was uploading the test packet to was running the same hard drive, and a 6 core CPU instead of a dual core i3. Nonetheless it’s interesting to note that the NAS performs better at pushing data to your computer than receiving data from it. Both of the result sets shown in the two graphs above were collected using a standard 15 metre CAT6 cable. 

I didn’t use the Home Plug adapters for this, I just wanted to put these results out there to demonstrate that the Home Plug / Powerline standard still has a long way to go before being a viable replacement for CAT5e/CAT6 within your home IF this option is available to you. Obviously if you can’t run ethernet cables in your home’s walls (like is the case for me) then the Home Plug standard is the ideal solution for you and with the new AV2 standard already released to developers earlier this year, we should see another boost in performance, however we still do have to bear in mind that this will likely still be very much dependant on the quality of your internal electrical circuitry throughout your home.

Additional Testing:

These tests are very good at giving us a broad overview of how a particular device works however I wanted to show you just a couple of screenshots of downstream and upstream speeds to and from my NAS through these TP-Link adapters (40m apart).



For comparison, the older Solwise 85mbps adapters had an upstream/downstream of around 600-750kb/s, and transferring data wirelessly from within Zone 1 of the house (wireless AP connected directly to Gigabit switch) gives me an upstream/downstream of around 9.5-11mb/s.

This makes the TP-Link kit roughly 6x faster when performing an upstream operation, and 10x faster in downstream.


Just out of curiosity I also tried the TP Link utility software provided with the unit, it also states an expected throughput of around 80mbps between the two units, the same reading as is displayed by the green powerline LED on the unit itself.

Concluding Thoughts:

This kit provides a very stable connection (tested for over a week now with no loss of connection) for a very affordable price tag, (http://www.amazon.co.uk – £52.70). I was particularly impressed by its ability to maintain a strong connection and upstream/downstream speed when connected roughly 40 metres apart especially considering that our internal electrical wiring isn’t the most modern around.

One thing that did have me scratching my head were the data rates than the powerline LED claims to be getting. I am getting a solid green even when the adapters are plugged 40 metres apart yet my LAN speed tests show me getting 23-37mbps which would indicate that I should be getting a red LED on my powerline (Red = data rate less than or equal to 48mbps). I realize that it could also be the wireless between my TP-Link WA901ND access point and my USB dongle but this adapter and access point has shown to be able to push and pull as much as 9.5-11mb/s across so I very much doubt that.

The AP is connected to the second TP-Link adapter with a CAT5e cable. At the end of the day it doesn’t really bother me as i’m really pleased with teh speeds I’m able to get already, and I realise that LAN Speed Test isn’t the be all end all of network speed test software out there, having said that I wouldn’t pay too much attention to the LED’s although it does make the unit look nice! I also liked the idea of the 128bit AES encrption on the units.

One thing that I mentioned at the start is that I’d consider changing is the inclusion of CAT5 cables, they can only transfer 100mbps max whereas CAT5e/6 can transfer up to 1000mbps and despite the units not actually able to surpass the 100mbps limit of those CAT5 cables, the unit itself has a gigabit switch, from a marketing point of view I’d include CAT5e cables, especially considering that the price difference between the two is a matter of pennies even for a customer only buying one cable so I’d imagine that bulk buying cat5e over cat5 really wouldn’t set you back by much more.

0 comments:

Post a Comment