Tech Blog‎ > ‎

Powerline Performance Test - TP-Link 500Mbps TL-PA4010KIT

posted Dec 26, 2013, 12:50 PM by Victor Zakharov   [ updated Dec 29, 2013, 7:51 AM ]
This is just an experiment I've planned for Christmas, to verify if powerlines are any good, a couple years after they became popular on the market.
So I picked a rather new adapter kit, which was supposed to deliver 500 megabits over a 100 Mbit connection, which sounded kind of interesting.
Namely this one: TP-Link TL-PA4010KIT (AV500 Nano Powerline Adapter Starter Kit).
It is advertised as "high-speed data transfer rates of up to 500Mbps", so let's see what the real figures would be.
As a side note, I paid 45 Canadian dollars before tax for the kit, which includes two adapters and two good quality cat5 cables (see below why I stressed that).


I did several TCP network speed tests using Passmark Performance Test v8 b1027, latest version at the moment of writing (no affiliation with these guys).
The measured PCs (later being referenced as A and B respectively) are an Acer 1810T laptop (year 2009, I believe) and my custom made PC, using built-in LAN, motherboard is P8Z68-V PRO/GEN3.
A is obviously a wandering asset, so I am going to stick it in every outlet I find, and B just sits there stationary in the bedroom, connected to 8-port Gigabit switch from Trendnet (good one, with a metal case).
Both computers are said to be capable of going as high as a GBit/Sec, but I don't think I've ever seen any of them go higher than 800 Mbit/Sec in different setups.

The first two tests were done without a powerline adapter at all, just running over a gigabit network, in order to test network cables and the infrastructure in general.
(1a) uses a cable, that would later be hooked up between A and a wall AC.
(1b) is a cable that's between B and a wall AC respectively.
In both of these tests, a cable goes to the aforementioned Gigabit switch, logically from A to B.

Other test titles are pretty much self-explanatory. So, here it goes:

Setup

Average, Mbit/Sec

Minimum, Mbit/Sec

Maximum, Mbit/Sec

1a. Gigabit network over cat5 (no powerline)

571.2

514.0

624.7

1b. Gigabit network over cat6 (no powerline)

509.6

416.6

588.2

2. Powerline – same socket

37.31

32.03

40.90

3. Powerline – same wall, different sides

34.03

29.31

36.38

4. Powerline – different walls, bedroom to main

28.75

18.87

31.13

5. Powerline – different walls, bedroom to kitchen

34.01

29.40

37.75

6. Powerline – opposite walls, bedroom

35.77

29.70

39.84

7. Powerline – opposite walls, bedroom, through UPS

34.47

27.53

39.57

8. Powerline – different walls, bedroom to bathroom

36.36

33.66

40.30


Points of interest:
- The average speed is hovering in the range of 30-40 Mbit/Sec, which is okay for most home needs.
- The speed is nowhere near the advertised 500Mbps. Let alone that, I would expect "same socket" setup to yield close to 100 Mbit connection (90+ would be okay).
- Cat5 cable included with the powerline kit performs slightly better than the Cat6 cable I purchased some time ago.
- Running through a UPS does not affect the setup (no considerable performance drop, the speed difference I got could be a measurement error).
Some people say it won't work at all, so now you have proof of the opposite.

Screenshots from all performance tests I did are included below, so you can double check the numbers.

If you have any relevant tests, which prove powerline adapters to be better than above, please feel free to post an article of yours and let me know.
Provided that it looks legit, I may consider linking your article from here. People need to know the truth, and the only way to get it is by experiment.
ą
Victor Zakharov,
Dec 26, 2013, 12:50 PM
ą
Victor Zakharov,
Dec 26, 2013, 12:51 PM
ą
Victor Zakharov,
Dec 26, 2013, 12:51 PM
ą
Victor Zakharov,
Dec 26, 2013, 12:51 PM
ą
Victor Zakharov,
Dec 26, 2013, 12:51 PM
ą
Victor Zakharov,
Dec 26, 2013, 12:51 PM
ą
Victor Zakharov,
Dec 26, 2013, 12:52 PM
ą
Victor Zakharov,
Dec 26, 2013, 12:52 PM
ą
Victor Zakharov,
Dec 26, 2013, 12:52 PM
Comments