Question Is a speed rate of 150 mbps okay for a TP-Link AV1200 WPA7517 powerline adapter ?

sanslash332

Honorable
Oct 27, 2017
10
0
10,510
Hello Guys,
I recently bought a TP-Link AV1200 WPA7517 Powerline Kit: https://www.tp-link.com/en/home-networking/powerline/tl-wpa7517-kit/ to get a more stable internet connection in a room where the WiFi suddenly drops sometimes and is very annoying (apparently caused by the WiFi disrupters in a jail that's near my street).

Well, I know that the expected performance of these things its of the 40-50% in best cases, and less in poorly situations. But, after do a lot of testing I got a stable range of speeds between 110 and 150 mbps, regardless of which outlets i've used.

The nearest setting that I got was put the first device on the same outlet of the router (the outlet just have two plugs, so I cant put the second device on that location) and the second device on other outlet, but in the same living room. I cant confirm that is the same circuit, but doing the same room I think that.
Well, the performance on that scenario its the same that when I've put the second device on my room (A few meters away of the living room)
The result is 150 mbps at best cases, and 110 at the poor cases.

Is that is the best performance I can get with a AV1200 device? Or is it faulty ?
And is it normal that the TPPLC app shows that the speed between the two devices fluctuates between 450-520mbps? I didn't see anything like that on my tests.
I repeat; I know that the "av1200" label its just marketing, but. 150mbps is only 12.5% of 1200. That is the regular performance that these kind of devices do?
Thanks so far for your answers :)

Just for clarity; the tests were done just doing speed checks through fast.com or speedtest with my laptop and with my phone to test to the wi-fy. The results are the same. My internet is 600/100 mbps, and through WiFi 5g I got 420mbps in the best cases. Through a direct LAN connection to the router, the best values are around 720 mbps.

When the secondary PC with WiFi is connected directly to the router, it got on the WiFi connection the same performance of the router (420 mbps) so its clear that something is wrong on the wires, or the PC system itself.

Thanks so far!
 
Solution
That is about the "expected" results. Even the ones that have a 2000 number on them do not get much more. TPlink in particular like to be deceptive about this. They used to call there unit based on a standard called av-500 av500 but when a newer form of powerline came out they just change the name to av600 even though the real standard is av2-600 which it not what the unit is. They did not even change the part number. Those units though do not even get 100mbps.....kinda why they only have 100mbps ethernet ports so you know the manufacture knows they can come nowhere close to the number on the box.

Wifi is almost as deceptive in their claims, adding the speeds of the 2.4 and 5g radios together even though you can't do...
That is about the "expected" results. Even the ones that have a 2000 number on them do not get much more. TPlink in particular like to be deceptive about this. They used to call there unit based on a standard called av-500 av500 but when a newer form of powerline came out they just change the name to av600 even though the real standard is av2-600 which it not what the unit is. They did not even change the part number. Those units though do not even get 100mbps.....kinda why they only have 100mbps ethernet ports so you know the manufacture knows they can come nowhere close to the number on the box.

Wifi is almost as deceptive in their claims, adding the speeds of the 2.4 and 5g radios together even though you can't do that. They are going to try something stupid like that in wifi7.

Then again both these technology would call a 1gbit ethernet cable 2gbit because they add the transmit and receive together but ethernet can actually send 1gbit and receive 1gbit at the same time.

The only technology I know that is closer are the newer moca units. They can run a ethernet cable at full 1gbit speeds....both up and down at the same time if you buy the so called 2.5g units.

Key here is what are you actually doing with your network connection. The huge bandwidths you see now days mean very little. Web surfing uses very little. Video streaming might user 25mbps if you watch netflix 4k. Only file downloads can use the total bandwidth and it depends how many hours in a day you do that. The key reason people use powerline over wifi is that it is much more stable, the latency is very consistent so online games run much better.
 
  • Like
Reactions: sanslash332
Solution

sanslash332

Honorable
Oct 27, 2017
10
0
10,510
That is about the "expected" results. Even the ones that have a 2000 number on them do not get much more. TPlink in particular like to be deceptive about this. They used to call there unit based on a standard called av-500 av500 but when a newer form of powerline came out they just change the name to av600 even though the real standard is av2-600 which it not what the unit is. They did not even change the part number. Those units though do not even get 100mbps.....kinda why they only have 100mbps ethernet ports so you know the manufacture knows they can come nowhere close to the number on the box.

Wifi is almost as deceptive in their claims, adding the speeds of the 2.4 and 5g radios together even though you can't do that. They are going to try something stupid like that in wifi7.

Then again both these technology would call a 1gbit ethernet cable 2gbit because they add the transmit and receive together but ethernet can actually send 1gbit and receive 1gbit at the same time.

The only technology I know that is closer are the newer moca units. They can run a ethernet cable at full 1gbit speeds....both up and down at the same time if you buy the so called 2.5g units.

Key here is what are you actually doing with your network connection. The huge bandwidths you see now days mean very little. Web surfing uses very little. Video streaming might user 25mbps if you watch netflix 4k. Only file downloads can use the total bandwidth and it depends how many hours in a day you do that. The key reason people use powerline over wifi is that it is much more stable, the latency is very consistent so online games run much better.


Thanks @bill001g For your reply :)


With that I can be calm about my units that they are OK and don't have problems :)


My main purpose to get these things is for get a backup connection when my wi-fy drops. To get more stable connection that will not be affected for these disruptors so, for that, they are accomplishing that task.


But, a final question, exist a method to fusion both connections on windows?
So with that I can get the speed advantages of 5g wi-fy when it works fine, but automatically fallback into Ethernet when wi-fy is unstable. The regular behavior of windows is that it keeps trying to use wi-fy until it disconnects or you turn of the adapter, but exist a other method (built into windows) that do a better management of both connections at the same time?
Googling I found the solution of put the network metric of the both devices in manual mode and both set to 15, but I don't know if it really works.
On my task manager I cant see any activity for both devices; for both it says 0kbps for download and upload, but internet works fine :)
So well, again, thanks for your answer!
 
I have no idea what happens when you set the metric the same.
The mac addresses are different and they likely have different IP addresses. That means something like a open session to a game server can not just move over because the router will create a different NAT entry because it thinks it really is a different machine. To work this would have to do it more like linux does and use the loopback IP and then the "router" would have to actually have real route entries so it knows that that IP is on both paths.

There is no simple way to do switching. The wifi never really disconnects and unlike ethernet you can't even see error counts because wifi uses data retransmission to correct them. You would have to have something that could measure the average latency and if it saw spikes change over.

There used to be programs like forcebind ip that let different aps use different interfaces but it does not work that well anymore. Microsoft constantly breaks it with new patches and you get stuff like chrome that does not use a single processid anymore.

Realistically the best way to do this would be to use the powerline for everything and when you needed to do a big download just disable the ethernet.
 
  • Like
Reactions: sanslash332