86U vs 68U - Wifi Throughput

[doczenith1]

I picked up an 86U last week and finally got around to doing some testing of the wifi throughput vs my 68U. The test parameters are listed in the screenshot. Both phones were restarted when switching routers and the wifi connection forgot and reconnected. The 86U really shined when the phones were at a longer distance from the router. I was kind of surprised at the results after what I saw when I compared the 68U to the 3100 a year ago. I'm a little skeptical of the Ookla test/results. 719 Mbps seems high over a 866 link. Just noticed a typo. 68U is running 380.68_4 firmware.

 

[maxbraketorque]

Those are some nice improvements. Perhaps I will pick up and AC86U and compare to my AC3100.

 

[Keenan]

If I'm reading your data correctly, it seems to indicate that the 86U does very well. Is that a correct assessment?

Thanks

 

[doczenith1]

If I'm reading your data correctly, it seems to indicate that the 86U does very well. Is that a correct assessment?

Thanks

Yes, the 86U is performing up to 3 times better (300% in the summary section) with download speeds than my 68U when the devices are not that close to the router. At close range it's between 1.2 and 1.5 times faster. I don't know if there is any range improvement with the 86U as I would need to go outside to test the max range and it's currently 20 degrees here. Maybe next week when it is supposed to hit 40 degrees F I'll get ambitious and test the range and throughput on each router. And as mentioned in other threads the 86U is a beast when it comes to vpn throughput.

 

[Keenan]

Thanks. Looking forward to those outdoor results.

 

[maxbraketorque]

My AC86U was waiting at my door when I arrived home from vacation today. I am in the process of performing a 3-way test between the AC68U, AC3100, and AC86U wifi quality. I've got the data on the AC3100, and I'm in the process of gathering data on the AC86U. Its beating the AC3100 in some areas but not all. One other tidbit is that the AC86U has greater moment-by-moment variability in signal quality and transfer speeds.

 

[maxbraketorque]

My AC86U was waiting at my door when I arrived home from vacation today. I am in the process of performing a 3-way test between the AC68U, AC3100, and AC86U wifi quality. I've got the data on the AC3100, and I'm in the process of gathering data on the AC86U. Its beating the AC3100 in some areas but not all. One other tidbit is that the AC86U has greater moment-by-moment variability in signal quality and transfer speeds.

Well this is turning out to be a less decisive test than I expected. I haven't completed testing yet, but another observation is that my AC68U just beat my AC3100 and AC86U in near field AC tests. The AC68U speeds are about 0-20% higher than the AC3100 and AC86U depending on DL/UL and whether its a single large file or a group of smaller files. And to be sure, I've done some selected repeat testing to verify. This is pretty unexpected and is much different from my earlier testing about a month ago when I picked up the AC3100. At that time, the AC3100 was more 30% faster than the AC68U in near field AC tests. What's different this time is that I did a factory reset on the AC68U. In near field 2.4 GHz N tests, all three are roughly equal with no obvious across-the-board winner. Anyhow, I'll create a new thread this evening after everyone has gone to bed so that I can tinker with wireless and finish the testing.

 

[Keenan]

As I currently have the AC86U but I'm not happy with its faulty network map issues so I've been looking at the AC68U(1900P) and the AC3100. I have until the end of Jan to return the AC86U if I decide to go with a different model.

Your testing is of great interest to me, thanks.

 

[sfx2000]

I'm not happy with its faulty network map issues

Network Map has been a ball of hate since, well, forever... I wouldn't worry to much about it.

If the rest is good...

 

[Keenan]

Yes, I'm learning to live with it. Been looking for a local support group...

 

[sfx2000]

Yes, I'm learning to live with it. Been looking for a local support group...

grant me the serenity to accept the things I cannot change...

 

[Keenan]

Indeed!

 

[doczenith1]

I finally got around to some long range testing with the mild weather that we had last week. For the tests I walked across the street to a location 300 feet away from the router/center of my house with clear line of sight to my house. I'm guessing that based on a straight line that there may have been around 4 walls for the wifi to penetrate before getting outdoors. All test used the 5 Ghz band. I recorded the link rates (Rx and Tx) and RSSI values from the router's GUI although the link rates don't seem to agree with the observed results. Although not my favorite method of testing I used the speedtest.net Android app as it is takes much less time to run the test vs the DSLReports web page test. I ran 5 tests for each configuration and took the average of the best four results. Test client was the Pixel 2 XL.

It was interesting that as I walked out to the testing location with my body in between the phone and router the connection dropped and then reconnected when I turned around so that my body was not in the way.

One last thing to add. I pulled up the wifi analyzer app and noted the signal strength reported by the app. For the 68U it was reporting -80 dBm and for the 86U it was reporting -72 dBm.

 

[maxbraketorque]

Matches up well with my comparison testing.

 

[doczenith1]

I came across a brand new RT-AC1900P still wrapped in cellophane on craiglist for $70 (and she had it in the original Best Buy bag). Yes, you read that right, $70. Of course I had to snatch it up. I decided to do another round of throughput testing.

Both routers were on 384.11_2 with identical settings. Only difference was the 1900P did not have any scripts running.

Format will be device down/up speed in Mbps.
Results are the average of 4 tests using dslreports speedtest.

Close range testing one room away with full 866 link speed:
86U 68U
Pixel 3 XL 523/690 404/584
Pixel 3a 551/695 480/569
Average 537/692 442/576

86U roughly 20% faster

Medium range testing outdoors.
86U 68U
Pixel 3 XL 143/180 131/87
Pixel 3a 267/247 174/132
Average 205/142 152/109

86U roughly 33% faster

What surprised me the most was the 3a besting the 3. Both on June security patch. Better antenna design or the plastic body on the 3a lets more signal through than the glass on the 3?

 

[sfx2000]

What surprised me the most was the 3a besting the 3

Different device... simple enough...

Take each one as it is - as every device is different, but the AP should be the same.

 

[Grisu]

Both routers were on 384.11_2 with identical settings.

both set to fix channels and bandwith?
When set to auto it wont be same config, then its luck what it will take!

 

[doczenith1]

Yes, both manually set to the same channel and bandwidth. At some point in the near future I'm going to run the tests again from a location further away from the routers.

 

[maxbraketorque]

Yes, both manually set to the same channel and bandwidth. At some point in the near future I'm going to run the tests again from a location further away from the routers.

I found that the best channel on the AC68U was different than for the AC86U.

 

[doczenith1]

Turning this into a full time job for me huh? I've heard arguments both ways on whether or not the control channel matters when using a 80 MHz channel bandwidth. I will at the minimum run some tests on the lower group and the upper group and look for any differences.