Potential problem RT-AC68U (originally TM-AC1900) and WiFi driver 6.37.14.126

[sirwifi]

Background
For the past couple of Merlin releases I noticed that I had problems with my phones making VOIP calls over WiFi, 2.4Ghz band. Very poor call quality, hard to understand the other party, etc. It prompted me to look at the WiFi connection. Didn't notice anything unusual, the signal was strong, the connection itself was stable, the clients were not being disconnected. I suspected channel interference, so I tried other channels (but mainly tried to stay with 1, 6, 11). Nothing helped. I thought it was just my phone. But it did it on other phones too.

Troubleshooting
I decided to connect my Macbook Pro (late 2013 model) to the same 2.4Ghz SSID (the Mac normally is on 5Ghz) and ran some iperf tests against a wired Linux server.
What I noticed was jitter all over the place and packet loss around 5%, sometimes lower, sometimes higher. That explained the crappy VOIP calls right away.

Fix
First attempt: I had read about the LTS fork before so I decided to use the latest one V27E5 (which was somewhat tricky to downgrade to from 380.68) because I had no problems in the past with older versions of Merlin. To my surprise, it didn't make a difference, same problems. Then someone mentioned in the forums that even LTS uses the latest drivers for RT-AC68U, so I posted the question which LTS version had the old wireless drivers. Answer: v26E4.

Working fix: I switched to LTS v26E4 which uses wireless driver "Apr 25 2014 11:16:33 version 6.37.14.86 (r456083)" and BINGO! back to a working WiFi. Very low jitter, no more packet loss, a few test calls sound good. Now, it's a bit early to declare total victory but it looks promising.

Just for kicks, I put once again 380.68 back on the router, reset NVRAM, clean configuration by hand. Same problems returned. Now back to LTS v26E4 (the painful way) and it's all fine again.

So there is something going on here, something about the new driver "Jun 1 2017 17:28:05 version 6.37.14.126 (r561982)" and my hardware. I think it's the driver, not sure what else to blame it on. Once again, with the "bad" driver the connection is stable, signal is strong, nothing unusual, except Tx/Rx results in high packet loss and jitter.

For completeness here are my hardware details:

TM-AC1900 (2014) flashed as RT-AC68U
Bootloader (CFE): 1.0.2.0

nvram get boardrev
0x1100

nvram get HW_ver
170

nvram get bl_version
1.0.2.0

nvram get wl_corerev
42

I'm not sure whether the CFE can have an impact in any of the above. Mine is a custom made one based on a guide from 3+ yrs ago.

Now this is one data point, it may be just me, but I'd be curious to hear from others who have genuine RT-AC68U and also modified TM-AC1900.

But if you run into any WiFi problems, you may want to try a firmware version before the driver change.

 

[sirwifi]

This is what a good iperf run looks like, with the "good' driver:

[ ID] Interval       Transfer     Bandwidth        Jitter   Lost/Total Datagrams
[  3]  0.0- 1.0 sec  18.0 MBytes   151 Mbits/sec   0.118 ms    0/12824 (0%)
[  3]  1.0- 2.0 sec  18.0 MBytes   151 Mbits/sec   0.096 ms    0/12824 (0%)
[  3]  2.0- 3.0 sec  18.0 MBytes   151 Mbits/sec   0.108 ms    0/12819 (0%)
[  3]  3.0- 4.0 sec  18.0 MBytes   151 Mbits/sec   0.111 ms    0/12816 (0%)
[  3]  4.0- 5.0 sec  18.0 MBytes   151 Mbits/sec   0.119 ms    0/12819 (0%)
[  3]  5.0- 6.0 sec  18.0 MBytes   151 Mbits/sec   0.082 ms    0/12826 (0%)
[  3]  6.0- 7.0 sec  18.0 MBytes   151 Mbits/sec   0.132 ms    0/12809 (0%)
[  3]  7.0- 8.0 sec  18.0 MBytes   151 Mbits/sec   0.082 ms    0/12831 (0%)
[  3]  8.0- 9.0 sec  18.0 MBytes   151 Mbits/sec   0.089 ms    0/12824 (0%)
[  3]  0.0-10.0 sec   180 MBytes   151 Mbits/sec   0.084 ms    0/128205 (0%)

vs. a terrible one with the "bad" driver, though the one below shows only packet loss, jitter is fine on this particular run, but sometimes it is a few ms:

[ ID] Interval       Transfer     Bandwidth        Jitter   Lost/Total Datagrams
[  3]  0.0- 1.0 sec  15.5 MBytes   130 Mbits/sec   0.018 ms  359/11400 (3.1%)
[  3]  1.0- 2.0 sec  17.2 MBytes   144 Mbits/sec   0.025 ms  614/12887 (4.8%)
[  3]  2.0- 3.0 sec  16.7 MBytes   140 Mbits/sec   0.023 ms  981/12887 (7.6%)
[  3]  3.0- 4.0 sec  17.3 MBytes   145 Mbits/sec   0.035 ms    0/12325 (0%)
[  3]  4.0- 5.0 sec  16.4 MBytes   138 Mbits/sec   0.021 ms 1394/13096 (11%)
[  3]  5.0- 6.0 sec  16.5 MBytes   138 Mbits/sec   0.026 ms 1051/12815 (8.2%)
[  3]  6.0- 7.0 sec  15.2 MBytes   128 Mbits/sec   0.017 ms 1581/12440 (13%)
[  3]  7.0- 8.0 sec  12.4 MBytes   104 Mbits/sec   0.022 ms 4135/13000 (32%)
[  3]  8.0- 9.0 sec  18.0 MBytes   151 Mbits/sec   0.022 ms  388/13218 (2.9%)
[  3]  9.0-10.0 sec  17.8 MBytes   149 Mbits/sec   0.061 ms    0/12699 (0%)
[  3]  0.0-10.1 sec   165 MBytes   137 Mbits/sec   0.088 ms 10502/128204 (8.2%)

For all these tests I used the same settings, forced channel 6 @ 20Mhz, RSSI about -55 dBm, Noise around -92 dBm.

 

[strangeluck]

I also have a converted TM-AC1900 running Merlin's 380.68 firmware and while I don't have a very good laptop to test with here are my iperf results:

[ ID] Interval           Transfer     Bandwidth       Jitter    Lost/Total Datagrams
[  5]   0.00-1.00   sec  6.60 MBytes  55.4 Mbits/sec  0.570 ms  0/845 (0%)
[  5]   1.00-2.00   sec  6.34 MBytes  53.1 Mbits/sec  0.837 ms  0/811 (0%)
[  5]   2.00-3.00   sec  7.52 MBytes  62.9 Mbits/sec  1.199 ms  0/963 (0%)
[  5]   3.00-4.00   sec  7.48 MBytes  63.0 Mbits/sec  0.773 ms  0/958 (0%)
[  5]   4.00-5.00   sec  7.55 MBytes  63.3 Mbits/sec  0.996 ms  0/967 (0%)
[  5]   5.00-6.00   sec  7.52 MBytes  63.0 Mbits/sec  1.218 ms  0/962 (0%)
[  5]   6.00-7.00   sec  7.51 MBytes  63.0 Mbits/sec  0.501 ms  0/961 (0%)
[  5]   7.00-8.00   sec  7.30 MBytes  61.2 Mbits/sec  1.759 ms  0/935 (0%)
[  5]   8.00-9.00   sec  7.21 MBytes  60.6 Mbits/sec  1.076 ms  41/964 (4.3%)
[  5]   9.00-10.00  sec  7.48 MBytes  62.8 Mbits/sec  0.893 ms  0/958 (0%)
[  5]  10.00-10.07  sec   432 KBytes  52.5 Mbits/sec  0.966 ms  0/54 (0%)

Ran the tests a couple times and the results are roughly the same on each run. While I do have some packet loss I can't say I have noticed any wifi calling related problems myself. I did test on another access point here in the house (TP-Link Archer C7 V2) and do not have any packet loss on that AP, so I don't think the problem is this laptop's radio (poor as it may be).

 

[sirwifi]

I assume you are using 2.4Ghz?

 

[strangeluck]

I assume you are using 2.4Ghz?

Yes, 2.4GHz. I actually just ran a couple of tests again. I am in a different part of the house from where I tested earlier. My last round of tests I was in the same room with the router, Now I am two floors up from the router's location (as evidenced by the slightly lower data rate). However, I am getting 0% loss in this part of the house. Signal is reported in the WebUI as -72 dBm.

[ ID] Interval           Transfer     Bandwidth       Jitter    Lost/Total Datagrams
[  5]   0.00-1.00   sec  5.76 MBytes  48.3 Mbits/sec  1.231 ms  0/737 (0%)
[  5]   1.00-2.00   sec  6.24 MBytes  52.4 Mbits/sec  1.187 ms  0/799 (0%)
[  5]   2.00-3.00   sec  6.59 MBytes  55.2 Mbits/sec  1.667 ms  0/844 (0%)
[  5]   3.00-4.00   sec  6.18 MBytes  51.7 Mbits/sec  1.118 ms  0/791 (0%)
[  5]   4.00-5.00   sec  6.75 MBytes  56.6 Mbits/sec  0.814 ms  0/864 (0%)
[  5]   5.00-6.00   sec  6.70 MBytes  56.4 Mbits/sec  0.822 ms  0/857 (0%)
[  5]   6.00-7.00   sec  6.19 MBytes  51.8 Mbits/sec  0.883 ms  0/792 (0%)
[  5]   7.00-8.00   sec  6.32 MBytes  53.1 Mbits/sec  0.790 ms  0/809 (0%)
[  5]   8.00-9.00   sec  6.62 MBytes  55.5 Mbits/sec  1.768 ms  0/847 (0%)
[  5]   9.00-10.00  sec  6.05 MBytes  50.7 Mbits/sec  1.528 ms  0/774 (0%)
[  5]  10.00-10.07  sec   448 KBytes  53.3 Mbits/sec  1.553 ms  0/56 (0%)

 

[sirwifi]

That looks very good. What hardware rev do you have? The commands are in my first post.

 

[strangeluck]

That looks very good. What hardware rev do you have? The commands are in my first post.

170, same as you. The only difference in results of those commands is that my bl_version is listed as 1.0.2.1

 

[RMerlin]

Disable Airtime Fairness.

 

[ColinTaylor]

170, same as you. The only difference in results of those commands is that my bl_version is listed as 1.0.2.1

Same here. (My test results were posted in the other thread).

 

[john9527]

What version of iperf is everyone using?

 

[strangeluck]

What version of iperf is everyone using?

I'm using iperf 3.1.3

Disable Airtime Fairness.

I have airtime fairness disabled on my router already.

 

[ColinTaylor]

What version of iperf is everyone using?

3.0.12 on the Linux server and 3.0.11 on the Windows client.

 

[john9527]

I have to join everyone else in not being able to detect a difference. Ran against a 2.4GHz only RA-Link USB adapter, and an Atheros dual-band Mini-PCIe card (on 2.4GHz). All substantially looked like this, with the USB having a slight higher jitter/variation.

[ ID] Interval           Transfer     Bandwidth       Jitter    Lost/Total Datagrams
[  5]   0.00-1.01   sec  7.03 MBytes  58.2 Mbits/sec  2.210 ms  0/900 (0%)
[  5]   1.01-2.01   sec  9.49 MBytes  79.8 Mbits/sec  1.318 ms  0/1215 (0%)
[  5]   2.01-3.01   sec  8.62 MBytes  72.5 Mbits/sec  2.238 ms  0/1104 (0%)
[  5]   3.01-4.01   sec  8.42 MBytes  70.8 Mbits/sec  1.532 ms  0/1078 (0%)
[  5]   4.01-5.01   sec  9.55 MBytes  80.3 Mbits/sec  1.501 ms  0/1223 (0%)
[  5]   5.01-6.01   sec  9.38 MBytes  78.8 Mbits/sec  1.331 ms  0/1200 (0%)
[  5]   6.01-7.00   sec  9.09 MBytes  76.4 Mbits/sec  1.328 ms  0/1164 (0%)
[  5]   7.00-8.00   sec  9.03 MBytes  75.9 Mbits/sec  1.282 ms  0/1156 (0%)
[  5]   8.00-9.00   sec  9.09 MBytes  76.3 Mbits/sec  0.932 ms  0/1163 (0%)
[  5]   9.00-10.02  sec  9.30 MBytes  76.9 Mbits/sec  1.405 ms  0/1190 (0%)
[  5]  10.02-10.22  sec  1.87 MBytes  77.2 Mbits/sec  1.496 ms  0/239 (0%)

 

[john9527]

I just had another thought though.....I seem to remember that Mac's were sensitive to the regulation mode (not exactly sure which way, but I think they liked it disabled/off). You may want to try toggling the setting on the Wireless > Professional page

EDIT: This is on my V27 fork....Merlin no longer has this setting.

 

[sirwifi]

Disable Airtime Fairness.

Where is this option supposed to be? I don't see it in the LTS fork GUI, neither under General nor Professional tab.

Here's how I change the WiFi settings from defaults:

Set N or N/AC only for the mode (I don't have any legacy a/b/g devices)
Set bandwidth to 20Mhz for 2.4Ghz and 80Mhz for 5Ghz
Set fixed channel (in my case I found 6 and 149 to perform best)
WPA2+AES encryption

Short preamble
Enable Reduce USB 3 interference (even though I don't have any USB devices attached to the router)

 

[sirwifi]

I just had another thought though.....I seem to remember that Mac's were sensitive to the regulation mode (not exactly sure which way, but I think they liked it disabled/off). You may want to try toggling the setting on the Wireless > Professional page

EDIT: This is on my V27 fork....Merlin no longer has this setting.

Regulation mode is off for me by default.

 

[john9527]

Where is this option supposed to be? I don't see it in the LTS fork GUI, neither under General nor Professional tab.

Down near the bottom of the professional tab (added on V27)....but I set it to default off, but worth a double check.

I'd also set WMM APSD to disabled.

 

[sirwifi]

Down near the bottom of the professional tab (added on V27)....but I set it to default off, but worth a double check.

I'd also set WMM APSD to disabled.

OK, so it's not in v26, I'm not blind.

This guide says WMM APSD should be enabled for VOIP applications (which I need from my phones):
https://routerguide.net/wmm-apsd-on-or-off/

 

[john9527]

This guide says WMM APSD should be enabled for VOIP applications (which I need from my phones):

Guides are like.....everyone has one. APSD is a power saving feature that uses a 'packet scheduling' algorithm. If both sides work together, great. But when I read about it, my first thought was there was a lot of opportunity to mess up. I'd still try with it off.......it's only a couple mouse clicks

 

[sirwifi]

Fair enough, I can try it though I'm staying with your v26 which seems to work best on my hardware.