in the interest of being thorough (and having time to futz with it,) from the manual I ran a scan on neighboring networks, most of them are on 1, 6, 11, so changed the router to channel 3. No improvement. Also my WMM support is grayed out, can't try turning it off per the manual.
During one of the slow times I ran traceroute on cox.net:
traceroute to cox.com (45.60.45.167), 30 hops max, 60 byte packets
1 gateway (192.168.0.1) 11.541 ms 22.788 ms 22.781 ms
2 10.142.0.1 (10.142.0.1) 22.758 ms 23.313 ms 23.308 ms
3 68.6.14.132 (68.6.14.132) 23.299 ms 23.282 ms 23.251 ms
4 100.120.108.12 (100.120.108.12) 23.224 ms 23.204 ms 24.226 ms
5 ae55.bar2.SanDiego1.Level3.net (4.16.106.5) 24.851 ms 25.326 ms 25.285 ms
6 ae-0-11.bar1.SanDiego1.Level3.net (4.69.146.65) 25.817 ms 23.681 ms 19.123 ms
7 * * *
8 * * *
9 4.7.26.34 (4.7.26.34) 25.848 ms 28.049 ms 27.584 ms
10 * * *
11 * * *
12 * * *
13 * * *
14 * * *
15 * * *
16 * * *
17 * * *
18 * * *
19 * * *
20 * * *
21 * * *
22 * * *
23 * * *
24 * * *
25 * * *
26 * * *
27 * * *
28 * * *
29 * * *
30 * * *
And that's where it dies, at Level 3 Communications, CA.
https://whatismyipaddress.com/ip/4.7.26.34
A test a minute later dies at 192.80.16.42, NTT communications, CO.
An interesting question is why the second two tracroutes show the tenth hop through JAPAN.
10 ae-10.r27.tokyjp05.jp.bb.gin.ntt.net (129.250.2.156) 144.007 ms 145.087 ms 147.728 ms
A normal traceroute should end with the last hop, like this one.
13 lax17s38-in-f14.1e100.net (172.217.14.78) 35.062 ms 35.061 ms 34.524 ms
