Does Cox artificially favor Speed-Tests? Real-World speeds slower than SpeedTest results.

Hi @jonathanj,

The specific test results that I posted here were from a desktop machine with a wired connection to my router. However, as I mentioned in my initial post, I performed a similar round of testing with my computer hard-wired directly to the modem. While the results were marginally better, the 1mbps difference was small enough to be attributed to standard network overhead.

I have run numerous tests on speedtest.net, specifically choosing servers which were geographically distant and NOT hosted on the Cox network, and results achieved through speedtest.net and fast.com consistently show results in the neighborhood of my plan's 150 Down/10 Up rates. I also specifically ran several tests to the BayNIC speedtest server in Fremont, CA, which is hosted in the same Hurricane Electric datacenter as iperf.scottlinux.com, the public iPerf server that I have been using to run my iperf tests.

Regarding iPerf and iperf.scottlinux.com, that server is confirmed to have at least 1 Gbps upstream and downstream, and in running off-site iPerf tests outside of my home network (i.e. via the T-Mobile network and my work's enterprise network)  I have consistently been able to achieve upstream and downstream results to that server well above what I see at home, indicating that there are no bottlenecks on the server side. 

Note that iperf is just the best tool I have found to specifically test speeds outside of the widely used speed test platforms, but the upstream results I see via iPerf are consistent with the transfer rates I see with real world loads such as Plex Media Server streaming, syncing to Google Drive, syncing my devices remotely via Syncthing, or performing large file transfers over an OpenVPN connection. Under all of these loads, my upstream transfer rates are consistently in the range of 2.5-4.5 Mbps up

I should also note that neither I nor the Cox support rep that I spoke to on the phone have found any evidence of packet loss on my modem, and in the past I have been able to consistently achieve 8-10 Mbps upstream using my current hardware configuration, so I am increasingly convinced that this is not a hardware issue on my end.

If you have any additional suggestions for further troubleshooting, I am all ears.

they do not artificially favor any speed test. however, the speed test they encourage (the one on the website) is to their own servers, which is one of the most direct node pathings. it is a bit disingenuous on their part. 

Go to speedtest or ookla or dslreports and set the server manually to literally anything that isn't a Cox hosted server, and you will see the real world slowdown by distance and node hops. 

"The problem is that these speeds are not holding up under sustained real-world loads, particularly on the Upstream side. I operate a Plex Media Server in my home and routinely stream from it throughout the day. In the past I was able to reliably stream at 8 Mbps, but in the last couple of months I have had severe buffering issues and had to bump down to 4 Mbps or less to get a stable stream. I have backed this up with a series of iperf3 tests, all of which show an initial transfer speed of 10 Mbps, which almost immediately drops to 4-4.5 Mbps after about a second of testing. Regular large transfers to other cloud storage services tell a similar story."

This sounds like your networking hardware is overheating, or more likely, a line in the ground got shifted somewhere and it cut one of the up lines. If you are running a media server then overheating is very likely, however. 

Alright, status update on this. In the interest of eliminating as many external variables as possible I connected to a public "Cox WiFi" hotspot hosted by one of my neighbors, and I was able to achieve stable 10Mbps uploads from my Plex server at home streaming to my phone. Results were similar for other workloads, meaning that there are no bottlenecks on my home network or the local Cox infrastructure. Since streaming is consistently slower on non-Cox networks, I'm thinking a bottleneck further up the chain and not directly related to my personal hardware or Cox internet service. While the issue still exists, I am going to consider this effectively solved as any solution would be well above my pay-grade.