OK I am completely stumped and am hoping someone can shed some light on this. When I do a speedtest to Tampa hiveocity on a gig I get 90 Mbps down and 20ish up when I change my connection to 100 I get about the same results (as expected)
If I do the same test to Atlanta at a gig I get 50 down and 8 up.
If I change my speed to a 100 I get 50 down and 26 up.
I have verified the results with wireshark and I am getting the speeds listed above. Can anyone explain to me why I'm getting a faster speed on 100 than a gig? The speed was changed on the switch directly (cisco 3750)
Thanks for your question tegguy.
Speed tests are very non-precise. Things that change the outcome are the ping, how many hops it has to travel, the distance it has to travel, what type of connection you are using (T1, T3, Cable, DSL, Fiber), and most importantly, the load on the server you are using for the speed test at the exact moment you run your test. For example, if 200 are people initiate a speedtest at the same time, the results for each will be different then if only one or two were doing the test at the same time.
Another thing to consider is, if you are using a DOCSIS3 Cable Modem, the change in connection speed may be forcing access to the cable modem with or without the channel bonding enabled. This could result in changing speeds both upload and download. I'm not aware of an ISP in the US that has channel bonding active on uploads but there has been talk about it. They may have it disabled for upload on gigabit connections and enabled on 10/100 connections in early testing.