That's not exactly how Internet traffic works. Your traffic is routed to and from the destination server via a series of hops across your ISP's infrastructure and then onto other organizations' networks. This shows you an example route: tracert

Routes are dynamic so that your traffic to the same site may take different routes at different times, depending on internet congestion and outages. These "routing tables" are maintained by ISPs - you have no control over them.

Speed tests are a way to measure your max bandwidth. They are not a good way to measure "how fast" a certain site will be. For example, for servers hosted in Seattle, that "very fast" Calgary route might wind up being slower than the Vancouver route.

 

try fast.com
But yeah your service servers have nothing to do with speed tests

 

1. When I upgraded my service with Lightspeed, I happened to ask the technician where the Lightspeed servers were located, and he said that was confidential information. Is this normally confidential info with ISPs?

Many companies, not just ISPs, do not advertise where their servers are for security reasons. I have been to several in my work and sometimes I have had difficulty finding them, as they didn't even have the street number on the building.

 

We provided data centre services for our clients for quite a number of years through a white label provider. While we knew the city the facility was located in (Richmond Hill), we didn't get the exact address until we signed an NDA.

 

conspiracy theory here
Internet companies know the speed test sites and give them priority so you get the false illusion you have fast internet.

 

There are several speed test sites and the speed will vary a bit with each of them. Search the web for options. Fast.com readings are typically very high and are not in the same ballpark as most other sites, so I take Fast with a lot of salt.

Various websites will have various speeds. The only way to accurately test a specific site is to download a large file and time how long it takes and do the calculation based on Mbps achieved. Although at that speed even large files don't take a lot of time.

Web browsing is not affected much once you're up to Gbps speeds.

 

Web browsing is not affected much once you're up to Gbps speeds.

I stopped noticing much of a difference around 150 Mb.

 

I use Speed Test by Measurement Lab

 

Thanks. My service tier is 1000Mbps down and 200 Mbps up.

This is what I got:

 

Speed test servers have nothing to do with the speed of the local internet service. They merely measure the speed to and from remote server sites. Some speed test servers are slower than others. That does not affect the local internet connection speed. It just indicates that server is overloaded, is routed through a slow connection or has a slow connection itself. The fastest speed test server on any day is a good indication of the local internet connection speed. Also, make sure that nothing else on the LAN is using the internet since that will return a lower sped test results.

 

Actually iut is "up to" those numbers usually.
In my case with Telus 940/940 is typical but it varies.
Hardwired ethernet to PC...right?

 

Yes, cable to the modem. Wifi is a bit slower.