@Pat2 -- the "variability" can be compared to driving your automobile from Vancouver (BC) to Seattle -- some roads are highly-congested at various times of the day. Also, the "wait" at the CAN/USA border, where BC Highway 99 becomes Interstate Highway 5, has great variability.
On the Internet, your Internet traffic takes various "segments" (each segment joins 2 routers). Example:
Tracing route to fast.com [104.86.233.81]
6 8 ms 9 ms 8 ms rd1lp-be1.gv.shawcable.net [66.163.72.22] (Greater Victoria)
7 11 ms 10 ms 11 ms rc1wt-be40.wa.shawcable.net [66.163.68.18] (actually Vancouver)
8 11 ms 13 ms 12 ms rc2wt-be18-1.wa.shawcable.net [66.163.64.82] (actually Vancouver)
9 11 ms 10 ms 11 ms rx0wt-korea-telecom.wa.shawcable.net [66.163.68.14] (actually Vancouver, not Korea)
10 14 ms 33 ms 13 ms ae34.r01.border101.sea01.fab.netarch.akamai.com [23.203.145.247] (Seattle)
11 * * * Request timed out. (the router chose to not respond to "traceroute" packets)
12 * * * Request timed out. (ditto)
13 * * * Request timed out. (ditto)
14 14 ms 12 ms 12 ms a104-86-233-81.deploy.static.akamaitechnologies.com [104.86.233.81] (a high-speed "co-located" server just one "hop" away from an Internet "backbone"
Trace complete.
>>> I understand that cable internet data is affected by the number of users at the cable aggregation point which explains why cable internet can slow down in your neighbourhood.
That is "false news" that Shaw's competitors use to promote their network.
For my Shaw "Total TV" package, Shaw's coaxial-cable carries over 200 channels to my home, plus Internet, plus Shaw Home phone. It does the same for each of my neighbours who also subscribe to Shaw.
As you can see from the above "traceroute", the latency gradually increases, at each router.
Another example, tracing across the Atlantic Ocean, to TV channel 2, in Norway:
Tracing route to www.tv2.no [5.44.65.150]
6 10 ms 10 ms 9 ms rd1lp-be1.gv.shawcable.net [66.163.72.22] (Greater Victoria)
7 11 ms 11 ms 10 ms rc1wt-be40.wa.shawcable.net [66.163.68.18] (Vancouver)
8 12 ms 10 ms 11 ms sea-b1-link.ip.twelve99.net [213.248.67.224] (Seattle)
9 55 ms 55 ms 54 ms chi-b23-link.ip.twelve99.net [62.115.132.154] (Chicago)
10 54 ms 54 ms 54 ms chi-bb2-link.ip.twelve99.net [62.115.126.158] (Chicago)
11 71 ms 71 ms 71 ms nyk-bb2-link.ip.twelve99.net [62.115.132.134] (New York)
12 * * * Request timed out.
13 158 ms 155 ms 154 ms oso-b1-link.ip.twelve99.net [62.115.116.106] (Oslo, Norway)
14 158 ms 154 ms 155 ms telenor-ic-335111.ip.twelve99-cust.net [213.248.84.58] (Telenor)
15 172 ms 158 ms 156 ms ti0001c360-ae7-0.ti.telenor.net [146.172.105.65] (Telenor)
16 154 ms 156 ms 155 ms ti0275c360-ae1-0.ti.telenor.net [146.172.22.37] (Telenor)
17 155 ms 156 ms 156 ms ti0300a400-ae0-0.ti.telenor.net [146.172.99.158] (Telenor)
18 157 ms 156 ms 156 ms 194.248.41.50
19 155 ms 154 ms 156 ms 5.44.65.150
Trace complete.
Again, increased latency at each "hop", especially the very-busy undersea cable between New York and Oslo.
Also, you must consider the number of connections to a "gaming-server" varies throughout the day/night. Sometimes, those servers cannot "keep-up" with their customer's demands. Compare to a first-day trainee [slowly] working the cash-register at McDonalds to the [full] speed of the shift-manager. Customers still enter the front-door of McD's at about the same rate, but the queues inside the store grow and grow, at a "rush-hour" (10 AM coffee, 12 PM lunch).
Anyone have a solution to this problem?
Yes, Shaw does monitor their routers, and adds capacity when a router is "too-busy". Routers are expensive components. So, they are "right-sized" when they are installed. Obviously, a router in Vancouver is faster at "packet-switching" than a router in Tofino -- and, of course, costs a lot more. Any solution costs money.
Also, those routers in the USA (and Norway) are not controlled by Shaw, and were not purchased by Shaw. So, any "bottle-necks" of routers outside of Canada cannot be remediated by Shaw. If a "game-server" is located in Las Vegas, the solution is to purchase a faster router to be deployed in Las Vegas. You probably will also need faster routers in Los Angeles and Seattle, to move the packets (Vancouver -> Seattle -> LA -> LV) across those segments of the Internet.
