Well... their servers aren't located in Canada for starters. GW2 is Texas, Planetside 2 is California and Virginia.
Trace to an IP listed for Planetside's Jaeger server (Virginia):
Code:
Tracing route to abepsn-liv-gw01.planetside2.com [199.108.194.38]
over a maximum of 30 hops:
1 1 ms <1 ms <1 ms LPTSRV [10.10.100.1]
2 1397 ms 2936 ms 187 ms cpe-098-025-064-001.sc.res.rr.com [98.25.64.1]
3 12 ms 11 ms 11 ms cpe-024-031-198-005.sc.res.rr.com [24.31.198.5]
4 23 ms 26 ms 28 ms 24.93.64.124
5 31 ms 31 ms 30 ms 107.14.19.20
6 30 ms 31 ms 30 ms 107.14.19.133
7 30 ms 32 ms 39 ms tge15-3.fr3.iad.llnw.net [69.28.158.181]
8 41 ms 29 ms 32 ms sony.ge4-12.fr3.iad.llnw.net [68.142.111.254]
9 42 ms 41 ms 40 ms abepsn-liv-gw01.planetside2.com [199.108.194.38]
Trace complete.
Now here's a trace to the IP I use for Midgard in FFXIV (Montreal, Canada):
Code:
Tracing route to 199.91.189.25 over a maximum of 30 hops
1 2 ms <1 ms <1 ms LPTSRV [10.10.100.1]
2 516 ms 1198 ms 66 ms cpe-098-025-064-001.sc.res.rr.com [98.25.64.1]
3 13 ms 11 ms 11 ms cpe-024-031-198-005.sc.res.rr.com [24.31.198.5]
4 24 ms 22 ms 26 ms xe-7-0-0.rlghncpop-rtr1.southeast.rr.com [24.93.64.40]
5 31 ms 30 ms 30 ms 107.14.19.44
6 82 ms 29 ms 31 ms ae-2-0.pr0.dca10.tbone.rr.com [66.109.6.169]
7 34 ms 33 ms 38 ms te0-16-0-23.ccr41.iad02.atlas.cogentco.com [154.54.10.209]
8 34 ms 34 ms 38 ms be2176.ccr21.dca01.atlas.cogentco.com [154.54.41.54]
9 39 ms 40 ms 41 ms be2151.mpd22.jfk02.atlas.cogentco.com [154.54.40.74]
10 43 ms 44 ms 44 ms be2106.ccr21.alb02.atlas.cogentco.com [154.54.3.50]
11 66 ms 66 ms 66 ms be2088.ccr21.ymq02.atlas.cogentco.com [154.54.43.17]
12 53 ms 53 ms 54 ms 38.122.42.34
13 44 ms 45 ms 46 ms 192.34.76.10
14 43 ms 44 ms 46 ms 199.91.189.242
15 44 ms 46 ms 46 ms 199.91.189.25
Trace complete.
The point is that you are dealing with unique routing for each one, and each can take vastly different paths, and thus can be subject to vastly different shaping issues. In this case, some key corridors leading to the limited entry points into Montreal have been having issues keeping up with the increased traffic. This can further compound issues SE has been having with some of their servers.
Fortunately, we've been able to get them to clean up a large part of our routing through the cogentco segments. That portion of our routing was once consistently breaking the 200ms mark and stalling a lot. They moved us to Level3 for while (which was a nightmare), but eventually they got us back into the much cleaner Cogent deal we are now using. Just can't get them to address that primary gateway yet at 98.25.64.1. When it's acting up like it is in these traces, the skip is really bad. When it's clean though, you'd be hard pressed to notice any lag until you hit instanced content (which is where SE is having the real trouble atm).