Lagging on perfect connection.

[flicken]

i must agree, i have been playing flawlessly since launch and now after the poetic cap increase, i now have lag on Primal/Malboro servers. i did a pathping and IP trace to the NA data center and i was around 73 ms average with no packet loss, but yet i get players and spells will stop, then speed up really fast to catch up, this is a SE server problem and i know this because why would my game run great for over a year and then after a emergency maintenance on the 17th for Primal, it starts acting up?

Also note that i had my ISP's Tier 3 tech on the phone and at my house checking things out and "its not on my end" and im not paying WTfast to play a game, thats just stupid. my route is only 5 stops to NA data center through Century link.
The server is having problems and needs to be looked at and i have a good deal of my FC that will prove the lag on Malboro just started. Please fix.

[Raist]

i must agree, i have been playing flawlessly since launch and now after the poetic cap increase, i now have lag on Primal/Malboro servers. i did a pathping and IP trace to the NA data center and i was around 73 ms average with no packet loss, but yet i get players and spells will stop, then speed up really fast to catch up, this is a SE server problem and i know this because why would my game run great for over a year and then after a emergency maintenance on the 17th for Primal, it starts acting up?

Also note that i had my ISP's Tier 3 tech on the phone and at my house checking things out and "its not on my end" and im not paying WTfast to play a game, thats just stupid. my route is only 5 stops to NA data center through Century link.
The server is having problems and needs to be looked at and i have a good deal of my FC that will prove the lag on Malboro just started. Please fix.

You may still have issues along your route. You stated 73ms average, but how bad is the jitter? That average is awfully high for you to be that close (Wisconsin?). I'm in South Carolina with a 17 hop route, and I stay well under that--and I sometimes wind up sling-shotting Toronto to get there. I'm talking <64ms all the way there, with under a 10% variance between each ping across each hop. That is what is important here... consistent latency.

downdetector is getting reports from areas around you and/or your potential routes that may be impacting you intermittently--which would be in line with your rubber-banding description:

https://downdetector.com/status/centurylink

Recent reports mainly originated from: Denver, Iowa City, Boise, Des Moines, Portland, Phoenix, Saint Paul, Ottawa, Chicago, and Los Angeles. View on map

The bulk of them appear to be internet complaints too (80%). The thumbnail used as a link to the live outage map tracks recent history, and it shows problems have recently been much more widespread than just what is showing on the more current map (live map is linked in that quote):


Edit: there's some threads over at the broadbandreports forum about some weirdness with their lines during foul weather and other complaints occurring as well. who knows, you may find some help/suggestions over there as well: http://www.dslreports.com/forum/centurylink

So there may be more there than what you are capturing with smaller snapshots. Pathping isn't going to track the jitter that may be occurring along the route--it just gives you the average response times. You need to see the min/max as well, and need to see it track over time so you can see how much it may be varying. You may get a hint of the impact of the jitter with something like pingtest.net (although, it doesn't have a site for Montreal, but it may demonstrate the issue--it pings 250 times and tracks the average jitter between each ping for you).

When jitter is bad enough to cause packets to be delayed/received far enough out of order you will get a lot of retransmits. If it gets far enough out of whack you can actually loose your session. This is the kind of data that needs to be collected (how much variance there is along the route) as it can be a good indicator of where there may be intermittent pockets of elevated congestion/packet loss--which is what will cause the rubber-banding phenomenon you described.

Oh, and just to be clear... I don't typically experience this on my line, and when I do my route gets changed rather quickly and it clears up. I have worked with TWC's team to track the entire route, which has ultimately resulted in our routing getting changed quite often. At least once or twice a month (about 3 weeks on average) I get swapped to one of their other peering partners when latency starts to climb. Time Warner is doing this... NOT Square Enix. It appears to be the worst when I get paired with Level3... and when I do, I usually get switched off in about a week, sometimes less. Cogent and TATA seem to be the better choices of the three in play from here.