It's Cogentco In Montreal that's causing the lag Guaranteed

[Pygar]

It is most certainly Cogentco that's responsible for the connection problems guaranted.

After two months now, I can say without a doubt it is their backend causing the problems.

With that said, I don't see a resolution to this mess unless SE move their DataCenter somewhere else because I have to imagine SE know by now it is them and have known for a while and there is still no resolution.

So you are saying everyone's latency issues are tied to Cogentco because everyone's routes go through them? Clarification please.

Cogentco can't handle the load, It is clearly their backend. During peak times, more will notice the issues and experience them.

Over the past two months, I have been working with Comcast (HQ in PA) to determine if it is their (Comcast's) network with their maint/escalation department since I raised the question of whether or not they are throtteling it as P2P and their engineer (read: not Tier 1, not Tier 2) and we have found nothing wrong with the routing between myself and at least them.

With that in hand, I have gathered tracerts from more than a few handful of people, spread across the U.S. who connect to the U.S. Data center (in Montreal) and they all exhibit the exact same behaviour. Time outs, or extrodinarily high latency once they reach Cogentco's backend.

The reason some don't experience issues, or less issues is most certainly the time of day and Cogentco's ability to spread the load to so many connections. Unless Cogentco address the problem, SE will continue to see these issues occur and I suspect, that's why after 140 pages of blaming our ISP's (the hundreds of them) it means, they either have not bothered to contact our ISP's (the hundreds of them) to remediate the issue or they know that it isn't our ISP's and know Cogentco can't handle the load and saying as much would be very damaging to the success of the game.

There are folks in NY and New England that suffer latency issues and a large amount of them have trace routes not through Cogentco/ATT.

NE pipeline is mostly Verizon (alter.net) and into Montreal Inteliquent (Tinet) unless your routing is off.

SE is using Ormuco Communications as far as Datacenter solutions.

Wouldn't it be ironic if the throtelling were P2P directly out the backend from Ormuco. I mean, maybe THEY are not provisioned for the correct loads.

Not likely I'm sure, since I would bet that's where they started looking.

[ntall1]

They should never have put all the servers in a single data center. At least now they have learned why no other company has ever done this.

[ReplicaX]

It is most certainly Cogentco that's responsible for the connection problems guaranted.

So you are saying everyone's latency issues are tied to Cogentco because everyone's routes go through them? Clarification please.

[Swishytail]

This wouldn't surprise me, nor would it be the first time:

Cogentco/ATT:
http://us.battle.net/wow/en/forum/topic/7592642473

Cox (mostly Arizona):
http://us.battle.net/wow/en/forum/to...3741008?page=1

I remember playing WoW and having the same garbage at one point. Same company. Plus Cox.
This went on for like a month before it was fixed, if I recall.

One person is quoted as to saying:

"While that is the problem, you cant blame it completely on ATT. Cogentco has a history of problems between peering partners and this isnt the first game that has fallen victim to cogentco latency. I currently also play SWTOR, guess what its unplayable as well because of the way my traffic is routed."

[ReplicaX]

There are folks in NY and New England that suffer latency issues and a large amount of them have trace routes not through Cogentco/ATT.

NE pipeline is mostly Verizon (alter.net) and into Montreal Inteliquent (Tinet) unless your routing is off.

SE is using Ormuco Communications as far as Datacenter solutions.

Wouldn't it be ironic if the throtelling were P2P directly out the backend from Ormuco. I mean, maybe THEY are not provisioned for the correct loads.

While I'm sure Cogentco is part of the issue for many players. I feel during peak hours and most complaints stemming from instance battles, its possible that the datacenter or its hardware isn't properly handling the loads either at times. Looking at many players pings into Ormuco's fiber network, they jump greatly during peak hours. This would add even more latency on top of players affected by Cogentco.

This is just another look at the matter. It might not be just one issue, but many. However, by no means am I , a network engineer or programmer (engine/netcode)

[Faelart]

And SE wouldn't fund creating a new datacenter while keeping this one up to keep the game online. And taking the datacenter offline while the databases are transferred across the network from datacenter a to datacenter b, and no way in hell they'd even take the game offline long enough to just take all the equipment to another location either, so their options are only to bleed cash or bleed even more cash, or let people piss and moan and blame our ISPs and slowly bleed money from subscription loss which when enough people quit the latency issue will resolve itself due to less network strain, lol.

[Ervin]

I'm gonna give SE until patch 2.1 most likely to get the lag issues sorted out aswell. If the lag is still there when the end of november comes I'll most likely quit playing. There are other options out there that are ok to play, without lag and without spending 11 Euro a month. I'm fine with paying the sub if the game works as it should, but right now with the lag issues, the game isnt really worth the 11 Euro per month.

It's one of the most important things with an MMO game, a stable reliable connection.

[Enyyx]

I'm going to disagree, at least in my case.

I never suffer the horrid rubber band lag at home, which routes through Cogent. In my other play location, which routes through Tata, play becomes unbearable around 2pm eastern every day. Today, for the first time, I was able to remedy this for myself: I VPN'd to my home office to hop on the cogent route and the lag went away.

Your results may vary

[Donoho]

For a few weeks now (since the ISP throttling accusations), this has had all the signs of Datacenter Bandwidth issue. I don't mean the pipe isn't big enough, I mean the routing can't handle the amount of traffic. Although visible symptoms take on a different form on a corporate network, the overall signs are still far too similar.

Although basically a guess, a couple of decades of network troubleshooting experience makes it (at least) an educated guess.

[svann]

http://forum.square-enix.com/ffxiv/t...ausing-you-lag

Its not difficult to check. If it's cogentco then you would expect that you will get no timeouts coincident with gamelag on any hop up until their server.