SE should consider providing its own VPN network to solve the lag problem

[Kowen]

Weird, not one of my FC members living in NA were hit by any "post 2.1 lag issue". I hadn't heard about any *new* lag issues since 2.1. The only lag issues I had were between the novemeber hotfix patch and patch 2.1 where I would get random lag spikes. But for me at least, that stopped after 2.1.

As far as I'm concerned, network issues have decreased post 2.1, not increased.

There's a whole thread about it here.

The problems have to do with Level3, one of the network points FiOS uses to send data to Montreal, frequently dropping packets. Anybody with an ISP that communicates with Level3 will experience the problem, not just FiOS customers.

That's why WTFast fixes the issues; it circumvents that whole network.

[Alhanelem]

There's a whole thread about it here.

The problems have to do with Level3, one of the network points FiOS uses to send data to Montreal, frequently dropping packets. Anybody with an ISP that communicates with Level3 will experience the problem, not just FiOS customers.

That's why WTFast fixes the issues; it circumvents that whole network.

Well it seems to me like this problem is mostly beyond SE's control then. The only thing they could do would be to relocate to a new datacenter or do something to force routing around the problem areas if Level3 can't fix them. I can say at least the former option would cost a fortune.

[Wargaszm]

Then go to: http://network-tools.com/

type in your servers IP address

you will get a read out possibly like this:

TraceRoute from Network-Tools.com to 199.91.189.26
Hop (ms) (ms) (ms) IP Address Host name
1 0 0 0 8.9.232.73 xe-5-3-0.edge3.dallas1.level3.net
2 40 40 40 4.69.145.190 vlan80.csw3.dallas1.level3.net
3 40 40 40 4.69.151.150 ae-81-81.ebr1.dallas1.level3.net
4 40 40 40 4.69.151.117 ae-14-14.ebr2.chicago2.level3.net
5 40 40 45 4.69.132.113 ae-1-100.ebr1.chicago2.level3.net
6 40 40 40 4.69.140.189 ae-6-6.ebr1.chicago1.level3.net
7 40 Timed out 40 4.69.153.86 ae-10-10.car2.montreal2.level3.net
8 40 40 40 4.59.178.74 ormuco-comm.car2.montreal2.level3.net
9 40 40 40 192.34.76.10 -
10 41 41 40 199.91.189.242 -
11 40 40 40 199.91.189.26 -

Trace complete

[Wargaszm]

The important part is this:
7 40 Timed out 40 4.69.153.86 ae-10-10.car2.montreal2.level3.net

Hop 7 is a server with a know problem, its timing out at that point in the connection users are experiencing packet loss.

This site monitors routing servers and has history of the problem.
http://www.dslreports.com/routerwatc...al2.Level3.net

Should SE provide a VPN ... Your crazy, but they should be pressuring level 3 to fix the problem because there game will suffer with poor internet connectivity.

WTFast is a solution but it comes at a cost and will reroute your connection and stabilize your ping spikes and reduce packet loss. Fio's customers pay for the "best" internet available in the US ... Fiber optic connections ... yadda yadda yaddda .. the routing should be fixed on one side or the other but SE is going to have enter into negotiations to solve the problem.

[Alhanelem]

not all MMO's have terrible netcode and require an amazing connection to do the end-game raiding.

Except it is well established that this isn't a netcode issue, it is a backbone ISP issue. It is not within SE's control, short of them badgering the owner of the infrastructure to fix it. This also explains why some people are not having any issues, because their connections are taking a different path to the target host machines.

If you read the posts above, you'll understand exactly why the VPN is effective. If it was "terrible netcode", then you would still be having problems even with the VPN. No amount of efficiency in networking would be able to compensate for bad netcode.

[Rekujen]

Except it is well established that this isn't a netcode issue, it is a backbone ISP issue. It is not within SE's control, short of them badgering the owner of the infrastructure to fix it. This also explains why some people are not having any issues, because their connections are taking a different path to the target host machines.

If you read the posts above, you'll understand exactly why the VPN is effective. If it was "terrible netcode", then you would still be having problems even with the VPN. No amount of efficiency in networking would be able to compensate for bad netcode.

That's very understandable and I'm aware of the situation, but other MMO's have back up features for this exact problem. For instance, in WoW my ping is 300 but they have lag compensation to help, and it really does work.. FFXIV has no lag compensation whatsoever and the netcode in regards to client/to/server ratio isn't optimized properly, meaning players with ping over 200+ have a very poor chance of dodging red AoE's compared to other games and their AoE mechanics.

3 options

1. implement lag compensation
2. tweak the client to server ratio
3. contact ISP's and fix the "p2p throttling" that is apparently plaguing this game.

I have made contact with my ISP but am still waiting for a reply...

[Squa]

If it was "terrible netcode", then you would still be having problems even with the VPN. No amount of efficiency in networking would be able to compensate for bad netcode.[/QUOTE]

well its a few things. SE (unlike other companies) use server side netcode. So, you are much more prone to issues such as these.
Secondly, I think absolutely SE should either pay for users VPN, or have their servers in an area such that they aren't needed. Compromising the gameplay experience (which is the solution now) I think is the least desired, especially for a subscription based game. People will eventually get sick of having to pay for 3rd party and subscription to play 'correctly' and it will be bad business.