Regarding high latency from Los Angeles/surrounding area

[dascorp]

I was wondering what was wrong with my connection... I just did the trace... and it isnt pretty..

Tracing route to 199.91.189.57 over a maximum of 30 hops

1 3 ms 1 ms 1 ms home [192.168.1.254]
2 13 ms 13 ms 13 ms bras13-l0.lsanca.sbcglobal.net [151.164.186.13]

3 11 ms 11 ms 17 ms 12.83.97.13
4 72 ms 43 ms 92 ms 12.122.84.69
5 47 ms 46 ms 46 ms ix-15-0.tcore1.LVW-LosAngeles.as6453.net [66.110.59.41]
6 93 ms 92 ms 92 ms if-3-2.tcore1.PDI-PaloAlto.as6453.net [66.198.127.25]
7 93 ms 91 ms 98 ms if-2-2.tcore2.PDI-PaloAlto.as6453.net [66.198.127.2]
8 91 ms 90 ms 90 ms if-11-3.tcore2.CT8-Chicago.as6453.net [66.198.144.58]
9 91 ms 90 ms 90 ms if-3-2.tcore1.W6C-Montreal.as6453.net [66.198.96.45]
10 91 ms 92 ms 92 ms 66.198.96.50
11 92 ms 90 ms 92 ms 192.34.76.2
12 92 ms 92 ms 92 ms 199.91.189.234
13 112 ms 90 ms 90 ms 199.91.189.57

Trace complete.

[ZohnoReecho]

I was wondering what was wrong with my connection... I just did the trace... and it isnt pretty..

Tracing route to 199.91.189.57 over a maximum of 30 hops

1 3 ms 1 ms 1 ms home [192.168.1.254]
2 13 ms 13 ms 13 ms bras13-l0.lsanca.sbcglobal.net [151.164.186.13]

3 11 ms 11 ms 17 ms 12.83.97.13
4 72 ms 43 ms 92 ms 12.122.84.69
5 47 ms 46 ms 46 ms ix-15-0.tcore1.LVW-LosAngeles.as6453.net [66.110.59.41]
6 93 ms 92 ms 92 ms if-3-2.tcore1.PDI-PaloAlto.as6453.net [66.198.127.25]
7 93 ms 91 ms 98 ms if-2-2.tcore2.PDI-PaloAlto.as6453.net [66.198.127.2]
8 91 ms 90 ms 90 ms if-11-3.tcore2.CT8-Chicago.as6453.net [66.198.144.58]
9 91 ms 90 ms 90 ms if-3-2.tcore1.W6C-Montreal.as6453.net [66.198.96.45]
10 91 ms 92 ms 92 ms 66.198.96.50
11 92 ms 90 ms 92 ms 192.34.76.2
12 92 ms 92 ms 92 ms 199.91.189.234
13 112 ms 90 ms 90 ms 199.91.189.57

Trace complete.

12.122.84.69 is AT&T's and looks like has a variable ping. You could try calling them.

[Rapture]

Traceroute taken just now. The massive latency spikes only on S-E's network have returned.

1 <1 ms <1 ms <1 ms 192.168.1.254
2 <1 ms <1 ms <1 ms 001.131-78-65.ftth.swbr.surewest.net [65.78.131.1]
3 <1 ms <1 ms <1 ms 172.22.4.13
4 <1 ms <1 ms <1 ms 172.21.1.9
5 <1 ms <1 ms <1 ms 172.21.0.254
6 1 ms 1 ms 1 ms te-7-1.car1.Sacramento1.Level3.net [4.53.200.37]
7 1 ms 1 ms 1 ms ae-11-11.car2.Sacramento1.Level3.net [4.69.132.150]
8 4 ms 4 ms 4 ms ae-4-4.ebr2.SanJose1.Level3.net [4.69.132.158]
9 4 ms 4 ms 4 ms ae-62-62.csw1.SanJose1.Level3.net [4.69.153.18]
10 4 ms 4 ms 4 ms ae-1-60.edge1.SanJose3.Level3.net [4.69.152.16]
11 5 ms 5 ms 5 ms te0-3-0-2.ccr21.sjc03.atlas.cogentco.com [154.54.13.49]
12 6 ms 6 ms 6 ms be2000.ccr21.sjc01.atlas.cogentco.com [154.54.6.105]
13 7 ms 7 ms 7 ms be2166.mpd21.sfo01.atlas.cogentco.com [154.54.28.69]
14 50 ms 50 ms 51 ms be2134.mpd21.mci01.atlas.cogentco.com [154.54.6.38]
15 56 ms 56 ms 57 ms be2157.ccr22.ord01.atlas.cogentco.com [154.54.6.118]
16 81 ms 80 ms 81 ms be2080.ccr22.yyz02.atlas.cogentco.com [154.54.42.6] - Toronto, Canada
17 83 ms 83 ms 83 ms be2090.ccr21.ymq02.atlas.cogentco.com [154.54.30.206] - Montreal, Canada
18 83 ms 82 ms 82 ms 38.122.42.34 - Montreal, Canada (Ormuco Fiber bandwidth provisioning from Cogent)
19 * * * Request timed out.
20 142 ms 186 ms 146 ms 192.34.76.2 - Montreal, Canada (Ormuco Fiber bandwidth provisioning to Square Enix NA Datacenter)
21 152 ms 157 ms 156 ms 199.91.189.234 - Montreal, Canada (Square Enix NA Datacenter)
22 156 ms 151 ms 153 ms 199.91.189.39 - Montreal, Canada (Square Enix NA Datacenter, FFXIV Behemoth Server)

Nominally these last three hops @ S-E's NA Datacenter are ~82ms as can be seen from my some of my previous traceroutes. This same latency spike starting from 192.34.76.2 Ormuco Fiber [AS20324] can be seen from Level 3 backbone's 10GigE+ Looking Glass trace: http://imageshack.us/a/img716/3203/r5h5.png

[Jeize]

I live in SoCal (Los Angeles), have TWC, and experience delays in my game which make Titan hardmode annoying because landslides and plumes can be hit or miss. I was googling ways to fix it and some suggested battleping and pingzapper, but I really didn't want to pay anything extra. So, I fiddled with the system configuration - graphic settings. It was set to Maximum or High/Desktop. I changed it to Standard/Desktop, and it made a HUGE difference for me. The game is actually playable! Hopefully, this helps some of you guys too.

[kurgan78]

Just FYI, I live in Florida, and I intermittently have these same routing lag issues when others in the game are not having any problems. It's the exact same symptoms, and traceroutes seem to show a network somewhere between my ISP and SE's data center as the source of the bottleneck (usually as6453.net, IIRC).

I got sick of the lag I was getting yesterday and gave WTFast trial a shot, and the problem went away immediately. However, after several hours, the WTFast proxy in FL that I was using started experiencing the same lag problem. After logging out and switching to a different WTFast proxy in GA the lag went away immediately.

Just so you SoCal guys know, it's not just SoCal. Affected here in SoFL, too!

[StrikeOut]

Well I am affected. I have WebPass in Southern California and the Palo Alto jump is killing my connection. This is horrible.

[Varra]

Also having this issue. SoCal - TWC. It hasn't gone away for over a month. SE please help.

[Rapture]

This is still an occasional issue, as I got a 1 second long pause->fast_forward lag around 8PM PST last night in Coil when no one else did. I was the only one from California during that run, but notably when I (in Sacramento) run with someone from the San Fransisco area on a completely different ISP, we always seem to have lag issues at the exact same time when no one in our party from other geographic locations have them. When those people on the East Coast NA have lag, we don't lag...

Yet again, these are occasional lags which usually coincide with peak times. No one in our group has chronic lag 24/7 caused by their ISP. The issue still seems to be related to the poor choice of backbones S-E is using for that NA datacenter, unless the S-E servers themselves have some sort of packet processing issue under heavy load that causes responses to become queued/delayed.

[Boatrider]

So, what I got from this thread is in order for my internet problems to go away, one or more of the following needs to happen:

• Tata Communications to fix their crap -
• Use a VPN service -
• Move the FFXIV data center to SoCal -

[Saikou]

What I can derive from what we know, the edits to the OP and the data that has been shown: Once your connection enters TATA airspace, ping goes to hell. Even if your connection doesn't go thought TATA, your ping will still go to hell once it reaches Montreal and likely even before that at whichever backbone connection is getting you into Canada. The problem with, er, the problem being TATA is that they do not talk to peons like us. It would take our ISP's or SE talking to TATA to get them to work on the packet loss issues with the routes controlled by them. And even if TATA fixed its issues, there's still the problem of severe packet loss happening at the datacenter itself.

I live in Washington State and my ISP is Comcast.

I experience positional update lag and input lag all the time. Occasionally there is rubberbanding. As an example of positional update lag, I can almost never avoid magitek cannon in Castrum Meridianum or Keeper of Halidom's targeted AOE in Wanderer's Palace.

As you can see, my latency isn't anywhere near as bad as the OP or anyone else in SoCal being routed through TATA's Palo Alto or San Jose routers. It is, however, notably worse than my experience with WoW, SWTOR, Rift, and Terra. In battles where seconds matter, like Titan and Binding coil, this is very frustrating.

Based on the following:

http://www.reddit.com/r/ffxiv/commen...th_lag_repost/

http://forums.timewarnercable.com/t5...rnia/td-p/9059

Some of the first chokepoints are owned by consumer grade ISP's (comcast, time warner, fios, etc). But the bigger problem comes with the routes controlled by business tier ISP's: TATA, Cogentco, and Tinet. Almost everybody is running into a router inside their network where packet loss, and therefore positional lag (hit by voidzones you weren't in), input lag (you hit the button and the ability doesn't go off until several seconds later), and rubberbanding (the game seems to freeze and then all the actions that happened in the frozen period happen rapidly as your computer catches up to everything it missed), is happening.

I fear that SE's intervention may be our only hope for getting both tiers of ISPs to do something, as most of us who have either been brushed off or given the run-around.

My own Tracert

Tracing route to 184.107.107.176 over a maximum of 30 hops

1 4 ms 4 ms 3 ms 192.168.1.1
2 11 ms 12 ms 11 ms 73.104.212.1
3 11 ms 14 ms 17 ms 68.85.240.37
4 16 ms 17 ms 15 ms ae-41-0-ar03.burien.wa.seattle.comcast.net [69.1
39.164.29]
5 87 ms 34 ms 15 ms ae-41-0-ar03.seattle.wa.seattle.comcast.net [69.
139.164.33] ((Comcast is having data loss issues of it's own))
6 17 ms 16 ms 15 ms he-1-12-0-0-11-cr01.seattle.wa.ibone.comcast.net
[68.86.93.177]
7 14 ms 16 ms * ix-7-2-2-0.tcore1.00S-Seattle.as6453.net [64.86.
123.33] ((Entering TATA airspace...))
8 18 ms 19 ms 20 ms if-13-0-0-4.core1.00S-Seattle.as6453.net [64.86.
124.13]
9 89 ms 87 ms 87 ms if-8-1-2-4.tcore2.CT8-Chicago.as6453.net [64.86.
124.22] ((My first major packet loss with TATA doesn't happen until we get to Chicago))
10 88 ms 96 ms 88 ms if-3-2.tcore1.W6C-Montreal.as6453.net [66.198.96
.45]
11 90 ms 90 ms 90 ms 66.198.96.58
12 89 ms 90 ms 89 ms te7-4.dr9.mtl.iweb.com [184.107.1.110]
13 88 ms 91 ms 97 ms 72.55.128.44
14 92 ms 92 ms 88 ms 184.107.107.176 ((Latency is still in hell on all the hops between Chicago and the datacenter, leading to the conclusion that there are still problems on SE's end))

Trace complete.