Experiencing problems with a specific MMO, traceroute data including comparison with TWC NY/NJ

[kelstonb]

I believe I am running 75/75 Verizon FIOS out of Bayside, NY (11361). Let me start by saying that I know Verizon FIOS has no issues with gaming that I know of as a policy as i've used Verizon FIOS to play World of Warcraft for years without a hitch as well as numerous other games including other MMOs.

However, recently, I picked up the MMO run by Square Enix. Final Fantasy XIV: A Realm Reborn. Their NA datacenter appears to be located in Montreal and operated by their partner Eidos Montreal. Now, I know also that Verizon has said they have no control over hops that don't belong to them which is why i'm bringing in the extra data from other users from other ISPs that also have performed traceroutes at my request that end up being routed through the same routers that I am and yet does not experience the problem Verizon FIOS users do. This would only indicate that the problem is on Verizon's end specifically to do with this application and/or the peering agreements Verizon has with vendors along the path to the Montreal datacenters.

Here is my WinMTR report to the datacenter during primetime while I am experiencing unplayability: https://i.imgur.com/pzXfHFD.png

I want you to note hops 6 through 14 indicating 10-15% packetloss. Most of these hops belong to TATA Communications.

First, I want to present a WinMTR report to the datacenter of a Roadrunner customer also in NYC: https://i.imgur.com/tA6WE1k.png

Take note that after he leaves Roadrunner's backbone, he runs through exactly the same TATA routers as I do from NY through Chicago through Montreal. All the same but no packetloss. Nothing. His connection is clean.

Here is a WinMTR report to the datacenter during primetime from a Roadrunner customer in Southern California: https://i.imgur.com/C5S8QPQ.png

While you can notice that most of the route isn't the same, you can notice that once it hits TATA's Montreal server, the route is the same. You'll also notice that despite this, the Montreal router onwards does not experience 10-16% packetloss despite being the exact same routers that i'm routed through as a Verizon FIOS customer.

Now here is where it gets interesting because here is a WinMTR report to the datacenter from a Verizon FIOS customer located in Dallas: https://i.imgur.com/cKsLHcX.png

You'll notice that once it hits the Chicago hop, he's routed exactly the same as I am, and yet, he doesn't experience the packetloss issues that I seem to experience.

Before you point the finger at TATA Communications, allow me also to submit a FIOS user from Rhode Island: https://i.imgur.com/X1bT9mn.jpg

What you'll note is that rather than be routed through TATA to the Montreal datacenter, instead he's routed through Level3, the vendor Verizon made a special bilateral agreement with. And the result is that rather than packetloss, he gets all of his hops spiking to 500ms.

So you'll say it might be a hardware issue with my ONT that you'll need to reset or maybe my Verizon router is problematic and send me another one, but I ask you to hold that thought for just a second!

I collected this data from a thread I started on Reddit hoping to get to the bottom of this: https://www.reddit.com/r/ffxiv/comments/3ea6a5/do_you_lag_do_you_not_lag_use_fios_dont_use_fios/

What you'll find if you peruse this thread is that a large number of East Coast FIOS users also experience the same problems I do. Here's a quote from user Lanbatal: "Northern VA, FiOS. Game is unplayable without a VPN. The number of lost packets is ridiculous."

Here's a ping report from another NoVA FIOS user SMYFFL reporting 14% packetloss: https://i.imgur.com/7lPIBFz.png

So from my end, what I can gather appears to be an issue with Verizon FIOS. It could be that Verizon FIOS has accidentally or inadvertently flagged FFXIV traffic as P2P traffic and throttles it during primetime in extremely high usage areas (which explains why NYC/NoVA area that routes through NY/Boston sees this but the Dallas guy doesn't). What I and likely the other customers ask isn't for Verizon to admit guilt but rather to finally figure out why this is happening only to this one specific application and only to Verizon customers and to help us use the service we've paid for.

I've gone through Verizon chat support before and got the standard song and dance of pointing the blame at the game itself. If Verizon wants to say it is the fault of the game's servers or is the fault of the routers along the pathway to the Montreal datacenter that hosts the NA servers for the game or if it's just my computer, Verizon needs to answer some simple questions:

1) Why doesn't Time Warner NY that goes through the same pathway see the exact same issue if it's the pathway?

2) Why doesn't Verizon FIOS Dallas or Time Warner NY that goes to the same datacenter see the exact same issue if it's the game?

3) If it's just my computer, why does Northern VA FIOS customers, other NY FIOS customers, customers from Massachusetts, and FIOS Florida also see this issue? Are they running my computer?

[smith6612]

Great comparison here! Definitely helps to bring out the issues that many of us have been suffering from for a long, long time.

[Joey-Zaza]

This topic needs more attention.  FFXIV lag for several months now!

All during peak hours.  I too go through LEVEL3.

I agree with your statement here:

It could be that Verizon FIOS has accidentally or inadvertently flagged FFXIV traffic as P2P traffic and throttles it during primetime in extremely high usage areas (which explains why NYC/NoVA area that routes through NY/Boston sees this but the Dallas guy doesn't). What I and likely the other customers ask isn't for Verizon to admit guilt but rather to finally figure out why this is happening only to this one specific application and only to Verizon customers and to help us use the service we've paid for.

I am in New Jersey, for the record.

You know what drives me crazy is that I have done EVERYTHING that I can to fix this issue on my own.  SE is not responsible for routing to their servers, that is an ISP thing.  But when you call up they ask you to reset your router and walk you through all these low tech steps like we're morons.  When they as for tracerts they conclude that it's past the 5th hop and they can't do anything about it.  But here it is plain as day.  This is right when the lag first started, it got worse later.  I took out the first few as they as irrelevent.

4 * * * Request timed out.
5 10 ms 10 ms 9 ms 0.ae4.XL4.NYC1.ALTER.NET [140.222.226.37]
6 9 ms 9 ms 8 ms 0.ae2.BR1.NYC1.ALTER.NET [152.63.19.198]
7 * * * Request timed out.
8 194 ms 166 ms 219 ms ae-11-11.car2.Montreal2.Level3.net [4.69.141.1]

9 311 ms 447 ms 434 ms ae-11-11.car2.Montreal2.Level3.net [4.69.141.1]

10 35 ms 31 ms 31 ms ORMUCO-COMM.car2.Montreal2.Level3.net [4.59.178.
74]
11 43 ms 41 ms 44 ms 192.34.76.10
12 44 ms * 41 ms 199.91.189.242
13 * * * Request timed out.
14 * * * Request timed out.
15 * * * Request timed out.
16 * * * Request timed out.
17 * * * Request timed out.
18 * * * Request timed out.
19 * * * Request timed out.
20 * * * Request timed out.
21 * * * Request timed out.
22 * * * Request timed out.
23 * * * Request timed out.
24 * * * Request timed out.
25 * * * Request timed out.
26 * * * Request timed out.
27 * * * Request timed out.
28 * * * Request timed out.
29 * * * Request timed out.
30 * * * Request timed out.

Trace complete.