Important Information Regarding North American Data Center Relocation

[Valkyrie_Lenneth]

They also have been complaining for years, and so did Euros before their DC was moved.
Almost like 200ms being perfectly playable is a lie.
130ms on the NA data center.
110ms on the EU data center.
Guess I'm EU now.

I dont recall seeing much complaining. I certainly never did. I moved from Florida to Washington State, and barely saw a ping increase. Most ISP's can't route network traffic for crap.

[KisaiTenshi]

I dont recall seeing much complaining. I certainly never did. I moved from Florida to Washington State, and barely saw a ping increase. Most ISP's can't route network traffic for crap.

Most Tier 1 backbones have no problem routing things. The problem is that large ISP's like AT&T, Verizon and Comcast own their own backbones, so it's cheaper for them to not use any peers until the last hop possible. However many of these Tier 1 ISP's are also really evil to each other and have been known to route traffic over unfair peering agreements. Cogent and level3 are known for "breaking the internet"

eg:
http://www.telecompetitor.com/comcas...pute-resolved/
https://arstechnica.com/business/201...ernet-quality/

It's important to note that level3 and cogent are usually the villains in peering disputes. In the case of Square-Enix, they primarily use NTT's backbone but that tends to still require people to peer over at least one other backbone.

Like using my non-residential connections, a connection from hurricane electric (Fremont California) gets 5ms to FFXIV's gateway servers, but has to over telia in San Jose. My connection from Arizona goes over Cogent at Phoenix NAP. My local connection goes over Level3 at Seattle.

If I use a few lookingglass systems:

3 60 ms 59 ms 77 ms ae-3.r20.miamfl02.us.bb.gin.ntt.net (129.250.2.110)
4 87 ms 87 ms 87 ms ae-4.r23.asbnva02.us.bb.gin.ntt.net (129.250.2.86)
5 87 ms 88 ms 86 ms ae-10.r22.snjsca04.us.bb.gin.ntt.net (129.250.6.237)

So Miami is going to Ashburn, Virginia before it's routed to San Jose, California. So needless to say that's several hundred miles in the wrong direction. Most other route paths go via Chicago, San Jose or directly to Los Angeles.

The only people who should have noticed a drastic change are those who live in the central-california area (where the servers now are) or the people who live in Montreal (where the servers were), everyone else would have been routed through the same gateway systems and with a few exceptions, likely still had the shortest best path.

[Mature]

An update about what is getting done or what's gonna get done would be fair not only for me but for the thousands having the same issue.

[Domeric]

The only people who should have noticed a drastic change are those who live in the central-california area (where the servers now are) or the people who live in Montreal (where the servers were), everyone else would have been routed through the same gateway systems and with a few exceptions, likely still had the shortest best path.

I live in the midwest and I see a patented difference in raids and in hunts or fates where a great many people have arrived. I didn't have this problem when the servers were in canada.

[asheryon]

Guess this is it for me, i would have taken some ping increase, but it went from 165-180 to 270. they mostly killed the south american players with this, shouldnt had bought SB damn it.

[Alistaire]

This thread got pointed out on reddit today, and I can't believe I missed this one quote:

FINAL FANTASY XIV was designed to be played in an environment featuring ping times of 200 – 220 milliseconds or less.

This is so blatantly false it's not funny. In 2.0 (when the game was designed), your character status was updated only every .3 seconds in instances. This was eventually changed to .1, but there's no way you can have us believe the game was designed to have a 50% chance of not updating your character for nearly half a second with the way aoe moves were designed in the game. A design of working under 200-220 would have at least guaranteed an update somewhere within that timeframe, while .3 seconds guarantees some of the time you're not updating for a whole round trip of data.

[Princess_Momoka]

This thread got pointed out on reddit today, and I can't believe I missed this one quote:

FINAL FANTASY XIV was designed to be played in an environment featuring ping times of 200 – 220 milliseconds or less.

This is so blatantly false it's not funny. In 2.0 (when the game was designed), your character status was updated only every .3 seconds in instances. This was eventually changed to .1, but there's no way you can have us believe the game was designed to have a 50% chance of not updating your character for nearly half a second with the way aoe moves were designed in the game. A design of working under 200-220 would have at least guaranteed an update somewhere within that timeframe, while .3 seconds guarantees some of the time you're not updating for a whole round trip of data.

That is why a lot of people are upset about this move. Now with people being hit with an increase ping, more people see the problems now and being outspoken about it in this thread. The servers do not update fast enough for the mechanics of the game. So it may be playable under optimal conditions, but once you see a ping spike or heck, most real world ping values, you see problems. THEN you had the high delays in some abilities that are hard coded on top of that, it makes it very frustrating to some people.