Game has become unplayable due to latency since the 18th

[Chap_]

Hello,

Writing here as a last resort before giving up.

I play on PS4 and had played this game for 6 weeks without one issue of latency. Ever since around the 18th this game has become very close to unplayable. Progressing in Level 50 content is impossible. Every minute, I will freeze for 5-10 seconds, and then all at once everything catches back up. This has resulted in countless deaths, and all other kind of frustrations. I can't interact with things for 5-10 seconds. Have to keep trying over. I've managed to use a limit break 3 times on my screen. All of my GCD moves will be used on my end and the timer resets but they don't get applied. Forget doing my Mudras. The game has become an absolute nightmare to load up.

I have 50 mbps download speeds and 18 mbps upload speeds. I can connect to share play on the PS4 and play FFXIV on a PS4 across the country lag free. I can't run an extra program as i've read suggested since I am on PS4 and I don't think I should be required to.

Is there anything that can be done before I have to cut my losses and stop paying for a game that is virtually unplayable?

Thank you

[Raist]

EDIT:
Just noticed in the Lodestone there is a message that Ormuco has maintenance scheduled for tonight. No details on it, but it could <hopefully> be something to address some of the potential bottlenecking we've seen in the past around the peering exchanges and/or VLAN's. They've gone through this before and were able to eliminate some congestion issues in the past. Keep your fingers crossed!

could simply be a stability problem upstream from your location, before you get to SE. TATA recently had a bout of big problems with routing. Level3 frequently has issues. Cogent is hit and miss... but basically, there are recurring issues with all three of them (as well as the other two top tier ISP's Ormuco peers with up there in Canada). Your ISP has agreements with these companies for routing you from their networks to Ormuco (SE's ISP in Canada). They also use some of them as well as other ISP's for getting us to Tokyo for the servers we hit there, as well as Sony's servers.

You'd need to get some diagnostics run against the routes to test for issues though. If you aren't comfortable doing it yourself, you should be able to get your ISP to run some tests for you. You'll need the addresses for some of SE's servers though, so I'll provide a list of the registered DNS names for the lobby servers that can be used for investigation. They aren't the IP's of the game servers, but they are within the subnet used, and will provide them information for gathering more data like the ASN's and such to conduct a more proper investigation if they deem it necessary. I will also provide a link to a post from Blizzard about running some basic tests yourself that can be applied to these server addresses in kind. You may in fact want to run tests against some of the addresses provided by Blizzard as well as to other known services like Google/YouTube, Netflix, Hulu, TwitchTV, or other games to have more data for comparison. It should help build a case for routing issues if you find significant variations between the different traces and such.

SE's lobby servers in Canada:
neolobby02.ffxiv.com
neolobby04.ffxiv.com
neolobby06.ffxiv.com

SE's lobby servers in Japan:
neolobby01.ffxiv.com
neolobby03.ffxiv.com
neolobby05.ffxiv.com

***note some of these assignments may change with introduction of the new EU based servers coming with 3.0

Blizzard's Article on testing connectivity issues with their servers:
https://us.battle.net/support/en/art...g-a-traceroute

[Chap_]

While I do appreciate the reply, you paste the same thing into everyones post. EVERYONE is having this issue on this forum. I understand you say its out of SE control but when a good chunk of the FFXIV population is experiencing these issues, clearly SE needs to take control. No other game has these issues.

[Chap_]

traceroute to neolobby02.ffxiv.com (199.91.189.74), 64 hops max, 52 byte packets
1 zxhn_h368c.home (192.168.200.1) 2.606 ms 1.076 ms 0.916 ms
2 nw1-dsl-208-102-244-1.fuse.net (208.102.244.1) 3.486 ms 3.496 ms 3.933 ms
3 172.17.122.34 (172.17.122.34) 4.150 ms 3.511 ms 7.266 ms
4 ws-zt-1.cin1.core.fuse.net (216.68.14.52) 4.920 ms 8.733 ms 3.575 ms
5 cin1.eve1.core.fuse.net (216.68.14.41) 28.363 ms 22.217 ms 24.225 ms
6 216.68.14.95 (216.68.14.95) 21.734 ms 26.523 ms 21.460 ms
7 xe-4-2-0.was14.ip4.gtt.net (173.241.131.157) 21.688 ms 25.074 ms 20.794 ms
8 xe-9-0-0.mtl10.ip4.gtt.net (141.136.105.162) 39.103 ms
xe-7-3-0.mtl10.ip4.gtt.net (141.136.106.106) 31.330 ms
xe-9-0-0.mtl10.ip4.gtt.net (141.136.105.162) 67.542 ms
9 ormuco-gw.ip4.gtt.net (216.221.156.110) 34.529 ms 36.240 ms 34.930 ms
10 * * *
11 * * *
12 * * *
13 * * *
14 * * *
15 * * *
16 * * *
17 * * *
18 * * *
19 * * *
20 * * *
21 * * *
22 * * *
23 * * *
24 * * *
25 * * *
26 * * *
27 * * *
28 * * *
29 * * *
30 * * *
31 * * *
32 * * *
33 * * *
34 * * *
35 * * *
36 * * *
37 * * *
38 * * *
39 * * *
40 * * *
41 * * *
42 * * *
43 * * *
44 * * *
45 * * *
46 * * *
47 * * *
48 * * *
49 * * *
50 * * *
51 * * *
52 * * *
53 * * *
54 * * *
55 * * *
56 * * *
57 * * *
58 * * *
59 * * *
60 * * *
61 * * *
62 * * *
63 * * *
64 * * *
neolobby04.ffxiv.com
traceroute to neolobby04.ffxiv.com (199.91.189.93), 64 hops max, 52 byte packets
1 zxhn_h368c.home (192.168.200.1) 2.582 ms 1.198 ms 0.917 ms
2 nw1-dsl-208-102-244-1.fuse.net (208.102.244.1) 2.629 ms 2.363 ms 2.809 ms
3 172.17.122.34 (172.17.122.34) 5.411 ms 3.109 ms 2.714 ms
4 ws-zt-1.cin1.core.fuse.net (216.68.14.52) 4.309 ms 5.391 ms 5.928 ms
5 cin1.eve1.core.fuse.net (216.68.14.41) 23.689 ms 21.900 ms 22.458 ms
6 216.68.14.95 (216.68.14.95) 22.096 ms 22.140 ms 23.749 ms
7 xe-4-2-0.was14.ip4.gtt.net (173.241.131.157) 22.613 ms 22.128 ms 22.116 ms
8 xe-9-0-0.mtl10.ip4.gtt.net (141.136.105.162) 34.543 ms 35.136 ms 37.571 ms
9 ormuco-gw.ip4.gtt.net (216.221.156.110) 35.904 ms 35.786 ms 35.011 ms
10 * * *
11 * * *
12 * * *
13 * * *
14 * * *
15 * * *
16 * * *
17 * * *
18 * * *
19 * * *
20 * * *
21 * * *
22 * * *
23 * * *
24 * * *
25 * * *
26 * * *
27 * * *
28 * * *
29 * * *
30 * * *
31 * * *
32 * * *
33 * * *
34 * * *
35 * * *
36 * * *
37 * * *
38 * * *


What would an ideal traceroute to that server look like? regardless after 3 hours on the phone with my ISP they say they are only responsible for the first five hops

[Chap_]

here is my route to my server goblin

traceroute to 199.91.189.26 (199.91.189.26), 64 hops max, 52 byte packets
1 zxhn_h368c.home (192.168.200.1) 4.918 ms 2.028 ms 2.003 ms
2 nw1-dsl-208-102-244-1.fuse.net (208.102.244.1) 2.719 ms 5.188 ms 2.937 ms
3 172.17.122.34 (172.17.122.34) 8.397 ms 7.224 ms 5.644 ms
4 ws-zt-1.cin1.core.fuse.net (216.68.14.52) 5.126 ms 4.554 ms 3.591 ms
5 cin1.eve1.core.fuse.net (216.68.14.41) 22.019 ms 23.145 ms 24.524 ms
6 216.68.14.95 (216.68.14.95) 22.781 ms 23.843 ms 22.444 ms
7 xe-4-2-0.was14.ip4.gtt.net (173.241.131.157) 21.806 ms 22.875 ms 22.007 ms
8 xe-9-0-0.mtl10.ip4.gtt.net (141.136.105.162) 110.147 ms
xe-1-2-0.mtl10.ip4.gtt.net (89.149.186.217) 118.324 ms
xe-1-3-0.mtl10.ip4.gtt.net (89.149.184.74) 109.815 ms
9 ormuco-gw.ip4.gtt.net (216.221.156.110) 113.636 ms 103.463 ms 103.138 ms
10 * * *
11 * * *
12 * * *
13 * * *
14 * * *
15 * * *
16 * * *
17 * * *
18 * * *

What stops are after the 9th one? How do I connect to these?

[Raist]

While I do appreciate the reply, you paste the same thing into everyones post. EVERYONE is having this issue on this forum. I understand you say its out of SE control but when a good chunk of the FFXIV population is experiencing these issues, clearly SE needs to take control. No other game has these issues.

Not true. Many other games have reported these EXACT same issues for nearly 20 years. It happens off and on, but if you did a little googling you will find issues cropping up for many games and see some common names keep showing up---Level3, TATA (AS6453), Cogent, GTT/TiNet/SPA, XO, and a few others. It will also be bound to some popular companies as well--like EA and Blizzard. TWC has had several runs on it's forums in the past year for WoW, Diablo3, GW2, LotR, LoL... just to name a few. And if it wasn't something funky on TWC's end, it was most frequently pointing at Level3 or TATA, occasionally Cogent.

Go ahead and google this phrase: "WoW high latency and disconnects" and notice how many hits you get--north of 20,000. Notice how many were fairly recent. You may also notice that while you type it auto-fill may try to point you to a query with THIS year in it. FFXIV is NOT some special snowflake with this issue.

Here is a link, just in case you can't be arsed with actually doing the query your self:
https://www.google.com/?gws_rd=ssl#q...%20disconnects
The top hit is a link to the exact page from Blizzard I posted earlier...updated Jan. 27, 2015. Take some time to read posts linked directly from that query. You will see some going back as far as 2011 that point the finger at specific ISP's that were having issues---and it wasn't always Blizzard or Microsoft at fault, but you will see it being discussed how it is the ISP's involved along the route.

Than there are all the filings at the FCC from both TATA and Level3 about ISP's not operating under fair practices, the Comcast/Netflix scandal... all the things that have led up to the great Net Neutrality debate... there is basically a big pissing contest amongst the ISP's. More specifically, our ISP's not wanting to expend the resources upstream from our modems to properly support all the bandwidth they've been selling to us locally.

And... yes... I do post it a lot, because I have a lot of experience with the issue. A lot of experience resolving the issue through my ISP I might add. If you truly are aware of all my postings on the matter, that should be readily apparent... as I and others have attested to the fact that going through our ISP improved the situation for many, and there are others that have asserted that when they can't make headway with their ISP they improve things by using a VPN. In either scenario, it affirms that there are routing issues in play--because that is what we are changing that is showing improvement.

As for the traces you recently posted just after the post I am replying to, there appears to be an issue at/around the exchange point from GTT/TiNet into Ormuco's network. That is the routing partner YOUR ISP has assigned to you for getting into Ormuco. It is a group that has been seen having issues frequently in the past. Haven't done a dig on your ISP yet, but they should have an agreement with at least one of the 4 other ISP's Ormuco peers with. If they are not willing to try to address any potential issues with the the GTT group, they can at least try to alter your routing to get around that. We have very recently seen more evidence of issues with specific ISP's. In just the past few days, there were posts about people not getting on at all with their ISP, but got on with a VPN. Others reported they got back on after resetting their modems and getting assigned new IP addresses (effectively, they forced a DHCP lease renewal--probably changed their gateway also). It was just confirmed there are outages tied to Charter that are currently causing problems as well. These all support the notion there are problems in route to the servers that our ISP's need to get involved with to address properly.

Edit:
Ok, just looked up an ASN for fuse.net, and they have peering with many other ISP's---one of them is Cogent, who Ormuco peers directly with. This was a lookup based on one of their name servers, so there is at least one potential alternate route that can be taken:

AS3257 Tinet SpA
AS2828 XO Communications
AS174 Cogent Communications
AS6939 Hurricane Electric, Inc.
AS30023 CyrusOne LLC
AS5580 TripartZ B.V.
AS9002 RETN Limited

***I did a second look up on your gateway IP, and it was in the same subnet so it returns the same ASN, so the IP4 peer list does apply to your connection as well.

If you look further into their peering arrangements, you will see that in a bit of a round-about way they can also get to TATA (AS6453), as well as Level3 (AS3459, AS3356)--two other ISP's that Ormuco peers with directly. FUSE.net also peers directly with Hurricane, which for some regions could open up a host of other paths into several of the 5 peers that Ormuco uses. So your ISP does in fact have options for helping to resolve an alternate path for you if they aren't going to work with GTT/TiNet on your behalf--even though they should because they have entered into a peering arrangement with them and that is who your ISP has put you with to get to Ormuco.



Source for all of this info: http://bgp.he.net/AS6181#_asinfo

Edit2:
EEP! Just noticed your ISP looks to be tied in with Cincinnati Bell. While they don't seem to get as many reports as others in Ohio (AT&T and TWC are a MESS)... Ohio itself has been fraught with issues for many ISP's for some time now. Infrastructure problems abound throughout the region, more specifically to the north and east. Unfortunately that is likely where you are getting routed through.

[Chap_]

Thats good info. Thanks for taking all of the time to do that for me, correcting me, and my apologies about that.

After multiple 2+ hour long conversations with my ISP, they say they can't control my route of traffic after I leave their network. Nothing can be done on their end.

Unable to even craft really at this point. Here is my current route for my server

traceroute to 199.91.189.26 (199.91.189.26), 64 hops max, 52 byte packets
1 dd-wrt (192.168.1.1) 2.366 ms 0.868 ms 0.836 ms
2 zxhn_h368c.home (192.168.200.1) 1.380 ms 1.078 ms 1.662 ms
3 nw1-dsl-208-102-244-1.fuse.net (208.102.244.1) 2.939 ms 2.220 ms 6.261 ms
4 172.17.122.34 (172.17.122.34) 3.378 ms 9.491 ms 4.288 ms
5 ws-zt-1.cin1.core.fuse.net (216.68.14.52) 2.718 ms 2.241 ms 3.178 ms
6 cin1.eve1.core.fuse.net (216.68.14.41) 21.930 ms 21.664 ms 21.289 ms
7 216.68.14.95 (216.68.14.95) 22.695 ms 21.356 ms 21.015 ms
8 xe-4-2-0.was14.ip4.gtt.net (173.241.131.157) 20.753 ms 21.835 ms 21.861 ms
9 xe-0-0-0.mtl10.ip4.gtt.net (89.149.187.226) 117.083 ms 110.989 ms 110.901 ms
10 ormuco-gw.ip4.gtt.net (216.221.156.110) 113.066 ms 111.461 ms 110.952 ms

89.149.187.226 being Tinet SpA like you had mentioned. It says this server is located in germany when I use google?

Is there anything else I should try since my ISP will not help me? I've tried to setup a PPTP vpn on my router and failed.

[Raist]

Hate to say it... but it may be time to shop around if you can. Perhaps the threat of loosing a customer will put a fire under their butts. There are other ISP's that lease lines from the locally installed ISP's. Not sure who is providing coverage in your area though. Earthlink may be an option. They lease access from TWC here. They even bill through TWC's system as well--the funny thing is it is $3 cheaper. Go figure. Only reason I'm still with TWC for at least one more year is that when they tried to jack my rates, I told them I was switching and they dropped me to $1 under Earthlink's promo rate for the next 12 months. heheh.

CenturyLink or Exede may be viable candidates as well. IDK... could try plugging your zipcode into something like this site to get a list of companies:
http://internetprovidersbyzip.com/in...iders-zip-code

All I can say is my TWC changes my route regularly. At first, I had to contact Tier3 directly about it (providing trace logs, signal pages and event logs from my modem and such)... but lately they've been catching it just as the congestion starts getting nasty. Once my latency starts breaching the 200ms mark coming through the north east (typically around the Ashburn/DC or NY/NJ areas), the route gets flipped in the next day or so without me having to even shoot them an email on the matter. On average, it's about 2-3 weeks between updates. I've gone from Tata, to Cogent, to Level3, and back to Tata again since Valentines day weekend. My gateway has remained the same throughout all of that too (in the past, they also switched me to another area in the process--so I've appeared to be moving all over the state if you were to look at my geolocations).

Don't know the dirty details of just how it is done... but it is done quite often in the business. Maybe it's because you aren't talking to the right people. But we've seen them do this kind of thing quite regularly for people. One thing for certain though is that it has always required getting Tier3 involved to investigate the route (why the tracert helps so much--gives them the roadmap), and then they call for the needed escalation to get it addressed. Maybe you need to request a superviser or higher level technician or something. SOMEONE in that organization is responsible for assigning the metirc that is picking specifically GTT's service to get you to Ormuco. Whether that decision is based on BGP data or something else--there is a system for making that prediction/assumption, and there should be a means to either influence the outcome or outright override it somehow. The trick is getting in touch with someone who knows what is what.

[Chap_]

Hate to say it... but it may be time to shop around if you can. Perhaps the threat of loosing a customer will put a fire under their butts. There are other ISP's that lease lines from the locally installed ISP's. Not sure who is providing coverage in your area though. Earthlink may be an option. They lease access from TWC here. They even bill through TWC's system as well--the funny thing is it is $3 cheaper. Go figure. Only reason I'm still with TWC for at least one more year is that when they tried to jack my rates, I told them I was switching and they dropped me to $1 under Earthlink's promo rate for the next 12 months. heheh.

CenturyLink or Exede may be viable candidates as well. IDK... could try plugging your zipcode into something like this site to get a list of companies:
http://internetprovidersbyzip.com/in...iders-zip-code

All I can say is my TWC changes my route regularly. At first, I had to contact Tier3 directly about it (providing trace logs, signal pages and event logs from my modem and such)... but lately they've been catching it just as the congestion starts getting nasty. Once my latency starts breaching the 200ms mark coming through the north east (typically around the Ashburn/DC or NY/NJ areas), the route gets flipped in the next day or so without me having to even shoot them an email on the matter. On average, it's about 2-3 weeks between updates. I've gone from Tata, to Cogent, to Level3, and back to Tata again since Valentines day weekend. My gateway has remained the same throughout all of that too (in the past, they also switched me to another area in the process--so I've appeared to be moving all over the state if you were to look at my geolocations).

Don't know the dirty details of just how it is done... but it is done quite often in the business. Maybe it's because you aren't talking to the right people. But we've seen them do this kind of thing quite regularly for people. One thing for certain though is that it has always required getting Tier3 involved to investigate the route (why the tracert helps so much--gives them the roadmap), and then they call for the needed escalation to get it addressed. Maybe you need to request a superviser or higher level technician or something. SOMEONE in that organization is responsible for assigning the metirc that is picking specifically GTT's service to get you to Ormuco. Whether that decision is based on BGP data or something else--there is a system for making that prediction/assumption, and there should be a means to either influence the outcome or outright override it somehow. The trick is getting in touch with someone who knows what is what.

I live in Cincinnati- For me I can choose between Cincinnati Bell Fioptics (FTTH) or Time warner cable. Unfortunately the only tech support Cincinnati Bell offers is outsourced overseas. I have escalated my same concern 4 times in a row, basically refusing their reply of "its out of our control" but it seems like they're fine with the cycle continuing. I can't say the seem to concerned. I am told only they can speak with their engineers, but this is what they relay to me. I told them I would certainly look at my other options (TWC) but they just offered to transfer me to someone who would help with cancellation. I was trying to steer away from TWC due to it flat out just dropping completely during peak hours when I had it in the past at the same location. At least for me the Cincinnati bell fioptics is a constant connection.

When you say you contacted Tier3, how did you go about doing this? I am willing to try whatever I can. Is it typical for me to be bouncing overseas to a server located in Germany/Netherlands before coming back to North America? I feel so helpless! Thank you for your continued help. It is much appreciated.

[Raist]

The parent company is German, but it appears to be a node physically in Montreal. Geolocation data is a mess for that hop. Depending which database you hit, it comes up as Canada, France, and Germany. Considering the DNS name hints at Montreal, and there is a record from 3/31/15 showing it as Montreal...pretty safe to say it is in Montreal.

As for getting to Tier3... it can be a bit of a song and dance doing it over the phone sometimes. Typical scenario I hear all the time:

They'll want to send techs on site to check your modem/lines/splitters and verify good signal locally. Then speed test you...but only to a fairly close node within their network, of course. Then you'll try to demonstrate the problem to the technician while there, and of course the game behaves....call them back later the same night to complain again. The great hamster wheel of futility...after a couple tries you finally get someone that actually listens to you when you tell them for the umpteenth time that you have data showing high latency spikes and packet loss along a specific route, only during primetime, and you need to talk to someone in Tier3 about it. You can forward the reports to an email or FTP or something if they'll just give you the contact info. <static, cricket noises> "Have you tried reseting your modem?" <then the rage finally sets in and you find yourself in the market for a new phone>.

Seriously.. the 800 number can be your worst enemy sometimes. If your ISP has a support option at their website, you should get better results. Worst case scenario you may be able to gain some ground by emailing the information to the admin contact listed for one of the hops within your ISP's segments. Will see if I can find it again... want to say it was Scott Pennington at Cincinnati Bell. IT really shouldn't be so hard to get to the right people when you have done the legwork for them and have actionable data for them to work with. It's not like you're dropping your car keys at the service center and simply telling them "Fix it" and walking away without describing your symptoms/concerns. Practically telling them exactly where to look and what to look for when you give them a tracert.

Edit:
Oh boy... looks like you've got a screwy support system in general up there. May be why it's hard to get through. Finding multiple brands with different contacts. Will just start listing them, may be updating it a couple times before I'm done scraping these guys.

***Added some more direct email contacts I found after digging a bit more, but they aren't really the right guys to contact directly (well, aside from the generic hostmaster@fuse.net email--that should be appropriate for virtually any networking related contact really), but you might be able to go through them to get to the proper techs if the twitter/facebook or website contacts don't get you anywhere.

https://twitter.com/CincyBell
https://twitter.com/hashtag/CinBellCare

https://www.facebook.com/cincinnatibell/110376283088
https://www.facebook.com/cincinnatib...32387213452269

http://www.cincinnatibell.com/customer_support/
https://www.cincinnatibell.com/contact-us/

socialmedia@cinbell.com (referenced for ZoomTown subscribers)

Not seeing a specific contact for their Network Operations Center just yet... ARIN is just giving up the generic one, but it may get to someone who can make something happen just the same: hostmaster@fuse.net

You could give the guys a shot that manage their ARIN entries as well I guess, they should at least have communication with the network admins if they aren't admins themselves. They list 513 area codes for the contacts, so should be local guys. I would hazard a guess if you voice your concerns about quality of service and your treatment thus far, it might gain some traction and maybe light a fire under someone's butt:

scott.pennington@cinbell.com Gregory Scott Pennington - organization administrator for ARIN data
wagner@fuse.net Rick Wagner - looks like another guy to manage registration data in ARIN

***just noticed a reference to Network Solutions with the wagner contact. I've dealt with them on reviving some downed DNS servers back when I was with Lyles Data (long time ago). Might actually be one worth contacting if he is in fact connected with them. Real savvy Linux geeks--which is pretty much what all this hardware runs on.

Found one more email buried for the generic abuse@fuse.net contact. Looks like a bit of a mismatch though. The real name given doesn't fit the pattern for the email address. Though it is used for contacting in regards to things like spambots and other malicious kinds of activity, these contacts are often the same guys that may deal with localized networking issues as well and might be able to at least forward on to someone more appropriate:
msantel@cinbell.com Timothy J. Beck