Guild Wars 2

Logged in for the second time in 3 weeks. Can’t WvW or PvP because of the lag.

It’s 2014. Your servers have been lagging significantly for MORE than a year, huge amounts of delay, and spikes, and rubber banding.

Yet you do nothing but update the gem store.

PvP is unplayable. Melee is unplayable. Dodging is unplayable.

Game servers back in 1998 in the 56k dial up days had less delay than 2014 GW2 servers. I’m not even joking about this. I have NEVER played on game servers that lag as much as GW2.

How do you manage to have such terrible servers, and no regard for fixing it, in 2014?

It is pathetic. It is amazing how big a nose dive a company can take between 2 product releases.

Sincerely,

One of the many frustrated customers still waiting for you to fix your servers/game. Logging in less and less as time goes on, eventually not going to bother since you don’t care about fixing the servers and bugs in your product.

/End rant. Get your act together.

Yep,its ANETs servers at fault here,that why this subforum is flooded with posts just like yours. (Heres a hint; this is called sarcasm)

You shouldnt post in the tech section with zero info about your PC/internet configuration,especially if you’re going to ignorantly point fingers at Anet.

Come back with more info about how you connect to the net,where you are,maybe do a Tracert and post the results here. We can figure out your problem.

I’ve never once experienced lag while playing since the day the game released. It is probably your own internet service provider at fault and not ANET.

I would recommend you to perform some speedtests at speedtest.net but don’t only test your speed locally see what your speed is like to other countries as your ISP may not have good peering to where ANET’s servers are situated.

If the OP will consider ALL the possible points of failure in the cause for his lag, he may realize that just assuming it’s the LAST point that MUST be the problem is not that intelligent.

If Anet’s server cause constant lag for every player (as would be the case if the Servers were just “bad”), this game would have never survived the first month. Your assumption of where the problem lies is the actual problem here.

OP instead of pointing fingers, why not try and actually PROVE the issue lays with Anet.

In command prompt do tracert to the following IP addresses

206.127.159.116
64.25.38.72

Ill be willing to bet your MS Spikes BEFORE you even reach the NCSoft server farm!

sirsquishy.8531:

OP instead of pointing fingers, why not try and actually PROVE the issue lays with Anet.

In command prompt do tracert to the following IP addresses

206.127.159.116
64.25.38.72

Ill be willing to bet your MS Spikes BEFORE you even reach the NCSoft server farm!

Edited to shorten a long story.
I proved to them some 4 months ago during weeks of back and forth emails that I wasn’t lag spiking until hitting their servers and they still told me it wasn’t on their end. So proving it isn’t on your end won’t actually do anything..

But I do agree you should try doing a trace because proving it is on your end is easier to fix than it is to just deal with it not knowing you could have.

Sir Kaboomski.1508:

sirsquishy.8531:

OP instead of pointing fingers, why not try and actually PROVE the issue lays with Anet.

In command prompt do tracert to the following IP addresses

206.127.159.116
64.25.38.72

Ill be willing to bet your MS Spikes BEFORE you even reach the NCSoft server farm!

Edited to shorten a long story.
I proved to them some 4 months ago during weeks of back and forth emails that I wasn’t lag spiking until hitting their servers and they still told me it wasn’t on their end. So proving it isn’t on your end won’t actually do anything..

But I do agree you should try doing a trace because proving it is on your end is easier to fix than it is to just deal with it not knowing you could have.

You have to realize, if it was their servers or their ISP every player would be affected. Right? And I do not recall ever having such bad lag to be disconnected as much as most of these posts are reporting. So that means, so far, the issue has never been at their end. It has ALWAYS been a transport issue on the way to their servers

Here is a summary of events since January;

France, FRA61,62,34 Nodes, Level3 out of Los Angeles, ATT Attens.net on the east cost, and that horrible DoS attack that was going after Blizzard back in February using the DNS reflection (SMURF) attack

All the above has been going on since January. And all of that is out of End User, End Companies controls. France ISP, ATT, Level3…ect are the ones that need/needed to fix the issues. As they are all BACK BONE providers and 90% of the traffic we use for everyday activities goes through one or more of their BGP paths.

sirsquishy, I’m pretty sure that you’d be tempted badly to create a thread or to submit a request if u suddenly got some permanent freezes and lags out of nowhere – u’ve had game running perfectly ever since, u’ve checked everything u could check and found nothing.
Wouldn’t u?

Brack.6172:

sirsquishy, I’m pretty sure that you’d be tempted badly to create a thread or to submit a request if u suddenly got some permanent freezes and lags out of nowhere – u’ve had game running perfectly ever since, u’ve checked everything u could check and found nothing.
Wouldn’t u?

Being in IT, and a network/systems engineer at that, I would first troubleshoot the issue from my end as much as possible first. As in 99% of the cases, its not the far end. Its almost always a Transport issue.

And the above is even more true, if I decide to look at the forums for multiple posts on my issue at the time. If there is no obvious evidence that the issue is wide spread, then its almost guaranteed to be on my end.

And to rule out the possibilities there are;

Fiber Cuts
DoS Attacks
Congestion due to a downed path
BGP poisoning
faulty routing hardware/interface (DSLAM issues)
Miss configured IPS/IDP (3com once took down DNS for 75% of the internet, when I worked for TippingPoint as a DV signature Engineer, because the IPS code tripped on DNS requests as well as the attack string, and QA didnt catch it. Blocking all DNS requests for new queries.)

… and so on.

So yea, you cannot assume just cause you are having latency issues with your application that the issue is at the servers you connect to. There is a lot of stuff between you and those servers. ALOT OF STUFF!!!

No way, someone that understands the majority of the time the end-user is at fault? I would actually say it’s an ID-10-T error, but that’s a venture onto the broken path of QQ.

Fact of the matter is, there have been NUMEROUS posts about ANET’s servers lagging, but if that were true it would affect a large population of people, not just a few. Similar to another user that came onto the forums actually typing in all caps saying how crappy and laggy ANET servers were, then after telling the kitten possible causes on his end to check for, 2 days later he actually deletes his first post to not look like a complete idiot and makes a final post of “It was just my ISP, it’s better now”.

SirSquishy is 100% right, 99% of the time it is an issue on the user’s end or on the route TO the servers, not the servers themselves. When people start to realize that and understand how routing/networking works, we can see a lot less of those types of posts saying how bad ANET servers are, when in fact, the servers are just fine.

sirsquishy.8531:

You have to realize, if it was their servers or their ISP every player would be affected. Right? And I do not recall ever having such bad lag to be disconnected as much as most of these posts are reporting. So that means, so far, the issue has never been at their end. It has ALWAYS been a transport issue on the way to their servers

All I’m saying is when you run a ping trace for 10 hours a day for a week straight and every single time you watch every hop stay at 70ms and as soon as lag happens in game you check the trace and lo and behold the only servers who’s response time has increased are the NCSoft IP’s, not a single hop before that, you get tired of the whole “it’s not on our end” argument. Unless it’s possible for a hop to not show up on a ping trace at all, if it wasn’t them it was their provider. The rep I was talking to was stumped to because I could prove it wasn’t on my end. So finally he said he would pass it to their network engineers but said that he couldn’t guarantee we would hear anything back.

Sir Kaboomski.1508:

sirsquishy.8531:

You have to realize, if it was their servers or their ISP every player would be affected. Right? And I do not recall ever having such bad lag to be disconnected as much as most of these posts are reporting. So that means, so far, the issue has never been at their end. It has ALWAYS been a transport issue on the way to their servers

All I’m saying is when you run a ping trace for 10 hours a day for a week straight and every single time you watch every hop stay at 70ms and as soon as lag happens in game you check the trace and lo and behold the only servers who’s response time has increased are the NCSoft IP’s, not a single hop before that, you get tired of the whole “it’s not on our end” argument. Unless it’s possible for a hop to not show up on a ping trace at all, if it wasn’t them it was their provider. The rep I was talking to was stumped to because I could prove it wasn’t on my end. So finally he said he would pass it to their network engineers but said that he couldn’t guarantee we would hear anything back.

They have network engineers? lol

From what I know, and what I have heard, Anet is a hosted solution. They outsource their network operations to the datacenter they lease. So the TCP Stack is not managed by Anet directly. And you know what that means, Billable. Every network based case they open would be billed by whatever team they source to. And if their SNMP/Syslogs data do not show anything obvious, I can see them ignoring a legit issue that is not wide spread.

Anet does have a server team, obviously the game’s design team (desktop Application all the way up to the Servers applications), and all that. But I do not believe they have any Network team at all.

Now, your Traceroutes are only showing from your direction to theirs. What they SHOULD have done was traceroute back to you to make sure TX, as well as RX, was working correctly. But that’s the thing about Traces, its only going to show data in one direction not both.

I kind of wish Anet had a tool available via Web that would let us do connectivity tests from their servers out. That would be very useful.

Dunno lol… maybe he just told me that cause he didn’t know what else to do.. It was a rather exhaustive exchange between the both of us.

What you say makes sense, it’s just a shame that at the time there were other people having the same problem as me posting on the forums and they still I guess didn’t feel compelled to inquire about it. I don’t know… but it was frustrating as hell lol.

sirsquishy.8531:

I kind of wish Anet had a tool available via Web that would let us do connectivity tests from their servers out. That would be very useful.

I’m not sure putting a utility that could directly saps server performance (no matter how small) accessible to the public would be a great idea….(the info would certainly be nice, but the potential for abuse is HUGE).

It’s not unusual for a Support person to escalate an issue to another companies employee (obviously the company that hosts the servers MUST have Network Engineers) and just not related the information that another company’s employee will actually be looking into the problem. His statement about not hearing back from them kind of supports that (the Network Engineer at the hosting company likely has no direct responsibility to get back to the CS Agent about a specific request unless that is covered by the support contract Anet has with the hosting company).

I’m sure it’s less of a “billing” issue as a single complaint not being high enough priority to pursue vigorously. Hosting companies do have some level of responsibility to maintain a spec connection that is free from interruptions without any extra “billing” being involved. From my experience with large hosting companies (on a very small scale), they are usually very responsive to such requests .

Brother Grimm.5176:

sirsquishy.8531:

I kind of wish Anet had a tool available via Web that would let us do connectivity tests from their servers out. That would be very useful.

I’m not sure putting a utility that could directly saps server performance (no matter how small) accessible to the public would be a great idea….(the info would certainly be nice, but the potential for abuse is HUGE).

It’s not unusual for a Support person to escalate an issue to another companies employee (obviously the company that hosts the servers MUST have Network Engineers) and just not related the information that another company’s employee will actually be looking into the problem. His statement about not hearing back from them kind of supports that (the Network Engineer at the hosting company likely has no direct responsibility to get back to the CS Agent about a specific request unless that is covered by the support contract Anet has with the hosting company).

I’m sure it’s less of a “billing” issue as a single complaint not being high enough priority to pursue vigorously. Hosting companies do have some level of responsibility to maintain a spec connection that is free from interruptions without any extra “billing” being involved. From my experience with large hosting companies (on a very small scale), they are usually very responsive to such requests .

In a hosted environment, where the client – ANet – has to pay per incident, you are going to find that companies will take singled out issues less serious due to the billable over anything else. The exception to this is if the singled out client happens to be a big chuck of income to the company in question.

Most hosting companies will charge 125/hr to investigate any issue they are in charge of. Some places do it in blocks of 4 hours, some less and others more. While to you and I that’s not a lot for 1 incident, If there were 20-30 of these a week..; they do add up.

GW2’s only income is the gem store, and new players(Accounts). I feel that’s why singled out issues are not really dealt with, if they require additional support that Anet cannot directly give.

And this is why I wish for more tools that the players can use to help diagnose the issue sourcing from Anet’s side.

For network issues, its pretty simple. A traceroute will tell you what you need to know, most of the time.

For system performance issues, having the game client pull performance statistics, while playing the game, then uploading them to a database (like 3dmarks uses) and rank your system against others in the Gw2 database would also be useful.

But those are just a couple ideas that would help. And you know, with out the ROI we will never see these tools.

It’s your connection.

Arcade Fire.4895:

It’s your connection.

^ I lol’d, I like this guy lmao