Guild Wars 2

Hi,

I’ve had ping problems on a regular basis but they mostly (though not entirely) went away once I started using Pingzapper a few months back. However, I am unable to play since Saturday due to :

1) Spikes up to 4k (usually about 300 to 500, rarely 1k) for a few seconds
2) Sometimes, ping averaging at about 100 to 600

Needless to say, it’s unplayable.

On Saturday and Sunday it started at 5 pm up until midnight and on Monday it started at noon which isn’t exactly prime time (and kept going the entire day, though 7am to 11am was roughly ok). Just for the record nobody’s downloading on my network and I was the only one on it this Monday. The problems remain even if I use a VPN. I tested both Pingzapper AND WTFast and tried different settings (automatically close to my country vs manually a bit further away). I used to be in Wi-Fi, now I’m not so that’s ruled out too.

Here’s a couple plots from Ping Plotter :

Pinging GW2 IPs :

I pinged some of the servers whose IP start with 64 : http://imgur.com/a/oPnpn
zoom on Packet Losses : http://imgur.com/a/cg38O

I pinged some of the servers whose IP start with 206 : http://imgur.com/a/atNQD

I believe this shows that I get some high average latency on NCsoft servers of about 200-400ms. I also get packet loss on those servers as demonstrated on the zooms.

I also pinged the IP address my VPN (Pingzapper only because WTFast provides about 10 IPs and I’ve no idea which one to ping) goes to when I’m using it in “Firewall mode” (ie all traffic goes through it).

PZ : http://imgur.com/a/uXZoi

PZ zoom on Packet Losses : http://imgur.com/a/hhes0

And then I’m confused. It doesn’t look like the NCsoft servers (or Telia to some extent) are the problem anymore. I’m getting high latency and packet loss here too? What the heck?

If anyone can make sense of this and direct me to my ISP or to Anet CS, I would be most grateful.

P.S. : I know Ping Plotter shows that I supposedly have 20ms average but trust me, it’s not quite right ._.

P.S2: The gw2 diagnosis tool used to work a few months back but now it doesn’t show anything past the second hop, dunno if that’s helpful information.

Looks like a Tier 3 issue (Telia, etc.), as usual.

You can contact the Tech CS Team via the ‘Support’ link above/below and ‘Submit a Ticket’ for assistance.

It appears that you packet loss is starting just past your modem. Once you have packet loss it carries to each node you connect to. I would start trying to eliminate it at your ISP, that should also clear it up at each node along the way.

Inculpatus cedo.9234:

Looks like a Tier 3 issue (Telia, etc.), as usual.

You can contact the Tech CS Team via the ‘Support’ link above/below and ‘Submit a Ticket’ for assistance.

That doesn’t add up for a number of reasons, starting with the fact that I tested several VPNs that I’d started using previously to fix, indeed, a Telia-induced Packet Loss problem.

Gator.5729:

It appears that you packet loss is starting just past your modem. Once you have packet loss it carries to each node you connect to. I would start trying to eliminate it at your ISP, that should also clear it up at each node along the way.

Mhm, can’t hurt to whine to my ISP. Thanks !

Gator.5729:

It appears that you packet loss is starting just past your modem. Once you have packet loss it carries to each node you connect to. I would start trying to eliminate it at your ISP, that should also clear it up at each node along the way.

Came here to say the same thing: in three of the four links, you have ~ 4.5 percent packet loss in the first hop. Talk to your ISP, because you should never see anything there, and that’s definitely in their ballpark.

Beyond that you may still have problems, but … what you care about is when you get a specific level of red packet loss at a hop and every hop past it.

If you see one hop with high packet loss, but the next hop is fine … it’s just not responding to your ping messages. That is totally legit, and pretty standard, for busy routers, they just don’t treat it as important.

When there is no packet loss following, that’s a sign that it is forwarding traffic just fine, but not responding to messages that end at “it” as a hop.

The 4.5% is actually an average on packet losses, upon closer inspection half the packet losses start from the ISP server on the third hop (and then the following hops are affected too). Some even reach 10-20% on the third hop. Soooo I guess you’re correct.

Thank you ! I’ve already tried calling my ISP but the tech guy I had on the phone had no idea what ping is so it’s gonna prove challenging.

Uhtameit.2413:

The 4.5% is actually an average on packet losses, upon closer inspection half the packet losses start from the ISP server on the third hop (and then the following hops are affected too). Some even reach 10-20% on the third hop. Soooo I guess you’re correct.

Thank you ! I’ve already tried calling my ISP but the tech guy I had on the phone had no idea what ping is so it’s gonna prove challenging.

It’s kinda cheating, but I find telling them that webpages are “stalling” while loading is often a useful trick to get past the script, and up to some reasonable debugging on their part.

Also, “it gets worse when it rains or is foggy” can help.