I’m getting quite high ping on Draenor right now so I ran a WinMTR test and this is what I got:
I’m assuming connection 11 is within Blizzard since it’s listed after a blizzard.com server?
And if you actually perform a DIG on that hostname (hop 11), it comes back with the registrant details mostly hidden.
I honestly don’t think Blizzard would need to hide their registrant details, which leads me to believe that its not a directly owned Blizzard server, but it is possible that Blizzard may be leasing the server (or even just paying for bandwidth through a server) from a CDN (content delivery network) company.
This is an important distinction, because it would mean that Blizzard have no influence on the servers configuration, nor the ability to fix it, other than contacting the company and asking them to fix it.
1 Like
Well home latency has gone down to normal now. But I’m still getting high world latency and random spikes where everyone but me freezes. Had to drop out of raid because it was unplayable.
Here’s another WinMTR log:
Log
Edit:
Spoke too soon. Home latency just went up again.
If you do ping google.com -l 1024 -t do you get anything higher than 10-15ms?
Do you ahve a VPN connection maybe? if so getting on or off on it makes a difference?
I don’t use a vpn. I got this result from ping:
Ping statistics for 142.250.74.46:
Packets: Sent = 175, Received = 173, Lost = 2 (1% loss),
Approximate round trip times in milli-seconds:
Minimum = 20ms, Maximum = 2963ms, Average = 64ms
There does seem to be random spikes here as well.
Could you maybe let that ping run for a minute and post a random section of the results?
However that maximum 2963 seems to be quite high and that 1% loss isnt too good either. 64ms average for google is also high, unless youre in very outlying remote rural place. It’s nothing noticable for browsing, even streaming (not 4k) could do fine with it but gaming is another thing.
Here are my results for comparison:
Packets: Sent = 71, Received = 71, Lost = 0 (0% loss),
Approximate round trip times in milli-seconds:
Minimum = 3ms, Maximum = 6ms, Average = 3ms
I guess if you experience lag spikes with this internet connection of yours alone, these results could more or less explain it. Couple that with peak times on high pop servers…
Have you tried using google DNS on your router settings instead of your ISP’s? or if youre using google dns try reverting to your ISP’s, or even better contact your ISP.
AFAIK they’re hosted in France and I’d guess they do not have their own servers instead they are hosted in DC providers. But that should not affect the quality of service, if anything it should improve it.
I don’t have access to the router dns. I get my net via my landlord wirelessly.
Random snip:
Reply from 142.250.74.46: bytes=68 (sent 1024) time=26ms TTL=114
Reply from 142.250.74.46: bytes=68 (sent 1024) time=27ms TTL=114
Reply from 142.250.74.46: bytes=68 (sent 1024) time=22ms TTL=114
Reply from 142.250.74.46: bytes=68 (sent 1024) time=27ms TTL=114
Reply from 142.250.74.46: bytes=68 (sent 1024) time=27ms TTL=114
Reply from 142.250.74.46: bytes=68 (sent 1024) time=23ms TTL=114
Reply from 142.250.74.46: bytes=68 (sent 1024) time=26ms TTL=114
Reply from 142.250.74.46: bytes=68 (sent 1024) time=29ms TTL=114
Reply from 142.250.74.46: bytes=68 (sent 1024) time=27ms TTL=114
Reply from 142.250.74.46: bytes=68 (sent 1024) time=30ms TTL=114
Reply from 142.250.74.46: bytes=68 (sent 1024) time=27ms TTL=114
Reply from 142.250.74.46: bytes=68 (sent 1024) time=27ms TTL=114
Reply from 142.250.74.46: bytes=68 (sent 1024) time=960ms TTL=114
Reply from 142.250.74.46: bytes=68 (sent 1024) time=1386ms TTL=114
Reply from 142.250.74.46: bytes=68 (sent 1024) time=651ms TTL=114
Reply from 142.250.74.46: bytes=68 (sent 1024) time=26ms TTL=114
Reply from 142.250.74.46: bytes=68 (sent 1024) time=30ms TTL=114
Reply from 142.250.74.46: bytes=68 (sent 1024) time=25ms TTL=114
Reply from 142.250.74.46: bytes=68 (sent 1024) time=24ms TTL=114
Reply from 142.250.74.46: bytes=68 (sent 1024) time=28ms TTL=114
Reply from 142.250.74.46: bytes=68 (sent 1024) time=26ms TTL=114
Reply from 142.250.74.46: bytes=68 (sent 1024) time=27ms TTL=114
Packets: Sent = 190, Received = 190, Lost = 0 (0% loss),
Approximate round trip times in milli-seconds:
Minimum = 21ms, Maximum = 1386ms, Average = 61ms
Resetting modem seems to have worked for now at least.
Maybe it was low on memory?
ISP supplied modem/routers do tend to skimp on memory, meaning that their ARP tables can become full, leading to latency.
But.
I don’t think that would have affected just one single hop. It would have affected all the hops in the traceroute.
If the issue comes back I’d give google DNS Servers a go. You’d need to change that in your router or on your pc.
Yeah it’s weird. Even now when everything is ok again, the latency to that server is awful. But I am between 40-50 to the actual WoW servers so it’s playable again at least.
I have no idea how networking works tbh. Why does high latency and packet loss in 11 and 12 not affect 13 and 14? They are also being sent less packets than the other hops:
And that will help even if can’t change dns on external router?
Ping (basically “hey are you there , respond plz”) packets have generally lowest priority so if the device responding to them is busy they will drop them and not respond at all or respond after everything else is done. When you send a ping it goes through several routers until it reaches the destination. The hops ping goes through pass it onwards normally as it didn’t require response from them.
Hops 11 and 12 had long response time to some of the pings and I guess you used a time limit rather than attempt limit for them so there was less time to send attempts to them because the program was waiting for responses.
Switching DNS server can be done either on your computer or router. Usually computers get their DNS information from the router automatically but it can be easily overridden in the settings. Switching DNS is unlikely to affect pings as if the DNS was the problem you would not be able to access web sites in general.
A lot of devices are configured to outright ignore ICMPEcho requests (pings). A practice that actually breaks several RFC’s, but happens non-the-less.
I have noted that a substantial portion of Blizzards infrastructure are configured this way. Bad Blizzard.
Just used the default settings in WinMTR.
I have similar problem but in my case blizzard servers ignores all my packets (100%) and ping was unmeasurable. After few years i get used to lags in inventory and pve content. Seems it getting worst when i have another players around (like in dungeons, raids and/or world events. Its weird but in completly empty locations game plays smoothly. To be honest I have those problems on all Blizz games in contrast to Steam or FF XIV where i didnt noticed any problems. Seems Blizz dont care about players from central Europe. Seems I must move to Belarus or Russia to play Blizz games smoothly.