3
Unacceptable Ping For Europe-Based Players
Started by MechWarrior414712, Mar 07 2015 09:42 AM
28 replies to this topic
#21
Posted 07 March 2015 - 04:31 PM
Having worked in telecom, there are times the services between you and your destination, MWO here, will change routes for various reasons.
#22
Posted 09 March 2015 - 01:07 PM
DaZur, on 07 March 2015 - 10:48 AM, said:
I "believe" their website and their game server reside in the same farm node. So if you use: http://mwomercs.com/ that pretty much gets you in the same ballpark.
Yep
Alistair Winter, on 07 March 2015 - 04:01 PM, said:
My ping is the same as it has been for a while. It was 150-160 when I started in 2012. Over the past few months, it's dropped down to 130-135.
Still good.
Still good.
That's what we've been working toward. Not everyone is getting better speeds but most are. It's a constant work to keep our network in good shape.
For anyone with unusually bad ping, please send a me traceroute so I can investigate. If I can't see what's going on, I can't help. : (
#23
Posted 09 March 2015 - 01:23 PM
DaZur, on 07 March 2015 - 10:40 AM, said:
While I appreciate the fact that it's seems logical that PGI would be singularly culpable for the increase in ping rate... They are actually near the bottom in terms of culpability...
ISPs are notorious for changing routes (particularly EU routes). I'd begin looking there first IMHO...
That said, this site can be handy sometimes: Online Visual Tracerout
If you have no baseline to gauge from it's more info than helpful but I've persoanlly seen the number of hops taken + / - 3, which in the grand scheme of things affected me up to 20 points one way or the other.
ISPs are notorious for changing routes (particularly EU routes). I'd begin looking there first IMHO...
That said, this site can be handy sometimes: Online Visual Tracerout
If you have no baseline to gauge from it's more info than helpful but I've persoanlly seen the number of hops taken + / - 3, which in the grand scheme of things affected me up to 20 points one way or the other.
Wow. Apparently my traffic likes to sneakily detour to France. You can't use company funds like that, traffic. I'm going to have to audit you. I don't care if you 'really like the bread'.
#24
Posted 09 March 2015 - 01:25 PM
just for measure, from germany, starting 1 hop behind my router:
2 17 ms 15 ms 15 ms 87.186.225.98
3 19 ms 16 ms 18 ms 87.190.186.102
4 21 ms 21 ms 21 ms 217.239.39.18
5 21 ms * * 94.23.122.153
6 25 ms 24 ms 24 ms 91.121.128.125
7 25 ms 24 ms 24 ms 37.187.36.155
8 * * * timeout
9 * * * timeout
10 115 ms 115 ms 115 ms 192.99.109.192
115 ping is around where it always was since the beginning, 100-120 average.
2 17 ms 15 ms 15 ms 87.186.225.98
3 19 ms 16 ms 18 ms 87.190.186.102
4 21 ms 21 ms 21 ms 217.239.39.18
5 21 ms * * 94.23.122.153
6 25 ms 24 ms 24 ms 91.121.128.125
7 25 ms 24 ms 24 ms 37.187.36.155
8 * * * timeout
9 * * * timeout
10 115 ms 115 ms 115 ms 192.99.109.192
115 ping is around where it always was since the beginning, 100-120 average.
Edited by 627, 09 March 2015 - 01:25 PM.
#25
Posted 09 March 2015 - 02:13 PM
PGI is probably running out of money and shutting servers down. Have you seen how many grab deals they are running lately? Reeks of desperation.
#26
Posted 09 March 2015 - 02:27 PM
I did a bunch of trace routes a while back and never heard anything; But i would agree with other posters that you should move hosts to a US based company;
For one; it would appear that the cross boarder hand off is done through New Jersey; What ever that provider is doing, they are not buying enough port speed for that hand off to cover peak time 3PM EST-7PM EST. Normally by 10PM EST things calm down and ping return to normal.
There are literally some games my ping is 75ms and then immediately the next one is 140; I run a trace and see that the delay is happening on OVH's handoff router in New Jersey.
So i dont know if that is their only hand off; or its over saturated, but it appears that they cant handle peak times with out packet loss; Coming from the west coast you would think that they would have other POP besides the east coast where the data can route though and avoid the extra 700+ miles and saturated links.
The other would be that your main audience for the game is located to the south of you; I see no reason to use saturated boarder links that are shared with Amazon, Hulu, etc;
I know there are considerations when it comes to currency conversion and US Colo pricing. But if that means gaining more users then those costs should offset.
If you want to send me a rack of servers, there is a HE about 2 miles from my house in Cali; My clients always see to have good luck with them.
For one; it would appear that the cross boarder hand off is done through New Jersey; What ever that provider is doing, they are not buying enough port speed for that hand off to cover peak time 3PM EST-7PM EST. Normally by 10PM EST things calm down and ping return to normal.
There are literally some games my ping is 75ms and then immediately the next one is 140; I run a trace and see that the delay is happening on OVH's handoff router in New Jersey.
So i dont know if that is their only hand off; or its over saturated, but it appears that they cant handle peak times with out packet loss; Coming from the west coast you would think that they would have other POP besides the east coast where the data can route though and avoid the extra 700+ miles and saturated links.
The other would be that your main audience for the game is located to the south of you; I see no reason to use saturated boarder links that are shared with Amazon, Hulu, etc;
I know there are considerations when it comes to currency conversion and US Colo pricing. But if that means gaining more users then those costs should offset.
If you want to send me a rack of servers, there is a HE about 2 miles from my house in Cali; My clients always see to have good luck with them.
#27
Posted 09 March 2015 - 03:25 PM
Since the day they started using their current SP there have been problems all over the world, they blame ISP, national carriers, everything except the cheap SP they use.
Do your traces and most of you will see the huge increase in ping as soon as it hits that SP..
It isnt rocket science to figure where the problem is..
Do your traces and most of you will see the huge increase in ping as soon as it hits that SP..
It isnt rocket science to figure where the problem is..
#29
Posted 09 March 2015 - 06:28 PM
Well at least they post their network map:
http://weathermap.ovh.net/#usa
I can actually trace where my connection comes in from comcast and see where it ends up; Or sort of; It would appear most of the equipment blocks responding to ping or is firewalled; But you can follow the lines and pretty much know what routers you are going through.
Whats funny is it looks like they have a comcast peer point in chicago; But me being on the west coast, is using the peer point in NJ.....
Go figure...
http://weathermap.ovh.net/#usa
I can actually trace where my connection comes in from comcast and see where it ends up; Or sort of; It would appear most of the equipment blocks responding to ping or is firewalled; But you can follow the lines and pretty much know what routers you are going through.
Whats funny is it looks like they have a comcast peer point in chicago; But me being on the west coast, is using the peer point in NJ.....
Go figure...
1 user(s) are reading this topic
0 members, 1 guests, 0 anonymous users