![]() ![]() ![]() |
|
If you dont want to move from spark using a VPN client can get around the issue as they can link endpoints over their own routes.
I have used Astrill with good success.
michaelmurfy:
I wouldn't imagine that a ping difference of a few ms would make much of a difference to a game like WoT.
It is no different to an FPS.
Michael Murphy | https://murfy.nz
Referral Links: Quic Broadband (use R122101E7CV7Q for free setup)
Are you happy with what you get from Geekzone? Please consider supporting us by subscribing.
Opinions are my own and not the views of my employer.
As for the OP you might have better luck making things known from the other end, my understanding is Wargaming's network partner (which looks to be G-Core Labs S.A.) puts in a lot of effort in regards to low latency routes they may be able to effect a resolution from their end. If you let them know via their NOC email they might even respond.
@michaelmurfy Usually your comments are helpful and insightful but when you offer opinions like "I wouldn't imagine that a ping difference of a few ms would make much of a difference to a game like WoT" perhaps you should refrain.
The OP mention an increase of ~150 ms, and while World of Warships (the game the OP originally mentioned) is less affected by latency than World of Tanks (WoWS firing is generally a lot longer range than WoT) 150ms would be noticeable.
I stand by my comments, as they are observations from real world usage. The only thing I didn't go into is that traceroutes are only showing how your data is getting to the destination, not how the destination gets data back to you. Sometimes that can be a completely different route (and impossible for your ISP to correct if its a higher latency path).
yitz: How is it incorrect? ISPs are free to suppress certain routes if they choose. As for the OP you might have better luck making things known from the other end, my understanding is Wargaming's network partner (which looks to be G-Core Labs S.A.) puts in a lot of effort in regards to low latency routes they may be able to effect a resolution from their end. If you let them know via their NOC email they might even respond.
Simply this:
Amosnz: Then I moved to Spark who with a lot more customers require a lot more bandwidth, and found their focus is on providing capacity at low cost rather than the lowest possible latency. When UFB became available I moved to another smaller ISP (where I also know someone) and once again have never had a problem getting increased pings looked into and corrected..
Global Gateway is used by many NZ ISP's and is considered a "premium" product (however as others rightfully pointed out sometimes not the best). Pings are normally quite low however the issue the OP is having can happen on any ISP. I can confirm I am getting the same route on a couple of other ISP's also using Global Gateway.
It does seem there is a problem with the routes at the moment (taken from another ISP):
murphy@pikachu:~$ traceroute 162.213.61.115
traceroute to 162.213.61.115 (162.213.61.115), 30 hops max, 60 byte packets
1 router (192.168.2.1) 0.758 ms 0.692 ms 0.662 ms
2 219.88.232.254 (219.88.232.254) 12.610 ms 14.833 ms 14.813 ms
3 122.56.60.70 (122.56.60.70) 13.110 ms 13.088 ms 13.065 ms
4 122.56.60.71 (122.56.60.71) 13.506 ms 14.584 ms 14.577 ms
5 ae11-201.tkbr11.global-gateway.net.nz (122.56.118.149) 14.533 ms 14.545 ms 14.869 ms
6 ae6-10.akbr7.global-gateway.net.nz (203.96.120.93) 15.467 ms 14.397 ms 14 .789 ms
7 122.56.119.18 (122.56.119.18) 38.823 ms xe0-0-1.sgbr3.global-gateway.net.nz (202.50.232.110) 35.908 ms 36.602 ms
8 ae2-10.sgbr4.global-gateway.net.nz (202.50.232.246) 35.512 ms 35.519 ms 3 5.410 ms
9 ae-3.a04.sydnau03.au.ra.gin.ntt.net (103.13.80.125) 35.437 ms 35.391 ms 3 6.806 ms
10 ae-4.r21.sydnau03.au.bb.gin.ntt.net (202.68.64.124) 41.162 ms 41.165 ms 4 1.117 ms
11 ae-11.r31.tokyjp05.jp.bb.gin.ntt.net (129.250.5.34) 149.469 ms 150.179 ms 149.445 ms
12 ae-2.r30.tokyjp05.jp.bb.gin.ntt.net (129.250.5.23) 149.396 ms 150.091 ms 148.034 ms
13 ae-4.r24.tkokhk01.hk.bb.gin.ntt.net (129.250.2.51) 191.281 ms 191.223 ms 191.175 ms
14 ae-1.r00.tkokhk01.hk.bb.gin.ntt.net (129.250.2.125) 191.907 ms 191.849 ms 191.003 ms
15 ae-0.a00.tkokhk01.hk.bb.gin.ntt.net (129.250.2.128) 190.895 ms 190.887 ms 190.840 ms
16 ix-ae-24-0.tcore1.HK2-Hong-Kong.as6453.net (180.87.112.153) 275.634 ms 275 .162 ms 276.571 ms
17 if-et-17-3.hcore1.KV8-Chiba.as6453.net (180.87.112.46) 247.347 ms if-ae-5-2 .tcore1.TV2-Tokyo.as6453.net (180.87.112.206) 273.712 ms 273.651 ms
18 if-et-21-2.hcore1.KV8-Chiba.as6453.net (120.29.217.67) 240.780 ms 236.865 ms if-ae-24-2.tcore2.PDI-Palo-Alto.as6453.net (66.198.144.56) 274.244 ms
19 if-ae-5-2.tcore2.SQN-San-Jose.as6453.net (64.86.21.1) 247.380 ms 246.416 m s if-ae-24-2.tcore2.PDI-Palo-Alto.as6453.net (66.198.144.56) 272.283 ms
20 if-ae-5-2.tcore2.SQN-San-Jose.as6453.net (64.86.21.1) 239.513 ms if-ae-1-2. tcore1.SQN-San-Jose.as6453.net (63.243.205.1) 282.360 ms 282.331 ms
21 216.6.33.122 (216.6.33.122) 246.564 ms 236.428 ms 238.035 ms
22 92.223.120.164 (92.223.120.164) 274.160 ms 216.6.33.122 (216.6.33.122) 239 .847 ms 92.223.120.163 (92.223.120.163) 252.128 ms
23 92.223.120.163 (92.223.120.163) 245.217 ms * *
24 * * *
Compare this to Voyager:
mmurphy@sever:~ $ traceroute 162.213.61.115
traceroute to 162.213.61.115 (162.213.61.115), 30 hops max, 60 byte packets
1 192.168.2.1 (192.168.2.1) 1.257 ms 0.972 ms 1.164 ms
2 114.23.3.254 (114.23.3.254) 25.042 ms 25.306 ms 25.631 ms
3 ae-0-942.cr1.mdr.vygr.net (114.23.3.248) 25.522 ms 25.800 ms 25.695 ms
4 114.23.3.243 (114.23.3.243) 25.974 ms 25.860 ms 26.091 ms
5 ten-0-7-0-5-396.bdr04.alb01.akl.vocus.net.nz (175.45.93.77) 27.186 ms 29.158 ms 29.044 ms
6 bundle-11.cor01.alb01.akl.vocus.net.nz (114.31.202.48) 160.855 ms 158.664 ms 158.829 ms
7 bundle-200.cor01.lax01.ca.vocus.net (114.31.202.45) 158.324 ms 157.987 ms 157.807 ms
8 100g-0-1-0-0.cor01.sjc01.ca.vocus.net (49.255.255.0) 158.489 ms 158.376 ms 158.693 ms
9 ten-2-0-0.bdr01.sjc02.ca.vocus.net (114.31.199.181) 158.578 ms 158.476 ms 158.684 ms
10 eqix-sj-sv5.gcore.lu (206.223.116.232) 161.427 ms 161.317 ms 161.554 ms
11 sv4-a9006-edge-1-be20-2000.fe.core.pw (92.223.120.130) 160.252 ms 158.810 ms 159.074 ms
12 92.223.120.164 (92.223.120.164) 158.981 ms 92.223.120.163 (92.223.120.163) 159.207 ms 92.223.120.164 (92.223.120.164) 158.757 ms
13 * * *
But I stand by saying this can happen to literally any ISP and I am sure Spark will get this resolved...
Amosnz:
@michaelmurfy Usually your comments are helpful and insightful but when you offer opinions like "I wouldn't imagine that a ping difference of a few ms would make much of a difference to a game like WoT" perhaps you should refrain.
The OP mention an increase of ~150 ms, and while World of Warships (the game the OP originally mentioned) is less affected by latency than World of Tanks (WoWS firing is generally a lot longer range than WoT) 150ms would be noticeable.
I stand by my comments, as they are observations from real world usage. The only thing I didn't go into is that traceroutes are only showing how your data is getting to the destination, not how the destination gets data back to you. Sometimes that can be a completely different route (and impossible for your ISP to correct if its a higher latency path).
Yes my apologies I did read that statement incorrectly (end of day at work and all that). I agree, a jump of 150ms will be noticeable but I initially read it as a ~60ms jump in latency.
Michael Murphy | https://murfy.nz
Referral Links: Quic Broadband (use R122101E7CV7Q for free setup)
Are you happy with what you get from Geekzone? Please consider supporting us by subscribing.
Opinions are my own and not the views of my employer.
michaelmurfy:
Amosnz: Then I moved to Spark who with a lot more customers require a lot more bandwidth, and found their focus is on providing capacity at low cost rather than the lowest possible latency. When UFB became available I moved to another smaller ISP (where I also know someone) and once again have never had a problem getting increased pings looked into and corrected..
Global Gateway is used by many NZ ISP's and is considered a "premium" product (however as others rightfully pointed out sometimes not the best). Pings are normally quite low however the issue the OP is having can happen on any ISP. I can confirm I am getting the same route on a couple of other ISP's also using Global Gateway.
The reason I say this is slower downloads are more noticeable to the average user than latency. A person who knows nothing about the internet can look at a download and say "I'm getting 150 number now but sometimes i get 600 number". So having larger capacity (even if its on higher latency/longer links) generally is better.
I haven't used Spark since about October 2014 so things may have changed since then.
For what it's worth, here's the same server on Vibe:
traceroute to 162.213.61.115 (162.213.61.115), 64 hops max, 52 byte packets
1 10.0.1.1 (10.0.1.1) 0.379 ms 0.358 ms 0.292 ms
2 * * *
3 lt-2-0-10-3.mdr-cr1.as45177.net.nz (14.1.50.36) 6.893 ms 6.364 ms 7.896 ms
4 equinix-sv1.as45177.net (206.223.116.96) 141.700 ms 143.185 ms 155.805 ms
5 eqix-sj-sv5.gcore.lu (206.223.116.232) 142.057 ms 142.615 ms 143.910 ms
6 sv5-a9006-edge-1-be20-2000.fe.core.pw (92.223.120.131) 143.367 ms 144.005 ms
144.027 ms
7 92.223.120.164 (92.223.120.164) 144.204 ms 142.578 ms 144.254 ms
8 * * *
Azzura:Linux:Sorry but it's more a mind thing than a slight change in latency
Linux
What study did you read that in?
FWIW:
Vodafone UFB Auckland: 162.213.61.115
Host Loss% Snt Last Avg Best Wrst StDev
1. 10.0.2.2 0.0% 31 0.3 0.3 0.2 0.4 0.0
2. 192.168.1.1 0.0% 31 1.5 2.1 1.2 4.3 0.4
3. 203-118-183-254.dsl.dyn.ihug.co.nz 0.0% 30 5.5 23.9 5.5 88.9 19.7
4. 10.123.80.70 0.0% 30 7.5 23.3 5.1 68.3 18.8
5. 10.123.80.69 6.7% 30 131.6 144.2 129.0 192.1 19.2
6. v422.core1.sjc1.he.net 0.0% 30 138.2 153.7 129.6 200.4 20.7
7. 100ge1-1.core1.sjc2.he.net 0.0% 30 140.4 147.6 128.5 195.9 17.3
8. eqix-sj-sv5.gcore.lu 0.0% 30 137.4 146.8 130.5 214.0 21.4
9. sv4-a9006-edge-1-be20-2000.fe.core.pw 0.0% 30 134.6 138.9 130.0 170.0 9.3
10. 92.223.120.164 0.0% 30 130.7 146.0 130.5 209.6 17.6
11. sv4-sl-b115.fe.core.pw 0.0% 30 130.9 145.0 129.1 184.4 14.3
Scotdownunder: While no IP routing is guaranteed, one must wonder why the international gateway routers are not using the most direct routes over Southern Crossing with fewer hops. Routes via Sydney are OK as only marginally longer.
Is this capacity too expensive ? SC have always maintained there was plenty of capacity (hence no new cable system required) but the market disagrees and a new cable system is finally under construction. Lets hope this increase in capacity (a hopefully lower costs) will mean direct AKL > West Coast hops will advertise lower costs than round the Pacific routes.
We live in hope
I don't think Spark are actively or purposefully routing this traffic via Hong Kong/Asia. Rather it is likely Wargaming have engaged their network partner/s to pick up as much of their gaming traffic from IXPs/ISPs in the Asian region as possible and send it to them over their own private transit links across the Pacific. It is likely Spark has been lumped into this group of Asian ISPs as they will also have their own arrangement for traffic into the Asian region.
It is really up to either or both parties to make tweaks to such a policy to give mutual customers the best experience.
seems the routing for this IP has been restored to its normal path.
|
![]() ![]() ![]() |