Gaming on multiplay

[Tarantella]

I find my connection To Multiplay, the UK's Premier gaming service provider, is suffering connectivity problems during peak periods with disconnects and erratic latency.

A resume of data taken recently showed this.

NeoTrace Version 3.25 - TRIAL Trace Results
Target: 85.236.100.100
Date: 27/11/2008 (Thursday), 19:09:49
Nodes: 13



Node Reg IP Address Location Node Name \Packet Data High Low Avg Tot Lost


3 62.30.241.39 Bradford osr02brad-v11.network.virginmedia.net 141 1 23 100 0
4 195.182.178.125 Leeds lee-bb-b-ge-110-0.network.virginmedia.net 50 10 22 100 0
5 62.253.187.185 Leeds lee-bb-a-ae0-0.network.virginmedia.net 46 11 22 101 1
6 213.105.175.26 Guildford bre-bb-b-so-200-0.network.virginmedia.net 72 18 27 100 0
7 62.253.185.74 Hounslow telc-ic-1-as0-0.network.virginmedia.net 37 16 27 100 0
8 213.248.100.97 Westminster ldn-b2-link.telia.net 43 20 27 100 0
9 80.91.250.214 Westminster ldn-tch-i2-link.telia.net 197 0 37 100 0
10 80.91.250.217 Hamburg ldn-tch-i1-link.telia.net 175 18 32 100 0
11 213.248.74.42 Westminster 113069-ldn-tch-i1.c.telia.net 220 22 75 100 0
12 195.72.129.14 54.083N,4.650W multiplay-gw.inx.datahop.net 229 24 40 101 1
13 85.236.100.100 Southampton beechbone.multiplay.co.uk 53 23 27 100 0


Whilst I'll accept that the average ping is ok the spiking is unacceptible for game playing.

On a second point I'd like to know why on reaching Guildford via VM my trace takes an about turn (enters the Telianet network)and disappears off to Hamburg instead of travelling a short 5 ms hop to the Multiplay servers on the South coast?

[Joxer]

Seem to be early on as well - if connected directly to the modem run ipconfig and then try pinging the default gateway a lot ping -t 100 (gateway ip).

If you have a router you will need to look at it's config page to find the default gateway, you could also try pinging the modem ping -t 100 192.168.100.1

[TheDon]

The high pings you're getting on that trace look like they're just from the routers having a low priority for icmp traffic. As the last hops pings are lower than hops 12 and 13, dispite having to be routed through them, it says that either those nodes are only spiking at the moment you're just pinging them (unlikely) or it's just a case of them delaying responding to icmp traffic whilst dealing with other forms of traffic.

As to your second point, it doesn't look like it's ACTUALLY going to hamburg (you're average ping to the destination being 27 would be pretty low for that), just that the IP is registered in Hamburg, the name of it also indicates it's in london (the ldn part says it's a london based router) and the next hop is also london based. This is the problem with using the location data from the whois records, it just indicates where the router is registered, not where it's actually in use.

[Noggo]

Originally Posted by Tarantella

Multiplay, the UK's Premier gaming service provider,

Not if you're got a VM connection it isn’t. I know cable connections have weird / bad routing sometimes, but 13 hops to a UK server . More hops = more places for it to go wrong.

I don't think any of the hops are in mainland Europe though, you wouldn't see minimum pings of 14ms (see attached ping plotter image).

I would suggest having a look at program like Ping Plotter software and look at the trace over a period of 10 mins or so, much better than a snap shot picture of whats going on.

• Test if you can at different times of the day, first thing in the morning then again peak time to see if there's any difference.
  
• Does the erratic latency happen to other games servers as well or is it just Multiplay?
  
• Does your download speed drop off when you're having problems?

[Tarantella]

Thanks for your interest.

Im using the modem directly and the second hop on the trace has a 10.153.0.1 address and ping cmd reports it as unreachable. I assumed for some reason VM modems dont respond to ping requests.

My IE browser brings up the cable info and status when I insert 192.168.100.1 and sticks on my BBC homepage when I insert 10.153.0.1!

[Joxer]

There isn't going to be a web server at 10.153.0.1 so it won't reply to any http requests.

Still interested in seeing pings to the default gateway.

[Noggo]

The command for pinging the modem is
ping 192.168.100.1 -n 100
and not `ping -t 100 192.168.100.1'

I'm assuming that it is windows command prompt you're trying to access it from.

[Joxer]

My bad, on Linux here so would use ping -c - easy to misremeber command switches you never use.

[Tarantella]

I'm off work with the -f -l -u today but managed to get some stats.


1600-1615 hours

Ping statistics for 62.30.241.39:
Packets: Sent = 100, Received = 100, Lost = 0 (0% loss),
Approximate round trip times in milli-seconds:
Minimum = 6ms, Maximum = 163ms, Average = 13ms

Ping statistics for 62.30.241.39:
Packets: Sent = 100, Received = 100, Lost = 0 (0% loss),
Approximate round trip times in milli-seconds:
Minimum = 6ms, Maximum = 185ms, Average = 11ms

Ping statistics for 62.30.241.39:
Packets: Sent = 100, Received = 100, Lost = 0 (0% loss),
Approximate round trip times in milli-seconds:
Minimum = 6ms, Maximum = 179ms, Average = 22ms

Very variable pings and I'll get some more around 8pm.

[Noggo]

Why are you pinging 62.30.241.39 on VM' network?

With a direct PC to VM cable modem connection ping the modem with:
ping 192.168.100.1 -n 100
and you should get lots and lots of 1ms

If you're going to use the ping command do it to the games server. This will give you round trip results:
ping 85.236.100.100 -n 20

and trace route test will show up any en-route problems, like packet loss:
tracert 85.236.100.100

[Joxer]

That looks like the router just low prioritising ICMP traffic - you get similar result pinging from elsewhere on the VM network.

PING 62.30.241.39 (62.30.241.39) 56(84) bytes of data.
64 bytes from 62.30.241.39: icmp_seq=1 ttl=253 time=2.46 ms
64 bytes from 62.30.241.39: icmp_seq=2 ttl=253 time=2.40 ms
64 bytes from 62.30.241.39: icmp_seq=3 ttl=253 time=2.18 ms
64 bytes from 62.30.241.39: icmp_seq=4 ttl=253 time=2.41 ms
64 bytes from 62.30.241.39: icmp_seq=5 ttl=253 time=62.1 ms

--- 62.30.241.39 ping statistics ---
5 packets transmitted, 5 received, 0% packet loss, time 4001ms
rtt min/avg/max/mdev = 2.185/14.329/62.176/23.923 ms

Hint: The default gatway will be similar to your ip address but end with a 1 (probably).

[Raistlin]

Originally Posted by Noggo

The command for pinging the modem is
ping 192.168.100.1 -n 100
and not `ping -t 100 192.168.100.1'

I'm assuming that it is windows command prompt you're trying to access it from.

From windows you could also do:

ping 192.168.100.1 -t

Which would continuously ping the modem until you hit Ctrl+C.

Don't know if it will help, but here is the ping test to that IP address from Leeds

Pinging 85.236.100.100 with 32 bytes of data:
Reply from 85.236.100.100: bytes=32 time=15ms TTL=55
Reply from 85.236.100.100: bytes=32 time=18ms TTL=55
Reply from 85.236.100.100: bytes=32 time=14ms TTL=55
Reply from 85.236.100.100: bytes=32 time=20ms TTL=55
Reply from 85.236.100.100: bytes=32 time=14ms TTL=55
Reply from 85.236.100.100: bytes=32 time=17ms TTL=55
Reply from 85.236.100.100: bytes=32 time=15ms TTL=55
Reply from 85.236.100.100: bytes=32 time=14ms TTL=55
Reply from 85.236.100.100: bytes=32 time=23ms TTL=55
Reply from 85.236.100.100: bytes=32 time=16ms TTL=55
Reply from 85.236.100.100: bytes=32 time=15ms TTL=55
Reply from 85.236.100.100: bytes=32 time=20ms TTL=55
Reply from 85.236.100.100: bytes=32 time=14ms TTL=55
Reply from 85.236.100.100: bytes=32 time=39ms TTL=55
Reply from 85.236.100.100: bytes=32 time=14ms TTL=55
Reply from 85.236.100.100: bytes=32 time=13ms TTL=55
Reply from 85.236.100.100: bytes=32 time=15ms TTL=55
Reply from 85.236.100.100: bytes=32 time=15ms TTL=55
Reply from 85.236.100.100: bytes=32 time=14ms TTL=55
Reply from 85.236.100.100: bytes=32 time=13ms TTL=55
Ping statistics for 85.236.100.100:
Packets: Sent = 20, Received = 20, Lost = 0 (0% loss),
Approximate round trip times in milli-seconds: Minimum = 13ms, Maximum = 39ms, Average = 16ms

And the tracert -
Tracing route to beechbone.multiplay.co.uk [85.236.100.100]
over a maximum of 30 hops:
1 3 ms 1 ms <1 ms
2 7 ms 7 ms 8 ms
3 18 ms 10 ms 7 ms seac-cam-1b-e911.0.network.virginmedia.net [80.0.55.25]
4 9 ms 10 ms 8 ms leed-t3core-1b-ae2-0.network.virginmedia.net [195.182.175.177]
5 9 ms 9 ms 20 ms lee-bb-b-so-020-0.network.virginmedia.net [213.105.175.69]
6 46 ms 9 ms 9 ms lee-bb-a-ae0-0.network.virginmedia.net [62.253.187.185]
7 14 ms 13 ms 14 ms bre-bb-b-so-200-0.network.virginmedia.net [213.105.175.26]
8 18 ms 19 ms 14 ms telc-ic-1-as0-0.network.virginmedia.net [62.253.185.74]
9 25 ms 15 ms 31 ms ldn-b2-link.telia.net [213.248.100.97]
10 15 ms 14 ms 15 ms ldn-tch-i2-link.telia.net [80.91.250.214]
11 13 ms 13 ms 15 ms ldn-tch-i1-link.telia.net [80.91.250.217]
12 25 ms 23 ms 23 ms 113069-ldn-tch-i1.c.telia.net [213.248.74.42]
13 25 ms 14 ms 38 ms multiplay-gw.inx.datahop.net [195.72.129.14]
14 34 ms 14 ms 18 ms beechbone.multiplay.co.uk [85.236.100.100]
Trace complete.

[Noggo]

Originally Posted by Rob M

From windows you could also do:

ping 192.168.100.1 -t

Which would continuously ping the modem until you hit Ctrl+C.

I know . I was trying to figure out what Joxer meant by `ping -t 100 192.168.100.1'. I translated into ping the modem 100 times, hence `ping 192.198.100.1 -n 100'.

[Noggo]

Just done another trace to the games server to show peak time and apart from one glitch up to 43ms and one incident of packet loss, there's nowt wrong with the server. I'll be happy playing FPS games on there.

May be you're got some local issues effecting you. Have you been in contact with VM about it and had your connection checked by tech support?