Menu
Reply
Highlighted
  • 4
  • 0
  • 0
Joining in
575 Views
Message 1 of 8
Flag for a moderator

Latency - here we go again

I wrote to VM CEO 5th Feb 2020, more latency issues, lutz.schueler@virginmedia.co.uk

I haven't included all the details, but here is a summary since I joined VM:
  • I'm a senior member of a gaming community, server in the USA.
  • I joined VM to upgrade to 100mb, also upload was better than other ISP's.
  • Latency was terrible, UK to Chicago USA, pings >148, previous ISP 110.
  • Complained April 2018, eventually went to VM CEO Complaints, spoke to top UK tech support team (and Liberty Global tech team), all agreed (verbally) pings seemed to be high. Sent them loads of latency data and proof the same route should actually be <100. 
  • Main issue pointed to VM's International Carrier Telnet. Contacted Telnet, refused to deal with it as I was VM customer, told me to raise it with VM!
  • Complaint dismissed by VM CEO Complaints as contract not breached.
  • Raised with CISAS, not upheld, mainly no breach of contract!
  • Ping did become stable at 148. Decided to stay with VM as just renewed contract. Perhaps change ISP when my contract ends.
  • Ping improved over time >135 and was fairly stable so stayed with VM. 
  • Contract due for renewal Jan 2020, spoke to Retention's team, good price for upgrade to 200mb, plus upload speed improved to 20mb. Confirmed ping should also reduce. 
  • Ping about 122, varied a bit to 130. Still not great but I was generally happy. 
  • Two weeks later sudden terrible lag on server. Tested ping jumping between 126 and 350 and wouldn't stabilise. 
  • Spoke to offshore so called tech support, they haven't got a clue, follow scripts, I've got a better understanding of broadband. 
  • Raised online complaint, no reply to date. As above, wrote to CEO, no reply to date!
  • Ping unstable, did settle to 148 a couple of times but now all over the place!

This is a table from a national and international carrier company:

  Cardiff Chicago Edinburgh Maidstone Newcastle

London5.331ms86.177ms10.559ms1.518ms8.945ms

I'm getting 28ms to Maidstone, live in west Essex?

I have read so many posts on this forum about Latency. When is VM going to wake up and realise the customer demand is not always about bandwidth, which is actually throughput, not speed? I know it's more complicated than this but these days with increasing bandwidths and demand for online services, streaming movies, gaming etc low latency is a critical factor for a good quality service.

0 Kudos
Reply
Highlighted
  • 10.96K
  • 730
  • 1.52K
Alessandro Volta
559 Views
Message 2 of 8
Flag for a moderator

Re: Latency - here we go again

If Latency is important to you why are you with virgin media?

0 Kudos
Reply
Highlighted
  • 3.29K
  • 177
  • 507
Wise owl
549 Views
Message 3 of 8
Flag for a moderator

Re: Latency - here we go again

For speed > Virgin Media 

For latency > Openreach 

0 Kudos
Reply
Highlighted
  • 2.41K
  • 152
  • 618
Problem sorter
535 Views
Message 4 of 8
Flag for a moderator

Re: Latency - here we go again

DOCSIS connections always suffer higher latency/jitter than VDSL. 140ms to Chicago doesn't seem too bad although 350ms definitely isn't normal. I've not heard of Telnet but VM tend to use the cheaper carriers like Telia and Cogent (and Liberty Global's own Aorta network) which can add to the latency. 

Are you able to share the IP/URL of the server then we could see if everyone gets the same problem?

0 Kudos
Reply
Highlighted
  • 4
  • 0
  • 0
Joining in
526 Views
Message 5 of 8
Flag for a moderator

Re: Latency - here we go again

VM in my area is the only company that provides a pretty stable fibre network, from a higher bandwidth point of view, even though it's not purely fibre all the way. Fact, higher grade copper from home to fibre cabinet. All other ISP's, in my area, have to use BT's very unstable old copper network to a fibre cabinet and bandwidth can be highly compromised. In my case it's a long run to the fibre cabinet.

As ISP's go VM has been the best service I have had (not customer service but most are not good either) from a broadband point of view, no major issues, no downtime etc, apart from latency.

It's called a trade off, I need a good bandwidth and upload speed for what I do, I need as low latency as I can get but I can live with it if it's constantly stable and I did mention it's been fairly stable since I last complained, mid 2018. My previous ISP was a lot better as far as latency is concerned but I had terrible network problems. My point is why doesn't VM acknowledge these latency issues and try and resolve them? Their advisers don't know what to say or do and as the bandwidth tests are ok they think a problem doesn't exist, read the posts. 

I do not know why it's become so unstable now but I do believe VM is aware that the International Carriers it uses do use lower quality networks, probably costs. The experts agreed with me but obviously wouldn't put it in writing! I know someone in Holland who's ping is 102, almost constant, to the same server! Check the table I posted, it confirms what the latency should be like!  

Now it's unstable, if it doesn't change, I may have to consider leaving but what a pain! I read somewhere that VM's Hub3 may be causing some of the local latency issues so I will try that first as I also have a Hub2.

Hope this answers your question?  

 

0 Kudos
Reply
Highlighted
  • 4
  • 0
  • 0
Joining in
524 Views
Message 6 of 8
Flag for a moderator

Re: Latency - here we go again

Sorry I did say Telnet I meant Telia!  

Here you go 74.91.119.166

Thanks for being helpful.

0 Kudos
Reply
Highlighted
  • 2.41K
  • 152
  • 618
Problem sorter
429 Views
Message 7 of 8
Flag for a moderator
Helpful Answer

Re: Latency - here we go again

I've tested it on my VM connection and also got a friend to try it on their EE connection and these are the results

 

Virgin Media

tracert 74.91.119.166

Tracing route to v-74-91-119-166.unman-vds.premium-chicago.nfoservers.com [74.91.119.166]
over a maximum of 30 hops:

1 <1 ms <1 ms <1 ms RT-AC86U-7358 [192.168.1.1]
2 12 ms 9 ms 10 ms 10.250.64.1
3 12 ms 13 ms 11 ms pres-core-2a-xe-714-0.network.virginmedia.net [80.0.160.65]
4 * * * Request timed out.
5 * * * Request timed out.
6 25 ms 26 ms 24 ms 86.85-254-62.static.virginmediabusiness.co.uk [62.254.85.86]
7 24 ms 26 ms 24 ms 213.46.174.46
8 26 ms 25 ms 24 ms be2868.ccr41.lon13.atlas.cogentco.com [154.54.57.153]
9 93 ms 101 ms 93 ms be2099.ccr31.bos01.atlas.cogentco.com [154.54.82.34]
10 92 ms 94 ms 92 ms be3599.ccr21.alb02.atlas.cogentco.com [66.28.4.237]
11 132 ms 132 ms 135 ms be2878.ccr21.cle04.atlas.cogentco.com [154.54.26.129]
12 132 ms 131 ms 132 ms be2717.ccr41.ord01.atlas.cogentco.com [154.54.6.221]
13 136 ms 140 ms 131 ms be2765.ccr41.ord03.atlas.cogentco.com [154.54.45.18]
14 131 ms 138 ms 140 ms cogent.e9.router1.chicago.nfoservers.com [64.74.97.254]
15 135 ms 130 ms 132 ms v-74-91-119-166.unman-vds.premium-chicago.nfoservers.com [74.91.119.166]

Trace complete.


EE

tracert 74.91.119.166

Tracing route to v-74-91-119-166.unman-vds.premium-chicago.nfoservers.com [74.91.119.166]
over a maximum of 30 hops:

1 <1 ms <1 ms <1 ms eehub.home [192.168.1.254]
2 4 ms 4 ms 4 ms 172.16.14.227
3 * * * Request timed out.
4 14 ms 14 ms 14 ms 213.121.98.136
5 23 ms 16 ms 15 ms 87.237.20.132
6 * 16 ms 15 ms lag-107.ear1.London1.Level3.net [212.187.166.145]
7 15 ms 14 ms 15 ms ae-2-3214.edge5.London1.Level3.net [4.69.148.222]
8 15 ms 15 ms 14 ms ae-3.a00.londen12.uk.bb.gin.ntt.net [129.250.9.125]
9 102 ms 101 ms 101 ms ae-16.r04.londen12.uk.bb.gin.ntt.net [129.250.2.137]
10 17 ms 16 ms 15 ms ae-4.r24.londen12.uk.bb.gin.ntt.net [129.250.4.125]
11 * 85 ms * ae-7.r20.nwrknj03.us.bb.gin.ntt.net [129.250.6.147]
12 * * * Request timed out.
13 103 ms 105 ms 103 ms ae-1.r20.chcgil09.us.bb.gin.ntt.net [129.250.2.166]
14 105 ms 104 ms 104 ms ae-6.r07.chcgil09.us.bb.gin.ntt.net [129.250.6.61]
15 102 ms 102 ms 101 ms ntt.e9.router2.chicago.nfoservers.com [74.91.119.253]
16 101 ms 101 ms 102 ms v-74-91-119-166.unman-vds.premium-chicago.nfoservers.com [74.91.119.166]

Trace complete.

EE is definitely better, some of which will be down to the technology (VDSL) but also better transit from NTT. Unfortunately there's nothing us customers can do about that. Virgin Media's international peering has always let it down, especially for gamers. 

Highlighted
  • 4
  • 0
  • 0
Joining in
381 Views
Message 8 of 8
Flag for a moderator

Re: Latency - here we go again

cje85,

Thanks so much for posting these results and it proves my points perfectly. The EE trace route is what I'd expect from the UK depending on where you are located. I know what you mean when you say customers can't do a lot about it but if enough continue to raise this with VM then just maybe! Either that or leave!

What does infuriate me is not matter how much proof we throw at VM they still ignore it. Maybe one day the regulators will demand that latency levels are part of ISP's contracts. I know they have been looking at it. 

I suppose I may need to have another close look at what is more important for me! 

It's no good paying for a larger water pipe, to deliver your water needs, if it's full of holes!

0 Kudos
Reply