Comparitive Latencies

I have come to the end of my tether it seems with Plusnet, so before I make my final decision to either leave or stay I'd just like to know what other people are experiencing in regards to latiences from Plusnet these days.

My Story: I first joined up with Plusnet over 2 years ago, and at which time I was happy with the service they provided - good bandwidth, good prices and great pings, consistently 12-14ms regardless of day or night.

However, today is a different story - with my bandwidth being all over the shop and my pings have got considerably worse, a price which has remained the same although amounts have decreased quite a bit and although I accept broadband heavy usage is not what it used to be, I still opt to pay for the highest price for the most bandwidth, which is only fair considering what I use.

I now suffer with this as my online ping:

Pinging www.jolt.co.uk [82.133.85.65] with 32 bytes of data:

Reply from 82.133.85.65: bytes=32 time=38ms TTL=55
Reply from 82.133.85.65: bytes=32 time=37ms TTL=55
Reply from 82.133.85.65: bytes=32 time=39ms TTL=55
Reply from 82.133.85.65: bytes=32 time=39ms TTL=55

Ping statistics for 82.133.85.65:
Packets: Sent = 4, Received = 4, Lost = 0 (0% loss),
Approximate round trip times in milli-seconds:
Minimum = 337ms, Maximum = 39ms, Average = 39ms

At this time of night, I'd say it was unacceptable considering usage dropping and all, and although its stable right now - when I play games during the day I'm often confronted with incredible amounts of packetloss.

I had suspected it to be my PC or the software I was using, so I reinstalled and double checked everything my end to no avail. I really cannot explain why my pings have almost tripled as a standard (which wasnt an instant switch but more of a progressive downfall) and have to tolerate such a poor connection for gaming.

My question: Could others of Plusnet, whom have both been here a short time and also longer term customers post some of their reports to jolt and see how mine compare, just to rule out anything wrong with my physical connections. Also I'm interested in people's opinions as to how latencies have changed for them, if at all, over the last 6-12 months.

Thanks in advance.

37ms is a fine ping for gaming. Plusnet's network is pretty much broke atm. It's useless for doing anything other than surfing and email at peak times IMO, though some pipes are better than others.

I'd highly recommend you migrate out because plusnet have no real strategy for solving this. Their current idea is to start classifying p2p encrypted traffic as bronze (and thus go slow as hell along with the rest of p2p) but this is nothing more than a stopgap measure as traffic will continue to rise, leading to this circumstance yet again.

Hi BM

C:\Documents and Settings\kitz>ping www.jolt.co.uk

Pinging www.jolt.co.uk [82.133.85.65] with 32 bytes of data:

Reply from 82.133.85.65: bytes=32 time=15ms TTL=56
Reply from 82.133.85.65: bytes=32 time=15ms TTL=56
Reply from 82.133.85.65: bytes=32 time=15ms TTL=56
Reply from 82.133.85.65: bytes=32 time=15ms TTL=56

Ping statistics for 82.133.85.65:
Packets: Sent = 4, Received = 4, Lost = 0 (0% loss),
Approximate round trip times in milli-seconds:
Minimum = 15ms, Maximum = 15ms, Average = 15ms


C:\Documents and Settings\kitz>tracert www.jolt.co.uk

Tracing route to www.jolt.co.uk [82.133.85.65]
over a maximum of 30 hops:

1 1 ms 1 ms 1 ms 192.168.7.1
2 15 ms 15 ms 15 ms lo0-plusnet.pte-ag1.plus.net [195.166.128.64]
3 15 ms 15 ms 15 ms ge0-0-0-403.pte-gw1.plus.net [84.92.4.1]
4 15 ms 16 ms 15 ms vl23.thn-gw1.plus.net [212.159.4.19]
5 16 ms 16 ms 15 ms g2-2-501.cr01.hx2.bb.pipex.net [193.203.5.14]
6 15 ms 15 ms 15 ms v3952.cr05.tn5.bb.pipex.net [62.72.137.9]
7 15 ms 15 ms 16 ms g1-1-6.er01.tn5.bb.pipex.net [62.72.140.142]
8 15 ms 15 ms 15 ms ge-0-0-0-3801.jolt-gw.cust.pipex.net [212.241.24
1.14]
9 15 ms 16 ms 16 ms secure.jolt.co.uk [82.133.85.65]

Trace complete.

----------------------------------

Firstly Ive got to ask the obvious - youre not on MAX and had interleaving applied have you?
fairysnuff that over and done with.......

37ms is fine and as long as they are stable then it shouldnt cause you any problems.
(yeah I know sorry its not much consolation to you - and I know its all about the numbers... but ask zen users if it affects their gaming)

Those pings seem in the same range.. and not much variation and makes me wonder if it has anything to do with your routing.
The reason I point this out is that at least 3 times since Ive had problems with latency over the 30ms.
The first time was in 2004 when overnight my latency went from 16ms to 32ms.
It was down to the BT routing and there was absolutely naff all I could do about it. It stayed like that for about 2 months, then suddenly cleared.
It happened another time mid 2005 after I'd been reRASed..
and again earlier this year - (yet another reRAS).. when once again my latency went up to 28ms.

Ironically this was to a much nearer RAS locationwise.. yet it took twice as long to get there.
You can see from the attached screen cap that it was BT routing because it took 25/26ms just to get to the RAS 50 miles away in Manchester.. nvm the additional 200+ miles down to telehouse.

Im not saying this is what has happened to you.. but it is a possibility.. and I have seen it happen to others.. and far worse - anyone remember poor rizlas problems no matter which ISP he went to :/
From what Ive seen in the past the largest portion timewise of latency seems to occur on the ATM backhaul.


----------

There is a problem with some of PNs gateways right now.. and Im not enamoured that we have to go pipe hopping.
It is obvious that there is a problem.. and although the overall latency may look ok... when you do a tracert you may see something like. (made up figures)

1. 1ms
2. 65ms
3. 15ms
4. 16ms
5. 16ms etc

From that its obvious that the Juniper is under pressure.. its passing packets through ok.. but its not actually responding to icmp pings as quick as it should... even though the overall figure may be fine.
Reconnect to a different pipe and things are ok.

Ironically I was one of the first saying that PN should cease the LLU transfers because of the problems experienced. However IMHO this has directly impacted on on centrals, since PN are moving some users back to IPStream, and its also not freeing up bandwidth on the centrals as new customers come onboard.
New pipe needed me thinks.

In the same boat as you, been a customer for about 2years and its only this last week or two that have been a problem.

I am working on a business account here where I can see the exchange (not interleaved) on a solid 8meg connection (so, there shouldnt be any management on the line).

Here are some stats:

You are currently connected to the "Telehouse North, Juniper 5" gateway. This gateway is sometimes known as thn-ag1.

H:\>tracert www.jolt.co.uk

Tracing route to www.jolt.co.uk [82.133.85.65]
over a maximum of 30 hops:


3 6 ms 13 ms 2 ms +++++++.plus.com [++.++.222.25]
4 17 ms 17 ms 17 ms lo0-plusnet.thn-ag1.plus.net [195.166.128.80]
5 21 ms 20 ms * vl303.thn-gw2.plus.net [84.92.3.178]
6 20 ms 19 ms 21 ms ge0-0-0-22.pte-gw1.plus.net [212.159.4.9]
7 21 ms 20 ms 20 ms vl23.thn-gw1.plus.net [212.159.4.19]
8 21 ms 21 ms 20 ms g2-2-501.cr01.hx2.bb.pipex.net [193.203.5.14]
9 * 22 ms 22 ms v3952.cr05.tn5.bb.pipex.net [62.72.137.9]
10 22 ms 20 ms 21 ms g3-1-5.er01.tn5.bb.pipex.net [62.72.140.14]
11 20 ms 18 ms 19 ms ge-0-0-0-3801.jolt-gw.cust.pipex.net [212.241.241.14]
12 21 ms 21 ms 22 ms secure.jolt.co.uk [82.133.85.65]

Trace complete.

H:\>ping www.plus.net

Pinging portal.plus.net [212.159.8.137] with 32 bytes of data:

Reply from 212.159.8.137: bytes=32 time=29ms TTL=246
Reply from 212.159.8.137: bytes=32 time=35ms TTL=246
Reply from 212.159.8.137: bytes=32 time=28ms TTL=246
Reply from 212.159.8.137: bytes=32 time=29ms TTL=246

Ping statistics for 212.159.8.137:
Packets: Sent = 4, Received = 4, Lost = 0 (0% loss),
Approximate round trip times in milli-seconds:
Minimum = 28ms, Maximum = 35ms, Average = 30ms


My home line stats (via a SSH connection test) also not interleaved (stable rate of 5500 and connected at 6500)

# ping plus.net
PING plus.net (212.159.8.137): 56 data bytes
64 bytes from 212.159.8.137: icmp_seq=0 ttl=249 time=50.0 ms
64 bytes from 212.159.8.137: icmp_seq=1 ttl=249 time=40.0 ms
64 bytes from 212.159.8.137: icmp_seq=2 ttl=249 time=40.0 ms
64 bytes from 212.159.8.137: icmp_seq=3 ttl=249 time=40.0 ms
64 bytes from 212.159.8.137: icmp_seq=4 ttl=249 time=40.0 ms
64 bytes from 212.159.8.137: icmp_seq=5 ttl=249 time=40.0 ms
64 bytes from 212.159.8.137: icmp_seq=6 ttl=249 time=40.0 ms
64 bytes from 212.159.8.137: icmp_seq=7 ttl=249 time=40.0 ms
64 bytes from 212.159.8.137: icmp_seq=8 ttl=249 time=40.0 ms
64 bytes from 212.159.8.137: icmp_seq=9 ttl=249 time=40.0 ms


Jolt.co.uk 30-40ms

--- jolt.co.uk ping statistics ---
16 packets transmitted, 16 packets received, 0% packet loss
round-trip min/avg/max = 30.0/30.6/40.0 ms


Over the weekend, it hit its worst where I was getting pings between 55-80ms to jolt.co.uk

37ms as a ping is not good enough in my opinion. Once you connect to a server and add overheads it gets worse. When I joined plusnet my pings were an avg of 16ms

Edited by deleted (Tue 12-Sep-06 10:39:24)

Thankfully (touching a major piece of wood) my first hop latency is around 16-17ms now.

A long time ago, my first hop latency was 43ms. I'm in the midlands, and the culprit was my RAS being in Edinburgh!

When they changed my RAS to Birmingham (I had to dance with the devil to get this done) it reduced to 23ms.

When I got moved to Max the must have moved me on to another DSLAM or VP too, as now I have 16-17ms, but with the same BRAS.

Edited by deleted (Tue 12-Sep-06 11:22:59)

Just to clarify, yes I am on MaxDSL and yes I have already had Plusnet remove interleaving in hopes that that would clear it, but it hasnt.

Unfortunately those pings taken were late at night and resonably stable, however during the day its a different story, with lots of packet loss in most cases. I've left one running to show the problem so I'll just post the summary since it would take up loads of space otherwise.

Ping statistics for 82.133.85.65:
Packets: Sent = 100, Received = 87, Lost = 13 (13% loss),
Approximate round trip times in milli-seconds:
Minimum = 35ms, Maximum = 45ms, Average = 36ms

I was having similar ping problems after my line was Max'd and interleaving kicked in. I requested interleaving be disabled on my line and was reassured that it was disabled, however my pings and modem stats contradicted this, pings were still higher than average and modem stats showed interleaved instead of fast, it drove me crazy. Then four weeks after Plusnet had apparently disabled interleaving, interleaving was turned off and all returned to normal. My advice would be to check your modem/router stats and determine whether interleaving has *actually* been disabled.

The problem was whilst the BT information sent to Plusnet confirmed interleaving was disabled, it seemed BT had not actually switched it off until four weeks later!

------

Tracing route to www.jolt.co.uk [82.133.85.65]
over a maximum of 30 hops:

1 16 ms 16 ms 16 ms lo0-plusnet.pte-ag1.plus.net [195.166.128.64]
2 17 ms 52 ms 17 ms ge0-0-0-403.pte-gw2.plus.net [84.92.4.2]
3 19 ms 16 ms 16 ms vl23.thn-gw1.plus.net [212.159.4.19]
4 18 ms 19 ms 16 ms g2-2-501.cr01.hx2.bb.pipex.net [193.203.5.14]
5 17 ms 17 ms 19 ms v3952.cr05.tn5.bb.pipex.net [62.72.137.9]
6 18 ms 17 ms 18 ms g1-1-6.er01.tn5.bb.pipex.net [62.72.140.142]
7 16 ms 17 ms 19 ms ge-0-0-0-3801.jolt-gw.cust.pipex.net [212.241.24
1.14]
8 15 ms 15 ms 16 ms secure.jolt.co.uk [82.133.85.65]

Trace complete.


Tracing route to www.jolt.co.uk [82.133.85.65]
over a maximum of 30 hops:

1 27 ms 18 ms 17 ms lo0-plusnet.pte-ag1.plus.net [195.166.128.64]
2 17 ms 16 ms 16 ms ge0-0-0-403.pte-gw2.plus.net [84.92.4.2]
3 18 ms 16 ms 16 ms vl23.thn-gw1.plus.net [212.159.4.19]
4 17 ms 16 ms 17 ms g2-2-501.cr01.hx2.bb.pipex.net [193.203.5.14]
5 17 ms 19 ms 16 ms v3952.cr05.tn5.bb.pipex.net [62.72.137.9]
6 19 ms 18 ms 16 ms g1-1-6.er01.tn5.bb.pipex.net [62.72.140.142]
7 17 ms 17 ms 17 ms ge-0-0-0-3801.jolt-gw.cust.pipex.net [212.241.24
1.14]
8 16 ms 16 ms 16 ms secure.jolt.co.uk [82.133.85.65]


Pinging [82.133.85.65] with 32 bytes ->bytes=32 time=17ms TTL=57
Pinging [82.133.85.65] with 32 bytes ->bytes=32 time=18ms TTL=57
Pinging [82.133.85.65] with 32 bytes ->bytes=32 time=17ms TTL=57
Pinging [82.133.85.65] with 32 bytes ->bytes=32 time=16ms TTL=57
Pinging [82.133.85.65] with 32 bytes ->bytes=32 time=17ms TTL=57
Pinging [82.133.85.65] with 32 bytes ->bytes=32 time=17ms TTL=57
Pinging [82.133.85.65] with 32 bytes ->bytes=32 time=17ms TTL=57
Pinging [82.133.85.65] with 32 bytes ->bytes=32 time=17ms TTL=57
Ping statistics for above hosts:
Packets: Sent = 8, Received = 8, Lost = 0 (0% loss)
Approximate round trip times (RTT) in milli-seconds:
Minimum = 16ms, Maximum = 18ms, Average = 17ms

------

past plusnet customer here...

heres the view from northern ireland using BTBroadband.

Pinging www.jolt.co.uk [82.133.85.65] with 32 bytes of data:

Reply from 82.133.85.65: bytes=32 time=30ms TTL=51
Reply from 82.133.85.65: bytes=32 time=30ms TTL=51
Reply from 82.133.85.65: bytes=32 time=29ms TTL=51
Reply from 82.133.85.65: bytes=32 time=30ms TTL=51

Ping statistics for 82.133.85.65:
Packets: Sent = 4, Received = 4, Lost = 0 (0% loss),
Approximate round trip times in milli-seconds:
Minimum = 29ms, Maximum = 30ms, Average = 29ms