Lowest Latency Openreach FTTP ISP?

What is the lowest latency openreach FTTP ISP?

Surely geographic location has a fair deal to do with it ?

I get 15ms to London from NW Ireland (Derry) - Always found BTs routing and latency great.

PING bbc.co.uk (151.101.192.81): 56 data bytes
64 bytes from 151.101.192.81: seq=0 ttl=58 time=15.605 ms
64 bytes from 151.101.192.81: seq=1 ttl=58 time=15.470 ms
64 bytes from 151.101.192.81: seq=2 ttl=58 time=15.506 ms
64 bytes from 151.101.192.81: seq=3 ttl=58 time=15.621 ms
64 bytes from 151.101.192.81: seq=4 ttl=58 time=15.541 ms

--- bbc.co.uk ping statistics ---
5 packets transmitted, 5 packets received, 0% packet loss
round-trip min/avg/max = 15.470/15.548/15.621 ms

Edited by BuckleZ (Sun 31-Oct-21 21:42:39)

Sorry, I forgot to mention I live in London.

I mean if your london based most the ISP's should have data centres there, I believe Zen, BT do, can't speak for talktalk although i would imagine so.

Take a look at the latest ofcom report not much data but there is a bit, it gives a vague general trend even if it included various technologies although won't be bang on accurate it just helps you see which ISP's overall have lower latencies.
Ofcom report march 2021 data

Chart

I live in the south west (not London) and always get exceptionally good latency given the fact my connection has to go all the way up to London and back. I can't imagine you'll have any issues with any ISP in London unless routing in their core network is somehow broken. Recently, I've seen a lot of latency and packet loss in BT's core network lately when running traceroutes/mtr to a VPS in London and putting my machine under a heavy load, there's none on my router which has an average of 0.1-0.2ms of latency but some on the next hop and this persists. I know they'll fix this, but just pointing it out.

Edited by aidanh (Sun 31-Oct-21 22:45:17)

Recently, I've seen a lot of latency and packet loss in BT's core network

How long ago did this start happening?

Within the past week and it only shows up under a high-load (just running a speedtest on speedtest.net or fast.com with mtr running in the background is enough to cause loss).

Some examples:

IPv4 mtr to my VPS whilst running a speedtest (no loss on the first hop this time, but I've seen that before).

Host Loss% Snt Last Avg Best Wrst StDev
1. AS??? OpenWRT (192.168.0.254) 0.0% 106 0.1 0.1 0.1 0.2 0.0
2. AS??? 172.16.16.87 (172.16.16.87) 0.0% 106 1.3 1.2 0.6 1.9 0.3
3. AS2856 62.172.102.69 (62.172.102.69) 87.6% 106 5.8 5.9 5.2 7.0 0.4
4. AS2856 62.172.102.76 (62.172.102.76) 0.0% 106 6.1 6.0 5.3 6.6 0.3
5. AS2856 peer3-et4-1-7.slough.ukcore.bt.net (62.172.102.49) 0.9% 106 6.4 8.8 5.8 36.1 6.6
6. (waiting for reply)
7. (waiting for reply)
8. (waiting for reply)
9. (waiting for reply)
10. AS14061 165.227.XXX.XXX (165.227.XXX.XXX) 1.0% 105 6.7 7.0 6.4 7.5 0.3

IPv6 mtr to my VPS whilst running a speedtest where loss is clearly visible on the first hop after my router and persists:

Host Loss% Snt Last Avg Best Wrst StDev
1. AS2856 2a00:23c7:XXXX:fe00::1 (2a00:23c7:XXXX:fe00::1) 0.0% 82 0.2 0.1 0.1 0.2 0.0
2. AS5400 2a00:2302::1102:203:4c8 (2a00:2302::1102:203:4c8) 1.2% 82 1.4 1.5 0.6 21.4 2.3
3. AS5400 2a00:2302::1115:100:1e (2a00:2302::1115:100:1e) 89.0% 82 5.5 5.5 5.1 5.8 0.2
4. AS5400 2a00:2302::1115:100:23 (2a00:2302::1115:100:23) 22.2% 82 5.3 5.8 5.3 6.4 0.3
5. AS2856 peer3-et3-0-6.slough.ukcore.bt.net (2a00:2380:106::19) 1.2% 82 20.3 9.0 5.9 49.0 7.5
6. (waiting for reply)
7. (waiting for reply)
8. (waiting for reply)
9. (waiting for reply)
10. AS14061 2a03:b0c0:1:d0::XXX:XXXX (2a03:b0c0:1:d0::XXX:XXXX) 1.2% 81 6.6 6.5 5.9 7.2 0.3

E1234TJ

Living in London your latency is likely to be 2-5 ms at max over any ISP to any London destination. To anywhere else depends on ISP interconnects and where you are testing too.

From London,once outside the UK sheer distance overcomes ISP differences within the UK.

In reply to a post by E1234TJ:

What is the lowest latency openreach FTTP ISP?

Well I also live in London, and when I used my BT connection we got in the range of 2ms to 5ms (with the odd 1ms) to places in the UK, EU about 5 to 10ms (WoW EU Servers) I cannot test now due to I am now using a different provider.

Paul

I'm up near Rotherham, about 160 miles from London, and get 8 or 9 ms from BBC.co.uk

On Cerberus networks FTTP connected via Amvia:

Pinging bbc.co.uk [151.101.0.81] with 32 bytes of data:
Reply from 151.101.0.81: bytes=32 time=8ms TTL=60
Reply from 151.101.0.81: bytes=32 time=9ms TTL=60
Reply from 151.101.0.81: bytes=32 time=8ms TTL=60
Reply from 151.101.0.81: bytes=32 time=8ms TTL=60

Ping statistics for 151.101.0.81:
Packets: Sent = 4, Received = 4, Lost = 0 (0% loss),
Approximate round trip times in milli-seconds:
Minimum = 8ms, Maximum = 9ms, Average = 8ms

I'm in South East London (Orpington area), using TalkTalk via OpenReach.

Ping results:
[~] # ping bbc.co.uk
PING bbc.co.uk (151.101.0.81): 56 data bytes
64 bytes from 151.101.0.81: seq=0 ttl=59 time=2.449 ms
64 bytes from 151.101.0.81: seq=1 ttl=59 time=2.263 ms
64 bytes from 151.101.0.81: seq=2 ttl=59 time=2.173 ms
64 bytes from 151.101.0.81: seq=3 ttl=59 time=2.127 ms

[~] # ping apple.com
PING apple.com (17.253.144.10): 56 data bytes
64 bytes from 17.253.144.10: seq=0 ttl=59 time=2.684 ms
64 bytes from 17.253.144.10: seq=1 ttl=59 time=2.106 ms
64 bytes from 17.253.144.10: seq=2 ttl=59 time=2.621 ms
64 bytes from 17.253.144.10: seq=3 ttl=59 time=1.878 ms

Looks like you’re hitting their CDN’s. Try the nytimes.com. It will ping faster than the beeb 🤣

Surely info about your current provider is useful to others too Paul ?

In reply to a post by Pheasant:

Looks like you’re hitting their CDN’s. Try the nytimes.com. It will ping faster than the beeb 🤣

Surely the point of CDN's is that they're 'local' caches of the content to be delivered...so they *should* be low latency (assuming that one's DNS setup picks up the relevant 'local' IP)

Anyway...not so fast on this one...

[~] # ping nytimes.com
PING nytimes.com (151.101.193.164): 56 data bytes
64 bytes from 151.101.193.164: seq=0 ttl=59 time=3.151 ms
64 bytes from 151.101.193.164: seq=1 ttl=59 time=3.262 ms
64 bytes from 151.101.193.164: seq=2 ttl=59 time=3.088 ms
64 bytes from 151.101.193.164: seq=3 ttl=59 time=2.994 ms

In reply to a post by E1234TJ:

What is the lowest latency openreach FTTP ISP?

There is no single answer to that question.
Different backhaul networks can and do have lower latency in 1 City compared to another city.
BT Wholesale might have lower latency than Talktalk in City A but higher in City B.

In general though BT Wholesale has 1 of the lowest latencies over OpenReach.
Much of Talktalks backhaul network daisy chains from exchange to exchange, adding distance and latency.

Up here in Edinburgh I get a minimum 13.9ms to London via BT Wholesale (over OpenReach FTTP).
On the same line I get 16.1ms to London over Talktalk backhaul.
I've no personal experience of Sky or Vodafones networks from my location.

There is likely less of a difference in latency between backhaul networks for anyone in London.

Here is mine gfast with talktalk business (unchainedisp)

Microsoft Windows [Version 10.0.22000.282]
(c) Microsoft Corporation. All rights reserved.

C:\Users\PC>ping bbc.co.uk

Pinging bbc.co.uk [151.101.0.81] with 32 bytes of data:
Reply from 151.101.0.81: bytes=32 time=8ms TTL=59
Reply from 151.101.0.81: bytes=32 time=8ms TTL=59
Reply from 151.101.0.81: bytes=32 time=8ms TTL=59
Reply from 151.101.0.81: bytes=32 time=8ms TTL=59

Ping statistics for 151.101.0.81:
Packets: Sent = 4, Received = 4, Lost = 0 (0% loss),
Approximate round trip times in milli-seconds:
Minimum = 8ms, Maximum = 8ms, Average = 8ms

C:\Users\PC>

This is what I get on Community Fibre at the moment (I am just over 5km birds eye from Telehouse)

Ping

ping -4 bbc.co.uk

Pinging bbc.co.uk [151.101.64.81] with 32 bytes of data:
Reply from 151.101.64.81: bytes=32 time=1ms TTL=59
Reply from 151.101.64.81: bytes=32 time=1ms TTL=59
Reply from 151.101.64.81: bytes=32 time=1ms TTL=59
Reply from 151.101.64.81: bytes=32 time=1ms TTL=59

Ping statistics for 151.101.64.81:
    Packets: Sent = 4, Received = 4, Lost = 0 (0% loss),
Approximate round trip times in milli-seconds:
    Minimum = 1ms, Maximum = 1ms, Average = 1ms

Trace Route

tracert -4 bbc.co.uk

Tracing route to bbc.co.uk [151.101.0.81]
over a maximum of 30 hops:

  1    <1 ms    <1 ms    <1 ms  yazoo [192.168.2.253]
  2     1 ms     1 ms     1 ms  x.x.x.x
  3     2 ms     2 ms     2 ms  94.177.138.59
  4     2 ms     1 ms     1 ms  94.177.138.170
  5     1 ms     1 ms     1 ms  195.66.225.91
  6     1 ms     1 ms     1 ms  151.101.0.81

Trace complete.

But if I recall I got about the same with BT when I last did a test on it.

Paul

Yes precisely. The CDNs are exceedingly optimised for that function.

In your pings above to bbc.co.uk and nytimes.com you're actually hitting Fastly's CDN (151.101.xxx.yyy) in both cases. From where I am, the times to both is effectively the same, and nytimes.com content on average (at the moment) is slightly faster (1.254 ms) than the beeb (1.278 ms) - but they are coming from the same delivery network (they had that rather 'spectacular' outage back in June..the 'OMG the internet is broken' outage)

I don't know about the OP, but the folks that typically obsess over broadband ping times are invariably gamers. So they're looking at CDN perhaps where games and patches are hosted for download, but also standalone / independent servers etc often or not outside of the major CDN's and located far and wide, not just in Slough and the Docklands.

In reply to a post by Zarjaz:

Surely info about your current provider is useful to others too Paul ?

Yeah, I posted a Ping and Traceroute to the BBC Site in a previous reply.

As my sig (assuming people have it display it) says, I am now with Community Fibre on their GIGAFAST package, along with TBB IPv4, IPv6 and Ookla Speedtest Results along with TBB BQM.

I still have BT (MikroTik Tik Router not connected to anything) but just waiting for the contract to run out for the broadband and will then only use BT as a Phone service until we get around to port the number.

Ping to BBC Site

ping -4 bbc.co.uk

Pinging bbc.co.uk [151.101.0.81] with 32 bytes of data:
Reply from 151.101.0.81: bytes=32 time=1ms TTL=59
Reply from 151.101.0.81: bytes=32 time=1ms TTL=59
Reply from 151.101.0.81: bytes=32 time=1ms TTL=59
Reply from 151.101.0.81: bytes=32 time=1ms TTL=59

Ping statistics for 151.101.0.81:
    Packets: Sent = 4, Received = 4, Lost = 0 (0% loss),
Approximate round trip times in milli-seconds:
    Minimum = 1ms, Maximum = 1ms, Average = 1ms

Traceroute to BBC Site

tracert -4 bbc.co.uk

Tracing route to bbc.co.uk [151.101.0.81]
over a maximum of 30 hops:

  1     1 ms    <1 ms    <1 ms  yazoo [192.168.2.253]
  2     1 ms     1 ms     1 ms  x.x.x.x
  3     1 ms     2 ms     1 ms  94.177.138.59
  4     1 ms     2 ms     1 ms  94.177.138.170
  5    <1 ms     1 ms    <1 ms  195.66.225.91
  6     1 ms     1 ms     1 ms  151.101.0.81

Trace complete.

---

Ping to TBB BQM Server

ping pingbox1.thinkbroadband.com

Pinging pingbox1.thinkbroadband.com [80.249.99.164] with 32 bytes of data:
Reply from 80.249.99.164: bytes=32 time=1ms TTL=57
Reply from 80.249.99.164: bytes=32 time=1ms TTL=57
Reply from 80.249.99.164: bytes=32 time=1ms TTL=57
Reply from 80.249.99.164: bytes=32 time=1ms TTL=57

Ping statistics for 80.249.99.164:
    Packets: Sent = 4, Received = 4, Lost = 0 (0% loss),
Approximate round trip times in milli-seconds:
    Minimum = 1ms, Maximum = 1ms, Average = 1ms

Traceroute to TBB BQM Server

tracert pingbox1.thinkbroadband.com

Tracing route to pingbox1.thinkbroadband.com [80.249.99.164]
over a maximum of 30 hops:

  1     1 ms    <1 ms    <1 ms  yazoo [192.168.2.253]
  2     1 ms     1 ms     1 ms  x.x.x.x
  3     1 ms     2 ms     2 ms  94.177.138.59
  4     1 ms     1 ms     2 ms  94.177.138.170
  5     1 ms     1 ms     1 ms  185.59.125.153
  6     1 ms     1 ms     1 ms  po11-13.bdr-rt3.thdo.ncuk.net [80.249.97.22]
  7     1 ms     1 ms    <1 ms  po4-31.core-rs4.thdo.ncuk.net [80.249.97.85]
  8     1 ms     1 ms     1 ms  pingbox1.thinkbroadband.com [80.249.99.164]

Trace complete.

I can do more tests if anyone wants any more information.

Paul

I just looked at all the pings being shown here and me with my connection in cambridgeshire its like being in another country.

ping -t bbc.co.uk

Pinging bbc.co.uk [151.101.128.81] with 32 bytes of data:
Reply from 151.101.128.81: bytes=32 time=49ms TTL=57
Reply from 151.101.128.81: bytes=32 time=64ms TTL=57
Reply from 151.101.128.81: bytes=32 time=42ms TTL=57
Reply from 151.101.128.81: bytes=32 time=50ms TTL=57
Reply from 151.101.128.81: bytes=32 time=56ms TTL=57
Reply from 151.101.128.81: bytes=32 time=41ms TTL=57
Reply from 151.101.128.81: bytes=32 time=41ms TTL=57
Reply from 151.101.128.81: bytes=32 time=48ms TTL=57
Reply from 151.101.128.81: bytes=32 time=42ms TTL=57
Reply from 151.101.128.81: bytes=32 time=46ms TTL=57
Reply from 151.101.128.81: bytes=32 time=49ms TTL=57
Reply from 151.101.128.81: bytes=32 time=45ms TTL=57

Ping statistics for 151.101.128.81:
Packets: Sent = 12, Received = 12, Lost = 0 (0% loss),
Approximate round trip times in milli-seconds:
Minimum = 41ms, Maximum = 64ms, Average = 47ms


added trace

Tracing route to bbc.co.uk [151.101.0.81]
over a maximum of 30 hops:

1 3 ms 3 ms 4 ms TP-SHARE [10.0.0.1]
2 * * * Request timed out.
3 44 ms 43 ms 36 ms nrth-core-2a-xe-1001-0.network.virginmedia.net [80.1.202.173]
4 * * * Request timed out.
5 39 ms 62 ms 46 ms eislou2-ic-3-ae0-0.network.virginmedia.net [94.174.238.226]
6 48 ms 53 ms 43 ms 157.52.127.6
7 38 ms 42 ms 48 ms 151.101.0.81

Trace complete.

Edited by RR_The_IT_Guy (Mon 01-Nov-21 23:06:16)

I switched from VM to BT.

bbc averaged 22-25ms on VM.

Now:
PING bbc.co.uk (151.101.192.81): 56 data bytes
64 bytes from 151.101.192.81: seq=0 ttl=58 time=15.506 ms
64 bytes from 151.101.192.81: seq=1 ttl=58 time=15.580 ms
64 bytes from 151.101.192.81: seq=2 ttl=58 time=15.492 ms
64 bytes from 151.101.192.81: seq=3 ttl=58 time=15.526 ms
64 bytes from 151.101.192.81: seq=4 ttl=58 time=15.554 ms

--- bbc.co.uk ping statistics ---
5 packets transmitted, 5 packets received, 0% packet loss
round-trip min/avg/max = 15.492/15.531/15.580 ms

In reply to a post by Pheasant:

Yes precisely. The CDNs are exceedingly optimised for that function.

Agreed.

In reply to a post by Pheasant:

In your pings above to bbc.co.uk and nytimes.com you're actually hitting Fastly's CDN (151.101.xxx.yyy) in both cases. From where I am, the times to both is effectively the same, and nytimes.com content on average (at the moment) is slightly faster (1.254 ms) than the beeb (1.278 ms) - but they are coming from the same delivery network (they had that rather 'spectacular' outage back in June..the 'OMG the internet is broken' outage)

I was about to say that they are via AS54113 - Fastly I just use those sites when others are using them to compare.
And yeah they was the exact same for both sites and so does twitch.tv

In reply to a post by Pheasant:

I don't know about the OP, but the folks that typically obsess over broadband ping times are invariably gamers. So they're looking at CDN perhaps where games and patches are hosted for download, but also standalone / independent servers etc often or not outside of the major CDN's and located far and wide, not just in Slough and the Docklands.

While I game a lot online and also do stream stuff now and then, latency doesn't need to be super low, but more of a stable one, I get about 8ms when I play on World of Warcraft EU Servers, so its fine for my needs.

Paul

In reply to a post by RR_The_IT_Guy:

I just looked at all the pings being shown here and me with my connection in cambridgeshire its like being in another country.

It's your provider rather than your county. Elon can get broadband from outer space in faster time

In reply to a post by PaulKirby:

... latency doesn't need to be super low, but more of a stable one

Agree. An often overlooked but very important signifier of connection quality and consistency, especially for all real time apps.

In reply to a post by Pheasant:

In reply to a post by PaulKirby:

... latency doesn't need to be super low, but more of a stable one

Agree. An often overlooked but very important signifier of connection quality and consistency, especially for all real time apps.

Yeah, I have played an online game with 1000ms (yes 1 second ping) but it was playable due to is was a very stable 1 second ping LOL

Paul

Is it possible if you can ping these and share your results?
qosping-aws-eu-west-2.ol.epicgames.com (london)
qosping-aws-eu-central-1.ol.epicgames.com (frankfurt)

Edited by E1234TJ (Tue 02-Nov-21 00:27:56)

In reply to a post by E1234TJ:

Is it possible if you can ping these and share your results?
qosping-aws-eu-west-2.ol.epicgames.com (london)
qosping-aws-eu-central-1.ol.epicgames.com (frankfurt)

Sure. Here you are:

[~] # ping qosping-aws-eu-west-2.ol.epicgames.com
PING qosping-aws-eu-west-2.ol.epicgames.com (52.56.149.83): 56 data bytes
64 bytes from 52.56.149.83: seq=0 ttl=27 time=4.021 ms
64 bytes from 52.56.149.83: seq=1 ttl=27 time=3.995 ms
64 bytes from 52.56.149.83: seq=2 ttl=27 time=3.922 ms
64 bytes from 52.56.149.83: seq=3 ttl=27 time=3.854 ms

[~] # ping qosping-aws-eu-central-1.ol.epicgames.com
PING qosping-aws-eu-central-1.ol.epicgames.com (35.158.209.186): 56 data bytes
64 bytes from 35.158.209.186: seq=0 ttl=48 time=15.700 ms
64 bytes from 35.158.209.186: seq=1 ttl=48 time=15.779 ms
64 bytes from 35.158.209.186: seq=2 ttl=48 time=17.364 ms
64 bytes from 35.158.209.186: seq=3 ttl=48 time=17.167 ms

They’re very solid results. I’ll post up some comparators.

In reply to a post by j0hn83:

…
BT Wholesale might have lower latency than Talktalk in City A but higher in City B.

In general though BT Wholesale has 1 of the lowest latencies over OpenReach.
Much of Talktalks backhaul network daisy chains from exchange to exchange, adding distance and latency.
…
There is likely less of a difference in latency between backhaul networks for anyone in London.

That was pretty much my experience moving from a BTW backed (Cerberus) to TTB - around a 1 to 2ms increase in baseline ping. Which for all intents makes no odds.

In reply to a post by E1234TJ:

Is it possible if you can ping these and share your results?
qosping-aws-eu-west-2.ol.epicgames.com (london)
qosping-aws-eu-central-1.ol.epicgames.com (frankfurt)

To lend some context, here's a quick summary based on some quick and dirty tests from three of my connections:

1. London AWS servers (in eu-west-2a)
- London EG AWS: 0 ms min/max/average (unsurprisingly!!)
- Frankfurt EG AWS: 14 ms min/average/max

2. London EAD connection
- London EG AWS: min/avg/max/stddev = 2.652/2.792/2.904/0.075 ms
- Frankfurt EG AWS: min/avg/max/stddev = 13.829/14.024/14.153/0.092 ms

3. Suffolk FTTP connection (TalkTalk Biz on OR)
- London EG AWS: 10 ms min/avg/max
- Frankfurt EG AWS: 21 ms min/avg and 22 ms max

On a 40/10 FTTC Utility Warehouse via Talk Talk.

Pinging qosping-aws-eu-west-2.ol.epicgames.com
Packets: Sent = 4, Received = 4, Lost = 0 (0% loss),
Approximate round trip times in milli-seconds:
Minimum = 17ms, Maximum = 19ms, Average = 18ms


Pinging qosping-aws-eu-central-1.ol.epicgames.com
Packets: Sent = 4, Received = 4, Lost = 0 (0% loss),
Approximate round trip times in milli-seconds:
Minimum = 29ms, Maximum = 31ms, Average = 30ms

I'm thinking about going with IDNet. Does anyone with experience with this ISP have any comments/anything I should be aware of? Also is it possible to have a dynamic IP address with IDNet or is static the only option?

Edited by E1234TJ (Sun 07-Nov-21 02:39:26)

There’s a whole sub forum for them:
https://forums.thinkbroadband.com/idnet.html

Maybe read the the most recent threads to get some views and/or post your own thread asking for experiences.

I'm in the process of leaving IDNet.

They advertise lowest latency in the UK but the reality is they use Zen wholesale, and the connection behaves just like a Zen connection would. I've had the same outages (and there has been a few, more than I've experienced since ADSL days), and the same congestion and variable latency issues as Zen customers have reported.

Latency varies with each connection as you can be randomly routed via various places and I've seen latency as high as 18ms and as low as 5ms, this is to the Gateway, so that is added to each and every destination, also depending on the route it can be more or less congested.

You can play gateway roulette and keep dropping the WAN and reconnecting and you can get lucky and get on a better route, but I think a lot of Zen customers play that game, so eventually everyone congregates onto the lowest latency route, causing congestion, so every few weeks in the early hours they drop PPP for a chunk of connections to try and load balance the network again, at least that is my assumption why I've not gone more than a few weeks before a disconnection overnight.

They offer good support but can't do anything about Zen backhaul, and you would have to ask them why they still use them.