FTTP/FTTH Vodafone Gigafast High Ping

Hey all,

We just have had our new full fibre (FTTP/FTTH – I will call it FTTP) connection installed on 22nd June 2020. The ISP is Vodafone Gigafast and we went for 900Mbps package.

We live in Milton Keynes.

We are a BT customer for over a decade and never had an issue with them. We are still with BT and the package is Infinity 2 (FTTC) so we still have not cancelled with BT just yet as we are still within 14 days cooling period with Vodafone. So, if we are not happy with Vodafone then we will cancel with them and stay with BT for a little longer until their BT FTTP becomes available in my area, or even Hyperoptic.

The main reason why we wanted to switch to FTTP from FTTC are because of faster upload speed and lower latency ping. I am already aware that there will not be a huge difference in ping latency when switching from FTTC to FTTP.

With our BT Infinity 2, we always get around 74Mbps download and around 15Mbps upload speed with consistent 9ms ping latency via speedtest.net by Ookla with the closest server selected (Clouvider Limited – Enfield) over wired and wireless connection (2.4GHz and 5GHz). Also, we have been using third-party router (currently ASUS RT-AC86U) connected to BT Openreach modem (Huawei HG612). On Rocket League, average ping is 16ms and on Rainbow Six Siege is around 9ms. Ping to bbc.co.uk is always 9ms.

With our Vodafone Gigafast 900Mbps, we get around 750Mbps download and 920Mbps upload speeds with the lowest ping of 15ms over wired connection to my high-end PC via speedtest.net with closest server selected (Vodafone UK - Manchester). This is not what I was expecting (ping latency) as I was expecting around 3-5ms ping latency. We have also tested on Vodafone’s supplied router and ASUS router, both shows same result. Rocket League average ping is 24ms and around 20ms on Rainbow Six Siege. Ping to bbc.co.uk is 14ms on average.

The weird thing is that the external IP address City/Region with Vodafone is showing Manchester whereas BT is showing Milton Keynes. I live in Milton Keynes and Manchester is 130 miles away from me. Whenever I go onto speedtest.net Ookla, it shows that the closest server is Vodafone UK - Manchester – 2 miles, but I’m in Milton Keynes? Whereas on BT, it shows that the closest server is Clouvider Limited – Enfield – 40 miles which is normal.

My friend who is also with Vodafone Gigafast for about a year and also lives in Milton Keynes (10 minutes drive from my house), he gets around 5ms ping and his external IP address City/Region shows Milton Keynes via speedtest.net Ookla with the closest server selected as Vodafone UK – Watford – 32 miles, which is normal.

I think Vodafone’s routing system is messed up and that they incorrectly assigned my IP address really far away from where I live which causes high pings. Is Vodafone able to change my IP address to Milton Keynes area? What can I do?

The CityFibre cabinet is only 80 metres away from my house so that is not the problem.

I have reset ONT as well as Vodafone Router and ASUS router – no change.

I have contacted Vodafone about this, and they told me that their dedicated tech team will be monitoring my line for the next 72 hours. I have requested to change my IP address to Milton Keynes as well, but still need to wait 3 days.

Any ideas?

So in regards to your IP address saying Manchester over Milton Keynes, this is irrelevant. IP's very often will have the wrong geolocation on them as this data is populated by 3rd party companies, this doesn't mean your internet is routing all the way up to Manchester rather than Milton Keynes.

The latency from FTTC to FTTP is generally 5-8ms of a difference, however this can be based on numerous different things. Can you please run a traceroute using command prompt to bbc.co.uk and then post a result? This will let us see if your internet is routing down to London properly or if it's going somewhere else, in which case it may be a proper fault.

Speedtest.net determining your location is also somehting that can be ignored, as it does it off your geolocation which by nature can be wrong. (Example being I'm in Scotland, yet I'm apparently in London according to Zen).

In reference to imgur.com/fnJPjF0:

tracert bbc.co.uk

So it looks like they actually are routing you up to Manchester, unfortunately I couldn't confirm if this is normal activity or not unless your friend can also share a traceroute with us. It's all based on where the VM core routers are, as your traffic needs to route through them before it goes anywhere else.

I will ask my friend to do the tracert as he is busy at the moment.

Thank you though.

In reply to a post by OfficialUmey:

I have reset ONT as well as Vodafone Router and ASUS router – no change.

How have you connected the vodafone router and the asus router?

In reply to a post by max1486:

The latency from FTTC to FTTP is generally 5-8ms of a difference, however this can be based on numerous different things.

It's usually much lower of a difference.

In examples I've seen (with the same routing) it's around 2-3ms difference at most.

There being a 5-8ms difference would suggest FTTC is adding that much latency.
There are FTTC connections with 5ms pings to bbc.co.uk so that's impossible.

Some ISP's will have different routing for FTTP and FTTC traffic which can make the difference between the 2 more emphasised.
Nobody should be anticipating a 5-8ms improvement in latency when switching from FTTC to FTTP though or they are likely to be very disappointed.

Of course as I said it depends on the conditions, my personal experience was a 5ms difference with the exact same ISP with no routing difference whatsoever. (BT, 23ms to 18ms, FTTC cable distance was super far, only ever got 20Mbps tops on FTTC).

Edited by deleted (Fri 26-Jun-20 22:50:04)

Currently, my ASUS router is connected directly to the ONT via Cat6 ethernet cable.

What I meant about resetting both ASUS router and Vodafone router, is that I have tested on both of them. Both still show the same results.

Test closest to where you are physically, who cares where the IP says as long as its in the same country.

Well....my gigafast IP thinks i'm part of vodafone india (while being in aberdeen) so at least you don't have that headache.

Edited by epyon (Sat 27-Jun-20 15:38:57)

FTTC sync speed is not a reliable indicator

e.g. 20 Mbps sync here and pinging our servers over an IDNet connection at 9ms

8 to 8.5ms to the first hop.

The real variable for FTTC is the depth of interleaving applied (i.e. error correction).

Not expecting any decrease when FTTP rolls around, maybe 1ms.

I've tested the closest one to where I physically live.

Closest server is Vodafone UK - Watford and still showing 15ms.

If I manually change the server to Vodafone UK - Manchester, I get 8-9ms ping.

I'm in Milton Keynes.


I've been comparing BT and Vodafone connections on Rocket League.

BT connected to ASUS RT-AC68U. Wired to PS4 only.

Vodafone connected to ASUS RT-AC86U. Wired to PC only.

Rocket League supports cross-platform, so I created a private match on PC and invited myself to my PS4. One of the game shows that PS4 (BT) gets around 12ms whilst PC (Vodafone) gets around 20ms. They're on the same game server as well.



What's the average ping you're getting at the moment?

The difference in pings between both providers is negligable and even then online gaming has balancing measures in place so those with the lowest ping are not at a competitive advantage to those with higher pings. Vodafone routes much of its traffic through Manchester because they have a large office there (former C&W building).

I'm getting FTTP installed next month, I'm not expecting a massive change in latency, if any at all .

They are routing your line via Manchester.


So for the majority of connections you make are going from Milton Keynes to Manchester, then to London (then onwards to the test of the world if not hosted in London.

Your friend with the lower ping is being routed direct to London.

So you have the added latency of all your traffic going up to Manchester and back down to London.
There's isn't a whole lot hosted in Manchester.

Post the results of a tracert to a couple addresses, it might show in the 1st couple hops where the traffic is going.

Get your friend to do the same if possible.

Ideally Vodafone get the routing changed as going via Manchester makes no sense.

You're totally right. I knew it.

Why would Vodafone route my line to Manchester though? Is this a rare case or?

My friend will do the trace route tomorrow afternoon, so I will share it with you all on here. I'm pretty sure his line is routed to London because his pings on online games are identical to my BT. My BT line is definitely routed to London.

Also, I will give a call to Vodafone Technical Team again and ask to be transferred to someone who knows what I am talking about (hopefully).


As requested, here are the results of tracert (CoD Warzone on PS4). I've tested on 2 different UK servers on the same game. I've also used BT connection on 2nd ASUS router for comparison, then did the tracert on 2nd PC connected to BT. All on wired connection only.


Warzone Server 1 on Vodafone: https://i.imgur.com/rOSJ3Vk.png

Warzone Server 1 on BT: https://i.imgur.com/jhO7QgH.png

Warzone Server 2 on Vodafone: https://i.imgur.com/OfBJPel.png

Warzone Server 2 on BT: https://i.imgur.com/hiebcFT.png

For those wondering how I got the IP addresses of Warzone's servers, I use a third-party software from PC to 'sniff' any IP addresses including game servers on PS4.

I wouldn't say it's rare, just unfortunate. I'd say it's incredibly unlikely the people on the phone will be able to help unless they can raise directly to their network admin team. I wish you luck though!

Hi all again,

Some of you have asked me to share the results of my friend's tracert. We have used the exact same IP addresses. It appears that his (lets call him Brad) line is being routed to Slough which is only 20 miles away from London whereas my line is being routed to Manchester. This explains why he gets much lower pings on online video games whereas mine is almost doubled.

We both live in Milton Keynes and we're 10 minutes drive away from each other.


The results:

Brad's tracert 151.101.0.81 (bbc.co.uk): https://i.imgur.com/3FLCKOs.jpg

My tracert 151.101.0.81 (bbc.co.uk): https://i.imgur.com/aQUJ6Gf.jpg

-

Brad's tracert 8.8.8.8 (dns.google): https://i.imgur.com/q48gyXl.jpg

My tracert 8.8.8.8 (dns.google): https://i.imgur.com/DJZigPh.jpg

-

Brad's tracert 78.129.201.32 (CoD Warzone): https://i.imgur.com/kj1z9zh.jpg

My tracert 78.129.201.32 (CoD Warzone): https://i.imgur.com/yivoPbL.jpg


Contacted Vodafone today to be told that their 3rd line support team is only open on Monday - Friday. They gave me a reference number so I can call directly to 3rd line team without having to explain everything again tomorrow Monday.

Also, I want to say thank you all for your posts

It is odd that you seem to be being routed via Manchester. But BT customers are routed via London, wherever they live. So those in the north like me don’t get good pings.

Did you ever get this sorted out? My Gigafast line (also in MK) has started routing up to Manchester as well -

% traceroute -I -a fastmail.com 
traceroute: Warning: fastmail.com has multiple addresses; using 66.111.4.147
traceroute to fastmail.com (66.111.4.147), 64 hops max, 72 byte packets
 1  [AS0] vodafone (192.168.1.254)  3.700 ms  2.587 ms  2.017 ms
 2  [AS5378] 84.65.128.1 (84.65.128.1)  11.727 ms  10.573 ms  11.551 ms
 3  [AS1273] 63.130.104.202 (63.130.104.202)  11.841 ms  11.807 ms  11.961 ms
 4  [AS1273] ae5-100-xcr1.man.cw.net (195.89.96.113)  22.933 ms  10.926 ms  11.486 ms
 5  [AS1273] ae1-ucr1.bsh.cw.net (195.2.27.193)  17.791 ms  18.685 ms  18.373 ms
 6  [AS1273] ae20-xcr1.lnd.cw.net (195.2.27.190)  16.798 ms  16.207 ms  17.633 ms
 7  [AS1273] ae3-xcr2.lsw.cw.net (195.2.28.182)  17.582 ms  18.077 ms  17.583 ms
 8  [AS1273] so-4-0-0-zar1.fri.cw.net (195.2.2.118)  17.634 ms  16.654 ms  17.687 ms
 9  [AS0] ae11.mpr1.lhr1.uk.zip.zayo.com (64.125.27.49)  18.144 ms  18.317 ms  17.492 ms
10  [AS0] ae10.mpr2.lhr2.uk.zip.zayo.com (64.125.31.194)  17.845 ms  18.358 ms  17.658 ms
11  * * *
12  * * *
13  [AS0] ae20.mpr2.ewr1.us.zip.zayo.com (64.125.26.143)  93.567 ms  86.924 ms  84.169 ms
14  [AS0] ae3.mpr2.ewr1.us.zip.zayo.com (64.125.31.238)  84.921 ms  84.681 ms  86.120 ms
15  [AS0] 208.184.34.238.ipyx-076763-900-zyo.zip.zayo.com (208.184.34.238)  86.148 ms  88.958 ms  86.647 ms
16  [AS11403] 199.83.60.138.static.nyinternet.net (199.83.60.138)  88.435 ms  85.663 ms  86.502 ms
17  [AS11403] www.fastmail.com (66.111.4.147)  84.557 ms  85.157 ms  85.272 ms

A trace route to 1.1.1.1 appears to show the routing going up to Manchester, down to London and then across to France instead of just going straight to London and then hitting Cloudflare's London datacenter!

% traceroute -I -a 1.1.1.1     
traceroute to 1.1.1.1 (1.1.1.1), 64 hops max, 72 byte packets
 1  [AS0] vodafone (192.168.1.254)  3.258 ms  2.183 ms  2.336 ms
 2  [AS5378] 84.65.128.1 (84.65.128.1)  12.092 ms  10.716 ms  11.641 ms
 3  [AS1273] 63.130.104.202 (63.130.104.202)  12.318 ms  12.144 ms  12.438 ms
 4  [AS1273] ae5-100-xcr1.man.cw.net (195.89.96.113)  13.287 ms  10.484 ms  14.396 ms
 5  [AS1273] ae0-ucr1.mcr.cw.net (195.2.28.70)  12.429 ms  13.092 ms  11.819 ms
 6  [AS1273] ae16-xcr1.lnt.cw.net (195.2.27.165)  17.439 ms  19.540 ms  18.495 ms
 7  [AS2914] ae-20.r02.londen03.uk.bb.gin.ntt.net (129.250.66.45)  18.703 ms  17.683 ms  18.677 ms
 8  [AS2914] ae-11.r21.londen12.uk.bb.gin.ntt.net (129.250.4.85)  19.603 ms  17.035 ms  18.772 ms
 9  [AS2914] ae-6.r21.parsfr04.fr.bb.gin.ntt.net (129.250.4.25)  28.494 ms  28.426 ms  28.255 ms
10  [AS2914] ae-5.r03.parsfr02.fr.bb.gin.ntt.net (129.250.2.207)  26.447 ms  26.756 ms  26.416 ms
11  [AS2914] 185.84.18.174 (185.84.18.174)  27.010 ms  26.663 ms  42.625 ms
12  [AS198949] one.one.one.one (1.1.1.1)  26.285 ms  26.953 ms  26.583 ms


Cant be sure when this started happening but can't be more than a few weeks. I tried restarting the ONT and router but no change...

Edited by Adrianuk (Sun 22-Nov-20 15:57:57)

In reply to a post by Adrianuk:

A trace route to 1.1.1.1 appears to show the routing going up to Manchester, down to London and then across to France instead of just going straight to London and then hitting Cloudflare's London datacenter!

I'd be very careful about assuming that 1.1.1.1 is supposed to be routed through London. Usually that's the case but sometimes Cloudflare re-routes entire data centres (due to outages or for maintenance). The whole of Cloudflare's UK data centres were re-routed through Amsterdam the other day and traceroutes to 1.1.1.1 on BT are still showing up as routed via Amsterdam for me even though Cloudflare's UK data centres are now operational again.

traceroute to 1.1.1.1 (1.1.1.1), 30 hops max, 60 byte packets
1 OpenWrt.lan (192.168.0.254) 0.219 ms 0.227 ms 0.244 ms
2 172.16.16.87 (172.16.16.87) 1.543 ms 1.552 ms 1.578 ms
3 * * *
4 31.55.186.184 (31.55.186.184) 6.461 ms 6.486 ms 6.498 ms
5 core4-hu0-8-0-3.faraday.ukcore.bt.net (213.121.192.68) 6.467 ms 6.913 ms 6.973 ms
6 62.6.201.144 (62.6.201.144) 6.981 ms 6.391 ms 6.395 ms
7 166-49-209-132.gia.bt.net (166.49.209.132) 6.046 ms 6.054 ms 6.140 ms
8 t2c3-et-7-3-0.nl-ams2.gia.bt.net (166.49.195.174) 11.155 ms 11.168 ms 11.200 ms
9 * * *
10 one.one.one.one (1.1.1.1) 12.425 ms 12.370 ms 13.106 ms

Ah, thanks for that, I'll keep an eye on it over the next few days - though that only accounts for the hop from London to France, I'm still generally detouring up to Manchester for some reason.

Edited by Adrianuk (Sun 22-Nov-20 17:08:07)

Post deleted by bluebull

In reply to a post by bluebull:

I want to bring this back up to the top because all of us on City Fibre lines actually have a bigger problem.

Look at the 2nd hop on your traceroutes. Notice how it's around 10ms (it should be much less), that is the latency that is added before it's handed over from Cityfibre to Vodafone. That might not seem like a high number on it's own, but when you add that onto where you're actually pinging it is.

If you compare your gigafast line to a standard vodafone ADSL line you will always see an extra 10ms on your pings.

The Vodafone network itself is actually very good with impressive latency across the board. It would definitely have lower latency than BT if it didn't have 10ms added on from City Fibre.

I've spoke with an engineer at City Fibre and even he agreed there is clearly an issue, the problem is they can't work on it until Vodafone 2nd line support issue City Fibre with a ticket.

Please people, encourage everyone to get this escalated. This will have a huge positive impact on our connections if we can get this looked at.

That is not true. CityFibre hand off to Vodafone and others at the equivalent place to Openreach, at the local POP.

Your characterisation of the Vodafone network is incorrect, it doesn't have impressive latency across the board, they're still in the process of upgrading it from the network that formerly ran Bulldog ADSL.

The latency is traffic getting from the Vodafone device on the interconnect to CityFibre to wherever Vodafone terminate the PPP session.

I appreciate you're misinformed, however if there were issues so heavy with the CityFibre leg they'd be impacting every CityFibre customer, and I would imagine one of the other CityFibre customers would've noticed by now, given a bunch of them are selling business-grade services.

There are a bunch of devices on the network you can't see on a traceroute. The CityFibre part is the OLT terminating the FTTP service and a switch, then you're onto the Vodafone network, and there are is a lot more to go wrong there before it arrives at a device that'll respond to a traceroute.

As the previous poster pointed out. I was incorrect (and have deleted the post to avoid confusion). However, I have a further issue I'd definitely like to mention.

Look at the 2nd hop on your traceroutes. Notice how it's around 10ms (it should be much less at this stage), that is the latency that is added before it's handed over from Cityfibre to Vodafone. That might not seem like a high number on it's own, but when you add that onto where you're actually pinging it is.

If you compare your gigafast (city fibre) line to a standard vodafone VDSL line you will always see an extra 10ms on your pings. Ping any server and this can easily be proven. Vodafone host their own speedtest.net server so the pings should be 2 - 3 ms but because of the CityFibre issue they are always > 10ms

I've spoken with an engineer at City Fibre and even he agreed there is clearly an issue, the problem is they can't work on it until Vodafone 2nd line support issue City Fibre with a ticket.

As the issue is within City Fibre's network (before it gets handed off to the Vodafone gateway), there's nothing Vodafone can really do apart from send support tickets to City Fibre.

Please people, encourage everyone to get this escalated. This isn't going to fix the routed by Manchester issue but it will fix the additional 10ms problem, which by looking at your trace routes you all seem to have.

I'm sure most people are fine with an additional 10ms, but I personally do a job that requires less than 13ms to the server and I'm sure the gamers here will agree 10ms is a very welcome reduction. On top of this, you upgrade from DSL to FTTH and expect to see lower pings, not higher. Further to this, City Fibre has a whole mission statement on how reducing latency is one of their core objectives of their product, so it's not unreasonable to request this be fixed. In 2021, it's not unreasonable to demand hops from Milton Keynes to Milton Keynes should take a lot less than 10ms, especially when Vodafone seem capable of getting us from MK to telehouse in less than 3ms.

Edited by bluebull (Sat 10-Jul-21 14:33:58)

In reply to a post by bluebull:

Look at the 2nd hop on your traceroutes. Notice how it's around 10ms (it should be much less at this stage), that is the latency that is added before it's handed over from Cityfibre to Vodafone.

You can't determine that from a traceroute.

The Cityfibre to Vodafone interconnect is entirely at layer 2 - it won't appear in any traceroute. The first IP address you hit in traceroute is Vodafone's BRAS at 84.65.128.1

https://bgp.he.net/ip/84.65.128.1

It will have been over some backhaul network. Whether Cityfibre provide that as a managed service to Vodafone, or Vodafone run their own backhaul network, is something we can't see.

It's quite possible that that has taken some sort of dog-leg path, and there's a better path that could have been chosen. But none of this means for certain that it's a problem with Cityfibre.

In reply to a post by candlerb:

It will have been over some backhaul network. Whether Cityfibre provide that as a managed service to Vodafone, or Vodafone run their own backhaul network, is something we can't see.

CityFibre hand off in the same way Openreach do - the local POP. This is why when you see ISPs come online with CityFibre it's on a city by city basis.

In reply to a post by CarlTSpeak:

CityFibre hand off in the same way Openreach do - the local POP. This is why when you see ISPs come online with CityFibre it's on a city by city basis.

Yes, but Cityfibre also have a national fibre network, which they bought from KCOM. They *could* be supplying service to Vodafone over this.

In reply to a post by candlerb:

Yes, but Cityfibre also have a national fibre network, which they bought from KCOM. They *could* be supplying service to Vodafone over this.

Nah - https://www.cityfibre.com/wp-content/uploads/2018/01...

How can you have both FTTC and FTTP simultaneously? Don't they remove the Copper line once the Fibre has been installed?

In reply to a post by bluebull:

I personally do a job that requires less than 13ms to the server

Are you in a position to hint at what this is, because that's a very specific requirement and would rule out people from doing that job from e.g. Newcastle, possibly Cornwall would not be an option either.

Quite possible to have both services, the copper line is not removed when FTTP is installed but may be at some unknown time in the future.

In reply to a post by LucBorg:

How can you have both FTTC and FTTP simultaneously? Don't they remove the Copper line once the Fibre has been installed?

In general, no. The FTTP service is separate from the FTTC one and there's no need to cease the FTTC to take FTTP.

For underground feeds, they just pull the fibre in alongside the copper. For aerial feeds, they can replace the existing copper cable with a "hybrid" cable that has both fibre and a copper pair.

Of course, there are some new-builds which are fibre only. And going forward, as FTTP is rolled out there will be an increasing number of places with a "stop sell" on copper services where fibre is available - initially no new provides, then no migrations back from FTTP to FTTC. But forcibly removing copper services is still a way down the road.

Separately from this is the switch-off of the PSTN, scheduled for 2025. By then the baseband analogue voice signal *will* be forcibly removed from everyone, but will be replaced by voice-over-IP (which will be over xDSL if you don't have a fibre service).

In reply to a post by jpm:

Are you in a position to hint at what this is, because that's a very specific requirement and would rule out people from doing that job from e.g. Newcastle, possibly Cornwall would not be an option either.

Likely professional gaming on a game with something with potentially an 80 tick rate - update every 12.5 ms.

Obvious solution to that requirement is to move to Docklands and only use servers hosted there.

I'm a futures trader.

Sorry for the late reply.

The high latency is caused by Vodafone routing customers via the Edinburgh BNG. It's done automatically as part of their load balancing.

I've spent many hours on the phone and the problem can't be fixed by their support team.

There are rumours of network upgrades but nothing has been confirmed.

Needing those sort of tight latency guarantees are what leased line SLAs are for. They don't even have to be internet circuits - you could get layer 2 straight back to your office if you wanted.

In reply to a post by max1486:

So in regards to your IP address saying Manchester over Milton Keynes, this is irrelevant. IP's very often will have the wrong geolocation on them as this data is populated by 3rd party companies, this doesn't mean your internet is routing all the way up to Manchester rather than Milton Keynes.

That ^^

The "location" data is very vague and often wrong, especially in the UK.

It has absolutely nothing to do with the routing you take as the previous poster said.

In reply to a post by MrSaffron:

FTTC sync speed is not a reliable indicator

e.g. 20 Mbps sync here and pinging our servers over an IDNet connection at 9ms

8 to 8.5ms to the first hop.

The real variable for FTTC is the depth of interleaving applied (i.e. error correction).

Not expecting any decrease when FTTP rolls around, maybe 1ms.

Quote - as an example:

idnet FTTC:
Reply from 8.8.8.8: bytes=32 time=13ms TTL=121
Reply from 8.8.8.8: bytes=32 time=14ms TTL=121
Reply from 8.8.8.8: bytes=32 time=13ms TTL=121
Reply from 8.8.8.8: bytes=32 time=13ms TTL=121

idnet FTTP:
64 bytes from 8.8.8.8: icmp_seq=0 ttl=121 time=13.934 ms
64 bytes from 8.8.8.8: icmp_seq=1 ttl=121 time=15.103 ms
64 bytes from 8.8.8.8: icmp_seq=2 ttl=121 time=14.514 ms
64 bytes from 8.8.8.8: icmp_seq=3 ttl=121 time=15.524 ms

Same geographic location, same underlying conditions/routers etc.

I've done a lot of digging since my eariler posts. The issue is actually caused by Vodafone using Path finder to balance their network capacity.

The way this is achieved is by changing the customers gateway to a less congested location. In the case of Vodafone, this often tends to be Edinburgh. If you are located down South, this means your traffic has to go all the way to Edinburgh and then all the way back down to their London data center. I'm in Bucks and this round trip ads around 20ms to my ping times.

There is a huge thread on the Vodafone forum. Unfortunately, it's basically just a handful of users handing out false information that a fix is coming. It isn't, the routing we now see is to ensure customers get their advertised download speeds.

For those interested, here is a trace route

Tracing route to www.google.com [142.250.200.4]
over a maximum of 30 hops:

1 1 ms <1 ms <1 ms RT-AX92U-6090 [192.168.50.1]
2 14 ms 13 ms 13 ms 90.247.128.1
3 24 ms 24 ms 24 ms 63.130.172.45
4 25 ms 27 ms 27 ms 72.14.216.236
5 27 ms 28 ms 26 ms 216.239.42.41
6 28 ms 33 ms 26 ms 108.170.234.231
7 24 ms 23 ms 25 ms lhr48s29-in-f4.1e100.net [142.250.200.4]

Let's just hope other ISP's don't adopt this model.

Edited by bluebull (Mon 06-Mar-23 13:30:53)

In reply to a post by bluebull:

Let's just hope other ISP's don't adopt this model.

AIUI, Zen beat them to it.

That sucks. I just have this horrible vision of the future. ISP's trying to sell us "gaming broadband", obviously more expensive than regular broadband.

Edited by bluebull (Mon 06-Mar-23 11:54:12)

Cheaper to redirect than increase capacity, I suppose from an ISP point of view, if they have a under utilised link and one thats full, they might see it as wasted money to upgrade the fully loaded link in preference to sharing the load with the lightly loaded link, so depends in the ISP values their balance sheet or service quality more.

This sort of stuff likely happens commonly on transit providers, but when its extreme like going to the opposite part of the country first is when consumers start to notice.

In reply to a post by bluebull:

That sucks. I just have this horrible vision of the future. ISP's trying to sell us "gaming broadband", obviously more expensive than regular broadband.

There are already ISPs that market themselves that way.

If you choose any "small" ISP, which typically only has one POP (or maybe two) in London, then the issue goes away. They'll rely on BT or Talktalk Wholesale to bring the traffic to their POP, and that routing is fixed.

If you buy from a large-scale ISP with rock-bottom pricing, then you get what you get.

In reply to a post by candlerb:

In reply to a post by bluebull:

That sucks. I just have this horrible vision of the future. ISP's trying to sell us "gaming broadband", obviously more expensive than regular broadband.

There are already ISPs that market themselves that way.

If you choose any "small" ISP, which typically only has one POP (or maybe two) in London, then the issue goes away. They'll rely on BT or Talktalk Wholesale to bring the traffic to their POP, and that routing is fixed.

If you buy from a large-scale ISP with rock-bottom pricing, then you get what you get.

What about the ISPs like IDNet who are migrating customers from BTW backhaul to Zen. This is presumably a measure driven by finance. They are definitely not in the rock-bottom pricing group but there is evidence if you search t'internet that some of their customers have been adversely affected by the change. Given the awareness of the average consumer in regard to technology matters the people posting on t'internet are probably part of a small subset of the total number of customers affected.

In reply to a post by GonePostal:

In reply to a post by candlerb:

In reply to a post by bluebull:

That sucks. I just have this horrible vision of the future. ISP's trying to sell us "gaming broadband", obviously more expensive than regular broadband.

There are already ISPs that market themselves that way.

If you choose any "small" ISP, which typically only has one POP (or maybe two) in London, then the issue goes away. They'll rely on BT or Talktalk Wholesale to bring the traffic to their POP, and that routing is fixed.

If you buy from a large-scale ISP with rock-bottom pricing, then you get what you get.

What about the ISPs like IDNet who are migrating customers from BTW backhaul to Zen. This is presumably a measure driven by finance. They are definitely not in the rock-bottom pricing group but there is evidence if you search t'internet that some of their customers have been adversely affected by the change. Given the awareness of the average consumer in regard to technology matters the people posting on t'internet are probably part of a small subset of the total number of customers affected.

Imagine that you have left, for example, Zen because of its "spotty" backhaul, and gone to an ISP which did not use Zen's backhaul only to be moved back onto it.