I’m constantly frustrated by my broadband speeds. They are not bad, but not as good as they should be either, especially not when marketed as “superfast broadband”. And the real-world speeds seemed to drop about 10% when I switched ISPs from PlusNet to Vodafone a couple of years ago. I think I’m out of contract now, so it’s a good opportunity to take a look at what is possible.
For reference, I’m on a Vodafone Superfast 1 business broadband package, which should give up to 38Mbps, with a claimed average of 35Mbps and 73% of customers able to achieve those speeds:
In the real world, I get around 22Mbps down and 7Mbps up, according to various speed checkers. The most I’ve ever seen on a speed test is around 25-26Mbps, with my previous ISP.
Step 1: Check basic broadband availability.
The Kitz Broadband Checker helps here, using BT Wholesale and SamKnows data to check what is available for a given phone number/postcode. There are basic details of the local telephone exchange and there is a rough indicator of how far away it is. It may be 618m as the crow flies, but I can tell you it’s more like 1500m by road/cable (assuming it follows the route I would expect around the local streets and doesn’t run along the alleyway behind my house). I don’t understand all of the acronyms and abbreviations but it seems to me that the site doesn’t (yet) understand whether Fibre To The Premises (FTTP) is available, or just Fibre To The Cabinet (FTTC) and earlier broadband options.
Step 2: A few more details from BT Wholesale
The BT Wholesale Broadband Checker was my next port of call. This tells me all sorts of things about my line, like that: I should be able to get around 42.5Mbps down (instead of my 22) – which is probably where the real-world claimed up to 38Mbps comes from; that FTTP is available (along with several other products); and that my current FTTC cabinet is Cabinet 10. I’ve spotted this cabinet on my walks around town and so I know it’s not particularly close to home – probably around 1000m by road, though it could be as little as 350m if the cable runs through the alleyway I mentioned earlier.
Step 3: Check the modem stats
I stopped using my ISP-provided modem a couple of years ago and switched to a DrayTek Vigor 130 modem (since discontinued by the manufacturer) with a Ubiquiti AmpliFi router.
Logging on to the router’s web interface told me that I was syncing at around 27Mbps down and 7 up and on VDSL Profile 17A with a signal to noise ratio (SNR) around 6:
That fits with my real-world performance in speed tests of around 22Mbps. Digging a little deeper via a telnet session gave me a whole load of stats:
> adsl status
---------------------- ATU-R Info (hw: annex A, f/w: annex A/B/C) -----------
Running Mode : 17A State : SHOWTIME
DS Actual Rate : 27400000 bps US Actual Rate : 7265000 bps
DS Attainable Rate : 28950732 bps US Attainable Rate : 7265000 bps
DS Path Mode : Fast US Path Mode : Fast
DS Interleave Depth : 1 US Interleave Depth : 1
NE Current Attenuation : 25 dB Cur SNR Margin : 7 dB
DS actual PSD : 4. 3 dB US actual PSD : 11. 7 dB
NE CRC Count : 935 FE CRC Count : 13805
NE ES Count : 261 FE ES Count : 7519
Xdsl Reset Times : 0 Xdsl Link Times : 7
ITU Version[0] : fe004452 ITU Version[1] : 41590000
VDSL Firmware Version : 05-07-06-0D-01-07 [with Vectoring support]
Power Management Mode : DSL_G997_PMS_L0
Test Mode : DISABLE
-------------------------------- ATU-C Info ---------------------------------
Far Current Attenuation : 28 dB Far SNR Margin : 6 dB
CO ITU Version[0] : b5004244 CO ITU Version[1] : 434da4a1
DSLAM CHIPSET VENDOR : < BDCM >Comparing that sync speed of 27.4Mpbs with the BT Wholesale test in step 2, and my connection is running as quickly as it ever has (though I’m not sure what period the BT Wholesale test ran over for its maximum observed speed).
(It also tells me that my cabinet has a Broadcom chipset, so it’s most likely to contain Huawei equipment).
So, why is my “Superfast broadband” not so… superfast?
So, I have lots of metrics, what’s the analysis?
- Possibly: my connection. Several years ago (pre-broadband) we had a second phone line put in the house (since disconnected) and I seem to recall that some jiggery-pokery was required at the exchange to accommodate that, possibly even running two phone numbers over one copper connection. I thought that might explain why I get about half the theoretical maximum bandwidth (and my neighbour three doors down gets the whole lot).
- Possibly: internal wiring on-premises. My modem is connected directly to a line into the house but it’s not the BT master socket – it’s connected by some other external route that I don’t understand. I’ve tried moving previous modem/routers to connect directly to the master socket and it’s not made any noticeable difference to sync speeds.
- Most likely: physics. Whilst researching this post, I found information about FTTC speed vs. distance from the cabinet (repeated on various forums). At 1km from the cabinet, the most I will get, even on a 17A (80Mbps) profile, is 28Mpbs and I’m syncing at around 27 (this table at ThinkBroadband suggests even lower). So what about my neighbour who gets 38? Maybe he’s just lucky, or perhaps his line takes a different route around town…
Assuming my analysis is correct, this is probably about as good as it gets, without an FTTP upgrade. And, as the service is cheap (around £28/month including a home phone calling package and with no line rental), we might just stay put for now – after all my Teams calls work in the day and my Netflix and YouTube work in the evening/at the weekend!
SamKnows?
As a little addendum (and there is nothing in this for me), if you’re trying to work out what’s going on with your broadband, I’d recommend checking out SamKnows. I have one of their “white boxes” on my network and have had for several years (ever since the late Jack Schofield pointed me in their direction). In exchange for some real-world performance monitoring, which is aggregated to assess ISPs, I get reliable stats on the state of my connection.
