Looks like different servers. Might have different numbers of people accessing them, or be on the network in a different way.

Do a traceroute to the IPs:
http-speedtest4.thinkbroadband.com = 80.249.103.8
and
ipv4.download.thinkbroadband.com = 80.249.99.130

for my Plusnet I get:

C:\>tracert 80.249.103.8

Tracing route to speedtest4.thinkbroadband.com [80.249.103.8]
over a maximum of 30 hops:

  1    <1 ms    <1 ms    <1 ms  <my router>
  2     9 ms     9 ms     9 ms  lo0.central10.pcn-bng01.plus.net [195.166.130.248]
  3    10 ms    17 ms    10 ms  411.be5.pcn-ir02.plus.net [84.93.253.71]
  4     9 ms     9 ms     9 ms  195.99.125.144
  5    10 ms    10 ms    10 ms  195.99.127.87
  6     *        *        *     Request timed out.
  7    10 ms    10 ms    10 ms  speedtest4.thinkbroadband.com [80.249.103.8]

Trace complete.

C:\>tracert 80.249.103.8

Tracing route to speedtest4.thinkbroadband.com [80.249.103.8]
over a maximum of 30 hops:

  1    <1 ms    <1 ms    <1 ms  <my router>
  2     9 ms     9 ms     9 ms  lo0.central10.pcn-bng01.plus.net [195.166.130.248]
  3    10 ms    17 ms    10 ms  411.be5.pcn-ir02.plus.net [84.93.253.71]
  4     9 ms     9 ms     9 ms  195.99.125.144
  5    10 ms    10 ms    10 ms  195.99.127.87
  6     *        *        *     Request timed out.
  7    10 ms    10 ms    10 ms  speedtest4.thinkbroadband.com [80.249.103.8]

Trace complete.

Got ya

In reply to a post by jchamier:

Looks like different servers. Might have different numbers of people accessing them, or be on the network in a different way.

Do a traceroute to the IPs:
http-speedtest4.thinkbroadband.com = 80.249.103.8
and
ipv4.download.thinkbroadband.com = 80.249.99.130

for my Plusnet I get:

C:\>tracert 80.249.103.8

Tracing route to speedtest4.thinkbroadband.com [80.249.103.8]
over a maximum of 30 hops:

  1    <1 ms    <1 ms    <1 ms  <my router>
  2     9 ms     9 ms     9 ms  lo0.central10.pcn-bng01.plus.net [195.166.130.248]
  3    10 ms    17 ms    10 ms  411.be5.pcn-ir02.plus.net [84.93.253.71]
  4     9 ms     9 ms     9 ms  195.99.125.144
  5    10 ms    10 ms    10 ms  195.99.127.87
  6     *        *        *     Request timed out.
  7    10 ms    10 ms    10 ms  speedtest4.thinkbroadband.com [80.249.103.8]

Trace complete.

C:\>tracert 80.249.103.8

Tracing route to speedtest4.thinkbroadband.com [80.249.103.8]
over a maximum of 30 hops:

  1    <1 ms    <1 ms    <1 ms  <my router>
  2     9 ms     9 ms     9 ms  lo0.central10.pcn-bng01.plus.net [195.166.130.248]
  3    10 ms    17 ms    10 ms  411.be5.pcn-ir02.plus.net [84.93.253.71]
  4     9 ms     9 ms     9 ms  195.99.125.144
  5    10 ms    10 ms    10 ms  195.99.127.87
  6     *        *        *     Request timed out.
  7    10 ms    10 ms    10 ms  speedtest4.thinkbroadband.com [80.249.103.8]

Trace complete.

Looks like you pasted the same traceroute twice but here is mine:

Tracing route to http-speedtest4.thinkbroadband.com [80.249.103.8]
over a maximum of 30 hops:

1 <1 ms <1 ms <1 ms RT-AC68U [192.168.1.254]
2 * * * Request timed out.
3 17 ms 11 ms 19 ms bmly-core-2a-xe-821-0.network.virginmedia.net [213.105.194.25]
4 14 ms 17 ms 16 ms bmly-core-2b-ae1-0.network.virginmedia.net [213.105.159.234]
5 * * * Request timed out.
6 15 ms 22 ms 14 ms eislou2-ic-2-ae0-0.network.virginmedia.net [62.254.84.62]
7 17 ms 18 ms 14 ms linx-gw2.thdo.ncuk.net [195.66.236.240]
8 15 ms 19 ms 16 ms te2-1-9.star10g.bdr-rt1.thn.ncuk.net [80.249.97.18]
9 15 ms 14 ms 13 ms speedtest4.thinkbroadband.com [80.249.103.8]

Trace complete.


Tracing route to ipv4.download1.thinkbroadband.com [80.249.99.148]
over a maximum of 30 hops:

1 <1 ms <1 ms <1 ms RT-AC68U [192.168.1.254]
2 * * * Request timed out.
3 12 ms 13 ms 13 ms bmly-core-2a-xe-821-0.network.virginmedia.net [213.105.194.25]
4 12 ms 12 ms 14 ms bmly-core-2b-ae1-0.network.virginmedia.net [213.105.159.234]
5 * * * Request timed out.
6 14 ms 22 ms 13 ms eislou2-ic-2-ae0-0.network.virginmedia.net [62.254.84.62]
7 12 ms 14 ms 17 ms linx-gw2.thdo.ncuk.net [195.66.236.240]
8 16 ms 15 ms 15 ms te2-1-9.star10g.bdr-rt1.thn.ncuk.net [80.249.97.18]
9 13 ms 17 ms 18 ms download.thinkbroadband.com [80.249.99.148]

Trace complete.

But since when did Virgin sell a 10 Mbps upload 50 Mbps package?

As for gap, yes a 5% difference will look wider to the eye as speeds increase

But as others on other providers can get over 100 Mbps it is not a limit per se, and also when we've had people testing where a limit was involved you got a flat line due to the limit and not a congested line

In reply to a post by MrSaffron:

But since when did Virgin sell a 10 Mbps upload 50 Mbps package?

As for gap, yes a 5% difference will look wider to the eye as speeds increase

I was talking about download speeds not upload (TBBx1 and HTTPx6 on the TBB speed tester).

Edit: I used the wrong test here is the correct one:
https://www.thinkbroadband.com/speedtest/14955570907...

Edited by deleted (Mon 29-May-17 16:29:25)

In reply to a post by thinksingle:

I was talking about download speeds not upload (TBBx1 and HTTPx6 on the TBB speed tester).

You're missing his point. The two meet, however the customer should, going by his upload, be receiving at least 150Mb down, not about 50.

The user in that test has an upload of > 10Mb. There's nothing below 150Mb on VM with that upload. That is not an example of a properly functional VM connection. It looks potentially like one that's being heavily limited by wireless performance.

It's pretty slow to ramp up, too. Wonder if that was done on a mobile phone?

In reply to a post by thinksingle:

I should have worded it better - I meant that if the single thread results that look like the one you just linked (if the person was on 50mb:

https://www.thinkbroadband.com/speedtest/14959816255...

it seems to look OK as the TBBx1 and HTTPx6 graph do meet, however in reality if they moved up to a higher package:

for example 100mb:
https://www.thinkbroadband.com/speedtest/14959820465...

and 152mb:
https://www.thinkbroadband.com/speedtest/14959808098...

Are these tests all being done on the same connection with rate limiting applied?

The 100/152Mb were done with a rate limit of about 3Mb/s on the upstream. The other test as I mentioned looks heavily limited by wireless performance, potentially with a sub-optimal TCP stack too.

Thats the issue with looking at a database of results, not knowing the OS, and hardware used.

This information can be grabbed somewhat via browser headers tho. Might be an idea for TBB to consider that.

In reply to a post by Ignitionnet:

In reply to a post by thinksingle:

I was talking about download speeds not upload (TBBx1 and HTTPx6 on the TBB speed tester).

You're missing his point. The two meet, however the customer should, going by his upload, be receiving at least 150Mb down, not about 50.

The user in that test has an upload of > 10Mb. There's nothing below 150Mb on VM with that upload. That is not an example of a properly functional VM connection. It looks potentially like one that's being heavily limited by wireless performance.

It's pretty slow to ramp up, too. Wonder if that was done on a mobile phone?

Apologies that was my error I used the wrong example for 50mb it should've been this one:

https://www.thinkbroadband.com/speedtest/14955570907...