Does anybody actually get the maximum FTTC download

I've just moved house, and I'm very close to the FTTC cabinet, syncing at close to 78 Mb/s
However, real speedtests etc, and even attempting ftp transfers as tests never get above 60 Mb/s. Upstream I'm syncing at 22 Mb/s, and I get 18-19 Mb/s real world values (As I'd expect)

I was in a pub the other day, not in the same area as me, but its FTTC cabinet is outside.
Doing casual speedtests on my phone sat next to the router, it's the same experience as at home, 58 to 60 max

Are me and pub unlucky, or is this par for the course (because the cabinets only have a single (GB ?) connection ?

I am 181m from my cabinet, get full 80/20 sync (you can�t have 22 up ?) and hard wired to decent device, I get 76 throughput.

So yes, it looks like it�s you or your provider.

Are you using a wired connection or wireless for the speedtests? If wireless then it could be the bottleneck.

Yes, we see people getting full speed

Using Wi-Fi does add extra potential for that to reduce the speeds e.g. picking the 2.4GHz band rather than faster 5 GHz band

http://tbb.st/1566829894158373255 74.4 Mbps down, and on a phone too.

The Gigabit split between users is plenty of capacity.

In reply to a post by Zarjaz:

I am 181m from my cabinet, get full 80/20 sync (you can�t have 22 up ?) and hard wired to decent device, I get 76 throughput.

So yes, it looks like it�s you or your provider.

I'm measuring using a wired LAN connection, and on more than one device, never more than 60.

Oh well, (60 is more than enough for my needs, and some folk are lucky to get a fraction of that, so I'll get me coat......)

Ah 20/75 sync, Max Obtainable is 22/77

Current Rate (Kbps) 20000 75708
Max Rate (Kbps) 22638 77768
SNR Margin (dB) 7 3.3
Line Attenuation (dB) 14 6.5
Errors (Pkts) 0 0

Edited by broadbandjockey (Tue 27-Aug-19 14:54:09)

In reply to a post by broadbandjockey:

I've just moved house, and I'm very close to the FTTC cabinet, syncing at close to 78 Mb/s
However, real speedtests etc, and even attempting ftp transfers as tests never get above 60 Mb/s. Upstream I'm syncing at 22 Mb/s, and I get 18-19 Mb/s real world values (As I'd expect)

I sync at 72 Mbps and regularly get the speeds below (wired) i have found different speed tests give different results and many vary a lot, my preference is the test files as i think they are the most accurate way to test.
(i am just a tyre kicker so could be wrong)

My test result (close enough for me) - https://i.postimg.cc/bJHQDx1Y/ffffffff.jpg

The speed test i used above is here - http://speedtest2.aa.net.uk/

I also use TBB test files which gives similar results - https://www.thinkbroadband.com/download

.

Edited by deleted (Tue 27-Aug-19 16:49:00)

Are you running anything else on anything, like a smart TV, at the same time? And presumably in the pub your phone is on wireless.

Is your ISP BT Wholesale based? That introduces an IP profile that knocks a bit (or a few ) out.

In reply to a post by RobertoS:

Are you running anything else on anything, like a smart TV, at the same time? And presumably in the pub your phone is on wireless.

Is your ISP BT Wholesale based? That introduces an IP profile that knocks a bit (or a few ) out.

Yes, Plusnet connection. Tests have been conducted with nothing else connected (WiFi or LAN) to my router and using a wired LAN connection,

The pub was via WiFi, and sat next to the router, naturally no idea of any other conditions other than it is a BT hub, so presumably BT service

In reply to a post by Chippy_Tea_:

In reply to a post by broadbandjockey:

I've just moved house, and I'm very close to the FTTC cabinet, syncing at close to 78 Mb/s
However, real speedtests etc, and even attempting ftp transfers as tests never get above 60 Mb/s. Upstream I'm syncing at 22 Mb/s, and I get 18-19 Mb/s real world values (As I'd expect)

I sync at 72 Mbps and regularly get the speeds below (wired) i have found different speed tests give different results and many vary a lot, my preference is the test files as i think they are the most accurate way to test.
(i am just a tyre kicker so could be wrong)

My test result (close enough for me) - https://i.postimg.cc/bJHQDx1Y/ffffffff.jpg

The speed test i used above is here - http://speedtest2.aa.net.uk/

I also use TBB test files which gives similar results - https://www.thinkbroadband.com/download

.

I've just run the AA test, 59.5, 18.5, ping of 11ms

It kind of feels like somewhere upstream I have a hard cap of 60!

Do Plusnet still use their "Current line speed" setting? I think it is still available using this link, but may not be.

That used to get stuck at times in the past, but if you've done a few re-sync's that's unlikely to be the problem.

If it is really stuck then should ne a nice flat line on https://www.thinkbroadband.com/speedtest if other variables are not affecting things

I am on PlusNet and basically get the full 80/20 upload/download IF I pick a server that I know has the capacity to max out my connection. I think it works out at about 78/19 averaged over ~1GB, I sync at 79999/19999 according to the stats from the modem with apparently a slight high attainable.

I am fortunate to work at a UK university looking after a HPC system and have access to servers with 10Gbps connections and the University has an even fatter pipe into JANET. Though truth be told I can max it out from my desktop at work with just a 1Gbps connection.

I have in the past had download speeds very close to 76Mbps with sync of 80 and max achievable at 90ish. However new installs have dropped my max achieveable to around 78 and sync to 76 ...

I am around 450m from the cabinet.

In reply to a post by RobertoS:

Do Plusnet still use their "Current line speed" setting? I think it is still available using this link, but may not be.

That used to get stuck at times in the past, but if you've done a few re-sync's that's unlikely to be the problem.

That link is dead. I've not done any re-syncs since moving in (8 weeks ago). The sync rate started at 58, and crept up to its present value after a few days. Perhaps I should reboot the router ?

Hi Broadbandjockey,

A way to identify the wat is causing the speed limit is to:

Check your BTW IP profile at:
http://speedtest.btwholesale.com/
Run the speedtest, then press the further diagnostics button, enter your phone number etc....

If the BTW IP profile is around 60 mbps then that is the cause of the speed limit.
If a modem resynch does not increase the BTW IP profile, then contact PN to raise the issue with BTW.

If the BTW IP profile is (obviously) higher than 60 mbps, then it is likely that the PN profile is the issue.
I find the easy way to get the PN profile checked and adjusted is to strat a topic on the PN community forums.

In reply to a post by broadbandjockey:

In reply to a post by Chippy_Tea_:

In reply to a post by broadbandjockey:

I've just moved house, and I'm very close to the FTTC cabinet, syncing at close to 78 Mb/s
However, real speedtests etc, and even attempting ftp transfers as tests never get above 60 Mb/s. Upstream I'm syncing at 22 Mb/s, and I get 18-19 Mb/s real world values (As I'd expect)

I sync at 72 Mbps and regularly get the speeds below (wired) i have found different speed tests give different results and many vary a lot, my preference is the test files as i think they are the most accurate way to test.
(i am just a tyre kicker so could be wrong)

My test result (close enough for me) - https://i.postimg.cc/bJHQDx1Y/ffffffff.jpg

The speed test i used above is here - http://speedtest2.aa.net.uk/

I also use TBB test files which gives similar results - https://www.thinkbroadband.com/download

.

I've just run the AA test, 59.5, 18.5, ping of 11ms

It kind of feels like somewhere upstream I have a hard cap of 60!

Have you tried the thinkbroadbasnd test files? - https://www.thinkbroadband.com/download

.

In reply to a post by skandia2:

Hi Broadbandjockey,

A way to identify the wat is causing the speed limit is to:

Check your BTW IP profile at:
http://speedtest.btwholesale.com/
Run the speedtest, then press the further diagnostics button, enter your phone number etc....

Which gives me the following message

1. Downstream Test:

Download Speed : 58.18 Mbps

Your speed test has completed and the results are shown above, however during the test an error occurred while trying to retrieve additional details regarding your service. As a result we are unable to determine if the speed you received during the test is acceptable for your service. Please re-run the test if you require this additional information.

In reply to a post by Chippy_Tea_:

In reply to a post by broadbandjockey:

In reply to a post by Chippy_Tea_:

... nested quotes trimmed ...

I sync at 72 Mbps and regularly get the speeds below (wired) i have found different speed tests give different results and many vary a lot, my preference is the test files as i think they are the most accurate way to test.
(i am just a tyre kicker so could be wrong)

My test result (close enough for me) - https://i.postimg.cc/bJHQDx1Y/ffffffff.jpg

The speed test i used above is here - http://speedtest2.aa.net.uk/

I also use TBB test files which gives similar results - https://www.thinkbroadband.com/download

.

I've just run the AA test, 59.5, 18.5, ping of 11ms

It kind of feels like somewhere upstream I have a hard cap of 60!

Have you tried the thinkbroadbasnd test files? - https://www.thinkbroadband.com/download

.

Yep, just over 3 mins for 1GB, so just under 60 Mb/s?

In reply to a post by broadbandjockey:

Yep, just over 3 mins for 1GB, so just under 60 Mb/s?

It's not quite so simple.

That file is 1 GiB (gibibyte), which is 1024x1024x1024 = 1073741824 bytes. But comms speeds are measured in powers of 10, e.g. 1Mbps is 1,000,000 bits per second.

Now, suppose your transfer took 3 mins and 5 seconds (185 seconds). The payload throughput (not counting headers etc) is:

1024x1024x1024x8 / 185 = 46432078 bps ~= 46.4 Mbps

Given that this is a TCP/IP stream, each 1500-byte datagram carries 1460 bytes of TCP data. So the throughput including TCP/IP headers is about 47.7 Mbps.

In reply to a post by candlerb:

In reply to a post by broadbandjockey:

Yep, just over 3 mins for 1GB, so just under 60 Mb/s?

It's not quite so simple.

That file is 1 GiB (gibibyte), which is 1024x1024x1024 = 1073741824 bytes. But comms speeds are measured in powers of 10, e.g. 1Mbps is 1,000,000 bits per second.

Now, suppose your transfer took 3 mins and 5 seconds (185 seconds). The payload throughput (not counting headers etc) is:
1024x1024x1024x8 / 185 = 46432078 bps ~= 46.4 Mbps

Given that this is a TCP/IP stream, each 1500-byte datagram carries 1460 bytes of TCP data. So the throughput including TCP/IP headers is about 47.7 Mbps.

Your calculation seems reasonable, but I was only going by the timing guides on the download page itself

In reply to a post by broadbandjockey:

I've just moved house, and I'm very close to the FTTC cabinet, syncing at close to 78 Mb/s
However, real speedtests etc, and even attempting ftp transfers as tests never get above 60 Mb/s.

I'm only syncing at just above 73 Mb/s and my speed test results are typically around 67 Mb/s.

Are you testing over a wired or wireless connection? The latter can often yield worse results than wired. Alternatively it could be a hardware issue, configuration issue or just a poor ISP.

Edited by Andrue (Thu 29-Aug-19 15:56:52)

In reply to a post by candlerb:

In reply to a post by broadbandjockey:

Yep, just over 3 mins for 1GB, so just under 60 Mb/s?

So the throughput including TCP/IP headers is about 47.7 Mbps.

The 200 mb file which i used regularly takes 28 seconds which according to the times next to it is slightly below 60 Mbps (it should take 27 seconds)

Are you telling us all the times given with each file are wrong?

.

In reply to a post by Chippy_Tea_:

Are you telling us all the times given with each file are wrong?

Seems so - they're not even consistent with each other.

For example, it says that 200MiB takes 27 seconds @ 60Mbps, but 1GiB (1024MiB) takes 3 mins = 180 seconds.

FWIW, I calculate:

200MiB @ 60Mbps to be 200*1024*1024*8/(60*1000*1000) = 27.96 secs
1024MiB @ 60Mbps to be 1024*1024*1024*8/(60*1000*1000) = 143.17 secs

Could be rounding errors - although 143 secs is closer to 2 mins than 3 mins.

Makes note to either add a note that timings are approximate or display them to the precise second.

In reply to a post by MrSaffron:

Makes note to either add a note that timings are approximate or display them to the precise second.

To be honest i usually just click the file and see where the download speed settles out at after a few seconds for me its usually around 60M/bs.

Edited by deleted (Thu 29-Aug-19 20:52:43)

In reply to a post by RobertoS:

Do Plusnet still use their "Current line speed" setting? I think it is still available using this link, but may not be.

That used to get stuck at times in the past, but if you've done a few re-sync's that's unlikely to be the problem.

Well, as I said a few days ago, I've not rebooted since the day the service was connected, two months ago.

I've just done a reboot now, and now I'm getting, 68-69 ish on speedtests, (via LAN, everything else disconnected) I'm syncing at 75.8, so probably not far off the theoretical max ?

Can someone explain what and how 'Max Obtainable' sync is calculated ?

In reply to a post by broadbandjockey:

In reply to a post by RobertoS:

Do Plusnet still use their "Current line speed" setting? I think it is still available using this link, but may not be.

That used to get stuck at times in the past, but if you've done a few re-sync's that's unlikely to be the problem.

Well, as I said a few days ago, I've not rebooted since the day the service was connected, two months ago.

I've just done a reboot now, and now I'm getting, 68-69 ish on speedtests, (via LAN, everything else disconnected) I'm syncing at 75.8, so probably not far off the theoretical max ?

Can someone explain what and how 'Max Obtainable' sync is calculated ?

That's the maximum sync speed possible if the router and DSLAM agree to ignore all the rules. Typically that would result in an unstable connection that failed in short order. To avoid that the router and DSLAM use various rules to adjust the parameters of connection trading off speed against stability.

Under some circumstances the current sync speed can be lower than is strictly necessary for stability but in practice what has been negotiated is usually the highest and most reliable speed possible. This will often be lower than the highest possible speed but is intended to try and achieve 24/7 connectivity.

Edited by Andrue (Sun 01-Sep-19 18:56:02)

In reply to a post by Andrue:

Under some circumstances the current sync speed can be lower than is strictly necessary for stability

One example is if you live very close to the cabinet. VDSL2 profile 17a is capable of 100Mbps, but since the product is defined by OpenReach as having a maximum sync speed of 80Mbps, they limit it to that - even if higher speeds would be reliable.

In reply to a post by candlerb:

In reply to a post by Andrue:

Under some circumstances the current sync speed can be lower than is strictly necessary for stability

One example is if you live very close to the cabinet. VDSL2 profile 17a is capable of 100Mbps, but since the product is defined by OpenReach as having a maximum sync speed of 80Mbps, they limit it to that - even if higher speeds would be reliable.

So in other words the router is extrapolating the SNR vs Sync curve to the threshold point?

Makes sense, that's why then my son's max obtainable is 62, but the sync speed is exactly 40, because he's on an 'up to 38' package. If he went for an up to 76 package, he'd probably be syncing at 57ish?

Something like that, though you don't have to be that close to the cabinet to get max obtainable greater than 80/20. I sync at 79999/19999 which I presume is the full 80/20 with a max obtainable of something like 89/23. My mother is even closer to her cabinet getting the same 79999/19999 with an even higher max obtainable. However I am sufficiently far away from the cabinet that I can't get G.fast despite the cabinet being enabled. I am sure profile 35b would have made much more sense that G.fast.