ISPs BB checkers and speeds

been looking at changing ISP and trying the line/speed checkers on their
sites.
Some give 1Mb/s; one gave 3Mb/s and some are 0.5Mb/s.

I've just 'phoned Plusnet, spoke to Tech. Support (on the basis that Sales
will either know nothing or tell me what I want to hear).

PN said that the check was Offcom's, so I'm suspecting that it's there to
protect ISPs, as I'm currently connected at 2.1M and have had 3.1M.

Am I right on this - that Offcom and the ISPs are being ultra-cautious and
I should be able to get my usual speeds?
--
Peter.
You don't understand Newton's Third Law of Motion?
It's not rocket science, you know.

PeterC wrote:

> been looking at changing ISP and trying the line/speed checkers on their
> sites.
> Some give 1Mb/s; one gave 3Mb/s and some are 0.5Mb/s.

The only useful speed test is to find a large file and download it.
Instantaneous speed tests won't give you a meaningful picture of your
broadband speed.

> PN said that the check was Offcom's, so I'm suspecting that it's there to
> protect ISPs, as I'm currently connected at 2.1M and have had 3.1M.

Which 'the check' are you talking about? Are those speeds your sync speed or
your actual measured throughput? Sync speeds tend to depend on line
quality, laws of physics etc and not the incompetence or otherwise of your
ISP/Ofcom/Thatcher ruining the country in the 1980s.

--
<http://ale.cx/> (AIM:troffasky) ((E-Mail Removed))
22:45:43 up 76 days, 57 min, 2 users, load average: 0.24, 0.10, 0.08
Sexy ladies, and nasty boys, all freaky freakin', to the robot noise

On Wed, 18 Feb 2009 22:51:30 +0000, alexd wrote:

> PeterC wrote:
>
>> been looking at changing ISP and trying the line/speed checkers on their
>> sites.
>> Some give 1Mb/s; one gave 3Mb/s and some are 0.5Mb/s.
>
> The only useful speed test is to find a large file and download it.
> Instantaneous speed tests won't give you a meaningful picture of your
> broadband speed.
>
>> PN said that the check was Offcom's, so I'm suspecting that it's there to
>> protect ISPs, as I'm currently connected at 2.1M and have had 3.1M.
>
> Which 'the check' are you talking about? Are those speeds your sync speed or
> your actual measured throughput? Sync speeds tend to depend on line
> quality, laws of physics etc and not the incompetence or otherwise of your
> ISP/Ofcom/Thatcher ruining the country in the 1980s.

My test is, at a quiet time, start a d/l of Ubuntu from Oxford (about the
closest), just give it 100MB and extrapolate to a full distro.

Yes, 'real life' speeds are the useful ones.
--
Peter.
You don't understand Newton's Third Law of Motion?
It's not rocket science, you know.

On Wed, 18 Feb 2009 23:13:42 +0000, PeterC wrote:

>> Which 'the check' are you talking about? Are those speeds your sync speed or
>> your actual measured throughput? Sync speeds tend to depend on line
>> quality, laws of physics etc and not the incompetence or otherwise of your
>> ISP/Ofcom/Thatcher ruining the country in the 1980s.
>
>My test is, at a quiet time, start a d/l of Ubuntu from Oxford (about the
>closest), just give it 100MB and extrapolate to a full distro.
>
>Yes, 'real life' speeds are the useful ones.

Agreed, but your sync speed provides the upper limit on what you can
possibly get (it's about 10% below the sync speed due to overheads).
You should be able to get that figure from your modem.

Bear in mind that your internal wiring can have an enormous effect on
this speed. If it's poor it can really clobber things. You should be
looking at the down sync speed and the signal to noise ratio margin and
see how the margin varies throughout the day. The target for it should
be about 6dB but if you are on a long line it can vary a lot due to
noise pickup at various times on the line from the exchange.

Also monitor the attenuation. That increases with line length or bad
joins in the line and as that increases the margin tends to fall.


--
Regards - Rodney Pont
The from address exists but is mostly dumped,
please send any emails to the address below
e-mail ngpsm4 (at) infohitsystems (dot) ltd (dot) uk

On Wed, 18 Feb 2009 23:50:47 +0000 (GMT), Rodney Pont wrote:

> On Wed, 18 Feb 2009 23:13:42 +0000, PeterC wrote:
>
>>> Which 'the check' are you talking about? Are those speeds your sync speed or
>>> your actual measured throughput? Sync speeds tend to depend on line
>>> quality, laws of physics etc and not the incompetence or otherwise of your
>>> ISP/Ofcom/Thatcher ruining the country in the 1980s.
>>
>>My test is, at a quiet time, start a d/l of Ubuntu from Oxford (about the
>>closest), just give it 100MB and extrapolate to a full distro.
>>
>>Yes, 'real life' speeds are the useful ones.
>
> Agreed, but your sync speed provides the upper limit on what you can
> possibly get (it's about 10% below the sync speed due to overheads).
> You should be able to get that figure from your modem.
>
Not without some software - it's a USB modem with a 'go-faster' firmware
upgrade.
The same PC on an ADSL modem gets the same speeds.
I can get 3M but Tiscrap limits it to 2.1M.

I'm 3.3km from the exchange (by road).

> Bear in mind that your internal wiring can have an enormous effect on
> this speed. If it's poor it can really clobber things. You should be
> looking at the down sync speed and the signal to noise ratio margin and
> see how the margin varies throughout the day. The target for it should
> be about 6dB but if you are on a long line it can vary a lot due to
> noise pickup at various times on the line from the exchange.

There's only a 'phone on the circuit and, apart from the PC there's no
other wiring near it.
>
> Also monitor the attenuation. That increases with line length or bad
> joins in the line and as that increases the margin tends to fall.


--
Peter.
You don't understand Newton's Third Law of Motion?
It's not rocket science, you know.

On Thu, 19 Feb 2009 13:25:33 +0000, PeterC wrote:

>Not without some software - it's a USB modem with a 'go-faster' firmware
>upgrade.
>The same PC on an ADSL modem gets the same speeds.

I take it the modem was in your socket, otherwise you can't really
compare the two.

I'm not a fan of the USB ones personally. I have a friend with one and
they usually get about 5meg but I plugged my modem in and it reported
8128 sync speed so they should get 7150Mb/sec with that. They don't
want the extra wallwart and power cable though... They are only a
couple of hundred metres from the exchange as well and are with
TalkTalk. The only problem they had was with voice on the phone and it
took ages to get it sorted. TT didn't really want to call BT and of
course BT wouldn't accept a fault from my friends because their
contract was with TT and not them.

>I can get 3M but Tiscrap limits it to 2.1M.

That's about right for a 3meg sync speed although maybe a little slow
but it does depend on the sync rate throughout the day, the BRAS
(maximum IP rate) will be set to the lowest speed you connect at. There
is a nice explanation here
http://aaisp.net.uk/kb-broadband-how-atm.html with a table near the
bottom showing how they match. This is for BT though and if you have
LLU it may have different sync speed/IP rate blocks.

>I'm 3.3km from the exchange (by road).

I have another friend, yes really - more than one, who are about 4km
from the exchange and they get 5meg, can't remember if that's sync or
download speed though. Their line is obviously in better condition than
yours but that happens and there isn't anything you can do to improve
matters.

About half a km for me but the wires go round a couple of fields to get
to me, don't know why they don't just come down the road because there
are no buildings over the fields and the road to the exchange is
straight. BT put a new conduit in a couple of years ago and that does
come straight down the road so if they ever replace the wiring I
presume they will use that since there is a manhole just outside my
drive. Easy run for fibre if I could afford the annual cost and needed
the speed :-)

>> Bear in mind that your internal wiring can have an enormous effect on
>> this speed. If it's poor it can really clobber things. You should be
>> looking at the down sync speed and the signal to noise ratio margin and
>> see how the margin varies throughout the day. The target for it should
>> be about 6dB but if you are on a long line it can vary a lot due to
>> noise pickup at various times on the line from the exchange.
>
>There's only a 'phone on the circuit and, apart from the PC there's no
>other wiring near it.

It sounds like the only thing that could be wrong with that then is no
filter for the phone but I assume you have one in there :-)

--
Regards - Rodney Pont
The from address exists but is mostly dumped,
please send any emails to the address below
e-mail ngpsm4 (at) infohitsystems (dot) ltd (dot) uk

On Thu, 19 Feb 2009 16:33:45 GMT, Ato_Zee wrote:

>>>There's only a 'phone on the circuit and, apart from the PC there's no
>>>other wiring near it.
>
> No extension wiring to the modem/router?
> Often extension phone or ADSL extension domestic wiring
> between floors follows the same route parallel to the mains.
> Put a low energy lamp, or anything with a switch mode
> PSU on the mains, and the noise from it will fall nicely
> in the ADSL spectrum, the extension will be a
> nice indoor aerial.
> I used the old quad as a draw wire to pull in CAT5,
> router now next to NTE5 and much faster.

Yes, but the master socket is next to the desk.
--
Peter.
You don't understand Newton's Third Law of Motion?
It's not rocket science, you know.