ADSL over coax (AKA I don't know anything about ADSL)

[David Paste]

As I understand it, ADSL signals are passed along the twisted pair (is
that right?) telephone cables which are the same used for voice calls.
At a much higher frequency, of course. I am also under the impression
that the length of this cable is critical for service quality and
overall data rate (apart from junction / joint quality).

Would using coax for the last stretch to the subscriber's premises
provide any better quality of service?

I only ask as my cable service uses coax and I wondered about the
difference.

Cheers,

David.

[Theo Markettos]

David Paste <paste...@gmail.com> wrote:
> As I understand it, ADSL signals are passed along the twisted pair (is
> that right?) telephone cables which are the same used for voice calls.
> At a much higher frequency, of course. I am also under the impression
> that the length of this cable is critical for service quality and
> overall data rate (apart from junction / joint quality).
>
> Would using coax for the last stretch to the subscriber's premises
> provide any better quality of service?

Yes and no.

Both ADSL and cable are adaptive technologies. In other words, they take
the bit of wet string that is your line, analyse it, see where the frequency
nulls are, and send the data in frequency bands to avoid them. As things
change (atmospherics, day/night, water, interference, whatever), they can
adjust the bands to cope.

In cable's case the network quality is much better so there are many fewer
nulls and you can get much higher bandwidth out of it. However you're also
sharing that coax with your neighbours - and it carries all the TV signals
too[1]. Plus the cable operator needs to agree with each modem what bands it
gets so they don't conflict.

So you could route ADSL over coax, but it's still a fundamentally different
technology. Plus you'd have to impedance-match the interfaces between
twisted pair and coax which would increase the losses. So a short coax run
might make it worse not better.

Theo

[1] back when analogue cable was still alive, you could stuff the cable in
the back of your TV and get FTA TV directly, no STB required. Same goes for
FM radio.

[Adrian]

On Sat, 08 Feb 2014 17:56:29 +0000, Theo Markettos wrote:

> In cable's case the network quality is much better so there are many
> fewer nulls and you can get much higher bandwidth out of it. However
> you're also sharing that coax with your neighbours - and it carries all
> the TV signals too[1]. Plus the cable operator needs to agree with each
> modem what bands it gets so they don't conflict.

Also worth mentioning that the cable is only split to each household at
the street cabinets, whilst ADSL goes all the way back to the exchange.

[c...@isbd.net]

David Paste <paste...@gmail.com> wrote:
> As I understand it, ADSL signals are passed along the twisted pair (is
> that right?) telephone cables which are the same used for voice calls.
> At a much higher frequency, of course. I am also under the impression
> that the length of this cable is critical for service quality and
> overall data rate (apart from junction / joint quality).
>
> Would using coax for the last stretch to the subscriber's premises
> provide any better quality of service?
>

Only if "the last stretch" was a significant proportion of the total
distance to the exchange, and even then you'd have to have impedance
balancing devices between the twisted pair and the co-ax.

> I only ask as my cable service uses coax and I wondered about the
> difference.
>

That's co-ax all the way - or at least until it reaches fibre optic or
some other very fast 'highway', so it's going to be able in principle to
provide a much faster connection.

--
Chris Green
·

[Graham.]

On Sat, 8 Feb 2014 09:24:37 -0800 (PST), David Paste
<paste...@gmail.com> wrote:

>As I understand it, ADSL signals are passed along the twisted pair (is
>that right?) telephone cables which are the same used for voice calls.
>At a much higher frequency, of course. I am also under the impression
>that the length of this cable is critical for service quality and
>overall data rate (apart from junction / joint quality).

Correct.

>Would using coax for the last stretch to the subscriber's premises
>provide any better quality of service?
>
>I only ask as my cable service uses coax and I wondered about the
>difference.

You have sort of answered your own question.

The whole point of ADSL is that it was invented as a means to send
high speed data over twisted pairs several km long that were only
originally intended for base band voice. The system is highly
adaptive, using digital line management.

ADSL also needs a modem at each end of each subscribers line, where
co-ax distribution systems like Virgin's can be largely passive.



--
Graham.


%Profound_observation%

[dennis@home]

The UK cable network does not quite do that.
There are splitter boxes in the footpath holes and a cable runs from the
street cabinet and is split off at the nearest hole in the ground.
There can be dozens of houses hanging off each cable coming from the
street cabinet.

[Rick Hughes]

On a d-i-y point - put Coax just on last part and most likely you will
have impedance mismatch, and degraded signal.
If coax were used to exchange or to fibre cabinet then yes that would
be better than twisted pair.




--
UK SelfBuild: http://uk.groups.yahoo.com/group/UK_Selfbuild/

[John Rumm]

On 08/02/2014 17:24, David Paste wrote:

> As I understand it, ADSL signals are passed along the twisted pair (is
> that right?) telephone cables which are the same used for voice calls.
> At a much higher frequency, of course. I am also under the impression
> that the length of this cable is critical for service quality and
> overall data rate (apart from junction / joint quality).

Yup pretty much. There are a few different flavours of DSL technology,
and give a relatively short decent line some will get noticeably better
performance than others. However once you get to a few km of wire they
all become similar in performance.

See the graph for the tradeoff:

http://www.internode.on.net/residential/adsl_broadband/easy_broadband/performance/

> Would using coax for the last stretch to the subscriber's premises
> provide any better quality of service?

Depends on if you mean "could the system be re-engineered to do this and
would it help" (yes, and probably not noticeably)[1], or could you just
"lashup a bit of coax in the place of twisted pair, and would that help"
(yes, and absolutely not).

So the short answer no.

> I only ask as my cable service uses coax and I wondered about the
> difference.

You can use co-ax or twisted pair for data transmission - each have pros
and cons but the mechanisms at play for keeping the signal in and the
noise out are different - so the electronics that drives and receives
these signals needs to be specific to the cable type.

[1] The other way round could work better; replace the miles of damp
string with co-ax and then use the damp string for the last few hundred
yards ;-)

--
Cheers,

John.

/=================================================================\
| Internode Ltd - http://www.internode.co.uk |
|-----------------------------------------------------------------|
| John Rumm - john(at)internode(dot)co(dot)uk |
\=================================================================/

[Brian_Gaff]

Well, I'd guess that there would be less crosstalk in a couple of coax
cables running near each other. It really is the rest of the route that is
the issue though, the twisted pair from your house to wherever the
multiplexer is. If you are proposing coax all the way, then you are
suggesting a very big job in many cases. The small bit from the pole or
junction box t to your house is only part of it.
I always thought adsl was ab bit of a bodge in any case, trying to use
cables probably installed before the internet was even launched, for such
purposes. Brute force in both directions comes to mind!
Brian

--
From the Sofa of Brian Gaff Reply address is active
"David Paste" <paste...@gmail.com> wrote in message
news:55494342-fa67-4487...@googlegroups.com...

[Dave Liquorice]

On Sat, 8 Feb 2014 09:24:37 -0800 (PST), David Paste wrote:

> As I understand it, ADSL signals are passed along the twisted pair (is
> that right?) telephone cables which are the same used for voice calls.
> At a much higher frequency, of course. I am also under the impression
> that the length of this cable is critical for service quality and
> overall data rate (apart from junction / joint quality).

Pretty much correct. People underestimate the bandwidth that can be
carried over a twisted pair. 40 odd years ago broadcast video (up to
5+ MHz) was sent over half a dozen bonded twisted pairs, not far only
a mile or so but shows what could be done with the twisted pair and
the technology back then. This is ten years before the home computer
and 20 odd before cheap general access to the internet.

> Would using coax for the last stretch to the subscriber's premises
> provide any better quality of service?

It might but not without changing the kit each end. You couldn't just
join the two together, twisted pair is balanced, coax unbalanced.
Making a passive balun (balanced to unbalanced) convertor of
suffcient bandwidth (30 odd MHz) wouldn't be easy.

--
Cheers
Dave.

[charles]

In article <nyyfbegfubjuvyypb...@srv1.howhill.co.uk>,

Dave Liquorice <allsortsn...@howhill.com> wrote:
> On Sat, 8 Feb 2014 09:24:37 -0800 (PST), David Paste wrote:

> > As I understand it, ADSL signals are passed along the twisted pair (is
> > that right?) telephone cables which are the same used for voice calls.
> > At a much higher frequency, of course. I am also under the impression
> > that the length of this cable is critical for service quality and
> > overall data rate (apart from junction / joint quality).

> Pretty much correct. People underestimate the bandwidth that can be
> carried over a twisted pair. 40 odd years ago broadcast video (up to
> 5+ MHz) was sent over half a dozen bonded twisted pairs, not far only
> a mile or so but shows what could be done with the twisted pair and
> the technology back then.

I've seen Riverside Studios to TVC, which must be getting on for 2 miles
allowing for detours to HAMmersmith & SHEpherds Bush exchanges.

--
From KT24

Using a RISC OS computer running v5.18

[Adrian]

On Sun, 09 Feb 2014 10:15:52 +0000, Dave Liquorice wrote:

> People underestimate the bandwidth that can be carried over a twisted
> pair.

Quite. Gigabit over twisted pair is almost ubiquitous, and ten gig is out
there - although 10gbe cable is stiffer.

[tony sayer]

In article <nyyfbegfubjuvyypb...@srv1.howhill.co.uk>, Dave

Liquorice <allsortsn...@howhill.com> scribeth thus

AFAIR in the local cable system here which is and has been right from
the word go fibre to the "cabinet" and then co-ax to the sub of varying
lengths, there was or is an upper limit on the co-ax of 1 Ghz odd and at
that end the losses are well on the go. I do believe they have
equalisers to render the whole co-ax channel a lot flatter then it
otherwise would be.....

But I do know that on our 30 Meg service each and every time I do a
speed test its just a shade over the 30 mark which is were it should be
supposed to be getting 60 'ere long:)...

--
Tony Sayer

[Andy Dingley]

On Saturday, 8 February 2014 17:24:37 UTC, David Paste wrote:
> Would using coax for the last stretch to the subscriber's premises
> provide any better quality of service?

Broadly, no. Twisted pair is surprisingly good as cabling for high bit rate digital.

Coax developed for analogue signals. It's good at avoiding lots of problems that affect high bandwidth analogue. When digital telephony began over the local loop (ISDN, early '80s) it was discovered that pairs worked surprisingly well and also also that the problems affecting pairs (dispersion for one) weren't problems that were quite so important to this type of signal. Closely spaced twisted pair (with terminations designed to work with it) is even better. As it's also far cheaper, this is one reason for 10baseT replacing both thick & thin coax for Ethernet.

[The Natural Philosopher]

Sasy rather that advanced adaptive signal processing and modulation
schemas made ADSL over existing copper possible in a way it wasn't before.

Not quite the story for Ethernet as that was always possible but the
cost of having a switch held it back a bit, until the cost of all that
coax and the unreliability in large networks made todays arrangement
vastly cheaper...





--
Ineptocracy

(in-ep-toc’-ra-cy) – a system of government where the least capable to
lead are elected by the least capable of producing, and where the
members of society least likely to sustain themselves or succeed, are
rewarded with goods and services paid for by the confiscated wealth of a
diminishing number of producers.

[Andrew Gabriel]

In article <a8scf99asija1jdh9...@4ax.com>,

That's not really fair. Cable has a multi-channel VHF modem at each
end - effectively something like 8 modems running in parallel on
different RF frequencies on the same cable.

--
Andrew Gabriel
[email address is not usable -- followup in the newsgroup]

[The Natural Philosopher]

On 09/02/14 23:04, Tim Streater wrote:
> In article <ld8l4h$khf$1...@news.albasani.net>, The Natural Philosopher

> <t...@invalid.invalid> wrote:
>
>> On 09/02/14 15:38, Andy Dingley wrote:
>> > On Saturday, 8 February 2014 17:24:37 UTC, David Paste wrote:
>> >> Would using coax for the last stretch to the subscriber's premises
>> >> provide any better quality of service?
>> >
>> > Broadly, no. Twisted pair is surprisingly good as cabling for high
>> > bit rate digital.
>> >
>> > Coax developed for analogue signals. It's good at avoiding lots of
>> > problems that affect high bandwidth analogue. When digital telephony
>> > began over the local loop (ISDN, early '80s) it was discovered that
>> > pairs worked surprisingly well and also also that the problems
>> > affecting pairs (dispersion for one) weren't problems that were quite
>> > so important to this type of signal. Closely spaced twisted pair
>> > (with terminations designed to work with it) is even better. As it's
>> > also far cheaper, this is one reason for 10baseT replacing both thick
>> > & thin coax for Ethernet.
>> >
>> Sasy rather that advanced adaptive signal processing and modulation
>> schemas made ADSL over existing copper possible in a way it wasn't
>> before.
>>
>> Not quite the story for Ethernet as that was always possible but the
>> cost of having a switch held it back a bit, until the cost of all
>> that coax and the unreliability in large networks made todays
>> arrangement vastly cheaper...
>

> Unreliable it is if you've ever had to make a tap on thick ethernet -
> or had smart-alec physicists think they can just add 50m of their own
> 2mm thick 50 ohm coax on the end of a segment of thin ethernet.
>
BTDTGTTS. Try 70 dumb users on one length of coax.

[dennis@home]

On 09/02/2014 23:04, Tim Streater wrote:

> Unreliable it is if you've ever had to make a tap on thick ethernet -
> or had smart-alec physicists think they can just add 50m of their own
> 2mm thick 50 ohm coax on the end of a segment of thin ethernet.
>

Its much more fun to add a tee and a few meters of coax onto a PC.
If you get it right you can make it so some PCs are invisible to others
while they work to the rest of the network. ;-)

[Andy Dingley]

On Sunday, 9 February 2014 23:04:44 UTC, Tim Streater wrote:
> Unreliable it is if you've ever had to make a tap on thick ethernet -
> or had smart-alec physicists think they can just add 50m of their own
> 2mm thick 50 ohm coax on the end of a segment of thin ethernet.

Are all the terminators on?

Yes.

Have you looked?

<walks off>

Yes. All three ends have terminators on. I told you our network was OK, it _must_ be your software's fault!


(It was a T shaped building. It made a sort of sense.)

[John Rumm]

Its probably fair to say that *any* broadband technology needs a modem
at each end, almost by definition.

(whether you would argue an optical fibre is broadband or not is another
matter!)

[John Rumm]

On 10/02/2014 08:35, Tim Streater wrote:
> In article <a218ffc8-c68c-4073...@googlegroups.com>,

> Not only did the physicists in question add a length of their own
> thin-thin-thin ethernet, they also, as you hint, added T-sections
> elsewhere so that instead of the thinnet going to the back of their
> Sun, and being connected with the shortest possible T (just the
> connector), they had a 15ft or so long T. It made the cabling much
> neater :-)

Kind of like when (some) electricians do TV systems...

[The Natural Philosopher]

On 10/02/14 11:11, John Rumm wrote:
> On 10/02/2014 08:35, Tim Streater wrote:
>> In article <a218ffc8-c68c-4073...@googlegroups.com>,
>> Andy Dingley <din...@codesmiths.com> wrote:
>>
>>> On Sunday, 9 February 2014 23:04:44 UTC, Tim Streater wrote:
>>> > Unreliable it is if you've ever had to make a tap on thick ethernet -
>>> > or had smart-alec physicists think they can just add 50m of their own
>>> > 2mm thick 50 ohm coax on the end of a segment of thin ethernet.
>>>
>>> Are all the terminators on?
>>>
>>> Yes.
>>>
>>> Have you looked?
>>>
>>> <walks off>
>>>
>>> Yes. All three ends have terminators on. I told you our network was
>>> OK, it
>>> _must_ be your software's fault!
>>>
>>> (It was a T shaped building. It made a sort of sense.)
>>
>> Not only did the physicists in question add a length of their own
>> thin-thin-thin ethernet, they also, as you hint, added T-sections
>> elsewhere so that instead of the thinnet going to the back of their
>> Sun, and being connected with the shortest possible T (just the
>> connector), they had a 15ft or so long T. It made the cabling much
>> neater :-)
>
> Kind of like when (some) electricians do TV systems...
>
>

<grin>

I had to extend a coax cable from where it was originally intended to
where SWMBO wanted it. Then ultimately she realise that where the
original socket was was indeed a better place, but by that time the
radio tuner which shares the same cable (VHF and TV are multiplexed down
the same cable from a distribution amp) had grown roots and was to stay
where it was.

I am afraid I simply 'tapped off' the cable in a short spur to the TV..

It all still works with no ill effects.

[David Paste]

Thanks for all the replies.

So ADSL is a technology developed to cope with the installed telephone lines, not a technology developed independently and then shoe-horned onto the existing lines? This makes much more sense to me now!

The one thing that annoys me about cable is the asymmetry of upload speed compared to download. I have a 60 meg DL, and 3 meg UL. I do enough uploading for this to be a bit of a pain in the arse!

ADSL seems to have MUCH better upload speeds. As I understand it, it is due to the bandwidth allocation of the cable system (DOCSYS?) and the 'traditional' bandwidth requirements of earlier consumer internet times.

I wonder if we'll ever get a Google fibre style company in the UK?

[tony sayer]

In article <b80d3ce0-eaaa-465a...@googlegroups.com>,
David Paste <paste...@gmail.com> scribeth thus

>Thanks for all the replies.
>
>So ADSL is a technology developed to cope with the installed telephone lines,
>not a technology developed independently and then shoe-horned onto the existing
>lines? This makes much more sense to me now!

About right..

>
>The one thing that annoys me about cable is the asymmetry of upload speed
>compared to download. I have a 60 meg DL, and 3 meg UL. I do enough uploading
>for this to be a bit of a pain in the arse!
>

Well does it really matter for the great majority of users ., the name
ADSL Asymmetric digital subscriber line implies that. Most everyone will
download faster than upload and unless your running a server what's the
real problem?, with anything I upload I just start it and let it get on
with it.

>ADSL seems to have MUCH better upload speeds. As I understand it, it is due to
>the bandwidth allocation of the cable system (DOCSYS?)

DOCSIS

http://en.wikipedia.org/wiki/DOCSIS

AIUI it is improving or will be when they change systems whenever that
will be..

>and the 'traditional'
>bandwidth requirements of earlier consumer internet times.
>
>I wonder if we'll ever get a Google fibre style company in the UK?

????....

--
Tony Sayer

[Dave Liquorice]

On Mon, 10 Feb 2014 11:07:09 +0000, John Rumm wrote:

> (whether you would argue an optical fibre is broadband or not is another
> matter!)

It could be some of the higher speed fibre connections use more than
one laser frequency (aka carrier), but then you are probably talking
silly data rates like a few Tbps... B-)

GPON uses two down a single fibre but in opposite directions.

Of course the vast majority saying that they have "fibre broadband"
haven't they have VDSL with the local head end fed by fibre. I reckon
all this crappy FTTC will come home to roost in ten years or so. When
streaming multiple channels of HDTV at sensible bit rates (10 Mbps or
more) becomes the norm, rather than the barely better than SD 2.5
Mbps or so that is used at present.

--
Cheers
Dave.

[Dave Liquorice]

On Mon, 10 Feb 2014 17:44:39 +0000, tony sayer wrote:

>> I wonder if we'll ever get a Google fibre style company in the UK?
>
> ????....

I think he means a company that installs FTTP. I doubt there will
ever be a national one that covers everywhere. FTTC sort of works OK
in densly populated areas. Move out into rural ones and the costs of
digging 2 miles to connect a customer aren't economic for a
commercial company.

There are a number number of community systems installed or very soon
to be. Fibre GarDen in Garsdale/Dentdale springs to mind.

--
Cheers
Dave.

[Capitol]

Fibre TTC doesn't have to be underground.

[John Rumm]

On 10/02/2014 16:42, David Paste wrote:
> Thanks for all the replies.
>
> So ADSL is a technology developed to cope with the installed
> telephone lines, not a technology developed independently and then
> shoe-horned onto the existing lines? This makes much more sense to me
> now!
>
> The one thing that annoys me about cable is the asymmetry of upload
> speed compared to download. I have a 60 meg DL, and 3 meg UL. I do
> enough uploading for this to be a bit of a pain in the arse!

Well that's better than any ADSL setup...

> ADSL seems to have MUCH better upload speeds. As I understand it, it
> is due to the bandwidth allocation of the cable system (DOCSYS?) and
> the 'traditional' bandwidth requirements of earlier consumer internet
> times.

The A of ADSL is the key! Typically most links will top out at about
1Mb/s upload even if getting the full ~24Mb/s download.

VDSL (i.e. FTTC) will do a much better uplink at over 8Mb/s

There is also SDSL (symmetric) available in some places (usually at much
higher cost) for those that need the outbound data rate.

> I wonder if we'll ever get a Google fibre style company in the UK?

Who knows... IIUC they have bought up loads of unlit fibre for future use.

[Andy Burns]

John Rumm wrote:

> The A of ADSL is the key! Typically most links will top out at about
> 1Mb/s upload even if getting the full ~24Mb/s download.
> VDSL (i.e. FTTC) will do a much better uplink at over 8Mb/s

I've recently been upgrading a network of private* ADSL1 connections,
some to ADSL2+M, some to EFM, some to FTTC, others to "proper" fibre,
depending on what's available at reasonable prices in various parts of
the sticks.

The FTTC ones have tended to be exceptionally close to the cabinet (as
in 5 yards outside the building) and have easily achieved 79Mbps down
and 19Mbps up speeds.

* i.e. they don't go to an ISP.

[John Rumm]

On 11/02/2014 08:19, Andy Burns wrote:
> John Rumm wrote:
>
>> The A of ADSL is the key! Typically most links will top out at about
>> 1Mb/s upload even if getting the full ~24Mb/s download.
>> VDSL (i.e. FTTC) will do a much better uplink at over 8Mb/s
>
> I've recently been upgrading a network of private* ADSL1 connections,
> some to ADSL2+M, some to EFM, some to FTTC, others to "proper" fibre,
> depending on what's available at reasonable prices in various parts of
> the sticks.

What options are available for areas that don't have local cabinets, and
the BT exchange can do ADSL max at best?

> The FTTC ones have tended to be exceptionally close to the cabinet (as
> in 5 yards outside the building) and have easily achieved 79Mbps down
> and 19Mbps up speeds.

Most of the FTTC installs I have seen have done 70 or better downstream
(sync speed, if not actual throughput)

>
> * i.e. they don't go to an ISP.

[The Natural Philosopher]

On 11/02/14 14:19, John Rumm wrote:
> On 11/02/2014 08:19, Andy Burns wrote:
>> John Rumm wrote:
>>
>>> The A of ADSL is the key! Typically most links will top out at about
>>> 1Mb/s upload even if getting the full ~24Mb/s download.
>>> VDSL (i.e. FTTC) will do a much better uplink at over 8Mb/s
>>
>> I've recently been upgrading a network of private* ADSL1 connections,
>> some to ADSL2+M, some to EFM, some to FTTC, others to "proper" fibre,
>> depending on what's available at reasonable prices in various parts of
>> the sticks.
>
> What options are available for areas that don't have local cabinets, and
> the BT exchange can do ADSL max at best?
>

essentially nothing.
Satellite maybe or a private wifi link to something better

>> The FTTC ones have tended to be exceptionally close to the cabinet (as
>> in 5 yards outside the building) and have easily achieved 79Mbps down
>> and 19Mbps up speeds.
>
> Most of the FTTC installs I have seen have done 70 or better downstream
> (sync speed, if not actual throughput)
>

cab get that at sub 400 meters usually I think.

>>
>> * i.e. they don't go to an ISP.
>
>


--

[David Paste]

On Tuesday, 11 February 2014 02:23:18 UTC, John Rumm wrote:

> VDSL (i.e. FTTC) will do a much better uplink at over 8Mb/s

Maybe this is what I was thinking of. A relation has BT Infinity,
installed last summer or thereabouts, and I was amazed by the upload
speed. I presume this will be VDSL (I did not know that VDSL was
different to ADSL).

> > I wonder if we'll ever get a Google fibre style company in the UK?
>
> Who knows... IIUC they have bought up loads of unlit fibre for future use.

Interesting!

[Andy Burns]

John Rumm wrote:

> What options are available for areas that don't have local cabinets, and
> the BT exchange can do ADSL max at best?

If you're lucky, EFM, ok it's out of most "home" budgets as it bonds
four PSTN lines but it gets 10-15Mbps symmetric, rather than whatever
ADSL1 can manage.

[John Rumm]

On 11/02/2014 17:52, David Paste wrote:
> On Tuesday, 11 February 2014 02:23:18 UTC, John Rumm wrote:
>
>> VDSL (i.e. FTTC) will do a much better uplink at over 8Mb/s
>
> Maybe this is what I was thinking of. A relation has BT Infinity,
> installed last summer or thereabouts, and I was amazed by the upload
> speed. I presume this will be VDSL (I did not know that VDSL was
> different to ADSL).

Yup that sounds likely.

VDSL is a relative of ADSL, but optimised for this particular
application, makes use of more sophisticated filtering and has much more
stringent length limits.

[John Rumm]

Interesting - that is one I had not investigated before. Prices are not
cheap (seems like £170/month is not uncommon for 10Mb symmetrical), but
compared with a couple of business class ADSL connections and bundled
phone lines its not a huge jump.

[tony sayer]

In article <FMOdndcnsN2OKmfP...@brightview.co.uk>, John
Rumm <see.my.s...@nowhere.null> scribeth thus

>On 11/02/2014 21:36, Andy Burns wrote:
>> John Rumm wrote:
>>
>>> What options are available for areas that don't have local cabinets, and
>>> the BT exchange can do ADSL max at best?
>>
>> If you're lucky, EFM, ok it's out of most "home" budgets as it bonds
>> four PSTN lines but it gets 10-15Mbps symmetric, rather than whatever
>> ADSL1 can manage.
>
>Interesting - that is one I had not investigated before. Prices are not
>cheap (seems like £170/month is not uncommon for 10Mb symmetrical), but
>compared with a couple of business class ADSL connections and bundled
>phone lines its not a huge jump.
>
>

I heard an ISP comment the other day that they, BT, were delaying the
roll out of Fibre to some industrial estates because why give them fibre
at 30 odd quid a month when you can supply leased lines and EFM for a
lot more moolah.....


--
Tony Sayer

[The Natural Philosopher]

that is one way to spin it. Another might be that they need symmetric
performance so staff can use VPNS for home, or upload big data to the
corporate (hosted) web site and ADSL isn't the best option. So why
invest in it when what the dues actually need is different?

FTTC is only a stopgap anyway. In the end we need full symmetric 100Mbps
or better. if ethernet over (two) twisted pairs is the next best thing
one wonders whey its not being pushed harder.

[Mike Humphrey]

The Natural Philosopher wrote:
>
> FTTC is only a stopgap anyway. In the end we need full symmetric 100Mbps
> or better. if ethernet over (two) twisted pairs is the next best thing
> one wonders whey its not being pushed harder.

Ethernet over twisted pair only goes 100m - and it needs to be a
high-quality twisted pair (at least cat 5) not phone wire. If you're
going to install fibre to within 100m of the premises, and replace the
drop wire, then you might as well just take the fibre all the way.
That's available now, Openreach will happily provide 10M or 100M
symmetrical Ethernet - but you might not like the price!

Mike

[tony sayer]

In article <t6GdnfjWyv7kY4fO...@brightview.co.uk>, Mike
Humphrey <nos...@goldberry.plus.com> scribeth thus

>The Natural Philosopher wrote:
>>
>> FTTC is only a stopgap anyway. In the end we need full symmetric 100Mbps
>> or better. if ethernet over (two) twisted pairs is the next best thing
>> one wonders whey its not being pushed harder.

Well we have VM's 30 meg service and really no complaints at all.
Richard and that runner lightning mate of his tell that its up to 60 in
the summer for the same price so can't grumble;)..

>
>Ethernet over twisted pair only goes 100m - and it needs to be a
>high-quality twisted pair (at least cat 5) not phone wire. If you're
>going to install fibre to within 100m of the premises, and replace the
>drop wire, then you might as well just take the fibre all the way.
>That's available now, Openreach will happily provide 10M or 100M
>symmetrical Ethernet - but you might not like the price!

No.. they do know how to charge for that after all why on business
estates implement FTTC when you can charge a lot more for leased
lines;?..

>
>Mike


Bin there dun just that. The 88 odd metres of CAT 5 we had in an
underground duct was shall we say, subject to rather slow running more
than anything else. Fixed up Fibre and it now goes at err the speed of
light;)...

--
Tony Sayer