nav search
Data Center Software Security Transformation DevOps Business Personal Tech Science Emergent Tech Bootnotes BOFH

back to article
DSL inventor's latest science project: terabit speeds over copper

Silver badge

It would be an interesting development

If it turned out those tired old twisted pair telco lines could allow for higher bandwidth than the cable company's HFC network. It would be funny if Verizon regretted selling off all their copper...

1
0
Silver badge

Actually no

This is a stopgap solution for very short lines, and this, if it will even work in practice, will only work for very short distances and with very complex equipment. You already need high double digit gigabit speeds in the backplane of a vectoring DSLAM, not for the data itself, but for all the data you need to send around to eliminate cross talk.

On the other hand, once you have a fibre, it's cheap to run 1 Gigabit over it. The lasers and detectors for that costs about 20 Euros per side. For 10 Gigabits that goes up to 50 Euros. Both optics can plug into bog standard Ethernet equipment. And for fibre the limit now actually is how to get the data away from the optics fast enough. So far only 100 Gigabit Ethernet uses QPSK which is about at the level of sophistication needed for 9600 bps modems.

11
0

Re: Actually no

@stopgap

And if you have to replace the copper with specially shielded, break free new lines, why would you not go fibre already? (I presume my rat eaten phone line, patched in numerous places would be unsuitable!).

9
0

Re: Actually no

I think 100Gb is a bit more than a 9600b modem - needs optical coherent detectors and dual polarisation, adding DWDM can get one to 10 - 20 Tb per fibre. However the ironic twist is that 40 years ago ( before the first public trial of optical fibre transmission in the UK that achieved 140Gb over 12 miles) the proposed solution for long haul high rate transmission was a waveguide made by lining a tube with copper wire. Plus ca change ...

4
0
Silver badge

Re: Actually no

But would it work in Milton Keynes? (Ally cables.)

Seriously, does this waveguide effect require a decent smooth bit of copper?

What about all the dinks in the conductor caused by crimps and IDC terminals in all those joints?

Security / crosstalk?

Etc.

Sounds a nice idea but I'd like to see some real world test results first.

2
0
Def
Silver badge

Re: Actually no

adding DWDM can get one to 10 - 20 Tb per fibre

That's pretty awesome in my book though. I remember reading about the new fibre based Internet backbone being laid through Europe back in the early noughties that managed 40Gb per fibre through multiplexing eight (I think, it could have been five) channels over a single fibre. The network itself was composed of 160 fibres, I believe.

The current record is apparently held by Nippon Telegraph and Telephone Corporation who transferred 1 Petabit per second over 50 kilometers over a single fibre in 2012.

I have two fibres coming into my house (one is for the apartment downstairs), but I can't justify upgrading my current 80/80 connection - I really don't need anything faster at the moment.

0
0
Silver badge

Exponential Drop off

you get to a Terabit system that Cioffi reckons can operate at 100 metres; 100 Gbps at 300 metres; and 10 Gbps at 500 metres

and at 2500 metres you get ?

I wonder what the "average" local loop distance is in Australia... and what "quality" of copper you need compared to the existing services.

2
0
Silver badge

Re: Exponential Drop off

It's a stopgap solution, designed to run those speeds over short runs of bad copper. The only usecase for this is inhouse cabling on old buildings, where you put a DSLAM in the cellar and bridge the last metres of copper. It's rather idiotic as running fibres will cost _much_ less than such a DSLAM.

2
0
Silver badge

Re: Exponential Drop off

The annoying thing is the technology is available to make low power gigabit repeaters for a couple of quid that could be cheaply dropped in every couple of Km on copper anyway.

But then BT had the capability 27 years ago to put in 10km of fibre and have 2.4Gb end to end for an equipment cost less than £100 per line but they 'didnt do research' anymore.

0
0
Silver badge

Pointless excercise

Diminishing returns continue.

Can we just, finally, let it die and put fiber in the ground.

7
1
Silver badge

Re: Pointless excercise

>Can we just, finally, let it die and put fiber in the ground.

Is it just the Oz press getting the wrong end of the stick based on DSL being a long(ish) range tech and this is by the same person?

Surely this is more for "inside-the-data-centre" where the exponential drop-off isn't an issue?

3
0
Silver badge
Boffin

Interesting stuff. I wonder if it fits in with the document that Ofcom published a couple of years ago? I think it was a (very theoretical) review of how far twisted pair could be taken in the future.

0
0

TB speeds, until BT/OpenReach/Virgin get their hands on it and you will be lucky to get dial up at daft prices !

3
0
Silver badge

Magic ...

It's all smoke and mirrors I tell you. When I were a lad, you needed waveguide for a 1GHz signal, and then they made computery things that ran at 100MHz and didn't swamp Radio 1 with noise - though I admit it was hard to tell the difference ...

4
0
Silver badge

Don't tell BT...

...they just use it as an excuse to not lay fibre to the home....

1
0

Re: Don't tell BT...

They tend not to do FTTP anyhow, mostly FTTC. FTTP is a major pain in the arse anyhow, the BT guy comes out and sticks the modem on the wall, then sticks on a battery backup for Fibre Phone which doesn't exist, but they stick the battery on anyhow and then nag at you when the battery backup breaks and they have to come out and replace it. "But you don't need it, so why replace it?" "you stick it in, you replace it if it breaks".

Then there is the ISP's that support FTTP, is it 4 or 5? And then they only give you FTTC speeds anyhow, the only bonus being you actually get the full fat 80Mb.

0
0
Silver badge

Re: Don't tell BT...

Then there is the ISP's that support FTTP

I think you will find that the ISP's either require a BT FTTP line, because they are in dispute with Ofcom over who should be supplying the battery backup, specifically they want BT to provide the battery for the FTTP line they installed and for which BT doesn't get any service fee's from, or they expect the house to have the BT POTS over copper service for fixed line telephony (which effectively leaves the power/battery issue with BT but this time BT do get some revenue).

FTTP is occurring more on recent larger new build developments (ie. developments of 20+ homes built in 2016 and later). I don't know if the cable being used is pure fibre or also has copper cores; it would make sense for it to be pure fibre, but then copper cores can aid fault location and diagnosis.

0
0

Showing my ignorance probably..

but I wonder how (When it eventually arrives) the much slower signal that goes down the center of the wire doesn't interfere with the data then being received from the edge?

0
0
Def
Silver badge

Re: Showing my ignorance probably..

Why would you still need the slower signal? You might keep the cable, but the equipment at both ends would change, surely.

0
0

Re: Showing my ignorance probably..

"Why would you still need the slower signal? "

You don't need it and you don't want it, but wouldn't it be there and interfere with the faster signal. I guess they must have a way of filtering it out.

0
0
Def
Silver badge

Re: Showing my ignorance probably..

No, it wouldn't be there. That was my point about changing the equipment at both ends. The new equipment wouldn't support the older signal.

1
0

This post has been deleted by its author

Silver badge

"Energy still leaks off wire if bent"

Typical telco wires go through several junctions that are one big rat's nest of crushed and untwisted wires. Even with a neighborhood distribution box, there's at least one more junction mess on the poles and a few more in the house. Each tiny disruption to the wire pair is an impedance mismatch, an antenna, and a low pass filter. This is why DSL is slow and this is why telcos can't just send plain old Ethernet down the wires.

Just replace it already.

0
0
Silver badge

This of course begs the Question: How much bandwidth do you need at home?

And as anyone knows, the answer is: More than I have now.

With all the point to point connections demanding bigger and bigger hoses for the data, consumers have demanded more and more data through inefficient connections. At one time we had "Broadcast" technology which if you summed up how much data was consumed by individuals would get an enormous number (sure it was all the same data, but that is what BROADcast means).

Now we have all these point to point links between providers and consumers that want anything and everything on an on-demand NOW basis. Satisfying the bandwidth habit will only get more expensive over time as "replace" happens over and over again.

Sad but most likely true. (*SIGH*)

2
0
Silver badge

Re: This of course begs the Question: How much bandwidth do you need at home?

It seems to be a zero sum game. At home I have 2Mb which is not enough to watch video over but is more than adequate to browse the web except where people fill page with unnecessary and uncompressed images by the thousand.

If you take into account what I ignore the real data rate is as fast as I can read most of the time, its just the noise level has gone up since 2400 baud dial up.

Upgrades can be scheduled to run overnight with minimal effort and caching proxies reduce my bandwidth volumes by the eight or ten machines I'm running regularly. Intelligent downloading of TV is possible - things I miss on BBC are get-iplayered overnight and if tv services did the same for box sets and series I'd guess the average family could live on less than 20Mb - people in my local village who have 70Mb say they never needed the fibre upgrade from 17Mb - apart from one mate who has more video than he can watch in his lifetime already.

1
0

Speed is of the essence, but obscene profits are much nicer.

Wouldn't matter if it could transmit 100 terabytes over infinite distances, it would never get implemented in the USA as the business model here is to minimize competition, and sell the minimum capacity at the maximum cost. I expect to see the merger of several wireless broadband suppliers in the near future, as T-Mobile has forced the cost of wireless data to slide, messing up AT$T and Verizon's decade long price fixing monopoly. why sell terabyte speed when you can force the end-user into accepting megabyte speed because the end-user has no choice?

The cable companies have used this model for decades and made very healthy profits, with little innovation, until Google fiber came along (of course, google is now getting out of the business) and forced higher speeds.

It's ever so much more profitable when you can create an artificial shortage.

OPEC showed them how it could create a viable, profitable business model based on a perceived shortage, and milk the Rubes for everything they could, all with governments covert approval.

Terabyte speed? Not gonna happen, over copper, over fiber, or over the air.Just pay your $10.00 per gig and be silent, (and Smile!) you shall be happy to accept our domination!

1
0
Silver badge
Happy

One thing for sure, copper with paper insulation and lead sheathing won't cut it!

There must still be hundreds of miles (even more kilometres) of the former Post Office TelephoneS (POTS) cabling lurking in Great Britain.

As a new fibre optic subscriber, with proper government regulation with respect to sharing, has to be better than trying to resuscitate an old technology. Apart from this there are environmental and health considerations to be considered these days.

China, and Indochina, have invested heavily in fibre optics with great success. Cambodia, a circular shaped country, has brought modern services to formerly isolated areas. VietNam, an elongated shaped country has almost all forms of communication using fibre, from TV programming, long-haul cell links between cell sites, etc.

All services (supports pure data, Data +Voice, Data+ Voice+ WiFi, etc. and can provide up to 1.25Gbps uplink and 2.5Gbps downlink transfer rates) are available. I chose TWO 100Mbs InterNet services from different suppliers emanate from a single modem attached to a single fibre optic.

The vertical 'house wiring' fibre optic that feeds each floor of 20 apartments, plus five spares, has multi-core feeds all in an external diameter of that of a lead pencil.

Putting old technology on life support will be for all the wrong reasons - many financial (which don't reduce consumer installation charges).

0
0
Anonymous Coward

Sigh

Another excuse for BT to refuse to deprecate their obsolete copper network...

0
0

POST COMMENT House rules

Not a member of The Register? Create a new account here.

  • Enter your comment

  • Add an icon

The Register - Independent news and views for the tech community. Part of Situation Publishing