Try this one...it has lots of good vectoring advantage info.

http://www.assia-inc.com/DSL-technology/DSL-knowledge-center/industry-articles/Vectoring-Compatibility-CISS-2012(Web).pdf

And I do not go to trade shows anymore, pretty much a waste of time (imho).

seven

Wrt. to the loop length - take a look at our results at 2500 ft at http://www.alcatel-lucent.com/events/2012/annual-technology-symposium/presentations/broadband-access.html

Wrt. power, the industry has come a long way. I am not going to argue AT&T's case on how old their chipsets might or might not be, or how depreciated their LTs are. These are not specific to AT&T, anyway.

I do hope to see you at BBWF (whoever you are)!

M. 

P.S. BTW, with F1/F2 I meant end of F1 (FTTN) or end of F2 (FTTC), not the boundary.

FTTN in AT&T (F1/F2) is 2500 foot loops...not a lot of vectoring value.  They are already calling out 25Mb/s but dedicate most of that to video.  To get 25Mb/s Data and 25 Mb/s Video AND not do new plant is a bigger problem.  They have 2 additional problems in AT&T.  First, they would need new linecards replacing those that are not fully depreciated (remember their VDSL chipsets are old).  Second, they have power issues in adding even more power to the cabinets (and cooling).

FTTC is shorter and better.  FTTC is at the F2/F3 point not F1/F2.  Also, the old Marconi kit is fully depreciated.  They wanted Erriccsion to replace it (which didn't happen).  So right now they can go to Tellabs and get an upgrade or do some GPON or something else.

seven

Note that much of AT&T's network is already VDSL2 FTTN (Fibre to the cabinet - F1 plant) - http://en.wikipedia.org/wiki/AT%26T_U-verse . Vectoring is something you add on top of VDSL2. Much like Belgacom in Belgium, who already had their FTTN plant before they decided to take a closer look at vectoring (http://www.belgacom.com/be-en/newsdetail/ND_20110927_alcatel_lucent.page).

Does it make sense to do construction of new active cabinets and F1/F2 fiber plant iso. FTTH? Probably not. Do it make sense to upgrade current cabinets if you have already the F1 (cabinet at SAI)/F2 (FTTC)? That would depend on the current capabilities of your VDSL2 plant, the location of your cabinets (F1/F2 and remaining distance to subs), competition and your view of (realistic) future bandwidth demand. 

The $xB question, indeed.

The US is very different here.

You would have to do significant construction to get value from vectoring in a huge portion of the network.  The exception would be the former BellSouth footprint with its FTTC bits.

Once you decide you are doing construction, the delta is am I doing just the F2 Plant or both F2 and F3, what do I get for that, and then how much time does Vectored DSL buy me.

Is it a wise investment to do fiber F2 construction, build a whole new bunch of cabinets, and then have to do FTTH in 5 years?  That is a hypothetical that needs to be answered.

seven

The question wether the vectoring innovation will delay FTTH investments appears to be relevant, indeed. But there is an implicit assumption at work here: namely that there is room and a workable investment plan for an FTTH roll-out in the first place.

If you challenge that assumption, I'd assert that vectoring, which gives people access to higher bandwidths and the associated innovation in services earlier, might even stimulate the development of such services. This would make the case of benefits to both the business, social and economical improvement that BB brings much stronger - and accelerate the next round of FTTH roll-out. 

And let's not forget that any vectoring = fiber-to-the-node roll-out contains in general less than 10% residual copper - it is indeed the first step to a fiber-rich diet.

P.S. And the Dutch, being a trading nation, have long since realized that importing good beer from Belgium is a very profitable business. No trade barriers in Europe! So that should not pose a problem either.