Is there a way of predicting FTTC speed based on distance from the cabinet?

[klipp]

I'm 0.8km away from my cabinet by road, or 0.5km as the crow flies.  Does anyone have similar distances who might know what that would equate to on 'up to 40' or 'up to 80' services?  I can't check online as my exchange isn't going live until next year lol.

Thanks.

[andrue]

I'm 0.8km away from my cabinet by road, or 0.5km as the crow flies.  Does anyone have similar distances who might know what that would equate to on 'up to 40' or 'up to 80' services?  I can't check online as my exchange isn't going live until next year lol.

Thanks.

http://www.buckconsult.co.uk/fttx/graphvdsl2.png

That's the 'best' graph I've found so far. It would suggest you'll sync at 50Mb/s, maybe 60Mb/s. Distance isn't the actual cause of the slow down though. It's just a useful guideline assuming a typical UK telephone line.

[klipp]

Oh thanks andrue.  What is the cause, if not the distance?

[Steve]

Certainly a signal degrades over a longer distance due to the resistance  of copper wire,plus a longer line allows more opportunities for interference and crosstalk, both of which reduce performance.

[Lance]

Quality of the line is important to, along with the number of junctions on the way. I've got a long bit good quality line which way outperforms my inlaws with their shorter but poor quality line.

[klipp]

Thank you.

[psp83]

Also, the distance to the cab (as you would walk it) does not mean your line is that long.

BT seem to take the longest route, like mine.

If I walk to the cab, its about 350m - 400m, but my line goes the other way round which adds another 150m (ish) on top.

Bloody BT 

[andrue]

Oh thanks andrue.  What is the cause, if not the distance?

What I meant as others have said is that distance is an indicator. The signal from the cabinet degrades as it travels down the cable (as does ADSL from the exchange). The rate at which it degrades depends on the cable quality including any junctions that the signal has to pass through. UK telephone cable is fairly standard gauge I think aside from some unfortunate souls with aluminium and possibly also the Hull area.

This means that it's reasonable to assume a particular loss rate per metre. That in turn allows you to guess at the affect on speed purely by using cable length.

As to how it actually works: Higher frequencies suffer attenuation more than lower frequencies. You can see this with radio waves where FM doesn't reach as far as MW which doesn't reach as far as LW. xDSL technologies work by pretending to consist of multiple modems - many hundred in fact. Each 'modem'(*) is assigned its own frequency. Those assigned the highest frequencies are the first to fade followed by gradually more and more frequencies. You can see the effect if your modem is unlocked and you plot a graph. There's some examples here:

http://www.geekzone.co.nz/forums.asp?forumid=49&topicid=105693

And more info here:

http://www.kitz.co.uk/adsl/adsl_technology.htm

(*)As far as I'm aware there aren't actually hundreds of modems in the box. Clever mathematics is used to 'pretend' that there are or to achieve the same effect.

[Bald_Eagle1]

As to how it actually works: Higher frequencies suffer attenuation more than lower frequencies. You can see this with radio waves where FM doesn't reach as far as MW which doesn't reach as far as LW. xDSL technologies work by pretending to consist of multiple modems - many hundred in fact. Each 'modem'(*) is assigned its own frequency. Those assigned the highest frequencies are the first to fade followed by gradually more and more frequencies. You can see the effect if your modem is unlocked and you plot a graph. There's some examples here:

(*)As far as I'm aware there aren't actually hundreds of modems in the box. Clever mathematics is used to 'pretend' that there are or to achieve the same effect.

& I have attached a FTTC example from my connection that is between 850m & 1000m from the cabinet, depending which actual route it takes.

FTTC as used in the UK (Profile 17a) splits downstream into 3 frequency/tone bands.

Huawei & ECI DSLAMS use very slightly different tones, but the priinciple is the same.

You will see from my graphs that attenuation at the highest frequency band (D3) for me to achieve any bit-loading at all & not that much from the middle band.

The best my connection has synced at is 35Mb downstream & is usually around 33Mb, now that the physical line problems have been fixed.
These speeds are apparently quite good for that distance but I have seen one or two with higher sync speeds over longer distances. These would be the exception rather than the rule though.