asbokid has graphed whatever it was I sent him and produced this:
http://img846.imageshack.us/img846/3836/statsop.jpg
Now i'm not sure what I'm looking at but does it appear a bit strange to anyone else? Seems the first image roughly correlates in shape to the 2nd one, except for the big "gash" in it.
I'm just trying to get my head around it.
If you really want to make
your head hurt there's a copy of ITU spec G993.2 (aka VDSL)
online.
BT are using VDSL profile 8c, annex B and according to
this document band plan 997 truncated at 7.05MHz.
This uses tones 32-695 for the 1st downstream band, 696 to 1182 for upstream, then 1183 to 1634 downstream again.
The modem reports what tones it's using with "xdslcmd info --show"
Discovery Phase (Initial) Band Plan
US: (0,95) (696,1183)
DS: (32,687) (1192,1627)
Medley Phase (Final) Band Plan
US: (0,95) (696,1183)
DS: (32,687) (1192,1627)
What I don't quite understand is what the initial US: 0,95) is about as it overlaps the (32, 687) interval - there is the US0 band (i.e same part of the spectrum as ADSL upstream) but I don't know what BT are doing with that (I don't think they support it though). Note that using US0 is needed for supporting long lines otherwise the upstream tones (between 3MHz & 5.1MHz) are going to get lost due to attenuation. If BT aren't using the US0 band then that would be one reason why FTTC lines will not "degrade to ADSL performance" on very long lines.
I also don't understand why the last upstream tone is 1183 as the ITU spec says 1182.
Finally - why the gaps? 688-695 appear not to be used, nor 1184-1191, maybe the modem decided it didn't need them to attain the required data rate??
So the graphs. I think asbokid has already said but they are 1) line SNR (in dB i presume) per tone but the "gash" is the upstream tones where the modem doesn't report a SNR. Note
not SNRM.
2) Tone attenuation (or rather negative log of) per tone, again gap for the U/S tones.
3) Line noise. In the higher tones this nicely mirrors the SNR as you'd expect. In the lower tone it doesn't - I suspect this is due to spectral power management.