Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Mick]

Hi All,

I noticed that suddenly my SNRM increased from 6dB to 7dB, yesterday morning.  At the same time the current sync speed shown by my router also increased (see attached graphs).

I wasn't at home to resync and see what I get at the time, but what would you say was the reason of this sudden increase in performance?  Today all is back to normal.

PS.  The readings are from a 2Wire router operating in bridged mode.  The green coloured area represents initial values obtained when I sync'ed some days ago.  The blue line represents current values.  I suspect that the sync speed is derived via some algorithm from the SNRM measured by the router.

[attachment deleted by admin]

[jeffbb]

Hi
Which is your ISP?.
This continuously changing Synch rate implies that you connection supports  Seamless Rate Adaption (SRA)

see SRA 

Regards Jeff

[Mick]

Thanks Jeff,

That would be nice, but I am still on 20CN.  I don't think that our exchange is in line for an upgrade until late 2011 if not later. 

[jeffbb]

Hi
Well then your synch rate changing like that is unusual .You would only expect to see Synch rate change after a reconnection to the exchange . This could be after losing synch or rarely initiated by the exchange when the target SNR margin changes. Neither of those 2 scenarios apply to your connection as the synch rate is continually changing .
Oh well somebody may know why ? something new every day

Regards Jeff

[Mick]

I think I may have confused matters with the title of my post ... 

As I explained in my first post the value of the green area of my sync speed graph shows the initial sync speed.  This won't change unless I resync.  In the attached router stats the initial sync value is shown as 7712kb/s.  That's the green area.  The blue line is the value described as Max2 in the stats and I think is a derived value (calculated by the router) based on the SNRM at the time.  This value goes up and down depending on the real time changes in the SNRM.

So, when during the morning the SNRM jumped up to 7dB, the corresponding Max2 value also increased as shown in the graph.  (Note that although the initial sync the router obtained was at 7712, the sync rate shown in the stats is 7616 because I think that the maximum I can get with interleaving On.)

I am thinking now that the jump happened because the cross talk I have on my line went away during the morning (because e.g. a neighbour switched off their mains) but I can't of course be sure.  Could it be anything else?

[attachment deleted by admin]

[roseway]

You can't read too much into that 1 dB step in the reported noise margin. 2-Wire routers only report the noise margin in increments of 1 dB, so all it may mean is that the noise margin was (say) 6.4 dB rounded down to 6 dB, and now it's 6.6 dB rounded up to 7 dB. The noise margin will vary by small amounts all the time, and this sort of apparent jump is the sort of thing you see with this type of router.

[jeffbb]

Hi
oops misread the graphs 
senile moment
Regards Jeff

[Mick]

You can't read too much into that 1 dB step in the reported noise margin. 2-Wire routers only report the noise margin in increments of 1 dB, ...

Of course!  I forgot that what I am looking at in these graphs is a step up/down in (relatively) large increments, which are averaged over 5 minutes by mrtg.  Also, I am using vanilla rateup as a backend in mrtg, which it rounds down to the
integer rate per second.  So, the step up that I saw and rubbed my eyes thinking someone switched off my Interleave setting at the exchange    was merely a reflection of my measuring tool and its lack of granularity.

Thank you both for helping me think this through.