In message <(E-Mail Removed). com>,
ale.cx <(E-Mail Removed)> writes
>
>Paul wrote:
>
>> I have had Max for about 4 months, been working very well, I could hold
>> around 6Mb/s sync with SNR margin down to 5 or 6 dB at times. The
>> attenuation is 45dB and constant as it should be. Download speed was
>> just over 5 Mb/s, all was going well.
>
>What was the attenuation? Has it changed recently?
No still the same.
>
>> The problem now is my SNR Margin has jumped to at least 12dB when I
>> reconnect (it used to be 6dB) and this has killed my sync speed and as a
>> result of that also my download speed.
>
>The margin is the difference between the minimum signal level you need
>for a given sync speed, and what you currently have. So the bigger the
>difference, the better. [lightbulb goes on in alexd's head] So I expect
>what has happened is that you've been dropped to a lower sync speed, in
>an attempt by the DSLAM to get you a more reliable connection. So, the
>minimum signal you need for the speed you are currently on is lower
>than the previous speed you were on, but the actual signal strength
>hasn't changed. Hence, your margin doubles, and you have a slower but
>theoretically more reliable connection. All makes sense when you think
>about it.
I see how that works, though my connection seemed OK before, download
speed at 5Mb/s and online gaming no problems, (low ping no packet loss).
>
>> I assume that it's the exchange equipment that has done this, as I have
>> tried a different router and it's the same high SNR Margin (12dB).
>
>Yep.
>
>> Prior to this problem I did have some random tripping of my house RCCB
>> for a week or so (now fixed), would that have made the exchange think I
>> have a bad connection as my router is not on my UPS (as I have moved it
>> to the master socket)?
>>
>> If that is the problem will it improve and be changed back to the lower
>> SNR Margin automatically now my connection is not dropping due to power
>> cuts.
>
>Hopefully.
Here's hoping then

>
>> Or could the problem be something else?
>
>The problem could be something else. Unplugged that ancient fax machine
>that was injecting horrific quantities RF noise into the line, from
>it's faulty switched-mode PSU?
Good point I will have to have a listen and check that out, off to find
MW radio and HF receiver.
>
>
>A sidenote: Some people question the veracity of any of the noise- and
>signal-related figures given by economically-made routers from the Far
>East, citing lack of calibration and whatnot; the suggestion is often
>made that only data from properly calibrated E-side kit can be taken
>with any degree of seriousity. If that's the case, it would be nice to
>be able for the consumer to see live data from the DSLAM [or wherever
>the stats come from] they're connected to. Perhaps any port 80 HTTP
>requests to the IP address of my-line-stats.bt.net could be intercepted
>by exchange equipment, with the requisite info delivered on an HTML
>page. Of course, it would only be of any interest to geeks like me, so
>it's not likely it's ever going to happen. Oh well, life goes on.
That would be very useful for fault finding, certainly more accurate
than the router. My router by the way is a Netgear 834pn and have been
plotting the stats with the RouterStats program.
>
>alexd
>
Cheers thanks for the reply
--
Paul