IP and actual speed?

[#1]

Hi Jojo
Many thanks for the info as the Gateway Tool is still "puzzled"  
For the record so far the measured speed has not increased!  The Margin is currently 3.8db, so I still remain kind of puzzled myself as to how it selected 3db band and holds it so nicely but the measured speed is stubbornly that measurably bit lower than when I was sync'ed at a siimlar speed but with a Margin in the 6db bracket and that was Fast Path compared to now Interleaved Path (on the DL only).
Is it worth a reset of the Margin to 6db by PlusNet to readjust the and expect the DLM to "see" the wisdom of the change  
dick:quote

[#2]

Wisdom, lol. I don;t think it would help to be honest. More likely to lower the speed. The sync is 3755 at the moment. What have you seen in the past?

[#3]

Technology & the ghost in the machine stories come to mind 
Off the top of my head when at 6db I used to sync >3500 (e.g. closer to the current 3755 ~ though would need to look back over older records/forum posts et al for the exact figures) and the measured speed was typically >3Mbps.  Having said that at its worst the sync was 3100 to 3200 and the measured speed was well under 3Mpbs
So I do take your point but latterly with the 3db margin I was sync'ed at 3801 and even at that speed I could not get measured speed of >3Mbps ~ yes 21CN has meant that the changes are more fluid but still every bit of speed I can get is worthwhile IMO 
Heh ho, roll on to one day moving house to a FTTC location or at the very least a shorter copper line???

[#4]

Bear in mind that although your target is 3dB, at the time of measuring your sync the actual SNR was 3.8dB, so that will shave a bit off

[#5]

Ah! but when I sync'ed at 3755 I am quite sure that the margin was way closer to 3db i.e. just checked the logs and it was floating between 2.9 and 3.1db on the 30th June start of 3755

[#6]

Well, it's ~50k so that's easily attributed to small changes like that

[#7]

The fact that your margin is currently ~3.6 to 3.8 dB is a good thing in the sense that it will help to reduce the potential errors that may otherwise give you an even lower throughput.
It's possible or even probable that your "environment" has got more noisy than it used to be so given a slightly slower speed for the same margin as well as increasing the errors a bit.

[#8]

Hmmm! environment getting (electrically) noisier???
Could it also be argued that connection kit degrades?  By that I mean I use an iPlate and of course a filter, so it it logical maybe to replace the iPlate with a new one and indeed the filter as well every so often.  I think I swapped filters approx 2 years ago and I am unsure just how long the iPlate has been in place
FWIW before the iPlate I had an ADSL Nation filtered faceplate, still got that in the cupboard somewhere!

[#9]

Just for the record

Ookla says
Down 3.11Mbps
Up      0.38Mbps
With a Ping of 38ms
DL
Noise Margin:     3.6   dB
Connection Rate:  3755  Kbps
Line Attenuation: 56.0  dB
Power:            0.0   dBm
Max Rate:         4732  Kbps

SuperFrames:      31412983
SF (CRC) Errors:  1674
Reed Solomon:     1020921966
RS Corrected:     27781257
RS Un-Corrected:  17190
HEC:              1538
Errored Seconds:  721
Severe ES:        5
Interleave Depth: 16
Bitswaps:         14990
Uptime
Total Uptimes (From SF counts):
WAN:   6 days, 04:21:34
LAN:   6 days, 04:21:34


CRC:  1674
LOS:  0
LOF:  0
ES :  721

[#10]

Those stats don't look too bad at all actually.
Should be no need to change i-plate unless you get struck by lightning, filter should be fine but that won't affect download speed directly. When they go wrong you see lots of other problems like sync speed problems, use of phone causes drops etc.
When I say your environment has possibly got noisier, I mean more electrical gadgets around, not just your house, neighbours etc., more people with broadband, you get a bit of noise from others in the same cable to the cab/exchange - all that sort of thing.