Print

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

[gt94sss2]

According to post on TBB:

the SNR, it can be 3dB, 4dB or 5dB. It doesn't have to jump right down to 3. It'll depend on the line. Obviously if your line is already on 10dB or higher it seems unlikely you'd ever see any benefit from this trial, DLM won't suddenly reduce the SNR to under 6 just because you happen to be in a trial area.

[Chrysalis]

thats why I asked if its 3db, as it may be possible he could be on 4 or 5db.

[niemand]

I would have expected a bigger jump in speed, seems a bit disappointing.

You are on a 3db snrm now?

I'm assuming a margin change is the cause of the increase. The timing suggests it was a configuration change of some description and certainly not been at 67Mb for about 300 other subscribers on the node. Not monitoring the modem for right now so no idea about the exact margin.

[burakkucat]

Earlier this afternoon, in a PM with Chrysalis, I mentioned that it does seem as if N*Star's circuit is operating with a 4 dB target SNRM.

Subsequently my attention was drawn to the comments made elsewhere.

Perhaps those avid MDWS watchers will make a note of which circuit appears to be operating with a 3 dB, a 4 dB and a 5 dB target SNRM? 

Do we have one or more representative cases of each of those targets? 

[NewtronStar]

Could you look at the SNRm graph at just after 00:00 to me that looks like crosstalker spikes 

[William Grimsley]

Definitely looks like crosstalkers to me, it's not that user who's been having line drop outs again?

[niemand]

I'm assuming a margin change is the cause of the increase. The timing suggests it was a configuration change of some description and certainly not been at 67Mb for about 300 other subscribers on the node. Not monitoring the modem for right now so no idea about the exact margin.

It was not a margin change it appears.

I suspect planned work was being carried out on the DSLAM and it was reloaded. The outage was somewhat long for a resync.

The increased sync would've been due to my modem coming online before some of my many disturbers (~287 of them on one DSLAM, ~200-240 on the other).

I restarted my modem earlier and not only was there no 6dB profile, DLM responded to the restart by switching off G.inp and falling back to default 8ms delay downstream interleaved profile.

No matter. Will leave it be and let it do its thing.

[Chrysalis]

this is the problem with no access to stats, lots of people will now assume any upward change of sync speed is now related to this trial.

[burakkucat]

Looking critically at the plot of N*Star's SNRM that I posted in the early hours of this morning we can see, a the far right hand end, that "something was happening". My initial thought was that perhaps another stepped reduction of the SNRM target was taking place. Clearly that was not the case.

Looking at the latest SNRM snapshot plot we can see that the target still appears to be 4 dB. (But we can't be absolutely sure without seeing the effect of a resynchronisation event.)

Having considered all the various graphs that have been plotted from the data harvested from N*Star's circuit, it does seem as if the circuit is somewhat abnormal. One to continue to watch. 

[S.Stephenson]

Surely the upload target is 4dB and the download one is 3dB,  otherwise the attainable when SNR is at 3.3 would be below the sync.

Maybe it only looks like 4dB on the download due to the service being 40/10.

[kitz]

I was able to look at NS's stats at the time it happened, at that particular moment in time, it was quite clear at the time it was 3dB on the downstream. 
If you zoom in real close you can see at that particular time the line had small bumps but was only varying by about 0.1dB.

ie prior to the resync there were small fluctuations of between 7.2 and 7.3 dB.    After the resync there were small fluctuations between 3dB and 3.1dB.

Most of his SNR variance doesn't start until 5pm but starts to rise again after 9:30.   
At the time of the resync aside from the odd 0.1db it was pretty stable and doing its usual thing. 
It does however look like the upstream is set at 4dB.

I guess a resync is in order, to double check.


----
attached
1) Zoomed in graph showing 3.1 dB and small fluctuations between 3db and 3.1dB for the downstream.
2) Zoomed in graph of downstream bands, at time of sync showing 2.9dB for D1 and 3.1dB for D2

[kitz]

Maybe it only looks like 4dB on the download due to the service being 40/10.

Like you say, he will be restricted to syncing at the max 39999, so there will be times when he goes above 3dB.


Between the hours 4am and 4pm the line is at its best, so will sync at 39999 with surplus SNRm.
The sure-fire way of finding out would be a resync at around 7/8pm when the line is at its worst. 

[NewtronStar]

it does seem as if the circuit is somewhat abnormal. One to continue to watch. 

My circuit has been abnormal for just under a year and even your good self has made comments in the past to my line stats it's like a SNRm battle against my crosstalkers.

I suspect it's all down to the badly corroded underground joint that effects all who use this distribution pole and until that has been sorted out then premises No 1 is still going to have random drop outs which effects my SNRm.

When on the 6dB profile if the SNRm went below 4.5dB it was an automatic DLM resync for my modem.   

[burakkucat]

I suspect it's all down to the badly corroded underground joint that effects all who use this distribution pole <snip>

Ah, yes. I had completely forgotten about that known poor joint. 

[NewtronStar]

Between the hours 4am and 4pm the line is at its best, so will sync at 39999 with surplus SNRm.
The sure-fire way of finding out would be a resync at around 7/8pm when the line is at its worst.

Would you like a quick HG612 resync or a power off and wait 30 mins & stick the 12v plug back in  think I can afford 1 MTBR, 21:00 looks like the lowest SNRm point at the moment