Kitz ADSL Broadband Information
adsl spacer  
Support this site
Home Broadband ISPs Tech Routers Wiki Forum
 
     
   Compare ISP   Rate your ISP
   Glossary   Glossary
 
Please login or register.

Login with username, password and session length
Advanced search  

News:

Pages: 1 ... 5 6 [7] 8 9 ... 21

Author Topic: 3dB Target SNRm Live?  (Read 76959 times)

NewtronStar

  • Kitizen
  • ****
  • Posts: 4898
Re: 3dB Target SNRm Live?
« Reply #90 on: April 19, 2016, 12:47:58 PM »

How long was it between the automatic resync, and the subsequent manual one you performed?

It was up for 3 days 20 hours, should have left the modem alone until an automatic resync was initiated by DLM.

I would have thought a trial would last much longer that this if you were selected it should take into account manual resyncs be it the user or DLM actions or even say power failures, if that is the way Openreach conduct their trials then they are not getting the full picture on end-users lines.
« Last Edit: April 19, 2016, 05:13:43 PM by NewtronStar »
Logged

garypower

  • Member
  • **
  • Posts: 16
Re: 3dB Target SNRm Live?
« Reply #91 on: June 08, 2016, 09:47:47 AM »

Looks like I may have been moved onto 3db profile - I wonder if it is still a short term trial or whether it will stay... time will tell.

My line with uptime of > 90 days dropped last night for over 5 minutes - dsl light went off on the modem.
It came back as per attachment.

Logged

Ronski

  • Helpful
  • Kitizen
  • *
  • Posts: 4302
Re: 3dB Target SNRm Live?
« Reply #92 on: June 08, 2016, 10:17:28 AM »

My works connection resynced at 1am this morning,  but the downstream synced at 5db, which gives me the max of 40 Mbps for the package we're on, upstream stayed at 6db but the speed improved. Stats are on MDWS under Ronskiwork.

No idea of this is related,  but just seems odd to sync at 5 instead of 6.
« Last Edit: June 08, 2016, 01:15:39 PM by Ronski »
Logged
Formerly restrained by ECI and ali,  now surfing along at 550/52  ;D

roseway

  • Administrator
  • Senior Kitizen
  • *
  • Posts: 43573
  • Penguins CAN fly
    • DSLstats
Re: 3dB Target SNRm Live?
« Reply #93 on: June 08, 2016, 11:40:04 AM »

I presume that was a typo in the last sentence. One possibility is that a 3 dB target has been applied and it ended up with 5 dB because it's capped at 40 Mbps.
Logged
  Eric

Balb0wa

  • Member
  • **
  • Posts: 44
Re: 3dB Target SNRm Live?
« Reply #94 on: June 08, 2016, 11:42:33 AM »

Those who are on the 3db, are you all on the HG612?
Logged

garypower

  • Member
  • **
  • Posts: 16
Re: 3dB Target SNRm Live?
« Reply #95 on: June 08, 2016, 11:55:23 AM »

I have an HG612.
Logged

WWWombat

  • Kitizen
  • ****
  • Posts: 1674
Re: 3dB Target SNRm Live?
« Reply #96 on: June 08, 2016, 12:42:58 PM »

Looks like I may have been moved onto 3db profile - I wonder if it is still a short term trial or whether it will stay... time will tell.

It looks like the downstream target has been set to below 6dB, for sure. Below 5dB is very likely. Hard to tell whether it has been set to 3dB or 4dB, when current SNRM is running at 4.5dB.

The reason it is hard to tell? The downstream speed has risen to 60Mbps, with an attainable of 63Mbps. That difference of 3Mbps perhaps represents just 1dB of margin ... suggesting a real target of 3.5dB. For reference, your actual speed rose by roughly 3Mbps too, and actual SNRM has dropped from 5.5dB before the resync to 4.5dB after the resync - a further confirmation.

The downstream speed has risen to exactly 60Mbps, and the upstream to a whisker shy of 17Mbps. It looks like you have been banded at speeds of 60/17 ... but it is impossible to know when those were put in place.

Upstream, however, does look like a 3dB target has been used. Your actual/banded speed is still well short of the attainable speed, but the picture gets muddied on upstream by the dynamic upstream power-backoff mechanism.

Now to look at the effect of this...
The underlying FEC rate seems to have increased, but there seems to have been very little change in the impact to retransmissions or to CRCs. Perhaps we shouldn't be surprised ... even though the target SNRM might have changed by 3dB, other factors mean your actual SNRM has only changed by 1dB.

NB:
You had a big burst of CRCs in the afternoon before the change, surrounded by a longer period of FECs. There was some noise source around then that doesn't appear in the days beforehand. Whatever happened, it doesn't look like it was enough to trigger a DLM intervention of any form.
Logged

Ronski

  • Helpful
  • Kitizen
  • *
  • Posts: 4302
Re: 3dB Target SNRm Live?
« Reply #97 on: June 08, 2016, 01:22:19 PM »

I presume that was a typo in the last sentence. One possibility is that a 3 dB target has been applied and it ended up with 5 dB because it's capped at 40 Mbps.

Yes, it was a typo, corrected now thanks. Could well be, attainable is 47, but we are interleaved.

I'm using an HG612 here at work.
Logged
Formerly restrained by ECI and ali,  now surfing along at 550/52  ;D

WWWombat

  • Kitizen
  • ****
  • Posts: 1674
Re: 3dB Target SNRm Live?
« Reply #98 on: June 08, 2016, 02:29:59 PM »

My works connection resynced at 1am this morning,  but the downstream synced at 5db, which gives me the max of 40 Mbps for the package we're on, upstream stayed at 6db but the speed improved. Stats are on MDWS under Ronskiwork.

Most of what I wrote for @garypower applies to you too, except you don't appear to have banding - but the package limit of 40Mbps cuts in instead, and has the same effect of a higher SNRM than target.

If a 3dB target was being used, then the 2dB you have "spare" ought to be worth around 4-5Mbps, by my rule-of-thumb reckoning. However, your attainable is currently 8Mbps above the actual speed. I think the difference comes down to the fact old-style FEC+interleaving is in place, where the attainable is artificially high.

In terms of errors, your line is reacting in a similar way too: More FECs, but not notably more CRCs or ES's. But, strangely, I see the same thing on your line as @garypowers from yesterday: a bigger burst of FECs and CRCs. A coincidence? Some kind of cause/effect going on?
Logged

garypower

  • Member
  • **
  • Posts: 16
Re: 3dB Target SNRm Live?
« Reply #99 on: June 08, 2016, 05:27:39 PM »


NB:
You had a big burst of CRCs in the afternoon before the change, surrounded by a longer period of FECs. There was some noise source around then that doesn't appear in the days beforehand. Whatever happened, it doesn't look like it was enough to trigger a DLM intervention of any form.

Thanks for the analysis - the burst of CRC's seem to coincide with the time of large thunder / lightening storms and lots of rain.
I can not think of anything else running in the house that is not usually on.
Logged

Ronski

  • Helpful
  • Kitizen
  • *
  • Posts: 4302
Re: 3dB Target SNRm Live?
« Reply #100 on: June 08, 2016, 07:58:06 PM »

There was also thunderstorms near us yesterday to.
Logged
Formerly restrained by ECI and ali,  now surfing along at 550/52  ;D

jasjeet

  • Member
  • **
  • Posts: 86
Re: 3dB Target SNRm Live?
« Reply #101 on: June 08, 2016, 08:19:58 PM »

HG612
Seems like the upstream maybe not on a 6dB profile anymore?
Logged

Ronski

  • Helpful
  • Kitizen
  • *
  • Posts: 4302
Re: 3dB Target SNRm Live?
« Reply #102 on: June 08, 2016, 09:25:01 PM »

Considering your uptime is 54 days, it's quite possible that the SNR has just drifted down overtime.

Edit. Although 54 days does take us back to the beginning of April, which is when NS started this thread, really need to know what your SNR was at time of resync.
« Last Edit: June 08, 2016, 09:27:06 PM by Ronski »
Logged
Formerly restrained by ECI and ali,  now surfing along at 550/52  ;D

garypower

  • Member
  • **
  • Posts: 16
Re: 3dB Target SNRm Live?
« Reply #103 on: June 09, 2016, 01:48:22 PM »

The upstream SNRM seems to have settled on 3db now - the downstream still at 4.3
I assume this is because of what looks like some banding set from a previous time.

I seem to remember 60000, 16999 from a long time ago before crosstalk dragged me well under these figures.

I wonder if this banding will remove itself at some point as the upstream looks like it can take some more and the downstream as well if it was allowed to go to 3db.

[Moderator edited to fix the typo. (upstream --> downstream)]
« Last Edit: June 09, 2016, 04:57:43 PM by burakkucat »
Logged

WWWombat

  • Kitizen
  • ****
  • Posts: 1674
Re: 3dB Target SNRm Live?
« Reply #104 on: June 15, 2016, 04:15:04 PM »

A week on, I've just gone back to check @garypower and @Ronskiwork's lines.

Both appear to be continuing as they were set up back then, with error rates continuing as was visible initially. Things look reasonable...
Logged
Pages: 1 ... 5 6 [7] 8 9 ... 21
 

anything