Print

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

[NewtronStar]

I turned of the HG612 last night for 15 minutes and noticed the DS INP: changed from 46:00 to 49:00 what sort of change will that do to my line ?

[adslmax]

U should have leave it powered off for at least 30 minutes not 15 minutes

[tommy45]

As long as you don't repeatedly  power off/on (re boot) the modem you should be fine, even if the connection is on the most stringent of the 3 DLM stability profiles (in theory at least) i would see the shift from 46 to 49 on the inp  wouldn't amount to much but as little is known it's just a guess

[Ixel]

It might be that re-syncing can result in a slightly different INP each time.

[jid]

I turned of the HG612 last night for 15 minutes and noticed the DS INP: changed from 46:00 to 49:00 what sort of change will that do to my line ?

May be that on resync the G.INP algorithm calculates the most suitable correction rates for the line.

Obviously all an assumption 

[AArdvark]

I found that DLM will keep re-syncing your line adjusting the DS / US INP over a period of days until it decides you are at the optimum settings. (For now  it is DLM.)

I had 2 or 3  re-syncs then my line has been stable for 18+ days. i.e. no more resyncs.

Still waiting for the line sync to recover back to the sync rate I was at, but more and more it is looking like there is a profile of 67/20 that people are getting stuck at even if they previously had syncs higher than 67M.

Would love to know if this is true ?

[tommy45]

Hasn't re-synced mine yet  but it may do if several steps have been taken by dlm in the past

[NewtronStar]

May be that on resync the G.INP algorithm calculates the most suitable correction rates for the line.

Obviously all an assumption 

I do remember when this line was interleaved the depth would change quite often after a reboot and depending on the current sync rate IE: with a higher sync rate the more interleaving was applied and less for a lower sync rate.

But I guess it depends on the amount of noise there is on the line during the re-sync.

[Balb0wa]

i was g inped 5 hrs ago, funny time indeed. My ping is a lot better, knocked 8ms off to the bbc  16-17ms now , not bad for a long line

Before



After

[danivtec]

i was g inped 5 hrs ago, funny time indeed. My ping is a lot better, knocked 8ms off to the bbc  16-17ms now , not bad for a long line

Before



After

Off topic but how are you getting 4db for the DS Snr?

[Balb0wa]

no idea, probably dropping as its night, my line is pretty poor to be fair, lots of ali cable and  about 1200 metres.

[jid]

no idea, probably dropping as its night, my line is pretty poor to be fair, lots of ali cable and  about 1200 metres.

You're seeing a lower snr margin as noise has increased since the resync that enabled G. Inp.

If you resynced now you'd drop sync speed so worth keeping an eye on error rates with a lower snr margin.

[WWWombat]

I turned of the HG612 last night for 15 minutes and noticed the DS INP: changed from 46:00 to 49:00 what sort of change will that do to my line ?

INP is set by DLM to specify how big a burst of SHINE must be protected against. Bigger numbers mean DLM wants more protection, so it affects the way in which the modem configures the retransmission queues. Likely to have little impact on your line speed or latency.

INP_rein defines the amount of protection against REIN. The modem adjust FEC and interleaving for this, and it has a small impact to speeds and latencies, as seen so far, anyway.

[theogeor]

no idea, probably dropping as its night, my line is pretty poor to be fair, lots of ali cable and  about 1200 metres.

Interesting.. My line is about 1200m as well but I get significant better speed but I am on a good quality copper line. I am about double that... One important point is that I am using a shielded cable from the master socket to the modem and that made massive difference. Modem was syncing (but unstable ) at around 22-25mbits and now I have a stable 32-34Mbits ... 

Theo

[WWWombat]

Here's my full-monty graph spread from the last 4 days.

There isn't really anything of note, except that the OHFerr and the ES, even though they were low, have now disappeared entirely.