Yes, I know most of that, but not all, so thanks for this also.

IIRC WWWombat did not very long ago post the figures DLM uses for its classification system, but I forgot to bookmark it .

In reply to a post by WilliamGrimsley:

No, banding no longer removes itself on VDSL2.

How sure are you? I had a master relocated a month ago, the engineer did a real nasty job. He took the incoming cable, jelly crimped in the original master and a new cable going from here to the new master. As he did his work he kept the original master connected, with my router plugged in.

I made the engineer cut out the original master, and the jelly crimps. & got him to do a direct connection to the new master.

At the end my router had banded (well my sync was stuck lower than usual), it did resolve itself within a couple of weeks.

In reply to a post by lee111s:

What people don't get to see is that even if a line is stable on a banded profile it may still fall into DLM's amber category, which would mean DLM wouldn't release or relax the banding.

Last year, we were getting cases where banding wouldn't relent, even with zero ES. And cases where banding would relent for a few steps, and then stop, even with zero ES.

Is DLM's amber category possible with zero ES?

Having said that, stuck banding was much more prevalent last year. Perhaps things have improved this year.

Agreed.

At best, all we have is guess work really. I doubt the true inner workings of DLM will ever come out.

I have an unusually low upstream compared to my downstream, currently around 47/6. I got quite a bit more downstream with my ZyXel but at the cost of the upstream. Switching back to the hg612 increased my upstream at the cost of my downstream.

The upstream was originally 12, but has been steadily dropping over the years.

I'm in the same boat as you, banded at 60/20 due to a HG612 modem that kept locking up / rebooting.

Just going to leave the line up and see if / what DLM does.

It's been up for 12 days so far.