3db target - what to expect?

I'm currently taking a talk talk business line which, for a few reasons, got banded at 67 mb/s. My line is capable of 81-82 mb/s according to my stats.

Like many people I'm powerless to do anything about the banding but I was wondering if the upcoming drop in target margin will provide me with a 'back door' to a new training period.That's if TalkTalk business lines are included.

During the trial were all lines given the chance to retrain or were banded lines excluded? My line stats:

DSL synchronization status:
Up
Connection status:
Showtime
Upstream line rate (kbit/s): 19999
Downstream line rate (kbit/s): 66999
Maximum upstream rate (kbit/s): 31143
Maximum downstream rate (kbit/s): 80703
Upstream noise safety coefficient (dB): 15.5
Downstream noise safety coefficient (dB): 9.4
Upstream interleave depth: 0
Downstream interleave depth: 0
Line standard: VDSL
Upstream line attenuation (dB): 17.3
Downstream line attenuation (dB): 8.8
Upstream output power (dBm): 3.9
Downstream output power (dBm): 12.5
Channel type: None
DSL up-time: 15 days 11 hours 6 minutes 25 seconds

The 3,4,5 trial is only going to play with those lines it thinks will go lower, and being banded and thus the likely 9dB target suggests no change for you

Unless you can get a DLM reset, your line won't be in the trial/rollout.

... and no-one has worked out a way to force this still?

I did. Just gotta know an Openreach engineer well.

When does the trial end?

When does the rollout begin?

28th February.

March.

Cheat!

I assume there is no point me retraining as the DSLAM will reset my line if it removes the banding on my line?

Not really cheating, but PM me if you want details.

No point, leave it as it is.

To previous answers, I'll add:

Our experience of the trial is limited to 1 line that I think we can classify as definitely included. Beyond the evidence there, most statements are our best guesses, even if they are educated ones.

That 1 line is currently showing large amounts of retransmissions, and almost no ESs. It shows that G.INP can really take a hammering...

It is my personal opinion that the current use of banding is too severe. And, IIRC, it came into play alongside the last attempt to put G.INP onto ECI cabs.

I therefore kinda expect the DLM process to change somewhat this year. Obviously the 5-4-3dB deployment needs a change for Huawei, and G.INP may merit changes on ECI. I'm sure that whatever we know today about DLM will prove to need a rethink. I hope the stickiness of banding gets the boot.

In reply to a post by WilliamGrimsley:

I did. Just gotta know an Openreach engineer well.

You've got to get past the ISP support staff and Openreach systems before knowing the engineer comes into play.

True.

I've been away from here too long (and knew nothing of Openreach actually trying to improve speeds for very stable lines).

Only came back to see if Openreach were trialling something.

Three days ago (and until now; I thought for no apparent reason) my connection rebooted (having been up for an entire month).
For about the past year to 18 months; my connection speed has only strayed between about 64-66Mps.

The line is ridiculously stable (only ever seems to resync when I do something this end).

Suddenly; 3 days 22 hours ago (so just after midnight Monday; actually around 40 minutes into Tuesday) my connection rebooted (showing "Network Uptime" 3 days, 22 hours - I have the latest HomeHub).

The connection speed has suddenly jumped by more than 7Mbps.
A rise, such as that, should not be expected for a connection resync in the middle of the night (convention normally states night = more noise = lower sync).

Unfortunately (as I have the lastest Home Hub) I've no idea what margin DLM has stuck me on, but it certainly looks like I've been chosen for the trial (either by random, or because the line is so stable).

I wish I knew my current margin (or a rough way of estimating the attainable sync increase for every 1dB margin change - maybe someone has worked this out with VDSL - probably not; I guess this is the first time any of us have seen target margins different than 6dB)

I guess we'll know more over the next few weeks/months (if BT's trail does, indeed, become a roll-out and DLM starts experimenting with lower margins).

After years of rubbish ADSL and Openreach's punishing DLM (sticking people on 15dB margins because of thunderstorms or power cuts and refusing to un-stick it) this does make a refreshing change

The smart hub should show the snr.

Somewhere in Troubleshooting I think.

In "Technical Log" (obviously ).
Typical of BT to get their units screwed up (apparently; my connection is 73.31kbps)
Anyhoo;
Data rate: 20.00 kbps / 73.31 kbps (yeah, right, kbps )

Maximum data rate: 20000 / 71987 (going by the actual "data rate" and the noise margin below; I figure I'm probably on a 5dB target (or still on a 6dB target and there's some other reason why the connection speed has suddenly jumped by 7Mbps, when every resync, over the last year or so, has seen the connection "speed" normally just shy of 65Mbps).
If the target has only dropped by 1dB, that's a heck of a speed increase.
If the target is actually 3 or 4dB, I can't see how the maximum/attainable rate is below the actual data rate (when the current margin is higher), but maybe I'm still thinking along the lines of how ADSL2 worked!

Noise margin: 10.8 / 4.5 (I'm not going to play, and resync during daylight, let's just see how this pans out)

Line attenuation: 15.2dB

VPI / VCI: 0/38

Modulation: G_993_2_ANNEX_B

Latency type: Fast Path

There's a thread with all the exchanges in the trial. If you're not in the trial areas then it was probably a DSLAM reboot. It's actually incredibly common to see a line randomly resync 5-10mb higher than usual.

Are you on a Huawei cabinet and have G.INP enabled? The stats you pasted show fastpath and I've no idea how the homehubs show G.INP.

You look more like a plain DSLAM restart.

Only line that shows complete 5-4-3dB traits did the resync around 12:30pm.

5db is the first step