Print

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

[kitz]

To know for sure you'd need to check immediately after a resync.

Im not saying that this is the case with your line, but its possible that
(a) Its still 9dB SNRM and youve lost 2dB of SNR.   OR
(b) Its now 6db SNRM and youve gained 1.2 dB of SNR.


The lack of CRCs is good even though its only been up for 3 hrs.   
It looks like the Err/Secs may be cumulative and not reset each sync.


Based on that one snapshot FAST mode is fine otherwise Id expect to see some CRCs...   but would need to be monitored properly.
Still think the line is way under performing for a 6dB atten....  but we are still in the waiting period.

---
Edited to add

Whilst on the subject of 6dB attenuation...   I confess I dont know much about the HH3 , but I suspect it may well be that the HH3 is one of those routers that reports the down attenuation at 300 kHz, (due to the figure not having changed when moving between adsl1 and adsl2+) ~ See attenuation

[matthewcl375]

Well, I'm now on what seems to be a 4.5db Target NM, the hub's been up since the end of the training period:

ADSL Line Status

Connection Information

Line state:   Connected
Connection time:   5 days, 13:38:33
Downstream:   18.56 Mbps
Upstream:   1.195 Mbps
 
ADSL Settings

VPI/VCI:   0/38
Type:   PPPoA
Modulation:   G.992.5 Annex A
Latency type:   Fast
Noise margin (Down/Up):   4.8 dB / 5.7 dB
Line attenuation (Down/Up):   6.2 dB / 3.3 dB
Output power (Down/Up):   20.1 dBm / 12.3 dBm
FEC Events (Down/Up):   0 / 0
CRC Events (Down/Up):   53 / 149
Loss of Framing (Local/Remote):   0 / 0
Loss of Signal (Local/Remote):   0 / 0
Loss of Power (Local/Remote):   0 / 0
HEC Events (Down/Up):   126 / 192
Error Seconds (Local/Remote):   123 / 236

[burakkucat]

Well, I'm now on what seems to be a 4.5db Target NM,

Eh?    I think not. The target SNRM would be set to one of the values in this range: 3, 6, 9, 12, 15, 18 . . . dB.

[kitz]

>>> Well, I'm now on what seems to be a 4.5db Target NM,

Unlikely..  Target SNR Margin is set in 3dB increases for the plain and simple reason that it takes 3dB of  'real' SNR to load a bit into a frequency bin.
ADSL1 needs a minimum of 2 bits per subchannel - hence 6dB target SNRm.
ADSL2/2+ has better handling of Error Correction overheads and only needs to load one bit per bin and still be happy - hence min 3dB target SNRm

- Yes it should be noted that we can tweak our routers and over-ride Target SNR to specific figures but this is just making use of the fact that some bins may have a little bit more 'real SNR' and is able to load sufficient bits.  The DLM set figures cant be messing with this and that and try tweaking to see what works and what doesnt..  hence it following the standards of 3,6,9,12,15 etc.  Ive certainly never seen BT using 4.5dB before.

As mentioned in my post above, the only way you can find out for sure is taking some stats immediately after a reboot.
 
Your line has been up for >5 days, so the figure shown as Noise margin is just that... ie the current Noise Margin at that moment in time...  and not the Target SNR

 

[Black Sheep]

This has me confused now ?. My SNR's are 5DS and 5.7US ?? I see these kind of variances day-to-day on my JDSU. What gives ??

Whoops, just seen the back end of the post ..... it's the current NM not the target. Gotcha. 

[matthewcl375]

Ah, most likely it's 3db then, and it has risen since being synced, my mistake 

[kitz]

Ah, most likely it's 3db then, and it has risen since being synced, my mistake 

As mentioned in my previous posts...  the only way you can tell is immediately after a resync.

Same as my post on the very top of the page ie the variance.

It can either be 3dB target SNRM and youve gained 1.5dB of SNR or
it could be 6dB target SNRM and youve lost 1.5dB of SNR

[matthewcl375]

Just a little update, my current stats:


Line state:   Connected
Connection time:   3 days, 13:00:26
Downstream:   19.81 Mbps
Upstream:   1.105 Mbps
 
ADSL Settings

VPI/VCI:   0/38
Type:   PPPoA
Modulation:   G.992.5 Annex A
Latency type:   Fast
Noise margin (Down/Up):   2.7 dB / 6.3 dB
Line attenuation (Down/Up):   6.4 dB / 3.3 dB
Output power (Down/Up):   20.1 dBm / 11.1 dBm
FEC Events (Down/Up):   0 / 0
CRC Events (Down/Up):   73 / 72
Loss of Framing (Local/Remote):   0 / 0
Loss of Signal (Local/Remote):   0 / 0
Loss of Power (Local/Remote):   0 / 0
HEC Events (Down/Up):   132 / 72
Error Seconds (Local/Remote):   68 / 62

I'm a bit worried about the noise margin, I had a PPPOA error and since I've restarted it the NM has stayed around 2.5db, I would've thought it should rise unless there's excessive noise

[kitz]

Speeds are looking much better.

>>> I would've thought it should rise unless there's excessive noise.

The SNRm on a decent line will stay stable...  mine often stays at 3dB for weeks on end without any variance.

[kitz]

btw I know we keep nagging you, but have you tried your TG582n on that line yet?

You may well find that whilst certain routers perform well on long lines..  these often dont work so well for very short lines.   In my own experience by using routers such as the TG582n and ST585 type, then I can get a much better sync speed than the routers recommended for long lines. Others on short lines also observe this too.

Your line attenuation is better than mine..  yet Im getting a higher downstream sync - regardless of the fact that Im on annex M, which should be lopping off my downstream to give to the upstream.