Print

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

[onion]

Absolutely no idea! My service provider have been great and are issuing a DLM profile reset in the coming days so just going to use the old modem again. They said they can see the line has been capped. Draytek support have been very poor for me. After issuing me the beta (RC1) firmware and that not making any difference, they said they have passed to their 3rd Line support engineers and will let me know on an update. The next email i get 2 days later from them says 'our engineers have suggested you try this firmware xxxx'. This was just a link to the original firmware (its actually the direct link from the website) which gave me the problems in the first place. I gave them my final stats, told them to close the ticket and wished them well. Seems like a totally different experience for you.

[j0hn]

The fact your inteleave depth is only 4 is odd as well. I would chat with your ISP as you are on fastpath.

Perfectly normal for G.INP for the interleave depth to be 4.
It's only 1 on ECI cabinets with no error protection enabled.

There's 3 members in this thread who have lost Downstream sync.
All have dropped sync because the DLM has increased the target SNRM. This happens due to an increase in errors.

If you don't have a 3, 4 or 5dB target then you wouldn't be affected.

[Newfie]

Absolutely no idea! My service provider have been great and are issuing a DLM profile reset in the coming days so just going to use the old modem again. They said they can see the line has been capped. Draytek support have been very poor for me. After issuing me the beta (RC1) firmware and that not making any difference, they said they have passed to their 3rd Line support engineers and will let me know on an update. The next email i get 2 days later from them says 'our engineers have suggested you try this firmware xxxx'. This was just a link to the original firmware (its actually the direct link from the website) which gave me the problems in the first place. I gave them my final stats, told them to close the ticket and wished them well. Seems like a totally different experience for you.

Sorry to hear that you’ve not had much luck. I’m holding steady and support did ask me to alter my SNR but I’ve asked some more questions and waiting to hear back.
I’m not expecting an instant fix, just have to keeping working with support to aid them.

[onion]

Perfectly normal for G.INP for the interleave depth to be 4.
It's only 1 on ECI cabinets with no error protection enabled.

There's 3 members in this thread who have lost Downstream sync.
All have dropped sync because the DLM has increased the target SNRM. This happens due to an increase in errors.

If you don't have a 3, 4 or 5dB target then you wouldn't be affected.

Now i've got the old modem back in i can see my SNRM for downstream is actually set at 8.7db!! (it was stable on 3 before), the downstream sync was just dropping consistently every single night. I couldnt keep holding out, may have ended up with adsl speeds!

Sorry to hear that you’ve not had much luck. I’m holding steady and support did ask me to alter my SNR but I’ve asked some more questions and waiting to hear back.
I’m not expecting an instant fix, just have to keeping working with support to aid them.

I really wanted to hold out, but the lack of true technical support from Draytek I personally received, and the consistent degredation of the line overnight has forced my hand somewhat. If i had of been stable at a reduced rate I'd have held but we are getting to the point where my actual day to day experience will be impacted from further drops so its just one of those things. Will be sure to keep an eye on the thread going forward!

[Newfie]

Support is what it is, many are the same unless you have direct contact inside where it’s easier. I’m in the same boat as you as unlike the other companies I know I have no way to get to the engineers so it’s the long route with DrayTek.

[mofa2020]

I see that bitswap is disabled for both DS/US, do not know if this could cause more frequent resyncs. if some tones are affected and make the DLM take action. if it can affect the line then it can be turned on by CLI command as I believe there is no setting for it in the GUI. If ECI cabinets support bitswap after all

[onion]

I see that bitswap is disabled for both DS/US, do not know if this could cause more frequent resyncs. if some tones are affected and make the DLM take action. if it can affect the line then it can be turned on by CLI command as I believe there is no setting for it in the GUI. If ECI cabinets support bitswap after all

Didnt notice that! My broadcom modem plugged in now says this - Bitswap:   1788/1794      288/288
Not actually sure about these technical things or what bitswap actually is(!) but seems like the broadcom is doing something on Bitswap, but the draytek wasn't.
Edit - then again, we actually have no idea if it was enabled, because the status pages are totally broken and misreporting!

[Guest_1]

Perfectly normal for G.INP for the interleave depth to be 4.
It's only 1 on ECI cabinets with no error protection enabled.

There's 3 members in this thread who have lost Downstream sync.
All have dropped sync because the DLM has increased the target SNRM. This happens due to an increase in errors.

If you don't have a 3, 4 or 5dB target then you wouldn't be affected.

Oddly I would have thought with such a loss of throughput G.INP would be changed to Interleave as Vdsl 2 can use much greater interleaving depths, the maximum interleaving depth is 3072 (for profile 17a) so 4 with that drop seems odd hence my mention of it. My Interleaving is 1087 when it was on fast path before the modem change, on an ECI cab. If the poster is on a Huawei cab I would have thought that interleaving would have been applied until the line was stable then G.INP reapplied at some point. Since not everyone is having re-syncs saying what cab they are on would help in this thread. What do you think?

[onion]

Oddly I would have thought with such a loss of throughput G.INP would be changed to Interleave as Vdsl 2 can use much greater interleaving depths, the maximum interleaving depth is 3072 (for profile 17a) so 4 with that drop seems odd hence my mention of it. My Interleaving is 1087 when it was on fast path before the modem change, on an ECI cab. If the poster is on a Huawei cab I would have thought that interleaving would have been applied until the line was stable then G.INP reapplied at some point. Since not everyone is having re-syncs saying what cab they are on would help in this thread. What do you think?

Mine is definitely Huawei - just checked

[Craig]

I am approximately 235m (as the crow flies) the ECI cab I'm connected to.

Here are updated stats from my 166

f8lure graph: https://f8lure.mouselike.org/proxyfirebrick.asp?ID=59492
thinkbroadband graph: https://www.thinkbroadband.com/broadband/monitoring/quality/share/3e3858033e7e057f7049e78b7ddcd21a5c55fa55-25-01-2021

Code: [Select]

DrayTek> vdsl status
  ---------------------- ATU-R Info (hw: annex A, f/w: annex A/B/C) -----------
   Running Mode            :      17A       State                : SHOWTIME
   DS Actual Rate          : 79992000 bps   US Actual Rate       : 20000000 bps
   DS Attainable Rate      : 81043380 bps   US Attainable Rate   : 24488100 bps
   DS Path Mode            :        Fast    US Path Mode         :        Fast
   DS Interleave Depth     :        1       US Interleave Depth  :        1
   NE Current Attenuation  :       13 dB    Cur SNR Margin       :        6  dB
   DS actual PSD           :     3. 5 dB    US actual PSD        :     3. 4  dB
   NE CRC Count            :        0       FE CRC Count         :        0
   NE ES Count             :        0       FE  ES Count         :        0
   Xdsl Reset Times        :        0       Xdsl Link  Times     :        1
   ITU Version[0]          : fe004452       ITU Version[1]       : 41590000
   VDSL Firmware Version   : 12-3-2-3-0-5   [with Vectoring support]
   Power Management Mode   : DSL_G997_PMS_L0
   Test Mode               : DISABLE
  -------------------------------- ATU-C Info ---------------------------------
   Far Current Attenuation :        0 dB    Far SNR Margin       :        9  dB
   CO ITU Version[0]       : 00b54649       CO ITU Version[1]    : 4e5406b2
   DSLAM CHIPSET VENDOR    : < ----- >
DrayTek> vdsl status more
  ---------------------- ATU-R Info (hw: annex A, f/w: annex A/B/C) -----------
                      Near End          Far End  Note
 Trellis   :      0     1
 Bitswap   :      0     0
 ReTxEnable       :      1     1
 VirtualNoise     :      0     0
 20BitSupport     :      0     0
 LatencyPath      :      0     0
 LOS   :      0     0
 LOF   :      0     0
 LPR   :      0     0
 LOM   :      0     0
 SosSuccess       :      0     0
 NCD   :      0     0
 LCD   :      0     0
 FECS   :      0     0 (seconds)
 ES   :      0     0 (seconds)
 SES   :      0     0 (seconds)
 LOSS   :      0     0 (seconds)
 UAS   :      0     0 (seconds)
 HECError   :      0     0
 CRC   :      0     0
 INP   :      0     0 (symbols)
 INTLVDelay       :    100   100 (1/100 ms)
 NFEC   :    255   255
 RFEC   :     16     6
 LSYMB   :   5410 20577
 INTLVBLOCK       :    255   255
 AELEM   :      0 ----

[Robbie]

Hey Robbie,

That's the EU version of the 166 yes?
It looks like it has the same 0 error counters as the UK model. Is that the case or was the screen grab taken immediately after sync?

Hi j0hn,

EU version with latest EU firmware.  It did appear to have the same 0 errors reported issue and the firmware itself seemed to be more of a rough beta than ready for production use.

I've not had a chance to run tests on the UK firmware yet so still running on the OR modem for now.  I was catching up on this thread to see if it was worth trying again this weekend but there seems to be only one other on this forum using the v166 with g.fast.

[Robbie]

I swapped out an MT992 with the 166 in the UK on BT network.

I am on the Fibre 100 g.fast product which gives me speeds of up to 160mbps downstream. Looking at the max achievable rate I could go up to around 210mbps if I had the Fibre 250 product.

The MT992 used to drop around once a day. Since installing the 166 I've not yet had a drop other than when I've manually rebooted. Much improved. DSL checker shows improved figures than with the MT992.

I do not seem to have any problems with packet loss whatsoever, with my MTU left at its default setting perhaps I'm just lucky?

Hi Oakserver - could you post your modem stats as you are one of the few running a 166 on g.fast?

Regards.

[smallal]

I'm running a 166 (UK version) on G.Fast on it's original firmware & it's been stable so far, run to cab is just over 100m
I noticed mention of a beta firmware 423RC1 for the UK model, anyone know what's changed & where it can be downloaded from?
My line stats:
ATU-R Information
Type:   
Hardware:   Annex A
Firmware:   12-3-2-3-0-5
Power Mngt Mode:   DSL_G997_PMS_L0
Line State:   SHOWTIME
Running Mode:   G.fast
Vendor ID:   00000000 00000000
ATU-C Information
Vendor ID:    00b54442 4d4390c1 [-----]
Line Statistics
        
                         Downstream     Upstream               
Actual Rate    328793Kbps   48157Kbps
Attainable Rate    457270Kbps   50991Kbps
Path Mode            Fast           Fast
Interleave Depth 0           0
Actual PSD    0. 0   dB   -5.-8   dB
                         Near End  Far End                   
Trellis            ON           ON
Bitswap            OFF           OFF
ReTx                    1       1   
SNR Margin    16dB   10dB
Attenuation    127dB   127dB
CRC                    0           0
Every stat below this shows as 0 with this firmware

[hushcoden]

I'm running a 166 (UK version) on G.Fast on it's original firmware & it's been stable so far, run to cab is just over 100m
I noticed mention of a beta firmware 423RC1 for the UK model, anyone know what's changed & where it can be downloaded from?

Here: https://www.draytek.co.uk/download/support/files/v166_bt_423RC1.zip
They say it should fix the stats showing '0' but not in my case...

[j0hn]

Oddly I would have thought with such a loss of throughput G.INP would be changed to Interleave as Vdsl 2 can use much greater interleaving depths, the maximum interleaving depth is 3072 (for profile 17a) so 4 with that drop seems odd hence my mention of it. My Interleaving is 1087 when it was on fast path before the modem change, on an ECI cab. If the poster is on a Huawei cab I would have thought that interleaving would have been applied until the line was stable then G.INP reapplied at some point. Since not everyone is having re-syncs saying what cab they are on would help in this thread. What do you think?

My thoughts... I'll try cover all your points.

The DLM doesn't ever change from G.INP to interleaving.
G.INP is the default on Huawei cabinets.
G.INP is now retained after a DLM reset or ISP migration (recent change).

New activations on Huawei cabinets have Interleaving applied for 2 days then G.INP is applied instead. That may also have changed and G.INP might even be applied immediately.

The Interleaving depth is usually 4 or 8 with G.INP applied.
It's only in the hundreds/thousands if INP/interleaving is applied instead.

The loss of throughput is due to the DLM increasing the target SNRM (noise margin).
The default target is 6dB down and up.
It's always 6dB on the upstream.
It's always 6dB on the downstream on ECI cabinets (due to no completed G.INP rollout).
If a line with G.INP in a Huawei cabinet is stable enough and has low enough errors the DLM lowers the target downstream SNRM to 5, 4 or 3dB, in 1dB steps.

There's 3 users in this thread with the Vigor 166 on VDSL2  who have had the DLM increase the target SNRM since connecting the Draytek.
This is due to an increase in errors. With the Draytek connected the DLM would never have lowered the target from 6dB in the 1st place.
At least 1 of the lines have gone from 3db to 4db to 5db all the way back to 6dB.
That's a considerable drop in stability compared to the users previous modem.

That's why having the error counters fixed is so important.
It's impossible to tell how stable the chipset is or how good/poor it is performing without being able to see the stats.
It's impossible to tell if each new firmware revision improves performance or not.

They say it should fix the stats showing '0' but not in my case...

Well that's a shame.
I'm very interested in seeing the full stats in the hope of seeing what's going on that's making the DLM so harsh.

It's a shame as the Vigor 130 is rock solid and lines regularly drop to 3dB using it.

I wonder if the Vigor 165 suffers the same 0 errors bug.
So far my Google search has been futile.