Print

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

[kitz]

As mentioned in the other thread last night  we were asking for your stats ASAP. 

I think you said you had a higher sync than you've ever had, so why not cap your sync rather than wait for interleaving. The idea was if you could act during that first day window, you may end up with a decent sync, no interleaving and a higher snrm which would help stabilise their err secs.

This is why I said yesterday you may be able to use the reset to your advantage, as long as you act quick enough.

[NewtronStar]

As mentioned in the other thread last night  we were asking for your stats ASAP. 

I think you said you had a higher sync than you've ever had, so why not cap your sync rather than wait for interleaving. The idea was if you could act during that first day window, you may end up with a decent sync, no interleaving and a higher snrm which would help stabilise their err secs.

The higher sync was when i was using the BB2 as modem the sync was 35600 kbps that gave me a thoughput of 33000 kbps but on the HG612 the sync is 33172 with a thoughput of 30192.

I am not interested to get a higher sync rate or lower latency all i want is the errored seconds close to what is was when G.INP was enabled a stable broadband on a long and noisy line.

Have a look at the stats kitz and will notice 17 for DS and 8 for US interleaving on a new DLM configuration (new profile) its the same profile after the line was reset on the old BT ISP way back in Feburary

xdslcmd info --stats
xdslcmd: ADSL driver and PHY status
Status: Showtime
Retrain Reason:   0
Last initialization procedure status:   0
Max:   Upstream rate = 6965 Kbps, Downstream rate = 32168 Kbps
Bearer:   0, Upstream rate = 6984 Kbps, Downstream rate = 33172 Kbps

Link Power State:   L0
Mode:         VDSL2 Annex B
VDSL2 Profile:      Profile 17a
TPS-TC:         PTM Mode(0x0)
Trellis:      U:ON /D:ON
Line Status:      No Defect
Training Status:   Showtime
      Down      Up
SNR (dB):    5.8       6.0
Attn(dB):    24.8       0.0
Pwr(dBm):    11.2       3.5
         VDSL2 framing
         Bearer 0
MSGc:      18      26
B:      243      223
M:      1      1
T:      64      38
R:      10      10
S:      0.2341      1.0185
L:      8680      1838
D:      17      8
I:      254      117
N:      254      234
         Counters
         Bearer 0
OHF:      31835353      234754
OHFErr:      14675      52
RS:      2037445635      1953759
RSCorr:      646603      318
RSUnCorr:   127378      0

         Bearer 0
HEC:      69326      0
OCD:      3343      0
LCD:      3343      0
Total Cells:   3339909035      0
Data Cells:   8076374      0
Drop Cells:   0
Bit Errors:   0      0

ES:      2774      39
SES:      0      0
UAS:      76      76
AS:      119713

         Bearer 0
INP:      0.00      0.00
INPRein:   0.00      0.00
delay:      1      1
PER:      3.76      9.71
OR:      51.06      26.35
AgR:      33223.29   7010.42

Bitswap:   66526/66573      28/28

Total time = 1 days 9 hours 16 min 29 sec
FEC:      646603      318
CRC:      14675      52
ES:      2774      39
SES:      0      0
UAS:      76      76
LOS:      0      0
LOF:      0      0
LOM:      0      0
Latest 15 minutes time = 1 min 29 sec
FEC:      43      0
CRC:      1      0
ES:      1      0
SES:      0      0
UAS:      0      0
LOS:      0      0
LOF:      0      0
LOM:      0      0
Previous 15 minutes time = 15 min 0 sec
FEC:      1340      0
CRC:      15      0
ES:      11      0
SES:      0      0
UAS:      0      0
LOS:      0      0
LOF:      0      0
LOM:      0      0
Latest 1 day time = 9 hours 16 min 29 sec
FEC:      28256      161
CRC:      995      28
ES:      536      22
SES:      0      0
UAS:      0      0
LOS:      0      0
LOF:      0      0
LOM:      0      0
Previous 1 day time = 24 hours 0 sec
FEC:      618347      157
CRC:      13680      24
ES:      2238      17
SES:      0      0
UAS:      76      76
LOS:      0      0
LOF:      0      0
LOM:      0      0
Since Link time = 1 days 9 hours 15 min 11 sec
FEC:      646603      318
CRC:      14675      52
ES:      2774      39
SES:      0      0
UAS:      0      0
LOS:      0      0
LOF:      0      0
LOM:      0      0
#

[Chrysalis]

is there anyone with a low sync on fast path? I wonder if the default profile changes if high attenuation?
Also on the hauwei's with g.inp removed from new DLM changes did they reactivate fast path?

[currytop]

Consider yourselves lucky for those lowly fees

AAISP charged me £100 to regrade from their ADSL service to FTTC. I assumed that was to pay for the OR modem and agent visit but never got an alternative option. Wires only would have suited me as the engineer gave me the plug to insert and then watched as I plugged in my router and configured it! Shame I got an ECI on a Huwaei cabinet as not long afterwards it seems G.INP may have been enabled at the cab!

Steve

[NewtronStar]

is there anyone with a low sync on fast path? I wonder if the default profile changes if high attenuation?
Also on the hauwei's with g.inp rmeoved from new DLM changes did they reactivate fast path?

Theres me thinking you had all the answers we all in the same boat we just don't know !
I am going to take time off and stop monitoring my line stats for while and just use the BB2 as it was intended to be used as an all in one VDSL2 modem and router.

Just before i go here are the BB2 DSL line stats 

[kitz]

I am not interested to get a higher sync rate or lower latency all i want is the errored seconds close to what is was when G.INP was enabled a stable broadband on a long and noisy line.

But irrc, the err/secs werent sufficient for the DLM to be bothered about (at least on the old system anyway).   
If you really do want less E/S then the way to go is capping your sync, which will reduce your Err/Secs.   There's a couple of regs that already cap their sync to stop the DLM applying interleaving (They get more E/Secs than you do). 

[kitz]

Just before i go here are the BB2 DSL line stats 

That looks like you are using it in router only mode - not combined vdsl modem/router.


--------
ETA

I note you've now changed the image and uploaded a different one.

[Chrysalis]

notice on the BB2 stats, the interleaved stats are -, he only has fast path data.  Interesting.

Newt is latency lower when using that device?

[NewtronStar]

notice on the BB2 stats, the interleaved stats are -, he only has fast path data.  Interesting.

Newt is latency lower when using that device?

No the latency increases by 10ms on the BB2 (35ms) the HG612 give me (25ms) but the BB2 gives me 2000kbps more thoughput on the DS and 900kbps more on the US i just don't understand why this is.

[Chrysalis]

Then I give up

[NewtronStar]

Consider yourselves lucky for those lowly fees

AAISP charged me £100 to regrade from their ADSL service to FTTC. I assumed that was to pay for the OR modem and agent visit but never got an alternative option. Wires only would have suited me as the engineer gave me the plug to insert and then watched as I plugged in my router and configured it! Shame I got an ECI on a Huwaei cabinet as not long afterwards it seems G.INP may have been enabled at the cab!

Steve

Thats kind of steep currytop when i moved from ADSL Max to VDSL2 the charge was £50 that was nearly three years ago.

[kitz]

Also on the hauwei's with g.inp removed from new DLM changes did they reactivate fast path?

Aardvark was previously g.inp enabled, but had it removed after a DLM reset by the engineer.  He's found himself back on interleaved path but depth 1.

[NewtronStar]

Kitz will you keep us upto date when G.INP is rolled out again as it's very unlikely i'll have any compatible stats monitoring hardware i'm just fed up with this G.INP On/Off and Fastpath On/Off stuff sometime it's better not knowing what your profile is as long as the internet works that all we need to know.

[kitz]

notice on the BB2 stats, the interleaved stats are -, he only has fast path data.  Interesting.

Newt is latency lower when using that device?

NS has changed the stats image so I can see what you mean now.

Thats the first time Ive seen a VDSL connection use FAST path, although that laylout of stats is typical of the Brightbox.
Note that the PATHS are entirely separate things to bearer channels so we must not get them confused.
It is entirely possible to use RS on FAST path, its just something Ive never seen BT do before, even back in the days of ADSL1.

It is interesting to note that only the upstream is seeing FECs and downstream is showing CRC but no FEC.  That to me would imply that NS's line has no interleaving but it does have RS encoding applied to upstream only.

Bearing in mind we know that BT are currently dabbling with removing interleaving from the upstream but those users on the trials are reporting no decrease in sync speed.

This would appear to be similar to what we are seeing here on NS's line with the BB2.   His downstream sync will be less because of the increased overheads of RS encoding.   His downstream is would appear to be running at full pelt because of no interleaving or RS encoding and why his downstream is able to achieve a higher sync.

In a weird way it looks like NS may have picked up one of the test profiles, because what we are seeing here sounds remarkably like what BT are 'playing with' when they are removing interleaving to replace previous g.inp lines.

No the latency increases by 10ms on the BB2 (35ms)

Now this is weird, because I cannot see any trace of interleaving in those (limited) stats.

NS can you post a tracert to the BBC please.

[kitz]

Kitz will you keep us upto date when G.INP is rolled out again as it's very unlikely i'll have any compatible stats monitoring hardware i'm just fed up with this G.INP On/Off and Fastpath On/Off stuff sometime it's better not knowing what your profile is as long as the internet works that all we need to know.

Yep I will do.   
I was promised some info last week, but now its all gone very quiet