Kitz ADSL Broadband Information
adsl spacer  
Support this site
Home Broadband ISPs Tech Routers Wiki Forum
 
     
   Compare ISP   Rate your ISP
   Glossary   Glossary
 
Please login or register.

Login with username, password and session length
Advanced search  

News:

Pages: 1 ... 4 5 [6] 7

Author Topic: G.INP on ECI cabs  (Read 36139 times)

GigabitEthernet

  • Kitizen
  • ****
  • Posts: 2243
Re: G.INP on ECI cabs
« Reply #75 on: August 11, 2015, 10:47:47 PM »

Kitz,

In my case I lost a few megabits from crosstalk and when G.INP was enabled this sync speed returned.

I'm now terribly confused :(
Logged

NewtronStar

  • Kitizen
  • ****
  • Posts: 4898
Re: G.INP on ECI cabs
« Reply #76 on: August 11, 2015, 11:30:41 PM »

In my case I lost a few megabits from crosstalk and when G.INP was enabled this sync speed returned.

I'm now terribly confused :(

That's because your line was using interleaving and when G.INP was enabled on the 21st of july the sync increased due to the way G.INP works and nothing to do with crosstalk.

if you looked at your MDWS that's confusion for ya, only messing but i have never seen line stats change so much in 10 days  :) apart from Starman's
« Last Edit: August 11, 2015, 11:39:12 PM by NewtronStar »
Logged

GigabitEthernet

  • Kitizen
  • ****
  • Posts: 2243
Re: G.INP on ECI cabs
« Reply #77 on: August 11, 2015, 11:42:19 PM »

Ah in that case I must have no crosstalk! >:D

With only 40 or so other lines on the cabinet perhaps that isn't surprising.
Logged

kitz

  • Administrator
  • Senior Kitizen
  • *
  • Posts: 33881
  • Trinity: Most guys do.
    • http://www.kitz.co.uk
Re: G.INP on ECI cabs
« Reply #78 on: August 12, 2015, 12:11:11 AM »

Kitz,
In my case I lost a few megabits from crosstalk and when G.INP was enabled this sync speed returned.
I'm now terribly confused :(

I havent had a look at your stats, but its impossible for g.inp to recover crosstalk sync.   As mentioned above crosstalk is a constant which affects your SNR so the noise is there at sync time.   Retransmission is a method of recovering data in the event of noise bursts.  They work at different levels. 

As NS says the increase will be because Interleaving - or more correctly RS encoding - was turned off when G.INP was applied.  G.INP can only recover any sync speed lost from RS overheads.

Think of it this way: Crosstalk is a constant and you only see your SNRm change with the disturbers line is either switched on or off.    REIN and other environmental type noises cause spikes in the SNRm. 
You should already be familiar with the fact that when SNRm spikes occur then data packets can be lost, which in turn trigger errors such as CRC/FEC/ErrSec/LEFTRS etc. 
In old money it was the job of ReedSoloman redundancy to attempt to recover from FEC. The downside it carried a heavy overhead which subtracted from your available sync speed.
These days the retransmission process only kicks in only when it detects a lost packet and needs to recover it.   But the good thing about g.inp is that it works on the fly without permanent overheads so if the line isnt erroring then it doesnt need any overheads... and hence why it can give back speeds lost through Interleaving and ErrorCorrection.

Hope that helps make things a bit clearer :)

 
Logged
Please do not PM me with queries for broadband help as I may not be able to respond.
-----
How to get your router line stats :: ADSL Exchange Checker

kitz

  • Administrator
  • Senior Kitizen
  • *
  • Posts: 33881
  • Trinity: Most guys do.
    • http://www.kitz.co.uk
Re: G.INP on ECI cabs
« Reply #79 on: August 12, 2015, 12:18:45 AM »

Ah in that case I must have no crosstalk! >:D

With only 40 or so other lines on the cabinet perhaps that isn't surprising.

I bet there is some... its just that you seldom notice it because it doesn't cause the traditional type of noise spikes from SNRm line monitoring.
But below I will attatch a classic sign of a cross-talker.    Here's one of mine rebooting their router.   Note how its an upwards spike.   Its very easy to see on my line because my SNRm is normally rock steady.   If you also have environment noise and your SNRm is spiking anyhow,  its much harder to spot.
Logged
Please do not PM me with queries for broadband help as I may not be able to respond.
-----
How to get your router line stats :: ADSL Exchange Checker

Dray

  • Kitizen
  • ****
  • Posts: 2361
Re: G.INP on ECI cabs
« Reply #80 on: August 12, 2015, 12:19:35 AM »

The question is how does crosstalk, which is noise, not cause packet loss but REIN, which is noise, does cause packet loss?
Logged

Weaver

  • Senior Kitizen
  • ******
  • Posts: 11459
  • Retd s/w dev; A&A; 4x7km ADSL2 lines; Firebrick
Re: G.INP on ECI cabs
« Reply #81 on: August 12, 2015, 12:27:05 AM »

The question is how does crosstalk, which is noise, not cause packet loss but REIN, which is noise, does cause packet loss?

I don't understand. Both will cause packet loss surely if sufficiently bad that its uncorrectable?
Logged

kitz

  • Administrator
  • Senior Kitizen
  • *
  • Posts: 33881
  • Trinity: Most guys do.
    • http://www.kitz.co.uk
Re: G.INP on ECI cabs
« Reply #82 on: August 12, 2015, 12:43:02 AM »

The question is how does crosstalk, which is noise, not cause packet loss but REIN, which is noise, does cause packet loss?

Because crosstalk its a constant noise. Its there at sync time, so during the sync process & analysis its already affecting the SNR so your line rate is set based on the current SNRm and in effect crosstalk noise already allowed for.   Its constant nature means that it doesn't affect your SNRm until the x-talk disturber switches their modem off - or back on.   


REIN/SHINE/PEIN  are all types of noise bursts of various lengths which can cause the SNRm to vary.   A REIN type noise burst causes the downwards spike in SNRm.  When there is insufficient SNRm for the data signal to be heard properly at the other end, the data packet can become corrupt or lost which is what triggers an Error. 

Its spikey nature of low SNRm that causes packet loss.  Anything constant & your modem will factor it in your sync.

 
Logged
Please do not PM me with queries for broadband help as I may not be able to respond.
-----
How to get your router line stats :: ADSL Exchange Checker

GigabitEthernet

  • Kitizen
  • ****
  • Posts: 2243
Re: G.INP on ECI cabs
« Reply #83 on: August 12, 2015, 12:49:29 AM »

Kitz,

I recall seeing somebody saying that after a power cut their sync speed hugely reduced temporarily when the noise was significantly reduced.

Would vectoring give the same sync speed as this situation or would it still be worse?
Logged

kitz

  • Administrator
  • Senior Kitizen
  • *
  • Posts: 33881
  • Trinity: Most guys do.
    • http://www.kitz.co.uk
Re: G.INP on ECI cabs
« Reply #84 on: August 12, 2015, 01:04:01 AM »

Both will cause packet loss surely if sufficiently bad that its uncorrectable?

Yep.   The time when it can happen with crosstalk is if you sync up and your disturber isnt online.  The following is an example when it may occur: .

Your neighbour goes away on holiday and turns their router off.. and your SNRm goes up to 9dB
Later the same day you do a resync of your modem and it will automatically sync up at 6dB SNRm at 'x' Mbps and is quite happy at 'x'Mbps and doing its normal 1dB varience of between 5dB - 6dB without many errors

A week later your neighbour comes home and turns their modem on.
Immediately your SNRm goes down to 3dB..  and stays at 3dB, but 3dB isnt always enough for an error free line, so errors will start racking up, and if the line then does its daily 1dB variance the line is now  down to 2dB..  and errors will really start going through the roof.


Note: I used an example of 3dB for the disturber but it can be any amount from a fraction to more.  However it will almost always be the same amount from the same disturber.  Some of mine are fractions of a dB, a couple of mine are around 1dB, and I have a few at around 1.5dB.   How much they disturb your line depends upon the proximity of their line to yours, how strong their signal strength is and any PCB or PSD masks in use.
Logged
Please do not PM me with queries for broadband help as I may not be able to respond.
-----
How to get your router line stats :: ADSL Exchange Checker

kitz

  • Administrator
  • Senior Kitizen
  • *
  • Posts: 33881
  • Trinity: Most guys do.
    • http://www.kitz.co.uk
Re: G.INP on ECI cabs
« Reply #85 on: August 12, 2015, 01:13:12 AM »

Kitz,
I recall seeing somebody saying that after a power cut their sync speed hugely reduced temporarily when the noise was significantly reduced.

Would vectoring give the same sync speed as this situation or would it still be worse?

Ummm.. that doesnt make sense, sync speed doesnt reduce when noise if reduced.    Sync speed goes up when noise is reduced.

If thats a typo..  and you mean what usually happens after a power cut....  if you sync first then you normally get a better speed.   When others start to come back online, then your SNRm starts going down.

In that case:-

Vectoring would be using noise cancellation to mitigate the loss of sync speed.  When neighbours power off or on, you are less likely to notice any difference between your speed variance.
Logged
Please do not PM me with queries for broadband help as I may not be able to respond.
-----
How to get your router line stats :: ADSL Exchange Checker

GigabitEthernet

  • Kitizen
  • ****
  • Posts: 2243
Re: G.INP on ECI cabs
« Reply #86 on: August 12, 2015, 01:15:41 AM »

Sorry yes that was a typo :)

I'm sorry but I don't quite understand your second paragraph. The sync speed would be increased because there was less noise in the environment.

Would vectoring simulate the same conditions as other lines would no longer be interfering with yours?
Logged

kitz

  • Administrator
  • Senior Kitizen
  • *
  • Posts: 33881
  • Trinity: Most guys do.
    • http://www.kitz.co.uk
Re: G.INP on ECI cabs
« Reply #87 on: August 12, 2015, 01:28:29 AM »

Quote
I'm sorry but I don't quite understand your second paragraph. The sync speed would be increased because there was less noise in the environment.

Yes.  If you are first to sync up then your sync speed would be increased because less noise.  However when those other users start coming online, your 6dB of SNRm will be no more, each time a disturber comes back on, they will gradually eat away at your SNRm

Quote
Would vectoring simulate the same conditions as other lines would no longer be interfering with yours?

I dont know how efficient vectoring is.  Its a noise cancellation technique to reduce crosstalk.   I should imagine the results may vary depending on how much noise the neighbouring line puts on to yours.   They dont all put the same amount of noise on.    Your next door neighbour may not be causing you any crosstalk, but the one across the road and 5 houses down may be causing 2dB.   I honestly don't know, but I dont think it's 100% efficient, just that it makes great improvements. 
Logged
Please do not PM me with queries for broadband help as I may not be able to respond.
-----
How to get your router line stats :: ADSL Exchange Checker

Mark07

  • Reg Member
  • ***
  • Posts: 251
Re: G.INP on ECI cabs
« Reply #88 on: August 12, 2015, 04:10:06 PM »

The OR engineer I had out yesterday said they were enabling all newly installed cabs with vectoring... anyone know if this is true?
Logged

GigabitEthernet

  • Kitizen
  • ****
  • Posts: 2243
Re: G.INP on ECI cabs
« Reply #89 on: August 12, 2015, 04:11:39 PM »

I don't think that's true.
Logged
Pages: 1 ... 4 5 [6] 7