Print

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

[konrado5]

The location of the site of an impedance change, relative to a known point, cannot be determined from an Hlog plot. It could be anywhere, across the entire length of the circuit.  I suspect, at a minimum, at least three separate changes

1. I've thought about what long circuit with changes can cause it? Is it possible that changing of 20 meters cable would improve it?

2. I've noticed very curious phenomenon. Hardly ever I gain synchronization when the ending tones 490-511 are disabled for a while only at first minutes of synchronization. It is curious because these tones are continously loaded with 8 bits therefefore there are no so much SNR fluctuations. Is it any problem with SNR measurement only during synchronization establishment?

[Loading]

I doubt that just the adding on 20m is going to make a massive difference? Unless it is an external problem. It could be that the appearing tones are due to bitswapping, do the same tone appear when bitswapping is off?

[konrado5]

Loading: I don't question aboput appearing tones. I question about some tones that doesn't immediately appear hardly ever if I get new synchronization. On these tones there are continously about 8 bits if I don't disconnect, therefore there aren't significant SNR variations.

[konrado5]

burakkucat: have you got any idea?

[NewtronStar]

burakkucat: have you got any idea?

Lets not target one member with this question you would like an answer to, I put it to you konrado5 to answer your own question first and then the answers may roll in 

[burakkucat]

burakkucat: have you got any idea?

Sorry, I have absolutely no idea. 

You have exhausted all my possible suggestions. 

[konrado5]

OK burakkucat. There is other issue. I've noticed, sometimes If I have higher upstream rate for example 1245 kbps I have starting 7th tone not used. Somethings disables starting tone if further tones are unusually high bit-loaded. Example:
I've had 1221 kbps with following bit-loading:

Code: [Select]

Tone number      Bit Allocation
   0 0
   1 0
   2 0
   3 0
   4 0
   5 0
   6 0
   7 7
   8 9
   9 12
   10 12
   11 13
   12 14
   13 14
   14 14
   15 14
   16 14
   17 14
   18 15
   19 15
   20 15
   21 15
   22 15
   23 14
   24 14
   25 14
   26 14
   27 13
   28 13
   29 12
   30 10
   31 9
Other synchronization 1245 kbps:

Code: [Select]

Tone number      Bit Allocation
   0 0
   1 0
   2 0
   3 0
   4 0
   5 0
   6 0
   7 0
   8 10
   9 13
   10 13
   11 14
   12 14
   13 15
   14 15
   15 15
   16 15
   17 15
   18 15
   19 15
   20 15
   21 15
   22 15
   23 15
   24 14
   25 14
   26 13
   27 13
   28 13
   29 12
   30 11
   31 11
I've written about it here:
http://forum.kitz.co.uk/index.php?topic=13446.0

What is cause?

Best regards
konrado5

[kitz]

Does it really matter?

I thought we had already discussed that at much length in that thread what we thought the possible explanations could be.   We are going round in circles again.
Its not fair to keep asking such intricate questions and keep picking away at people and putting them under pressure to answer questions they dont know.

The questions never cease, weve done it for months going round in circles covering the same topics repeatedly.  Im tired of it now.  We dont have all day to sit and answer questions that require depth of master thesis.

Whatever we answer its just not good enough. 
The fact is no line will ever sync exactly the same, with exactly the same bit loading each time.

If you continue to call out me or b*cat, then I will start locking your threads as repetitive.

[konrado5]

I thought we had already discussed that at much length in that thread what we thought the possible explanations could be.

You gave me posssible explanation: new PSD masks. However, this explanation can be excluded because I have 7th tone not bit-loaded only when I have very highly bit-loaded remaing tones.

We dont have all day to sit and answer questions that require depth of master thesis.

On this forum people has much knowledge.

[konrado5]

Does it really matter?

It is very curious.

[NewtronStar]

You gave me posssible explanation: new PSD masks.

Konrado5 I don't even know what a PSD mask is on broadband ? if you would like to help the less knowledgeable members like me understand this would be most grateful   

[konrado5]

NewtronStar: PSD mask is power output setting for particular tones. If you have noticed U-shaped bit-loading and SNR graphs you see particular tones PSD-masked.

[konrado5]

PSD - power spectral density

[NewtronStar]

NewtronStar: PSD mask is power output setting for particular tones. If you have noticed U-shaped bit-loading and SNR graphs you see particular tones PSD-masked.

yeap I have a few of those U shaped bit-loading and SNR graphs but I don't understand what a PSD mask is and how it effects the tones (cause and effect) please could you put this in a simple format.

Cheers

[konrado5]

NewtronStar: PSD mask is particular frequiences higher power cut-back than other frequiences.