Print

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

[razpag]

That'll do for me. 

[burakkucat]

I got the REIN situation slightly wrong due to not being trained in that particular area, but am due to be at the end of this month.

Mr Pag,

Once that has occurred and you have a few "good" REIN stories noted down in your log book, perhaps you could provide us with some interesting technical anecdotes?

Something for the future, please.

[razpag]

I think informed information from myself, regarding REIN, will be a good while away yet B*Cat. Would you believe that the course is only 5 hours long ??!!!!!!!

I'm hoping that just to cover the 'legals', as I'm sure one of my old friends on OR who also happens to be a trainer, mentioned that the trainee has to assist on a 'live' REIN fault before gaining full accreditation. I do hope so, as I have been on a couple of jobs where i've sourced the REIN myself, but they were easy. I want to be assisting on an extremely hard-to-find fault.

TBH, I think i'll acquire more relevant info from you guys on here, until i'm au-fait with it all. 

[jeffbb]

Hi
quote :until i'm au-fait with it all

That may take a long time ,its one of those odd things . The more you learn the more you realise how little you Know 

Regards Jeff

[razpag]

That is so true Jeff. That's why I'm hoping to prise info from you guys in the know !!!

The problem is, actual 'live' REIN faults aren't as common as you might think within an engineers PWA (Preferred Work Allocation). I'll probably be lucky to get to 3/4 in a year. 

[roseway]

I think people often believe that they are subject to REIN when it's actually just ordinary electrical interference, which is common, probably universal.

[jeffbb]

Hi
quotes from Talk talk
quote 1 :If your router keeps switching off, DLM can see this as instability and consequently slow your speeds to the slowest available in order to 'stabilise' the line.

quote 2 : If you do want to switch it off, it will in most cases cause no ill effects with DLM as long as there is a period of several hours between such power down/power up

quote 3 :if the router is left 'on',DLM will see a stable line and speed the connection up to the fastest available profile that your line can handle, however if the router is always switched off/on,then it will remain always on the default profile which is somewhat conservative for many lines in terms of speed.

quotes  1 and 2 are as understood by most people and are the sort of advice given from this site .

quote 3  well that's a little different but its correct as far as it goes . Again it states as we all understand that continually switching "on"and "off " will result in the line being seeing as unstable .The  quote is correct  as the DLM will see the line as stable ,it does not imply that switching off once a day will jeopardise that .

Again this is totally in line with the normal advice given from this site

There are other factors to be considered which may have a bearing on the case . Some areas get more problems with stability during different times of the day  and it may be better to switch OFF during those hours ,if the router is not required .
Mr RP
 this link http://uk.answers.yahoo.com/question/index?qid=20101105095925AAZn083

as far as I can see the consensus was that there was no reason to link the slow down with the switching off the router nightly .

although I have noticed one of the responses is from BT Engi.

this one  http://www.diynot.com/forums/viewtopic.php?p=1969280&sid=b34fbc06ac850b9e9a66b92db9ed08fe

I must have missed something its about telephone sockets ?

Regards Jeff

[razpag]

Hi Jeff, whilst I appreciate your input into this debate, and can see where you are coming from ....... I will stick with my own thoughts on this subject, that turning the router off every evening does have a detrimental effect on performance on most lines controlled by DLM.

Again, the 'evidence' provided are just peoples personal views, and whilst they do form an arguement/alternative view to my own, they are just opinions. I can only go from what I know to be fact, and part of that fact is that (copied from my previous posting) ......  'I have a Field Log Book detailing umpteen EU visits where the cause of their 'slow-speeds' has indeed been due to switching off ones router. There has not been a single re-visit on these jobs that I am aware of, since making the EU conscious of the need to leave the router powered up 24/7.'

Add to the other fact that the ISP's technical people (not sales or front-line) I speak with on regular co-op calls, also have this same opinion, then who am I to argue

I really don't know what more I can add to this subject .  .There are always going to be scenarios that fly in the face of what I'm saying, but I reiterate that my particular stance on this appears to bring about long-standing results for the EU's. That, at the end of the day, is what I'm employed to do.

Cheers Jeff.

[jeffbb]

Hi
Something interesting  from
 EP 2 169 980 A1    EUROPEAN PATENT APPLICATION

Table 2
Stability                           Metric                   Very Poor                     Poor             OK                  Good
Aggressive             Re-Trains                    >10 per hour              mtb<3600      mtb<8640       mtb≥8640
Aggressive             Errors                                  -                         mtb<10          mtb<8640       mtb≥8640
Normal                   Re-Trains                    >10 per hour              mtb<7200      mtb<8640        mtb≥8640
Normal                   Errors                                  -                         mtb<300         mtb<8640       mtb≥8640
Stable                    Re-Trains                     >10 per hour             mtb<28800     mtb<86400     mtb≥86400
Stable                    Errors                                 -                          mtb<1000       mtb<28800     mtb≥28800

[0081] In table 2 "mtb" stands for "mean time between" and thus corresponds to the normalised metrics calculated by
dividing the total time in seconds for which the respective line has been in synchronisation over the past 24 hour period
of the monitoring by the number of re-trains or errors recorded in that period. For all cases, in the present embodiment,
if there are more than 10 re-trains in any one hour period, the line is assumed to be very poor, regardless of the number
of errors recorded. For lines operating at an aggressive stability level, if the average time between retrains is less than
once per hour (=3600 seconds) (e.g. 6 re-trains in less than 5 hours of "up-time") or if the average time between errors
is less than one per 10 seconds of up-time, then the line is deemed to be poor; if the average time between re-trains is
less than once every 2.4 hours (but more than once every hour) or the average time between errors is less than once
every 2.4 hours (but more than once every 10 seconds) then the line is deemed to be ok, whereas if the average time
between re-trains is greater than or equal to once every 2.4 hours or if the average time between errors is greater than
or equal to once every 2.4 hours, then the line is deemed to be good. From Table 2 above, it is clear what the thresholds
are for the other stability levels in the same way.


[0082] In an alternative embodiment, the stability levels could operate such that for the most aggressive stability level
the DLM function attempts to keep sync loses to below 12 per 24 hour period (including switching off modems/routers
which count as a sync loss) and to keep the line error free for 98.3% (59/60 seconds) of uptime measured over a 24
hour period; for the normal stability level the DLM function attempts to keep sync loses to below 6 per 24 hour period
and to keep the line error free for 99.8% (599/600 seconds) of uptime measured over a 24 hour period; and for the stable
stability level the DLM function attempts to keep sync loses to below 3 per 24 hour period and to keep the line error free
more than 99.98% (5999/6000 seconds) of uptime measured over a 24 hour period.

Regards Jeff

[mikew650]

Hello

I will give you my experience as to the effects of routine router switch off.

I never used to leave my router on 24/7 because I was a light user and always had a 63+ down attn so was happy to get 1mb speeds. In February this year the broadband became very poor due to HR fault which OR engineer fixed enabling me to connect at 3mb+ with a 53 db attn and 9 db TSRN. I decided to leave HGV2700 on 24/7 because of the improved speeds. However I am on a long rural line and experienced a few random disconnects which could knock my IP Profile down below 2mb.

In May the disconnects became worse and my TSRN had increased to 12 db. BT call centre could see no fault, but said would do some thing to help the stability. I managed to remain connected for 6 days at about 3400 but the profile was still at 2mb. Then I had another noise spike which knocked me down to 1500 IP. After 2 days the IP had increased to 2.5 Mb as I was connecting at 3000 after which I began switching the router off. That speed is all I need and as I connect at about 3400 during the afternoon, it should leave enough leeway.

So I am switched on for about an hour most days. I’ve done this for the last 2 weeks and today I checked the IP Profile which has gone up to 3 Mb which goes against all I’ve heard about needing to  leave the router on 24/7 for 3-5 days to get an increase. I don’t think my local exchange has been updated to the new profiling system.

[roseway]

There's no doubt in my mind that, with a long and slightly dodgy line like yours, switching off at night is a much better option than leaving the router switched on and risking instability during the night. Switching off the router is only a single loss of sync event, whereas a burst of interference when it's connected could result in multiple connection drops.

[waltergmw]

Hi Mike W,

I don’t think my local exchange has been updated to the new profiling system.

If you are referring to BT's 21CN you can tell if you have been changed over as the BT Speedtest will now show both a downstream and an upstream entry, whether or not you have asked to remain on ADSL1.

Kind regards,
Walter

[jeffbb]

Hi
quote I checked the IP Profile which has gone up to 3 Mb which goes against all I’ve heard about needing to  leave the router on 24/7 for 3-5 days to get an increase.

I have not ever thought that, . It might take 3 to 5 days to reset IP profile (which is NOT the same as resetting Target SNR etc .) The actual time it takes is dependant on  the % difference between current profile and the final IP profile you could expect.But this is not dependant on being connected  24/7. As far as the IP profile is concerned your synch rate is the governing factor . If you connect with Low synch rate (because of temporary noise on line) then your profile drops . Then within a few days either still connected  or if you connect as you require at a good synch rate the the profile will go back up . 

Regards Jeff

[waltergmw]

@ Jeff,

I am perplexed about the different ways IP Profiles work depending upon whether or not the line is a 21CN one.
For a 21CN line, I believe that the IP Profile now reacts almost immediately and precisely as 88.125 % of the sync speed, presumably whenever the modem re-syncs.

If I'm mistaken somebody please tell me !

Kind regards,
Walter

[mikew650]

Hello all

Walter, my exchange is using G.DMT Annex A and only downstream on BT speedtest.

Jeffbb, I support  what you say, I'm only going off the BT forum advice which always says to never disconnect the router in order to get the IP profile higher. It has been  better for me, with my unstable line, to turn off the router when not in use, usually off about 23 hrs a day. Of course you need to be re-connecting in a speed range to support the higher profile and my best time is during daytime as long as the weather is not stormy. I know this would not suit a lot of people but it needs to be pointed out more that switching off can be a benefit to those on unstable lines.

Regards Mike