I ask, because having switched from line A with ISP A, gaily synching at 6dB margin and achieving >4000 synch rates, despite having noise margin reset LINE B with ISP B, is now on 12dB margin synching reliably at just over 3000Kbps..This is at the same premises.
The ISP claiming that BT cant, or wont, hardwire my noise margin.
I am wondering if indeed they simply cant supply adequate bandwidth to the exchange..and this is policy, not technical..
Because the switchback always occurs at precisely 9:30 a.m. Whereas I would have thought a noise margin reset would have been more or less when an arbitrary counter gets to a value..
>> I ask, because having switched from line A with ISP A, gaily synching at >> 6dB margin and achieving >4000 synch rates, despite having noise margin >> reset LINE B with ISP B, is now on 12dB margin synching reliably at >> just over 3000Kbps..This is at the same premises.
>> The ISP claiming that BT cant, or wont, hardwire my noise margin.
>> I am wondering if indeed they simply cant supply adequate bandwidth to >> the exchange..and this is policy, not technical..
>> Because the switchback always occurs at precisely 9:30 a.m. Whereas I >> would have thought a noise margin reset would have been more or less >> when an arbitrary counter gets to a value..
>I always thought the noise margin is a measured value at the receiving end >and is determined by the line quality. Your line B sounds as though it might >be locked to 3000Kbps because it could go faster with a reduced noise >margin. You're not asking the correct question.
The margin is what the value is at the time you look. It may be that it goes down to 6dB at night and that gives the upper limit of 3000Kbps but in the daytime the noise decreases. This makes it look as though the line could do better whereas the speed is set by the worst case noise so that it's stable.
Mine has varied between 7.5 and 10.5 over the last 24 hours but I have a constant sync of 8128.
-- Regards - Rodney Pont The from address exists but is mostly dumped, please send any emails to the address below e-mail ngpsm4 (at) infohitsystems (dot) ltd (dot) uk
> >> I am wondering if indeed they simply cant supply adequate bandwidth to > >> the exchange..and this is policy, not technical..
With a compatible router and DMT Tool you can set a lower SNR margin, and the line does indeed go faster. Unfortunately if you do it during the day or when noise is low, and the noise increases, your lower DMT Tool set SNR, will be reset back by BT. The DSLAM looks at the reported figures from the customers modem and adjusts the SNR to reduce the error rate. Some routers perform better than others and give better speeds, and above all it pays to put your router next to the master socket. My sync speed was 2528kbps, has been as high as 5920kbps, and is at the moment 5408kbps. Most of the improvemet came from moving the router to alongside the master socket, chaining a couple of filters in series on the phone side, and finding a better modem. Ask here for opinions on the best modem/router, if you want to start a long thread, followed by a flame war tagged on the end of the thread. It's not a BT conspiracy. Your sync speed is the size of the pipe, whether your ISP can fill it, well that is another story.
In article <atcfzvasbuvgflfgrzfygqhx.ksrf030.pmin...@ouse.infohitsystems .ltd.uk>, Rodney Pont <aspami...@infohitsystems.ltd.uk> writes
>Mine has varied between 7.5 and 10.5 over the last 24 hours but I have >a constant sync of 8128.
Can I check whether I have this right? A higher margin is desirable because it ensures data quality at the expense of sync speed, yes/no?
It doesn't matter whether I set the margin to 1dB or 10dB, I always get a sync of 8032kbps. So I just set the margin slider at 100%, which gets me a pretty constant reported margin of 10-11dB.
Ato_Zee wrote: >>>> I am wondering if indeed they simply cant supply adequate bandwidth to >>>> the exchange..and this is policy, not technical..
> With a compatible router and DMT Tool you can set a lower > SNR margin, and the line does indeed go faster. > Unfortunately if you do it during the day or when > noise is low, and the noise increases, your > lower DMT Tool set SNR, will be reset back by BT.
This is esentially what is happening.
> The DSLAM looks at the reported figures from > the customers modem and adjusts the SNR to > reduce the error rate. > Some routers perform better than others and give > better speeds, and above all it pays to put your > router next to the master socket. > My sync speed was 2528kbps, has been as > high as 5920kbps, and is at the moment 5408kbps. > Most of the improvemet came from moving the > router to alongside the master socket, chaining > a couple of filters in series on the phone side, > and finding a better modem.
done all that.
> Ask here for opinions on the best modem/router, > if you want to start a long thread, followed by a > flame war tagged on the end of the thread.
don't want to change. The router is perfectly capable of synching much higher: Bt keeps increasing target noise margin so it wont. Why?
> It's not a BT conspiracy. > Your sync speed is the size of the pipe, whether > your ISP can fill it, well that is another story.
Why is it so hard to understand the point that:
My router synchs and holds happily at 4K or more and reported SNR of 7-11dB (presumably 9dB margin target). then BT decides it will run to a 12dB target. At 9:30 a.m. precisely, the router resynchs at 12dB or more SNR, and the speed drops concomitantly.
I've already GOT the gain turned fully up. Packet loss and errors are not that high. total noise is no worse than before.
i.e. is BT using a mechanism that is supposed to negotiate down to a reliable speed for *your ADSL line*, to negotiate down to a reliable speed their backhaul can cover?
By artificially increasing the target noise margin when there is no real need to do so?
I had two lines here: I have moved ISP and a new router to a different line. The performance noise wise is no worse that the old one was, the only difference is that BT are setting a far higher target noise in this one than they did on the last one and strangling its performance
The Natural Philosopher wrote: > Mike Tomlinson wrote: >> In article <qgjJm.130730$R52.60...@newsfe24.ams2>, Ato_Zee >> <ato_...@hotmail.com> writes
>>> chaining >>> a couple of filters in series on the phone side,
>> What does that do?
> reduces RF into the phone side of things a bit more.
> But its still hissy.
If you're hearing hiss, with or without filters in line with the phone, it's because there is something causing intermodulation on the line. That is - a non-linearity is causing a mixing of the frequencies in the DSL signal, and is generating ADDITIONAL components down in the audible range. The most likely cause is a high resistance connection. Oxidation on the contact surfaces forms a diode, which will act as a pretty good mixer.
The Natural Philosopher wrote: > My router synchs and holds happily at 4K or more and reported SNR of > 7-11dB (presumably 9dB margin target). then BT decides it will run to a > 12dB target. At 9:30 a.m. precisely, the router resynchs at 12dB or more > SNR, and the speed drops concomitantly.
No idea why 9:30am should be the time the DSLAM decides it knows best, but my experience is that a high target SNR takes about 3 weeks of VERY stable line conditions before it will drop to a lower value. In my case, I tweaked the modem coding to give an artifically low sync speed (you can do this with a Draytek). I let it run for 3 weeks (NO resyncs !), and reset the coding. DSLAM then resync'd at a target SNR of 6dB.
(It doesn't take much in the way of repeated noise bursts to push the Target SNR back up again - such as happens in the summer during lightning storms).
>I ask, because having switched from line A with ISP A, gaily synching at >6dB margin and achieving >4000 synch rates, despite having noise margin >reset LINE B with ISP B, is now on 12dB margin synching reliably at just >over 3000Kbps..This is at the same premises.
> The ISP claiming that BT cant, or wont, hardwire my noise margin.
> I am wondering if indeed they simply cant supply adequate bandwidth to the > exchange..and this is policy, not technical..
> Because the switchback always occurs at precisely 9:30 a.m. Whereas I > would have thought a noise margin reset would have been more or less when > an arbitrary counter gets to a value..
On 8-Nov-2009, Mike Tomlinson <m...@jasper.org.uk> wrote:
> > chaining > >a couple of filters in series on the phone side,
> What does that do?
By experiment I've found that an extra stage of filtering gives more isolation and greater sync speed stability without any adverse effect on phone audio quality. Two stage filtering gives the same result as some of the high priced filters. After all the ADSL goes straight through the first one, and the resistance with shunt capacitors that block the HF don't do much at audio frequencies since the phone is not exactly HiFi.
On Sun, 8 Nov 2009 13:05:35 +0000, Mike Tomlinson wrote: >In article <atcfzvasbuvgflfgrzfygqhx.ksrf030.pmin...@ouse.infohitsystems >..ltd.uk>, Rodney Pont <aspami...@infohitsystems.ltd.uk> writes
>>Mine has varied between 7.5 and 10.5 over the last 24 hours but I have >>a constant sync of 8128.
>Can I check whether I have this right? A higher margin is desirable >because it ensures data quality at the expense of sync speed, yes/no?
A higher margin is desirable but only up to a limit. You don't really want to push the margin up above 6dB if it pushes your sync speed down. You also want the margin high enough so that it doesn't go to 0 at the worst time of the day.
My 8128 sync speed is the maximum for ADSL so a higher margin won't help me. When the exchange is upgraded and I'm on ADSL2+ it should help me but I know I won't get the full 24meg.
>It doesn't matter whether I set the margin to 1dB or 10dB, I always get >a sync of 8032kbps. So I just set the margin slider at 100%, which gets >me a pretty constant reported margin of 10-11dB.
>Netgear DG834GT with DGTeam firmware.
I've never seen 8032 as a sync speed. Is that the maximum the DGTeam firmware runs at on ADSL?
-- Regards - Rodney Pont The from address exists but is mostly dumped, please send any emails to the address below e-mail ngpsm4 (at) infohitsystems (dot) ltd (dot) uk
Gaius wrote: > The Natural Philosopher wrote: >> Mike Tomlinson wrote: >>> In article <qgjJm.130730$R52.60...@newsfe24.ams2>, Ato_Zee >>> <ato_...@hotmail.com> writes
>>>> chaining >>>> a couple of filters in series on the phone side,
>>> What does that do?
>> reduces RF into the phone side of things a bit more.
>> But its still hissy.
> If you're hearing hiss, with or without filters in line with the phone, > it's because there is something causing intermodulation on the line. > That is - a non-linearity is causing a mixing of the frequencies in the > DSL signal, and is generating ADDITIONAL components down in the audible > range. The most likely cause is a high resistance connection. Oxidation > on the contact surfaces forms a diode, which will act as a pretty good > mixer.
That I had considered. The question is, how hissy before BT wont say 'well they are all like that'
On Sun, 08 Nov 2009 15:27:06 +0000, Gaius <b...@127.0.0.1> wrote: >The Natural Philosopher wrote:
>> My router synchs and holds happily at 4K or more and reported SNR of >> 7-11dB (presumably 9dB margin target). then BT decides it will run to a >> 12dB target. At 9:30 a.m. precisely, the router resynchs at 12dB or more >> SNR, and the speed drops concomitantly.
>No idea why 9:30am should be the time the DSLAM decides it knows best, >but my experience is that a high target SNR takes about 3 weeks of VERY >stable line conditions before it will drop to a lower value. In my case, >I tweaked the modem coding to give an artifically low sync speed (you >can do this with a Draytek). I let it run for 3 weeks (NO resyncs !), >and reset the coding. DSLAM then resync'd at a target SNR of 6dB.
My personal opinion is that problems like the OP's are more down to faulty equipment than a BT conspiracy. Despite my line being stable at a 6dB margin BT increase my target margin up to 12dB /exactly/ 48 hours after it is lowered, every time. I've only seen my target margin drop when it has been done manually and it's been on 12dB for months at a time. My ISP is quite good at asking BT to drop the margin, however.
If you have the DG834GT router you can use the DGteam firmware and you can manually lower the target margin. However I have found this makes my connection unstable, despite it working OK at that margin when not using this feature. YMMV, of course.
And then there's always LLU if this available at your exchange. -- (\__/) M. (='.'=) Due to the amount of spam posted via googlegroups and (")_(") their inaction to the problem. I am blocking most articles posted from there. If you wish your postings to be seen by everyone you will need use a different method of posting. [Reply-to address valid until it is spammed.]
In article <atcfzvasbuvgflfgrzfygqhx.kst9td0.pmin...@ouse.infohitsystems .ltd.uk>, Rodney Pont <aspami...@infohitsystems.ltd.uk> writes
>I've never seen 8032 as a sync speed.
I know, it's odd. It consistently syncs at that speed no matter how much I fiddle with the margin slider. It also did it with the standard Netgear firmware before I decided to give DGTeam a go. Results confirmed using speedtest.net, and I regularly get consistent 800kBps downloads, which fits.
I used to get 7616 or 75-something with a Tiscali-supplied Siemens SE584(?) router.
> Is that the maximum the DGTeam >firmware runs at on ADSL?
Pass. Fairly sure I have seen others claim to obtain 8128kbps with the same router. They don't specify the firmware though.
Thanks for the explanation about the margin. May try setting it to 6dB to see if the sync creeps up. I don't think there is much point though, since if the sync really is 8032 I should be getting a 7Mbps BRAS.
> On Sun, 08 Nov 2009 15:27:06 +0000, Gaius <b...@127.0.0.1> wrote:
>>The Natural Philosopher wrote:
>>> My router synchs and holds happily at 4K or more and reported SNR of >>> 7-11dB (presumably 9dB margin target). then BT decides it will run to a >>> 12dB target. At 9:30 a.m. precisely, the router resynchs at 12dB or more >>> SNR, and the speed drops concomitantly.
>>No idea why 9:30am should be the time the DSLAM decides it knows best, >>but my experience is that a high target SNR takes about 3 weeks of VERY >>stable line conditions before it will drop to a lower value. In my case, >>I tweaked the modem coding to give an artifically low sync speed (you >>can do this with a Draytek). I let it run for 3 weeks (NO resyncs !), >>and reset the coding. DSLAM then resync'd at a target SNR of 6dB.
> My personal opinion is that problems like the OP's are more down to > faulty equipment than a BT conspiracy. Despite my line being stable > at a 6dB margin BT increase my target margin up to 12dB /exactly/ 48 > hours after it is lowered, every time. I've only seen my target > margin drop when it has been done manually and it's been on 12dB for > months at a time. My ISP is quite good at asking BT to drop the > margin, however.
A much more likely explanation is that the OP is fantasising or trolling again. This is not the first time he has claimed to have some obscure problem that does not exist. He was adamant that the make or model of router/modem made no difference and when possible solutions were suggested he became abusive. Best ignored IMHO.
>>>> My router synchs and holds happily at 4K or more and reported SNR of >>>> 7-11dB (presumably 9dB margin target). then BT decides it will run to a >>>> 12dB target. At 9:30 a.m. precisely, the router resynchs at 12dB or more >>>> SNR, and the speed drops concomitantly. >>> No idea why 9:30am should be the time the DSLAM decides it knows best, >>> but my experience is that a high target SNR takes about 3 weeks of VERY >>> stable line conditions before it will drop to a lower value. In my case, >>> I tweaked the modem coding to give an artifically low sync speed (you >>> can do this with a Draytek). I let it run for 3 weeks (NO resyncs !), >>> and reset the coding. DSLAM then resync'd at a target SNR of 6dB. >> My personal opinion is that problems like the OP's are more down to >> faulty equipment than a BT conspiracy. Despite my line being stable >> at a 6dB margin BT increase my target margin up to 12dB /exactly/ 48 >> hours after it is lowered, every time. I've only seen my target >> margin drop when it has been done manually and it's been on 12dB for >> months at a time. My ISP is quite good at asking BT to drop the >> margin, however.
> A much more likely explanation is that the OP is fantasising or trolling > again. This is not the first time he has claimed to have some obscure > problem that does not exist. He was adamant that the make or model of > router/modem made no difference and when possible solutions were suggested > he became abusive. Best ignored IMHO.
> Peter Crosland
Peter FOAD.
Just because you simply aren't capable of following complex arguments doesn't mean they are fantasy.
Just that you are cross because no one listens to your crap anyway.
I didn't spend ten years of my life installing internet links across the world to have you to patronise me.
The Natural Philosopher wrote: > I ask, because having switched from line A with ISP A, gaily synching at > 6dB margin and achieving >4000 synch rates, despite having noise margin > reset LINE B with ISP B, is now on 12dB margin synching reliably at > just over 3000Kbps..This is at the same premises.
If it you had your SNR reset to 6db then /something/ must have resulted in it creeping back up to 12db, the most likely culprit being instability/loss of DSL signal due to intereference. One line isn't necessarily the same as the next even if they do serve the same premesis.
> The ISP claiming that BT cant, or wont, hardwire my noise margin.
AFAIK, Wholesale haven't really a mechanism for hardwiring the target SNR on their rate adaptive products. They can reset it to 6db but that won't stop the DLM systems from pushing it up again if they see fit.
> I am wondering if indeed they simply cant supply adequate bandwidth to > the exchange..and this is policy, not technical..
> Because the switchback always occurs at precisely 9:30 a.m. Whereas I > would have thought a noise margin reset would have been more or less > when an arbitrary counter gets to a value..
My target SNR was recently lowered from 12db to 6db (in 3db increments) without any intervention from myself at all. Make what you will of that.
-- |Bob Pullen Broadband Solutions for |Support Home & Business @ |Plusnet Plc. www.plus.net +--------------- twitter.com/plusnet ----------------
On 10-Nov-2009, Plusnet Support Team <supp...@plus.net> wrote:
> > Because the switchback always occurs at precisely 9:30 a.m.
Time of day can be critical because lots of things are on timers. I have a picture of a phone drop wire resting against the light sensor of a high pressure mercury vapour street light, which gives a lovely noise spike on startup, and clag all through the night. You can imagine what it does to SNR and sync speed, which is how I got involved. Wasn't always like that, was only a flashing orange beacon Zebra, but council decided it needed to be lit, so up went two massive mercury lamps, one on each side. Being taller than the BT pole, up in the air went the drop wire. Council "We can't shift the lamps, it's BT's problem" and BT "We can't shift our pole for one user, and there isn't any other pole - anyway it's fine when we test it (during the day, so no noise) What are you complaining about? It's no fault found" How do you get that one fixed?
Plusnet Support Team wrote: > The Natural Philosopher wrote: >> I ask, because having switched from line A with ISP A, gaily synching at >> 6dB margin and achieving >4000 synch rates, despite having noise margin >> reset LINE B with ISP B, is now on 12dB margin synching reliably at >> just over 3000Kbps..This is at the same premises.
> If it you had your SNR reset to 6db then /something/ must have resulted > in it creeping back up to 12db, the most likely culprit being > instability/loss of DSL signal due to intereference. One line isn't > necessarily the same as the next even if they do serve the same premesis.
>> The ISP claiming that BT cant, or wont, hardwire my noise margin.
> AFAIK, Wholesale haven't really a mechanism for hardwiring the target > SNR on their rate adaptive products. They can reset it to 6db but that > won't stop the DLM systems from pushing it up again if they see fit.
>> I am wondering if indeed they simply cant supply adequate bandwidth to >> the exchange..and this is policy, not technical..
>> Because the switchback always occurs at precisely 9:30 a.m. Whereas I >> would have thought a noise margin reset would have been more or less >> when an arbitrary counter gets to a value..
> My target SNR was recently lowered from 12db to 6db (in 3db increments) > without any intervention from myself at all. Make what you will of that.
>> Bob Pullen Broadband Solutions for >> Support Home & Business @ >> Plusnet Plc. www.plus.net > +--------------- twitter.com/plusnet ----------------
Friends in high places Bob? ;) -- Heard melodies are sweet, but those Unheard are sweeter flying...@tiscali.co.uk FN 2°°9 +, Mungo Brandybuck of Buckland
>On 10-Nov-2009, Plusnet Support Team <supp...@plus.net> wrote:
>> > Because the switchback always occurs at precisely 9:30 a.m.
>Time of day can be critical because lots of things are on timers. >I have a picture of a phone drop wire resting against the light >sensor of a high pressure mercury vapour street light, which >gives a lovely noise spike on startup, and clag all through the >night. You can imagine what it does to SNR and sync speed, >which is how I got involved. >Wasn't always like that, was only a flashing orange beacon >Zebra, but council decided it needed to be lit, so up went >two massive mercury lamps, one on each side. >Being taller than the BT pole, up in the air went the drop wire. >Council "We can't shift the lamps, it's BT's problem" and >BT "We can't shift our pole for one user, and there isn't any >other pole - anyway it's fine when we test it (during the day, >so no noise) What are you complaining about? It's no >fault found" >How do you get that one fixed?
Indeed. And if they can get a faulty line stable by increasing the target noise margin to a high value then it's also not a fault.
-- (\__/) M. (='.'=) Due to the amount of spam posted via googlegroups and (")_(") their inaction to the problem. I am blocking most articles posted from there. If you wish your postings to be seen by everyone you will need use a different method of posting. [Reply-to address valid until it is spammed.]
> On Tue, 10 Nov 2009 16:48:54 GMT, "Ato_Zee" <ato_...@hotmail.com> > wrote:
>>On 10-Nov-2009, Plusnet Support Team <supp...@plus.net> wrote:
>>> > Because the switchback always occurs at precisely 9:30 a.m.
>>Time of day can be critical because lots of things are on timers. >>I have a picture of a phone drop wire resting against the light >>sensor of a high pressure mercury vapour street light, which >>gives a lovely noise spike on startup, and clag all through the >>night. You can imagine what it does to SNR and sync speed, >>which is how I got involved. >>Wasn't always like that, was only a flashing orange beacon >>Zebra, but council decided it needed to be lit, so up went >>two massive mercury lamps, one on each side. >>Being taller than the BT pole, up in the air went the drop wire. >>Council "We can't shift the lamps, it's BT's problem" and >>BT "We can't shift our pole for one user, and there isn't any >>other pole - anyway it's fine when we test it (during the day, >>so no noise) What are you complaining about? It's no >>fault found" >>How do you get that one fixed?
> Indeed. And if they can get a faulty line stable by increasing the > target noise margin to a high value then it's also not a fault.
Which is exactly what some ISPs do, until it is choked down to some silly speed, then they decide to get OR to site.
Meanwhile, at the uk.telecom.broadband Job Justification Hearings, Mike Tomlinson chose the tried and tested strategy of:
> In article <atcfzvasbuvgflfgrzfygqhx.kst9td0.pmin...@ouse.infohitsystems > .ltd.uk>, Rodney Pont <aspami...@infohitsystems.ltd.uk> writes
>>I've never seen 8032 as a sync speed.
> I know, it's odd.
Is it perhaps an interleaved speed?
-- <http://ale.cx/> (AIM:troffasky) (UnSoEsNpE...@ale.cx) 22:31:29 up 26 days, 7 min, 4 users, load average: 0.52, 0.42, 0.29 "Stupid is a condition. Ignorance is a choice" -- Wiley Miller
|
