I've now been running a 2700 for 26 days. During that time, I've had two re-syncs. What interests me is that they have happened when the noise margin reduced to 5db on both occasions. My Netgear would hold sync until the reported noise margin reached -2db.
This, coupled with the much steadier noise margin, suggests to me that the 2700 may 'dampen' the reported margin, while the Netgear let it through 'raw'. Both routers have shown a tendency to drop the sync on a Sunday afternoon, around 4-5pm, which suggests to me some local noise source (which I haven't been able to track).
The 2700 consistently re-syncs at higher speeds than the Netgear, so I'm still happy. :)
A thought Rik (another one! ::));
If someone was trying to access your computer wirelessly, would this pull your connection down? Even if they were then unable to crack the security?
If so it could explain a lot of things.
I don't have/never had wireless connection, so excuse if nimby question. :idea:
The re-sync must be a noise thing Rik, or a difference in software my 2700 holds at 0-1 without problem but I'm letting her have a run to see how long it will run now, as you say it holds sync at a higher speed and re-syncs higher, and is more stable than the netgear so I would say thus far I am happy also. :)
My 2700 holds sync for weeks on end and then today dropped it twice :o First time it re-syncd at a lower speed so rebooted and held at 7501kbps ;D
Quote from: MoHux on Dec 24, 2007, 19:34:36
If someone was trying to access your computer wirelessly, would this pull your connection down? Even if they were then unable to crack the security?
It shouldn't do, Mo, and the 2700 has an LED which lights when there's a wireless connection established, so that might give a clue. The problem has existed, though, since I went onto Max, even with a wires-only router. Worth a thought, though. My own theory is that it's mains borne noise.
I just don't know whether it's that some routers are "better" or whether they report the SNRM differently. Like you say, Rik, the Netgear is all over the place but would happily hold sync right down to the minus figures. My SpeedTouch (at the moment while my target SNRM is 15dB) would vary between about 13.5-15dB, whereas the 2700 has stuck at 15-16dB.
It's probably a combination of some routers handling lines better, but also using a different method of reporting the SNRM.
I agree, Sebby, it's probably impossible to compare router reports between brands. With the 2700, it may be impossible to compare between the two firmwares. What does seem clear for my one is that 5db will trigger a re-sync, which makes me think the indicated 11-13db margin may, in fact be indicating 3db higher than it should be.
Quote from: Rik on Dec 25, 2007, 13:14:05
I agree, Sebby, it's probably impossible to compare router reports between brands. With the 2700, it may be impossible to compare between the two firmwares. What does seem clear for my one is that 5db will trigger a re-sync, which makes me think the indicated 11-13db margin may, in fact be indicating 3db higher than it should be.
I would say thats not true for mine Rik, as it held all night at 0db which would be -3db then, and on my line that's impossible the netgear could not hold at 0db for long so some software differences must come into play I would imagine :-\ its a odd one to work out
Not sure what your firmware is Rik but hear is a mention of one of the version 4's
Early 2700HGV
fw: v4.25.27
hw: 2700-100588-005
Often reysncs every few days. Appears to try to return to highest line speed if the speed had dropped for any reason earlier. Tested for 9+ weeks. Router never locked up. Most reyncs often occured in early hours of morning. Detailed Logs sometimes emptied suggesting perhaps a reboot occured.
Quote from: Killhippie on Dec 25, 2007, 13:20:25
I would say thats not true for mine Rik, as it held all night at 0db which would be -3db then, and on my line that's impossible the netgear could not hold at 0db for long so some software differences must come into play I would imagine :-\ its a odd one to work out
That was partly my point, Gary. You have, I think, the BT firmware, so the reporting my be very different. If you take two Sky boxes and compare their signal readings on the same input, they will vary, so it's entirely possible, I guess, that every router has a unique 'signature'.
Whatever, the lowest sync I've obtained on the 2700 has kept my 2500 profile, rather than dropping me to 2000 as the Netgear would. That's a result as far as I'm concerned. :)
Quote from: Killhippie on Dec 25, 2007, 13:24:16
Not sure what your firmware is Rik but hear is a mention of one of the version 4's
Early 2700HGV
fw: v4.25.27
hw: 2700-100588-005
Mine is 4.25.19 and says there is no update.
Quote from: Rik on Dec 25, 2007, 13:26:12
That was partly my point, Gary. You have, I think, the BT firmware,
Mine is 4.25.19 and says there is no update.
yes I have the BT firmware Rik 5.29.107.19 it updated from .12 even through DNS poisoning which I found funny as it has to connect to a time server at BT's end to set the date etc, there are some versions of firmware out there for your one I think that will take it to.27 there are so many 2700 links but its all hit and miss, so best to take a peak over at the scream or safer still don't fix what is not broken since yours runs really well
Or just leave well alone and let it do its thing. ;)
Quote from: Rik on Dec 25, 2007, 13:38:15
Or just leave well alone and let it do its thing. ;)
agreed ;D
look here:-
http://your routers ip address here/xslt?PAGE=J42&THISPAGE=A02_POST&NEXTPAGE=J42
Lots of info the most interesting being the Impulse noise protection: This should be well below 1. mine goes up to 10-15 when it's icy and the trains go by.
As far as i am aware this is one of the best things about this router-most do not have this protection.
hope this helps identify if disconnects are noise relaated
Dean G7PKF
I have three showing at the moment, Dean. I have to confess I was wondering what they were. :)
Mine at present are
DSLAM: Country: {0x0F} Vendor: {ALCB} Specific: {0x00}
DSL Line (Wire Pair): Line 1 (inner pair)
Downstream Rate Cap: 8128 kbps
Downstream Atten. at 300kHz: 47.4 dB
Uncancelled Echo: -16.4 dB Ok
VCXO Frequency Offset: -10.4 ppm Ok
Final Rx Gain: 29.1 dB Ok
Impulse Noise Comp. Tones: 0 Ok
Excessive Impulse Noise: 0 Ok
Impulse noise protection: 0.64
Delay of latency path: 4.00 ms
I'm at:
DSLAM: Country: {0x0F} Vendor: {ALCB} Specific: {0x00}
DSL Line (Wire Pair): Line 1 (inner pair)
Downstream Rate Cap: 8128 kbps
Downstream Atten. at 300kHz: 55.3 dB
Uncancelled Echo: -16.1 dB Ok
VCXO Frequency Offset: 18.6 ppm Ok
Final Rx Gain: 34.5 dB Ok
Impulse Noise Comp. Tones: 1 Suspicious - impulse noise detected
Excessive Impulse Noise: 0 Ok
Impulse noise protection: 0.57
Delay of latency path: 4.00 ms
:) Suspicious - impulse noise detected :)
says it all really
It's the suspicious that I am puzzled by, Dean. What does it judge as suspicious?
1 or over it seems.
mine stays well below 1 then goes stupid 10-13+ then line disconnects.
I am at an advantage i know what causes it big long things that cause large sparks.
So that would be rugby players, then? ;)
Mine looks like this
Never really noticed it before, Ill keep an eye on it ;)
OOPS no pic. Be right back
[attachment deleted by admin]