Print

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

[burakkucat]

So...

T'was merely a faulty PSU!

So simple!    :clap2:

I notice that on the advanced menu of the Echolife HG520b, there's some stats that look like bit loading or somesuch, which interests me very much as I've not played about with plotting that sort of stuff before.   

Bald_eagle1 would be able to bore you to death guide you on that general topic. 

Now that their new modem is nice and stable, I wonder what DLM will do to their speeds in due course?  Looked like they were on a 15db profile when I left, but I've no idea what they used to have prior to all these problems.

If they are registered with the TT Members' Forum, they could make a request for DLM to be turned off for their line and to be put onto a 6 dB target SNRM profile. (Both of which I have done.)

Attached are Routerstats-Lite stats for the modem with original firmware, updated, with different cable, and back to my usual Modem.  I reckon it's fixed now, don't you? 

Er . . . yes. 

[sheddyian]

If they are registered with the TT Members' Forum, they could make a request for DLM to be turned off for their line and to be put onto a 6 dB target SNRM profile. (Both of which I have done.)

Their line is quite long, think the BT predicted speed was 5,500 and I saw a fair bit of variation in the S/N margin, so perhaps 6db is optimistic?  And as their prime interest is in a stable line, I wonder if DLM is a good thing in their case?

That said, I'll give it a week or so and see what the line is doing, and maybe get TT to bump the profile down a notch or two and see what happens. 

Ian

[kitz]

Glad you got it sorted.   

I read your post the other evening yesterday and thought power adapter.
I had even started & typed to respond, when I saw that the b*cat just beat me to it and gave a bit more info..  so I scrapped my reply.   

[burakkucat]

That said, I'll give it a week or so and see what the line is doing,

That reads as an excellent idea. If you could arrange for even just 24 hours of RouterStats monitoring, it would give us some idea for any potential tweaking of the profile.

There again, are your relatives still going to migrate away from TT? If yes, the results of the monitoring will still be useful but any tweaking would be pointless.

[sheddyian]

Glad you got it sorted.   

I read your post the other evening yesterday and thought power adapter.
I had even started & typed to respond, when I saw that the b*cat just beat me to it and gave a bit more info..  so I scrapped my reply.   

Well, I should have realised too, because I noticed that after powering off their old modem, it took a very long time to even show a power light when switching back on again, let alone begin booting.

The unreliable wireless should have been a clue as well, though I did also find that there were 8 or so other modems nearby ALSO using channel 1 (as this was), AND one of the PCs had a dodgy wireless card anyway, so the blame for the wireless fell elsewhere.  But failing PSU can't have helped much.

Ian

[sheddyian]

An update on this problem : It's been gloriously stable since replacing the modem, although I think the synch rate is perhaps a little low.

        DOWN   UP
SNR    12.9    14.7
ATT    47.2    25.9
SYNCH 4519   736
MAX    5088   752

I forgot to instruct about leaving routerstats running for 24 hours, but I ran it for a few hours whilst I was there, and saw variations of no more than approx +/- 1.5 SNR.  A lot more stable than it was. 

Did see a lot (65500) of CRCs a few minutes after rebooting - not sure what happened there.

Here's a curious thing.  BT speed estimator reckons 5500 for the line speed. They're not quite getting that (4519).  I assume this is down to length of line.

I got them to run speedtest.net a few times on their computers, and those of neighbours.

Speedtest results :

                       Down      Up
relatives1           3790       470   (ISP : TalkTalk)
relatives2           3850       500
relatives3           3970       580
neighbour A        4000               (ISP : Sky)
neighbour B       12000               (ISP : TalkTalk)


(relatives1 2 3 are same relatives, different tests.  Sadly they did not record time of tests)

Neighbour A and Neighbour B are 2 different neighbours, but all adjacent to each other.  No big difference in distance from exchange.

So why does neighbour B seem to get far higher throughput?

My first thought was FTTC, but they apparently have had the same setup for some time, and nothing has been upgraded.

Any thoughts as to why this might be?

Ian

[burakkucat]

It does look, to me, as if your relatives have their target SNRM set to 12 dB. If that was changed to 6 dB, then they might achieve around another 1000 Kbps in the DS sync.

Without seeing a 24 hours graph of statistics for the SNRM, I am unable to say if it would be wise to take it that low. If it was for your own line, then no problem, I'd say just do it and we'll work through any issues that may arise.

All that said, I am using my line with a DS target SNRM set to 4 dB. But then I am a techno-cat!