Can anyone explain the low reading just before the 40 tone mark please? I am trying to learn how to interpret the graphs.
Jane
Not sure Ann. But is the yellow portion the upstream
Oh sorry. Top part on the left in green is the upstream (which is low because I have had the snr increased for stability reasons). Then the blue part of the graph is the downstream which has a number of gaps. All of this I understand and can just about follow after reading Kitz's guide. It is when there is a low bar or a partial gap, in this case near the 40/172 mark on the upstream. I was wondering if it was indicative of interference at that frequency. Funny thing is, the graph has been like that for weeks. Just had a power cut and restarted the router and that first partial gap has gone? I was hoping maybe someone here understood these things and could give me an idiot's explanation (me being the idiot) as to whether such graphs could help diagnose any potential problems. I'm not saying that there's anything drastically wrong, I would just like to know what to look out for. I shall keep reading and maybe it will all sink in :)
Jane
Let us know when you find out ;D All I know is that some specific frequencies are not used.
That's about all I know too. Okay, if I learn anything useful I'll be back but don't hold your breath :dunno:
This is my understanding: During DSL Synchronisation (Channel Analysis phase) the modem will analyse the usability/quality of each channel and negotiate the number of bits assigned to that channel. The better the quality (i.e. the lower the noise, the higher the SNR, the lower the attenuation) the more bits can be loaded onto that channel. If you're seeing some channels with a low bit loading it's possibly because there was a higher level of noise/lower SNR on that particular frequency during channel analysis. I think DSL modems can also save a few channels with spare bits in the event one of the other channels becomes nosiy it can bitswap to spare channels instead of having to perform a DSL resynchronisation and new channel analysis.
That's sounds very plausible and logical , now if I could only remember it. ;)
I have a better understanding now. As well as the sections on the Kitz site I found this very useful
http://www.scribd.com/doc/24131175/ADSL-Basics
I was also wondering if the particular frequency/tone where the line was struggling a bit could be used to help diagnose any problems. The graphs I provided do not show any major problems, that's true, but I found it interesting that the 'blip' appeared to move further up the frequency scale after the power outage. I need a refresher course in this area methinks.
Jane
I've read somewhere that some frequencies are not available and are masked out to avoid interference. Be used to tell people that one such was to avoid conflicting with marine distress frequencies. Given that my exchange is about as far from the ocean as it's possible to get in the UK it always seemed a dubious explanation :eyebrow:
Hi Andrue. Yes I think that applies to the complete 'gaps' and also they will vary from country to country. They have some discussions over on the whirlpool forums where Australian radio frequencies can be detected from obviously unusual looking graphs. The more I look at it, it seems that the 'low bar' areas are probably unique to each person's line and are purely arbitrary rather than indicative of particular fault because the graph as a whole in my case look normal.
Jane
I had a chance to look at my stats - I'm also missing some bits (tone 207 in my case). However, I did a resync before I took this screenshot and the missing bits were in channel 108 instead. I suppose it's possible when I did the resync channel 207 was a bit noisy, but I honestly don't know. My personal theory is the router is keeping one tone clean in case it needs to swap bits.
Interestingly enough, tones 476 to 499 are completely unused. Tones in this area usually seem to be unused. I can't explain that at all! The only theories I have is - horrible noise on those frequencies, or they are in 'reserve' for bit swapping, or something far beyond my understanding!
That is an excellent insight and makes a lot of sense. I too did a resync today (switched off over night) and the missing bits moved further up. Wait a minute, it looks like mine have moved along to tone 207 as well! Our graphs otherwise are quite different of course but you may well be on to something.
I have managed to get RouterStats working on the Billion at last.