Can meaning be gotten from router's "SNR margin" figure?

Bert,

Late last October, my 1.5/1.0 Qwest line (in the White Bear Lake area) was reprogrammed; my Cisco 678 was now training at 7168/896, and the SNR Margin dropped to 13dB, from it's usual 35-36dB. I would have been happy with the higher speeds, but I discovered my line just couldn't handle it, since the

678 kept retraining due to line quality. The occasional long lags when surfing was my tip-off that something wasn't right with my DSL service.

My assumption (guess) is the higher speeds is now using the full frequency spectrum that DMT DSL runs at, and the SNR margin is an average of all the frequencies currently in use. As for the 9dB with the Actiontec, it may work fine with the 6M speed that it's training up at; where as my older Cisco had problems holding 7M with a slightly higher SNR. The DSL modems will negotiate download and upload speeds, based on line quality. Even though the central office is set to 7M/896K, your modem can only link up with the slightly slower speeds.

If stability becomes a problem, Qwest can drop your line down to 5M, and you will see an improvement in your SNR margin. Hopefully it won't come to that.

I've had a newsgroup message bookmarked which goes into detail on the difference between SNR from the older CAP and newer DMT technologies. If the link wraps badly, do a search for "Qwest DMT line quality question" in this newsgroup back from around 2001.

Hope this helps. Bob.

formatting link

Reply to
Bob
Loading thread data ...

I just upgraded my Qwest DSL service from 1.5Mb to 7Mb, although it turns out that my Actiontec GT701 only trains to 6144/896.

The router's reported "SNR margin" dropped from 31/xx (I think the xx [upload] was around 20-25, but I've already forgotten) to 9/12 which seems disturbingly close to zero.

When I had 640K service I had SNR values in the 40db range, but the drop to 30db was dismissed as being due to the way DMT circuits behave compared to the old CAP technology.

The switch only happened this morning, so I won't be able to exercise the line 'til I get home this afternoon, but I wonder how well I should expect it to work with this reading.

Other than error counters, the only other numbers the Actiontec shows are Transmit power (18/12) and Attenuation (50/54). I have no idea what the significance of these might be.

Reply to
Bert Hyman

Bert,

I tried using the wireless in the Actiontec at a client's site; it worked OK till we moved a large metal wire shelving unit into the office to hold a server and various remote computers. After adding the shelf unit, the wireless signal went into the dumps. I ended up running a network cable to the remote computer, and turned off the wireless.

I've seen and read about people putting a Linksys wireless router after their Actiontec and getting much better results and signal strength, so you are not alone with your comments on the Actiontec's wireless performance.

Regards, Bob.

Reply to
Bob

Thanks for the link and for relating your experiences.

I had a Cisco 678 and experienced the same problems as you, but even at only 1.5Mb speed. I found that the Actiontec hooked up well at 1.5Mb and never retrained.

So far, I'm also seeing good performance at 6Mb.

My only complaint with the Actiontec is in the wireless area, but even there it works "well enough".

Reply to
Bert Hyman

Yeah; I've considered turning off the wireless in the Actiontec and cascading a second wireless router or a plain access point.

Reply to
Bert Hyman

Cabling-Design.com Forums website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.