Can some check my modem readings does anything look wrong ?

[deadbeat12]

Well i have been getting very unstable speeds ever since i got virgin media 20mb

from speeds of 0.2 mb - 1.2 mb then sometimes 15mb to 19mb (very rare)
have had one techy already still broken me thinks well heres the readings.

Cable Modem Downstream
Downstream Lock : Locked
Downstream Channel Id : 3
Downstream Frequency : 331000000 Hz
Downstream Modulation : QAM64
Downstream Symbol Rate : 5056.941 Ksym/sec
Downstream Interleave Depth : taps32Increment4
Downstream Receive Power Level : 2.5 dBmV
Downstream SNR : 39.3 dB


Cable Modem Upstream
Upstream Lock : Locked
Upstream Channel ID : 1
Upstream Frequency : 34000000 Hz
Upstream Modulation : QAM16
Upstream Symbol Rate : 2560 Ksym/sec
Upstream transmit Power Level : 39.0 dBmV
Upstream Mini-Slot Size : 2


Hopefully someone can help me i'm ringing them 2moz anyway YAY RAGE DOWN THE PHONE FOR FUN

[moaningmags]

All of those readings are spot on :)

[deadbeat12]

Ty for the reply must be something wrong some were though lol
https://www.cableforum.co.uk/images/local/2009/10/68.png (http://www.speedtest.net)

:D:D 20 mb for 30 pond Hmm.....

[Kymmy]

The upload looks like it's STM'd, the download looks like it's congested..

[Sephiroth]

I ws surprised to see QAM 64 for your 20 mbps circuit. My modem modulates QAM 256, although both modulations will handle > 20 mbps (I think its 27 & 38 mb[s respectively).

There's a useful thread here: http://www.cableforum.co.uk/board/12/33603655-what-is-qam64.html

IMO, you've either got a faulty modem or a duff connexion at the cabinet.

[deadbeat12]

yeah my line just cut out and came back now this is are downstream readings

Cable Modem Downstream
Downstream Lock : Locked
Downstream Channel Id : 2
Downstream Frequency : 402750000 Hz
Downstream Modulation : QAM64
Downstream Symbol Rate : 5056.941 Ksym/sec
Downstream Interleave Depth : taps32Increment4
Downstream Receive Power Level : 0.6 dBmV
Downstream SNR : 38.7 dB

[Welshchris]

Phaps its constipdated like my connection only comes through its spits and spurts!

[Sephiroth]

yeah my line just cut out and came back now this is are downstream readings

Cable Modem Downstream
Downstream Lock : Locked
Downstream Channel Id : 2
Downstream Frequency : 402750000 Hz
Downstream Modulation : QAM64
Downstream Symbol Rate : 5056.941 Ksym/sec
Downstream Interleave Depth : taps32Increment4
Downstream Receive Power Level : 0.6 dBmV
Downstream SNR : 38.7 dB

The readings are fine. That's why I think it's the cabinet or possibly the modem.

[bomber_g]

if there was a problem at the cab, would get low SNR and your modem would drop out randomly (sync light flashing means your modem aint sync'd up)

try downloading win MTR and running a trace to your DNS servers ( Virgin's are 194.168.4.100 ) and see if there is packet loss anywhere between your PC and the UBR / CMTS your modem is plugged into

[Sephiroth]

if there was a problem at the cab, would get low SNR and your modem would drop out randomly (sync light flashing means your modem aint sync'd up)

try downloading win MTR and running a trace to your DNS servers ( Virgin's are 194.168.4.100 ) and see if there is packet loss anywhere between your PC and the UBR / CMTS your modem is plugged into
I'm not the expertin HFC, but from my adjacent network knowledge, if the RF terminating stage in the cabinet is fine but the optical stage in the cabinet is misbehaving then, SNR - being measured to the local RF terminator - will not be adversely affected.

Tell me if I'm wrong, please.

[Ignitionnet]

I'm not the expertin HFC, but from my adjacent network knowledge, if the RF terminating stage in the cabinet is fine but the optical stage in the cabinet is misbehaving then, SNR - being measured to the local RF terminator - will not be adversely affected.

Tell me if I'm wrong, please.

SNR issues can be traced to anywhere from the CMTS card to the cable modem. :)

On the downstream side the laser is feeding the coaxial network - It goes CMTS - Forward Path Matrix (Coaxial) - Downstream Laser transmitter - Node optical receiver - coaxial trunk amplification cascade - customer drop - customer equipment. So problems with the laser feeding the node will impact everything coaxial after the node.

That said there are probably laser issues in the area - note that both the main 402.75MHz downstream and the dual docsis downstream at 331MHz are both 64QAM which makes no sense unless there are reasons why 256QAM can't be used.

[Sephiroth]

SNR issues can be traced to anywhere from the CMTS card to the cable modem. :)

On the downstream side the laser is feeding the coaxial network - It goes CMTS - Forward Path Matrix (Coaxial) - Downstream Laser transmitter - Node optical receiver - coaxial trunk amplification cascade - customer drop - customer equipment. So problems with the laser feeding the node will impact everything coaxial after the node.

That said there are probably laser issues in the area - note that both the main 402.75MHz downstream and the dual docsis downstream at 331MHz are both 64QAM which makes no sense unless there are reasons why 256QAM can't be used.

Well - you were careful not to say I was wrong! I believe that the downstream SNR seen at the modem is just the one hop to the RF side of the cabinet node. Since the OP's SNR and power levels are fine, you can see where I'm heading.

So, specifically, am I right? The SNR reported by the modem is for the first hop to the cabinet?

[Ignitionnet]

So, specifically, am I right? The SNR reported by the modem is for the first hop to the cabinet?

No. First remember it's a downstream SNR so that run in between modem and cabinet would be the last hop, and it's a measure of signal quality from CMTS right to modem. It's a measure of how accurately the received signal hits the centres of the constellations it's being decoded to.

I did say you were wrong in a very roundabout way :)

SNR issues can be traced to anywhere from the CMTS card to the cable modem.

Ah I think I get why you were querying. The downstream lasers do not transmit digitally on the cable networks, they are actually transmitting analogue usually. Even if the optical leg were digital it would still report an SNR partly influenced by the analogue run before and after the laser and the quality of the ADC and DAC either side of the optical run.

[Sephiroth]

So we're no further forward with the OP's problem. His readings are nominal. Apart from VM finding something logical in the circuit, we're almost at the point of suggesting the OP never uses that damn unreliable SPEEDTEST.NET!

[Ignitionnet]

Given the extremely low upstream perhaps packet loss is to blame due to poor upstream SNR, that one parameter we can't see.

If those numpties hadn't taken away SNMP could check that too, in a roundabout way, through retransmits :(