QPSK Vs. QAM

[Harlow CM20]

Hello all, quick Q. regarding upstream modulations.

I've heard that if you can only get QPSK then there are upstream issues in that area, I've been on QPSK for most of the the time but recently noticed that I was on QAM16. When It was QAM16 the service was improved but this only lasted a few days and now it's back to QPSK.

Sometimes if I repeatatively reboot the modem I can get QAM16 but as I said only sometimes.

Should I phone VM and report this or will they just say my connection is fine and say that there's nothing they can do?

All suggestions welcomed. (except bad ones)

[pip08456]

As Igni posted in anoither thread "There is an upstream problem in your area. Hopefully they'll get it fixed toute suite.

For future reference for others, you can tell there's an upstream problem by this:

Quote:
Upstream Modulation : QPSK
Upstream Symbol Rate : 5120 Ksym/sec


The Advanced Spectrum Management has had to kick in and downgrade the upstream to QPSK as the network is currently not able to support 16QAM in that area. "

[caph]

I've been on QPSK for the last year or so and I constantly max out my upstream and downstream when I speedtest. Pings are great too.

Now that's a problem I don't mind having. It must be the longest upstream problem ever though!

[deathtrap3000]

I used to be on 16QAM but then it changed to QPSK. Not had a problem with the upstream though.

[Kymmy]

Always been on QPSK (I think it's UBR dependant) and never had an issue and have got good signal levels

[pip08456]

Perhaps Igni may post on here and add further education.

[Harlow CM20]

For future reference for others, you can tell there's an upstream problem by this:

Quote:
Upstream Modulation : QPSK
Upstream Symbol Rate : 5120 Ksym/sec



My upstream symbol rate is 2560 Ksym/sec. Does this mean the problem is twice as worse or only half as bad?

[Foo Fighter]

Mines the same
Upstream Modulation : QPSK
Upstream Symbol Rate : 2560 Ksym/sec
Upstream transmit Power Level : 53.5 dBmV
Upstream Mini-Slot Size : 2

Not had any trouble other than last year when i couldnt host mw2 games but i might have been on QAM then

[Mick Fisher]

I was always on 16QAM. Only noticed it had switch to QPSK when the BB lost it's upstream last week.

It's fixed now but I notice I am still on QPSK. Still it's working alright :) so I don't care.

Upstream Lock : Locked
Upstream Channel ID : 1
Upstream Frequency : 45800000 Hz
Upstream Modulation : QPSK
Upstream Symbol Rate : 5120 Ksym/sec
Upstream transmit Power Level : 55.2 dBmV
Upstream Mini-Slot Size : 2

[Harlow CM20]

Mines the same
Upstream Modulation : QPSK
Upstream Symbol Rate : 2560 Ksym/sec
Upstream transmit Power Level : 53.5 dBmV
Upstream Mini-Slot Size : 2

Not had any trouble other than last year when i couldnt host mw2 games but i might have been on QAM then


I pretty much can't host anything online unless I host just one or two people that are on a UK adsl connection. when my US modulation was QAM16 recently I was able to host upto 12 players regardless of connection type and location. Online play was very smooth in general even if I wasn't host and I could even connect well to players from Mexico. Atm I am unable to play online again.

As was mentioned earlier in this thread, if you can't get QAM then there's an upstream issue in your area, I would like this issue resolved if possible please.

Here's a few local pingtest results if it helps matters

http://www.pingtest.net/result/22806858.png (http://www.pingtest.net)

http://www.pingtest.net/result/22806913.png (http://www.pingtest.net)

http://www.pingtest.net/result/22806941.png (http://www.pingtest.net)

[Harlow CM20]

Well, I phoned VM yesterday to speak about this and all they were willing to do was send round "another" tech. I've roughly had 10 VM techs to my property for power level problems amongst other things in the last 18 months. Had 3 visits since May 2010 and all say that everythings as good as it could possibly get.

Why must I suffer with poor performance and a B class connection that turns into an F class connection as soon as distance becomes a factor?

I would like someone to look into this as I believe there must be a problem elsewhere.

Also, one last question, does the VM 50Mb service use QPSK or QAM16 for upstream modulation

[pip08456]

Qam16 or to be pedantic 16QAM

[Peter_]

Qam16 or to be pedantic 16QAM
On our all our work systems it is listed as qam16.;)

[Harlow CM20]

On my modem configuration page it's simply listed as QPSK

[pip08456]

Acronym Definition
16-QAM 16-State Quadrature Amplitude Modulation but merely semantics. :D:D

[Harlow CM20]

xx

[Sephiroth]

Igni may well come in behind me (in the right sense, of course!).

Upstream is rather sensitive in terms of RF noise. It operates on a lower frequency band then downstream and there is analogue TV somewhere in the middle of the range (those frequencies are avoided by cable broadband).

As many of you know, downstream modulates at 256QAM (bit density = 8 per symbol), or at 64QAM (6 bits/symbol). The decision on 256 vs 64 is established according to your CMTS and its plant determined thresholds. That means the settings defined into the CMTS to define what it needs to do for your line to keep a stable connection. That's a human decision, so far as I am aware.

On top class plant, modulation can be much higher, e.g. FTTH with good lasers and so on.

That explanation leads us to upstream. In the current HFC infrastructure (i.e. FTTC), the sensitivity to RF noise means that 16QAM (4 bits/symbol) or QPSK (2 bits/symbol) are deployed. I a technical article I read a superb analogy for this. I'll paraphrase.

Upstream is like a road with potholes. You have to drive carefully and slowly over then to reach your destination. The more/larger potholes, the slower you have to drive. Hence faster with 16QAM and slower with QPSK. As your symbol rate at 3.2 MHz frequency is 2560 Ksymbols/sec, the channel is capable of 10 Mbps at 16QAM or 5 Mbps at QPSK. You're given ½ or 0.75 or 1.75 Mbps within that channel according to your service tier.

Whenever you see the on-the-fly UCD change, that could be because the CMTS is switching you to an upstream channel that will let your modem communicate with the CMTS and you may find yourself modulated at QPSK where previously you were at 16QAM.

In many areas, VM are quietly rolling out 6.4 MHz upstream frequency (4120 K symbols/second). This doubles the above given upstream speeds.

Hope that explains enough for y'all.

[pip08456]

So we can safely say that if the CMTS automatically switches your upstream modulation from 16QAM to QPSK it is because of potholes (read problems) in the upstream needing a slower rate of transfer to navigate.

[Sephiroth]

If pip requires confirmation, see what Igni says here (http://www.cableforum.co.uk/board/12/33634797-new-ucd-in-effect.html).

[Nedkelly]

Great way of explaining the upstream :) This why we dont like self installs and customers doing there own stuff because of the noise they put on the network:rolleyes:when its not done to spec and the trouble it can cause to everyone on that part of the network /ubr card :D

[Sephiroth]

Great way of explaining the upstream :) This why we dont like self installs and customers doing there own stuff because of the noise they put on the network:rolleyes:when its not done to spec and the trouble it can cause to everyone on that part of the network /ubr card :D

Excellent point. Users can put noise on the network via a Hoover, or a drill or a noisy fridge - and they do. Anything that arcs and isn't suppressed.

[Chrysalis]

QAM16 double the capability over QPSK, I would like to think QPSK areas have their user load halved appropriately, whether that happens I have no idea.

My sister is on QPSK, her off peak UBR load is 'lower' than mine evident by a lower night time jitter/latency. However it hits saturation much more and more excessively, probably due to the lower capacity of QPSK. Things like ack saturation are less evident on QAM16.

[Sephiroth]

I'm not sure that's right, Chrysalis. For example, if you were at only 1 bit per symbol, a 3.2MHz channel (DOCSIS 1) then your maximum speed if you were configured to use the bandwidth would be 3.2 Mbps.

So saturation on the tiered QPSK/16QAM won't be a function of the modulation. You hit saturation when you get to your tiered upstream value irrespective of modulation.

I reckon it's more likely that differences between customers are locational rather than due to bit densities.

Where have I gone wrong?

[pabscars]

Excellent point. Users can put noise on the network via a Hoover, or a drill or a noisy fridge - and they do. Anything that arcs and isn't suppressed.

Interesting,,, so is there any way to reduce, eliminate or control noise created in ones gaff.

[Chrysalis]

I'm not sure that's right, Chrysalis. For example, if you were at only 1 bit per symbol, a 3.2MHz channel (DOCSIS 1) then your maximum speed if you were configured to use the bandwidth would be 3.2 Mbps.

So saturation on the tiered QPSK/16QAM won't be a function of the modulation. You hit saturation when you get to your tiered upstream value irrespective of modulation.

I reckon it's more likely that differences between customers are locational rather than due to bit densities.

Where have I gone wrong?

keep it simple

useable bandwidth on shared qpsk upstream channel is 4.5mbit.
useable bandwidth on shared qam16 upstream channel is 9mbit.

so qam16 channels are twice the size. on my ubr port all my legacy channels are qpsk, but the overlay is qam16. On legacy I speedtest higher on uploads but get all sorts of problems such as ack saturation, connection stalling, and sporatic timeouts. qam16 seems to have some immunity to this. Which isnt surprising as the fatter the pipe the harder it is to outright saturate it.

[caph]

The point I was trying to make is that this is all pointless.

A 20Mbit user on QPSK will get 768kbit upload. A 20Mbit user on QAM16 will get 768kbit upload speed. You're getting 768kbit either way and neither user is any better off than the other.

If you're getting sub-standard upload then it needs looking at whether you are on QPSK or QAM16.

Just because you are on QPSK you should not assume that your service is degraded. One does not dictate the other.

[pip08456]

The point I was trying to make is that this is all pointless.

A 20Mbit user on QPSK will get 768kbit upload. A 20Mbit user on QAM16 will get 768kbit upload speed. You're getting 768kbit either way and neither user is any better off than the other.

If you're getting sub-standard upload then it needs looking at whether you are on QPSK or QAM16.

Just because you are on QPSK you should not assume that your service is degraded. One does not dictate the other.

So, although there are potholes on the road which need slower and more careful avoidance QPSK is not a downgrade from 16QAM?

To my understanding the network supports 16QM but if there are problems in the upstream then the modem or CMTS will switch to QPSK. (Or have I fallen into a pothole? :D:D:D)

16QAM=No potholes

QPSK= Potholes

i.e. QPSK=Degradation

Am I missing something here?

[caph]

Am I missing something here?

Pip, if you're getting a good 768 up on QAM10000000 or a good 768 on QABC then what does it matter? You're maxing out your upload. That's my point and that's what you are missing.

Imagine the following support call:-

> Hi, I'm suffering from upstream degradation on my 20Mbit connection.

>> OK sir, what exactly is the degredation you are experiencing?

> Well, I'm on QPSK on my upstream.

>> OK, well what is your current upload speed?

> 768Kbit with no packet loss.

>> Well, actually sir, you're currently getting the maximum allowable speed for your connection. Why exactly are you wasting my time?

> Because Pip from CableForum told me I was degraded! Err, it was something about potholes and QPSK?

>> Go to bed sir, and one more thing, could you let me have Pip's address, I'd like to pay him a little visit...

[pip08456]

> Because Pip from CableForum told me I was degraded! Err, it was something about potholes and QPSK?



a) Where did I say that? I merely progressed on the summations made in this thread which point to QPSK being an indication of a problem.

b) As I am on 50MB I would be complaining if I only got the upstream bandwidth you quote.

If you wish to have my address then just PM me.

[caph]

> Because Pip from CableForum told me I was degraded!

a) Where did I say that?

You said it here Pip:-

QPSK=Degradation

Go to bed.

[Chrysalis]

The point I was trying to make is that this is all pointless.

A 20Mbit user on QPSK will get 768kbit upload. A 20Mbit user on QAM16 will get 768kbit upload speed. You're getting 768kbit either way and neither user is any better off than the other.

If you're getting sub-standard upload then it needs looking at whether you are on QPSK or QAM16.

Just because you are on QPSK you should not assume that your service is degraded. One does not dictate the other.

there is no garuantuee a user will get 768kbit at all, your statement implies that.

if the user load is equal on a qpsk channel to a qam16 channel then effectively the contention ratio is doubled. To some of us contention ratio is very important.

if VM do indeed reduce the numbers on qpsk channels then thats fair enough.

[pip08456]

I actually said

i.e. QPSK =Degradation.

If you wish to quote I have no problem as long as you get it right.

[caph]

You win Pip. Now go to bed.

---------- Post added at 23:50 ---------- Previous post was at 23:44 ----------

there is no garuantuee a user will get 768kbit at all, your statement implies that.

if the user load is equal on a qpsk channel to a qam16 channel then effectively the contention ratio is doubled. To some of us contention ratio is very important.

if VM do indeed reduce the numbers on qpsk channels then thats fair enough.

Jesus Chrysalis, I'm saying that if you have a good upstream connection on QPSK, QAM16 or ZXYZ then there's no need to worry. Is that so hard to understand???

If a user has upstream problems then ring it in. If they don't then don't. What I'm saying is that no-one should be getting worked up about QPSK if they have a good upstream connection. QPSK per se is no reason to assume you have a problem that needs addressing.

I really feel like I've managed to get marooned at the bus stop but it's the last bus home and I'm reduced to trying to pacify the drunk imbeciles.

[pip08456]

You win Pip. Now go to bed.

---------- Post added at 23:50 ---------- Previous post was at 23:44 ----------



Jesus Chrysalis, I'm saying that if you have a good upstream connection on QPSK, QAM16 or ZXYZ then there's no need to worry. Is that so hard to understand???

If a user has upstream problems then ring it in. If they don't then don't. What I'm saying is that no-one should be getting worked up about QPSK if they have a good upstream connection. QPSK per se is no reason to assume you have a problem that needs addressing.

I really feel like I've managed to get marooned at the bus stop but it's the last bus home and I'm reduced to trying to pacify the drunk imbeciles.

If you had bothered to absorb what the thread was about you would realise that it has nothing to do with anyone's connection but the difference between the 2 modulations and the possilble consequences.

Yes if your connection is OK don't worry about the minutia of modulation in the same way as if your power levels are not "in spec" but all is still well.

Please also note:-
a) I am not drunk.
b) I am not an imbecile nor are any others in this thread.
c) Yes I'm off to bed now to give you time to reflect on your offensive remarks. (and I'm tired)

[caph]

If you had bothered to absorb what the thread was about

Hello all, ...now it's back to QPSK ...Should I phone VM and report this

It's you who need to re-read the original post.

I'm quite obviously answering the OP. So obviously in fact that arguing about it is imbecilic. If you don't want to get likened to one. Stop acting like one.

You know, I think I do feel better for reflecting on my remarks, thanks.

[Sephiroth]

keep it simple

useable bandwidth on shared qpsk upstream channel is 4.5mbit.
useable bandwidth on shared qam16 upstream channel is 9mbit.

so qam16 channels are twice the size. on my ubr port all my legacy channels are qpsk, but the overlay is qam16. On legacy I speedtest higher on uploads but get all sorts of problems such as ack saturation, connection stalling, and sporatic timeouts. qam16 seems to have some immunity to this. Which isnt surprising as the fatter the pipe the harder it is to outright saturate it.
I don't agree. Take the Underground System Analogy (just to keep it unsimple!).

Northern Line from Kings X to Oval.

16QAM = Western route with 7 potholes (stations)
QPSK = Eastern route 8 potholes (stations)

Anything can happen because there is no common treatment of these routes. The 16QAM route could take longer if more people got on at each station.

Again, where am I wrong?

---------- Post added at 10:43 ---------- Previous post was at 10:37 ----------

So, although there are potholes on the road which need slower and more careful avoidance QPSK is not a downgrade from 16QAM?

To my understanding the network supports 16QM but if there are problems in the upstream then the modem or CMTS will switch to QPSK. (Or have I fallen into a pothole? :D:D:D)

16QAM=No potholes

QPSK= Potholes

i.e. QPSK=Degradation

Am I missing something here?

64QAM = Less potholes
16QAM = Potholes
QPSK = More potholes

DOCSIS = The solution to assure tiered service at the given modulation.

So, no degradation because of the DOCSIS solution.

However, infrastructure that will only work at QPSK will be limited for upgrade purposes to the maximum speed that 2 bits/symbol can deliver. If you want to call that "degradation", so be it, but it is not degradation in current operational terms.

HTH.

---------- Post added at 10:47 ---------- Previous post was at 10:43 ----------

QAM16 double the capability over QPSK, I would like to think QPSK areas have their user load halved appropriately, whether that happens I have no idea.

My sister is on QPSK, her off peak UBR load is 'lower' than mine evident by a lower night time jitter/latency. However it hits saturation much more and more excessively, probably due to the lower capacity of QPSK. Things like ack saturation are less evident on QAM16.
On this question of your circuit versus your sister's. I used the Northern Line analogy to cover the difference between the two of you.

My own case, having been moved for resegmentation and not put back to my original UBR, is that I move between 16QAM and QPSK on the fly. When I notice this, I do a load of quick upstream tests and compare with before. They're the same-ish with no pattern of degraded behaviour on QPSK.

This is all empirical, But DOCSIS is the governor IMO.

[Harlow CM20]

My own case, having been moved for resegmentation and not put back to my original UBR, is that I move between 16QAM and QPSK on the fly. When I notice this, I do a load of quick upstream tests and compare with before. They're the same-ish with no pattern of degraded behaviour on QPSK.

Wow, this thread has turned into quite the discussion and also is educational.

My 3 stbs and cm all lost connection to the vm network about 30 mins ago and came back alive 10-15 mins ago.

Since that brief loss of service my us modulation is now qam16, or to be pedantic 16qam.

Sephiroth, do you use your vm connection to play online games with because if you don't then you probably won't be affected by being on qpsk modulation.

[Sephiroth]

I don't use it for gaming but the boy does (CS) and he's oblivious to which modulation we're on. But I'll specifically ask him to note this weekend. When we're on QPSK, we don't get mved on the fly to 16QAM. That happens if I force it and get allocated a 16QAM channel (50/50 hit).

The factor that matters for gaming is your timeslot being available. If you're on QPSK you need more symbols to complete a transaction. I'm at work right now and don't know how many symbols are in a mini-timeslot. But I see your point. We might be lucky here because we are less than 500 homws passed per optical node and so the upstream is available.

But I'll look at that again in the light of your question.

[Harlow CM20]

Thankyou Sephiroth.

It would be kind of good in a way if you do get the same issues as I get as it would put my mind to rest regarding this situation and may lead to a fix or an improvement to the service available where I live but to be perfectly honest, I hope your son doesn't suffer from this type of us modulation as it can be quite annoying.

I even got host banned by the game server I use because my connection would usually drop players if I was hosting, the game server has an anti cheat mechanism which I kept triggering.

Anyway, hoping for a great weekend for us both and well, anyone else who wants a great weekend tbh.

[pip08456]

I

64QAM = Less potholes
16QAM = Potholes
QPSK = More potholes

DOCSIS = The solution to assure tiered service at the given modulation.

So, no degradation because of the DOCSIS solution.

However, infrastructure that will only work at QPSK will be limited for upgrade purposes to the maximum speed that 2 bits/symbol can deliver. If you want to call that "degradation", so be it, but it is not degradation in current operational terms.

Thanks for that Seph I can see the point now, I think. :D

[Chrysalis]

You win Pip. Now go to bed.

---------- Post added at 23:50 ---------- Previous post was at 23:44 ----------



Jesus Chrysalis, I'm saying that if you have a good upstream connection on QPSK, QAM16 or ZXYZ then there's no need to worry. Is that so hard to understand???

If a user has upstream problems then ring it in. If they don't then don't. What I'm saying is that no-one should be getting worked up about QPSK if they have a good upstream connection. QPSK per se is no reason to assume you have a problem that needs addressing.

I really feel like I've managed to get marooned at the bus stop but it's the last bus home and I'm reduced to trying to pacify the drunk imbeciles.

of course, but I dont have a good upstream :)

there is also a pattern in some areas where users channel hop to get qam16 to avoid ack saturation. (some areas mix qpsk with qam16 on legacy) , my main point tho is to address the issue that qam16 is double the capacity of qpsk and they not equal in capability.

even if it speedtests ok it doesnt mean there is no problem, my sister gets a good upload speedtest but has obvious upload congestion.

---------- Post added at 21:14 ---------- Previous post was at 21:08 ----------

I don't agree. Take the Underground System Analogy (just to keep it unsimple!).

Northern Line from Kings X to Oval.

16QAM = Western route with 7 potholes (stations)
QPSK = Eastern route 8 potholes (stations)

Anything can happen because there is no common treatment of these routes. The 16QAM route could take longer if more people got on at each station.

Again, where am I wrong?

---------- Post added at 10:43 ---------- Previous post was at 10:37 ----------



64QAM = Less potholes
16QAM = Potholes
QPSK = More potholes

DOCSIS = The solution to assure tiered service at the given modulation.

So, no degradation because of the DOCSIS solution.

However, infrastructure that will only work at QPSK will be limited for upgrade purposes to the maximum speed that 2 bits/symbol can deliver. If you want to call that "degradation", so be it, but it is not degradation in current operational terms.

HTH.

---------- Post added at 10:47 ---------- Previous post was at 10:43 ----------


On this question of your circuit versus your sister's. I used the Northern Line analogy to cover the difference between the two of you.

My own case, having been moved for resegmentation and not put back to my original UBR, is that I move between 16QAM and QPSK on the fly. When I notice this, I do a load of quick upstream tests and compare with before. They're the same-ish with no pattern of degraded behaviour on QPSK.

This is all empirical, But DOCSIS is the governor IMO.

ok so I am not misunderstood.

I am not saying QPSK will mean a degraded service, obviously if the port isnt oversubscribed things will be fine. Some areas will have better service on QPSK than other areas on QAM16.

However to say QPSK and QAM16 have no operational differences I think is wrong there is various documents that show the differences, QAM16 doubles the size of the upstream channel over QPSK. The unknown is as I said is if VM reduce the number of users on QPSK channels.

[caph]

Sephiroth, do you use your vm connection to play online games with because if you don't then you probably won't be affected by being on qpsk modulation.

Harlow, I also do a LOT of online gaming and I'm on QPSK and have been for a long time with the same upstream power level as you. I have zero problems with online gaming but then my ping times are 21ms with a jitter of 2ms, which is much better than yours. This is why I'm saying that QPSK is no reason to assume there is a problem.

If you do drop back to QPSK then try checking your speeds as well as you pings at various times. If you can get a history of the problem then ring it in again but only quote the speed and ping issue and let them troubleshoot it that way, just saying that you want to be switched from QPSK to QAM16 will get you nowhere. You might also want to check your UBR by clicking the connection link at the top of this page and keep your eye on which UBR you are on. You may be being switched from an oversubscribed UBR to a non-oversubscribed UBR as VM address oversubscription issues.

[Sephiroth]

Well I've been shifted to another sub-net address range on rdng20 (was CPC9 and moved from rdng21 in July). So they're messing around at the Reading hub. I'm back at 16QAM and for all this shuffling around, my 20 meg service is better than ever at peak times.

cpc9-rdng20-2-0-custXXX.15-3.cable.virginmedia.com

I've concluded (not actually knowing) that QPSK is not configured in poor infrastructure areas. It's clear now that channels can be allocated to QPSK modulation to cope with any upstream SNR issues that might arise on the fly. I got the same channel number & frequency on QPSK as I did on 16QAM on many occasions. This is where I get jealous of Igni who used to work for them and has real first hand knowledge.

I'm also certain that DOCSIS is soundly designed to allow for a wide range of ongoing line conditions. And to confuse you all, look at 16QAM and 64QAM Digital Freeview modulation!

[Harlow CM20]

Harlow, I also do a LOT of online gaming and I'm on QPSK and have been for a long time with the same upstream power level as you. I have zero problems with online gaming but then my ping times are 21ms with a jitter of 2ms, which is much better than yours. This is why I'm saying that QPSK is no reason to assume there is a problem.

If you do drop back to QPSK then try checking your speeds as well as you pings at various times. If you can get a history of the problem then ring it in again but only quote the speed and ping issue and let them troubleshoot it that way, just saying that you want to be switched from QPSK to QAM16 will get you nowhere. You might also want to check your UBR by clicking the connection link at the top of this page and keep your eye on which UBR you are on. You may be being switched from an oversubscribed UBR to a non-oversubscribed UBR as VM address oversubscription issues.

Thanks for the suggestions caph, my UBR has remained the same for a very long time now, only difference apparantly is that it changed from cpc3-bsfd2-0-0-custxxx.cable.virginmedia.com to cpc3-bsfd2-0-0-custxxx.5-3.cable.virginmedia.com something to do with rDNS and shouldn't affect routes taken and ping times according to Kymmy.

Played a few games online last night for the 1st time since being on qam16 and the difference is miles apart for me, played about 6 or 7 and only lost 1 as opposed to playing that many and only winning 1. lol

TBH, I wouldn't care what US modulation I am on if I get a good level of service, it used to be qpsk for many years for me without problems but then something changed for the worse and now it just seems broken if i get qpsk.

Hopefully things stay this way for now and only get better in the future then I can remain a happy and satisfied customer :)

[IGR]

Hi all, very fascinating if a little confusing, but hey that's just me :confused:
anyhow I am having some lag problems whilst gaming and wondering if it's related to this QPSK V QAM thingy..
Looking at the sticky it would seem that some of my power levels are askew too and ask if someone would take a look at my modem levels and advise if I need to call out an engineer.. many thanks Ian ;)
PS on the 20Mb and in Lincoln..

Downstream Lock : Locked
Downstream Channel Id : 75
Downstream Frequency : 299000000 Hz
Downstream Modulation : QAM256
Downstream Symbol Rate : 6952 Ksym/sec
Downstream Interleave Depth : taps12Increment17
Downstream Receive Power Level : 0.3 dBmV
Downstream SNR : 41.8 dB

Upstream Lock : Locked
Upstream Channel ID : 2
Upstream Frequency : 45800000 Hz
Upstream Modulation : QPSK
Upstream Symbol Rate : 5120 Ksym/sec
Upstream transmit Power Level : 39.7 dBmV
Upstream Mini-Slot Size : 4

[Sephiroth]

For what my opinion's worth, you should take this to your own thread so this one can stay on topic. Your problem would get better attention because it won't be the QPSK thing that's the p[roblem.

When you do put up your own thread, please also provide your event log (taken when you're having lag problems). Tell us also if it's at certain tiomes of day and what ypu're doing when you have the lag problems. BTW, those levels are exemplary!

Cheers

[Chrysalis]

Well I've been shifted to another sub-net address range on rdng20 (was CPC9 and moved from rdng21 in July). So they're messing around at the Reading hub. I'm back at 16QAM and for all this shuffling around, my 20 meg service is better than ever at peak times.

cpc9-rdng20-2-0-custXXX.15-3.cable.virginmedia.com

I've concluded (not actually knowing) that QPSK is not configured in poor infrastructure areas. It's clear now that channels can be allocated to QPSK modulation to cope with any upstream SNR issues that might arise on the fly. I got the same channel number & frequency on QPSK as I did on 16QAM on many occasions. This is where I get jealous of Igni who used to work for them and has real first hand knowledge.

I'm also certain that DOCSIS is soundly designed to allow for a wide range of ongoing line conditions. And to confuse you all, look at 16QAM and 64QAM Digital Freeview modulation!

if your subnet changed it is possible something like a reseg occured and the same modulation etc. may be a new UBR.

[Sephiroth]

I'm sure I'm in the middle of a reseg. I'm still on the same UBR though (rdng20) as since July. And an on the fly UCD change occurred to QPSK.

Anyway, I'm researching these mechanisms and will have an answer soon!

[Duke`]

My CM just rebooted and has changed from QAM16 to QPSK.

[Harlow CM20]

My CM just rebooted and has changed from QAM16 to QPSK.


My CM did the same yesterday afternoon after being on QAM16 for a whole week. During this small time period I thought "Wow, I have Broadband again" but has now reverted back to the broken QPSK modulation and because of this I was unable to play any online games because the connection is so laggy and unresponsive.

11pm, attempt to download from a Microsoft server (confirmed that this server had no problems at all) and a 70MB file took 3-4 minutes to download on a wired connection and this is because of the change in upstream modulation.

STM couldn't be reponsible either so don't go blaming it on that or my PC or anything because we all know the problem lies on VirginMedias network.

Why does it do this? Is it because the cabinet I connect to is 17 years old and is in dire need of replacing?

Just recently upgraded to 20Mb from 10Mb because I noticed that my connection "appeared" to be fixed, now it's gonna be another phone call to put me back on 10Mb so I don't have another agonising 12 month contract of this appalling service.

Hardly any wonder that most of my neighbours have a Sky dish on their house and with this in mind shoudn't there be plenty of available bandwidth on the cabinet I connect to?

I gave you a second chance VirginMedia and you messed up again!

My contract expires in May 2011 but I'll have a complete Sky package way before then, it's not a question of money but principle.

I wanted broadband, not failband! Why does your network have to be this way? :mad:

---------- Post added at 12:56 ---------- Previous post was at 12:49 ----------

Tue Sep 28 15:50:33 2010 Tue Sep 28 15:50:33 2010 Information (7) The s/w filename specified in the config file is the same as ...
Tue Sep 28 15:50:33 2010 Tue Sep 28 15:50:33 2010 Information (7) A software upgrade filename was specified in the config file.
Tue Sep 28 15:50:32 2010 Tue Sep 28 15:50:32 2010 Information (7) Authorized
Tue Sep 28 15:50:32 2010 Tue Sep 28 15:50:32 2010 Information (7) Registration complete!
Tue Sep 28 15:50:32 2010 Tue Sep 28 15:50:32 2010 Information (7) We registered with a DOCSIS 1.1 config file!
Tue Sep 28 15:50:32 2010 Tue Sep 28 15:50:32 2010 Information (7) Received a REG-RSP message from the CMTS...
Tue Sep 28 15:50:32 2010 Tue Sep 28 15:50:32 2010 Information (7) Sending a REG-REQ to the CMTS...
Tue Sep 28 15:50:32 2010 Tue Sep 28 15:50:32 2010 Information (7) CableModem SNMP configure complete
Tue Sep 28 15:50:32 2010 Tue Sep 28 15:50:32 2010 Information (7) IP init completed ok
Tue Sep 28 15:50:32 2010 Tue Sep 28 15:50:32 2010 Information (7) CableModem TFTP init ok
Time Not Established Time Not Established Information (7) CableModem DHCP client init ok
Time Not Established Time Not Established Critical (3) DHCP WARNING - Non-critical field invalid in response.
Time Not Established Time Not Established Information (7) MAP w/initial maintenance region received
Time Not Established Time Not Established Critical (3) No Ranging Response received - T3 time-out
Time Not Established Time Not Established Information (7) MAP w/initial maintenance region received
Time Not Established Time Not Established Critical (3) No Ranging Response received - T3 time-out
Time Not Established Time Not Established Information (7) MAP w/initial maintenance region received
Time Not Established Time Not Established Information (7) Downstream sync ok
Time Not Established Time Not Established Information (7) Beginning initial ranging...
Time Not Established Time Not Established Information (7) downstream time sync acquired...
Time Not Established Time Not Established Information (7) Downstream sync ok
Time Not Established Time Not Established Information (7) starting ds time sync acquisition...
Time Not Established Time Not Established Information (7) Locked on the downstream. Waiting for UCDs...
Time Not Established Time Not Established Information (7) Downstream lock ok
Time Not Established Time Not Established Information (7) Sync Start
Time Not Established Time Not Established Critical (3) DHCP WARNING - Non-critical field invalid in response.
Time Not Established Time Not Established Critical (3) No Ranging Response received - T3 time-out
Time Not Established Time Not Established Critical (3) No Ranging Response received - T3 time-out
Tue Sep 28 13:54:43 2010 Tue Sep 28 13:54:43 2010 Critical (3) DHCP WARNING - Non-critical field invalid in response.
Tue Sep 28 13:54:28 2010 Tue Sep 28 13:54:28 2010 Critical (3) No Ranging Response received - T3 time-out
Tue Sep 28 13:54:20 2010 Tue Sep 28 13:54:20 2010 Critical (3) No Ranging Response received - T3 time-out
Tue Sep 28 13:54:18 2010 Tue Sep 28 13:54:18 2010 Critical (3) No Ranging Response received - T3 time-out

---------- Post added at 12:57 ---------- Previous post was at 12:56 ----------

Cable Modem Upstream
Upstream Lock : Locked
Upstream Channel ID : 2
Upstream Frequency : 22200000 Hz
Upstream Modulation : QPSK
Upstream Symbol Rate : 2560 Ksym/sec
Upstream transmit Power Level : 56.0 dBmV
Upstream Mini-Slot Size : 2

[Ignitionnet]

This isn't an official Virgin Media forum - you're looking for http://community.virginmedia.com if you want to contact them directly.

[Harlow CM20]

Thankyou Ignition but should I really need to resort to this?

So many phone calls and tech visits etc... I really feel let down by this company, I have made them aware several times of the issues and problems that I get but does anything get sorted?

Okay over there I go...

Hopefully I can come back here and post about a positive outcome regarding this appalling service.

[Ignitionnet]

No, but it would be far more productive than posting on here, that is the official forum and your post reads like a direct message to Virgin - that's where they are!

[Harlow CM20]

Thanks for the advice, only reason I posted on here was because when I had problems in the past they were usually resolved by a visit to this forum but that was before the official VM community forum existed.

Have started a thread on that site regarding this.

http://community.virginmedia.com/t5/Fibre-optic-broadband-cable/Unusable-Internet-for-online-gaming/td-p/142890