You mean here. Where potential customers expect detailed information they get it, instead of glossy pictures and having to hunt for it in a fairly expert way.

That's the one!

Thanks to Mr Saffron & everyone else who has tried to help me through this technical maze - even those who think it�s ok to shout �FALSE!� at any passing Newbie!

Thanks especially to vimto_girl who I felt had actually read my posts rather than just reacted to them.

I will keep this thread updated if / when there are any significant developments. Meantime I continue swotting for my Advanced Broadband exams.

vimto_girl is quite the authority on broadband

Just to show it can go the other way I am on Infinity 2 and get 50/16. So around a 3:1 ratio - but I can't use that to ask BT to increase my downstream to match a ratio to the upstream as the VDSL is using whatever frequencies are available within the specs and just happens to be hitting this speed.

In my case the downstream is within BT's estimates but the upstream is a bit quicker than their highest estimate. The actual ratio for a specific line is down to physics and not mathematics.

Thanks ian72, a very interesting example. Yes, I certainly now accept it's physics not maths. However! While, as you say, you can't use these numbers to persuade BT to up your Down (so to speak!) - I still believe that if an engineer / DLM techie was prepared to go the extra mile and personally "tweak your frequencies"... that a higher Down could probably be achieved. Such individual intervention and tweaking is presumably not commercially viable, so I accept it ain't gonna happen. So: yes it's physics, but also, as always, money.

Despite all the education I have received here and elsewhere, I remain convinced that - for reasons I have speculated about previously - the DLM system has simply not tried anything higher than 3Mbps on my line, and that if it did, the line would cope, and about 6Mbps that could be achieved.

But I accept it ain't gonna happen, and I can already hear the cries of "FALSE" echoing across these mountains!

I guess my subtext here is about how much our lives are now ruled and regulated by poorly executed programs and imperfect algorithms. "I'm sorry sir, but the system won't let me do that..." "I'm sorry that's the maximum limit the computer's allowing me..." etc etc

All the progress in my case has been through conscientious individuals... Latest score: Humans: 4, Computers: 0!

The bins that are allowed to be used for up and down are set in the standards - this is to ensure that crosstalk is limited as it means that the same frequencies are working in the same directions for all lines.

The only way we would know for sure if there is capacity for the DLM to increase your upload would be to get the full stats of attenuation, SNR sync rate, and power for the connection. This would show if there was any potential for increases via a kick to the DLM. However, others more knowledgeable on here around stats (like MrS) suggest that your sync speed is in line with what would be expected for the length of the line.

In reply to a post by mayavirupa333:

I still believe that if an engineer / DLM techie was prepared to go the extra mile and personally "tweak your frequencies"... that a higher Down could probably be achieved. Such individual intervention and tweaking is presumably not commercially viable, so I accept it ain't gonna happen. So: yes it's physics, but also, as always, money.

As ian72 says, this sort of individual tweaking can't be done, and the reason is physics not money.

The VDSL2 bandplans have each frequency bin allocated for downstream or upstream use. No bins are available for "either way on demand", as that would cause the performance of all lines in the bundle to degrade owing to higher crosstalk. For acceptable crosstalk performance, you need the strong signal from the generating device to be either at the user end or the DSLAM end of the pair. If the strong signal in a bin could be at either end of the line, the strong signals would seriously degrade the weaker signals in the bin running in the opposite direction.

The bandplans and power masks have been carefully chosen and fixed by regulators and standards in order to ensure the best possible performance overall.


As ian72 says, if you post the line statistics (including per bin plots if you have them), people will be able to comment on the performance of your connection, though it seems that your connection is likely performing as fast as expected.

Unfortunately for you, it seems your line's performance is limited in first large block of upstream bins and has likely petered out before reaching the second large block of upstream bins, hence the relatively slow upstream connection. A substantial improvement will require bonding multiple connections together or a future enhancement to the network that is not likely in the near term (such as deployment of FTTrN).

In reply to a post by mayavirupa333:

I remain convinced that - for reasons I have speculated about previously - the DLM system has simply not tried anything higher than 3Mbps on my line, and that if it did, the line would cope, and about 6Mbps that could be achieved.

I tend to agree that more is probably available. It was just your insistence on a 4:1 ratio despite several people saying it didn't work like that which brought me in on the same lines. Whereupon unnecessary and unhelpful to you mayhem erupted. Water under the bridge I hope.

A few people have mentioned the bin plots and so on.

If you obtain an HG612 Openreach modem, there are usually several around on eBay, then you can unlock and use that and many diagnostic stats statistics can be produced. There are two different sets of programs well known on these forums that can automate collection, and produce useful graphs.

The BT Home Hub can be fed from that via the WAN port. If you decide to try it, do not pay well over the odds for one that is already unlocked, as it's a simple process. (Not so the ECI modem). You would need a 3B version.

At a simple level, as I don't use either at the moment though I did use one a while ago, just using telnet the bins in use can be identified, though not how much is used in each:-

# xdslcmd info --pbParams
xdslcmd: ADSL driver and PHY status
Status: Showtime
Retrain Reason: 0
Last initialization procedure status: 0
Max: Upstream rate = 14810 Kbps, Downstream rate = 55932 Kbps
Bearer: 0, Upstream rate = 14832 Kbps, Downstream rate = 57095 Kbps

Discovery Phase (Initial) Band Plan
US: (7,32) (871,1205) (1972,2782)
DS: (33,859) (1216,1961) (2793,3970)
Medley Phase (Final) Band Plan
US: (7,32) (871,1205) (1972,2780)
DS: (33,859) (1216,1961) (2793,3970)
VDSL Port Details Upstream Downstream
Attainable Net Data Rate: 14810 kbps 55932 kbps
Actual Aggregate Tx Power: 7.4 dBm 13.5 dBm
====================================================================================
VDSL Band Status U0 U1 U2
U3 U4 D1 D2 D3
Line Attenuation(dB): 5.2 29.1 44.0 N/A N/A 14.3 35.8 55.0
Signal Attenuation(dB): 5.2 28.1 42.9 N/A N/A 18.8 35.5 55.1
SNR Margin(dB): 6.1 6.1 6.1 N/A N/A 5.8 5.9 5.8
TX Power(dBm): 0.4 -13.6 6.4 N/A N/A 11.0 8.1 5.5
#

The discovery phase is finding the available bins. The medley phase is after which ones are viable has been decided. As you can see, mine are nearly identical. Yours might have significant differences.

Very helpful input from ian72, David_W & RobertoS. These are definitely the kind of answers I was hoping to find here. Planning to get an HG612 shortly so I can boggle everyone with stats & charts! Meantime I've got some intensive plumbing, so back in a week or so.