New advertised speed rules for ISPs

[roughbeast]

I have just read in Broadband Choices that the advertising regulators are to tell ISPs that they cannot advertise a particular speed unless at least 10% of customers can actually get that speed. Why have they set the bar so low? That means that 90% will not get the advertised speed.

Even though we get plenty moans about VM here I'm certain much more than 10% of cable customers get what is advertised. Would this be true of BT and other ISPs delivering ADSL on copper?

What would be a reasonable percentage to set?


http://www.broadbandchoices.co.uk/article.asp?id=new-rules-to-end-misleading-speed-claims-290911&partner=SpeedDemonEmail~Oct11~TEST2~21~CID786891&utm_source=speeddemon&utm_medium=email&hq_e=el&hq_m=1192023&hq_l=20&hq_v=6c6aaf1480


Sorry if this info has been posted elsewhere.

[jb66]

The ISP has little control over speed, it's the quality of your bt phone line so I ink it's fair

[roughbeast]

The ISP has little control over speed, it's the quality of your bt phone line so I ink it's fair


This is not about the speed ISPs can muster. It is about the speed they advertise.

[Sirius]

This is not about the speed ISPs can muster. It is about the speed they advertise.

Indeed,

So if BT or any other provider advertise up to 21 meg then it means 10% of there customers must be able to get that speed or they cannot advertise that speed, I think the adsl providers are going to have problems as the copper cabling will have a significant impact on there ability to achieve high speeds for over 10% of there customer base.

So how are they going to prove what speeds are provided.

[jb66]

So why advertise a speed on adsl, it means nothing. As fast as your line can handle seems better

[roughbeast]

So how are they going to prove what speeds are provided.


I guess some standard such as that provided by SamKnows will have to be used. How accurate is that?

They would also have to have a large proportion of customers monitored to arrive at statistically sound data. SamKnows is in a good position to do this with its embedded testing software. http://www.samknows.com/broadband/how-it-works I do wonder at the expense of such a project just so the ISPs can be forced to tell the truth.

[jb66]

I guess some standard such as that provided by SamKnows will have to be used. How accurate is that?

Send them out to folk with short lines

[roughbeast]

Send them out to folk with short lines

Surely the regulator would have to have a role in that decision.

[Sirius]

I guess some standard such as that provided by SamKnows will have to be used. How accurate is that?

They would also have to have a large proportion of customers monitored to arrive at statistically sound data. SamKnows is in a good position to do this with its embedded testing software. http://www.samknows.com/broadband/how-it-works I do wonder at the expense of such a project just so the ISPs can be forced to tell the truth.

I am on the samknows trial.

The spikes every hour is the testing and the results are for a 50 meg shub in router mode

https://www.cableforum.co.uk/images/local/2011/10/10.png (http://www.thinkbroadband.com/ping/share/948acfaeb3ff4407e84c67a07348a607.html)

[kwikbreaks]

Fortunately for VM I'm not on the SamKnows trial..

Download Failed (1)

The spikes and frequent periods with speeds of less than 10Mbps on a 50Mbps connection prove the efficiency of VM advertising or alternately their inefficiency in matching capacity to customer numbers. I did no downloading of any note during that 24 hour period.

[AbyssUnderground]

CAP also published new rules to prevent providers from advertising “unlimited” broadband that comes with fees or penalties if you download too much.

So now they MUST state you will be throttled if you download too much? If so, GOOD. I'm happy they will finally stop getting around that fact with very vague small print.