I'm not saying you're wrong as I haven't seen any empirical evidence but have you got any facts to back that up?
Seems to me that the only customers who could get an improvement fall into three groups:
- customers who were too heavily capped manually
- customers with a high sync and very high error levels but stable
- customers with unstable connections who haven't yet been manually capped
The first is just undoing Sky's earlier heavy-handedness. The anecdotal evidence I've seen tends to show that more often than not DLM whacks an even higher cap on the connection.
The second would be quite unusual as high errors and high sync don't often go together as the router will usually train down the connection itself. There are of course connections that have a high sync but low throughput due to errors but they're rare, as again, the ADSL protocol (in this instance) has error correction and can/will retransmit unrecoverable data. If you do the maths you'll see that except in very gross cases a reduction in sync to remove errors will actually result in a lower throughput than previously. You may well see an improvement in stability and latency though.
The third would be an interesting group: they presumably feel their connection is fine but Sky takes pre-emptive action to "improve" it resulting in lower sync and probably lower throughput. You've got to be pretty confident of your algorithms to ensure that only customers who will actually benefit get this action. Again, I think there's plenty of anecdotal evidence to show that Sky didn't get that right.
In the absence of hard facts I'd certainly agree that
some customers have benefited but "a lot"?
Linear Mode
