View Single Post
Old 21-02-2012, 14:20   #57
Andrewcrawford23
Inactive
 
Join Date: Aug 2005
Location: Scotland
Age: 42
Services: Virgin Media - XL Plus package with XXL broadband SKY HD Multiroom Freeview HD Freesat HD
Posts: 2,816
Andrewcrawford23 has reached the bronze age
Andrewcrawford23 has reached the bronze ageAndrewcrawford23 has reached the bronze ageAndrewcrawford23 has reached the bronze ageAndrewcrawford23 has reached the bronze age
Re: Speed test sites and why you shouldnt take them for granted

Quote:
Originally Posted by qasdfdsaq View Post
Thing is, if you're getting 27mbps on a 100mb line, it's not intentional, it's because of congestion that your line can't manage more than that. As with everything else congestion, effects will vary by time of day and path to server. A VM only server would probably be sat on the VM network, and therefore not be affected by VM's inability to provide sufficient outside capacity, and the fact that 6am is hardly peak time.

---------- Post added at 13:36 ---------- Previous post was at 13:33 ----------


I'd agree partly. You'll see from another thread of mine that congestion to/from TBB's network on VM is actually a lot lower than to/from other locations (I only tested BBC and Multiplay) - TBB not only has a significantly lower ping but also much lower average ping and standard deviation. It looks to me VM have deliberately prioritized or expanded the path to TBB.

On the other hand congestion control, or in general TCP rate control algorithms, as you say aren't the same everywhere. While there are standards (and 4-5 "standard" algorithms) not everyone uses the same standard. Indeed, different browsers on the same system can yield vastly different results because of how they set their buffers differently.
i think you might be right that vigi might have done something to TBB because i only average about 20mb on there before last year, but now i am averageing 45mb the hgihes ti have ever got on it
Andrewcrawford23 is offline   Reply With Quote