In reply to a post by kwikbreaks:

I'll triplicate that advice. I've no idea why but TBB always gives poor results on my 50Mbps while speedtest.net and real world downloads show it performing close to advertised rate most of the time.

The TBB founders have said there is an issue at VM's side of the link; that VM need to fix. Its been known for quite a while. See the TalkToTheStaff (TTTS) forum.

(Does make me wonder how many other issues VM need to fix).

In reply to a post by jchamier:

In reply to a post by kwikbreaks:

I'll triplicate that advice. I've no idea why but TBB always gives poor results on my 50Mbps while speedtest.net and real world downloads show it performing close to advertised rate most of the time.

The TBB founders have said there is an issue at VM's side of the link; that VM need to fix. Its been known for quite a while. See the TalkToTheStaff (TTTS) forum.

(Does make me wonder how many other issues VM need to fix).

speedtest.net don't seem to have any problems with VM. makes me think that maybe TBB are just passing the buck.

You can think we are passing the buck, but can assure you we are not.

I've seen speedtest.net report 150Meg on a VM 100Meg link, interesting result. Wonder how many of the speedtest.net servers are actually hosted on 100Meg versus GigE.

http://blog.thinkbroadband.com/2011/01/net-neutralit...
The methodology used in our tester is more likely to show that congestion up, than what is used on speedtest.net

How to fix TBB speedtest ? I try port 80 and without port 80 , no different. Getting wrong speed 17 meg out of 30 Meg on cable ?



http://www.speedtest.net/result/1154832722.png

Edited by adslmax (Sat 12-Feb-11 20:01:04)

In reply to a post by Chrysalis:

retest with the london server as we know that one is on a gigabit port, many of them are only on 100mbit ports. If london varies its most likely congestion on your UBR port. (VM problem).

The gloucester server is also gigabit.

In reply to a post by danmeah:

In reply to a post by Chrysalis:

retest with the london server as we know that one is on a gigabit port, many of them are only on 100mbit ports. If london varies its most likely congestion on your UBR port. (VM problem).

The gloucester server is also gigabit.

not bad a lower latency one from london. Iplayer was running during this test so about a mbit down.

http://www.speedtest.net/result/1155118593.png

In reply to a post by adslmax:

How to fix TBB speedtest ? I try port 80 and without port 80 , no different. Getting wrong speed 17 meg out of 30 Meg on cable ?

Ask Virgin to upgrade their transit and peering rather than shaping and hoping that will alleviate the need to?

http://www.speedtest.net/

What do you get on that one?.

In reply to a post by MrSaffron:

You can think we are passing the buck, but can assure you we are not.

I've seen speedtest.net report 150Meg on a VM 100Meg link, interesting result. Wonder how many of the speedtest.net servers are actually hosted on 100Meg versus GigE.

http://blog.thinkbroadband.com/2011/01/net-neutralit...
The methodology used in our tester is more likely to show that congestion up, than what is used on speedtest.net

the least you could do is put a notice on the speedchecker page to let VM users know that they're likely wasting their time. I know of a number of users who have spent time on the phone with support purely because of the erroneous results produced by the TBB checker.

In reply to a post by Ignitionnet:

In reply to a post by adslmax:

How to fix TBB speedtest ? I try port 80 and without port 80 , no different. Getting wrong speed 17 meg out of 30 Meg on cable ?

Ask Virgin to upgrade their transit and peering rather than shaping and hoping that will alleviate the need to?

I test it after 9pm & overnight as there shouldn't be any shaping trafic using TBB from virgin media's end.

I think TBB need to look into this and get it fixed asap.