Speed test differences

[joe]

I have been doing several speed tests recently both thinkbroadband and BT. This afternoon TB showed



BT (attached) showed d'ld 58.64Mbps. Does that imply that the BT test is using IPv6?

Athough IPv6 shows a higher speed when I d'ld a 1Gb test file from TB and it took a long time (10mins) although it purports to using IPv6 as the default.

[Steve]

I fairly certain the BT speedtest is IPv4 only.

[Technical Ben]

"The HTTP x6 test carries out six downloads at the same time to help maximise the usage on the connection."
It's not IPV6 that effects the speed/test results is it? Or am I missing something? I do often read things back to front (causes no end of trouble  ), but I think the test is independent of IP protocol and routing (can use either/or). Though I note it does show the type of IP use in the speed tester. But this is just to get the correct route/location, and does not effect speed AFAIK (well, perhaps some tiny overhead difference?).

[edit], Ah, turns out you can get small improvements in IPV6. So I guess the differences could be on account of that at times. Though I'd assume server/routing differences are going to be the biggest impact these days.

So I think I'd better go to sleep and stop trying to decipher posts and graphs at this time of night!