Virgin to give new Superhub out. - Page 3

BenMcr · 07-12-2010, 10:39

Quote:

Originally Posted by Ignitionnet

No you got it spot on, see above.

I said it was upstream bonding that was the difference, rather than the downstream which you said

Ignitionnet · 07-12-2010, 10:41

Quote:

Originally Posted by BenMcr

I said it was upstream bonding that was the difference, rather than the downstream which you said

Ah, then you fail for misreading my post and I fail for misreading your misquote of my post.

I'm full of man-flu this morning so I'll stick to blaming that.

It's all about the downstream though there won't be anything straining the 4 upstream channel bonding capability for a while yet.

pip08456 · 07-12-2010, 10:49

Now I understand, I blame Ben's misquote!

I had an idea you'd be about and thanks for chipping in and explaining.

Cure for man flu:- Whiskey honey and hot water. Plenty of it!!!

**Hugh** · 07-12-2010, 11:10

The hot water is optional....

pip08456 · 07-12-2010, 11:20

Quote:

Originally Posted by Hugh

The hot water is optional....

And the plenty referred to the whiskey!

qasdfdsaq · 07-12-2010, 12:49

There's still an S and an M but you can't buy either of them, the only thing currently available to buy or be switched to is L XL XXL or XXXL. Though apparently they've called it 100meg instead of XXXL.

whizzard · 07-12-2010, 13:31

Quote:

Originally Posted by qasdfdsaq

There's still an S and an M but you can't buy either of them, the only thing currently available to buy or be switched to is L XL XXL or XXXL. Though apparently they've called it 100meg instead of XXXL.

'100' being its proper name.

pip08456 · 07-12-2010, 13:36

Quote:

Originally Posted by qasdfdsaq

There's still an S and an M but you can't buy either of them, the only thing currently available to buy or be switched to is L XL XXL or XXXL. Though apparently they've called it 100meg instead of XXXL.

Slight aside as I couldn't find the post where you mentioned it.

You said clearing the browser cache would make no difference to results from speedtest.net.

Care to explain why the following is happening?

All before clearing the cache. Now after doing so.

Comments welcome.

MakaveLee · 07-12-2010, 14:37

Quote:

Originally Posted by pip08456

Slight aside as I couldn't find the post where you mentioned it.

You said clearing the browser cache would make no difference to results from speedtest.net.

Care to explain why the following is happening?

All before clearing the cache. Now after doing so.

Comments welcome.

I remember seeing somewhere that Kaspersky can cause this, try disabling it if you have it?

pip08456 · 07-12-2010, 14:53

Don't be silly!

Yes it does and no I don't use it. Still doesn't explain how clearing the cache corrects the disparity.

If I had Kaspersky it would make no difference unless I disabled it during the test.

jb66 · 07-12-2010, 14:59

Maybe it's because the file used to test the speed is stored in your web browser cache so it isn't actually downloading from the net

pip08456 · 07-12-2010, 15:08

Hence the need to clear the cache which qasdfdsaq said was unnecessary and why I quoted his post for him to explain why.

Both I you and Masque know why, I just wanted his reasoning behind it.

I just wish I could remember which thread his original post was in where he said it was not necessary as speedtest.net had worked around that problem (not exact words).

Found it.

Quote:

Originally Posted by qasdfdsaq

Browser cache has nothing to do with it. People should stop suggesting to empty the browser cache when speedtest.net's technology already bypasses it. There's no way any browser cache would ever affect the results.

On the other hand Kaspersky (at least with any of the networking components turned on) does make your download results jump higher than they should, but does not usually affect upload.

But I do think the first guy was actually on the 200mb trial by the sounds of his post.

qasdfdsaq · 07-12-2010, 19:17

Coincidence? Some weird interaction between Kaspersky and your browser? I can't explain what's happening on your computer without seeing your computer.

My reasoning is that Speedtest.net does it's speed tests in a flash app, not in the browser, and it does not do file transfers through the browser, so in both respects the browser does not do the data retrieval (though it does do the socket handling) so shouldn't be caching it.

The speed tester itself also has built in "cache-prevention" mechanisms that explicitly preventing browsers caching the data. However they do admit " In rare circumstances some security software can act as a proxy, delivering the speed test payload from cache, meaning from the local memory on the client machine"

So:
a) The web browser isn't processing the data and shouldn't cache it
b) Browsers that do process it are specifically told not to cache it
c) Browsers that ignore that are still subject to "cache prevention" measures, whatever they are
d) Antivirus/firewall/filtering programs may ignore this still

As for Kaspersky we know it does d) but in relation to your tests, Kaspersky may clear it's own cache when you clear your browser cache - this would make sense - but this is just a random guess as to an explanation.

It's still the one (known) exception though, without Kaspersky browser cache should have no effect on speedtest.net results. If you have Kaspersky, exiting Kaspersky will more reliably fix the bogus results than clearing your browser cache. Cache clearing may have worked for you, but never did for me as long as I left Kaspersky on. You still don't need to clear the cache.

Ignitionnet · 07-12-2010, 20:08

Quote:

Originally Posted by qasdfdsaq

Coincidence? Some weird interaction between Kaspersky and your browser? I can't explain what's happening on your computer without seeing your computer.

My reasoning is that Speedtest.net does it's speed tests in a flash app, not in the browser, and it does not do file transfers through the browser, so in both respects the browser does not do the data retrieval (though it does do the socket handling) so shouldn't be caching it.

Nope.

GET /st/speedtest/random2000x2000.jpg?x=1291755889659-3 HTTP/1.1
Host: trancefm.co.uk
Connection: keep-alive
Referer: http://cdn.speedtest.net/flash/speedtest.swf?v=330847
Accept: */*
User-Agent: Mozilla/5.0 (Windows; U; Windows NT 6.1; en-US) AppleWebKit/534.10 (KHTML, like Gecko) Chrome/8.0.552.215 Safari/534.10

The timing is done by the app, the download is a standard browser initiated HTTP GET.

pip08456 · 07-12-2010, 20:29

Quote:

Originally Posted by qasdfdsaq

Coincidence? Some weird interaction between Kaspersky and your browser?

Quote:

Originally Posted by pip08456

Don't be silly!

Yes it does and no I don't use it. Still doesn't explain how clearing the cache corrects the disparity.

If I had Kaspersky it would make no difference unless I disabled it during the test.

As I don't have Kasapersky it would be interesting to see how!

FYI I have ESET Smart Security v4.xxx Business Edition all bought and paid for.