|
|
On the usal think big speed tester.
down 30.8mbps up 7.1mbps
On the new flash speed tester
down 35.68mbps up 8.15mbps
Tried a few times a marked difference every time why?
Checked on other flash based tester they are comparable to think b flash tester.
Edited by deleted (Fri 12-Nov-10 01:41:57)
|
|
|
Difference between java and flash, for some Java implementations it may be the bottleneck depending on the PC
Also not 100% sure but flash tester may use multiple threads
|
|
The author of the above post is a thinkbroadband staff member. It may not constitute an official statement on behalf of thinkbroadband.
|
|
|
I get pretty much the same speeds up and down on both. I prefer the Java one simply for the results being logged in my profile.
|
|
Register (or login) on our website and you will not see this ad.
|
|
|
Just to be difficult, tests a few minutes ago gave me 28Mbps with Java and 22Mbps with Flash... comms is weird
|
|
The author of the above post is a thinkbroadband moderator but it does not constitute an official statement on behalf of thinkbroadband.
|
|
|
|
just this morning did 3 spedtests - am on fttc
first was on bt speedtester and it gave approx 32 down / 7 up
next was speedtester.net 34/6
then thinkbroadband it gave 13/2 !!!!
why is there ALWAYs this sort of difference when i run TBB speed tests ??
|
|
|
Myself and many others have brought this discrepancy up countless times and got nowhere, good luck
Sky
Any ISP that thinks that selling my click traffic is acceptable is MisinPHORMed
|
|
|
|
I have no problem with my speeds but recently, in a quite moment, I did several tests, thinkbroadband results were by far the lowest. Why should this be?
|
|
|
The only answer I can give to that is- I have no idea
For me it goes in phases- a few days when it shows speeds as expected, then perhaps a week or so showing about 60% of those speeds. I know others (not on this bbs) who get not dissimilar results.
It's not the tbb speedtester itself, that's perfectly capable of handling FTTC speeds. My hunch is that there's something odd about the routing of data from VDSL lines, I never saw this effect on ADSL2+. But it's no more than a hunch, I don't know enough about networking on a national scale for it to be more than that.
|
|
The author of the above post is a thinkbroadband moderator but it does not constitute an official statement on behalf of thinkbroadband.
|
|
|
I have no problem with my speeds but recently, in a quite moment, I did several tests, thinkbroadband results were by far the lowest. Why should this be?
TBB's Java speedtest is only single threaded.
Try the new Flash speedtest:
http://www.thinkbroadband.com/speedtest/broadband.html
Some ISPs and possibly some security software show marked slowdowns on single threads.
--
James - be* pro, on THFB exchange with a Draytek 2820Vn and a BeBox (585v7) BQM
|
|
|
It's not the tbb speedtester itself, that's perfectly capable of handling FTTC speeds. My hunch is that there's something odd about the routing of data from VDSL lines, I never saw this effect on ADSL2+. But it's no more than a hunch, I don't know enough about networking on a national scale for it to be more than that.
I don't think its the routing tech or the DSL that affects it. I'm sure its more local to PC setup, either security software or [censored] drivers etc. Or some ISP rate limiting per stream.
I can occasionally get a 20mbps sync on my line (3db margin, fast path, adsl2+) and with BE and my hardwired Core i7 desktop there is virtually no difference in the testers.
So I have no problem with the owners saying the tester works (and I've seen it work on a 20megabit corporate service without problem) and I suspect its just congestion. Maybe multiple streams handle congestion better.
Edit to add - multi streams simulates much more general web surfing; so to say the single stream is better assumes downloading without any accelerator. Many sites force you to use their accelerator (e.g. Adobe, or Microsoft).
--
James - be* pro, on THFB exchange with a Draytek 2820Vn and a BeBox (585v7) BQM
Edited by jchamier (Sat 13-Nov-10 17:50:57)
|
|
|
I'm pretty sure it's not my end (iMac, quad core i5, Airport Express, LAN is gigabit), I never saw it on ADSL2+ either, IDNet guarantee no congestion on their network and don't throttle anything, and if it's not the tbb end that only leaves the bit in the middle- BT
Re your reply to Sylcol about tbb's Flash speedtester- it's slower than the Java one for me
eta- Congestion... at 3am?
Edited by billford (Sat 13-Nov-10 18:04:17)
|
|
The author of the above post is a thinkbroadband moderator but it does not constitute an official statement on behalf of thinkbroadband.
|
|
|
I'm pretty sure it's not my end (iMac, quad core i5, Airport Express, LAN is gigabit), I never saw it on ADSL2+ either, IDNet guarantee no congestion on their network and don't throttle anything, and if it's not the tbb end that only leaves the bit in the middle- BT 
Similar to my setup, Dell Core i7 PC, Win 7 Pro x64, Gigabit connected to port on Draytek 2820. No congestion here on BE.
Re your reply to Sylcol about tbb's Flash speedtester- it's slower than the Java one for me 
eta- Congestion... at 3am?
I doubt congestion, especially at 3am. Still makes no sense.
On my link right now:
Flash = 14.86 Mbps download, 1.02 Mbps upload
Java = 14.5 Mbps download, 1.1 Mbps upload
So pretty much the same, given rounding differences at the 2nd decimal.
--
James - be* pro, on THFB exchange with a Draytek 2820Vn and a BeBox (585v7) BQM
|