Broadband speed tests questioned......

Here we are folks,

http://news.bbc.co.uk/1/hi/technology/7669713.stm

now we have the truth.....!!!

Oddi wrote:
> Here we are folks,
>
> http://news.bbc.co.uk/1/hi/technology/7669713.stm
>
> now we have the truth.....!!!
Well they would say that. wouldn't they?

Mark McIntyre wrote:
> The Natural Philosopher wrote:
>> Oddi wrote:
>>> Here we are folks,
>>> http://news.bbc.co.uk/1/hi/technology/7669713.stm
>>>
>>> now we have the truth.....!!!
>> Well they would say that. wouldn't they?
>>
> On the other hand, do you not think they have a point? And if you read
> through the article you'll see that speed test site owners agree to an
> extent.


"tests show that car speedometers are not accurate"

In the small print "most car speedometers over read by around 5%-7% to
remove any possibility that cars will inadvertently exceed speed limits"


From 'DailyBroadbandcars"

"Motorists have been amused by the sight of Virgin Broadband Trabants
running at 45mph on the motorway: The police on stopping one of them
questioned the driver who replied that he was sorry to have been doing
80mph, as he had not looked at his speedometer until the police flagged
him down"

"The Natural Philosopher" <(E-Mail Removed)> wrote in message
news:(E-Mail Removed)
> Oddi wrote:
>> Here we are folks,
>>
>> http://news.bbc.co.uk/1/hi/technology/7669713.stm
>>
>> now we have the truth.....!!!
> Well they would say that. wouldn't they?


I tried a few last week as it happens and I wrote the speeds down,
because I've had my doubts about the accuracy of them for a bit
because I use Internet Download Manager, and whenever I've tried a
broadband speed test they always seem to underestimate the speed
compared to what the speed says in Internet Download Manager. I
usually use iTunes to measure the speed, because its servers tend to
be fast. These are teh speeds I saw last week:

iTunes 4713 kbps (that's calculated by dividing the file size by the
time taken to download the file, so the actual download speed should
be say 10% or so higher than that if you take packet headers into
account)

http://www.speedtest.bbmax.co.uk/
3554 kbps
3790
2741
3952

http://www.top10-broadband.co.uk/speedtest/
5930

http://www.thinkbroadband.com/speedtest/
2187
2279

http://www.speedtest.net/
3967

http://resources.zdnet.co.uk/speedtest/
3858
3334




--
Steve - www.digitalradiotech.co.uk - Digital Radio News & Info

The adoption of DAB was the most incompetent technical
decision ever made in the history of UK broadcasting:
http://www.digitalradiotech.co.uk/da...ion_of_dab.htm

DAB sounds worse than FM wrote:
> "The Natural Philosopher" <(E-Mail Removed)> wrote in message
> news:(E-Mail Removed)
>> Oddi wrote:
>>> Here we are folks,
>>>
>>> http://news.bbc.co.uk/1/hi/technology/7669713.stm
>>>
>>> now we have the truth.....!!!
>> Well they would say that. wouldn't they?
>
>
> I tried a few last week as it happens and I wrote the speeds down,
> because I've had my doubts about the accuracy of them for a bit
> because I use Internet Download Manager, and whenever I've tried a
> broadband speed test they always seem to underestimate the speed
> compared to what the speed says in Internet Download Manager.



Internet download manager uses multiple streams. I'm pretty sure
Thinkbroadband only uses a single stream.

If a connection is heavily congested a speedtest will show a higher
total from using multiple streams. While this is ok for Download Manager
a lot of real life internet applications can only make use of a single
stream.

The figures you post tend to suggest that your line is congested.

I actually tested a VirginMedia line and a Zen line simultaneously
against Thinkbroadband.

In the morning the VM ran at ~19Mb/s and Zen ran at 4Mb/s.

At peek times VM ran as low as 1Mb/s where as the Zen line continued to
show 4Mb/s.



I
> usually use iTunes to measure the speed, because its servers tend to
> be fast. These are teh speeds I saw last week:
>
> iTunes 4713 kbps (that's calculated by dividing the file size by the
> time taken to download the file, so the actual download speed should
> be say 10% or so higher than that if you take packet headers into
> account)
>
> http://www.speedtest.bbmax.co.uk/
> 3554 kbps
> 3790
> 2741
> 3952
>
> http://www.top10-broadband.co.uk/speedtest/
> 5930
>
> http://www.thinkbroadband.com/speedtest/
> 2187
> 2279
>
> http://www.speedtest.net/
> 3967
>
> http://resources.zdnet.co.uk/speedtest/
> 3858
> 3334
>
>
>
>

DAB sounds worse than FM wrote:
> http://www.thinkbroadband.com/speedtest/
> 2187
> 2279

This URL is wrong: should be
http://www.thinkbroadband.com/speedtest.html. Incidentally, I couldn't
help but notice that cited results with this service are very
significantly lower than the others. Any reason why?

Also, I asked about the SamKnows methodology
(http://www.samknows.com/broadband/pm/PM_Summer_08.pdf) discussed in the
BBC article in this ng a few weeks back. Nobody posted. Any good?
Better? Ofcom seem convinced anyway...

Chris

"Nick" <(E-Mail Removed)> wrote in message
news:(E-Mail Removed)
> DAB sounds worse than FM wrote:
>> "The Natural Philosopher" <(E-Mail Removed)> wrote in message
>> news:(E-Mail Removed)
>>> Oddi wrote:
>>>> Here we are folks,
>>>>
>>>> http://news.bbc.co.uk/1/hi/technology/7669713.stm
>>>>
>>>> now we have the truth.....!!!
>>> Well they would say that. wouldn't they?
>>
>>
>> I tried a few last week as it happens and I wrote the speeds down,
>> because I've had my doubts about the accuracy of them for a bit
>> because I use Internet Download Manager, and whenever I've tried a
>> broadband speed test they always seem to underestimate the speed
>> compared to what the speed says in Internet Download Manager.
>
>
>
> Internet download manager uses multiple streams. I'm pretty sure
> Thinkbroadband only uses a single stream.
>
> If a connection is heavily congested a speedtest will show a higher
> total from using multiple streams. While this is ok for Download
> Manager
> a lot of real life internet applications can only make use of a
> single
> stream.


Fair enough then. But Thinkbroadband's speed test results were both
far lower than the other results. I did the tests in quick succession,
and some other test results were all over the place, so I think it's
fair to say that they're not very accurate.


> The figures you post tend to suggest that your line is congested.
>
> I actually tested a VirginMedia line and a Zen line simultaneously
> against Thinkbroadband.
>
> In the morning the VM ran at ~19Mb/s and Zen ran at 4Mb/s.
>
> At peek times VM ran as low as 1Mb/s where as the Zen line continued
> to
> show 4Mb/s.


I'm on VM ADSL, which according to the results from the Samknows
report is one of if not the worst ISP to use.


> I
>> usually use iTunes to measure the speed, because its servers tend
>> to
>> be fast. These are teh speeds I saw last week:
>>
>> iTunes 4713 kbps (that's calculated by dividing the file size by
>> the
>> time taken to download the file, so the actual download speed
>> should
>> be say 10% or so higher than that if you take packet headers into
>> account)
>>
>> http://www.speedtest.bbmax.co.uk/
>> 3554 kbps
>> 3790
>> 2741
>> 3952
>>
>> http://www.top10-broadband.co.uk/speedtest/
>> 5930
>>
>> http://www.thinkbroadband.com/speedtest/
>> 2187
>> 2279
>>
>> http://www.speedtest.net/
>> 3967
>>
>> http://resources.zdnet.co.uk/speedtest/
>> 3858
>> 3334



--
Steve - www.digitalradiotech.co.uk - Digital Radio News & Info

The adoption of DAB was the most incompetent technical
decision ever made in the history of UK broadcasting:
http://www.digitalradiotech.co.uk/da...ion_of_dab.htm

DAB sounds worse than FM submitted this idea :
> http://www.speedtest.bbmax.co.uk/
> 3554 kbps
> 3790
> 2741
> 3952
>
> http://www.top10-broadband.co.uk/speedtest/
> 5930
>
> http://www.thinkbroadband.com/speedtest/
> 2187
> 2279
>
> http://www.speedtest.net/
> 3967
>
> http://resources.zdnet.co.uk/speedtest/
> 3858
> 3334

All but the zdnet one(5.1) produce 6.3 to 6.5 here with an 8Mb line
which is syncing at 7616.

--
Regards,
Harry (M1BYT) (L)
http://www.ukradioamateur.co.uk

DAB sounds worse than FM wrote:
> "Nick" <(E-Mail Removed)> wrote in message
> news:(E-Mail Removed)
>> DAB sounds worse than FM wrote:
>>> "The Natural Philosopher" <(E-Mail Removed)> wrote in message
>>> news:(E-Mail Removed)
>>>> Oddi wrote:
>>>>> Here we are folks,
>>>>>
>>>>> http://news.bbc.co.uk/1/hi/technology/7669713.stm
>>>>>
>>>>> now we have the truth.....!!!
>>>> Well they would say that. wouldn't they?
>>>
>>> I tried a few last week as it happens and I wrote the speeds down,
>>> because I've had my doubts about the accuracy of them for a bit
>>> because I use Internet Download Manager, and whenever I've tried a
>>> broadband speed test they always seem to underestimate the speed
>>> compared to what the speed says in Internet Download Manager.
>>
>>
>> Internet download manager uses multiple streams. I'm pretty sure
>> Thinkbroadband only uses a single stream.
>>
>> If a connection is heavily congested a speedtest will show a higher
>> total from using multiple streams. While this is ok for Download
>> Manager
>> a lot of real life internet applications can only make use of a
>> single
>> stream.
>
>
> Fair enough then. But Thinkbroadband's speed test results were both
> far lower than the other results. I did the tests in quick succession,
> and some other test results were all over the place, so I think it's
> fair to say that they're not very accurate.
>

That assumes the other speedtesters also only use one connection. I'm
almost certain that http://www.speedtest.net/ uses at least two connections.

>
>> The figures you post tend to suggest that your line is congested.
>>
>> I actually tested a VirginMedia line and a Zen line simultaneously
>> against Thinkbroadband.
>>
>> In the morning the VM ran at ~19Mb/s and Zen ran at 4Mb/s.
>>
>> At peek times VM ran as low as 1Mb/s where as the Zen line continued
>> to
>> show 4Mb/s.
>
>
> I'm on VM ADSL, which according to the results from the Samknows
> report is one of if not the worst ISP to use.
>
>
The Samknows stuff looks interesting and I agree it would be better to
form an opinion based on this kind of testing.

Mark McIntyre wrote:
> The Natural Philosopher wrote:
>> Mark McIntyre wrote:
>>> The Natural Philosopher wrote:
>>>> Oddi wrote:
>>>>> Here we are folks,
>>>>> http://news.bbc.co.uk/1/hi/technology/7669713.stm
>>>>>
>>>>> now we have the truth.....!!!
>>>> Well they would say that. wouldn't they?
>>>>
>>> On the other hand, do you not think they have a point? And if you
>>> read through the article you'll see that speed test site owners agree
>>> to an extent.
>>
>> "tests show that car speedometers are not accurate"
>
> If my car went faster than 120mph my speedo wouldn't be at all accurate
> - since it can't go higher than that.
> Are you trying to make VM's point for them?
>
>> "Motorists have been amused by the sight of Virgin Broadband Trabants
>
> I've no idea what point you were trying to make with this.

Really? surely you cant be that slow..must be on Virgin..