Equivocal Results from Network Testing Applet

[Tom Wagner <tw...@snarkboojum.com>]

I get results from the link on the OM login spalsh which are widely at 
variance with upload, download and other stats posted by other 
browser-based utilites like Ookla and TestMySpeed.  Is this feature in 
face working, and if so what exactly does it measure? ...

  * the speeds between the OM server and the client browser?
  * the speeds between the client and the web at large?

My main concern is that number posted by the OM test seem very 
pessimistic vs connectivity number for a given clients general internet 
connectivity.  E.g. A webinar participant with Comcast cable reports 
upload/download number of 3/18Mbps from Ookla.   The OM test executed 
within 5 minutes of the Ookla test posts 0.34 / 1.52Mbps.

At other client sites, quite frequently, the upload and download 
proportions are dramatically reversed.  E.g. 3/18Mbps Ookla vs 
4.2/1.3Mbps OM.

Any guidance would be appreciated, as I am trying to establish and easy 
method that potential webinar participants can qualify their internet 
hookups to set expectations for their webinar experience.

Cheers,
         Tom Wagner, Jackson Orthopedic Foundation IT

Re: Equivocal Results from Network Testing Applet

[Maxim Solodovnik <so...@gmail.com>]

Actually it measures how fast Adobe Flash client can
download/upload binary data
The test is "synthetic" random set of bytes is being uploaded/dowloaded

I can point you to the source code of these tests, maybe you can give me
the hint how this can be improved

On Fri, Jul 29, 2016 at 11:51 PM, Tom Wagner <tw...@snarkboojum.com> wrote:

> I get results from the link on the OM login spalsh which are widely at
> variance with upload, download and other stats posted by other
> browser-based utilites like Ookla and TestMySpeed.  Is this feature in face
> working, and if so what exactly does it measure? ...
>
>    - the speeds between the OM server and the client browser?
>    - the speeds between the client and the web at large?
>
> My main concern is that number posted by the OM test seem very pessimistic
> vs connectivity number for a given clients general internet connectivity.
> E.g. A webinar participant with Comcast cable reports upload/download
> number of 3/18Mbps from Ookla.   The OM test executed within 5 minutes of
> the Ookla test posts 0.34 / 1.52Mbps.
>
> At other client sites, quite frequently, the upload and download
> proportions are dramatically reversed.  E.g. 3/18Mbps Ookla vs 4.2/1.3Mbps
> OM.
>
> Any guidance would be appreciated, as I am trying to establish and easy
> method that potential webinar participants can qualify their internet
> hookups to set expectations for their webinar experience.
>
> Cheers,
>         Tom Wagner, Jackson Orthopedic Foundation IT
>



-- 
WBR
Maxim aka solomax