Re: [hylafax-users] what is an acceptable rate of failure ?

["Andrew Rinaldi" <andrew.rinaldi@xxxxxxxxxxxx>]

Hi George,

>Your absolutely right on how we should test it. Though I can only influence
the faxserver itself and the modem used. Hmm I'd probably have to build
>another faxserver for testing and somehow get it on a POTS (which in my
field requires a bunch of paper work :( 

I fully understand that you can only control certain parts of this setup.
In my opinion you should be able to get 'your' part working 'perfectly'...
however, the only way I know of doing this in any meaningful timescale is to
be methodical in your testing.  It's also important to stress that every
time you add a variable which creates 'distortion' in the audio path between
fax machine and fax server then you reduce your chances of a successful
deployment (VoIP is a good example of this).

I also think it's important to understand that everyone in the fax business
measures 'success' differently.  Indeed, I know of several fax servers and
'high end' fax boards which don't even report some errors to the user...
indeed, some systems simply retry a failed attempt a number of times before
reporting any error to the user.  As a company, we try to eliminate as many
variables as possible in our comparison testing.... so, we use the same fax
machines, line emulators, test systems, and ITU-T test documents, when
comparing fax applications on our hardware... we even go to the extent of
analysing the audio to determine call duration in some test scenarios.

Unfortunately, when moving into the 'real' world all you can really do is
test with as many different fax machines and line types as possible.... here
the biggest single factor in successful deployments is a willingness and
ability to resolve problems.

Regards

Andrew Rinaldi
Mainpine Developer Support
USA +1 503 822 9944 | UK +44 8458 909438
andrew.rinaldi@xxxxxxxxxxxx | www.mainpine.com

<<attachment: winmail.dat>>