Possible Bug

[David]

Hello

I encountered a possible bug using the following equation:
ret_loss=20*log10((vswr-1)/(vswr+1))

Solving for ret_loss with a value of 1.22 for vswr gives the correct
result of -20.0786 but solving for vswr the result is 0.00100007. In
fact for any value entered for RL, results in a value of 0.00100007
for vswr. Can you help?

Thanks,
David

[hoymoon]

Hello David,

Thank you for pointing out the problem with the equation you're
solving.

We investigated, and found a pretty elusive bug that is triggered with
a particular combination of a non-continuous equation (like yours),
the standard intervals in which Formulaic searches for a solution, and
how division by zero is handled on the iPhone.

When Formulaic can't find a solution using the standard ranges in
which is searches, it should prompt for a user guess. It didn't in the
case you pointed out.

We've amended Formulaic to correctly detect this type of problem, and
augmented the standard ranges to include a case which will find a
solution to your equation. The upgraded version (V2.0.3) will be
available as a free download from App Store shortly.

Kind regards,
Hoy Moon