IP3 and blocking are related, but not closely enough to tie them together.
Any two amplifiers that are built using similar parts and similar circuits
will probably have these two parameters track, but you can do amazing things
with feedback to hold off the 3rd order products until the amplifier is just
about to shatter.
I haven't had to calculate this parameter myself, so hopefully someone with
real experience will answer. Try posting to rec.radio.amateur.homebrew --
there are some real pros who lurk there (really real pros - guys who build
radios and get paid for it).
I'm pretty sure that for most amplifiers desense will start when the
amplifier starts to clip. If you're getting a module with all the guts
hidden then you're going to have a hard time figuring it out. If you can
get your hands on a sample amplifier you should hook it up and crank up the
input until the output starts to level off, or dig down into the specs if
you have them. If you actually have a data sheet see if they either spec
the 1dB compression point or if they plot the IP3 -- usually the IP3 plots
also show the power out vs. power in, so you can look for the bend in that
line and know.
Failing that, for a class A (low-power linear) RF amp the _theoretical_
maximum output power with sine-wave input is 50% of the supply power before
it starts to clip. For a real amplifier that's not designed with reaching
theoretical maximums the actual will be somewhat to much less -- I wouldn't
be surprised if it were only 5% of the supply power, although I would be
offended. So you can get an idea just looking at the supply current and
voltage rating, but your milage _will_ vary.
