True, but it gets to the heart of the matter in that it's
then identifying the problem part so it can be replaced,
instead of having to hunt down various parts (assuming some
people don't have spares of many parts and that for some
parts it can be fairly laborious or at least time consuming
to swap some things like a motherboard including repair or
reinstall of windows to get it working) and swap each.
A typical consumer PSU tester would not accomplish much, it
is quite common for a PSU that is beginning to fail but not
totally dead yet (such that it doesn't yet shut itself off
or stay off from being extremely out of spec) to have
unloaded or barely loaded rails that are within spec. The
needed measurement is one when the PSU is powering the same
load or even better, the very PC it needs to be used in, or
even best, that very PC while the PC is exhibiting the
symptoms of a problem.
Most PSU testers are practically worthless except if one
plugs it into the PSU and PSU doesn't turn on at all, in
which case one can assume PSU is dead but putting a hard
drive on a connector and jumpering PS-On to Gnd will tell as
much with most PSU (excluding the rare ones that monitor and
shut down if there's no load on other rails like the 3.3V).
More to the point, the typical PSU tester can't indicate a
PSU is working properly to power a system.
And yet, having the knowledge is what allows to decide on
the needed remedy of replacing the PSU so it's puzzling what
issue you have with the methodology. I would speculate most
people don't have another equally *good* PSU lying around,
that one would tend to use the best PSU they have available
to power their system instead of leaving it unused... and
many people don't have several newer systems such that they
can swap an appropriate PSU out of one to put it in the
other system. Thus, the question is one of what to do next
- should the troubleshooter then seek a good replacement
PSU, suffering higher cost to get it locally or longer wait
to have it delivered from an online purchase, or should the
troubleshooter focus on other parts instead.
If one can't find the problem any other way, yes shotgunning
is a last resort, but one that takes quite a bit more time
than the initial voltage measurements. How many people do
you expect to have access to a known good PSU, one proven
suitable for the same load the questionable system causes?
What if the known good PSU isn't suitable for the system
even though it otherwise works, then the troubleshooter
could be led astray thinking some part other than the PSU is
the problem. If you are saying the known good PSU is known
to power this system, why was it swapped out at all for the
other psu? That seems unlikely, IMO most people will buy
one PSU for the system, install it and leave it alone unless
there is a problem. Some might have a PSU from an older
system but it may not be suitable since systems have moved
to requiring quite a bit more 12V current and that with
shorter recovery times than in the past. CPU and GPU going
from idle to full load can be a quite large and rapid change
in current.
