[Scott Dorsey]

geoff wrote:
On 11/11/2016 8:20 a.m., Scott Dorsey wrote:

And it's also very easy to make a lot of those decisions. If component X
gives a very low distortion spectrum but component Y gives an even lower one
and costs no more, then by all means I'll pick component Y even if the actual
sonic effect may be zero. On the other hand, if component X has a spectrum
with even harmonics dominating and component Y has a spectrum with odd
harmonics dominating and both have distortion down near zero, I might have
to think about how these harmonics might add to all the other ones in the
device to make a competent decision, even if the end result is very close to
zero.

However if said signal is passing thru 20 components the that minor
distortion is multiplied, so it may well be worth the extra cost unless
ginormous for esoterica and/or snake-oil components.

Right. On the other hand, if the signal is passing through 20 components of
a different sort, the distortion produced might be completely masked by that
of the other parts. Which is why watching that FFT analyzer and listening
are important.

It's all totally swamped by the distortion from any transducer anyway ...

Depends... transducer distortion is very high indeed, but usually worst at
lower frequencies and usually has lower odd harmonics dominant. So it will
tend to hide other lower odd harmonic distortion but it won't do anything
much to hide 6th harmonic in many cases. God only knows how this plays out
for intermodulation effects.
--scott

--
"C'est un Nagra. C'est suisse, et tres, tres precis."