[elfary]

I'm wondering lately how does distortion add up in the playback chain.
 
I mean...
 
If i have a 16Ohm iem with 0,1% THD and i feed it with an amp that has 0.07% THD...
 
The overall THD that reaches my ear is 0,1 + 0,07 ? Or just 0'1 since it masks anything below ?
 
My subjective experience lead me to believe that somehow even using 0'1% THD capable iems i can feel difference between using an amp with 0,0035 THD or using one with 0,06 THD. But then placebo could be playing tricks on me.
 
Any insight on this?

 

[EthanWiner]

Distortion adds as decibels by the individual harmonic (and IM) components. But two devices may not have the same types of distortion. Some distortions sound thick or grungy while others may be "buzzy" sounding due to higher order harmonics. In the grand scheme of things the distortion of electronic devices is insignificant compared to the distortion of transducers: loudspeakers, headphones, and microphones.

--Ethan

 

[bigshot]

  My subjective experience lead me to believe that somehow even using 0'1% THD capable iems i can feel difference between using an amp with 0,0035 THD or using one with 0,06 THD. But then placebo could be playing tricks on me.

 
Placebo. Anything under 1% non linear distortion sounds good. Much below that and distortion will be completely inaudible.

 

[elfary]

lately i'm trying smartphones with much more THD and IMD driving low impedances but subjectively the differences between them and my iPhone 4S (ultralow THD &IMD even with low loads) are far more subtle than objective data might suggest.
 
Thus i'm leaning to think that placebo is playing tricks with me big time.
 
Thanks for the input mates

 

[xnor]

That's one reason why people dismiss objective data.
 
They see a cheap amp with 0.03% THD and a supposedly better and more expensive one with 0.1% THD. When comparing both by listening they may prefer the supposedly better one, so they think the measurements are wrong or something, while in fact they completely ignore that our ears are far from perfect.
 
Actually hearing low amounts of distortion, especially harmonic distortion with a few low harmonics, is very difficult or simply impossible. Hearing tiny but easily measurable frequency response deviations is similarly impossible.

 

[jcx]

there are some spec games being played too - distortion is level and frequency dependent - if marketing wants a poorly built SS amp to give good numbers they spec THD near the full output level at 1 kHz - and compare to a tube amp at the same high level
 
but THD vs level and frequency plots could easily reveal zero crossing dead band crossover distortion or slew limiting
 
http://www.gedlee.com/distortion_perception.htm shows low level linearity is far more important to human perception
 
which makes "adding" any crossover distortion a instant degradation that completely swamps any smooth distortion that continuously declines with level