It's interesting what specs manufacturers believe will help them sell product these days. In the 1980s those distortion numbers would be seen as horrible. BUT, the STRH190 owners manual provides additional information:
Minimum RMS Output Power (8 ohms, 20 Hz – 20 kHz, THD 0.09%) 90 W + 90 W
So 90 watts both channels driven at .09% RMS, not bad. I believe .03% is Amir's measurement standard, but .09% is probably not audible on music, if it is accurate.
BUT, the Sony webpage marketing folks chose the 100w spec even though it is 10 times more distorted. Because everyone needs 100 watts, right? It probably looks better on the Bestbuy tag.
Its not too surprising these value priced components are advertised this way, what surprises me more is the number of higher end components that are too.
Example:
PSAudio Sprout 100 ($599)...
4 Ohms: 100 watts per channel, both channels driven (no distortion spec given) and
THD Speaker Output: <0.01% -1W, 1KHz into 4Ω... (at 1 watt 1KHz? that certainly reflects how I listen to music)
Great points - and thanks for that more detailed dive into the specs!
RE the Sprout, I am in no way, shape, or form surprised that that's how PS Audio communicates them - I'm quite skeptical of everything Paul McGowan says. The company clearly makes a lot of high-quality gear, but I remain totally unconvinced that they offer good value for the money. They're not quite at Audioquest levels for me (a company I consider to have an unacceptably high hype-to-value ratio), but they're not far off either. All IMHO of course.
At any rate, I guess what I'm curious about with regard to Amir possibly testing more traditional component-style amps is this: If inexpensive class A/B amps cannot exceed (or even meet) the specs of the better low(ish) cost Class D desktop amps, then where is the price range at which component amps (whether Class A/B or D)
do outperform the value desktop amps - specifically, how much do you have to spend to get at least CD-quality, 90+dB SINAD, reliably low THD, and so on - all with a bit more power than the 25-50wpc one typically can expect from a desktop amp?
Related to this, unless I am mistaken, it seems that one cannot simply buy a pre-assembled Hypex NCore-based amp and be confident it will perform as well as the best homemade amp(s) Amir has measured - in terms of SINAD, ultrasonic filtering, quality of power supply and power-noise isolation, and so on. If we knew of a couple such amps, then the purchasing decision would be easy, assuming those amps were not super-expensive.