So I'll add the
Harman/Kardon hk670 to the thread because ... there are measurements! Ironically the measurements likely take it out of the running for "SOTA" compared to today's amps. (Just on power alone -- rated at 60W per channel, measured at 74W per channel -- this unit doesn't stand a chance.)
- In the spirit of full disclosure I own one of these.
Measurements
Some highlights taken from the
April 1979 edition (pages 47-50) of
Stereo Review magazine.
*Please note*: I cannot speak to the equipment that was used for these tests, nor how it compares with Amir's test equipment.
- "The receiver's outputs clipped at 74.4 watts per channel into 8 ohms (IHF clipping headroom = 0..94 dB). The output into 4 and 16 ohms at clipping was 112.4 and 43 watts, respectively. The IHF dynamic headroom was 1.05 dB, corresponding to a short-term output of 76.5 watts into 8 ohms. The closeness of the clipping- and dynamic- headroom ratings indicates the use of well regulated power supplies in the hk670."
- "The harmonic distortion of the hk670 at 1,000 Hz was extraordinarily low at most usable power levels. From less than 0.002 per cent at 0.1 watt, it increased smoothly to 0.003 per cent at 1 watt, 0.01 per cent at 20 watts, and 0.028 per cent at the rated 60 watts. The intermodulation distortion was about 0.028 per cent at power outputs from 1 to 10 watts. It increased at lower and higher outputs, to 0.08 per cent at 0.1 watt and 0.095 per cent at 60 watts."
- "At rated power, harmonic distortion was under 0.03 per cent through the mid frequencies, rising to 0.04 per cent at 20 Hz and 0.05 per cent at 20,000 Hz. At reduced power, the distortion was consistently lower than at full power, with typical readings between 0.006 and 0.01 per cent over the entire audio band at normal listening levels."
- "The hk670 evidently has a very effective system [in the FM tuner section] for removing the 19 -kHz pilot carrier from its audio circuits (no schematic was supplied, so we do not know if it uses a filter or a canceling circuit). The frequency response was almost ruler-flat-within +0.1, -0.2 dB from 30 to 10,000 Hz, rising to +0.4 dB at15,000 Hz. In spite of the fully maintained high -frequency response, the 19 -kHz leakage into the audio was a very low -69 dB. The tuner hum was an inaudible -71 dB."
View attachment 74777
View attachment 74781
(not my image)