I don't know if this is still true, but back in the dark ages one of the hard decisions to be made when designing an amplifier was how much bias to apply, ie how much idle current to run through the output stage. The idle current is what makes the amp a class AB rather than a pure class B, ie the output transistors are always on just a tiny little bit. This results in waste heat but reduces or eliminates the "crossover notch", a non-linear response around zero caused by the fact that transistors need a bit of voltage (~0.7 ish volts) to turn on so a pure class B amp will have some distortion at low levels.

If you have the bias too low the idle current is very small or non existent but the distortion is pretty bad at low signal levels. if you crank up the bias "enough to be sure" you get a very clean sound but the amplifier gets hot even when it is sitting around doing nothing.

All the HK amps seem to run hot; I presume this is a "good thing" caused by having a high enough bias on the output stage to push it solidly into class-AB land at the cost of wasting some power at idle.

Again, this may no longer be true and I may not be remembering it correctly after 25+ years, but...


M60ti, VP180, QS8, M2ti, EP500, PC-Plus 20-39
M5HP, M40ti, Sierra-1
LFR1100 active, ADA1500-4 and -8