I think this is a more accurate explanation.
It has been reported for years that tube amplifiers seem to play louder than similarly rated SS designs.
Bart Locanthy (now deceased) a former president of the AES society, borough to my attention a test that validates this contention. The test method uses pink noise rather than a single sine wave as a test signal.
The rational is that pink noise, consisting of random amplitude and frequency components over the audio spectrum is a much closer approximation of music and speech than a sine wave of any single frequency. The test consists of comparing the maximum NOISE POWER (or music power) output between SS and tube amplifiers with similarly rated single frequency maximum RMS power output.
The test method is as follows:
The noise signal is applied to the first amp with the output connected to a parallel combination of a load resistor, the vertical input of an oscilloscope and a true RMS voltmeter. (Not an average responding meter calibrated in RMS).
The signal level is increased until the occasional clipping of signal peaks is noted. It is a matter of judgement to determine a rate of clipping over a unit of time that can be reasonably duplicated.
The reading on the meter is noted and the power calculated by the formula -Voltage squared over the value of load resistance. The same test is then repeated on the second amplifier. You will discover that a well designed tube amplifier will deliver 2 to 3 X more power than the solid state unit. The test described was with a Wolcott Presence in comparison to a well respected SS amp of 300 Watt rating, and revealed a 3 to 1 power advantage with tubes, delivering 75 watts to the SS output of 25 watts.
Another type of test was conducted by "Stereophile" magazine several years ago using tone bursts instead of pink noise, showing almost identical results.
Other advantages of tubes, particularly in the output stage, is that they do not exhibit slew rate limiting or reduction of gain with increasing frequency except for electron transit time considerations.
Transistor current gain falls off with frequency as their alpha factor approaches unity or zero gain. From a designers standpoint, tubes are easier to work with because they are characterized by only three main parameters:
Mu factor, Mutual conductance and plate resistance, whereas transistors require 22 parameters to fully characterize them.
This complication, in conjunction with their less desirable transfer function makes them prone to the generation of many more forms of distortion in music reproduction than are possible from vacuum tubes.
