I should point something out here. The big difference between tubes and transistors is that tubes tend to sound smoother, with greater actual detail rather than brightness masquerading as detail.
But tubes have measurably greater distortion- so how is this possible?
The answer has something to do with the ear's masking principle; that is where a louder sound can mask the presence of a quieter sound.
It also has to do with how the ear perceives distortion, and in this case I'm talking about distortion where the amplifier is not being overloaded. The ear interprets the 2nd and 3rd harmonics as 'body', 'warmth', 'fullness' and the like, all terms to describe their presence. If these harmonics are present in sufficient quantity, they will mask the presence of the higher orders. Since tube amps generally have more of the lower orders (2nd, 3rd and 4th) than solid state amps, the result is that the higher orders are masked and so tubes sound smoother even though they have greater higher ordered harmonic content than solid state.
So solid state tends to sound brighter and harsher (harder) than tubes because the ear interprets the higher orders in this manner, simply on account of the fact that the higher ordered harmonics (to which the ear is keenly sensitive as it uses them to gauge sound pressure) are exposed.
But there is more, and this bit is really fascinating. The lower orders, if in sufficient quantity, also contribute to soundstage width and depth, as well as lower level detail! This is why tube amps tend to have a wider and deeper soundstage than solid state. Oddly, this is not an exaggeration, but in order to understand that this is so it is helpful to have master tapes or files and also to have been present at the recording to know how it is supposed to sound.
The bottom line is this: unless the amplifier has **no** distortion and by that I really mean no distortion at all and not just vanishingly low, the inclusion of a bit of 2nd and 3rd in sufficient quantity can actually result in the amplifier sounding more 'neutral' to the human ear. Again, this is a topic that deserves more research, but this phenomena has been known for quite some time.
So its going a bit out on a limb to use the word 'accuracy' as in the comment below.
Put another way, if the amplifier designer is aware of how the human hearing perceptual rules work, and is pragmatic about the simple fact that building an amplifier that truly has no distortion is impossible, then the next best thing to do is to include some of the lower harmonics for the perceived benefit they bring, even though it might look bad on the spec sheet. In this way its easy to show that the spec sheets are a good example of the Emperor's New Clothes, as for the most part they ignore human hearing perceptual rules as they have been ascertained in the last 40 years.
Nelson Pass seems to have sorted this out. He is one of the few solid state designers to have done so, and no surprise, his amps sound more 'natural', 'neutral' and musical (IMO) than his competition with lower distortion as a result.