Without any scientific data to support this, I would say that, while I agree that sibilance is a real artifact, that it's not limited to simply microphones. The reason I question this is, why does it manifest primarily and while not exclusively, at least mostly in vocal recordings.
Massed strings, which because of the focus of thier collective bandwith, would seem to be very prone to this same effect, yet I don't find it in the same amounts.
That, plus as I listen to the remakes of Winston Ma's works, recordings which sound so unbelieveably sibilant in thier original forms, sound extraordinarily less so after he performs his 'magic'. Winston told me personally, that his process is one of (of course many more things than this and I am simplifying) removing layer after layer of noise within the recording.
The harmonics of say a female voice, an octave, two, and three above the dominant in that female voice reside in the 1khz, 2kz, 4khz frequency bandwidth. This, not too coincidentally is spread over the range of many mid to HF driver crossovers.
With time and phase 'smear' in this region, coupled with great amounts of noise in SOME recordings the result is a heightened distortion level.
As to tubes smearing this information, I don't think that's it. Tubes, by most, even solid state designers are by consensus thought to have better low level resolution. That greater resolution may simply give more information over the entire spectrum allowing for a fuller more complete therefore less sibilant sound.
I offer up the one theory, crossover issues in dynamics, as one thought after living with the Sound Labs for quite a while and not having anywhere near the same level of sibilance introduced. Also, the Quads, and even the Maggies full panel units. None of these panels, to my ear share in these sibiant issues to nearly the same degree.
This group of theories has, again, no proof offered, and may be completely wrong. But given hundreds of hours, make that thousands, of listening, it seems that some or all of these may contribute to the annoying artificacts we call sibilance.
Massed strings, which because of the focus of thier collective bandwith, would seem to be very prone to this same effect, yet I don't find it in the same amounts.
That, plus as I listen to the remakes of Winston Ma's works, recordings which sound so unbelieveably sibilant in thier original forms, sound extraordinarily less so after he performs his 'magic'. Winston told me personally, that his process is one of (of course many more things than this and I am simplifying) removing layer after layer of noise within the recording.
The harmonics of say a female voice, an octave, two, and three above the dominant in that female voice reside in the 1khz, 2kz, 4khz frequency bandwidth. This, not too coincidentally is spread over the range of many mid to HF driver crossovers.
With time and phase 'smear' in this region, coupled with great amounts of noise in SOME recordings the result is a heightened distortion level.
As to tubes smearing this information, I don't think that's it. Tubes, by most, even solid state designers are by consensus thought to have better low level resolution. That greater resolution may simply give more information over the entire spectrum allowing for a fuller more complete therefore less sibilant sound.
I offer up the one theory, crossover issues in dynamics, as one thought after living with the Sound Labs for quite a while and not having anywhere near the same level of sibilance introduced. Also, the Quads, and even the Maggies full panel units. None of these panels, to my ear share in these sibiant issues to nearly the same degree.
This group of theories has, again, no proof offered, and may be completely wrong. But given hundreds of hours, make that thousands, of listening, it seems that some or all of these may contribute to the annoying artificacts we call sibilance.
