Recorded Piano Sound (Was Rudolf Serkin: Columbia Legends)

'It must be said that Columbia NEVER was able to capture the sound Serkin produced in the concert hall. I heard him a dozen times and was always astonished by the dynamic range and the luscious tone. The first movement of the Moonlight left the audience enraptured. I will never forget it; the best I ever heard, bar none! But Columbia never got the sound. RCA never got Rubinstein's right either. And Decca has never captured Ashkenazy's tone.'

I'm just wondering how far this should be generalized. To some extent I guess it must be true of almost all great pianists, but are there other egregious examples people know of where modern recording engineers haven't come close to capturing the true sound of the pianist?

Gould, but that was his fault.

Show me one recording of a piano, monaural or stereo, analog or digital, that sounds as much like a real piano as recordings of voices and orchestras sound like real voices and orchestras. I have yet to hear it.

-david gable

Let's understand the nature of the problem first. The piano has an extremely wide frequency range and dynamic range compared with other instruments, plus an incredibly rich spectrum of overtones (harmonics). And the problem is compounded by the use of the pedal, which by lifting the dampers on all the strings in the instrument at the same time allows every possible harmonic mode to vibrate at the same time. The combined effect of all these factors is that the recording process is pushed very close to the envelope in every respect, and the recording engineer has very little margin for error. The most common method for acquiring a wider safety margin during recording is the use of dynamic range compression. This was common practice throughout most of the analog recording era and the early digital one. Unfortunately, it does affect the sound in subtle ways, though not always immediately recognizable as side-effects of dynamic range compression. One of the best examples to illustrate this phenomenon is Richter's peformance of Chopin's 4th ballade on a DG recording from the early '60s. There is a sudden and very audible drop in the noise level and a dramatic lowering of dynamic range at the beginning of the coda. The timbral quality of the sound changes and the sound stage collapses to a much smaller space. It sounds almost like the recording was spliced from two different performances. Listen to this performance and see if you notice what I described.

When recording an orchestra or chamber ensemble, one can use as many mikes as one wants, up to one mike per player. This allows every instrument to be recorded on a separate channel, and makes it easier to avoid tape saturation. The relative levels of the parts and/or players can then be conveniently adjusted during the mixdown to 2 channels. For obvious reasons this cannot be done when recording a single instrument, and with a modern grand under the hands of a Richter, Gilels, Cziffra or Berman it is almost impossible to avoid tape saturation during the loudest passages while at the same time keeping the softest pianissimi above the noise floor. I hope you get an idea why recording the true sound of a piano has been so difficult.

On the other hand, one should not imagine for a moment that the true qualities of a pianist's sound would be distorted beyond recognition

There's nothing to agree or to disagree about. You listen through your ears, and I listen through mine.

Plus we may not have heard the same recordings or attended the same concerts.

This is a classic leftist style attempt to take control of a debate!

I have heard many less pianists live than dk. but I think the 'problem' of capturing the true sound on recordings is exaggerated.

I heard Solomon live and was drawn to him first by his recordings...... same beautufl sound and touch live as on his records also Gieseking's touch comes through both in records and live. Lupu and Michelangelli as well(from my experience) most recently Volodos whose full, non-percussive sound is the same on record or live, (and on all his pirates also)

I also heard Serkin, and IMO, his was lousy live as on records.....

I particularly recall that goofy LP he made of Sibelius piano music (now there's a red flag for you!), where the sound perspective changed between movements, and sometimes even between phrases. It reminds me of early stereo recordings where microphone set-ups may not have remained consistent between sessions, or where engineers had fun twiddling with the orchestra spread in mixdown, leading to the phenomenon which my old friend James Langdell called 'Johnny Green and His Amazing Dancing Orchestra!'

Gould wanted to record his hummmmm... :-P

Gulda's recordings on Amadeo are usually very much 'on-mike,' whose sound, I think, regular audiances never able to hear unless you stick your face into the piano. But it was his liking.

I don't believe music recordings, piano or otherwise, should or could sound like the real thing. Recordings are entities unto themselves, perhaps bearing most resemblance to what their composers heard in their 'mind's ear' and that's just the way it is (and should be for me). I hate to have to disagree with you here, but a listener sitting in an adjacent room unable to look inside is very much more likely to believe there is a piano than an orchestra therein.

then why do you bother to listen to all those recordings....))))