Third-hand descriptions of the Welte recording process and the
speculation of how it may have been done have not convinced me that
there was an instant playback capability or that dynamics were
"recorded" in real time.
Where are the first hand accounts? There were scores, if not hundreds,
of people present at these mysterious recordings -- the pianists! Do
any of them, in their memoirs, describe instant playback or other
technical miracles of the day?
Paderewski's autobiography, to mention one prominent example, is
detailed to the point that it is loaded with trite anecdotes: he drops
names of everyone worth mentioning he meets; he gives details on the
hardness of his hotel bed in Rio where he toured in 1910. However,
recording piano rolls is not even mentioned once in the whole lengthy
book. Doesn't this border on denial?
Why no mention of Welte, Duo-Art or any other player piano? Perhaps
it is because he does not count his recordings onto rolls among the
achievements he wants to be remembered for. Perhaps the rolls do not
really represent true likenesses of his playing. Perhaps, in later
years, he was ashamed for taking money and becoming part of a ploy
that duped the public?
One name not mentioned here yet is Hugo Popper. If he was not the
father of the reproducing piano, then surely at least its midwife.
Based in the cosmopolitan city of Leipzig, an important centre of
music, mechanical and otherwise, this visionary had the means and the
contacts and the charisma to enlist famous pianists to record for
Welte. This was the key to breaking into the market.
Many recordings took place in Poppers' "salon" in Leipzig until his
death in 1910. It seems the artists needed some enticement to put
their signature to the rolls that bore their name but perhaps did not
truly represent their performance. Huge sums of money were undoubtedly
paid out, as well as "gifts" including at least one luxury automobile.
Regarding marketing -- 100 years ago advertisements were at least
as manipulative as they are today, using all the tools of propaganda
to lure in the potential customer. Reading a Weber orchestrion ad,
for example, leaves the impression that there are at least three or
four more ranks of pipes in their "Maesto" than are actually there.
Victor's motto, "His Master's Voice", should have read, "Reminiscent
of His Master's Voice". Similarly, the Welte Licensee ad promises
"The master's fingers on your piano", which perhaps should read
"Reminiscent of the master's fingers on your piano".
It would have been in any manufacturer's interest to create a veil
of mystique around his product and the process of achieving it. What
better way to titillate the public interest? Welte and Bockisch, the
inventors, and Popper did an excellent job of this -- long after they
died the mystery lingers on. We are still intrigued today by the
question, "How did they do it?"
Those knowledgeable in the history of technology can speculate on
whether or not it would have been possible to record dynamics or to
play back a recording instantly. The question remains, why not simply
hire two or three "editors" trained in music, possibly budding
pianists, who could follow along the score of the recording artist's
playing, taking note of dynamics of the piece in "real time" as it was
played. These notes would then be compared, possibly discussed with
the pianist, and later converted into dynamic coding for the roll to
The pianists would be "sworn to secrecy" by the signing bonuses,
honorariums and possibly royalties they received. These costs were
quickly retrieved by the high price of the Welte rolls, which were many
times more expensive than "regular" piano rolls, but could have been
only marginally more costly to produce.
The description of how the mercury trough captured dynamics is highly
questionable to someone with a working knowledge of the piano key and
action assembly. To put it into one sentence, to imply that "the
deeper the key is depressed the louder the note will sound" is simply
not what happens in a piano. If a piano were rigged in such a way, you
couldn't get the world's top pianists to play on it! The instrument
has to have the right "feel" and be regulated properly for an artist
to really perform.
As for the "rate over distance of the (dynamic roller ink) line
expanding", the duration of a hammer blow is measured in hundredths of
seconds. How fast must the paper be traveling to differentiate a
dozen different dynamic levels? At a fast rate, no doubt, perhaps as
much as 1 meter or 3 feet per second. Could a 10-minute piece possibly
have a master roll of a hundred or more meters, i.e., up to a thousand
feet in length? Please tell me I'm wrong!
My point is this: how valid is a second-hand account of the description
of a process by the inventor, who wants to shroud it in secrecy? If
the Welte system of recording was such a technical marvel as many would
like to believe, why did Welte and Bockisch not document it later in
life? The interest was surely there, with little danger of competitors
appropriating the by-then 50-year-old technology.
While it is intriguing to speculate and wonder how the Welte rolls
came to be made, it is tempting to ignore the "easy solution" and look
for an answer that fulfills our desire to believe in the extraordinary,
However it was done, whether the results are a _true_ rendition of the
early century's great pianists' musical interpretations or "only" an
approximation thereof, it is a privilege that we can enjoy this
wonderful music today.
[ The Welte T-100 rolls examined by knowledgeable collectors in
[ recent years -- including production rolls on red paper, production
[ masters on white paper, and a few rare rolls with ink traces -- are
[ all the same width, and all appear to be intended for use at the
[ standard paper speed of about 8.2 feet per minute, or about 1.6
[ inches per second (4.2 cm/sec).
[ Without a technique like the helical scan used by Ampico (and also in
[ video cassette recorders) it's difficult to imagine how key velocity
[ could be reliably measured. -- Robbie