The Naxos TOSCANINI CONCERT EDITION So Far...
Matthew B. Tepper
Don'tEvenTry...@Anywhere.com pondered what I'm pondering as
follows:
>
>As someone who recalls listening live to the broadcast of the famous
>Dec. 1953 "Eroica", and who has been collecting Toscanini records
>since my first childhood copy of the "Nutcracker" Suite in the early
>fifties, and as a broadcast host (and engineer) who devoted four
>continuous years to the weekly transmission of all the Toscanini
>material issued up to the year 1974, I have spilled some thousands of
>lines in previous posts in this n. g. discussing the technical and
>musical qualities of innumerable performances, recordings, and
>airchecks of the Maestro's work.
[enormously long, but enormously engrossing, learned and useful post
snipped for space]
Aha! If you're who I think you are, I was particularly hoping that
you'd make the acquaintance with this series! I'm glad to see your
comments at last! May one hope for a running commentary as you hear
more of the issues?
--
Matthew B. Tepper: WWW, science fiction, classical music, ducks!
My personal home page -- http://www.deltanet.com/~ducky/index.htm
My main music page --- http://www.deltanet.com/~ducky/berlioz.htm
And my science fiction club's home page --- http://www.lasfs.org/
To write to me, do for my address what Androcles did for the lion
A Fanatical Toscaniniphile
Dec. 1953 "Eroica", and who has been collecting Toscanini records
since my first childhood copy of the "Nutcracker" Suite in the early
fifties, and as a broadcast host (and engineer) who devoted four
continuous years to the weekly transmission of all the Toscanini
material issued up to the year 1974, I have spilled some thousands of
lines in previous posts in this n. g. discussing the technical and
musical qualities of innumerable performances, recordings, and
airchecks of the Maestro's work.
I have been assured off-line by some persons who regularly contribute
here that my technical impressions of the new Naxos edition would be
greatly appreciated, so here goes...
THE NAXOS TOSCANINI CONCERT EDITION
Not since the days of the ill-fated Clyde Key Arturo Toscanini Society
has there been such a well-conceived attempt to document the broadcast
performances of the NBC Symphony under Toscanini. I felt that the
during the early nineties the Maestro's commercial disks AT LONG LAST
had been accorded a comprehensive and respectful treatment by their
copyright holder, BMG: despite a few lapses, the RCA-BMG Toscanini
Collection CDs, made in pristine authentic mono for the most part,
generally offered the best transfers yet released off the authorized
commercial recording sessions and a few approved broadcast items.
The Clyde Key society LP issues of the late sixties/early seventies
were a mixed bag of successes and failures: some were improperly
transferred to LP, with severe loss of presence (such as the 1940
Grieg Holberg Suite); others were magnificently superb (a series of
live broadcasts from the late forties and early fifties, some of the
finest being the all-baroque concert of Nov. 22, 1947; the live Brahms
Second of Feb. 10, 1951; and the live Elgar "Enigma" Variations of
Feb. 17, 1951): made from amateur high-speed off-air FM aircheck
tapes, these are nearly the quality of the best commercial sessions,
and indeed often surpassed them in relatively uncompressed dynamic
range. In numerous cases, the original "atmosphere" of the radio
broadcasts was preserved by including the announcements of Gene
Hamilton or Ben Grauer. Later issues of the series began to appear in
fake stereo, or with heavy dynamic noise filter processing and audio
expansion (such as the remarkable April 1940 all-Debussy concert) and
culminated in the Turnabout issue of the old 1940 Verdi Requiem
program that resulted in copyright protection action being taken
against Key.
Over the years, with the proliferation of "legal" compact disk
recordings of the unauthorized broadcasts, ostensibly made in Italy
from "radio broadcast masters" (in reality, being merely tapes sent to
Italy by collectors, played on FM stations, and then pressed into CDs
in a land where copyright law is more lenient than in the USA) the
digitized catalog began to fill up the interesting compositions that
Maestro occasionally played in his broadcasts, but chose not to record
(or, as in the case of Martucci, was discouraged from preserving by
RCA Victor and its minions.) And there were the occasional transfers
from the U.K. (such as the outstanding, high quality Dell'Arte LPs or
their less numerous CDs, or the EMI editions of BBC Symphony
broadcasts) and Switzerland: the near-purist Relief digitals, using
the barest of analogue noise reduction and preserving about 98% of the
sound on the original acetate sources.
But in the past half-dozen years, the CD market has been invaded by a
large quantity of shoddy, ill-produced, undocumented, incorrectly
labelled, and wretchedly reprocessed faked "stereo" transfers of
perfectly decent monaural broadcast recordings, issued primarily by
Iron Needle and Gramofono 2000. Usually the recordings have been run
through early- generation Cedar processing software, using both the
declicking and noise filtering modules at maximum warp; then a layer
of phony stereo echo has been added; the highs have been boosted on
the left channel, and the lows on the right; ping-pong digitized
repetitive delays are introduced between the audio channels; and
finally a strange, warbling, wobbling rotating phase shift (at a rate
of about 3 to 4 Hz per second) has been added to try to create a
spurious L-R and "open up" the mono soundstage to simulate genuine
stereo: any or all of these artifacts may be added in any proportion,
depending on the exact track of any given CD.
An audiophile who has grown up with good, decent mono is completely
scandalized and shocked by the ineptitude of such reprocessing, which
totally falsifies what Toscanini's radio listeners heard during the
original broadcasts, as well as the conductor's own goals in balancing
his recordings. In the case of some of the more egregious phony
reprocessing efforts, the digitized delays "duplicate" or double some
rapid-fire percussive notes, like snare drum rolls, or completely
muddy the instrumental definition, which was almost always exceptional
and transparent, even in the broadcasts of the late thirties.
My comments about the ghastly Iron Needle transfer of the almost
unrecognizably falsified and distorted Haydn "Hornsignal" Symphony
seem to have fallen on interested ears at Iron Needle: their
subsequent issues of the 1938 Mussorgsky Pictures, and 1938/46 Verdi &
Martucci album, have been found by grateful collectors to be in
genuine mono sound, with only a minimum of computer-generated noise
reduction artifacts, in the range of tolerability. But the bad old
issues are still in print, and continue to lurk in record store bins
to confront the unwary.
Thus, one eagarly looked forward to the promised reissue of the
Toscanini programs by Naxos, from semi-authorized sources licensed
from the "Immortal Performances Society", whatever THAT is! Though
the issues were announced in British publications like the Gramophone
and ClassicCD, nary a word turned up in American press releases or on
the HNH website; an email by me to Naxos annotator Keith Hardwick was
quickly answered, notifying me that no American issues were
contemplated. Later, Klaus Heymann explained (in the press release
material from Naxos sent to record dealers like Tower and Blockbuster)
that he had no desire to go toe-to-toe with BMG or the Mertropolitan
Opera, who were "relying on court decisions from the fifties" to
continue to object to domestic issues of the radio broadcasts; Heymann
said he eventually hoped to bring the matter to the attention of NAFTA
and GATT.
But the exceptionally consistent quality of the thousands of Marco
Polo and Naxos CDs promised a professional competence in the handling
of historical material that is sadly lacking in the lousy productions
from the Maestro's own native land.
My first afternoon of auditioning six individual releases in the
series promises that -- for the most part -- that confidence in Naxos
was not misplaced.
The full and authoritative annotations explain, in addition to
pertinent details about the performances and music, the facts of
history that led up to the conception and execution of the release.
RCA's Richard Gardner, "the conductor's favourite sound engineer",
collected the sound recording masters or tape dubs from Walter
Toscanini and other sources, providing them to Richard Caniell of the
Immortal Performances Society (we are told), starting in 1948. When
Gardner was diagnosed with cancer, he completed the transfer of all
the collection. I guess this is duplicated at the Toscanini Rodgers
and Hammerstein Library Collection and the metal parts and existing
mastertapes that are in the RCA archives in New York: Don Drewecki can
provide all the details, one assumes, or correct what I am summarizing
if it is in error. How these exact source materials differ -- if at
all -- from the John Corbett transfers or authentic RCA source disks,
one cannot surmise.
The modern digitizing process was accomplished for Naxos by
transferring to CD "the first strata of recording directly from the
original discs via a Prism Sound 20 bit Analogue to Digital Converter
to digital tape and then Noise Shaped to 16 bits using the Prism Sound
SNS 2 process." I know absolutely nothing about the Prism algorithms,
other than what my ears reveal to me, in comparison to the "Cedar",
"Cedar II" or "Cedar-Weiss" processing on diverse European import
productions on Iron Needle, Sirio, or Gramofono, or the analogue
noise-gating used on earlier Nuova Era or Relief CD transfers. For
the most part, with exceptions as noted below, the Prism method, as it
is implemented by "Archivist and Restoration Producer" Richard Caniell
or his unnamed associates, is vastly more transparent and less
invasive than the noise-filtering modules of the Cedar methods of
competitors, and also creates less audible noise-modulation "hiss
tails" than the analogue noise gates used by Nuova Era or Relief
transfers of the late eighties, if not having the absolute precision
and accuracy of the vaunted prototype Orban autocorrelator system.
Thus, the overtones of string partials and wind instrument "chiff",
and the precious details of what small amount of dry Studio 8H
ambience managed to be captured by the microphones, tend to be audible
on these new Naxos transfers, with the exceptions noted below.
"Indeed, we have allowed the grit and ticks to remain unfiltered" say
the producers, and -- more important -- no radical attempts have been
made to smooth out sound differences between source disks so that the
series has a consistent "identifiable" sonic signature: in contrast,
the Gramofono disks tend to sound quite alike, up to the issues made
fairly recently before complaints started to be registered: there was
always lots of extra echo, a great softening of high frequency detail,
a boosting of upper midrange, and even an occasional compression of
the already limited dynamic range. The Gramofono set of the 1939
Beethoven Ninth, for example, has fake stereo, bass boost,
steely-sharp strings, piercing brass, and clearly audible "gain ducks"
for loud passages, all missing on the absolutely pristine Relief CD
set, which faithfully preserves disk groove distortion, tinny lows,
and a waterfall of ticks and pops as well as the glorious
music-making.
Naxos runs a sane middle ground between these two extremes. One can
listen with headphones to the single-channel source without being
troubled by the spurious dimensionality of phony stereo echo: the
soundstage remains solidly centered, and retains the integrity of the
simple monaural microphony. Yet the intrusive clicks are essentially
missing, magically edited out by the clever software with a far
greater transparency than previous methods (tape splicing, Burwen/SAE
real-time switching, or just grossly rolling off the transients by
removing all the highs.) This has been done about as well as it is
typically accomplished on the latest generation of Pearl and Biddulph
transfers by those two reigning geniuses of historic refurbishment,
Mark Obert-Thorn and Ward Marston, using Cedar de-clicking but NOT
excessive filtering.
To someone who grew up with 78 shellac sets, early Victor LP copies,
Victrola transfers of the Toscanini material, FM mono broadcasts, and
innumerable other treasurable items of audio memorabilia from the
forties and fifties, the honest original sound of Studio 8H programs,
sans any of the "second thoughts" of questionable engineers or
tin-eared producers, is a long-sought anodyne, assuaging the pangs of
frustration caused by disk after disk of butchery from other labels.
Clyde Key's original intentions are being continued here, with the
benefit of the techniques and insights provided by modern engineering
applied intelligently by experts whose conceptions are justifiable if
not universally "right": no one can say which is the correct approach,
to "fix" the recordings, or to faithfully convey them with all their
inherent faults. But one can surely aver that the WRONG approach is
to wreck the recordings with all kinds of specious alterations, and
then to hype them as being "refurbished" for modern listeners. I hope
that the complaints of purists like myself, who have condemned this
misguided ruination, has helped influence Naxos. And I hope that this
enlightened company searches out each and every recording that has
been botched by the European bootleggers, and sets matters right with
transfers of at least the quality of this initial Concert Edition.
At times, during some of these antique Toscanini programs of the
forties, the listener soon forgets the technical artifacts of the
actual recording, loses himself or herself in the music-making, and
enters into that temporally-liberating "time machine" experience of
seeming to travel back to that intense world, when people seemed to
really CARE about playing the C-Minor Concerto, Op. 37! Unlike modern
bland run-throughs in the total perfection of all-DDD sterility, these
performances have true human character, warts as well as
virtues: sometimes the tempi run a bit over the top, or the execution
slips below dull perfection; surely there are no cavernous billows of
soothing cathedral-like echo to cover up indifferent articulation or
imprecise second-violin sections. We have here in musical
truthfulness the equivalent of a stark, innocent declaration by a
defiant Harry Truman, who proclaimed what he felt and didn't give a
damn what anybody thought about it!
Thus, though one starts out to critique an individual recording, soon
we give up and surrender ourselves to Beethoven and Toscanini, and
just ride along, hanging on for dear life, stimulated and thrilled by
the experience, dropping our critical pencil in the rush of wind.
A RANKING OF THE FIRST SEVEN ISSUES:
I did not order all the first Toscanini sets on Naxos, as I had only
recently shelled out $20 per disk for many of the costly Relief CDs,
particularly the 1939 Beethoven Symphony cycle (infinitely better than
the grotesque Gramofono 2000 monstrosities.) Of the ones that I just
acquired by mail order from the U. K., I rank the following in order
of sonic success and desirability:
8.110804 BEETHOVEN: Third Piano Concerto in c, Op. 37, with Myra Hess;
also Coriolan Overture, and Wagner's Rhine Journey, from 24 Dec. 1946.
Sound is much better than the Clyde Key-ATS disk: fuller and richer,
with -- of course -- less of the original acetate noise. Better, in
fact, than the decent sound of the 1944 Beethoven c-minor with
Rubinstein. Compare the recent Piano Society transfer of the
Rubinstein performance with the BMG: it is totally dessicated and
lifeless, so I would expect that their issue of the Hess will be just
as poor. Ben Grauer commentaries are included, with an 18 minute
extract of the Coriolan rehearsal, where not nearly as much noise
reduction has been done in order to preserve details (the rehearsal
disks were almost always sonically inferior and noisy, anyway.) I
haven't played the rehearsal track all the way through yet, but I
heard it in the eighties, and it is a cherished item: I encountered at
least one bracing "Toscanini roar" in the section audited so far. For
extended passages in the Concerto and the Rhine Journey one hears
virtually the quality of a clean LP record, though the original source
is a set of acetate "linecheck" disks. Musically, the Coriolan is for
me the most valuable, as it is better executed than the 1953 NBC
performance (with the 'squeak' near the end), and is less aggressive
than the 1945 shellac. As to the concerto, I like both the Hess and
Rubinstein interpretations, and other commentators have said all that
needs to be conveyed about the performances.
It should be noted that in that work we can hear one of the artifacts
of computer noise reduction processing: note the "clack" at 0:02 in
the opening movement of the concerto, where a violinist's bow
apparently clatters on a music stand (surely it couldn't have been
Maestro's baton!): suddenly at the crisp transient sound, the computer
"gates" open up the highs for a few milliseconds and then shut down
again, and we can hear a noise tail and a strange, strangled echo in
perfect resonant "high fidelity" before the filtering kicks in again.
The Prism folks should use this as a "test case" add a few more lines
of code to cope with stuff like this in a less obvious fashion.
8.110813 BEETHOVEN: 25 November 1939 broadcast, part of the famed '39
symphony cycle. Gene Hamilton's commentary is included; major works
are Eighth Symphony, Leonore Overtures 1 & 2, excerpts (mislabelled)
from the Creatures of Prometheus, and the Lento assai & Vivace of the
last String Quartet, played by full string ensemble (as also recorded
commercially by Toscanini in a rather gritty sounding shellac set, on
BMG Toscanini Collection Vol. 45.) One who was stuck with only the
appalling old muddy Olympic disk of some of these items will be bowled
over by the near high-fidelity quality of these transfers. I think I
still prefer the nearly unavailable live 8 Nov. 1952 broadcast of the
Symphony, which is less aggressive and more goodhumored (and has one
of those "lovable" flaws, a bad clam from trumpeter Harry Glantz in
the finale.) This was available for a while on a Hunt CD, to the best
of my recollection, which I have never heard: one hopes it was as good
as the "samizdat" aircheck tape that circulated among collectors. The
1939 performance of the Eighth Symphony is tumultuous; it will
surprise listeners who think of the even-numbered symphonies as
"slighter works" than the great odd-numbered one.
The exquisite Adagio & Allegretto from the "Creatures of Prometheus"
Ballet is one of the supreme examples of the nuanced art of Toscanini
in shaping a miniature and wringing from it the maximum in rapt
concentration and character. The old Olympic LP had 5-kHz dull and
fuzzy "radio network" sound; right after the commencement it was
defaced by a loud "scrunch" which I would guess was a fouling of the
cutting stylus by the chip of acetate (the disk was undoubtedly made
by an amateur, some distance from New York.) Thus, the absolutely
perfect and clear Naxos transfer is one of those "time machine
moments" when the mechanism disappears and we are standing at the
podium next to Maestro, as the magic unfolds. Even the applause at
the end is crisp and "live" sounding; Gene Hamilton's voice sounds as
real as if he were in the room with us. How nice to hear him again: I
have never heard a modern radio commentator present classical music
with Gene's elegant blend of simplicity and culture, lacking even the
slightest trace of the impersonal pomposity of the otherwise
distinguished Ben Grauer, his successor.
8.110809 MOZART: Haffner Symphony plus rehearsal excerpts of nearly 5
minutes; Piano Concerto No. 27 w/Horszowski; Marriage of Figaro
Overture; 5 Dec. 1943. I have not heard this broadcast since the
early nineties, and I have not yet had the time to give this the
attention it deserves. But the Symphony is played with more grace and
less fire than his rather hard-pressed commercial disks of 4 Nov.
1946; timings are as follows:
1943 1946
Movements: 1 5:20 5:01
2 4:29 5:51 (more repeats)
3 3:35 3:13
4 3:35 3:28
Except for the second movement, the situation of repeats taken or
ignored is similar in '43 and '46: those extra seconds in the earlier
performance were needed to allow a little wiggle room for expression.
Pacing of the slow movement is more flexible in '43, and -- despite
the actual timing of the movements -- is slightly slower by a few
beats per minute (approx. quarter note=99, varying considerably down
to 94 or even a bit slower), with a noticeable expansion of some
phrases, and very marcato bowing in 'espressivo' style, not the
slightest bit clipped or impersonal. In '46, Toscanini started the
Andante at quarter note=96 beats per minute, which is fairly
consistently maintained, with less "interpretation" of different
sections. The opening movement is more relaxed than the shellac disks
of 4 Nov. '46, if not the equal of the radiant live broadcast of
November 3 (which I DO hope that Naxos chooses to issue!)
Those of us who have struggled through the wretched sound of the great
Serkin/Toscanini performance of the Concerto No. 27 with the PSNY
(Feb. '36) on Radio Years in noisy, distorted fake stereo will be
pleased to have a more ingratiating sound quality in this Studio 8H
broadcast with Horszowski.
One thing must be noted about this recording: it could well do with
what B. H. Haggin would have called "a step of bass": in fact, two or
three steps! The Naxos folks have obviously decided not to "fix"
this, but the sound is pretty thin, and can be easily corrected, to
the taste of the final auditor, by tone controls or graphic equalizer
if desired. Otherwise, the Concerto and Symphony are very clear and
clean "in house" transcription disk sound (with NO noise), though the
announcements and music recording of the Figaro Overture are in very
wiry and compressed AM-aircheck sound; Grauer's voice and the music
dramatically improve after the overture. Yet, at 4:17, this '43
broadcast of the overture is more graceful and yielding than the
blinding race preserved in the brilliant performance of the 8 November
1947 broadcast, in almost modern "hi fi" sound on Vol. 10 of the BMG
Toscanini Collection. While both versions are more driven than any by
a modern conductor in memory, the earlier has a much broader and
expressive beginning and an almost swaggering quality to the
diaglogues of winds and strings, if not possessing the playful
skittering of the '47 concert reading.
8.110817 WAGNER: Parsifal Prelude & Good Friday Spell; MENDELSSOHN
Violin Concerto, with Heifetz; 9 April 1944. While I am very partial
to the BBC broadcast of the Wagner, this is an entirely satisfying,
rapt, and committed reading (so is the commercial version from '49,
but BMG used noisy 78 rpm disks instead of the clean magnetic tape
employed in the earlier RCA Victrola transfer on VIC-1278.) The
concerto is sweetly played by Heifetz, but the reading is not exactly
a revelation, since there are so many fine editions extant. Was there
a live 1940 Beethoven concerto broadcast by Heifetz/Toscanini, before
their commercial recording sessions? My old edition of Marsh does not
so stipulate. If there were a live broadcast, it would be a valuable
alternative to the commercial shellacs (and so would the slower takes
unearthed by Winner from the RCA archives!) At any rate, Toscanini's
accompaniment to the Mendelssohn is ineffably superior to Munch's for
the Heifetz stereo Red Seal recording, in many ways that are hard to
speak of in general: let's merely say that there is greater intensity
and celerity than in that familiar chestnut of the Living Stereo
collection. But Beecham's view of the orchestral part is equally
valid, as documented in the 1949 HMV edition with Heifetz, or in the
1935 Szigeti performance; and Walter's version with Milstein from '45
is also treasurable. However, since I have never laid eyes on an LP
or CD bootleg of this broadcast before (though I heard a collectors'
tape) it must be said that this is an exceedingly rare item, and
should interest many folks. The sound of both works is similar: a bit
"one off" from the clarity of the '43 Mozart concert, with some
hissing noises, but entirely tolerable if not utterly realistic.
8.110801 BEETHOVEN: Sym No. 5, 4 April 1933; Triple Concerto with
Piastro, Dorfman, Schuster, 1 May 1942; both P-SNY.
Wasn't this Fifth supposed to have been a continuous film recording,
or was it the previous unapproved 1931 Philharmonic version?; I cannot
recall, and haven't any documentation at hand: Don or others can
correct me if I am in error. Naxos states that the source material is
16" lacquer disks cut at the East 24th Street Recording Studio. The
sound is very bright and clean compared to the really awful CBS radio
broadcasts by the orchestra under Toscanini from the thirties. The
emphatic performance is one of the Maestro's best, if not totally
supplanting the great 1939 commercial shellac set.
The Triple Concerto was recorded during an unbroadcast public concert
at Carnegie Hall, and is sonically substandard compared to the studio
work at RCA and NBC, or to the forties broadcasts from Carnegie
(Brahms and Verdi) that are available from Music&Arts. The Clyde Key
LP disk retained ALL of the original surface defects, including
swishes, clicks/pops, wavery high frequencies and phase distortion,
and overcut or damaged groove shattering. Yet I found it possible to
play the record with a slight treble rolloff, and to be fully engaged
in the great performance, one of the fastest ever done, but not
especially intense: it was as classical as Weingartner's, but at a
higher velocity of execution. Walter's studio recording with the same
orchestra, made seven years later, retains the same creamy elegance of
Philharmonic string sound, but unfolds the work at "traditional"
Germanic tempi.
Lest you suspect that the garbled sounds at the opening of the
concerto are primarily caused by the source disk damage, I would
suggest that instead the Prism software is interacting in a somewhat
strange way with the VERY high noise platform. The Gramofono 2000
transfer adds fake stereo to confuse the issue, but in playing back
their disk in L+R mono, I hear LESS of this effect, though the
Cedaring is more objectionable during soft passages with less groove
damage. I no longer have the Key LP issue for comparison, so I cannot
say for sure, but there are noticeable differences in the way the two
competing solftware programs try to "solve" this noise problem.
Frankly, who cares? I would rather not ever hear any added artifacts,
and rely on my brain to filter out the scratches. But the very heavy
noise at the beginning of the concerto soon subsides on ALL copies,
and one can more-or-less live with any of them; I imagine that this
Naxos is -- since it is in genuine mono -- really better than the
Gramofono, at least for headphone listening. If we have not heard a
better source recording by now, one probably does not exist.
8.110811-2 DEBUSSY: The Concert of 13 April 1940. I have listed this,
of the six sets I purchased, last in order of desirability because I
am STILL somewhat let down by the transfer, but perhaps Naxos is NOT
to blame; since I have never heard the available source disks, I am
not entirely certain. But the Dell'Arte CD release of the Marche
Ecossaise on CD DA 9021, processed by John Baily at EMI in London,
reveals what is missing on the Naxos production: highs and orchestral
transparency.
The old Olympic two-LP set, produced by the idiots at American Everest
(not the original classy company owned by Harry Belock, but the shoddy
catchall for junk and bootlegs that bought out the Everest name and
rights in the sixties), used an audio expander circuit to produce big
"swellings" in the loud moments of the music, and to recede the sound
quiety into the distance during the soft ones. This gave the "La Mer"
broadcast an almost "Mengelbergian" quality, and actually enhanced the
mystery of the opening, though it altered the original radio program
dynamics and possibly exceeded Maestro's intentions. At least the
sound was in genuine mono.
Then Gramofono brought out a CD of part of this program, adding the
1936 P-SNY La Mer in place of the NBC version, with sound that was so
wretched and falsified that it was WORSE than Toscanini's 1920-1 La
Scala acousticals!
So I had high hopes that the interesting program would at last be
available in "in house" fidelity, and possibly in the crisp timbral
accuracy of Seth Winner's astonishing transfer of the 1940 Petruchka
tableaux in Vol. 28 of the RCA/BMG Toscanini Collection CD series.
But no...the sound is muddy for the most part, comparable to the
rather dull Music&Arts Tchaikovsky Manfred and Pathetique on their
2-disk set, number CCD-956, and has peak-limited loud passages.
Now, not ALL copies of at least SOME of the preserved archival
material of this broadcast sound dull: the Dell'Arte transfer of the
Marche, mentioned above, is brilliant and similar to the Petrushka,
but is noisy, with impulse ticks, sustained crackles, and other small
defects. One ignores these, and enjoys the brilliance of the brass
overtones, the clear string section sound, and the full bass. On
Naxos, the comparable performance (the first of the 2-disk set) is
very muted and rolled off: there is nary a tick or pop, so the
computer software is working fine (or the disk source is different,
quiet but duller.) Yet some of the incandescent vitality of the
reading is gone along with the noise and highs.
I would just guess that the ENTIRE concert was recorded away from
Radio City, after some distance on the network lines: there is a trace
of that long-distance telephone line equalized circuit phase
distortion in the Naxos transfer that is TOTALLY missing from the
Dell'Arte copy of the Scottish March. Perhaps one might not like to
jump from "in house hi-fi" back to network mush, and back to "hi fi"
again, as the source material was alternated, so Naxos may have
decided to use a consistent set of source disks, if not the BEST copy
of any one selection.
This is not to say that the sound is in any way really bad...just that
it is not up to the examples we have from acetates made in 1940 at
NBC. Perhaps Seth Winner could elaborate on this, or another expert
could fill us in on the secrets...I am 3,500 miles away from the
Grail, writing from the west coast, as a mere record buyer and not a
"connected" member of the coterie of tape-traders in New York and New
Jersey.
"Nuages" has some of those "scrunches" that are audible on amateur
disk sets that I mentioned in the commentary on the '39 Creatures
excerpts above: this seems to indicate that the actual source material
is exactly the same as the Key/ATS edition that was mangled by
Olympic, and later by Gramofono, and NOT in-house professionally-cut
transcription acetates done on a lathe with a vacuum attachment to
keep the chip from clogging. At a duration of 6:42, Toscanini is
quite a bit more expansive in this broadcast than in his 1952
televised performance, which played in a quick 5:30. But I think that
Monteux's traversal of 7:00 (BSO, RCA 09026-61900-2) is even more
effective. There is, though, a wonderful intimacy in this performance
-- and in the entire program -- and a sense of steady, rapt
concentration that rises to a luminous inspiration in "La Damoiselle
Elue" and in parts of the "Iberia" and "La Mer" (which has a few
missing notes at 01:15 on track 3 of CD 2, in the "Dialogue of the
Wind and the Sea"; I am not sure if they were present on the Olympic
set, but assume they weren't. The producers have apparently chosen
not to paste in a replacement from, say, the BBC or one of the later
broadcasts in the manner that Music&Arts corrected the loss of
passages in the La Scala Beethoven 9th, or DGG repaired the 1943
Furtwaengler live BPO Beethoven Fifth. There is no "right" way to
handle this kind of problem; one can leave it alone or try to fix it;
Naxos has chosen the purist approach, the former.)
So I guess we will not hear all the works in flawless in-house hi fi
sound this side of the Elysian Fields. No matter: the entire
magnificent program, with Gene Hamilton commentary and valuable (and
much crisper sounding but crackly) rehearsal excerpts of 3 pieces,
brings a more rounded view of the Maestro's skills as a comprehending
Debussy interpreter than the few authorized commercial recordings.
CONCLUSION.
Naxos, Klaus Heymann, Series Producer Jonathan R. Wearn, archivist
Richard Caniell, and consultant John Ardoin are all to be
congratulated and encouraged. And thanks be to the Recording Angel
for the dedication of Richard Gardner. No one better than a complete
curmudgeon and Toscanini-hater would give less than grateful
Hosannahs.
Now, go forth and bring us MORE! And work with the powers that be to
issue these in the United States, the adopted land of Toscanini.
Yours,
A Fanatical Toscaniniphile [you can guess who]
John Wilson
(A Fanatical Toscaniniphile) wrote:
>As someone who recalls listening live to the broadcast of the famous
>Dec. 1953 "Eroica", and who has been collecting Toscanini records
>since my first childhood copy of the "Nutcracker" Suite in the early
>fifties, and as a broadcast host (and engineer) who devoted four
>continuous years to the weekly transmission of all the Toscanini
>material issued up to the year 1974, I have spilled some thousands of
>lines in previous posts in this n. g. discussing the technical and
>musical qualities of innumerable performances, recordings, and
>airchecks of the Maestro's work.
>
Remaining very thorough and excellent evaluation snipped. Many thanks
to the poster.
The answer to his question regarding the Heifetz/Toscanini NBC
Beethoven Violin Concerto recording is no. There was no performance
or broadcast before or after the recording. This is one of the rare
instances of a work being recorded by AT that had not been prepared
for a concert or broadcast.
John
Toscanini Web Site: http://home.earthlink.net/~jw3/index.html
nfna...@nospam.rocketmail.com
Metropolitan Operas transfer Richard Caniell's tapes too high. (Thank
God I held on to that variable speed CD player.)
Alrod
hows...@my-dejanews.com
ducky兀deltanet.com (Matthew B. Tepper) wrote:
> In article <361c0979...@nntp.ix.netcom.com>,
> >As someone who recalls listening live to the broadcast of the famous
> >Dec. 1953 "Eroica", and who has been collecting Toscanini records
> >since my first childhood copy of the "Nutcracker" Suite in the early
> >fifties, and as a broadcast host (and engineer) who devoted four
> >continuous years to the weekly transmission of all the Toscanini
> >material issued up to the year 1974, I have spilled some thousands of
> >lines in previous posts in this n. g. discussing the technical and
> >musical qualities of innumerable performances, recordings, and
> >airchecks of the Maestro's work.
> you'd make the acquaintance with this series!
Aw, Ducky, quit being so coy, and call an Old8H-Haggis and Old8H-Haggis,
willya? Welcome back!
K. Howson-Jan
-----------== Posted via Deja News, The Discussion Network ==----------
http://www.dejanews.com/ Search, Read, Discuss, or Start Your Own
Don'tBothe...@anywhere.com
Sorry to say that I do not have absolute pitch, and before writing my
review I did not think to get out a pitchpipe or go to our (constantly
retuned) pianos to check.
I have exquisitely sensitive sense of relative pitch, or the ability
to detect wow or flutter, and when I hear transcriptions joined
together that are even 2 Hz off pitch, I can detect it (as a record
producer and engineer, I once had to constantly admonish a lutenist to
retune during a recording session; he couldn't detect when his lute
went out, but I could)...the BMG set is very good as regards constant
pitch, the Gramofono and Iron Needles vary more than a semitone in
some works, such as the Martucci First Symphony, and the Naxos set did
NOT jump out and alert me to pitch problems in side-joins...I just had
time to listen to 5 of the 6 albums all the way through and to sample
the sixth, so what I heard in order to be able to write down my
quickest impressions of transfer quality did not have demonstrate
noticeable flaws to my non-perfect-pitch trained ears.
Even my wife, a concert pianist, pedagogue, and keyboard instructor of
30 yrs. experience, does not have perfect pitch, and cannot instantly
identify a speed error...we both consider this a sort of blessing when
investigating historical piano recordings!
There is a VERY slight speed variation in part of the second movement
of the Mozart Concerto No. 27, so slight that I did not want to water
down my positive review, since it was in the master disks they used
(former copies on tape of the program had this same defect): in
general, I wanted to concentrate on problems introduced by Naxos, if
any, rather than just the admitted defects of the source material, and
I was thrilled to find the Naxos production not only competently, but
also excellently accomplished.
John Corbett's transfers of Toscanini transcriptions, broadcast on
WRVR, had considerable pitch problems at side joins (he must have used
a table driven by a hysteresis-synchronous AC motor); of course,
anybody worth his salt these days uses a variable-speed electronic
servo turntable, or something like the monstrosity of Seth Winner
described in the notes to the Sony Masterworks Heritage series.
I did synch the Naxos transcription of the Leonore Ov. #2 (1939, on
8.110813) with the RCA / BMG transfer on Vol 45, track 6 (same
performance) and found that they agreed to the Hertz.
I repeated this test by synching the VPO/Abbado DDD recording of Leo
#2 on DGG 429 762-2 with the Naxos AT transfer, and found that either
the VPO tunes their A about 2 Hertz higher, or BOTH the Naxos and RCA
/ BMG digitizings are slightly lower in pitch.
Beyond that, I do not care about absolute pitch problems, as we don't
intend to play or sing along with these old Toscanini broadcasts.
AFT
nfna...@nospam.rocketmail.com
wrote:
>On Thu, 08 Oct 1998 15:38:00 GMT, in rec.music.classical.recordings
>you wrote:
>
>>Did you notice if all the pitching was accurate? Some of the Naxos
>>Metropolitan Operas transfer Richard Caniell's tapes too high. (Thank
>>God I held on to that variable speed CD player.)
>>
>>Alrod
>Beyond that, I do not care about absolute pitch problems, as we don't
>intend to play or sing along with these old Toscanini broadcasts.
>
>AFT
Ah, the old whimsy is back, too.
No, the problem with the Met broadcasts was not so much the side
joins, though these are occasionally problematic. I don't have perfect
pitch either. I do find that a whole performance up a half-step can be
disorienting, as in Bach's C-minor Mass or his Air on a G-sharp
String.
Alrod
Donald Rice
hows...@my-dejanews.com wrote:
> In article <6vh7c0$8...@journal.concentric.net>,
> ducky兀deltanet.com (Matthew B. Tepper) wrote:
> <snip>
> > Aha! If you're who I think you are, I was particularly hoping that
> > you'd make the acquaintance with this series!
>
> Aw, Ducky, quit being so coy, and call an Old8H-Haggis and Old8H-Haggis,
> willya? Welcome back!
>
> K. Howson-Jan
>
I would like to add my welcome to AFT and my sincere appreciation for his
efforts in assessing the Naxos releases for us. The authoritative technical
broadcast-engineer audio analysis, penetrating musical insights, the piano
teacher wife all resemble the sorely missed anonymous "Old 8H". Of course
this COULDN'T be him because he signs his post "AFT".
AFAI am concerned, AFT is a damned fine substitute for 8H and "he'll have to
do till the real thing comes along."
Don
A Fantatical Toscaniniphile
wrote:
>Did you notice if all the pitching was accurate?... the problem with
>the Met broadcasts was not so much the side joins, though these
>are occasionally problematic. I don't have perfect
>pitch either. I do find that a whole performance up a half-step can be
>disorienting, as in Bach's C-minor Mass or his Air on a G-sharp
>String.
>Alrod
I did some further analysis, and came up with results you may find
useful.
First, my wife can actually trace a certain amount of musical lineage
back to Dame Myra Hess, who taught her most prominent teacher. My
wife has the C Minor concerto in her repertoire, and has performed it
in public.
I played the Naxos transfer of the Hess/Toscanini last movement for
her in our music studio. I asked if she thought it was OK, or if
there was "anything obviously wrong" (remember, she does not have
perfect pitch.)
She scrunched up her face and said, "It sounds sort of strange!" She
went to our grand and played a few passages: our recently tuned Yamaha
C-7 was very much flatter sounding than the CD. So, on a careful
comparison the CD was at least a semitone sharp; we confirmed with our
Casio tuner that our own grand piano was OK.
Then I checked excerpts of a few other items as time permitted:
The Mozart Concerto No. 27 finale: it is bang-on with our piano.
The Mozart Haffner: out of about 6,000 CDs in our collections, it was
very hard -- if not impossible -- to find an all DDD recording by an
American orchestra that tunes to A=440. So I had to settle for the
DDD Capriccio recording of 1990 by Graf and the Salzburg Mozarteum.
It was very close to the Naxos Toscanini; maybe off a Hertz or so.
But you may note, if you are a very sensitive listener, that in the
opening of the Naxos Haffner and at least a bit of the 2nd movement of
the Concerto that the original transcription disks had a VERY slight
wow; I have heard MUCH worse than this on competing labels' issues of
Toscanini, and even in Victor's own LP transfers of certain material:
I believe I recall that a sixties LP reissue of the Heifetz/AT
Beethoven Concerto had very bad wow during some of the 78 sides,
though I checked my LP pressing and found it to be perfectly on
center.
Beethoven Eighth: my ONLY DDD copy was Kegel's; all our other CDs were
ADD; Kegel and Toscanini's '39 broadcast were about as far off as the
Haffner situation: close but not absolutely dead on: this could be
different orchestral tuning, European vs. American. But it was
nowhere near a semitone off.
Beethoven Fifth: Bang on in comparison to Kegel.
Beethoven Triple Concerto: agreed perfectly with the Barenboim DDD on
Angel, or the Walter ADD on Sony.
Debussy La Mer, Second Movement: Naxos '40 is perfectly in tune with
RCA/BMG Toscanini of '50.
Mendelssohn Violin Concerto: Heifetz on Naxos seems nearly a semitone
sharp compared to Heifetz/Munch on RCA 5933-2-RC compact disk
transfer; I don't have the latest Living Stereo version.
So the only performance that is so noticeably off speed as to POSSIBLY
make a slight difference in musical impression is the Beethoven 3rd
Concerto; I don't think the Heifetz pitch error is quite as much, but
it is still somewhat sharp. I could NOT find an all DDD American
recording of the Mendelssohn in my library...just scads of great old
violinists from Stern back to Szigeti and Kreisler. I did not have
time to check everything with the piano, or to try all the movements
of each work: that would have been a day-long task.
Now, I have a MODEST PROPOSAL for digital software engineers: an
automatic electronic wow-corrector.
It seems to me, as a former dabbler in designing analogue servo
circuits for audio control devices, that it would be possible to make
an all-digital or hybrid analogue/digital gadget that could be used to
correct pitch errors in old historic recordings.
The audio signal can be passed through a very sharp filter with about
a very sharp passband, tunable by the producer for the prime key of a
musical work, or set to detect an A if it falls within a given
frequency band (i. e. 440, or whatever one decides the orchestra was
tuned to; it would be helpful to know some historical facts about the
ensemble's tuning preferences.)
It would probably be necessary for a musician to supervise this
process, using a score so that only certain parts of the music would
be sampled, since modulations and key changes might confuse the
circuitry. The last "good pitch" could be maintained until the
sampler had another appropriate musical section to work with.
Using a phase-locked loop, the filter output capturing the the pitch
can be sampled and "grabbed" and used to lock a servo control that
corrects either the analogue playback motor speed, or preferably a
high-sample-rate digital pitch shifting circuit. My own personal
engineering experience with many all-analogue feedforward or feedback
audio servos, and with one digital servo used to control a commercial
microscope focusing device, would indicate that complete correction
and lock could be achieved in perhaps 10 to 20 milliseconds...I
believe that the ear of the average listener cannot discriminate pitch
in less than at least a few tens of milliseconds to a hundred
milliseconds.
If the phase lock method was not practical, then perhaps the recording
could be played through once while a "map" was created of the major
tonality; then that map could control the pitch-shifting circuitry as
the recording was dubbed through the system.
This could quickly and very transparently compensate for the wow of
off-center pressings, or disks cut on transcription tables that were
unstable and had unpredictable speed fluctuations, or tapes that were
made on poor transports or warped reels. For example, the 1951 dress
rehearsal transcription of Meyerbeer's L'Etoile du Nord opera overture
by Toscanini is severely wowed in a very unstable and unsteady manner:
such a pitch corrector could theoretically fix it precisely!
Any takers, you young whippersnapper digital design gurus?
For me, one of the joys of CDs is the absence of the usual end-of-side
wow of offcenter LP pressings (when I did my analogue tapes of
classical music programs for radio play, I often fixed offcenter disks
by removing the spindle of my table and mechanically centering the
grooves on the platter so that the tone arm did not wobble back and
forth. Even the slightest typical pressing wow used to drive me
crazy, as it gets worse and worse as the stylus plays through the
record. After a 30-minute side, I would be climbing up the walls in
frustration, unless I were just listening very casually for background
music purposes.)
Any modern CD transfer of an old analogue tape or disk can have the
wow inherent in the original recording, and other than by re-centering
a transcription, I know of no current manner of dealing with it: no
engineer could reliably fiddle with a digital pitch shifting control
circuit or turntable speed control "on the fly": after a few seconds
of concentrated effort, the results would become sloppier and sloppier
until the cure was worse than the disease.
I have not been involved with any digital tape editing since around
1990, so I do not know the state of the art in this particular
situation, but I assume that the problem has not been solved, because
we are still getting disks from all manufacturers that have pitch
variations, not just simply mispitched steady speeds, in historical
analogue transfers.
AFT
A Fantatical Toscaniniphile
HARDWICK'...my brain meant to write ANDERSON, his proper last name,
but for some reason my fingers typed Hardwick...I must have been
looking at an EMI historical CD transfer when I wrote this article.
I emailed Keith Anderson when I heard last year about the planned
historical series because his personal address was the one most
prominently featured on the HNH website.
In my suggestions for repertoire selection, I advised issue of the
great 1941 concert including the Pastorale, which has never been
widely distributed; the Schubert-Joachim "Gastein" Symphony; the
delectable Vieuxtemps Ballade and Polonaise, and many other items
which were issued on bad private labels with poor sound, or played in
modern times only on the WRVR concert retrospective broadcasts. I have
no idea if these works are contemplated, but I do hope that they are
included in future releases.
The program notes on the Naxos Toscanini Concert Edition are
absolutely extraordinarily fine. Bill Newman's analysis of La Mer is
astonishing!
AFT
Don Drewecki
I haven't heard the Naxos titles yet, so I thank the former head
of the Unofficial Toscanini Idolators Club for reviewing them.
For me, it's a thrice-yearly pilgrimage to the Toscanini Archives
to hear the actual masters, recently transferred in the analog
domain and with no post-production to provide the acid-test
reference for comparing bootleg CDs such as these. Against this
standard I will eventually make my own judgment -- right now, my
time is spent marveling at the great video and audio work of the
BBC Proms 98 telecasts and Radio 3 stuff.
Anyway, it just seems unlikely that, however sparing and careful
these CDs are, they can truly match the masters. My reference
point is the NBC broadcast of February 4th, 1939, which never
sounded good in the WRVR broadcast or all boots, but which on the
original NBC discs just blows me out of the room -- albeit with
surface noise that occasionally is strong -- just fantastic mono
sound. I just am not sure the Naxos CDs will ever sound as good
as this. Not to throw a wet blanket on this, of course, but I
have my doubts.
Also, the Library of Congress had the "unknown" alternate take of
the 1940 Beethoven Concerto with Heifetz, not RCA. But RCA does
have the legal right to issue it, and should do so now in the
20-bit series we've talked about here months ago. It would be
great if Naxos had such access, but it doesn't, though it may
produce better results than Key's ATS, or Turnabout or Olympic.
--
Don Drewecki
<dre...@rpi.edu>
Lani Spahr
>
> I repeated this test by synching the VPO/Abbado DDD recording of Leo
> #2 on DGG 429 762-2 with the Naxos AT transfer, and found that either
> the VPO tunes their A about 2 Hertz higher, or BOTH the Naxos and RCA
> / BMG digitizings are slightly lower in pitch.
I'm sure this is because the VPO tunes higher, as do all orchestras in
Europe compared to US orchestras.
Wonderful info, BTW. Much appreciated. Now if we could only buy the damn
things in the US.
--
----------
Cheers,
Lani Spahr
Bruckner Symphony Versions Discography
http://www.mv.com/ipusers/reingold
A Fantatical Toscaniniphile
>I haven't heard the Naxos titles yet, so I thank the former head
>of the Unofficial Toscanini Idolators Club for reviewing them.
I appreciate it.
>For me, it's a thrice-yearly pilgrimage to the Toscanini Archives
>to hear the actual masters...My reference
>point is the NBC broadcast of February 4th, 1939, which never
>sounded good in the WRVR broadcast or all boots, but which on the
>original NBC discs just blows me out of the room
This is EXCEEDINGLY tantalizing, as the Maestro's interpretation of
the Mozart Prague Symphony was wretchedly distorted on the Turnabout
LP editions produced in, I believe, a single disk release, and in a
collection of historic conductors. I am not sure if there was any
other LP release, but the CD by Iron Needle is just as distorted, but
is also very mushy with lots of fake stereo echo, phase shifting
between channels, digitized audio delay, and so forth. To think that
the original lacquers sound as good as, say, the Winner transcription
of the Petruchka tableaux (maybe not QUITE as good, since it was a
year earlier and I am sure the miking was a bit different) is
instructive.
I have not had the privilege of hearing any of the recordings at the
R&H Museum collection...Don, which of the RCA/BMG disks, or the best
of the current CDs, begin to live up to the authentic archival source
material?
My impression as an auditor is that the Dell'Arte copy of the Marche
Ecossaise must surely be absolutely second-generation: magnetic tape
or digital dub copied right from the source acetates; ditto the BMG
'39 Eroica, Petruchka, Sibelius 2nd, Shostakovich 7th; same for the
Music&Arts transfer of the Heldenleben, Don Quixote w/Feuermann,
Sibelius 4th.
On the other hand the Dell'Arte Graener suite is undoubtedly taken off
a wideband AM tuner (hetrodynes are a giveaway); the Key/Olympic plus
the Naxos transfer of the 1940 Debussy concert, the Music&Arts 1940
Manfred, and the Key/ATS and the Iron Needle transfer of the Martucci
Canzone dei Ricordi are probably taken from narrower bandwidth AM
transmission.
The BMG copy of the Leopold Mozart Toy Symphony is emphatically taken
off long-distance telco lines, possibly as far away as Los Angeles:
you cannot replicate that tell-tale phase distortion any other way
(and we know that there are 2 better copies: the R&H newly restored
version, plus an old collectors' tape that was taken off W2XMN-FM but
had EXCRUCIATING distortion on the cymbal crashes (overdeviation of
the FM transmitter? Too narrow an IF bandwidth on the FM receiver?
Perhaps...)
I am told that there are amateur offair 15 ips tape copies of the
post-1950 radio programs that are actually sweeter sounding and wider
in dynamic range with improved presence compared to the RCA / BMG
transfers of things like the Gluck Orfeo, Saint-Saens Organ Symphony,
and '53 Eroica; and that there are archival linecheck tapes made close
to Rockefeller Center that sound better than the 15 ips Walter
Toscanini collection copies of things like the live '51 Beethoven
Fourth.
So it is frustrating to know that the only way to hear the ultimate
reality is to go to NY...a friend of mine did that on a regular basis,
too, and saw the televised concerts years before I could get the BMG
videocassettes.
Still, we record collectors out here in the wide world come closer --
for the most part -- than ever to the Toscanini broadcasts by means of
this new Naxos series, to judge from the collective value of the six
sets I sampled. There is ALWAYS a better copy, according to somebody
in the know...but if I can just hear MOST of the music, minus all
kinds of audio fakery, I am willing to settle for that, since a 7,000
mile round trip to NY is not a practical alternative.
Incidentally, I confirmed my recollection that the Danse by Debussy
(Ravel orchestration) in the all-Deb. 1940 concert on Naxos
8.110811-2, is a completely different transcription source than the
other extant compact disk transfer, in rather ghastly fake stereo on
Music&Arts CD898. On the former (as well as the Key/Olympic double LP
set) Gene Hamilton talks over the beginning of the music; on the
latter, an earlier-level feed before the NBC network transmission mix
is employed, sans the announcer and with no talking over the opening
bars. Yet the M&A does not sound nearly as good as the superb
Dell'Arte transfer of the opening work of the program, the Scottish
March. I suppose the R&H collection has it all in glorious MONORAMA.
AFT
Matthew B. Tepper
>
> So it is frustrating to know that the only way to hear the ultimate
> reality is to go to NY...a friend of mine did that on a regular basis,
> too, and saw the televised concerts years before I could get the BMG
> videocassettes.
Actually, the televised concerts were toured to various museums across
the country in the Spring of 1986; I saw all of them except for the
two-part _Aida_ at the Walker Art Center in Minneapolis.
Sacqueboutier
While pondering the mystery of our "Fanatical Toscaniniphile",
I first thought it might be yourself (I still think you are
knowledgeable enough to have written the article and you
are sneaky enough to respond to your own post with your
real name just to throw us off).
But the only one who seems to want his privacy as
of late is Doc Gonzo. I saw a very long list of CDs
for sale by the good doctor last week. It was a fantastic
list. A perfect example of what such a post should be
like. I also remember his knowledge and appreciation
of AT's work. Could the doc be trying to make a
covert comeback?
--
Don
)**********************************************(
)* Don Patterson *(
)* Asst. Principal Trombonist *(
)* "The President's Own" *(
)* United States Marine Band *(
)* don...@erols.com *(
)**********************************************(
http://www.marineband.hqmc.usmc.mil
)**********************************************(
)* DCP Music Printing *(
)* Professional Computer Music Typeset *(
)* Music Arrangements *(
)* don...@erols.com *(
)**********************************************(
The views expressed are my own and in no way
reflect those of "The President's Own" United
States Marine Band or the United States Marine Corps.
Matthew B. Tepper
>
>
>While pondering the mystery of our "Fanatical Toscaniniphile", I first
>thought it might be yourself (I still think you are knowledgeable
>enough to have written the article and you are sneaky enough to
>respond to your own post with your real name just to throw us off).
>
>But the only one who seems to want his privacy as of late is Doc
>Gonzo. I saw a very long list of CDs for sale by the good doctor last
>week. It was a fantastic list. A perfect example of what such a post
>should be like. I also remember his knowledge and appreciation of
>AT's work. Could the doc be trying to make a covert comeback?
>--
>
>Don
Nope, it ain't Doc Gonzo and it sure as heck ain't me! I've been in
touch with Gonzo and 8-H lately (the latter values his privacy, and I
won't violate it), and I can't see them as two sides of the same coin.
A Fantatical Toscaniniphile
In fairness to Naxos and Richard Caniell, the producer of the
Toscanini Concert Edition, I would like to add a few clarifications to
the comments I made about the 1940 All-Debussy concert. I wrote:
>I would just guess that the ENTIRE concert was recorded away from
>Radio City, after some distance on the network lines: there is a trace
>of that long-distance telephone line equalized circuit phase
>distortion in the Naxos transfer that is TOTALLY missing from the
>Dell'Arte copy of the Scottish March. Perhaps one might not like to
>jump from "in house hi-fi" back to network mush, and back to "hi fi"
>again, as the source material was alternated, so Naxos may have
>decided to use a consistent set of source disks, if not the BEST copy
>of any one selection.
>This is not to say that the sound is in any way really bad...just that
>it is not up to the examples we have from acetates made in 1940 at
>NBC.
The information I wanted was, in part, right under my nose, and I
missed it! I finally got through the astonishing 20-page booklet that
accompanies the Debussy set (imagine Grammofono or Iron Needle doing
something like that...we are lucky if they get the tracks labelled
right!) On my first reading I had somehow skipped the brief comment by
Bill Newman on p. 10:
"The All Debussy broadcast stems from a tape given to Richard Gardner
by Toscanini during the period he worked with the Maestro at
Riverdale. Gardner, in turn, made a gift of it to Richard Caniell."
Thus, at least for this particular set in the series, the transfer was
not made direct from the "first strata of recording" to digital tape,
but from an already existing magnetic tape copy, dated obviously
during the lifetime of Toscanini (likely made post-1950.) On other
issues of the Naxos Toscanini set that I acquired, the technical notes
indicate that transfer was done direct from lacquer disks; here as
stated above, it is prepared from an already produced (obviously
analogue) audio tape.
Working from impressions only of the sound quality, in contrast to the
Dell'Arte issue of the Scottish March, I could detect no substantial
HF response in the Naxos or Olympic transfers above about 4 kHz,
though there is some hiss that is consistent with the random 1/f noise
of tape. I also hear the peculiar lack of transient accuracy and
clarity that results from extremely high-order filtering and its
consequent heavy phase shift: if, say, the source disks were processed
with very hi-Q multipole filters to cut off scratch noise, the phase
distortion would be evident down into the midrange, which I hear. On
the other hand, this type of loss of phase coherency is obvious on
equalized narrowband 5 to 8 kHz phone transmission lines that were
used for radio: 5 kHz for network transmission, and 8 kHz for AM
transmitter audio connections; later 15 kHz lines were available for
FM. At least the 5 kHz lines, if not so much the 8 kHz circuits,
would cause this same phase distortion. So it may not be likely that
the broadcast came down as far away as I suggested, but it surely DID
come some distance from immediate in-house connections to Studio 8H,
and did pass through some equalizers. I think this may explain the
similarity in sound of the Key/Olympic and the current Naxos
historical transfer, as Gardner may have supplied the tape to Clyde
Key back in the sixties.
Knowing that the Maestro himself, through Gardner and Caniell,
ultimately provides us with this invaluable and cherished tape makes
its presence in the Naxos series all the more treasurable!
Caniell notes, in a touching remembrance of the engineer: "I remember
Gardner now as clearly as if I saw him yesterday, pulling out 16"
lacquer disk copies of rehearsals to share with me some newly
discovered and inspiring work by 'the old man', as Gardner
affectionately termed the Maestro. I remember, too, Robert Hupka's
beaming and beneficenyt expression and the underlying religiosity
whcih made for so real a humility in his personality and lent such
transparency to his photographic art."
It is deeply satisfying to read these lines and know that the
comprehending hand of a real expert and enthusiast, who was actually
"Present at the Creation", has guided this project.
AFT
A Fanatical Toscaniniphile
That sort of thing is what happens when I try to spell something other
than my own name, which is NOT Dr. Gonzo!
AFT
Joe Salerno
Toscanini/Horowitz. Has a recording survived? Where is it?
--
Joe Salerno
Location Production in Betacam-SP Format
Real Time Non-Linear Editing on Matrox Digisuite/Speed Razor 4.0
PO Box 1487
Bellaire TX 77402 USA
jsal...@electrotex.com
Please note new new addresses:
http://www.angelfire.com/biz/ivs/index.html
http://members.tripod.com/~joesalerno/index.html
ICQ#: 20105039
A Fantatical Toscaniniphile <Don'tBo...@Nowhere.com> wrote in article
<361e4455...@nntp.ix.netcom.com>...
tlst...@tpgi.com.au
nfna...@NOSPAM.rocketmail.com wrote:
> Did you notice if all the pitching was accurate? Some of the Naxos
> Metropolitan Operas transfer Richard Caniell's tapes too high. (Thank
> God I held on to that variable speed CD player.)
>
> Alrod
Being digital, I'll be most surprised if changing the speed changes the
pitch. Let's know how you go.
--
Cheers!
Terry
Matthew B. Tepper
tlst...@tpgi.com.au pondered what I'm pondering as follows:
You're digital?
>--
>Cheers!
>
>Terry
A Fanatical Toscaniniphile
Fantatical Toscaniniphile) wrote:
>NOTE and ADDITION to my Review of Debussy Concert:
>>Radio City...
My first impressions, posted on 10.7.98, were taken after listening to
the entirety of 5 of the 6 sets purchased, on speakers; I audited and
compared several sections of the sixth, the Debussy concert, but had
not played every note of all the tracks.
Now that I have heard the Debussy concert of 13 April 1940 nearly 3
times (once on speakers, once on 'phones, and once on 10.7.98 as
above, on speakers in samples of selected tracks) I can give a
slightly more accurate account of the sound quality.
The first 5 tracks (Marche Ecossaise, Danse, La Damoiselle Elue and 2
of the Nocturnes movements) are rolled off above about 4 kHz, but have
audible hiss in the spectrum above that. However, as stated, the
sound is pure, solid mono throughout the set: no faked stereo, added
ambience (either mono OR stereo); no "squizzy" computerized noise
reduction.
Thus, the sound of the "Damoiselle" is vastly superior to the faked
stereo equivalent in Music&Arts CD-898, disk 2, where a digitized
audio delay has been applied that duplicates the signal, delayed about
50 milliseconds from the left to the right channel; in addition there
is a platform of phony stereo ambient echo, and a wobble in relative L
& R channel phase (which I assume has been done by something like a
Urei digital processor): the result is so ghastly that I cannot STAND
to listen on stereo over 'phones; even summing to L+R mono fails to
cancel all of the added artifacts. This is described in a box on the
last page of the booklet as "virtual stereo" but for me, it is virtual
crap!
The "Danse" from that M&A set (very expensive, for ONE usable monaural
disk -- no. 4, with Franck -- and 3 hideous phony stereo ones that
cannot be tolerated) is indeed sans any trace of Gene Hamilton's
pre-announce, which is heard over the first bars on both the
Olympic/ATS LP copy and the Naxos edition. The "Marche" has very soft
ticks in the M&A edition, totally missing on the Naxos. But, the
Naxos transfer has one of those disk-defect "scrunches" between 4 and
5 seconds' elapsed time on track 1, totally missing on the M&A: this
scrunch is also present at the same place in the music on the
Dell'Arte transfer, which has higher surface noise, but more
transparency. I do NOT believe that Cedar de-clicking or de-crackling
has removed it from the M&A, but that the source disk is different;
that the Naxos/Dell'Arte transfers have the same identical
transcription disk as their source, though a different one was used by
M&A.
Now, the 4-kHz bandwidth sound of the Naxos continues all the way
through the first 5 musical selections, but slightly and noticeably
improves for the remainder of the program: Iberia and La Mer. Highs
are extended up to about 7 kHz now, and there is a very distinct
change in audible spectrum at the start of track 7 (1st movement of
Iberia): more highs, and more audible hiss and soft, swishing groove
noise. In the "Fetes" the cymbals are so indistinct and mushy that
they do not stand out from the midrangy, clattery sound of the
climaxes; but in Iberia, the castenets and other percussion
instruments are suddenly clearer, sharper, and more well defined. The
sound is about as good as the best extant copies I have heard of the
Feb. 2, '48 broadcast, and at least as clear as the RCA / BMG transfer
of the 1941 Philadelphia shellac set, where -- however -- the mike was
further back with less close-up presence than in 8H in 1940. There is
NO trace in this 1940 Iberia transfer of the phase distortion effect
that I can hear in tracks 1-5. Perhaps this work was dubbed from a
totally different in-house lacquer set, augmented with linechecks or
airchecks of the prior pieces (with their announcements) when the tape
was prepared during Maestro's lifetime.
(Thanks, Naxos, for not cranking up Cedar to remove the very faint
"swishes" and then by doing so ladelling on all kinds of "squinchy"
ringing effects, as are heard on diverse Sirio, Iron Needle, and
Grammofono historical transfers.)
With the La Mer on Disk #2 of the Naxos set, we sort of "split the
difference" in fidelity and transparency between the good Iberia and
the muddy earlier tracks: there is less distinction and clarity than
in the Iberia, but a bit more transparency than in S. 1 tr. 1-5. I
also hear analogue tape irregularities (very tiny, tiny HF dropouts on
sustained passages) as well as the missing notes of music in the 3rd
movement, described in my original post.
Now, I no longer have the Olympic set for immediate comparative
reference, but it was muddier than this Naxos transfer and suffered
from the artifact of having had heavy-handed audio expansion applied:
the levels swell artificially on every note above a certain volume
threshold. The entire LP production was so indistinct that these fine
shadings that I can hear, comparing track to track on the Naxos
transfer of what is probably the same source tape, were not audible,
since they were overwhelmed by the EQing and level processing done by
Everest "engineers" (butchers) in L. A. for the issue of the original
vinyl set c. 1973.
So, until some other transfer surfaces of the first five tracks of the
initial disk of the Naxos collection, I am sure that this new set
(8.110188-2) in the Toscanini Concert Series is likely to be the
"state of the art" for the entire broadcast.
Between you, me, and the lamp-post, I would guess that the Dell'Arte
is PROBABLY a transfer of somebody's copy of the WRVR rebroadcast of
the early sixties, which is likely to have been made from acetate to
tape transfers by John Corbett, sans any aggressive attempt at
re-EQing or noise reduction. The Toscanini tape given by the Maestro
to Richard Gardner, and thence to Richard Caniell, and issued by
Naxos, is obviously prepared originally as a compilation of as many as
3 sources: AM aircheck audio for S. 1 tr. 1-5; in-house but not
perfect disks of Iberia; and fairly wideband offair (peak-limited)
acetates of La Mer.
Somebody once took me to task in some comments I posted here about
comparisons of THE REINER SOUND in various transfers, when I stated
that I used my headphones for the auditioning of subtle differences.
Well, unless one played the whole broadcast at about 95 dB SPL on
speakers, these very slight subtleties that I have detected in the
Naxos set will just go unheard...but on 'phones, one can immediately
tell the differences. In fact, I usually use 'phones to listen to my
classical CDs, so any artificial stereo, rechannelling, ambient depth,
or whatnot stands out as a disembodied presence, hovering around the
virtual soundstage. For example, in the M&A transfer of the
Furtwaengler 1951 Brahms First (CD-941) we hear a nice, solid, and
very transparent MONO orchestral recording of the orchestra centered
"right between our ears", but it is "surrounded" by a disembodied,
blowsy stereo echo. However, in certain Grammofono 2000 Toscanini
releases, such as the Brahms German Requiem, the ENTIRE orchestral
recording is dispersed helter-skelter, bit by bit, across the stereo
soundstage, and wobbles around for the whole performance due to the
ghastly delays and continual unstable phase-shifting.
But, play these Naxos Toscanini Concert Series CDs on 'phones, and you
hear a nice, solid, centered, identical left & right MONAURAL: it
takes about 2 seconds to mentally adjust to this, and then one forgets
about stereo or sound PER SE and concentrates on the wonderful music.
I will continue to rant and rant and complain to record producers
about their ridiculous and idiotic notions that fake stereo is
"better" for modern listeners than real, authentic single-channel
monophony; so far, attempt it as they may, they have not proved it to
my satisfaction.
Thus, one is somewhat distressed to learn that RCA & BMG are going to
remaster their excellent nearly all-mono Toscanini Collection with
what I have read as being described as "added ambience"...surely this
will be the stereo echo platform that is audible on about 3 or 4 of
the disks in the series (Dvorak New World, Strauss Don Quixote, and
the light music disk, Vol. 40) while almost all of the rest are
genuine mono, as they should be.
Why do so many buyers seem to tolerate what the crass producers are
doing to our historic pre-stereo repertoire? Is is a case of
tin-earitis, or a lack of comprehensive experience with the realism of
single-channel recordings? Do folks REALLY prefer to hear a bunch of
modern echoes and phase shifts added to properly balanced old monaural
recordings? If so, then why has Mercury been scrupulous in issuing
their domestic Living Presence monaural material; why has Sony
retained mono it its old shellacs, lacquers, and early tapes issued on
the Masterworks Heritage Series; why has EMI preserved genuine mono in
its References series; and why has DGG retained mono in its
appropriate pre-stereo Originals? Do these PROFESSIONALS in the
record business know something about sound that the amateurs at
Grammofono, Arkadia, Iron Needle, M&A have not yet learned?
Praise be to Naxos for giving us MONO Toscanini!!!!!!
AFT
Rtsina
>
>Praise be to Naxos for giving us MONO Toscanini!!!!!!
>
>
Interesting. What else do you do these days, Steve, other than write ten
screen postings? Usenet may run out of bandwidth.
Greetings to Frau Blucher.
DT
Curtis Croulet
>You're digital?
He probably has approximately 10 fingers and approximately 10 toes, so, yes,
he's digital.
Martin J Haller
A Fanatical Toscaniniphile wrote in message
>On Sat, 10 Oct 1998 02:53:13 GMT, Don'tBo...@Nowhere.com:
(Excellent description of sonic quality of Naxos Toscanini Debussy
issue much snipped.)
>sound is about as good as the best extant copies I have heard of the
>Feb. 2, '48 broadcast, and at least as clear as the RCA / BMG transfer
>of the 1941 Philadelphia shellac set, where -- however -- the mike was
>further back with less close-up presence than in 8H in 1940
Thank you for engaging my interest about these Naxos issues enough that
I
will be ordering two of the Beethoven and the Debussy programs. The
Philadelphia recordings mentioned have always been among my favorites ever
since RCA issued the complete collection in 1976. (Indeed, the Schubert C
major has spoiled me for any other recording.)
>So, until some other transfer surfaces of the first five tracks of the
>initial disk of the Naxos collection, I am sure that this new set
>(8.110188-2) in the Toscanini Concert Series is likely to be the
^^^
I believe you originally posted the catalog number as 8.110811-2. At
any
rate, although I am normally only a lurker here, let me join the others who
have welcomed you back to r.m.c.r. I was pleased to see your first post a
few days ago. Fate (as well as Ducky, no doubt) must have had a hand in it
as I had just come upon a print-out I had made of your "Seminar On Classical
Recording Noise Reduction Techniques" only a few hours before seeing your
new post!
Marty
Martin J. Haller (mjha...@acsu.buffalo.edu)
"As you gain experience, you'll realize that all logical questions are
considered insubordination." - Dilbert
A Fanatical Toscaniniphile
of the Toscanini collectors, Dick Tsina, who certainly helped
stimulate my own interest in the rare broadcasts that I had not heard.
His musical perspectives and technical ear make him a valued source of
knowledgeable opinion, should he have the time or inclination to
present any alternatives to my comments.
AFT
Rtsina
Oh my....I haven't the strength.
Cheers,
DT
MT
– with a new giveaway nickname and the same inimitable style.
Regards,
MrT
emf
About the problem of altered pitch, I found the Beethoven's 3rd Concert
with Myra Hess recorded One half tone up. BTW the recording is very good.
The Mozart 27th with Horszowski seems related to three different takes.
The triple concerto seems to have the same source as Grammofono 2000 (out
a few months ago) but with better results (a pity the beginning noise of
the source surface). Some source problems also in Beethoven's 2nd
symphony. However I'm very glad about the Naxos operation (as a whole). I
did not get the Debussy as it was out in the Music and Arts 4CD package.
Bye
Ezio Ferdeghini
nfna...@nospam.rocketmail.com
>Being digital, I'll be most surprised if changing the speed changes the
>pitch. Let's know how you go.
Well, a little surprise is good for all of us from time to time.
Yes, it changes speed and pitch together, just like variable speed
turntables and cassette decks do. Altering pitch does mute the digital
output, so you have to use the analog outs for the defective
transfers.
Alrod
Don Drewecki
The Toscanini/Horowitz Brahms concerto from Lucerne DID survive and it
is in the Toscanini Archives. I have never heard the recording.
--
Don Drewecki
<dre...@rpi.edu>
Matthew B. Tepper
nospam.che...@geocities.com pondered what I'm pondering as
follows:
>
Different performances of the Debussy from Music and Arts, where it's a
much earlier Philharmonic-Symphony performance, put through the wringer
by Lowell(owellowellowellowell) Cross(ossossossossossossoss).
A Fanatical Toscaniniphile
comments: so I have cancelled my messages about the Naxos Toscanini
Concert Series to allow others to provide a balance to my very
exhaustively expressed views.
One could look them up on Dejanews, I suppose, to read the original
critiques. But at least the servers who respond to "Cancel" messages
will now have a lot more room for such interesting threads as the one
attached to Achindoir's screed.
AFT
Paul Penna
<Don'tBo...@Anywhere.com> wrote:
> I apologize for perhaps too eagarly hogging bandwidth for my voluminous
> comments: so I have cancelled my messages about the Naxos Toscanini
> Concert Series to allow others to provide a balance to my very
> exhaustively expressed views.
In my view, the best use of bandwidth is to exchange interesting and
valuable information, which is what you were doing. I'm in favor of your
continuing to post as often and in as great a length as you see fit.
Thanks!
Paul Penna
A Fanatical Toscaniniphile
>
> In article <362244...@Anywhere.com>, A Fanatical Toscaniniphile
> <Don'tBo...@Anywhere.com> wrote:
>
> > I apologize for perhaps too eagarly hogging bandwidth for my > >voluminous comments: so I have cancelled my messages about the Naxos > >Toscanini Concert Series to allow others to provide a balance to my > > >very exhaustively expressed views.
>
> Paul Penna
Thanks and in that spirit I'd like to add some documentation to my
claims about the fidelity of the old Toscaninis as heard, at best, on
this new Naxos series...it confirms Don Drewecki's great enthusiasm for
the authentic, original sound of unfalsified Studio 8H broadcasts as
recently transferred at the R&H Museum. (I'LL BE AS BRIEF AS POSSIBLE,
and leave this up for a bit and then cancel it; you can get it on
Dejanews later if desired.)
You might also note that this sort of data applies to the transcription
equipment used by Columbia Records after 1939 to master many, if not
most, of their classical albums at 33.3 rpm, copied to 78 disk media for
commercial distribution (the superb, crisp, clear quality being
available for us to hear on such modern CD issues as Bruno Walter's 1945
Mahler Fourth, 1947 Mahler Fifth, the Beecham NYP recordings of 1942,
the Mitropoulos Mahler First, and the Budapest String Quartet Beethoven
performances post-1939, all currently available in authentic mono in
either the Bruno Walter series, or the Sony Masterworks Heritage
releases.)
The type of portable recorder that could be used to take down
transcriptions at 78.26 or 33.3 rpm, probably similar to what was used
by the Carnegie Hall Recording Service that made the original 78 rpm
disk masters of the Toscanini 1942 Beethoven Triple Concerto (Naxos
8.110801) would have been something like a Fairchild 539-G, a unit with
a driven traverse cutting head and separate playback tone arm: under
ideal conditions, it could achieve a 44 dB S/N ratio below standard
maximum cutting velocity at 1 kHz, with a response that was inferior to
a studio cutter (probably extended at best up to 7 kHz.) It could cut
OUT-IN or IN-OUT but had no vacuum equipment to suck up the lacquer
"chip" so OUT-IN cuts were extremely hard to supervise (ergo, the
"crunches" of fouled cutters that I believe I hear on many of the
'amateur' aircheck disks.)
The slightly better RCA 73B pro cutter, with an MI-11850-C head, could
achieve unsupervised OUT-IN cuts, because an optional vacuum system,
RS-1A pump and reservoir could be attached. However, the resulting
system was not portable. It had a much better frequency response,
provided -- in part -- by its completely adjustable inductance
compensator pack for matching recording heads: measured optically under
a microscope, the unit could achieve a recorded response at 96-lines per
inch, 33.3 rpm cuts, on a medium-hardness aluminum based blank at about
75 degrees Fahrenheit operating temperature, of +/- 2dB from 30 Hz to
10,000 Hz, driven by a nominal 1 w. level.
HOWEVER, this could not be achieved in playback on the systems of that
time, due to irregularities in the networks used to compensate for the
recording characteristic, which was known as "constant velocity"
recording, using a variable "turnover" equalization to avoid overloading
the low end of the spectrum: the whole 'system' of record/playback,
optimally tuned, seldom worked better than +/- 3 to 5 dB above nominal 1
kHz reference. So, a dub made back in the forties from a perfect ET
would not sound as good as a modern copy made with today's phase-linear
precise solid-state equalizers: if the disk had not deteriorated, our
modern playback could be adjusted to get nearly to the specified
response of the "perfect" system (this is with tones only; never mind
the room acoustics, mikes' pickup, errors of bandwidth in mixing systems
or limiters; etc.
In general, one could fairly reliably obtain about 30 or 40 Hz to 8,000
kHz, within about 4 dB variation, at a low 1-2% harmonic distortion, or
perhaps 4% IM distortion, at nominal level. Potential recording level
on a constant velocity mechanical disk system varies all over the map,
due to differences in groove displacement with respect to audio
frequency, the "land" between the grooves, and the highest peak level:
that is why all, 100% ALL commercial big-time disk mastering in those
golden days was done with vacuum-tube peak limiters in the chain.
Sometimes they were set up to just "jump" a bit on the loudest note in
the piece, or sometimes -- as in the case of the Monteux/SF Symphony or
many of the Toscanini/NBC records of the mid-forties -- the limiters
were cranked up to about 20 dB gain reduction for the peaks providing
the loudest level on the cut.
As you know, "live" acoustical classical orchestral music often exceeds
60 dB dynamic range. So, if the absolute peak S/N ratio is 45 dB
nominally, allowing 6 to 8 dB for safety headroom, that leaves us 37 dB
or thereabouts to record the loudest to softest sounds of the NBC
Symphony. Too loud: groove shatter, cutting amp overload; horrible
results. Too soft: a waterfall of what was delightfully described in
the ng recently as the "frying bean sound". No WONDER they had those
old tube limiters in the chains of the recording studios!
What you hear on the Naxos disk of the '39 Beethoven Eighth was probably
cut on the current RCA studio transcription mastering machine that
achieved this kind of performance: yet the Naxos disk is what I would
call "near high fidelity" and blows me away: Don, I hope you get to hear
it to report if it sounds inferior to the Feb. 1939 "Prague" Symphony
that you love in the original source at the R&H Museum.
Fairchild's deluxe 523 Studio Recorder of the late forties could achieve
a S/N ratio of 55 dB (better than the earliest magnetic tape, if one did
not count the inevitable impulse noises of playback!) at a recorded
ideal response up to 10,000 kHz. Just above this top level, the cutter
head's physical resonance created a huge phase shift before a gigantic
plunge in frequency response, equal to a multipole lowpass filter.
Without the most painstaking adjustment of the impedance-cancelling
network, the depth of cut, and the damping of the armature, one would
not achieve this "laboratory" response quality. But if one DID, as in
the very well-controlled studio sessions at CBS with Walter, Budapest,
etc., the results sounded just about as good as 15 ips tape up to the
early fifties! I would guess that the RCA recorder used to take down
the 1940 Toscanini Petruchka provided nearly this performance: that is
why Seth Winner & Adrian Cosentini's dubs of that set of disks, or the
CBS Walter lacquers, or Ward Marston's transfer of the '39 Toscanini
Eroica sound so GREAT!
Now, by the standards of the type of equipment that I document here (the
reference being the technical data printed in THE RECORDING AND
REPRODUCTION OF SOUND, by Oliver Read, Sams, 1952, augmented with
information provided to me by various audio engineers, such as Bill
Griffin of Allegro Music, Edward W. Meece of early Pacifica Radio, or
the late Vernon Berlin) one surmises that the so-called "in house"
masters in the R&H Library's acquisition of the Riverdale- Wave Hill
collection and at RCA archives, like the 1940 "Petruchka", are indeed
indisputably better than the dubs, alternate sources, bootlegged
multi-generation taped copies, and airchecks used by Grammofono, Iron
Needle, some of the ATS LPs, and so forth. The "authentic" disks are
STILL in RCA's possession and under their control.
Stuff like the Madison Square Garden concerts, the live Carnegie Triple
Concerto of '42, and so forth, were done -- I assume -- for Walter
Toscanini and the family's private use, and made on pairs of
not-very-well-matched portable cutters like the one I described above.
The aircheck disks that Paneyko (spelling?) and others took down,
including some of the material collected by Richard Gardner, was
mastered on such machines, but under better supervision; some of the big
Blue and Red Network NBC stations would have employed pro-quality RCA
cutters, and they (or the Armed Forces Radio) would have used the best
quality gear to cut the disks, but were forced to take audio feeds from
5 kHz phone network lines: thus, the funny pinched quality of the "telco
grade" recording of the Leopold Mozart Toy Symphony from 15 Feb. 1941 on
Vol 40 of the Toscanini Collection (60308-2-RG).
By the measure of the equipment available to make the disks of the
Toscanini broadcasts, we are LUCKY to have what we can get!
Let me give you an example of what it took to obtain a usable cut:
1. You had to match the cutter impedance with your output amplifier if
you wanted a predictably flat response. This meant changing some taps
on a network, cutting a test, examining the grooves with a microscope,
measuring the displacement, and trying again (or comparing with a test
disk, a much poorer reference check.) Ergo: not all the pairs of cutters
used for a continuous recording of a live program were well matched:
this is one of the reasons why we often hear RADICAL differences in
response at side-joins of ETs made on 2 lathes.
2. The recording level was hard to set; it was of course easier if a
peak limiter were in the circuit. Many test cuts had to be done with a
variety of material. It was really an art to find a usable level that
would play back at least a few times before the grooves got distorted by
the multi-ounce tone arms of the day. A cut made "for the archives"
would have been low and clean; if done for test playback to simulate the
sound of a commercial pressing, maximum groove modulation would be used:
this kind of ET could be played about TWICE before it began to sound
wretched!
3. You had to compensate what you were doing according to the air
temperature, as the lacquer blank surface changed its flexibility and
softness.
4. What angle did you want to use for your cutter stylus? Shattering
and disk squealing (perhaps just what I hear on the first 78 rpm side of
the Carnegie Hall private recording of the Beethoven Triple on Naxos
8.110801 and all other earlier transfers) would occur if you got beyond
a certain point in setting the vertical angle: 5 degrees off, and you
were a goner.
5. The stylus was lowered into the groove by a micrometer adjustment on
a lever; presuming that you had made a test cut earlier to set the depth
and pressure, you knew what the setting should be.
6. The feed screw is engaged, and off you go. OUT-IN cut? Well, you
had that darn chip problem: without a vacuum, you had to start a good,
clean chip, and then use a camel-hair brush and carefully trail it
behind the cutting head to the center alignment pins that hold the blank
in place: start a nice spiral wind, and keep gently pushing the chip all
during the 15 minutes of time of the 16" 33.3 cut, while you
simultaneously watch the VU meter and set the level attenuator.
7. To match the recording quality all through the duration of the cut,
you might want to get fancy and adjust the compensator settings: every
few minutes, you change a tap on an RC-LC network to increase the highs
for an OUT-IN cut, or decrease them for an IN-OUT cut. If you didn't do
this, then the linear velocity of the stylus displacement caused the
high frequencies to decrease by one-half as the frequency was doubled.
That is another reason why so many of the rather casually-made linecheck
disk sets do not easily dub together for a consistent sound during the
entire performance: you can hear this in the finale of the Martucci Sym
#1 on the Grammofono transfer (made from the same tape source as the
Clyde Key/ATS LP set): at a side change, there is a huge increase in
highs during the coda of the symphony. A professional cutter, working a
commercial recording session, would make the adjustments of the
compensator, or use one that was mechanically linked to the cutting
screw. But the archival linecheck disks were almost NEVER made this
careful way, so the highs change radically at side-joins.
Now, friends, consider the SOURCE audio. It could be a nice hardwired
balanced pair right from the distribution patchbay at Studio 8H; or
along the 5 kHz equalized telephone line at a radio station anywere in
the USA or Canada; or a shortwave receiver picking up RCA's
international transmissions; or somebody's old Meissner FM tuner
receiving W2XMN-FM or WNBC-FM; or a TRF tuner dialed to an NBC AM
station. If an air signal, it passed through a limiter at RCA-NBC in
Rockefeller Center, innumerable telco equalizers and line amps (using
Western Electric triode tubes that were made at the time of World War
I), hundreds of 111-C repeat coils; on and on: that gets it to the radio
station sending it through a mixing console, distribution amp, and LOTS
of transformers, to a crude old peak limiter at the studio, more phone
lines, a limiter at the transmitter building, and then a guy sitting
there watching a scope and holding his hand on the knob that feeds the
transmitter (see this in the marvelous montage in the OWI film "THE HYMN
OF THE NATIONS"! That's just the way it was done, in the days when the
FCC mandated a First-Class Ticket engineer at the transmitter at every
moment!) Along the way, guys are sitting around at consoles and meters,
bored out of their skulls, with their hands on the pots trying to look
"busy" and to justify their wages to the program supervisors.
It's a wonder that we hear such fine sound on many of the recordings
made off the premises of NBC!
So, once again, I thank the late Richard Gardner for his diligence in
assembling all of this material (and, by extension, all the nameless
drudges who cut the magic into acetate grooves); I thank Richard
Carniell; and I especially thank Klaus Heymann for having the temerity
to try to provide Toscanini for us fuss-budgets!!
AFT
Owen Hartnett
<Don'tBo...@Anywhere.com> wrote:
>I apologize for perhaps too eagarly hogging bandwidth for my voluminous
>comments: so I have cancelled my messages about the Naxos Toscanini
>Concert Series to allow others to provide a balance to my very
>exhaustively expressed views.
>One could look them up on Dejanews, I suppose, to read the original
>critiques. But at least the servers who respond to "Cancel" messages
>will now have a lot more room for such interesting threads as the one
>attached to Achindoir's screed.
Please don't allow a sour note to discourage you from your excellent
efforts. I'd rather read a barrel of your bandwidth than a character from
some others. The miscreants must learn to use a kill file, rather than
just be an occupant of one.
There are devices to block posts from certain people on the net. We can
only surmise that those who choose not to use them have a destructive
agenda.
-Owen
alhe...@my-dejanews.com
MT <matr...@sprintmail.com> wrote:
> We welcome back our friend old 8-H, the walking Cyclopedia Toscaniniana
> – with a new giveaway nickname and the same inimitable style.
>
I, too, am glad to see him back, being a fanatical Toscanini-phile myself.
But please don't make too much of it, or we'll scare him off again!
--
August Helmbright
-----------== Posted via Deja News, The Discussion Network ==----------
http://www.dejanews.com/ Search, Read, Discuss, or Start Your Own
Don Drewecki
One small note of correction -- Seth Winner did the disc-to-tape
transfer of the Toscanini 1939 Eroica on the BMG transfer, taken from
the NBC acetates at the AT Archives. I think Ward or someone else got
the flat dub from Seth, and worked on it for the CD remastering in 1991.
The reason why I think the AT Archives sound is best is because the goal
Walter Toscanini had was to collect the best possible sources of every
last surviving sonic scrap of music his father conducted. However,
there have been good dubs of some of this material, and some of those
were made by engineers in their studios or homes at the time. The best
of these come very close, but the prime sources will always be the
original lacquers, and practically all of them are in the Archives. It
is a shame that no traveling exhibit, or broadcast series on public
radio, has been made of any of this material. But still, if you can get
to NYC, and you're a fan of the Old Man (as I certainly am), get to the
Archives, which for the next two years are located at 40th Street near
10th or 11th Avenues (I'm not sure because I haven't been to the
relocated site yet).
--
Don Drewecki
<dre...@rpi.edu>
Don Drewecki
Jack Pfeiffer's "production" of the Toscanini Collection often left out
information, or wrong attributions. Seth and Adrian are the engineers
at the Lincoln Center Rodgers and Hammerstein Archives, of which the
Toscanini Legacy is a part. They did the actual work of transferring
the NBC reference lacquers of the 1939 Eroica. Jack didn't make that
clear, just as he never spelled RCA's producer Macklin Marrow's last
name correctly -- he mispelled it Morrow, when it was Marrow, as in bone
marrow. He also screwed up the credits in the video releases,
misspelling Don McCormick's last name too.
Seth is also responsible for most of the transfers of pre-tape materials
in the Sony Masterworks Heritage series and most of the Bruno Walter
edition too, and worked with the long-stashed Columbia lacquers. he
also did the recent Beecham set for Heritage.
--
Don Drewecki
<dre...@rpi.edu>
A Fanatical Toscaniniphile
>One small note of correction -- Seth Winner did the disc-to-tape
>transfer of the Toscanini 1939 Eroica on the BMG transfer, taken from
>the NBC acetates at the AT Archives. I think Ward or someone else got
>the flat dub from Seth, and worked on it for the CD remastering in 1991.
Thank you for the inside info. All I was able go by was the precise
statement "Disc to digital tape transfer: Ward Marston"
in the program note booklet for Vol. 23 of the BMG Toscanini
Collection, which is followed by the statement: "R&H Archives
assistance: Adrian Cosentini and Seth Winner". From that, I assumed
exactly the opposite of your information. Since you're a close
associate, I assume that you're correct, and the info in the program
booklet is therefore wrong. Right? We never hear from Ward here on
this newsgroup; perhaps Mark O-T or someone else can confirm.
AFT
A Fanatical Toscaniniphile
>
>Jack Pfeiffer's "production" of the Toscanini Collection often left out
>information, or wrong attributions.
>Don Drewecki
><dre...@rpi.edu>
Don, I am sure you are precisely right about this, though I had no
idea whom to attribute the problems to: for example, Vol. 37 states
that the recording used of the Debussy Fetes is from 27 March 1948 in
Studio 8H...well, the recording is absolutely continuous from the
precending "Nuages" of Carnegie in '52, agrees in timing with my video
copy of the '52 telecast, agrees precisely with the M&A transfer
attributed to '52; furthermore, it sounds entirely different from a
collector's tape I heard several years ago of the '48 broadcast. But
when the set was reviewed in the "Gramophone" the critic dutifully
noted the date of "1948" since he had only the liner note to go by.
I hope that these little anomalies will be corrected in the next
go-round, but I doubt it...betcha they just introduce some new ones!
AFT
Brendan R. Wehrung
Don Drewecki (dre...@rpi.edu) writes:
> Jack Pfeiffer's "production" of the Toscanini Collection often left out
When are we going to get some more Beecham on Masterworks Heritage?
Brendan
Rtsina
>go-round, but I doubt it...betcha they just introduce some new ones!
>
>
I'm with you. Case in point: Toscanini's 1953 Beethoven "Egmont" overture,
recorded commercially.
It first appeared on 45 RPM ERA-249 derived obviously from a rejected tape. In
the coda Harry Glantz produces a terrible clam.
This was corrected in the version that appears on the LP "Toscanini Plays Your
Favorites."
The CD reissue in the Toscanini series has yet another take, this time with no
clam but a noticeably slower coda.
The saga continues...
DT
Matthew B. Tepper
ck...@FreeNet.Carleton.CA pondered what I'm pondering as follows:
>
>When are we going to get some more Beecham on Masterworks Heritage?
Like the Berlioz Te Deum?
Don Drewecki
Toscanini approved the 1952 Nuages and the 1948 Fetes for
commercial release on LP, and that's how it was issued on the CD.
He never approved any videos, obviously, so the 1952 Fetes was
issued without his consent, but with the approval of the Toscanini
Estate, so BMG could issue it legitimately as a home video.
That is the distinction: Even if AT himself rejected something,
his Estate (which I presume is now just Walfredo or his agent)
CAN grant approval for BMG to issue those titles legitimately.
That is why we can still hope that Walfredo does grant clearance
to some or much of this stuff. Obviously, he must have had some
input into the live Philharmonic stuff from 1935-36 that was
issued in the boxed set, and that's a good sign.
--
Don Drewecki
<dre...@rpi.edu>
A Fanatical Toscaniniphile
>Toscanini approved the 1952 Nuages and the 1948 Fetes for
>commercial release on LP, and that's how it was issued on the CD.
When I bought the video, I made an audio cassette copy to play for
myself in the car: according to the timer in my Sony cassette deck, it
played for an elapsed time that was within about 2 sec. of the BMG CD
of the Fetes, or around 5:30 +/-2 seconds. The BMG disk lists tr. 9
(Fetes) as 5:30; on the CD player it times in a 5:33, but measured
from the first note to the last, it clocks at just about 5:30:30 as
close to the second as I could measure on a stopwatch after about 3
tries.
The M&A transfer, which has been "Cross-ized" with LOTS of echoes, is
pitched at the opening bars exactly in agreement with the BMG: I
compared by playing the two disks simultaneously on two players at the
same apparent volume, and switched instantly between them. From the
first to the very last note, tr. 3 of the M&A set on CD898 times out
nearly the same as the BMG; each time I did this comparison, I came up
with a fraction of a second different timing, so my several counts of
both transfers ranged from 5:30 to 5:31, worst cases. So the
performances are within 1 second, over the entirety of the Fetes (but,
as I found, they will differ more at specific points, which COULD be
indicative of performance variations, or MIGHT be evidence of unsteady
pitch of at least one of the transfers of the same reading.)
In BOTH recordings, there is a little "audience" or "musician's
squeak" JUST before the first note of the Fetes. It is perfectly
clear on the transparent BMG; however, it is somewhat muted on the M&A
but it seems to be at the same place timewise.
I wired my two Sony players as follows: I connected the left channel
of one to the L channel of my preamp, and the left channel of the
other to the R channel of my preamp, and donned 'phones. I set the
volume control on the adjustable player so that it gave an apparent
level match to the other machine that had a fixed output. I put the
M&A disk in the first machine, and the BMG disk in the second.
Carefully, I cued the BMG disk to the exact moment of the beginning of
Fetes, and then repeated the process with the M&A disk. I could hear
one out of one side of my 'phones, and the other one in the other ear.
After having cued them to the very instant of the first actual note of
the music, I starting both players at PRECISELY the same moment; I was
able to sync them after a couple of tries. As you can imagine, due to
the phase shifts of the different pickups, I could hear for the first
few seconds of play a kind of strange "stereo" effect; every note was
nearly in sync, but by the cymbal crash at around 0:42 elapsed time,
the BMG had moved just slightly ahead of the other by about 1 second.
By around 02:00 elapsed time there was a 1.5 second delay in the M&A
compared to the BMG; the horn calls at 03:03 were off by that much.
By around 03:45 there was a disparity of nearly 2.0 seconds; by 04:20
it was more than 2 seconds; BUT the oboe solos around 05:00 seemed to
sound just a bit flat on the M&A compared to the sharper pitch of the
BMG. At the end of the playback, there was a slight disparity of
about 1 to 1.5 seconds.
I compared the cymbal crashes, hall tone (affected by the M&A by the
added echo), the instrumental balances, and -- allowing for the
difference in general sound, the M&A being duller -- there really is a
remarkable agreement in sound quality, instrumental pickup, and so
forth.
I THINK that the time disparity is due to somewhat long-term pitch
shifts in the M&A transfer. Doing the test over again, I stopped the
M&A with the pause control to re-sync it at several spots: once again
I could obtain for periods of 15-30 seconds elapsed time an
astonishing synchronization, when the phase of the music is nearly
equal, and the note values line up vertically, until inevitably they
drift off...
I did such a test many years ago, using the Olympic transfer of the
Leo Overture #3 and the Victrola LP copy of the same b'cast, and I
believe that I derived actual stereo for periods of time: it was my
guess that the two recordings were taken from different mikes that fed
the SW and the domestic feeds, and they happened to be displaced at
least enough to create a binaural effect. However, the old 78 or 33
acetate disks did not maintain as constant a pitch as the new CD
transfers I tested today.
So, we have a peculiar situation. There was less than a 2 second
difference from first to last note in the playback of the two
recordings: one stated to be the telecast of '52 (M&A) and one having
been the "1948" aircheck.
When Toscanini approached a work after the lapse of several years, he
seldom recreated his interpretation within an elapsed time of 1 to 2
seconds!
Going back to normal CD player hookup, I auditioned the BMG CD of the
Nuages & Fetes with extreme care and attention. I think you will
agree that Nuages from '52 is indeed a magnetic tape source: it sounds
spectacularly transparent! The miking is just about as close and
clinical as it was for the 8H broadcasts, though it was done in
Carnegie. The string partials are as clear and pure and extended in
the high frequency spectrum as a modern digital recording (and the HF
clarity is better than the '53 Eroica.)
At the end of the Nuages, there is hall tone, musicians' or audience
members' fidgeting, and then after a short pause the Fetes begins.
There is continuous tape hiss audible at a "bias" of perhaps -50; at
no point during the interregnum between the N. or the F. music does
this change in ANY way. The hall presence does not change. The HF
response does not change. There is no increase in surface noise
during the Fetes. The highs are still PRECISELY as creamily, eerily
realistic. During this faster second Nocturne, the orchestral balance
and instrumental perspective is precisely as it was in the Nuages.
The dynamic range (and lack of compression) does not change. The
level of quiet passages is about the same, though of course Fetes
becomes much louder in the peaks.
If you play the BMG CD of the two Nocturnes, even at excruciating
volume, it does NOT reveal the slightest bit of sonic difference in
comparing the two pieces.
The cymbal crashes (the muted one at 03:15 and the full-bodied one at
03:34) sound EXACTLY like the ones in the M&A transfer, excepting the
difference in general sonic quality of the latter transfer (I have
played the cymbals in orchestra and band performances, though that
happened many decades ago, so I have at least some kind of an ear for
such details.)
If someone who has the '48 broadcast can carefully time the Fetes with
a stopwatch, we can have the comparative numbers. But no '48
recording in 8H (and I use the Japanese Victor LP or the RCA-BMG
transfer of the March '48 Otello Ballabile as my benchmark) sounds as
clean and transparent as the '52 BMG Nuages. Not even Cedar could
remove clicks to the perfection of what I hear on the BMG "1948"
Fetes; was NBC using magnetic tape by March 27, 1948?
Incidentally, in 1940, Maestro played the Fetes in 5:39 to 5:40,
timings I derived from test playbacks of the Naxos disk 8.110811-2.
He played the Nuages in '40 in 6:28; in 1952 according to my timings
of the BMG, he took 5:27, shaving off a whole MINUTE of the time in
1940! Extraordinary!!
I carefully compared the pitch of these two transfers, and they
perfectly agree. You can hear the Maestro's incredible change in
interpretation from the very first wind passages: in '40 he conducts
slowly, with inflection, and with a sort of hushed, rapt expression
from the players; in '52 the wind solos emerge like a flowing stream
that has always been meandering along under the listener's feet: I
struggle for a metaphor, but that's as good as I can do at the moment!
So, in 1940 he has a very different conception of the Nocturnes in
comparison to his '52 telecast; amazingly different in the Nuages. It
is anomalous that the "1948" and "1952" tracks on the respective BMG
and M&A transfers are so very, very close. Hmmm....
At some future time, I will conduct an experiment like this,
synchronizing the BMG edition of the Leo #1 with the new Naxos
transfer with the dual players in "alternating stereo" hookup, to see
if in syncing these agreed-upon identical performances taken from
different transfers, we get the same slight disparities along the
vertical score at diverse points.
AFT
Matthew B. Tepper
>
> That is the distinction: Even if AT himself rejected something, his
> Estate (which I presume is now just Walfredo or his agent) CAN grant
> approval for BMG to issue those titles legitimately. That is why we
> can still hope that Walfredo does grant clearance to some or much of
> this stuff. Obviously, he must have had some input into the live
> Philharmonic stuff from 1935-36 that was issued in the boxed set, and
> that's a good sign.
Not to disagree with your conclusion, but Wanda was still alive at the
time the New York Philharmonic box was being assembled.
Brendan R. Wehrung
Hey, the start of a new discography, the "Toscanini alternate issues"
site. This could be almost as thrilling as Furtwaengler!
Brendan
Don Drewecki
The attribution for the BMG CD is correct -- Toscanini's Nuages
comes from the simulcast of March 15, 1952. Then, listen
carefully and you hear the woodwinds pushed further back, and the
acoustic flattens out more for the Fetes, which is indeed from
March 27, 1948, in 8H. This kind of music generally did not work
well in 8H, in my opinion, and the sound is just too dry and flat.
The question then becomes, when BMG issued the home video version
of Toscanini's March 15th concert, did they sync the video track
to the 1952 audio track, or from 1948? That's a question I can't
answer because that is one of two AT videos I never purchased. I
will have to now, I guess.
--
Don Drewecki
<dre...@rpi.edu>
Don Drewecki
Maybe Wanda herself allowed the release of the Philharmonic stuff,
perhaps because Heifetz was involved in the Brahms concerto. I
don't know.
--
Don Drewecki
<dre...@rpi.edu>
A Fanatical Toscaniniphile
surmise from liner notes or even the (incorrectly judged) sound of the
recordings, and I now fully agree that the Fetes is from '48.
At Don's suggestion, I carefully audited the differences in the last
note of each copy: the "1948" edition on BMG, and the "1952" telecast
tape on M&A.
In '48, on the authorized release, there is a tiny pause of silence
after the very last soft note, and then the musicians relax and seem
to stretch a bit; this is followed after a brief second by a "pot
down" to utter blackness.
On the M&A, at the termination of the last note, there is almost
instantly a perfectly-timed clangourous clapping.
The little squeak and the between-movement noises from the '52
recording have apparently been edited right on to the very first note
of the '48 Fetes.
I played this several times, on speakers and 'phones, and honestly
felt that it was continuous in sound quality from the '52.
AFTER I had compared the last notes, and re-reading Don's description
of the woodwind balances in 8H vs. Carnegie, I repeated my playback of
the Fetes: NOW I could hear what Don hears: it is indeed the '48 and
not the '52 on the videotape!!!
It fooled me...Seth & Adrian & Arthur Fierro did a great job of
editing these approved takes together and matching them. Once more we
have evidence of FABULOUS recording quality at 8H for the pre-tape
era, if we can only hear the original masters (as we do in this BMG
CD!)
It is not so much that I was wrong in describing what I heard, but
that in the tests I just reported this afternoon, I did not hear ALL
of the details that Don has helped me perceive.
I wonder why old AT did not approve the '52 Fetes? It surely would
have saved a lot of work for RCA, BMG, and Seth & Adrian! Since the
two readings differ only by about a total of 1 to 1.5 seconds, it was
not a matter of tempo. My mystification just proves that I neither
have Toscanini's ear, nor that of Don Drewecki!
AFT
Rtsina
>site. This could be almost as thrilling as Furtwaengler!
You sound as if you could use a thrill.
DT