From: Mike Gray <[log in to unmask]>
> Michael makes a good point ref. time alignment vs. electrical phasing.
> My point is that unless one is using a sync track or mechanical linkage,
> two separate turntables not have enough linear similarities (unlike two
> pressings made from the same metal part, of course!) to usefully create
It is easily possible to tell the difference between identical masters
or not, to see if it is worth a try. Some masters will, even if just for
a few seconds, easily sync and cancel when combined phase reversed, and
others will be impossible to cancel out anywhere in the recording.
Listening to the Ellington Program Transcriptions, transferred the first
time by Brad Kay and Steve Lasker for the Everybodys LP and again by
Seth Winner for BMG CDs, they sound pretty convincing. Brad and Steve
had to resync 4 or 5 times per side, but Seth reports that much to his
surprise, one side maintained sync from start to finish and the other
needed only one resync. The results are released and available to be
I've ordered the CALA CD and hope to be able to report on it soon. But
even listening to the on-line samples, they sound VERY convincing on
speakers, especially at :39 in the Beethoven Sym 7. And with headphones
that have a mono-stereo switch, the difference is striking!! All of you
can check them for yourselves:
I just had Leah dub the samples onto Audacity and you can see the
differences in the instrument pick-ups in the waveforms and the
metering. In mono I don't hear phasing problems, and the notes do not
have dual-attacks. I haven't yet figured a way to phase-reverse and
combine here. I'm not in my studio so I am working on computers, not
> In addition to having Tony Griffith's
> testimony, I also have copies of HMV recording sheets (the earliest
> Abbey Road from 1937, the earliest French from 1934), including sessions
> at which 1 and 1-A machines were in use. Neither the electrical data,
> cutting data nor the microphones in use substantiate the idea that
> unintentional stereo was intended or created.
I should have made myself a little clearer that 99% of sessions using
dual-mastering have both recorders fed by the identical signal. That
there are a few -- very few -- sessions done differently is the real
point of interest. And from what I remember, they dated from before
your 34 and 37 dates, back into the 20s before Abbey Road, but even into
the early Abbey Road years.
> Finally, EMI didn't keep the staggered-head tape machine, so they
> had to use digital delay. Luckily, that machine was soon replaced
> with standard in-line head machines.
It would not have cost all that much to re-do a head block on one of the
fancy machines they still had. But heaven forbid they lower themselves
to try playing the tapes on an Otari!
> Besides, editing the staggered-head tapes was almost impossible.
> Mike G.
Edit after transferring! But yes, that is a major fault of the system,
much like SECAM TV which can't be edited or switched because there is no
such thing as a "frame" because every frame depends on the one before
and the one after it. (Their studios were PAL!)
Mike Biel [log in to unmask]
Michael Biel wrote:
> From: Mike Gray <[log in to unmask]>
>> Actually, I have heard the Cala disc through high quality headphones ...
>> I remain unconvinced that it's the real thing.
> Did you phase reverse one of the channels and combine them?
>> Separate tracks on mechanical carriers that contain 'identical' signal
>> content will be just different enough to create 'stereo' - read
>> out-of-phase - effects when they are aligned.
> There is a distinct difference between electrical phasing and time
> allignment phasing based on the mechanical wavelengths of the sound
> waves. If you play two identical records on a pair of tables they can
> be fused to create a perfect common signal or phase reversed to cancel
> 100%. If the cancellation cannot be achieved, there does have to be a
> difference. Differences in flatness of frequency curves -- such as in
> two different cutter heads in a 1 and 1-A pair of masters -- might
> change the levels at which any particular frequency will be heard in one
> recording vs. another, but if they are both fed from the same source
> they will cancel noticably. All wavelengths will be time-alligned even
> if there are slight level differences at different frequencies. If they
> were fed by different mics are at different locations from each other,
> they will not cancel. (They might cancel if the different mics were in
> a co-incidental array with each other, a distinct possibility in England
> because that was one of Blumlein's stereo arrays. Quite possibly his
> stereo tests recorded with this array might phase cancel despite us
> knowing that they ARE stereo.)
> Try talking into two combined phase-reversed microphones that are right
> together in the same vertical plane in front of you. Even though there
> are mechanical differences between any two microphones, the cancellation
> is almost complete, compromised only by frequency flatness differences.
> Then move ONE of the mics further away from you so that there is a time
> differential. Electrical equalization will not bring the mics back into
> cancellation because some frequencies will be hitting the mics sooner or
> later than the other at a different rate based on wavelength of the
> frequencies. I am not sure what will happen with time delay of one the
> channels -- you might be able to get back to cancellation if the mics
> are only a foot or two in front and back of each other -- but the
> further apart and aside the mics are from each other, the less likely
> the two can be recombined. Add in several voices or instruments coming
> from different locations, and the two mics can't be combined
> out-of-phase to cancel.
>> Even sources recorded in separate channels on the same carrier can
>> prove problematic: when EMI engineers tried using digital delay on
>> their earliest two-track staggered-head Stereosonic' tapes, they
>> discovered that they couldn't entirely overcome HF comb filtering.
> Why does everybody now seem to think that digital is the perfect answer
> to everything??? Did they try the "old fashioned" way where they could
> adjust mechanical azimuith on two separate heads?? As I have stated
> here before, the head separation on some Otari models between the record
> and reproduce heads is identical to the original staggered standard. I
> have a few of Robert Oakes Jordan's masters that sounded fine this way.
> There are also phase problems with the 1956 Fantasia stereo transfers
> because they were done thru phone lines from Burbank to Hollywood, but
> there is no doubt that they are rather effective stereo. It is why the
> DVDs offer a separate mono soundtrack because the stereo can't really be
> properly combined to mono.
>> As for undiscovered stereo, the late Tony Griffith told me without
>> qualification that there was no such thing - in other words, a fantasy.
> Yes, I have him on videotape saying that, but he also said that he based
> his claim on what he had been told years earlier by an oldtimer because
> he himself had not been there. At that point he had not done any
> personal listening or first hand research. I don't have my file where I
> am now, but a couple of books I was reading a few months ago gave some
> specific quotes from session sheets and memos from the time citing quite
> the opposite. I'll try to dig that up when I get back in a few weeks.
> Try an experiment with re-creating accidental stereo. Make two
> recordings on two separate machines from one mic feeding both. Then do
> it with a different mic in a different location feeding the second
> machine. You will be able to combine the two recording machines
> out-of-phase to cancel when they are fed by the same mic but you will
> not be able to do it for the recording with the two machines being fed
> by different mics in a different location.
> It is THIS test that can prove if any 1 and 1-A master are fed from the
> same or different mics. From what I see on the videotape, Griffith had
> not done these tests but that a very uneasy Gerald Plano sitting next to
> him had (on an oscilloscope, I think).
> Mike Biel [log in to unmask]
> Michael Biel wrote:
>> From: Mike Gray <[log in to unmask]>
>>> For "left" / "right" to produce 'binaural' would require two entirely
>>> independent microphone mixers feeding two separate disc-cutters.
>>> Anything else is a fantasy. Mike Gray
>> Since Stokowski had already participated in many stereophonic recording
>> sessions since the early 30s, there is a high likelyhood that these ARE
>> stereo recordings. The proof would be on Cala 551, of course. But also
>> consider, recording in multiple "angles" was common practice in several
>> Hollywood movie studios by the late 30s.
>> On another related note, while you have certainly far more experience
>> with the EMI archives than I, I have been reading recently that
>> paperwork there DO indicate some recordings made in what is now termed
>> "accidental stereo" despite their protestations to the contrary in the
>> Mike Biel [log in to unmask]
>> Dave Lewis wrote:
>>> Edward Johnson, in his notes for Cala 551, "Stokowski Beethoven Symphony
>>> No. 7 and Other First Stereo Releases on CD" states:
>>> "In 2004, Anthony Fountain, Classical Archivist at Sony Music Studios in
>>> New York, found many lacquer masters that Stokowski and the All American
>>> Youth Orchestra had recorded in Hollywood after their 1941 summer tour.
>>> The most significant part of the discovery was that all the recordings
>>> were made in duplicate, with each pair of discs labeled "Left" and
>>> "Right" respectively. [...] It was an exciting discovery and the Leopold
>>> Stokowski Society wished to license a complete CD of these AAYO
>>> 'binaural' recordings. However, the Sony powers-that-be decided that
>>> such a discovery should appear on their own label instead, along with
>>> any other records of the period that had been recorded binaurally. These
>>> included the Stravinsky/New York Philharmonic sets of the early 1940s in
>>> which the composer conducted his own 'Rite of Spring' and other works
>>> [...] However, it all came to nought in 2006 when the senior executives
>>> in charge were dismissed due to the poor sales of both their new and
>>> historic releases. The Stokowski/AAYO lacquers were sent off for storage
>>> and the transferring equipment dismantled, so it seems that the
>>> opportunity for hearing more of these historic recordings binaurally
>>> has, tragically, now gone."
>>> Okay - I'm assuming that these notes, published with the final Stokowski
>>> Society release that appeared in November, speak the truth. But just
>>> last week I heard a Sony producer protesting on NPR that "people should
>>> not take it on themselves and reissue classic recordings. First we have
>>> to locate the original master recording, then we have to find the legal
>>> holder of the performance rights, etc." The NPR commentator added that
>>> Sony has transferred about 10,000 classic recordings since 1994 or so,
>>> but is doing so in the face of the realization that only minimal
>>> financial gain is likely to be made in such endeavor.
>>> However, if they have "dismantled" [...] "the transferring equipment"
>>> then all that the Sony producer said is mere bluster; one has to assume
>>> that they aren't doing any of that kind of work now, based on what is
>>> said in the Stokowski notes. Who is telling the truth?
>>> David "Uncle Dave" Lewis Assistant Editor, Classical Rovi Corporation