Perspecta Sound
Scott Marshall
<dsc...@dbworld.net.au> writes:
>My Perspecta references are not ready to hand, but... I do think that
>compression and expansion is quite possible with the system.
All my references suggest that dynamic range compression and expansion is a
good description of an aspect of Perspeca sound.
There is certainly expansion during playback. When the audio is soft, the
control tones are at a low level and the gain is low. When the control tones
are at a high level and in synchrony with a loud audio signal, the gain is
raised and the resulting volume is raised, masking the increase in noise just
like any other compression/expansion noise reduction system.
Schematics of Perspecta encoders have not turned up, so I have not been able to
prove beyond reasonable doubt the use of compression during recording. Critical
study of the Perspect tracks that I've done and my studies of sonograms made by
others convinced me that the Perspecta encoder did use compression.
An excellent reference of Perspecta is in the SMPTE Jurnal April 1955 Volume 64
"Advantages, Scope and Limitations of the Perspecta Stereophonic System" by
Norman H. Crowhurst, formerly at Fairchild, purported inventors of Perspecta.
<< Perspecta was supposed to be a
single inventory product, like Dolby SVA, and it should sound reasonable on non
Perspecta systems. Now, if there had been considerable compression involved in
a Perspecta mix, playback of such a mix would sound lacking in a regular
theatre. Consequently, mono computability might limit the extent to which this
aspect of the system was utilised >>
I am told that hundreds of movies were made with Perspecta tracks in the 1950s,
and it is typical that the only remaining copies of many of these films have
Perspecta encoding in their tracks, ready to be decoded, and no-one is the
wiser, so it may indeed be comparable to playing Dolby Stereo in mono resulting
in acceptable reproduction without decompression.
Scott Marshall
Editor, Wide Gauge Film and Video
http://members.aol.com/widegauge/
PeterH5322
>>
I am told that hundreds of movies were made with Perspecta tracks in the
1950s,
and it is typical that the only remaining copies of many of these films have
Perspecta encoding in their tracks, ready to be decoded, and no-one is the
wiser, so it may indeed be comparable to playing Dolby Stereo in mono resulting
in acceptable reproduction without decompression.
>>
Or, perhaps there was no compression in the first place.
The Bruin in estwood (L.A.) had no Perspecta, yet it often played the same
features as the Village just across the street, which did have Perspecta,
according to a recent account on this NG.
No one seemed to complain in either theater.
David Scanes
> An excellent reference of Perspecta is in the SMPTE Jurnal April 1955 Volume 64
> "Advantages, Scope and Limitations of the Perspecta Stereophonic System" by
> Norman H. Crowhurst, formerly at Fairchild, purported inventors of Perspecta.
Really good article, this one. Started to build a decoder based upon the specs in
this back around 85 when I should have been studying for uni maths exams.
Not that many pages, but just about everything you wanted to know. Time constants
for the gain control circuits were about the only thing missing (from memory.)
Unfortunately I only got as far as building the centre filter and the auto
detector. (With filter switching for 24fps/25fps (being from a 50Hz country.))
The DBX2150's I wanted to use (right control law) were too expensive / hard for me
to procure at the time. Did once play a Tom and Jerry cartoon from free to air TV
into it and got the auto detector to light up though!
David Scanes (Wishing he had the time to build the other two filters and wire up
the 2150's now sitting on the shelf.) :-(
JohnW248
(Scott Marshall) writes:
>I am told that hundreds of movies were made with Perspecta tracks in the
>1950s,
>and it is typical that the only remaining copies of many of these films have
>Perspecta encoding in their tracks, ready to be decoded
Chance Productions has built a card for this (slides into the Dolby
Unit)--don't know if it's offered for sale, they sent one over to the Gene
Autry Theatre for an SMPTE meeting where we were going to see "Silk Stockings"
until they ran the masking closed on the screen--and they also decode these
tracks as part of their "restoration" process.
John
Scott Marshall
(JohnW248) writes:
>Interesting theory [of preparing Persecta tracks], but not the method
described by Loren Ryder.
Have you got any references for this? How did Ryder describe it?
The Perspecta track I studied most thoroughly was Forbidden Planet (1956, MGM
CinemaScope). I've been told that there were 4-track mag prints of this title.
If so, the Perspecta mix could, in theory, have been derived from that master.
PeterH5322
>>
The Perspecta track I studied most thoroughly was Forbidden Planet (1956, MGM
CinemaScope). I've been told that there were 4-track mag prints of this title.
>>
The 4-track version is the only one I've seen.
L.A. was not exactly a "stronghold" for Perspecta installations.
Perspecta was an East-coast development, and CinemaScope/Ampex 4-track mag was
a West-coast development, ergo most L.A. theaters had Ampex 4-track mag
installations (or Philips/Ampex AA-II installations).
If The Village theater received a mag-optical print from MGM, they would more
than likely play it magnetic and not optical/Perspecta.
The Village was a premier CinemaScope house in L.A. (better IMO than The
Chinese). And it was one of the few Fox West Coast Theatres installations to
have surrounds (most had just the three screen channels).
The Pickwood, the only other L.A. installation yet identified as having
Perspecta, also had surrounds. The Pickwood was one huge f*ckin' theater.
gordon mcleod
compression present when played back and toggling between prespecta and non
on the motiograph unit I still have
gordon mcleod
On 27 Jul 1999 14:20:51 GMT, Scott Marshall wrote:
>In article <379DB24D...@dbworld.net.au>, David Scanes
><dsc...@dbworld.net.au> writes:
>
>>My Perspecta references are not ready to hand, but... I do think that
>>compression and expansion is quite possible with the system.
>
>All my references suggest that dynamic range compression and expansion is a
>good description of an aspect of Perspeca sound.
>
>There is certainly expansion during playback. When the audio is soft, the
>control tones are at a low level and the gain is low. When the control tones
>are at a high level and in synchrony with a loud audio signal, the gain is
>raised and the resulting volume is raised, masking the increase in noise just
>like any other compression/expansion noise reduction system.
>
>Schematics of Perspecta encoders have not turned up, so I have not been able to
>prove beyond reasonable doubt the use of compression during recording. Critical
>study of the Perspect tracks that I've done and my studies of sonograms made by
>others convinced me that the Perspecta encoder did use compression.
>
>An excellent reference of Perspecta is in the SMPTE Jurnal April 1955 Volume 64
>"Advantages, Scope and Limitations of the Perspecta Stereophonic System" by
>Norman H. Crowhurst, formerly at Fairchild, purported inventors of Perspecta.
>
><< Perspecta was supposed to be a
>single inventory product, like Dolby SVA, and it should sound reasonable on non
>Perspecta systems. Now, if there had been considerable compression involved in
>a Perspecta mix, playback of such a mix would sound lacking in a regular
>theatre. Consequently, mono computability might limit the extent to which this
>aspect of the system was utilised >>
>
>I am told that hundreds of movies were made with Perspecta tracks in the 1950s,
>and it is typical that the only remaining copies of many of these films have
>Perspecta encoding in their tracks, ready to be decoded, and no-one is the
>wiser, so it may indeed be comparable to playing Dolby Stereo in mono resulting
>in acceptable reproduction without decompression.
>
Scott Marshall
peter...@aol.com.org.gov (PeterH5322) writes:
>The 4-track version [of Forbidden Planet] is the only one [version] I've seen.
Then it is conceivable that the 4-track mix was made first, then the Perspecta
mix was derived from that via a simple, hands-free automated process. My
critical listening of the "Forbidden Planet" Perspecta track (in a large
theater and later with headphones) suggests that it had been made in precisely
this way. The channeling changes were far too rapid and precise to have been
done by hand.
There certainly could have been more than one way to mix a Perspecta track. I
understand GWTW was Perspecta-ized once. Mixing an existing mono track to
Perspecta may have used a completely different technique than from scratch or
from a 4-track CinemaScope master.
Martin Hart
wide...@aol.com says...
>
> There certainly could have been more than one way to mix a Perspecta track. I
> understand GWTW was Perspecta-ized once. Mixing an existing mono track to
> Perspecta may have used a completely different technique than from scratch or
> from a 4-track CinemaScope master.
The MGM Perspecta mix for GWTW was made from the individual music,
dialog, and effects sound stems provided by D.O.S. for that purpose.
Marty
--
Relive the panoramic past:
Visit The American WideScreen Museum
http://www.simplecom.net/widefilm/
Scott Marshall
m.b.har...@worldnet.att.net (Martin Hart) writes:
>The MGM Perspecta mix for GWTW was made from the individual music,
>dialog, and effects sound stems provided by D.O.S. for that purpose.
Now, that's interesting. Others have assumed that Perspecta steering was done
after a mono mix. Since they already had a mono mix from the original release,
why did they remix from the stems for Perspecta? Perhaps they DID start by
mixing the stems to 3-track disctrete that was then fed into the Perspecta
encoder.
Do you know any more about this mix, Marty, or have iot documented on your
site?
PeterH5322
>>
>The MGM Perspecta mix for GWTW was made
>from the individual music, dialog, and effects
>sound stems provided by D.O.S. for that purpose.
>>
One report has Selznick providing only the composited negatives and the sound
track negative to MGM in compliance with his contract with Loews.
IOW, he provided the absolute minimum necessary for compliance.
Apparently that's not true.
Martin Hart
peter...@aol.com.org.gov says...
Yes, it is true. Selznick sold MGM exactly what they paid for, a useable
cut negative and a master sound mix. And that's all. Everything else was
retained by Selznick.
When MGM began work on the '54 re-release of the film, Selznick got
himself involved because he was concerned that his greatest project might
not fare too well in the conversion to 1.66:1 widescreen and the
channelling of the sound for Perspecta. That is when he turned over all
the sound stems to MGM, at no charge.
Martin Hart
wide...@aol.com says...
>
> Now, that's interesting. Others have assumed that Perspecta steering was done
> after a mono mix. Since they already had a mono mix from the original release,
> why did they remix from the stems for Perspecta? Perhaps they DID start by
> mixing the stems to 3-track disctrete that was then fed into the Perspecta
> encoder.
Perspecta steering would be done after a mono mix, of sorts. Doing a
Perspecta conversion of an existing mono soundtrack would be disastrous.
Lots of considerations have to be made when cueing the sound movement.
You might, for instance, want to place a sound effect on the right
channel but you can't because the music is also blaring and it would be
obvious if it suddenly moved. However if you have separate dialog, sound
effects, and music tracks, you can tone down one part so that it is not
noticed that it has moved briefly. You just have a lot more control over
how things will sound. Doubtless the three GWTW sound tracks were
transferred to a three track mag full coat and the Perspecta was mixed
while the levels of the individual stems were adjusted to keep things
from sounding too lousy. In our discussion on SVA recently, Ray mentioned
that a similar practice was common when dialog had to appear at the same
time that a loud sound effect was used in the left or right channels, you
just have to work around the dialog and make sure that the steering logic
doesn't suck an actors voice into the wings.
> Do you know any more about this mix, Marty, or have iot documented on your
> site?
The information, as I included it in a follow up to Peter's post, is
widely documented. Information on it is available in Ronald Haver's
exhaustive Selznick books and in many other sources.