Posted on 06/07/2004 12:26:29 PM PDT by weegee
Week of May 29, 2004; Vol. 165, No. 22 , p. 339
Groovy Pictures: Extracting sound from images of old audio recordings
Peter Weiss
Songs and words preserved on antique vinyl records and wax cylinders become more precious with each passing day. They also grow increasingly fragile and are especially vulnerable to damage if played.
Now, researchers using optical-scanning equipment have made exquisitely detailed maps of the grooves of such recordings. By simulating how a stylus moves along those contours, the team has reproduced the encoded sounds with high fidelity.
Libraries with collections of old recordings "don't want to queue up an antique piece of material every time you want to hear it," notes particle physicist Carl H. Haber of Lawrence Berkeley (Calif.) Laboratory, codeveloper of the new scanning approach. Instead, those institutions seek to extract sound from delicate recordings and preserve it electronically. In that form, it can be played back repeatedly without harming the original and also made available on the Internet.
A few years ago, Haber heard on the radio that archivists needed ways to nondestructively extract sound from old recordings. He and his Berkeley lab colleague Vitaliy A. Fadeyev, who make arrays of sensors for tracking minute particles in powerful accelerators, realized that their own work was relevant. To align their arrays, they scan sensor surfaces by using a microscope with submicrometer resolution. After hearing of the archivists' problem, "we thought, 'Wow! Why don't they just do it optically?'" Haber recalls.
The scientists used their microscope to make a two-dimensional map of the grooves on a 78-revolutions-per-minute shellac disc of a circa 1950 recording of "Goodnight Irene." They also wrote software that calculates the velocity with which a stylus would move in the mapped grooves. A sound clip from the virtual disc sounded better than the same section played back from the original disc with a stylus did, Haber and Fadeyev reported in the December 2003 Journal of the Audio Engineering Society.
Recently, the Berkeley physicists, along with other U.S. and British researchers, showed that the approach also worked well on a 1909 wax-cylinder recording. Because wax cylinders store sound as up-and-down undulations of the groove, rather than as side-to-side ones, as discs do, the team turned to three-dimensional scanning using an instrument known as a confocal microscope.
Haber described their work on May 25 in New York City at a meeting of the Acoustical Society of America.
The new mapping technique could enable archivists to retrieve recordings from damaged or broken records and discs, comments Peter G. Alyea of the Library of Congress in Washington, D.C., which is funding some of Haber and Fadeyev's work. The method could also solve the looming problem of playing archived recordings after old types of playback equipment become unavailable, he adds.
Using existing equipment, the scientists would require an entire day to scan one full wax cylinder, they estimate. One of their next tasks is to find ways to speed the process.
Vinyl PING
This past weekend I transferred some of those cassettes to CD-R. Eh, it's better than nothing.
A friend of mine has a Brunswick player that can play all 3 types of flat platters (Edison, Pathe, and "modern").
Then there are 7" 45s, 10" eps (at 33RPM and 45RPM), 16RPM records (and even 8RPM!).
There isn't a "one stop" solution to playing old records. Besides theirs will permit them to rescue cracked/broken records. Even accetates can only be played a finite number of times before they are destroyed.
The player you cite does look good though (7"-12", 30RPM-50RPM and 60RPM-90RPM). Multiple speeds of "78" are more important than holding the range between 30-50, 78s actually are recorded between 70RPM and 80RPM depending on the manufacturer and era:
The LT-1LA steps through its 30 rpm to 50 rpm range at 0.1 rpm for LP's and 45's. The LT-XA has a range of 60 rpm to 90rpm at increments of 0.2 rpm and can play 78's in addition.
Now that we are in the digital realm, we have 3 formats of CD/DVD audio that are not compatible (although combi-players exist). Add to that CD-R, and CD-RW. My 1986 laserdisc player is pretty good at being able to play CDs, the music tracks on CDs with CD-ROM data (like videos) which give my car CD player a lot of headaches, and even CD-Rs. It is a better "CD" player than most CD players these days.
I never knew there were three types of cylinders. Learned something new.
I've got a Sony CD player made back in '89 that plays homemade discs just fine, so I've always found it weird about how some have problems with the latest players. But it's amazing how we can burn them to begin with.
I'd have loved to had that laser player several years back. I too have gotten rid of my collection, such as it was, nothing the Smithsonian will miss.
I seem to have read once that Baird made some attempts to record his TV signals on discs back then. Maybe this program mentioned in the article could reveal another piece of history.
John Logie Baird recorded TV on discs back in the '20s.
http://www.tvdawn.com/tv1strx.htm#RWT115
Radiovision was his term for what we call "television". Television was his term for what we call "cable television".
Radiovision was a broadcast signal. Television was a cable signal. He could not lay out sufficient high grade wire to implement "television".
Radiomovies was his term for closed audience "broadcasts" (at say a theater). This was "pay per view" broadcasts of major sporting events (like boxing). As movies go digital, they will be beamed by satallite to theaters to reduce shipping costs that much more.
Jenkins studied the flapping of a flag on a flagpole and came up with a patent for movie projectors (pre-1900) that kept the film from flapping as it went past the lens. Thomas Edison bought that patent from Jenkins' company and used it to shut down a lot of other projector manufacturers.
There are a lot of inventors who get overlooked. I'll have to look at that website sometime. Looks interesting. I just hope that it doesn't spin the hyperbole of "first" too much.
I think that one type is made of wax (very fragile) while the other is bakelite (blue plastic).
Even "records" are made of different materials and different things are used to clean them (what is good for one isn't for another, eg. water is bad for another while alcohol is bad for a different type).
There will always be some laserdisc that just cannot be had on DVD or some later format. There are commentary tracks that will forever be "out of print" (like on the first edition Criterion James Bond discs); the studio was not happy about some of the comments. There are prints like the original theatrical release of Blade Runner (or even the Star Wars trilogy).
Then there are small edition titles like one I have of Space Shuttle missions from the early 1980s. It has day launches, night launches, night landings, different mission tasks, 16mm film of the cockpit activity during a launch. I found that for $7 at a used bookstore near Johnson Space Center in "Houston". It was a marketed item but not by NASA or even any sizeable company. It doesn't have all of the studio music or naration that overproduced documentaries do.
The question is how much will it cost to reconstruct some of this information in a laboratory setting?
The Zapruder film of the JFK assassination has been optically scanned. Since the film was shot 8mm film, there is no negative. The film had additional image data around the sprocket holes that no one would ever "see". There is a video release of that footage with the additional data shown (and in several ways).
Also, when it comes to "preserving" data, the historical measure has been for recorded tape. It is the standard for archivists. Digital recordings are not analog. Also they could have dropout within a matter of years (how good are the copies?).
The costly solution (which I have not heard discussed) is optical soundtrack on filmstock. They were actually recording some albums this way in the 1960s.
Too much of the culture has been deemed "disposable", with it history is being discarded.
Philo Farnsworth didn't get the credit he deserved for all electronic TV either, thanks to Sarnoff.
Wasn't there a man in Kentucky who supposedly made the first real radio transmission? (my memory isn't very sharp today, and it's been hard lately for my brain to shift gears into tech mode). I swear, so much of history of electronics is still tainted by the RCA revisionists of the time.
Albums recorded in something akin to Cinemascope? Never thought of it, though back then it likely would seem a "sound" enough idea. (I'm probably wrong. Cinemascope and Cinerama was even then magnetic recording, can't remember for sure).
BUMP!
This reminds me of something I was curious about: to what extent does the bias signal on an analog tape get recorded. I know that if one plays a tape with a conventional head, the bias signal, even if present on the tape, will be significantly attenuated on playback, but if the bias signal is physically recorded on the tape, I would think that a digital readout of an analog tape could yield audio that was better than anything that could ever have been gotten off the tape by conventional means even when the tape was new.
I wonder if George Méliès knew about it when he produced his Long Distance Wireless Photogaraphy in 1908?
I'm familiar with that Méliès film (I have a laserdisc filled with his work). It's possible.
I see that I was wrong with regards to it being optical soundtrack (it is magnetic) but according to this description, it gave superior sound:
http://www.musicboxtheatre.com/70mm.html
70mm SOUND
Left - Left Extra - Center - Right Extra - Right - Surround.
Six discrete, magnetic tape, channels of sound. Two are on the outer edges of the film and one between the sprocket holes and the picture on both side of the film stock (the black vertical lines shown above). Until recent advances in Digital sound reproduction, 70mm magnetic tape stereo was the apex of state-of-the-art audio reproduction.
The Music Box utilizes new state-of-the-art JBL theater speakers behind the screen for Left, Left Extra, Center, Right Extra and Right sound channels. 6 Altec Lansing / Seaburg presentation speakers are installed for surround, plus four 16" Yamaha Sub Woofers for Dolby, Dolby SR and DTS sub-bass channels; powered by 1000 watt Crown power amplifiers.
Not all 70mm film presentations were photographed using 70mm cameras. It was not unusual for studios to photograph movies in less expensive 35mm format and process them in 70mm just to utilized the marvelous sound reproduction of 70mm. Although the quality of picture clarity suffered slighty in the transfer from the smaller frame to the larger (enlarge a picture on your computer screen by 3x) the difference in sound presentation more than made up for the minor disparity of image. New digital sound process allows full frame 70mm film to be remastered in DTS digital sound. The Music Box Theatre has been priviledged to present new 70mm prints of Dr. Zhivago, Patton and now Lawrence of Arabia in 70mm and DTS digital sound reprocessed from the original magnetic sound track.
Not all films photographed in 70mm were seen in 70mm. Outside major market areas most films shot in 70mm were presented in 35mm Cinemascope format. Very few theaters outside of city centers had 70mm projectors. Today's re-release of 70mm restorations are even rarer "special event presentations" since even fewer major market area cinemas have 70mm projection capabilities.
70mm film projection begins with a large, full frame cell to achieve the projected image. In surface area, this 70mm cell is about 3 times bigger than a standard 35mm cell. The larger film cell allows for a brighter picture (movies are light projected through film and reflected off the movie screen) and sharper focus, corner to corner, with no distortion of image as found in anamorphic presentations (i.e.: Cinemascope).
Disclaimer: Opinions posted on Free Republic are those of the individual posters and do not necessarily represent the opinion of Free Republic or its management. All materials posted herein are protected by copyright law and the exemption for fair use of copyrighted works.