Slashdot Mirror
5000 Cylinder Recordings Placed Online
Jon Noring writes "The Department of Special Collections at the University of California Santa Barbara (UCSB) Davidson Library recently placed online, with free access, over 5000 sound recordings as part of its Cylinder Preservation and Digitization Project. These recordings date from the 1890's to the 1920's, all transfered from Edison cylinders using state-of-the-art equipment. The restorations are first-class, using CEDAR tools. Besides MP3 and streaming audio, the raw transfers are also available for diy'ers to try their own hand at audio restoration. For those who like their music 'hot', there's not much there since most of the cylinders predate the start of the Jazz Era (ca. 1917), but there is some early 'mouldy fygge' dance-type jazz, like 1920's 'Peggy' by Lopez and Hamilton's Kings of Harmony Orchestra."
Seriously, though, I've always thought that was an interesting song. Remember that the Wright Brothers flew only in 1903, so the whole concept of "flying machines" was incredibly new and exciting. There's a certain innocent romance to the song that's so... impossible to recapture today.
Sometimes it's best to just let stupid people be stupid.
When I was in Niagara On the Lake, or Niagara Falls [I don't remember which], I was in a museum with an Edison player, and a wax cylinder mounted in it. They wouldn't take it off, for fear that it would fall apart, even though they had a few other ones. They started it up for me and my friend, and I kicked myself for not bringing my digital vid. camera with me to record it. The music was over 90 years old, and recorded live! Cool; all those people are dead.
Saskboy's blog is good. 9 out of 10 dentists agree.
Fortunately, the U.S. Government, via the National Park Service (I believe) are the owners of the Edison masters, and so the recordings appear to be public domain. Or at least the U.S. Government won't attempt to claim state-level copyright on the recordings (which I suppose they could.) Note that sound recordings made before 1972 are NOT covered by Federal Copyright Law, rather they are covered by a patchwork of state copyright laws (both common and statutory), anti-competitive laws, etc. It's a mess. Pre-1972 recordings (other than those whose ownership is lost) will not come under Federal Copyright protection (and thus, hopefully, public domain status) until 2067! There are some early Columbia cylinder recordings from 1890 (technically owned by Sony-BMG) which will not become public domain until 2067, a whopping 177 years after they were 'waxed'.
Marion's Attic radio show is a good source:
http://members.aol.com/marionweb/
The velocity of the "needle" across the surface is inherently constant with a cylinder. With a disk the RPMs are constant but it spirals in so you have to compensate for this frequency drift when recording. I wonder how well that worked? I've heard that when cylinders competed with disks they were regarded as having higher fidelity. The reason they failed is eerily similar to the beta vs. VHS debate: cylinders couldn't record as long. Also, if you do the math you find it's much harder to pack the same ammount of surface area into a box of cylinders than it is for disks. So cylinders were more expensive and could hold less music. The difference in quality wasn't enough to overcome that, and disks won.
For all intensive purposes, "whom" is no longer a word. That begs the question, "who cares"?
Even though they are considered "surprisingly good quality" by the standards of the time they are still horrible by any other standard today. True, many of these recordings are quite rare and there is only perhaps one copy known. However, I'd imagine that for many of the more popular things it should be somewhat easy to find more than one copy. Couldn't you just digitize more than one cylinder and compare the waveforms to remove anything that doesn't appear in both of the copies thereby removing a great majority of the hiss/pops? It would kind of be the audio counterpart to the optical technique of cosmic ray rejection.
- "Hear that?! The percolations are imminent! Cease your ingress!"
Since submitting this SlashDot item, I discovered in the cylinder collection an even jazzier recording (from 1924) that some may enjoy: "Why Did you Do It?" by the Georgia Melodians.
By the 1920's, Edison was mastering onto vertical cut disc masters (and issued as "Diamond Discs"), and then producing cylinder masters by dubbing master disc pressings. So the sound quality of the cylinders issued in the 1920's was lower than the comparable discs, such as the above recording. It should be noted that disc records pretty much took over the market by 1915, so by the end of World War I cylinders became almost like the "8-track" of its day. Edison still issued cylinders until it went under in 1929, but the 1920's cylinders are quite rare compared to the Edison discs (as a side note, in 1928 Edison released lateral cut records to play on regular phonographs, and they sounded quite good. Edison was also late to switch to electrical recording, strangely enough.)
That's a bit misleading. Nobody is massproducing cylinder readers today. I suppose in 50 years you could build a "compact disk" reader.
:-)
Though the longevity of the media is another issue. Keep in mind we are not comparing the same thing in that regard. The density of bits per inch [yes, you could say those analogue recordings are storing "information" and hence can be coded in bits] is FAR less than that of a CD or hard disk.
Let's see a 400GB cylinder made in 1890?
I'm sure it's possible to make a 1MB CD that is HIGHLY redundant [you could do this in software] and likely to last a long time even as pits and scratches form on the media. If you have 650 copies [or 700 or whatever] of the same megabyte on the disk the likelyhood of them all being unreadable [with specialized equipment] is fairly low.
Average 8-track [because I won't be bothered to figure this out for a cylinder but the same point applies] ran at most 60 minutes [or so]. With a dynamic range of about 60dB or so [10-bits] with a freq response bandwidth of about 12Khz [24,000 samples per second] that's 29.29KiB/sec [thereabouts] , say 60KiB/sec for stereo. That's 210.93MiB of storage, at ~3.5ips that's 13500 inches of tape in bands of 1/16th of an inch for a total of 843 in^2 of material or a density of 2,098,943 bits per inch.
A CD can store about 695MiB of data in far less area with a density of about 46 million pits per in^2.
So you're saying something that stores 23 times the information having a lower longevity is a negative quality? It's even worse for a cylinder where the ratio is likely over 100x in favour of CDs [let alone DVDs].
All I'm saying is you can make a DVD or CD have an exteremely long shelf life if you record with pre-1920 densities
Tom
Someday, I'll have a real sig.
I used to work in Special Collections at UCSB. It's a nice facility and my boss was really cool. (He introduced me to Mogwai and Do Make Say Think.) He was responsible for getting some (all?) of the cylinders being presented online. He really went out of his way, too. Some of the cylinders he got were turn-of-the-century recordings from some middle eastern country (Afghanistan? I forget which). The labels were all in Arabic. It was really neat to see them up close! I got to play with some nice reel-to-reel players too.
The sibling post has it right. There would have been different settings for every cylinder. And the less you process the sound, the less of a chance there is of worsening the distortion. It might sound cleaner, but you might have messed up the signal a little while clearing out the noise. It was for this reason that when we made CDs of our old acrylic and aluminum 78s, the only processing done was to amplify it as much as possible without letting it clip.
I know the some of the cats that are responsible for this project, and it is entirely a labor of love. They know the subject and have done their damnedest to make sure everything is legal. This is the sort of project that the music industry should laud, and use for favorable pr.
Anthion Thrandocles, Prophet of the Oil
http://www.archive.org/audio/audiolisting-browsear tists.php?collection=78rpm
A lot of these are transfers from the flat Diamond Discs, not the cylinders dubbed from Diamond Discs. Some of those transfers are pretty freakin' amazing. Lots of history here. Hear Irving Berlin sing. Hear why people raved about Enrico Caruso...makes Pavarotti and Domingo sound like punters. Hear Fanny Brice do her schtick. A lot of what is referred to as "Jazz" is actually more like Ragtime. But that can be pretty amazing too.
I came here looking for cartoony music that had passed into the public domain for my upcoming podcast series The Cartoon Geeks. There's lots of it here. Here's the tune that's going to be the theme music. Yowza yowza.
Knowledge is power. Knowledge shared is power multiplied.
Get this -- if you download one of those files and view its ID3v2 info, the "copyright" field says "© 2004 Regents of the University of California".
...
Now, they can't legally claim a new copyright on public domain material unless they've modified the material in a sufficiently creative way for it to qualify as a new "derivative" work. An MP3 doesn't qualify, because there's no creativity involved. This is a bogus claim.
However, given the Creative Commons license on the site's text, the copyright factoids they have in the sidebar, and the fact that this claim would hold up for all of five minutes in court, I'm guessing that this is just a SNAFU.
Densities are nothing, it was the physical elegance of the design itself that lended it to longevity. Hard drives store alot of data, but you can't just pick up a few things lying around the house to pull that data off.
I'll never forget the first time I saw a vinyl lp as a child. It was incredible for me to think that with a sewing needle taped to the end of a cone-shaped tunnel of paper, I could clearly hear the recording by spinning the lp and holding the needle in the groove.
It's nostalgic stuff like this that keeps records around (that and DJ's). Vinyl is much more physical and 'real' than a chunk of shiny plastic you throw in a mystery device (cd player) and have music coming out of your speakers. You can SEE how records work. Records were just an extension of Edison's wax (or in some cases metal) cylinders; same principles at work and just about identical mastering methods.
One fascinating fact that's been touched on before is that during WWII vinyl records were used for telephone encryption. An operator on one side of the connection played a specially recorded white noise record while the operator on the other end did the same. Both records were identical and perfectly synchronized. Then, people on either end would start talking. Using analog sound hardware, both operators could subtract the noise and have the voice transmission come through clearly on both ends (realtime post-processing of sorts)..anyone listening in, without the precious 'decoder' records and settings, would hear nothing but noise. Since the white noise recordings were more than just loops, you needed the entire record and not just a 4 second snippet to decode the audio. It may even hold up in this day and age, sort of an audio one-time-key.
Here is some information about Vinyl:
Westrex 45/45 stereo system - Left channel modulates inner groove. - Right channel modulates outer groove. - A mono signal causes lateral only movement
- An out of phase mono signal causes vertical movement.
http://ultrasonicdesigns.com/audio/vinyl/4.JPG
There are 86 square inches of surface on which to cut.
- More Time = More Space
- More Level = More Space
- More Bass = More Space
Space is measured in lines per inch (lpi). This is called the pitch of the lathe.
- This is the number of grooves (lines) per inch of radius. - More Time requires higher lpi - More Level requires lower lpi - More Bass requires lower lower lpi
Pitch = (Run Time x 33.3 rpm)/Radius (3 inches)
- Max Pitch about 300 lpi - Minimum groove width is 1 mil. - Maximum groove width is 6 mils. - Average groove width is 2.5 mils. Gw = [(1000/lpi) + 1] / 2
An increase in lpi should be accompanies by a decrease in depth. An increase in depth should be accompanies by a decrease in lpi.
Pitch and depth (groove width) are controlled by a cutting computer. The pitch must be changed before the loud parts to prevent over cut. A one half revolution delay is required for the preview channel.
The variable pitch control receives right channel information from the preview system so that the pitch can be increased before loud signals that might cut into the previous groove. Left channel information comes from the program system. A difference signal from the preview system is also sometimes provided.
http://ultrasonicdesigns.com/audio/vinyl/5.JPG
The variable depth control receives the difference (left minus right) signal from the preview system.
http://ultrasonicdesigns.com/audio/vinyl/6.JPG
RIAA Curve
1953 RIAA instituted an EQ curve that narrowed the grooves and improved play time.
Boost high freq. 17 dB at 15 kHz and cut the low freq. 17 dB at 50 Hz.
- RIAA pre emphases is automatically added.
- Post emphases is done at the phono pre amp.
- Inner groove distortion causes high frequency loss (scanning loss).
- A compensation system was tried but mostly abandoned.
- Avoid putting bright (sibilant) cuts in this area.
- A low frequency crossover is almost always used to prevent lift out.
- The effect is to move low frequency signals into the center.
- The frequency below which this happens is variable.
- Cutting head is a moving coil device powered by Cutting Amps.
- Cutting stylus is a heated sapphire
- The cut produces a chip that is vacuumed up for safety.
- The Master Lacquer is an aluminum disc covered in lacquer cellulose nitrate.
The cutting console has four channels of everything 2 preview, 2 program. All controls are stepped for resetting purposes. A reference lacquer may be cut to test settings. A Master Lacquer may not be
played. An Eqed Master tape was made for other Mastering Labs.
http://ultrasonicdesigns.com/audio/vinyl/7.JPG
http://ultrasonicdesigns.com/audio/vinyl/8.JPG
http://ultrasonicdesigns.com/audio/vinyl/9.JPG
http://ultrasonicdesigns.com/audio/vinyl/010.JPG
http://ultrasonicdesigns.com/audio/vinyl/011.JPG
Libertas in infinitum