4 ms·

Wax cylinder recordings digitized

47 points by tintinnabula 11y ago

marincounty 11y ago
A few hundred years ago someone threw out an Edison voice recorder. You would speak into it and it would record on the wax cylinders. I brought it home and looked up its value on the Internet. I was suprised this particular model was not bringing in much money. I didn't have much room, and my girlfriend at the time talked me into throwing it out. Before I threw it out, I took the AC/DC motor out of the device. I just couldn't throw out that beautiful little motor. I now have that motor on my desk as a paper weight. People look at it, and ask why does it have a DC switch. I don't actaually know, but my thoughts were Edison was unsure of our electrical future?
- GFK_of_xmaspast 11y ago
  Edison had strongly held views about the superiority of DC over AC : https://en.wikipedia.org/wiki/War_of_Currents https://en.wikipedia.org/wiki/War_of_Currents
raverbashing 11y ago
Very interesting I wonder if it would be possible to enhance the reproduction by compensating for things like frequency response, wow and flutter, and perform some kind of denoising (some have a periodic noise to it which I think may be attributed to scratches or damage to one side of the cylinder)
- jonah 11y ago
  On the old version of their website you could download the raw rip, a cleaned up version, and the settings file for whatever software they used for the audio processing. You could download the original rip and the settings and adjust them to your own taste. I haven't looked to see why those other files are no longer available.
markbnj 11y ago
This is a great project and a fascinating archive to browse in.
keithpeter 11y ago
These sound great. I've not heard any cylinder recordings before. I'll fish out some of the popular music of the time. I like old recordings: a couple of sources... Early classical recordings available with search function http://www.charm.rhul.ac.uk/sound/sound.html http://www.charm.rhul.ac.uk/sound/sound.html The first few bars of this are wonderful... https://archive.org/details/rhapblue11924 https://archive.org/details/rhapblue11924
- johansch 11y ago
  Yeah, the sound quality is much better than I would expect...
borski 11y ago
I've tried to buy this domain a hundred times, but there's a guy who digitizes these here too, once a month: http://www.tinfoil.com http://www.tinfoil.com
- paxcoder 11y ago
  That design is a delightful blast from the past. tinfoil.com/tinfoil.htm especially triggers my nostalgia. The .ram format on other pages, however, does not.
Animats 11y ago
Those recordings aren't bad. The mechanical cylinder recording process was reasonably good, as was the duplication process; it was mechanical playback that introduced most of the distortion, because the playback mechanism was too massive to reproduce high frequencies well. The Library of Congress has a machine which can read scratched and broken records, by optically scanning and building a 3D model of the surface.[1] But undamaged cylinders don't need it. They can be read out with a stereo cartridge and a suitable stylus. Edison cylinders and records are vertically recorded; the groove bottoms go up and down, not left and right. Later records (RCA Victor, etc) were horizontally recorded. Stereo records are both; the groove format is called 45-45 Westrex, with two axes, both 45 degrees from vertical. This has the nice property that mono records played on stereo players, although one channel is phase-reversed. If you want a cylinder phonograph, they're easily available on eBay for about $500. [1] http://www.npr.org/templates/story/story.php?storyId=11851842 http://www.npr.org/templates/story/story.php?storyId=1185184...
sohkamyung 11y ago
Interesting. Checking the overview page [ http://cylinders.library.ucsb.edu/overview.php http://cylinders.library.ucsb.edu/overview.php ] shows how they got the recordings. At first, I though the recordings might have been based on the optical imaging technique outlined in this article [ http://www.symmetrymagazine.org/article/june-2015/lhc-physicists-preserve-native-american-voices http://www.symmetrymagazine.org/article/june-2015/lhc-physic... ] but it is apparently been used for a separate effort to preserve native American recordings.
S_A_P 11y ago
I wanted to download all of the available recordings so I built a tool to do so: https://github.com/CraigAlbright/hotwaxresidue https://github.com/CraigAlbright/hotwaxresidue

47 points by tintinnabula 11y ago

A few hundred years ago someone threw out an Edison voice recorder. You would speak into it and it would record on the wax cylinders. I brought it home and looked up its value on the Internet. I was suprised this particular model was not bringing in much money. I didn't have much room, and my girlfriend at the time talked me into throwing it out. Before I threw it out, I took the AC/DC motor out of the device. I just couldn't throw out that beautiful little motor. I now have that motor on my desk as a paper weight. People look at it, and ask why does it have a DC switch. I don't actaually know, but my thoughts were Edison was unsure of our electrical future?

GFK_of_xmaspast 11y ago
Edison had strongly held views about the superiority of DC over AC : https://en.wikipedia.org/wiki/War_of_Currents https://en.wikipedia.org/wiki/War_of_Currents

raverbashing 11y ago

Very interesting I wonder if it would be possible to enhance the reproduction by compensating for things like frequency response, wow and flutter, and perform some kind of denoising (some have a periodic noise to it which I think may be attributed to scratches or damage to one side of the cylinder)

jonah 11y ago
On the old version of their website you could download the raw rip, a cleaned up version, and the settings file for whatever software they used for the audio processing. You could download the original rip and the settings and adjust them to your own taste. I haven't looked to see why those other files are no longer available.

markbnj 11y ago

This is a great project and a fascinating archive to browse in.

keithpeter 11y ago

These sound great. I've not heard any cylinder recordings before. I'll fish out some of the popular music of the time. I like old recordings: a couple of sources... Early classical recordings available with search function http://www.charm.rhul.ac.uk/sound/sound.html http://www.charm.rhul.ac.uk/sound/sound.html The first few bars of this are wonderful... https://archive.org/details/rhapblue11924 https://archive.org/details/rhapblue11924

johansch 11y ago
Yeah, the sound quality is much better than I would expect...

borski 11y ago

I've tried to buy this domain a hundred times, but there's a guy who digitizes these here too, once a month: http://www.tinfoil.com http://www.tinfoil.com

paxcoder 11y ago
That design is a delightful blast from the past. tinfoil.com/tinfoil.htm especially triggers my nostalgia. The .ram format on other pages, however, does not.

Animats 11y ago

Those recordings aren't bad. The mechanical cylinder recording process was reasonably good, as was the duplication process; it was mechanical playback that introduced most of the distortion, because the playback mechanism was too massive to reproduce high frequencies well. The Library of Congress has a machine which can read scratched and broken records, by optically scanning and building a 3D model of the surface.[1] But undamaged cylinders don't need it. They can be read out with a stereo cartridge and a suitable stylus. Edison cylinders and records are vertically recorded; the groove bottoms go up and down, not left and right. Later records (RCA Victor, etc) were horizontally recorded. Stereo records are both; the groove format is called 45-45 Westrex, with two axes, both 45 degrees from vertical. This has the nice property that mono records played on stereo players, although one channel is phase-reversed. If you want a cylinder phonograph, they're easily available on eBay for about $500. [1] http://www.npr.org/templates/story/story.php?storyId=11851842 http://www.npr.org/templates/story/story.php?storyId=1185184...

sohkamyung 11y ago

Interesting. Checking the overview page [ http://cylinders.library.ucsb.edu/overview.php http://cylinders.library.ucsb.edu/overview.php ] shows how they got the recordings. At first, I though the recordings might have been based on the optical imaging technique outlined in this article [ http://www.symmetrymagazine.org/article/june-2015/lhc-physicists-preserve-native-american-voices http://www.symmetrymagazine.org/article/june-2015/lhc-physic... ] but it is apparently been used for a separate effort to preserve native American recordings.

S_A_P 11y ago

I wanted to download all of the available recordings so I built a tool to do so: https://github.com/CraigAlbright/hotwaxresidue https://github.com/CraigAlbright/hotwaxresidue