Print your own [vinyl ?] record with a 3D printer....

Neat... I had no idea that 3D printing had that sort of resolution.

George H.

It hasn't

"600dpi in the x and y axes and 16 microns in the z axis"

The maximum amplitude of modulation on an L.P. is around 0.005" and 600 dpi corresponds to 0.00167", so the waveform would be quantised into 3 levels each side of zero. The description that the sound is "recognisable" seems almost optimistic.

It might give Edison's original first-generation cylinders a run for their money. Even old shellac-based 78s would sound better, and as for a well-pressed LP on low-noise vinyl, I don't think the sound quality would even be on the same continent.

There are some interesting micrographs of LP grooves at

If the scales here are to be believed, these grooves aren't all that much wider (or deeper) than the 16-micro Z-axis resolution being quoted. The minimum audible groove deflection and feature size has got to be far below that.

On a sunny day (Fri, 21 Dec 2012 10:41:50 -0800) it happened snipped-for-privacy@radagast.org (Dave Platt) wrote in :

Have you listened to the examples? I think it is pretty good, especially the bass. I have read the resolution is about 5 to 6 bits. In German:

Plastic will smooth the ride'. Amazing the needle even tracks correctly, altough you hear some wining sounds in sync with rotation of the disk.

It is likely that because of pickup inertia the dots between samples will be smoothed. Very long time ago I invented a compression scheme for audio, and actually coded it in Z80 asm. It had a huge compression, the way it worked is just looked for maxima and minima, stored time and amplitude, and on playback replaced that with half sine waves of the same amplitude and time. Amazingly it sounded recognizable. Then later mp3 came and that was much better:-) Both systems, and this one, throw away the lower amplitude difference details.

I think the record printing thing is cute, and of course only the beginning. Bit of signal processing could make it better I am sure.

Anybody do a CD? LOL

Even just two level quantisation of sound can produce a result that is recognisable. I tried it once with the "The Liberty Bell" march (a.k.a. Monty Python tune) on a Sinclair Spectrum. Unsurprisingly the playback contained a hell of a lot of quantisation noise, but it was still recognisable.

Sylvia.