Why doesn't someone make a modern analog optical disc format for audio?

[MaximRecoil]

Using a modern type of optical disc such as Blu-Ray — given its high storage capacity — it seems to me that an extremely high-bandwidth analog audio encoding scheme could be developed for it (possibly surpassing the bandwidth of the 2-inch tape that is traditionally used for analog master recordings).

The master tracks and master mixes could be recorded on hard drives using the same (or superior, given their even larger capacities) analog audio encoding scheme.

An analog optical disc would have the benefits of digital media (e.g., audio CDs), such as random access, contactless playback, and suitability for playback in portable hardware (e.g., in a vehicle); and none of the drawbacks of vinyl records, such as having to flip them over to hear them in their entirety, wear & tear resulting from normal playback, inevitable "coloration" of the audio to one degree or another by the needle/tonearm, cleaning requirements, and lack of portability.

Since digital audio data has to be converted to analog eventually in order to hear it (i.e., in order to drive speakers), and it is analog when it is created, it seems like the best approach would be to leave it in an analog format for the entire chain; as opposed to analog --> digital --> analog as is the chain of a typical audio CD (or analog --> analog --> digital --> analog for older recordings that get digitally remastered for release on CD).

[Iain McNeill]

I would suspect that the optical system lacks linearity and dynamic range.

An analog "table of contents" to allow true random access would also be challenging.

[MaximRecoil]

Some of the LaserDiscs that had analog audio had good sound quality (the quality varied though, depending on the quality of the player and/or the mastering), and that was on an old format which only had the data density of a CD, and having to share space with video and several other audio channels (sometimes both analog and digital).

I don't see how dynamic range and linearity cababilities have anything to do with the fact that the audio information is being read by a laser. Shouldn't those be properties of the audio information itself?

Edit: I didn't word that well. Those properties would be determined not only by the audio information itself, but also by the laser's ability to read it accurately, and the quality of the output circuitry.

LaserDiscs had random access, even when used with the analog audio tracks (and originally they only had analog audio tracks). I don't know exactly how it was accomplished.

I'm thinking LaserDisc audio on steriods, due to exponentially greater capacity to work with (this allowing extreme bandwidth).

[Glowbug]

I've wondered why the laser pickup idea for LPs never became more widespread/cheaper...there had to have been some technical issue with it, otherwise it doesn't make sense why more companies didn't try to implement it.

[wwenze]

Reliability of detecting continuous-varying light levels at high speeds probably reduces its usefulness in data storage and hence halted its development, audio is such a small and niche market.

Did laserdiscs have analogue audio tracks?

[MaximRecoil]

Originally Posted by Glowbug

I've wondered why the laser pickup idea for LPs never became more widespread/cheaper...there had to have been some technical issue with it, otherwise it doesn't make sense why more companies didn't try to implement it.

I like the idea of those too, except you still have some of the disadvantages of vinyl records (e.g., having to keep them extremely clean, lack of portability, and I assume you still have to flip them to hear both sides). And of course, they are extremely expensive. They maintain a niche market though, and according to Robert Orban and Greg Ogonowski (from their article Maintaining Audio Quality in the Broadcast Facility):

The authors have thoroughly evaluated the ELP and we can recommend it as delivering higher audio quality than any other vinyl playback device known to us.

Additionally, vinyl records are limited with regard to their potential accuracy relative to the source (the "acetate"). The acetate goes through various metal coating processes to get to the "stamper" stage, which results in some degree of loss.

[MaximRecoil]

Originally Posted by wwenze

Did laserdiscs have analogue audio tracks?

Yes. Originally they only had analog audio, and later they had both analog and digital audio. They had analog video too.

[thoriated]

I think that they should define an analog format for blu-ray. It doesn't have to be linear in amplitude, just time. That would allow a FM based analog format. And to take care of timing variations due to rotation, there could also be a digital track which could have time base information, as well as standard types of digital track info, etc.

If this blu-ray audio standard was specified so that it could be practically implemented in both record and playback with vacuum tube circuitry up to the laser record/playback diode, then that would be close to being a 'best of all worlds' format. And after a few years or a decade, when everybody is streaming their digital video from online or storing it in solid state memory instead of using a moving media format, blu-ray as a format would be still in business because of analog.

Well, I can always hope, can't I?

[jcx]

I think that DSD is the closest thing to "analog modulation" (with a spread spectrum carrier but descretized amplitude & zero crossing times) that can be encoded given the tweaking of the media&players for digital storage/retrieval

a big question is why anyone would want to go with mechanical timing for audio today:

"Every analog magnetic recording starts with a motor dragging a rusty strip of plastic over various rollers and guides, and across scraping metal parts. Unfortunately, every inconsistency in the speed of the tape as it traverses this obstacle course serves to distort the music being recorded. The transport's various imperfections create an ever-changing matrix of speed variations, slow or fast, subtle or severe. At worst, this results in the familiar warps and warbles known as "wow" and "flutter." It can even (as in the famous case of Miles Davis' "Kind of Blue") cause a perceptible tuning change over time.

Even the very best analog recordings are affected by varying and shifting patterns of constantly overlapping high frequency flutters, causing random beat frequencies to be introduced... all of which seriously interferes with the natural harmonic structure of the musical material.
"
Plangent Processes

spinning disks still have this problem - the optical head has a radial tracking servo and IMD at the DVD rotation frequency will be bigger in a "pure analog" encoding than it is in buffered PCM where even today some rotation speed IMD is sometimes still visible, even with digital fifo/reclocking (although this might be ps/motor/servo current coupling to the crystal)

[thoriated]

I think the primary reason is that the discrete process of A/D/A conversion ( the so-far intractable bugaboo of digital where real world precisions exceeding 18 bits quantization monotonic without missing codes are extremely rare) is completely sidestepped by keeping the process analog. As could be the typical RF contamination problem endemic with almost all digital equipment. Also, SQ restrictions due to standardized word length and quantization rate limitations are no longer relevant.

Then, it should go without saying that the amplitude of speed and vibration variations during the record and playback process can be continually reduced as the mechanisms are refined. Who knows, maybe somebody will come up with a method of recording and playing blu-ray media where the disc does not move at all and all motion of the record/playback beam is handled with optical precision which has already been refined to precisions of far better than parts per million.