Half a croissant, on a plate, with a sign in front of it saying '50c'
h a l f b a k e r y
Eureka! Keeping naked people off the streets since 1999.

idea: add, search, annotate, link, view, overview, recent, by name, random

meta: news, help, about, links, report a problem

account: browse anonymously, or get an account and write.

user:
pass:
register,


                                             

Better Analog Music Recording

A laser based analog audio recording and playback system
  (+4, -1)
(+4, -1)
  [vote for,
against]

The problem with digital recording is the need for high sampling rates and great attention to the digital filtering. Currently, there are some high end alternatives which provide good sound, but the add on processors require thousands of dollars of investment. There is also the potential that DVD-Audio just won't cut it....and also that SACD won't make it....for years to come.

There is no reason why a high tech version of analog recording can't be perfected at low cost. Today, the Japanese are marketing a laser turntable. Fortunately, I'm going to garage sales and adding to my already over 8000 collection of vinyl....in anticipation that the day will come where this is marketable for $1000 or so (currently it's $10,000 cause it's not selling).

Anyway, envision a means of going directly from the microphone transducer to a laser based recording syste4m and media. Can't light intensity levels be recorded? Even if these had to be recorded using "ones and zeros" imagine a "direct" means of A/D, which could then be "demodulated" during playback in a direct manner.

The storage media is a big issue. You could have lasers potential cut some material to potentially make a small "record disk".

In keeping in the spirit of "half-baked" I'll stop here.....

BobWade, May 07 2002

[link]






       recording using ones and zeros *is* digital recording. Period.   

       I don't really see the particular advantage that laser technology would be offering. (More generally I don't see the need for analogue recording once digital recording becomes good enough - and it's already good enough that to really be able to tell analogue is better, you need one hell of a turntable - but that's an opinion)   

       Can we have more of the details of the possible solution? Otherwise it sounds a bit WIBNI-ish, IMHO.
yamahito, May 07 2002
  

       It's possible: using lasers rather than needles to cut records and play them back. I believe laser turntables have serious problems if there is any dirt on the surface however - that's one of the major benefits of digital technology, that it allows for error detection and correction.   

       But most recording nowadays is digital, so there's little benefit in analog compared to digital media (DVD-Audio at least has the benefit that it plays music back at the same sampling rate, 48kHz, as it is typically recorded, unlike CDs.)
pottedstu, May 08 2002
  

       "Better analog" is an oxymoron.
waugsqueke, May 08 2002
  

       To be more half-baked, I wonder if you could have a recording format which uses ultrasonic signals instead of lasers. Using sound to play back and record sound seems infinitely more appealing.   

       (How is "better analog" an oxymoron? Reel to reel tape is a better analog recording technology than wax cylinders. Nothing oxymoronic there.)
pottedstu, May 08 2002
  

       Picky picky. What I mean is that it's pointless to try to improve an inherently deficient technology. No matter how much better one makes analog recording, it can never be better than digital. In that sense, the phrase is an oxymoron.
waugsqueke, May 08 2002
  

       Au contraire [waugs] we live in an analogue world, digital processing approximates a waveform by splitting it up and taking a value at time intervals. As the sampling frequency increases you get closer to the true sound but it will always be an approximation of the original analogue wave, albeit a pretty damn good one.
IvanIdea, May 08 2002
  

       //we live in an analogue world//   

       This will piss off my QM lecturers... They'll have to get real jobs.
yamahito, May 08 2002
  

       I'd imagine your QM lecturers could happily point out that we live in a world which is both/neither/either* analogue and digital, depending on which way you look at it.   

       * I have a much more elegant description of QM but I don't have space to write it in this margin.
vincebowdren, May 08 2002
  

       //don't have space to write it in this margin//   

       Like Fermat's Last Theorem, huh?
yamahito, May 08 2002
  

       Given the amount of information that can be put on digital media now (which in the labs is several orders more than on the street, remember) i don't see the point of analog.   

       // high sampling rates // What do you think analog really is, if not a high sampling rate? Atoms, even quanta, aren't infinitessimally small in the detail they can record, so looked at closely enough any form of analog recording will always be digital. The extra 'feature' of analog is just that this degrades so easily when they get dust or scratches on it.   

       [yamahito] hey, is that you?
sadie, May 08 2002
  

       [Yamahito] I think you are confusing discrete with digital. I feel we are greatly in danger of bringing Shrodinger into the discussion and before you know it we'll be postulating cat powered toast recording technology with butter thickness defining the captured sounds. [Sadie] Quite agree with you about benefits of digital, especially degradation. but that wasn't the point.
IvanIdea, May 08 2002
  

       Just a joke, guys...   

       Bsides, is there really a difference between discrete and digital? I guess it's all semantics from here on out..
yamahito, May 08 2002
  

       There is a wealth of information concerning why digital music today is not perfect. I recommend doing basic searches on digital recording....how it has to be done right....the use of digital filters....of all diffent types......and the current upsurge of digital fixes now which are provided by upsampling....etc etc....once you get started you'll get in real deep very fast. People are buying them and noting the absolute improvement in truth provided by alleviating some of the phase distortion.   

       The bottom line is digital's current inability to achieve as fast a transient response as analog and also to alleviate phase distortion makes it only a facimile of the truth. Analog suffers from mechanical and magnetic distortion....but it provides a "window of truth" beyond which current 44.1 can provide.   

       One is left perplexed to figure how the human ear is so sensitive to such subtleties. The answer is simply...a well trained person....exposed to the good and the bad...can tell the difference. I work in acoustics at the US Army Research & Development Center. I know of work by the Navy that shows the human can discern PHASE differences of nanoseconds!! The common audiophile obsessed with harmonic distortion and frequency response has no idea what the phase problem is or how acute our senses are to it.   

       No...you don't need the ultimate analog equipment to achieve sound superior to 44.1KHz CD....that's easy to do. The court is still out on how analog competes with DVD-Audio and SACD, however.   

       The most basic math tells us that 44.1KHz CD cannot achieve the rise time of low microseconds that cartridges and speakers can achieve. DVD-A at 192KHz is certainly a step in the right direction in terms of slew rate....but have they solved the problem of phase distortion? SACD is much higher...but I haven't heard reviews that this is much better than DVD-A...so there's something possibly hidden in the digital implementation that's still a problem.   

       Read umtiump reviews on the latest digital enhancement, upsamplers, etc...and see what you can make of it.   

       BTW...early versions of the laser turntable were not good (not sure why)...but recent Japanese version is....and it is very listenable after the record is vacuumed...and I recommend sending for the free CD just to get some sense of what's going on (it makes for a good CD even though it's 44.1).   

       My question though....is not to prejudge the merit of the idea based on a prejudice of whether analog is better than digital...but to conceive of a simple, effective, efficient means to accomplish a "direct level recording and/or playback"....without the traditional magnetic tapes or other archaic means. Can't technology provide a new answer to this?   

       For those of you that can't get past digital....I would recommend that you conceive of the perfect digital filter...then you might have something.   

       Having digital as part of a "process" of defining light intensity or its use as defining an "analog level" would not necessarily introduce the historic problem of A/D and D/A. In other words....if the analog signal was truly converted in a pure analog way to a light intensity...and if this light intensity was recorded onto a CD as 1s and Os...then deciphered using an analog "conversion" to voltage...such a direct representation of the signal and its conversion back to analog would not be subject to digital filter distortion in the D/A process.   

       In fact...with such a playback system....current CDs could even be played back using this "analog converter" reading the 1s and 0s as direct signal levels.   

       BTW....there is quite a force out there which believes that 44.1KHz contains sufficient information and can produce considerable fidelity....provided the D/A process is "digitally enhanced" beyond current levels. Thus...you wouldn't necessarily have to impose the "direct analog" process at both recording and playback ends (although you potentially stand to gain alot in terms of bandwidth)....one could also employ it using current CDs....and possibly get as good as the current $5000 to $10,000 digital enhancement boxes do.   

       Now that I think about it....this is probably the most logical place to start considering the momentum of the industry...etc. Conceive of a simple efficient way of reading the "analog levels" represented by the 16 bits on a CD...then convert directly to voltage.   

       Do this and you compete directly with the high priced processors out there...and still maintain the use of a high volume of existing CDs.
BobWade1, May 08 2002
  

       I disagree with [waugsqueke] about "better anaolg" being an oxymoron. However, I do appreciate where the notion likely came from. Namely, that analog technologies to date have failed to achieve robustness. Analog recordings as we know them today are subject to degradation owing largely to environmental interaction with the recording and transmission media. Digital formats cause a small amount of degradation up front (quantizing noise, band limiting, bizarre filtering artifacts), but protect from further degradation by de-coupling the program content (now an abstract list of numbers, basically) from the underlying medium.   

       If some new analog technology were developed which could protect itself from environmental influence, then there would be no up-front penalty *and* there would be no back-end degradation. Hence you could in theory have a higher quality recording (and maintain it that way) than the best possible digital recording, regardless of sampling rate and bits per sample.   

       Unfortunately, [BobWade] has missed some critical facts about the nature of digital and analog formats and the conversion between them. I will attempt to fill in some gaps.   

       First of all, any "ones and zeros" format is inherently discrete-time and discrete-value. This is the definition of digital. Analog is, by definition, both continuous-time and continuous-value. Unless the entire path is pure analog, you are necessarily paying the penalty for a digital system. At that point you may as well do it all digital, to prevent the analog portions from degrading your output further.   

       Second of all, optical technologies do seem the likely candidates for building an all-analog recording medium that can remain unaffected by its environment. But the medium would have to be able to change continuously in value, such as with a smoothly changing surface reflectivity. Also, the medium would have to support continuous-time access, such as with an unbroken linear path (if tape) or spiral (if disk).   

       Such a format might be feasible if CD-R technology produces a continuously graded response to a writing laser whose output power can be varied with the analog signal. Anybody here know if this is the case?
BigBrother, May 08 2002
  

       WIth a sufficient sampling rate and resolution, digital audio can achieve performance superior to any analog system. Of course, recording audio as 24-bit "floating-point" samples [7-bit exponent, 1+16-bit mantissa, 1-bit sign with denormalization support] at a 14.3818MHz would be a bit expensive in both equipment and storage media, but the technology exists to do it.   

       The facts that most digital recording implementations use 16-bit samples at 44Khz, and that sufficiently-expensive analog systems can be made better than those, do not imply that the best analog system that could be produced is better than the best digital system that could be produced.
supercat, May 08 2002
  

       [supercat], I'm talking about theoretical limits of *ideal* digital and analog systems here. An ideal (i.e. "the best possible") digital system has an inherent loss built in. Consider an analog input to an ideal ADC, which feeds an ideal DAC, which drives an analog output. This loss in quality is a function of the sampling frequency and the number of bits in a sample. Even if you increase the sampling frequency and bit count to insanely high levels, there is still some loss. An ideal pure analog system has no inherent losses.   

       Now back to the real world. Real analog systems in use today are no match for real digital systems in use today. That is because analog systems suffer from external influences coupling into the system. You get hiss and hum coupled in through every stage of the process. Digital systems are largely immune to these effects because they can take advantage of redundancy and lossless regeneration techniques.   

       Now if some *magic technology* existed to shield analog systems from unwanted external coupling, then analog systems would achieve their theoretical limit and surpass the performance of all conceivable digital systems.
BigBrother, May 08 2002
  

       //Now if some *magic technology* existed to shield analog systems from unwanted external coupling, then analog systems would achieve their theoretical limit and surpass the performance of all conceivable digital systems.//   

       There exist digital systems which quantize beyond the degree of accuracy of the associated analog electronics. Any loss in signal quality from the act of digitizing is small compared to the loss of quality in the analog input stages. Since adding digital processing can remove the need for a certain amount of analog processing, digital is an all-around win if the quantizing limits don't approach the original signal.   

       BTW, ordinary analog recording tape is continuous-time but discrete-value, much like a half-toned photograph. The only methods of recording which are not noticeably discrete-value are gramaphone records and optical movie soundtracks.
supercat, May 08 2002
  

       If we are going analog, remove the 60 (or 50) Hz hum from the signal and make the recording acoustic and make the playback mechanical.
Amishman35, May 09 2002
  

       In the consumer market, a convenient but flawed digital solution is always likely to win out over a complex, hard-to-use but sonically superior analog system. Aside from quality issues, the advantages of digital media are many: easy storage on magnetic and optical media; electronic transmission; cheapness of components; lower bandwidth (possibly via compression); forward error correction for robustness; theoretically infinite duplication; ease of signal processing (both in design and operation); compatibility with existing digital electronic sound sources, recorders and signal processors; guaranteed similarity of mass-produced media. All these factors mean a return to analog recording in either professional or consumer products is very unlikely.   

       Of course the fact that I'm a DSP (*digital* signal processing) engineer in no way influences my opinions.
pottedstu, May 09 2002
  

       Hey things got going pretty good....I wish we could focus on the "direct light" recording and media...which
Bob Wade, May 09 2002
  

       Sorry, I got up for a second and junior pressed the OK button. No...I do understand obviously the discrete sampling nature of digital recording. I thought my comments on slew rate exposed that. Curious, that the new upsamplers and enhancers...coupled with even 44.1 are allegedly yielding outstanding fidelity....matching at least...(in many reviewer's opinions)....DVD-Audio. Has anyone out there heard DVD-Audio? An a-b on that compared to standard 44.1? I don't know of anyone that hasn't noticed a marked improvement even when compared to 96/24. Case closed...digital has not reached some "highest level" of fidelity. The fact that the human can discern nanosecond phase differences is going to be hard to fool.   

       Yes, Bigbrother....look at the bandwidth potential of optical...now all we have to do is record it direct. I've thought about the mechanical cutting, the mechanical this and that....and haven't had a brainstorm yet. There has to be a simple answer which avoids the usual drawbacks. Something which is "infinitely (yes I know there is no such thing) adjustable" in its implementation.....capable of storing this wide bandwidth.   

       But the latest idea I thought might have more promise....at least in defining a more practical solution at the playback side only. In other words....live with the digital recording world for the time being.....which still has years of growing pains to get through. Digital can be enhanced by being "cleaned up" on the playback side. Take a DVD-Audio...pretty good recording...then use the digital recorded levels in a "direct to analog" playback system....and you can compete with the $10,000 players/processors that are out there.
Bob Wade, May 09 2002
  

       Whatever happened to ol' Bob? He was a fixture for a whole week & most of his ideas evaporated - leaving only this one. Croissant
thumbwax, Jun 12 2002
  
      
[annotate]
  


 

back: main index

business  computer  culture  fashion  food  halfbakery  home  other  product  public  science  sport  vehicle