Sound preservation as pertaining to what? The storage of? The restoration of? The manipulation of and the requirements in which to do so?

Digital recording is in and by itself merely a representation of the audio. Something that has been minutely scrambled and reassembled in various ways. The most popular way and one of the earliest methods is what we still currently and mostly utilized today known as Pulse Code Modulation a.k.a. PCM. But as technology marches on, there are now also other variations known as DSD created by Sony/Philips. PCM is based upon a sampled " word " consisting of numerous bits and sampled at different speeds. DSD is the most recent refinement of digital audio not yet in widespread use or acceptance. This concept is not that of a digital word but of a single bit that has been sampled at millions of times per second instead of at thousands of times per second. Still the need for sushi style audio. Nothing today is like a continuous analog stream of old-fashioned disc cutting or analog magnetic recording.

Nevertheless, sound is preserved in time in a similar manner that film was to analog videotape and now digital video. All the same aspects apply.

If you are speaking in terms of recovery and/or restoration, current technology offers up capabilities not before capable in purely the analog realm. Kind of like computer animation as compared to the original consecutive freehand drawings of cellular animation (think Mickey Mouse on the boat). Computers today have offered up the capabilities of having virtual reality. Both visually and sonically. And for instance, this explanation of mine in print may last long after I am dead.

I have done a lot of archiving from 78 rpm-45-33 1/3 discs. A lot from full width, 1/4 inch monaural magnetic recording tape to 2 inch 24 track magnetic recording tape and even 1/8 inch 4 track cassette tapes to the digital realm. Digital specifications originally and today far exceeds that of any analog recording of the past. Because of this wider working latitude we can stretch otherwise dynamically reduced, dynamic range. We can eliminate residual noise levels. We can enhance poor quality recordings. We can (partially) restore overloaded and distorted early recordings. All of which are made possible by carefully crafted mathematical algorithms. Most of which, only geeks can create. We are just the end-users of these algorithms. Some of these algorithms can be extremely expensive to obtain. Others give them away free on the Internet. Some algorithms have been so carefully crafted so as to emulate specific pieces of equipment. Specific places and spaces. Effects for pop music where the previous hardware cost many thousands of dollars which today can be obtained in a piece of free shareware.

In reference to your question regarding telling the computer what volume to reproduce at what range of frequencies and maintaining the power? That is essentially a non sequitur. But it is also correct in your naïveté. Computers produce no power. Speaker amplifiers a.k.a. power amplifiers can reproduce signals and modulate power to deliver to speakers. The computer merely deals with theoretical math. So if you want to tell the computer that the sound which you have captured should be dynamically exaggerated in the high frequencies, you can do that. If you want to tell the computer to dynamically exaggerate the low frequencies, you can do that. If you want to tell the computer to play your captured signal at the same tempo but at a different pitch, it can do that. If you want to tell the computer to play your captured signal at the same pitch with a different tempo, you can do that. If you want to tell the computer to differentiate a signal from residual noise and remove the residual noise, it can do that. I can only do that however if the proper algorithm has been written by a geek who understands what you want of them. Because the computer can do nothing without a set of instructions from someone who has written instructions that a particular computer and its related operating system that is integrated with the operation of that computer, understands.

That book looks to be very interesting. Unfortunately, due to time constraints and lack of supply funding, I am not currently able to direct you to any other possible online resources that could be helpful. Especially since your question is rather vague as to what your question is really all about? This particular article looks to be aimed towards restoration of older analog recording sources. And your question really has nothing to do with the restoration of old recordings. We utilize in our recordings and in many restorations modifications of dynamic range and frequency content. Even the addition of mathematically formulated artificial ambience a.k.a. reverb can be utilized to help mask distortion components. While adding yet another dimension to the restored source. You never get something for nothing in anything that you do. Understanding the range and limits of what you are trying to do and with what kind of programs you're trying to do them with is of paramount importance. It's certainly not for any entry-level people to be able to produce professional results in the restoration of old recordings. Software may provide the power to do so but only experience and technique will enable one of the power to properly utilize the algorithm is provided in which to do so. In other words, you can't make a hit recording because you do not yet know how to. You may know how to drive but you are not ready to win the Indianapolis 500 much less the Daytona 500. I can make a hit record and I may know how to fly a plane but I am not yet ready to do battle in an F-22 (even though I've flown a Boeing simulator of that plane). I mean I could but I'd probably meet the Loch Ness monster before you ever have a chance to. LOL or so to speak.

So what really is your question?
Mx. Remy Ann David

I think there are several sources of confusion in your original post. For the purposes of recordings or preservation when there is a relatively high signal-to-noise ratio, the concept of power has little meaning, as Remy mentioned. The important unit is information, and this is usually tied in with the [[url=http://[/URL]="http://en.wikipedia…"]Shannon-Hartley theorem[/]="http://en.wikipedia…"]Shannon-Hartley theorem[/].

When you replay a recording, it is indeed up to you how "loud" you play it, and thus how much power is involved at that stage. Note, however, that if you are dealing with fields such as radar or medical sonography where the signals can have similar power levels to the system noise, then you have to approach this topic differently.

The Peter Copeland paper you referenced provides an historical perspective, but many of the limitations he assumes have been overtaken by development of technology, not only in the accuracy (number of bits) of digitization, but also in the amount of storage now available as well as the type of algorithms developed for lossless compression.

Boswell, post: 391987 wrote: I think there are several sources of confusion in your original post. For the purposes of recordings or preservation when there is a relatively high signal-to-noise ratio, the concept of power has little meaning, as Remy mentioned. The important unit is information, and this is usually tied in with the [[url=http://[/URL]="http://en.wikipedia…"]Shannon-Hartley theorem[/]="http://en.wikipedia…"]Shannon-Hartley theorem[/].

When you replay a recording, it is indeed up to you how "loud" you play it, and thus how much power is involved at that stage. Note, however, that if you are dealing with fields such as radar or medical sonography where the signals can have similar power levels to the system noise, then you have to approach this topic differently.

The Peter Copeland paper you referenced provides an historical perspective, but many of the limitations he assumes have been overtaken by development of technology, not only in the accuracy (number of bits) of digitization, but also in the amount of storage now available as well as the type of algorithms developed for lossless compression.

I will get back to the original issue later, I'm kinda reading through some other things now, but I wanted to ask whether you know of any more up to date source I could use?

Would it be valid to presume that the methods for extracting the information from the carrier and the considerations involved haven't changed that much and that the changes have been in the digital side of things?

Unregistered, post: 392011 wrote: Would it be valid to presume that the methods for extracting the information from the carrier and the considerations involved haven't changed that much and that the changes have been in the digital side of things?

I'm puzzled as to why you are suddenly talking about a "carrier". If you mean the word in the signal processing sense, then it's not relevant here as all the signals we are concerned with are baseband. If you are talking about it in marketingspeak to mean a CD or some other method of containing a finite block of information, then the answer to what I take to be your question must be no. The methods of storage and extraction of information are continually changing and developing, whether the information is in analog, digital or some other parametric form.

We can give you pointers and can set you straight on some direction and terminology, but, however interesting it would be, I'm sorry to say we are not here to do your homework for you.