Skip to main content ↑ ↓

Notice

Please note that most of the software linked on this forum is likely to be safe to use. If you are unsure, feel free to ask in the relevant topics, or send a private message to an administrator or moderator. To help curb the problems of false positives, or in the event that you do find actual malware, you can contribute through the article linked here.
Topic: High Resolution audio - interesting paper (Read 3824 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

High Resolution audio - interesting paper

This is an old paper, but upon returning to it today I was struck by how well written and carefully argued it is. You'll need some spare time to do it justice...

http://www.meridian-audio.com/ara/coding2.pdf

There are several discussions of this subject in the HA FAQ.

Cheers,
David.

High Resolution audio - interesting paper

Reply #1 –
Quote from: 2Bdecided on
This is an old paper, but upon returning to it today I was struck by how well written and carefully argued it is. You'll need some spare time to do it justice...

http://www.meridian-audio.com/ara/coding2.pdf

There are several discussions of this subject in the HA FAQ.

Cheers,
David.


This paper is more of an op-ed piece than a scholarly monograph.    It doesn't present any new research or insights but simply functions as a sounding-board for the author's views.  He tries to dress it up iin academic robes and a mortar board by referring to himself in the third person (no fewer than 18 times!)  and having a long list of references, but many of his references are from non-published sources (or as he says, "private publications") and internet threads.   

The encoding standards we choose are a compromise between a need to record every last possible detail in the music and practical and cost limits of storage, bandwidth, and the desir to have some unified standards so music can have a wide distribution.    The idea that we should design our encoding standards to accomodate the .01% of the music-critical population, or whatever it is, who can hear 26 KHz makies about as much sense as saying that all cars should be designed to comfortably accomodate that portion of the population who are 7'4" or who weigh 400 lb's.

High Resolution audio - interesting paper

Reply #2 –
His goal (like many on HA) is transparent reproduction of audio.

Given the cost of a blank DVD-R, I hardly think the cost of recording at or greater than 56kHz 20-bit is comparable to the cost of designing all cars "to comfortably accommodate that portion of the population who are 7'4" or who weigh 400 lb's"!

The paper doesn't claim to present new research - it's a lengthy summary of the reasons we might need more than 44.1kHz/16-bit in exceptional circumstances. Refreshingly, he doesn't claim that most people will hear a difference, and he passionately argues that going beyond the known limits of hearing is benign nonsense, and downright damaging to the listening experience if excesses of sample rate or bitdepth force us to stick with 2 channels.

It is, of course, marketing. However, he has the advantage of not just quoting a long list of references, but actually knowing and conversing with many of the people on that reference list.

Cheers,
David.

High Resolution audio - interesting paper

Reply #3 –
Quote from: 2Bdecided on
His goal (like many on HA) is transparent reproduction of audio.

Given the cost of a blank DVD-R, I hardly think the cost of recording at or greater than 56kHz 20-bit is comparable to the cost of designing all cars "to comfortably accommodate that portion of the population who are 7'4" or who weigh 400 lb's"!

That would only be true if you are recording your own live music.  For 56K/20bit to become a commercial standard would require that the whole pipeline support it, from the recording studio/concert-hall to the mixing and post-processing, to the manufacturing, to the distribution, to the playback devices.    That's a huge investment and changeover for the teensy proportion of the population who can hear any difference under ideal conditions with the current standard.

High Resolution audio - interesting paper

Reply #4 –
Quote from: 2Bdecided on
His goal (like many on HA) is transparent reproduction of audio.
[..]
It is, of course, marketing. ...

It may be worth noting that a lot of this, however interesting and good points are raised, seems targeting the use of MLP (Meridian Lossless Packaging). And Lossless compression does nothing for (or against) transparent reproduction compared with PCM.
In theory, there is no difference between theory and practice. In practice there is.

High Resolution audio - interesting paper

Reply #5 –
True.

The argument is that the required sample rate, bit depth, and number of channels don't quite fit within the DVD maximum datarate - unless you use lossless packing.

I'm not sure this is a strong argument, given that B-format ambisonics only requires 4 channels.

However, I think the argument that it's better to use 96kHz sampling with lossless packing, than raw PCM at a lower arbitrary sample rate, is a strong one. The argument that you might as well use lossless packing at whatever sample rate, simply for efficiency, is even stronger.


Quote from: plnelson on
That would only be true if you are recording your own live music.  For 56K/20bit to become a commercial standard would require that the whole pipeline support it, from the recording studio/concert-hall to the mixing and post-processing, to the manufacturing, to the distribution, to the playback devices.    That's a huge investment and changeover for the teensy proportion of the population who can hear any difference under ideal conditions with the current standard.


I don't know about "huge investment". I don't know where you could go to buy an ADC that was only 44.1Hz 16-bit these days! Apart from bottom end consumer sound cards, such devices have disappeared over the last decade. Everyone is working with more bits (or less, e.g. 1, or 5 etc), and everyone is working at 48kHz or 96kHz. (or more!).

Cheers,
David.