Sampling: Ana vs. Dig

Cpyder

reeltrouble1 wrote: »

let me try to rephrase, since the original recorded analog wave has an infinite number of samples it cannot ever be perfectly interpretated by any number of samples. the sampling rate of redbook is felt by many to be inadequate, I would suggest a reading and study of the work done by Mr. Robert Thomas.

The potential of the recorded analog vinyl signal is greater than the potential digital signal. A representation of the signal is by definetion not the signal.

RT1

Sorry to bring this thread out from the graveyard but I was reading over it and read this statement. This statement is overly simplified. If a signal was a straight line, it could be perfectly represented by sampling any two points. It's true that the more complex the signal, the higher the sampling rate needs to be. But you can't just say it has infinite points therefore you need an infinite sampling rate. That is simply not true. In fact, if you COULD perfectly band-limit a musical recording, you could perfectly represent it digitally. That's not an opinion, it's a fact. Although, in the case of music, you can't perfectly band-limit a signal and hence for this argument a recording wouldn't be perfect. For the sake of fairness even an analog recording such as a vinyl can't represent all signals. Perfect squarewaves for example cannot be played back with perfect reconstruction since they contain components with infinite frequencies and no needle can change position instantaneously.

turbopantera

You can get a little more complex than a straight line for sampling. At low frequencies, the 44K works ok since you may get 10 or 20 samples per cycle and the reconstructed wave will be pretty good. Now lets go up to 16khz or higher where you are getting harmonics and natural frequencies that make things like a cymbol or bell sound correct or even a transient hit from something like a tom drum or rim hit with drum sticks. At higher frequencies, you have fewer samples. Lets say 3. Three samples works well to reconstruct a sine wave because it follows well to analog filtering which is what you get after the DA converter. But lets say the 16khz signal was a square wave or a pulsed signal. The filter will try and round off the signal and not return the signal to the original sharp edged signal of the square wave. The result sounds flat, muddy, closed in. Much he way a poor quality amplifier would sound compared to a quality amplifier. You really need about a 5 to 10x sample rate to get close enough that the human ear has a hard time identifying the difference. Thus why DVD audio is sampled at a higher rate and more number of bits. 24 bit and 96khz or perhaps 192Khz yields a much better restoration of non sine wave signals. But what stinks about DVD audio is that it illegal to play it over a digital interconnect other than HDMI. Probably why it isnt too popular.

And you can try this at home boys and girls. If you have a signal generator that is capable of sine and square waves, you can sample them in a sound card, then play them back through the audio and look at the signal with an o-scope. If someone is interested, I can do this an post pics for various sample rates.

steveinaz

The "failures" of digital, I believe, are in the studio. As long as we have that variable to contend with, speculation will continue as to why many feel digital lacks. I've heard too many outstanding 16/44 redbook CD's that get it "right."

You can only truly judge a sources potential when it's been mastered correctly.

Many of us have grown up with distortion as an ingredient. With tubes, as with vinyl, it was pleasing distortion---but distortion, nonetheless. I grew up around a flabby, warm sounding RCA console that was heavenly in it's lumpy bass notes and smooth treble. This is really about our expectations of how we want our music reproduced, it's not about detail in the nth degree. I think this causes alot of confusion with audiophiles (so-called) as they struggle between what they want, versus what they know to be true. It's kind of like the teenager who listens to a rap CD on your rig and wonders "where did all the bass go?" he's accustomed to distortion, not bass as it truly sounds--so he feels your "reference" rig is lacking...

Could it be that many feel the same way about digital, for the same reasons?

madmax

steveinaz wrote: »

Could it be that many feel the same way about digital, for the same reasons?

That wouldn't make sense. Most all of us have long been used to digital, it is analog we have recently found. I can tell you the cheap analog in my past was something I wanted to forget. Nothing but snaps, crackles and noise.

sucks2beme

It's like digital photography. Or your HD TV. The CD technology was a
compromise of late 70's technology. The newer high sample rate
does a much better job. 44.1 is not as good as the analog signal.
True, there's pops and other lovely crap that comes out in analog,
but that aside, it's better than 44.1. New higher rate recordings
close the gap and if the industry ever adapts them, would likely kill the
digital VS. analog debate. But the ability to have great quality digital
recordings gives RICA a heart attack.
A good CDP or pc with a good DAC can sound very good.
That point I'd like to make. But I have no illusions of it being
as good as analog. It damned convenient, though.

CoolJazz

People working to close that gap are fooling with and even removing the filtering. It's pretty well know that the filtering is one of the biggest problems. Here is some interesting reading along those lines...

http://www.sakurasystems.com/articles/Kusunoki.html

CoolJazz

Fongolio

turbopantera wrote: »

But what stinks about DVD audio is that it illegal to play it over a digital interconnect other than HDMI.

No offense but that statement is somewhat misleading. My Pioneer DV-79Avi Sacd and DVD-Audion player connects to my AVR via six analog out interconnects and I know of no "law" prohibiting the use of any other digital connection to output dvd-a. I don't know why you would want to when the analog out sounds fantastic.

vc69

Fongolio wrote: »

No offense but that statement is somewhat misleading. My Pioneer DV-79Avi Sacd and DVD-Audio player connects to my AVR via six analog out interconnects and I know of no "law" prohibiting the use of any other digital connection to output dvd-a. I don't know why you would want to when the analog out sounds fantastic.

Hi Res Codecs (BD, HD-DVD)... yes. 24/96... no. No HDCP handshake is required to output bitstreamed data from a DVD-A (PCM, DD, DTS)

AFAIK, there is no law or copy protection covering DVD-A except the standard dcss.

Cpyder

turbopantera wrote: »

You can get a little more complex than a straight line for sampling. At low frequencies, the 44K works ok since you may get 10 or 20 samples per cycle and the reconstructed wave will be pretty good. Now lets go up to 16khz or higher where you are getting harmonics and natural frequencies that make things like a cymbol or bell sound correct or even a transient hit from something like a tom drum or rim hit with drum sticks. At higher frequencies, you have fewer samples. Lets say 3. Three samples works well to reconstruct a sine wave because it follows well to analog filtering which is what you get after the DA converter. But lets say the 16khz signal was a square wave or a pulsed signal. The filter will try and round off the signal and not return the signal to the original sharp edged signal of the square wave. The result sounds flat, muddy, closed in. Much he way a poor quality amplifier would sound compared to a quality amplifier. You really need about a 5 to 10x sample rate to get close enough that the human ear has a hard time identifying the difference. Thus why DVD audio is sampled at a higher rate and more number of bits. 24 bit and 96khz or perhaps 192Khz yields a much better restoration of non sine wave signals. But what stinks about DVD audio is that it illegal to play it over a digital interconnect other than HDMI. Probably why it isnt too popular.

And you can try this at home boys and girls. If you have a signal generator that is capable of sine and square waves, you can sample them in a sound card, then play them back through the audio and look at the signal with an o-scope. If someone is interested, I can do this an post pics for various sample rates.

I would love to see those results. If you have time, please go ahead. Otherwise I may do them in the coming week.