Ask an Audio geek

[IATIA]

I will draw up some pictures and documentation this weekend. I was not arguing that digital cannot be good enough only that it cannot be a perfect replica of an analog signal. It can never be. Every time the clock goes positive one gets a sample. What is missing is what is between the clock pulses. 50% at best. A picture is worth a thousand words.

[AFTT47]

I will draw up some pictures and documentation this weekend.  I was not arguing that digital cannot be good enough only that it cannot be a perfect replica of an analog signal.  It can never be.  Every time the clock goes positive one gets a sample.  What is missing is what is between the clock pulses. 50% at best.  A picture is worth a thousand words.

Draw up all the pics you want: You're wrong. It should be obvious that you're wrong. How could we achieve a measured THD of a tiny fraction of a percent of we we're only capturing half the signal? Impossible.

First of all, consider it from a strictly mathematical sense. if it were as you say, doubling, tripling, quadrupuling the sample rate would do nothing. Even though we sample more and more data, we never go over 50%. Does that make the tiniest bit of sense to you? This is like saying that if you have a distance to cover, every step you take can only get you half-way there. You can never get more than half-way to your destination.

I don't know if you have a problem conceptualizing your error but it shouldn't matter. This is the beauty of science. You can infer it by logic. If it were as you say, digital sampling would be virtually useless. It would be impossible to achieve the measured results we get. It just doesn't work like that.

This is like evolution. You might not be able to make the connection between random, genetic mutations and the result that a dog-sized creature evolved into the horses we see today but you don't have to. The fossil evidence is there. Similarly, the objective measurements are there to prove that digital sampling works far better than you think. It's undeniable.

I don't understand the mathematical theorem that Alex mentioned myself. I'm aware of it though and I know it is a done deal. The universe doesn't care about my lack of ability to understand it. Digital audio sampling and playback goes right on being far superior to any analog technique ever developed. It does so just the same as evolution trumps creationism - regardless of my or anyone else's ability to understand it. So drawing up your pictures and documentation is like a creationist trying to prove the flood. We've done that already. This is fact, not opinion. Sixteen bit digital audio is measurably superior in every way to any common analog storage technique. I say "common" because I have no way of knowing if someone has a laboratory magnetic tape machine with one meter-wide tape with a speed of 50 meters/second.

[IATIA]

You will see.

[AFTT47]

The only thing I will see is if you accept reality like an atheist or deny it like a creationist. As I said, this is not a debatable subject. The facts are all in. Nothing new to see here.

[IATIA]

Apparently you are wrong.

[AFTT47]

Creationist then. I hope you get better.

[ignoramus]

Guys, it comes down to the psychoacoustic effect.
This is how/why we can compress 90% of the information out of a piece of music into an mp3 and it still sounds "normal".
Sony/Philips know this when developing the CD.

Maybe the digital info isn't a perfect mathematical equivalent of an analogue piece of music.
No-one really cares if the human ear cannot detect the difference.
Our brains also do a lot of the number crunching (psychoacoustic effect).

No different to cinema. They've always been just moving pictures, right?

[Alex K]

No no, it has nothing to do with the psychoacoustic effect (that's relevant for dropping in-between frequencies in mp3 compression), it has to do with the limited frequency range of hearing.

The point of the sampling theorem is that with a signal of limited highest frequency, these discrete points uniquely determine what goes on between them. If the signal you send in is cut off sharply at say 20 khz using a steep filter, it will not have any features which are so quick that they can be missed if you only measure 44100 times per second. Such a signal can only change its direction of up-down movement so fast, because changing faster would require containing higher frequencies. If you have ever seen what a low pass filter does to short pulses, it might become clearer.

[AFTT47]

No no, it has nothing to do with the psychoacoustic effect (that's relevant for dropping frequencies in mp3 compression). The point of the samplinh theorem is that with  a signal of limited highest frequency, these discrete points uniquely determine what goes on between them.

Compression was also used in a failed recordable format called minidisc in the mid 90s. It was a magneto-optical technology.

But yeah, CD records the whole thing. Some people seem to have a mental block in accepting it.

[Alex K]

Sure! Remember DCC?