psychotropic
Well-Known Member
hey stevie,
nothing in that passage suggests that the nyquist shannon ntheorem "does not work", "in real life". as ajinkya has explained, it works in several applications in engineering and this perfection that the wikipedia article claims it is not capable of, is not necessary.
Using the limitations of nyquist shannon to criticise CDs is about as meaningless as using THD figures to criticise tube amps. The fact is that it does not deteriorate the sound. If it did, then it would deteriorate the sound of all modern LPs as well, because of the reasons i've mentioned.
@Rajiv,
Digital LP's sound good/better than CD because,in many cases the full digital signal can be represented on the LP whereas it must be compressed to 16 bits for standard red book CD.
Rajiv, when it's digital it is always assumed that the signal is not 'complete', the question is whether the sampling frequency and the bit-rate are sufficient to cross the nyquist threshold and that's a yes/no issue. If it is sufficient, then we are not hearing compromised sound at all. The same argument can be held against 24bit audio as well, because digital at the end of the day is not 'complete.'
CDs mastered from these recordings must either downsampled, or be converted to analog and then back to digital. There is an inherent loss of fidelity with both of these processes.
A lot of recording/mixing/mastering these days happens "in the box" where it is digital from end-to-end through mastering and pressing. In the case of vinyl it has to go through a DAC before it is pressed to vinyl. I would assume that the DAC-ing process will have at least as much of a detrimental effect on fidelity as any variation of sampling rate may have. So this is not really a works-for-all criticism in the CD v. vinyl debate.
- With vinyl, there are many more user-adjustable components in the signal path (stylus/cartridge/tonearm/table/pre-amp) than CD. Each of these components can have a very dramatic effect on the sound. Vinyl enthusiasts can tune their systems to their liking via their choice of components. Take the same sound, and you'll get different responses from different people: one man's 'natural and neutral' is another man's 'flat and lifeless'.
This is why I posit that perhaps vinyl also adds a special sauce to the music. It's not a bad thing at all, but perhaps it is the absence of this special sauce and not the failure of the nyquist-shannon theorem that vinyl-heads don't like about CDs.
- The limitations of vinyl actually help make it easier on the ears. This is both in terms of EQ (i.e. excessive treble and out-of-phase low frequencies must be avoided to prevent severe mistracking) and sharp attacks and transients. Case in point: a pure square wave is harsh and nasty. You can't record it onto vinyl and have it still be a square wave, for it's not possible for the cutting head to instantly go from the top of the waveform to the bottom. The cutting process smooths it down a bit, and also makes it easier on your speakers (which, like your stylus, can't instantly move from top to bottom). CDs are limited only by the nyquist frequency. You can put a full-on square wave on digital, and hear the resulting nastiness.
And here again, one man's limitation is another man's detail, and the other man's limitation is the first man's 'easier on the ears'. It is conceivable that someone would want to hear every single accurate detail of the recording including excessive treble, out of phase low frequencies and any other gremlins or deviations from the norm. Since vinyl is a medium where a lot of this has to be sacrificed, would it not be valid to suggest that vinyl is a compromised medium from a fidelity perspective and therefore inferior to CD? (on much the same grounds that the vinyl-heads point the finger at nyquist). No I don't believe in the previous statement, but I am just trying to say that technical criticisms are sometimes backworked from a conclusion and need not really count for too much. At the end of the day it's a preference.
Usually mastering is digital,but not always. Digital mastering will often have analog stages, but again not always. 90% or more of new recordings are digital and the analog ones almost always have a digital source somewhere, a sample or digital reverb. Vinyl still needs to be cut and the dynamics of it and bass summing are going to be different than a cd, plus you are going to have the inherent sound of vinyl in the mix. Often they will master the vinyl from a higher res digital than 16/44.1 and you will get a bit of a different sound than the downsampled redbook.
Absolutely. Vinyl has the "inherent sound of vinyl" and this seems to be what the vinyl heads love. And that's great. It can be called musicality or emotion or any of these things, but it certainly doesn't sound accurate to call it superior fidelity as many vinyl-heads claim.
Really it's all a matter of what sounds good to you. Most everything is hybrid these days. I hope this helps.
Yes sir! agreed.
nothing in that passage suggests that the nyquist shannon ntheorem "does not work", "in real life". as ajinkya has explained, it works in several applications in engineering and this perfection that the wikipedia article claims it is not capable of, is not necessary.
Using the limitations of nyquist shannon to criticise CDs is about as meaningless as using THD figures to criticise tube amps. The fact is that it does not deteriorate the sound. If it did, then it would deteriorate the sound of all modern LPs as well, because of the reasons i've mentioned.
@Rajiv,
Digital LP's sound good/better than CD because,in many cases the full digital signal can be represented on the LP whereas it must be compressed to 16 bits for standard red book CD.
Rajiv, when it's digital it is always assumed that the signal is not 'complete', the question is whether the sampling frequency and the bit-rate are sufficient to cross the nyquist threshold and that's a yes/no issue. If it is sufficient, then we are not hearing compromised sound at all. The same argument can be held against 24bit audio as well, because digital at the end of the day is not 'complete.'
CDs mastered from these recordings must either downsampled, or be converted to analog and then back to digital. There is an inherent loss of fidelity with both of these processes.
A lot of recording/mixing/mastering these days happens "in the box" where it is digital from end-to-end through mastering and pressing. In the case of vinyl it has to go through a DAC before it is pressed to vinyl. I would assume that the DAC-ing process will have at least as much of a detrimental effect on fidelity as any variation of sampling rate may have. So this is not really a works-for-all criticism in the CD v. vinyl debate.
- With vinyl, there are many more user-adjustable components in the signal path (stylus/cartridge/tonearm/table/pre-amp) than CD. Each of these components can have a very dramatic effect on the sound. Vinyl enthusiasts can tune their systems to their liking via their choice of components. Take the same sound, and you'll get different responses from different people: one man's 'natural and neutral' is another man's 'flat and lifeless'.
This is why I posit that perhaps vinyl also adds a special sauce to the music. It's not a bad thing at all, but perhaps it is the absence of this special sauce and not the failure of the nyquist-shannon theorem that vinyl-heads don't like about CDs.
- The limitations of vinyl actually help make it easier on the ears. This is both in terms of EQ (i.e. excessive treble and out-of-phase low frequencies must be avoided to prevent severe mistracking) and sharp attacks and transients. Case in point: a pure square wave is harsh and nasty. You can't record it onto vinyl and have it still be a square wave, for it's not possible for the cutting head to instantly go from the top of the waveform to the bottom. The cutting process smooths it down a bit, and also makes it easier on your speakers (which, like your stylus, can't instantly move from top to bottom). CDs are limited only by the nyquist frequency. You can put a full-on square wave on digital, and hear the resulting nastiness.
And here again, one man's limitation is another man's detail, and the other man's limitation is the first man's 'easier on the ears'. It is conceivable that someone would want to hear every single accurate detail of the recording including excessive treble, out of phase low frequencies and any other gremlins or deviations from the norm. Since vinyl is a medium where a lot of this has to be sacrificed, would it not be valid to suggest that vinyl is a compromised medium from a fidelity perspective and therefore inferior to CD? (on much the same grounds that the vinyl-heads point the finger at nyquist). No I don't believe in the previous statement, but I am just trying to say that technical criticisms are sometimes backworked from a conclusion and need not really count for too much. At the end of the day it's a preference.
Usually mastering is digital,but not always. Digital mastering will often have analog stages, but again not always. 90% or more of new recordings are digital and the analog ones almost always have a digital source somewhere, a sample or digital reverb. Vinyl still needs to be cut and the dynamics of it and bass summing are going to be different than a cd, plus you are going to have the inherent sound of vinyl in the mix. Often they will master the vinyl from a higher res digital than 16/44.1 and you will get a bit of a different sound than the downsampled redbook.
Absolutely. Vinyl has the "inherent sound of vinyl" and this seems to be what the vinyl heads love. And that's great. It can be called musicality or emotion or any of these things, but it certainly doesn't sound accurate to call it superior fidelity as many vinyl-heads claim.
Really it's all a matter of what sounds good to you. Most everything is hybrid these days. I hope this helps.
Yes sir! agreed.
