[mitchellshow]

Also, since most everything is recorded digitally these days, there are almost no "true analog" mixdowns, which means you're pressing a stair-stepped digital frequency onto vinyl.

With the exception of wealthy purists (read: just Radiohead), that vinyl you're listening to is actually the digital recording pressed onto vinyl, not an analog recording pressed to vinyl.

[teilo]

There is no "stair-stepping" in digital recordings.

The sample-rate of a CD exceeds the effective resolution of even the best vinyl pressing. Furthermore, the Nyquist limit guarantees that it is impossible to perceive the sample rate of a CD, and even with a bit-depth of 16, you have 50% more dynamic range than vinyl is capable of.

Add to that the fact that a master is generally 96Khz, 24-bit.

[flavio81]

This is theoric considerations, not practical.

For frequencies from about 2Khz to the limit, the dynamic range of the LP is over 100dB. The quoted 60-66dB dynamic range is only because takes into account low frequency rumble.

CDs are mastered with 44KHz sampling rate which is now generally accepted as too low for high quality audio. Yes according to Nyquist, you could capture up to 22KHz with no problem with such sampling rate, but in real life you run into all sorts of problems related to filtering on both the ADC and the DAC. It is too low for implementing a good DAC and ADC.

As for the LP record, they can extend beyond 20KHz easily, in fact i have records with info up to 50KHz which can be then played back correctly -- CD4 quadraphonic records.

CD- quality audio, in real life use, introduces non linear, non-musical distortions that detract from the sound, while vinyl records mostly have the typical 2nd harmonic distortion and in controlled levels.

Furthermore, LP record distortion diminishes with program level (softer sounds are less distorted), while PCM DACs, by nature, do the opposite. This isn't nice to the ear.

[teilo]

> This is theoric considerations, not practical.

With all due respect, there is nothing remotely practical about the theoretical ability of any medium to record frequencies above 22Khz. Do you play your CD4 quadrophonic records to your pets? On what? Ribbon tweeters?

Furthermore, the ability to perceive dynamic range attenuates with frequency. So once again, you are back at the limits of human hearing, even if I accept what you say about the theoretical DR of vinyl. And what if what you say is true? Can you show me any recording with a dynamic range exceeding, say, 70dB?

It is also relevant that the dynamic range of vinyl is a function of the circumference of a track, and decreases as the record plays. It also decreases every time you play a record.

And what you say about DACs has not been true for many years. These days, even run-of-the-mill DACs employ oversampling.

[flavio81]

> And what you say about DACs has not been true for many years. These days, even run-of-the-mill DACs employ oversampling.

Oversampling does not solve the issue i have commented, which is that 44KHz is too low. Read about filtering at the ADC and DAC stages. Read about the pros/cons of using FIR-digital, IIR-digital, brickwall-analog filters at them. Oversampling with a FIR filter was already used on the second CD player on the market, the Marantz CD63 aka Philips CD-100 of 1980. All through the eighties all kinds of DAC+filter combinations have been used and now in 2017, despite any combination you use, higher-resolution (say, 96Khz sampling rate) audio sounds better. 44KHz isn't enough.

You see, what happens is as you approach the 22KHz limit, all sorts of nasty stuff will happen with the audio. No matter what filter you use, the upper octave (say 10-20KHz) will have any of the following ills:

- ringing - rippled frequency response - wild phase shifts. - pre-echo or post-echo

So, again, let me repeat: 44KHz isn't high enough. Using a higher sampling rate like 96KHz or more, allows to move such problems away from the audible range.

> With all due respect, there is nothing remotely practical about the theoretical ability of any medium to record frequencies above 22Khz. Do you play your CD4 quadrophonic records to your pets? On what? Ribbon tweeters?

Yes there is. What this means is that a system that can resolve up to 50KHz will have clearer reproduction of the 10-20KHz range, which is so important to add definition to the sound. "Clearer" as in "less artifacts and distortions."

> It is also relevant that the dynamic range of vinyl is a function of the circumference of a track, and decreases as the record plays. It also decreases every time you play a record.

What decreases on the inner grooves is the ability to record higher levels for the high frequency ranges (>14Khz). And even this is a limitation that is self-imposed by record cutting engineers to make the record easily playable on the cheapest kind of stylus, the 0.7mil spherical stylus.

Advanced stylus shapes (line-contact, shibata, hyperelliptical, Fritz Gyger FG70, JICO SAS) wouldn't have any problem with this.

[teilo]

Uh huh. And if you reverse the polarity of the neutron flow, all these problems go away.

Seriously, you’re just spouting audiophile mumbo-jumbo and confusing issues encountered during mastering with limitations of the final recorded media.

I’m done here. Arguing with audiophiles is pointless.

[flavio81]

This is not audiophile stuff, this is audio engineering stuff, that you would find on AES papers for example.

[Dylan16807]

> CDs are mastered with 44KHz sampling rate which is now generally accepted as too low for high quality audio.

According to who?

[flavio81]

According to a lot of music lovers and studio engineers, if you are within that scene.

[teilo]

Read “gullible audiophiles.”

[protonfish]

Digital waveforms are not stair-steps https://www.youtube.com/watch?v=cIQ9IXSUzuM

[jccc]

[absolutely sincere question]

If you did two simultaneous recordings of a performance -- one a modern digital and the other a modern analog -- what would be the appreciable difference between the two once pressed to vinyl?

I just reflexively assume that there wouldn't be much difference at that point. Unless you're thinking of the loudness-war compression that's done to digital, but that doesn't happen to the recording until you're making the CD or something else for digital distribution.

[protonfish]

There is not much difference, though digital can represent a wider range of frequencies than vinyl (higher highs and lower lows) so it can be said that it is generally higher quality. For more information, I think this is a fairly good objective comparison: https://skeptoid.com/episodes/4303

[cr1895]

>With the exception of wealthy purists (read: just Radiohead), that vinyl you're listening to is actually the digital recording pressed onto vinyl, not an analog recording pressed to vinyl.

I'm not even sure Radiohead do AAA. Even the new OKNOTOK is supposedly sourced and remastered from digital. Not that there's a thing wrong with it, supposing it's done well.

[flavio81]

The problem is not digital audio, the problem is CD- quality digital audio. It isn't as good as state of the art digital audio.