Vinyl is equivalent to which digital bit-depth and sampling rate? Options

[KMD]

What sample rate and bit depth , dithered, would create a digital recording with the same bandwidth and signal to noise ratio as vinyl

[dhromed]

Somewhat below Redbook.

[Kees de Visser]

This old thread gives some answers:
http://www.hydrogenaudio.org/forums/index....showtopic=35530

[KMD]

Ok so I'm getting 44Khz 12 Bit from there. People are happy listening to Vinyl but would they be happy listenning to 44/12Bit Digital ? This is relevant to the formats discussion. Would anyone like to convert a CD to 44/12 dithered so people on the forum can hear what it sounds like.

[pdq]

There was a listening test here a few years back. Someone took a 44/16 clip and converted it, with dither, to 15, 14, etc. bits. As I recall, most people could distinguish 11 bits, some 12 bits, and very very few could tell 14 or 15. (I could be wrong about the details, but I'm too lazy to search for the thread).

[2Bdecided]

Ok so I'm getting 44Khz 12 Bit from there. People are happy listening to Vinyl but would they be happy listenning to 44/12Bit Digital ? This is relevant to the formats discussion. Would anyone like to convert a CD to 44/12 dithered so people on the forum can hear what it sounds like.

People have been listening to 32kHz 14-bits companded to 10-bits throughout the UK and much of Europe for years...
http://en.wikipedia.org/wiki/NICAM

lossyWAV often keeps less than 12-bits, without audible degradation...
http://wiki.hydrogenaudio.org/index.php?title=LossyWAV

That thread linked by Kees de Visser almost has it nailed. You'd need some weird pre/de-emphasis and maybe companding or noise shaping to make the digital as bad as vinyl at the frequency extremes. Even then, the digital would still be distortion-free (unless you clip).

Cheers,
David.

[KMD]

The question is do very low signal to noise ratio digital formats sound the same as analogue formats of the same signal to noise ratio

This post has been edited by KMD: Mar 13 2012, 15:04

[2Bdecided]

The question is do very low signal to noise ratio digital formats sound the same as analogue formats of the same signal to noise ratio

Not unless you shape the noise to match. And even then, no, because analogue will have distortion and clicks/pops that the digital won't.

The lispy tracing distortion of vinyl is generally unwanted, but some people like the minor harmonic (they may say euphonic) distortion and other colourations of particular playback equipment.

Cheers,
David.

[KMD]

If people think vinyl LP sounds better than 12 Bit dithered digital then the signal to noise based analysis of HD audio that is being discussed on the other thread is not sufficient. It would be interesting to hear an 8 bit dithered file compared to a 16 Bit file with analogue noise added to make the signal to noise ratio the same as the 8 bit file.

[2Bdecided]

If people think vinyl LP sounds better than 12 Bit dithered digital..

Who said this? All the comments above are to point out the opposite.

Cheers,
David.

[John_Siau]

What sample rate and bit depth , dithered, would create a digital recording with the same bandwidth and signal to noise ratio as vinyl

DSD demonstrates that 1 bit is sufficient to achieve a SNR that is much better than vinyl. The answer is that the word length must be at least 1 bit.

The bandwidth question is not quite as straightforward. 44.1 kHz is the minimum sample rate that will fully reproduce all audible frequencies while providing some room for the filter transition band. The answer is that the sample rate must be at least 44.1 kHz.

The combination of sample rate, bit depth, and dithering method will determine the what combinations produce sufficient SNR within the audio band. A 1-bit system at 44.1 kHz would produce a 6 dB SNR at best. For every bit-depth, and dithering method, a sample rate can be chosen to produce the desired SNR. If we choose the minimum sample rate, 16-bits are more than sufficient to capture the full SNR of the vinyl recording.

Please note that 44.1/16 is capable of delivering all frequencies up to 20 kHz at full amplitude. Vinyl disks cannot deliver high frequencies at full amplitude. For this reason, high frequencies were always limited when mastering vinyl. 44.1/16 has far more dynamic range at high frequencies than can be recorded on a vinyl record. The engineer can create a brighter mix on a CD, and this is often done.

The source material recorded on vinyl records must be mastered to fit the physical limitations of the format. The physics of the cutting head and the playback needle put significant constraints on the amplitudes that can be recorded at various frequencies. For this reason, vinyl masters are usually mixed differently than CD masters.

This post has been edited by John_Siau: Mar 13 2012, 15:44

[KMD]

listening tests at super high quality are hard to perform and the target differences are minute and thus lead to argument and uncertanty. But if the 8bit dithered file and the 16 bit file with analogue noise added have a different quality to the listener then it follows that a signal to noise based analysis alone to define the optimim quality of digital is not sufficient. It would be informative to both sides of the discussion. Then go to 4 Bits dithered. At some point a particular quality of the digital file will or will not be revealed. If it exists at low quality then it follows that it exists at high quality and if not why not.

[Rotareneg]

If it exists at low quality then it follows that it exists at high quality and if not why not.

I agree, the existence of heavily scratched and warped gramophone records clearly indicate the obvious poor quality of phonographs in general.

[krabapple]

listening tests at super high quality are hard to perform and the target differences are minute and thus lead to argument and uncertanty. But if the 8bit dithered file and the 16 bit file with analogue noise added have a different quality to the listener then it follows that a signal to noise based analysis alone to define the optimim quality of digital is not sufficient. It would be informative to both sides of the discussion. Then go to 4 Bits dithered. At some point a particular quality of the digital file will or will not be revealed. If it exists at low quality then it follows that it exists at high quality and if not why not.

Are you actually reading the answers you're getting on this thread?

[Arnold B. Kruege...]

listening tests at super high quality are hard to perform and the target differences are minute and thus lead to argument and uncertanty.

I would take that difficulty and uncertainty as a message from Mother Nature: There is very little here to hear.

But if the 8bit dithered file and the 16 bit file with analogue noise added have a different quality to the listener

Which is like saying that since battery acid and soda pop are both acid... The error here is excluding both the middle and also the fact that we are comparing extreme opposite ends of a spectrum of stimuli.

then it follows that a signal to noise based analysis alone to define the optimum quality of digital is not sufficient.

That's a truism because SNR does not include the effects of linear and nonlinear distortion.

It would be informative to both sides of the discussion.

This sort of thing is only informative to people who haven't done their homework.

Then go to 4 Bits dithered.

That's a non sequitor and it is also just plain nuts.

At some point a particular quality of the digital file will or will not be revealed.

Yet another trivial truism.

If it exists at low quality then it follows that it exists at high quality and if not why not.

You seem to need education on the concept of the law of diminishing returns and the facts of the thresholds of reliable human perception. You want links to help you get started on your homework, or what?

[Arnold B. Kruege...]

What sample rate and bit depth , dithered, would create a digital recording with the same bandwidth and signal to noise ratio as vinyl

Maybe 11 or 12 bits, and maybe 35 or 40 KHz.

This is a little hard to say exactly because the digital performance limits implied are hard walls, while the limits of analog are sort of like padded walls. However, the analog room is very much smaller and can never be enlarged. The digital room can be made as large as you want to pay for.

For example, while a LP can be made with 35 KHz bandwidth, it can't carry much data at that frequency and the data is highly corrupted not to mention difficult to play usefully and easy to destroy during playback. The LP format starts loosing dynamic range and adding audible distortion beginning around 5 KHz. Vinyl has severe power bandwidth problems at both ends of the audible spectrum.

Vinyl also has massive inherent jitter, and is typically made from analog tape recordings that have relatively massive jitter. The best analog tape recorder ever sold as a commercial product has about 1,000 nanoseconds or a million picoseconds of jitter on the best day of its life, and needs frequent (weekly, daily) maintenance to perform that way. Every recording that any regular consumer ever listened on vinyl was much worse.

A comparable digital recording is scrupulously clean very close to Nyquist, has full bandwidth at high levels, is very easy to recover fully, does everything right down to DC, has inherent jitter that can be reduced arbitrarily, and is very durable.

[icstm]

^^^ so isn't quite funny that all those ppl who were happy with Vinyl are now seeking better than CD sound quality through their SACDs and the general discussion on >16/44 formats!

[greynol]

While CDDA only goes to 22k, I don't believe it has been demonstrated to be insufficient for non-placebophile consumption.

[Porcus]

jitter

Is that synonymous to what we used to call wow & flutter, or is there more to it?

[Nessuno]

^^^ so isn't quite funny that all those ppl who were happy with Vinyl are now seeking better than CD sound quality through their SACDs and the general discussion on >16/44 formats!

On the contrary: it is perfectly coherent from vinyl worshippers to praise something that, to theirs eyes, pushes digital signals further towards the "infinite resolution" of analogic...

This post has been edited by Nessuno: Mar 22 2012, 15:54

[pdq]

jitter

Is that synonymous to what we used to call wow & flutter, or is there more to it?

Yes, wow and flutter are both jitter, just in different frequency ranges.

[Arnold B. Kruege...]

jitter

Is that synonymous to what we used to call wow & flutter, or is there more to it?

Wow, flutter, and jitter are all examples of FM distortion.

[Glenn Gundlach]

jitter

Is that synonymous to what we used to call wow & flutter, or is there more to it?

Wow, flutter, and jitter are all examples of FM distortion.

Jitter is comparatively easy to process out. Wow and flutter more difficult since there is no timing reference in the signal.

G²

[Woodinville]

jitter

Is that synonymous to what we used to call wow & flutter, or is there more to it?

Wow, flutter, and jitter are all examples of FM distortion.

Jitter is comparatively easy to process out. Wow and flutter more difficult since there is no timing reference in the signal.

G²

Except for being able to sense the bias signal and resample that out of a tape signal, or analyze the signal for periodic jitter components for an LP.

[DonP]

Jitter is comparatively easy to process out. Wow and flutter more difficult since there is no timing reference in the signal.

If the jitter happened in A/D, what is your timing reference?