[*Scotty*]

See representative image of a -90 16 bit  sine-wave from Stereophile.
 
Only three voltage transition levels are available to describe the sine-wave.
 It's kind of hard to figure out the amount that the waveform is distorted from the picture but it has to be a lot. The -90 sinewave looks a squarewave. This is the distortion I was referring to and it is real and not false a impression of rising distortion.
 As the signal level drops there are increasingly fewer bits available to completely describe the waveform hence the rising distortion level that tracks a falling signal level, there is an inverse function at work here.
Scotty

[orthobiz]

There! Feel better now Paul? 

Finally, recognition! Thanks,

Paul

[Elizabeth]

Thanks *Scotty* for the pic. I was going to comment on seeing pics like that and say how can anyone call that a musical waveform?
And yeah the real small signal stuff, the edges around all the music is what sucks in digital.
I live with it by using tubes to smooth that out a bit.

[InfernoSTi]

And I was a Digital Format Hater early on and used to quote people like Neil Young and others who were very vocal vinyl/analog proponents at that time.

Even Neil Young gets into Hi Rez these days....

A MESSAGE FROM NEIL YOUNG
May 17, 2011

2012 will be the year that record companies release High Resolution Audio. This is huge for our industry. Since the advent of the CD, listeners have been deprived of the full experience of listening. With the introduction of MP3s via online music services, listeners were further deprived.

The spirituality and soul of music is truly found when the sound engulfs you and that is just what 2012 will bring. It is a physical thing, a relief that you feel when you finally hear music the way artists and producers did when they created it in the studio. The sound engulfs you and your senses open up allowing you to truly feel the deep emotion in the music of some of our finest artists. From Frank Sinatra to the Black Keys, the feeling is there. This is what recording companies were born to give you and in 2012 they will deliver.

-- Neil Young

http://www.neilyoung.com/messagepage.html

[TONEPUB]

I think he was kidding.   

Just playing along...

[konut]

This is the distortion I was referring to and it is real and not false impression rising distortion.
   

Thank you Scotty, this is an excellent example. Now lets apply a little common sense. You're in one end zone at a football stadium, and I'm at the other. You say something in a normal speaking voice and I will probably be able to discern that you said "something". Now we apply 3% distortion to what you say. Am I going to be able to tell a difference? Not likely. This is digital. By shifting the distortion where it is least likely to be heard we take advantage of the peculiarities of the human auditory system. Vinyl does just the opposite. 

[*Scotty*]

The amount of distortion that can be tolerated and the amount or lack thereof, that is desirable are two very different quantities.
We live with the former and seek the latter.
Scotty

[Quiet Earth]

I think we all know how difficult it is to judge the true value of something based solely on analytical measurements or theory of operation. For example, if THD was really such a big deal, then why would so many of us still enjoy low powered tube amplifiers? Do you remember the THD wars of the 70's and 80's? Remember the ongoing promise of better fidelity with lower and lower published THD numbers? It didn't take us too long to realize that achieving vanishing low THD levels was no guarantee of a better music experience. In fact, it didn't guarantee much at all. It was mostly a marketing ploy. And so it goes with some of the digits in the digital world.

I can see why Wayner is miffed by the ongoing promises of superior sound quality with ever rising bit/sample rates. Where does it all stop before it is good enough??? I get that part of his rant. But I think most of us are smart enough to realize that its not just how many bits and samples you use, it's what you do with them and how you retrieve them. I doubt if most Hi-Res people enjoy sample and bit rate up-conversions just for the sake of getting the numbers up as high as possible. They want the original recording to be done in hi- res don't they? And there are lots of us old school red-bookers that are continually amazed at just how much music is really hiding in those lowly 16 bits at 44.1kHz. It can be mind blowingly good!

But like anything else in this life, you have to make a serious effort to get good results.

Try this nonsense on for size. If high Res folks are just vinyl wannabies, then vinyl folks must be reel to reel wannabies because that format is surely better than any electronically equalized and compressed analog record. My statement is obviously nonsensical too, just to make a point. We are all trying to have the best possible music experience that we can with the resources that are available to us. I think the more sources that we have access to, the better.

Now get yourself out there Wayner, and find yourself a nice transport and DAC. 

[Napalm]

There hasn't been any major breakthrough in sound *recording* for decades now.

We're obsessing with improving *storage*, but at this point it looks just like polishing a turd.

Nap.

[*Scotty*]

konut,you are going to have to explain where you put it,(ie, the distortion).
Looking at the Stereophile image it's fairly obvious that it is not being hidden.
 This not to say that 16bit has not been improved since it was first introduced.
Fortunately I can enjoy playback of 16 bit wave-files from my computer as much as I enjoy vinyl.
Obviously if I had a choice I would prefer to listen to the 24/96 master just as would like to listen to the master tape from which my records were made.
Scotty

[konut]

konut,you are going to have to explain where you put it,(ie, the distortion).
 
 

I though I did. Not hidden at all, just at so quiet a level as to be inaudible.

[*Scotty*]

What I would like to see are a series of THD measurements taken during CD playback of sinewave test tones taken every 10dB starting at negative 10dB and culminating in a measurement done at negative 90dB. The same series of measurements covering the octaves between 20Hz and 20kHz would also be interesting. This would take into account the effect of dither added to the signal and its effect on THD during playback.
 Regardless of the varying amounts of distortion tolerated by those of us participating in this hobby I don't believe anyone will make a serious argument in favor having more distortion in their system rather than the alternative of having less.
 At the end of the day distortion is not our friend.
Scotty

[konut]

What I would like to see are a series of THD measurements taken during CD playback of sinewave test tones taken every 10dB starting at negative 10dB and culminating in a measurement done at negative 90dB. The same series of measurements covering the octaves between 20Hz and 20kHz would also be interesting. This would take into account the effect of dither added to the signal and its effect on THD during playback.
 Regardless of the varying amounts of distortion tolerated by those of us participating in this hobby I don't believe anyone will make a serious argument in favor having more distortion in their system rather than the alternative of having less.
 At the end of the day distortion is not our friend.
Scotty

Agreed. The same measurements for vinyl would also be instructive.

[*Scotty*]

The vinyl would show an increasing ratio of noise to signal as the signal level dropped. The distortion would drop as the signal level dropped until the noise level swamps out the distortion.
The noise floor always limits how low a distortion measurement can be accurately taken. Now days analogue distortion measurements have to made taking into account the increased RF contamination that is prevalent.
 Unless special steps are taken this will swamp out your THD measurement because the RF noise floor of your environment exceeds that of the circuit you are measuring and the noise floor of your test gear.
 This is why in a meaningful THD spec you will always see it stated as THD plus N for noise.
Scotty

[dlaloum]

Has anyone considered the linearity issues with vinyl?

It is a different form of distortion...

The coils/magnets in the cartridge are effectively a "transformer" - and their abilities and efficiency vary with frequency.... they tend to be most efficient at lower frequencies.... and the efficiency is dropping off measurably by 10kHz...

Further you can "bias" a cartridge - the cartridges frequency response changes with the voltage going through it....

What this means is that the frequency response linearity of the cartridge is varying with 1) Frequency itself (level drops as frequency rises) 2) Self Biasing effects..... the voltage of the signal being replayed (the music) becomes a biasing voltage which in turn changes the frequency response of the cartridge.

And we haven't talked about basic frequency linearity of cartridges, and phase linearity...

Competing priorities...

How sensitive is the ear to THD?
How sensitive is the ear to Amplitude Linearity? (frequency response)
How sensitive is the ear to phase linearity?

In vinyl the result of the magnetic cartridge system is that the actual frequency response varies with level - it will be different at -20db and at -40db....
A frequency response that remains the same regardless of level, can be EQ'd, but what happens with vinyl is really a form of frequency & level sensitive compression/expansion....
(is this one of the signatures of the "vinyl sound"?)

Then there is distortion - 3% distortion at 15kHz is considered very good indeed for vinyl...
In fact I am currently looking at the distortion graph for an ATOC9 at Paul Miller audio.... the distortion rises above 2% at 3kHz by 10kHz we have 6% distortion rising to 14.5% at 20kHz.

And this is for a cartridge considered one of the better transducers, and a true budget high end contender.

The ortofon 2M Black (a current high end MM design) is marginally better (at around twice the price...) 2%@1.6k, 4%@10k and peak of 5.3%@15k before dropping back to 2%@20k

And these are all measured at -8db not at -80db......

Like I have said before - the fact that vinyl works at all is astounding - the fact that it works so well is absolutely miraculous.

Part of the reason why it works well is that it's flaws serendipitous match areas where the ear (and our psycho-acoustic way of processing what we physically hear) is least sensitive - and it tends to do well in areas where the ear is sensitive.

bye for now

David

[*Scotty*]

dlaloum,do you have any idea what the inherent distortion was of the test tones recorded into the grooves of the test record. One assumes that the initial test tone before cutting the lacquer master
was low enough to be considered adequate for measurement purposes,do they state the distortion floor of the record as part of its specs so that you can subtract it from the total distortion figure you measure? 
 Given the non-linearities of the phono cartridge it's a wonder a flat response curve is possible.
At least I have the basic phase problem due to coil inductance solved.
By any chance did you happen to measure the distortion produced by the AT440ML?
Scotty

[dlaloum]

Not the foggiest idea - Paul Miller audio do not advertise what test records they use!

But the distortion figures appear consistent with results published in golden age reviews where they were very careful to specify the test record used for each track....

Phase still bugs me.... LCR phase is easy enough - it is minimum phase and equalising with minimum phase filters will fix it - no worries! (RIAA EQ does just that...)

But cantilever resonance phase effects.....

LuckyDog on VE is adamant that the phase effects of cantilever resonance are minimum phase - there are documents out there that say that damped cantilever resonances are mixed phase. (if you search on mixed phase cantilever microscope.... it is from scientific work on electron microscopy)
His contention is that although the behaviour is not strictly minimum phase, it is so close to minimum that for our purposes in audio, we can treat it as minimum phase.

I'm just not comfortable with that!

I sort of got sick of measuring and decided to do a bit of listening for a while... so no distortion measurements, and I have some efforts still to come with respect to attempting to measure phase...

How did you resolve coil inductance phase?

bye for now

David

sitting at my desk, headphones on, listening to Holst the Planets.... delivered digitally from my music server, 16/44 flac, upsampled at my PC into M-Audio 2496 card then to my Firestone Audio cute curve Headamp...finally into a pair of 30 year old Koss Pro4x headphones - wonderful!

[Quiet Earth]

It didn't take us too long to realize that achieving vanishing low THD levels was no guarantee of a better music experience.

Guess I was wrong about that. Maybe only some of us gave up on the distortion wars.

I doubt if most Hi-Res people enjoy sample and bit rate up-conversions just for the sake of getting the numbers up as high as possible.

I guess I was wrong about that too. Perhaps audiophiles are more concerned with tests results than the way something actually sounds.

I sort of got sick of measuring and decided to do a bit of listening for a while...

There's hope yet ! 

[JohnR]

I think there's a difference between being rational and being obsessed.

[Edit]: let me rephrase that. I mean that statements like "Perhaps audiophiles are more concerned with test results than the way something actually sounds" just invoke this same old silly divide that does not, in my opinion, actually exist! The difference of opinion is in what sounds good, and how you might get there. But would you buy equipment designed by someone who eschewed the use of any kind of measurement equipment? (How they would be able to design it in that case is an interesting question.) Conversely, would you (knowingly) buy equipment from someone who didn't actually listen to it before putting it into production? No, of course not, in either case. (I hope.)

As for the original question (such as I can gather), my personal take is that I have vinyl and CDs and I am going to play them, whether anyone likes it or not. Buying them all over again would be silly. In time I expect to acquire high res digital material as a matter of course.

[woodsyi]

I would like to think I am obsessed with music and rationally go about playing them in all formats in which I find my favorite music.  Occasionally I will obsess over a gear but then a man can't be free of all vices, can he?