Digtal vs Analog; Why Isn't Digital Better? (long)

[Mallette]

No doubt this one will go on for a while! At least it is civilized so far...

Whether in the reproductive chain or in the media, the major problem with analog is the the fact that, so far, no one ever devised a way to store it or amplify it without storing and amplifying the container and pipes. By that, I am (very poorly) trying to say that when you play a tape, you hear the tape as well as the signal. When you play vinyl, you hear the vinyl as well as the signal. When you send the signal through analog circuits, you hear the circuits as well as the signal. When you copy an analog signal, you add the noise from the source medium to that of the new medium. Ad nauseum.

An earlier post mentioned the idea of making an LP of a CD and that it would not sound as good as the CD. No doubt. However, I have yet to hear an LP, even a 78, transfered to CD that sounded as good as the original properly played back.

As many of you know, I could go on. However, I'd prefer to introduce what I believe to be the answer. Current "standard" digital media are compromises to achieve file sizes for established distribution methods, media, and agenda. At least for audio, blu-ray and HDVD disks promise more than adequate space to play back completely untranscoded and uncompressed files. The media powers are gonna hate it and try to stop it, as Sony does with SACD, because it will allow mom and pops (like me) to distribute at will. The technology is in place and the high density media are beginning to flow into the marketplace. Unless the RIAA and such move to stop it somehow, disks with capacity to store and playback multiple channels of 24/196 or greater resolution will be available by the end of this year.

I stand ready to put disks in those trays...

Digital is not the problem, it is the solution. Once we eliminate the transcoding and compression, we can concentrate on getting the mikes in the right place and recording the right stuff.

Dave

P.S. Rather than add another post, I will just add to this one. To debate the "superiority" of digital vs. analog is actually pretty silly. It is the delivered result that counts. The first hundred years of recorded music was all analog, and digitizing will not improve it. It is what it is and will remain so. The next hundred years or whatever will be digital, and will be what it will be. When the Missouri class battleships were returned to service under Reagan, hundreds of millions were spent on modern digital battle systems. One thing that was not touched were the big gun computers...analog, and so accurate that no "improvements" could be justified. All things should be justified by results, not by the means or medium. As Duke Ellington said, if it sounds good, it IS good!

[sfogg]

Rick,

You added this after I read your posting..

"It represents what will be seen in a quiet passage at 16 bits resolution."

No, it doesn't represent that at all. Rip some of your songs with 'quiet passages' (places where music is playing at a low level) onto your computer and look at their level in any digital audio editing software.

You will find that those 'quiet passages' will be *far* above -90dB. Some of the most dynamic recordings have around 40dB of dynamic range. That means the 'quiet passages' will be recorded at around -66dB. That is a BIG difference from -90dB as far as signal integrity is concerned. Many CDs coming out today are lucky to have 10dB of dynamic range on them with the current vogue of ultracompressed 'loud is better' mindset.

And again, what happens at -90dB is irrelevant in the vast majority of most peoples systems. At -90dB this information will be totally swamped by the noise floor in just about everyones listening room.

Say someone sets their system up such that the peaks hit 105dB (which most would say is pretty loud). To be able to actually reproduce a recording that has 90dB of dynamic range the noise floor of their entire system (which includes their listening room) would have to be greater then 90dB. Which means the noise floor in their listening room would have to be below 15dB.

Almost no one has listening rooms anywhere near that silent. The typical household listening room is probably somewhere between 45 to 65dB noise floor. Those details recorded on the CD at -90dB would be 40dB *below* the noise of the room itself. IOW, those details are lost just by trying to reproduce them in the listening room among its background noises.

And again, to claim this is some huge problem for digital when analog playback can't even come close to the same playback at -90dB is pretty misleading.

Shawn

[sfogg]

"Then Shawn, why don't you explain why an analog recording sounds better than any digital format currently available?"

You forgot to include the "In my opinion" to that statement.

You are stating your personal preference, not a fact. What 'sounds better' to you may not to someone else.

You may prefer the analog and there is absolutely nothing wrong with that.

Others may prefer the digital and there too is nothing wrong with that.

Analog (vinyl) very much has its own 'sound' (or the dreaded coloration). They can be very pleasing and some people very much enjoy them. Digital doesn't have the same coloration and some people very much miss them.

To say nothing of the fact that in many many cases the people comparing a digital copy of a recording vs. the vinyl copy of the recording have no idea if they are actually made from the same mix or not. With two different mixes a person has no idea if the differences are due to the differences in delivery formats or due to differences in the mix.

It really is as simple as that.

Some people like horns, some don't. Some like SS, some don't. Some like tubes, some don't. Some like SETs, some don't. Some like surround sound, some don't. None of the above are 'facts', they are personal preferences.

Shawn

[3dzapper]

If I drew a picture of you, it is not you, it is a representation. Clearly not the original. The available images are all I have to work with.

You assume that Mr Nyquist is correct in his two sample theory. Practicality has proven him wrong. His two samples sound like dodo do do v analog thus the current 96KBS DSD/PWM formats sound better than Red Book although upsampling Red book is an improvement over 44.1KBS. Unless and until digital sampling achieves the compliance of a high quality phono cartridge, it will continue to do so.

[sfogg]

Rick,

"If I drew a picture of you, it is not you, it is a representation. Clearly not the original. The available images are all I have to work "

And a needle dragging its way through a bit of vinyl with grooves on the walls is not the original either. It is a representation too. One that can't give full bandwidth reproduction without enormous amounts of EQ that had to be built right into the system from the start.

"You assume that Mr Nyquist is correct in his two sample theory. Practicality has proven him wrong."

No, it hasn't. Nyquist is used in far far more then digital audio. If it was wrong an awful lot of things today wouldn't be working. You do realize that Nyquist isn't exactly new? It was put forth in 1928 and mathematically proven in 1949.

It has yet to be proven wrong.

Just because a person does not like the results of a system that uses it doesn't mean the Theorem isn't accurate.

"His two samples sound like dodo do do thus the current 96KBS DSD/PWM formats sound better than Red Book."

You mean all those albums that were remixed/remastered then put out on the new format? How do you know that the difference isn't in the remixing of the material? Ever heard a remastered CD that sounded better then the original CD version? How about different pressings of vinyl sounding different? Again, without knowing you are comparing identical mixes on two different formats you have no way of knowing if the difference is due to the mix or to the format.

Have you ever read any of the tests where 44.1/16 A/D-D/A were put in the middle of 'hi-res' feed to see if the listeners could detect it?

Ditto for the same thing done with vinyl playback? Such as outlined here...

http://www.bostonaudiosociety.org/bas_speaker/abx_testing2.htm

" although upsampling Red book is an improvement over 44.1KBS. "

DACs have been performing upsampling internally for well over a decade. All that buys you is the ability to place the low pass filter at a higher frequency so that it is further away from the audible range and so that it can be at a gentler slope. Those that think upsampling 'helps fill in the dots' don't understand how digital actually works.

Upsample a 1kHz signal recorded at -90dB and compare it to the original without upsampling on your scope and you won't see any smoother a signal. Not unless the unit was also adding dither or altering the signal in other ways.

"Unless and until digital sampling achieves the compliance of a high quality phono cartridge, it will continue to do so."

A digital audio system will *never* sound the same as vinyl playback. Not without artificially distorting things like inter-channel phase. BTW, that is why CD can sound 'flat' in comparison to vinyl. Vinyl has around 45 degrees of phase shift between the channels, it acts like a low level primitive form of inter-aural cross cancelation.... IE somewhat like what Carver's 'Sonic Holography' or Lexicon's 'Panorama' does. Carver built a box decades ago (called the Digital Time Lens) which intentionally altered digital playback in an attempt to mimics vinyls 'sound.' Although some thought Carver overdid the effect in that box a little too much to make it more obvious for instore demos.

Bottom line is you like what you like.

Shawn

[edwinr]

I did not like digital when it first came out. Until recently, my music has been largely concentrated in vinyl rather than CD's. But the last couple of years, I've noticed I'm buying more CD's and liking the sound.

Whatever the technical reasons may be, I think CD has got better. In my view, 2 channel SACD is another step closer to very good analogue. I've been less impressed with DVD/A. It'll be just my luck that the technical people and the manufacturers get CD right, then drop the thing for something new - and we start that big ole learning curve again.

[formica]

----------------

On 5/8/2005 9:43:08 PM sfogg wrote:

You are stating your personal preference, not a fact. What 'sounds better' to you may not to someone else.

You may prefer the analog and there is absolutely nothing wrong with that.

Others may prefer the digital and there too is nothing wrong with that.

Analog (vinyl) very much has its own 'sound' (or the dreaded coloration). They can be very pleasing and some people very much enjoy them. Digital doesn't have the same coloration and some people very much miss them.

----------------

I really do think it's a personal preference. Mike (DrWho) and Shawn are right about how many people enjoy the "vinyl" sound... but this doesn't indicate accuracy. And it is true there is nothing wrong with that.

Even simple measurements will show Digital's advantage when it comes to accurately reproducing the original signal versus vinyl. It's not due to the lack of accurate measurement equipment... but rather vinyl's KNOWN limitations. You shouldn't shoot the messenger if you don't like the message.

Don't worry about it, and enjoy your favourite recording... after all.... that's what it's all about.

Rob

PS: Another annoyance, which was barely mentioned, is the presence of dust and static that many people claim their vinyl collection doesn't have, but does. I know you get used to it, but it's very distracting if it isn't intentional.

PPS: my listening room has a normal noise floor of about 55 to 60db. 15db signal is inaudible.

[DrWho]

And for those of us that never grew up with the hissing and popping, it's even more distracting.

[Mallette]

"It'll be just my luck that the technical people and the manufacturers get CD right, then drop the thing for something new - and we start that big ole learning curve again."

Not a problem. DVD-A and CD are just storage media. Inherently, they are the same thing. I use the same skills to record DVD as for CD, no difference. What we WILL have very soon is digital storage media with enough capacity that will not require compression or transcoding to fit the music. That will be a major improvement on all fronts, and return the issue of getting it right to where it belongs...where the mike meets the wavefront.

Dave

[3dzapper]

"You are stating your personal preference, not a fact. What 'sounds better' to you may not to someone else.

You may prefer the analog and there is absolutely nothing wrong with that.

Others may prefer the digital and there too is nothing wrong with that."

Quite frankly you are wrong. I prefer the convience of digital and appreciate it's superiority over the medium it replaced, the cassette tape. My hope is that soon, digital can meet the promise that it made to us over thirty years ago.

"PPS: my listening room has a normal noise floor of about 55 to 60db. 15db signal is inaudible"

That's why we use amplifiers.

Rick

[Klewless]

"Nature is analog"

That is true.

That does not mean that any device conceived by mankind can capture that analog perfection.

However, I would be willing to bet a cup of coffee on MY OPINION about just how important this is to your ear/brain interpretation of that perfect analog sound.

Here goes. Take a perfect copy of a sound, if your could actually do such a thing, and run it through some kind of (perfect) chopping filter. This filter would simply kill any sound for a very short window of time.

Then increase the frequency of your chopping machine.

You will eventually reach a point where you cannot tell the difference between the original and your "chopped" sound. Your ear/brain will perform the duties of the digital DAC.

One could actually send this chopped sound to a speaker and it will fill in the gaps for you; no DAC required.

All this just makes my head hurt thinking about it.

We are blessed indeed to have systems which can do as well as they do.

[sfogg]

Rick,

""You are stating your personal preference, not a fact. What 'sounds better' to you may not to someone else.

You may prefer the analog and there is absolutely nothing wrong with that.

Others may prefer the digital and there too is nothing wrong with that."

Quite frankly you are wrong. "

How is the above thing you quoted from me wrong? I simple said you *may* prefer the analog, others *may* prefer the digital. Unless you think other peoples *personal* preferences are wrong I don't understand what you are trying to say.

"My hope is that soon, digital can meet the promise that it made to us over thirty years ago."

If you want it to sound the same as your vinyl setup you are going to continue to be left wanting.

""PPS: my listening room has a normal noise floor of about 55 to 60db. 15db signal is inaudible"

That's why we use amplifiers. "

To reproduce a signal recorded at -90dB above the noise floor in a room that has a noise floor of 55 to 60dB means the peaks in the playback would be above 150dB with an average playback level of on the order of 130dB or higher. That is above the threshold where that volume level will cause physical pain and hearing damage will be just about immediate. Literally that volume level would be about the same as sticking your head in a jet engine on takeoff at Green.

To say nothing of the fact that that playback level is about 30dB (a huge amount) beyond what a K'Horn can reproduce.

That is why in the vast majority of rooms and systems how a system handles playback at -90dB is utterly meaningless.

BTW, check out the signal to noise ratio of your phono stage. If the rest of the vinyl playback system was totally silent (which it isn't) that alone limits the dynamic range of vinyl playback. According to the specs on the Blueberry its phono stage has a SNR of about 70dB which is quite good for a phono stage.

The line stage has about 80dB SNR so just passing a CD signal through your pre-amp means what the CD does at -90dB is irrelevant as it will be lost in the internal noise/hum of the pre-amp itself.

Shawn

[psg]

Shawn,

Let me take a minute to say a few things:

- You should be commended for not taking Rick's bait. With just about anyone else, this would have been an all-out flame war a few posts ago. Lurkers who read your rational posts will be much more likely to be convinced by your calm explanations than by a flame war.

- It's great to see someone who understands Nyquist defend the digital side. Most people look at a 10 KHz sine wave and wonder how can 4 samples reproduce that? The math works. In my field, it gets used for sea-level and currents analysis (tides) and I use it to quantify turbulence and mixing rates from 100 Hz measurements of water temperature. I think the prrof of the misunderstanding is well summerized by Rick saying:

No matter how mant bits you add, the waveform will still be a semblance of the original until the sample rate reaches into the mega BPS range.

To him, the only way to reproduce a sine wave digitally is to sample it infinitely. All he sees are the increasingly smaller square staircases as we increase the sampling frequency or the bit rate. Most people don't understand the reconstitution of the original as a series of sine waves with different frequencies.

I'll give it a shot, for those who don't understand Nyquist. (This is not how Nyquist is usually explained in school). Say you have an analog square wave and you want to represensent it as a series of sine waves of various frequencies. You'll start with a sine wave with the same wave length as the square wave. Then you'll add a harmonic (2x) to start capturing the squareness of it, then a higher harmonic (3x) to get a better fit, etc. The square wave is decomposed into an infinite series of pure sine waves. Now, if you go in the opposite direction and start from a crude digital sample. It represents a basic sine wave and a series of higher harmonics. If you know that there's no content about some cutoff frequency, then you can filter them out of the digital squareness and leave only the frequencies less than half the sampling rate. They will be correctly reproduced.

Well, I gave it a shot...

[DrWho]

----------------

On 5/9/2005 7:45:32 AM Klewless wrote:

You will eventually reach a point where you cannot tell the difference between the original and your "chopped" sound. Your ear/brain will perform the duties of the digital DAC.

----------------

And this happens at a sample rate of twice the highest frequency to be played back

I think one of the major reasons that higher sample rates sound better is due to combination tones. If you have two frequencies, then you will also hear the sum and difference of those two frequencies (so you have f1, f2, f1 + f2 = f3 and f1 - f2 = f4). A flute for example has frequency information around 30kHz, which technically should be inaudible. However, combine this 30Hz with an 18kHz tone and you've got yourself a 12kHz subharmonic and a 48kHz harmonic. Granted the 48kHz in most likely inaudible, the 12kHz isn't. If you sample at 44.1, the 30kHz tone doesn't get reproduced so you no longer have the 12kHz. The end result? A grainy sounding flute. I find it very interesting though that analog equipment that can't even record at those higher frequencies doesn't introduce the graininess...I attribute it to the very steep filters they're using to make tje 44.1kHz sampling rate work up to 20kHz. Analog (even reel to reel) has a steady decline in the HF information, which tends to get swamped out with the distortion of the device.

I wanted to mention one more thing about the Nyquist thing that I don't think a lot of people realize. Whenever you try to sample a frequency higher than half the sampling rate, you end up with pieces of a waveform "randomly" scattered. I've got a picture in my textbook, I'll see if I can't reproduce it and upload it. It's just interesting to note that when the sampling rate isn't successful you don't end up with a bad representation, but rather you end up with a totally scattered wave form (which you will hear as pops and ticks...not a garbled tone).

[psg]

I don't think you will hear these frequencies as distinct sounds, but rather as low frequency beating of the amplitude of the main frequencies. This effect is digitised correctly.

As for sampling frequencies that are higher than half the sampling rate, that is called aliasing. it doesn't lead to a random effect of pops and ticks but rather to low frequencies. If I had a scanner close by I'd scribble an example...

[sfogg]

DrWho,

". If you have two frequencies, then you will also hear the sum and difference of those two frequencies (so you have f1, f2, f1 + f2 = f3 and f1 - f2 = f4). A flute for example has frequency information around 30kHz, which technically should be inaudible. However, combine this 30Hz with an 18kHz tone and you've got yourself a 12kHz subharmonic and a 48kHz harmonic."

Wouldn't this occur when the flute is playing in the real world? As such wouldn't that 12kHz subharmonic be recorded at the event and reproduced by Redbook CD or whatever you used to record it?

"...I attribute it to the very steep filters they're using to make tje 44.1kHz sampling rate work up to 20kHz."

Are you talking on the recording or playback side? On playback the analog filters aren't all that steep anymore and they are far above the audible range specifically because DACs up/oversample on playback.

I feed my system 48kHz material and the DACs oversample it by 8x to end up at 384kHz.

I feed my system 96kHz material and the DACs oversample it by 4x to end up at 384kHz.

If I could (S/PDIF transmitters in it don't support 192kHz) feed my system 192kHz material and the DACs would oversample it by 2x to end up at 384kHz.

The analog filter after the DAC is built around that oversampled signal. I don't know where they are actually placed in my equipment but in the reference material on the DAC chip itself they specify placing the analog filters at around 75kHz as I recall.

That is a far cry from the original non-oversampling DAC days when the analog filter would have to be around 20kHz.

Shawn

[Klewless]

What I love about this forum is that --- if your not careful you'll end up learning something. I try not to be too careful!

[jacksonbart]

Shawn

Thank you for the posts; they are very well written and educational IMO. Nice to see facts employed as opposed to just the typical stereo typical impressions of digital vs Analog mediums.

[TBrennan]

How much distortion and nonlinearity is caused by the mechanical action of a phono cartridge?

[D-MAN]

It seems to me that the issue is electro-mechanical vs. electro-mechanical-digital. In simplest terms, how much pure signal is actually there and can we get it back off the source material (cd or vinyl) in a less-damaged form?

Where vinyl is completely electro-mechanical, there is virtually no signal processing or converting going on. Probably about as "pure" as one can get, signal-wise, only one conversion applied (from the microphone). The signal from the vinyl is not actually a conversion, in the strictest sense. However in digital, the signal from the microphone (analog) is digitally converted, first and then is extracted via laser, and is then converted again to analog. I personally believe that in every conversion operation, the quality of the equipment doing the conversion and the resulting quality of the converted signal is rather indeterminate in that we have no control over it whatsoever. It just ends up on the source disk, take it or leave it.

The bottom line is in digital, some mechanical problems are reduced and other electrical problems are added. The benefit is that the drawbacks of environment (dust, static electricity, etc.) as compared to a vinyl rig is reduced, the overall dynamic capability is increased, but the trade-off is more electronics are involved. Anytime that more electronics are employed, the chances of signal degradation increases, depending on the quality, design and implementation, with the resulting cost increasing (looks exponential to me) the better the reproduction capability.

So the digital is a tradeoff in reducing environmental concerns and increasing dynamic range vs. a purely electro-mechanical signal that is intrinsically linked to environment with a dynamic limit.

It can be assumed that a signal that is converted in one or more ways will display artifacts that result however minute, and that the signal that is not subjected to a conversion process is therefore "better" in that it will not display any conversion artifacts. In other words, digitally converted signals are faximilies of the true analog signal, in a manner of speaking.

There is no sense in arguing about which is better in the end. It is JUST a matter of convienience vs. inconvienience, or environmental concerns (i.e., you won't want vinyl in your car) or how much dynamic headroom you think that you need.

It can even be resolved down to what type of music you listen to. So no arguing about that....

But I will forward this opinion: a cheap vinyl rig WILL sound better technically than a CD player of the same calibre. But if you want to talk about above $1500 cd players, which is the bottom limit (IME) of a "good" cd player, then the performance gap lessens quite a bit, and other considerations come into play.

Just my opinion, of course.

DM