Analog Thinking in a Digital World
Mechanical analog playback systems and signal routing are very dependent on machines with tight tolerances and signals that are protected from electrical degradation. Digital systems are fundamentally different. Think of a digital representation of an audio signal as the blueprints that can be used to recreate the sound when reproduced. When so-called authorities denigrate “bits is bits” engineers and drift into the mindset of analog properties affecting the sound of a digital playback, they are holding onto beliefs that should have been abandoned ages ago. But I guess it helps sell magazines and brings eyeballs to websites.
The author of the piece on the sonic value of $2000 glass discs wants us to believe that in spite of acknowledged fact that both the glass disc and polycarbonate versions result in exactly the same bit stream, there is an audible difference in the reproduced sound. But his rationalization goes on:
“Getting back to glass-versus-polycarbonate CDs, it’s worth noting that polycarbonate can introduce optical distortions that affect the playback laser beam. Specifically, polycarbonate can cause a phenomenon called ‘birefringence’—a double refraction of the playback beam introduced by variations in the refractive index of the material through which the beam is passing. These variations in the refractive index are caused by localized stress on the polycarbonate introduced during injection molding of the disc. That is, the liquid polycarbonate didn’t flow properly into the mold, creating areas that introduce birefringence. Obviously, a glass-substrate CD doesn’t suffer from this problem.”
Does any of this technical explanation make any difference to the transference of a series of digital data bits from a disc to the internal workings of your player? No. He already stated that the bits that come from the “perfect” glass disc are the same as the ones from the standard disc.
Think of it this way…kind of like Morse code (which I learned as a boy scout and then refined under the guidance of my HAM radio operator father WA8RSL). I encode my message into a series of dots and dashes or short and long pulses. Using a flashlight or beep tone, I deliver the Morse code stream to a recipient who writes down an equivalent stream of short and long pulses or whatever. The new received message is informationally identical to the transmitted original but could be representationally somewhat different. The receiver’s handwriting or paper or the size of the dots and dashes might be different. But somehow the transmitted message is identical.
This is how digital information is conceptually passed from one place or format to another.
We audiophiles want to believe that disc treatments, special back-coatings, expensive cables, mechanical isolators, and disc “lathes” will deliver a better set of digital ones and zeros. It’s part of the passion and the hobby. I get it.
Finally, if I played the $2000 glass disc into the memory of my computer would the “magic” sonic improvement attributed to the glass surface translate to the copied file? The only thing being stored in the memory of the computer is the raw digital data and the clock.
There is no conspiracy in all of this. Analog is analog and digital is digital…separate and not equal.
All true, all true, but one thing we have also learned is that the digital audio signal is not 100% robust, actually more fragile than one wants to think. I have had instances of transfer where “there shouldn’t be any difference” simply wasn’t true. Jitter is usually the cause even if claims of “clean clock” are made. Transferring digital requires care just as in transferring analog signals; just a different type of care.
The places where the recording and playback chains can affect the sound are in the analog circuits before and after the digital transfer, and in the conversion circuits between digital and analog (I know that the conversion circuit on playback is called a DAC for digital to analog converter – what’s the analog to digital encoder called?)
There’s still room for differences in sound quality, but not in the purely digital realm.
Here’s a question to ask them: Does the kind of paper on which sheet music is printed affect the sound quality of the performance?
Phil, ADC or Analog to Digital Converter
Hello
I searched online and found this, http://www.theabsolutesound.com/articles/the-2000-cd-made-from-glass-1/, article which, in short, states that birefringence causes jitter and that this is measurable as well as audible.
Oh, I just noticed, this the article you reference from the so-called authority.
Is there some industry wide organization that could test all of this to come to a definitive answer?
And can they get this done before I order any more platinum shm-cd’s?!!
Sincere thanks for all the thought provoking posts!
Yes, the disk treatments are probably silly, but the attack is misplaced. It’s not about identical bit streams; it’s about fewer mistakes in reading the pits. In short, minimizing the need for Reed-Solomon error correction is why most people resort to special disks, coatings, etc.. In a world of 20 k$ cables, worrying about the CD media quality seems reasonable. Of course, it’s snake oil.
I’m surprised that disks aren’t numbered from 0001 to the last press from a stamp. I bet people would pay extra for then least-worn stamping.
And for the ultimate in fantasy comments, checkout the series
http://www.psaudio.com/pauls-posts/misdirection/
http://www.psaudio.com/pauls-posts/reading-clues/
http://www.psaudio.com/pauls-posts/downstream/
http://www.psaudio.com/pauls-posts/it-sure-aint-all-the-same/
Well, what a load of gobbledygook, but as Paul also pretends to be a novelist, I guess the line between his novels and audio is just too fine !
Hilarious is the last word here !
I’ll have to take a look…thanks.
How would you account for an explosion between the sender an receiver of the morse code message that interferes with not the sending but the receiving of the transmitted light signals? Wouldn’t that then require the receiver to calculate the missing information? How do we know that calculation is correct?
Admittedly I know nothing about electronics so I do not understand how intent is transmitted digitally with 1’s & zeros”? By that I mean: “HOW are you” or “how ARE you” or “how are YOU”. Is there also a time domaine variance in reading optical discs by different digital readers? Aren’t the nuances in the music performance dependent on the interpreter & their understanding & preferences of the musical performance? Doesn’t that lead to “who designed the DAC We are listening through?
My point is, while these times are extremely exciting imagine the state of “Audio Art” 50 years from today.
“The only thing being stored in the memory of the computer is the raw digital data and the clock.” – The raw digital data may be the same, but what about the clock data? I’ve heard little explained about the robustness of clock data. And what about systems that throw the clock data away and provide new super stable clock data? Are they now wrong?
All that aside for the moment, I’ve been using an old Yamaha DVD player to play my CD’s through a very, very moderate system using the digital output directly into the DAC equipped receiver. Because of the bits is bits discussion I brought in two more players and connected the digital outputs in the same manner as before, then listened; I used the same modest digital cable for all three. I was not terribly surprised to hear minor differences between the players. In fact with the most “modern” (youngest) of the players, a four year old OPPO, the sound was distinctly more open but also a little bright sounding on top. I was able to tame the brightness slightly by using a different digital cable.
I will not pretend to understand what is happening. All I know is what I hear and can describe. I also have no idea which is most “correct.” Make of this what you will (or nothing at all).
Mark, you said:
” The only thing being stored in the memory of the computer is the raw digital data and the clock.”
Actually, only the digital data is stored. The timing of each individual bit/byte is not.
Regarding Dennis’ and Preston’s concerns about data integrity:
The CD standard already allows for up to 220 “raw” (EFM) symbol errors per second off a CD as being acceptable, and modern manufacturing typically results in rates of 1 per second or less. The error correction copes with this range just fine. Reducing the raw error rate even further won’t make any difference to the accuracy of the resulting data.
Regarding jitter, the DAC must do two things: (1) For each sample, accurately produce the analogue voltage represented by the sample value; (2) Perform this function at precisely spaced intervals. It must do these in the face of external influences such as electrical noise and supply voltage fluctuations, and variations in the timing of the incoming digital samples (jitter). The state of the art is such that there is no excuse for a DAC not to be able to do this to a level of accuracy orders of magnitude better than the most conservative limits of audibility. If you hear a difference, the DAC is broken. The solution is not to fix the noise or jitter, but to fix the DAC.
To Preston’s question about conveying “intent” with 1s and 0s:
Do you know the art technique known as “pointillism”? Delicate gradations of shadow and colour (“intent”) are represented by tiny points (“bits”) of solid colour. If the points are fine enough, and you view at a normal distance, you see an apparently solid image. The artist’s intent is conveyed with 1s and 0s (colour and not-colour). It may help to think of digital audio in the same way. (Strictly speaking, DSD would be the nearest equivalent to pointillism in digital audio. The nearest analogy to PCM would be half-tone printing, where the dots form a regular pattern but can vary in size.)
Don, I may have misrepresented the inclusion of the clock information in the memory of the computer. The data is separate from the clock but the mechanism of I/O for data is a clocking function that exists in the computer or AD or DA.
A bizzaro world indeed–this hobby which uncomfortably tries to make bedfellows of unyielding physics and the emotional vagaries of music. Unfortunately the science of converting analog to digital and back again is just too much for many to understand leaving a vacuum to be filled by the audiophile degree-less high priests of our day who pedal non-sense theories and products to the masses.
Assume Paul has discovered a deficiency in poly disk production. Does the conversion to glass then make the digital process perfect, or are there yet other problems? If bubbles are a problem on CDs, I take it that music on a “stick” ought to sound better than music on a CD. Can you tell, really?
Past years have produced every possible “improvement” in sound from using “polarized” copper in cables to painting CDs and everything in between. What is invariable lacking is the scientific method. No one takes a hearing test to calibrate their ears, measurements are eschewed, double-blind tests are nearly non-existent, but the subjective adjectives roll off the tongue like water; openness, tight, loose, spacious, moving, enthralling, mid-range-y, warm, cold and borrowings of every sort. I would just like these gurus to precede their proclamations with “I believe.”