The Audiophile Debate

[AnotherSpin]

There is a difference between people who listen critically and those who don't. Thing is, all of the objectivists are as critical in their listening as the subjectivists. Their ears are golden, too. There's no skill advantage for subjectivists, either.

In the case where the audeo set is too bad, listening becomes critical. It takes some level of fidelity of the playback system to stop paying attention to sound quality.

[AnotherSpin]

Yeah, it is a bit more complex than that, but I simplified things for a reason. What I said is the stuff that counts.

Agree. I'm all about simplicity too, which is why I strive to remove wrong or bad quality components from an audio system.

[71 dB]

Agree. I'm all about simplicity too, which is why I strive to remove wrong or bad quality components from an audio system.

Billionaires can remove and replace "wrong or bad quality components" with stuff made of diamonds, gold, snake oil and unicorn farts of course, but us normal people need to be more strategic and spend money where it counts the most.

Do you think expensive high end audio cables are used in studios? One would think studios where the music of the biggest stars in the World or the soundtracks of 300 million dollar movies are mixed should have top notch gear and in fact they do, but the cables are quite basic professional cables. Mechanically robust (because the installation has to be reliable), but no snake oil is needed. It is not that the people mixing there don't hear the finest differences in sound if there is any. Those people have really really good analytic hearing. They also have education and know about place effect and the importance of blind testing. How about you?

[AnotherSpin]

Billionaires can remove and replace "wrong or bad quality components" with stuff made of diamonds, gold, snake oil and unicorn farts of course, but us normal people need to be more strategic and spend money where it counts the most.

Do you think expensive high end audio cables are used in studios? One would think studios where the music of the biggest stars in the World or the soundtracks of 300 million dollar movies are mixed should have top notch gear and in fact they do, but the cables are quite basic professional cables. Mechanically robust (because the installation has to be reliable), but no snake oil is needed. It is not that the people mixing there don't hear the finest differences in sound if there is any. Those people have really really good analytic hearing. They also have education and know about place effect and the importance of blind testing. How about you?

Fortunately, I'm very far from being a billionaire. I have very limited resources, and for many years I have been trying to remove everything unnecessary from the sound within the limits available to me. My current system is quite modest, not a single diamond used, as far as I know, and I like the sound. Eventually, I got to the point where I could give attention not to the imperfections of the playback, but to the music. It was not easy, it took a lot of trial and error, but I am satisfied with the result.

[71 dB]

Fortunately, I'm very far from being a billionaire. I have very limited resources, and for many years I have been trying to remove everything unnecessary from the sound within the limits available to me. My current system is quite modest, not a single diamond used, as far as I know, and I like the sound. Eventually, I got to the point where I could give attention not to the imperfections of the playback, but to the music. It was not easy, it took a lot of trial and error, but I am satisfied with the result.

Imperfections aren't much of an issue in the 21st century. We are not dealing anymore with phonographs  and shellac records. Digital audio matured to audible transparency quarter of a century ago etc. What imperfections we still have are connected to speakers, headphones and room acoustics, but affordable speakers and headphones can provide quite satisfactory sound. For me the impecfections are elsewhere, in the production, mixing and mastering of music. Whenever the sound is bad for me, it is those things to blame.

[AnotherSpin]

Imperfections aren't much of an issue in the 21st century. We are not dealing anymore with phonographs  and shellac records. Digital audio matured to audible transparency quarter of a century ago etc. What imperfections we still have are connected to speakers, headphones and room acoustics, but affordable speakers and headphones can provide quite satisfactory sound. For me the impecfections are elsewhere, in the production, mixing and mastering of music. Whenever the sound is bad for me, it is those things to blame.

In general, I liked the sound of LPs in the 20th century much more than the digital sound of the 21st. The main problem with LPs was that they could easily be damaged mechanically. CDs and digital files eliminated the problem of mechanical damage, but in return they provided lifeless sound.

[71 dB]

In general, I liked the sound of LPs in the 20th century much more than the digital sound of the 21st. The main problem with LPs was that they could easily be damaged mechanically. CDs and digital files eliminated the problem of mechanical damage, but in return they provided lifeless sound.

What you or me "like" tells nothing about transparency. You like the distortions introduced by LP format and because CDs and digital files lack those distortions (are more transparent), you like them less and find the sound lifeless. Nothing prevents digital formats having same distortions as LP, but you have to generate them separately to the mix/master. The problem is in the production, mixing and mastering as I said.

[AnotherSpin]

What you or me "like" tells nothing about transparency. You like the distortions introduced by LP format and because CDs and digital files lack those distortions (are more transparent), you like them less and find the sound lifeless. Nothing prevents digital formats having same distortions as LP, but you have to generate them separately to the mix/master. The problem is in the production, mixing and mastering as I said.

Are you saying that digitizing an original pre-80s analog recording makes it more "transparent" and sounds better? I prefer analog, just as I prefer analog photography rather than its digitization. Likewise, I would prefer real leather hand stitched shoes to plastic ones, and a natural cotton shirt to synthetic one. Even if leather and cotton are less practical.

[71 dB]

Are you saying that digitizing an original pre-80s analog recording makes it more "transparent" and sounds better? I prefer analog, just as I prefer analog photography rather than its digitization. Likewise, I would prefer real leather hand stitched shoes to plastic ones, and a natural cotton shirt to synthetic one. Even if leather and cotton are less practical.

Of course I am not saying that. I'm saying the digitized version of an analog recording sounds identical because it is audibly transparent (doesn't change anything).

[AnotherSpin]

Of course I am not saying that. I'm saying the digitized version of an analog recording sounds identical because it is audibly transparent (doesn't change anything).

Identical to what? Help me understand. I recorded something on tape. Then I digitized it. Does a digitized copy sound more identical than the analogue original? We probably understand the word "identical" differently.

[Irons]

In general, I liked the sound of LPs in the 20th century much more than the digital sound of the 21st. The main problem with LPs was that they could easily be damaged mechanically. CDs and digital files eliminated the problem of mechanical damage, but in return they provided lifeless sound.

hear! hear!

[71 dB]

Identical to what? Help me understand. I recorded something on tape. Then I digitized it. Does a digitized copy sound more identical than the analogue original? We probably understand the word "identical" differently.

Identical to what you digitize of course. If you digitize a tape, it will sound audibly identical to that tape for all the good and bad. Of course if you keep playing the tape it will wear off and start to sound worse and worse in time, but they are identical at the moment of digitizing. For me the word "identical" is well defined. I don't know why you struggle with it this much. Sure, English isn't your native language either, but the concept of something being identical to something else isn't that hard. Surely Ukrainian language has also such Word? According to Google translate the word is Ідентичні. Very close to English (just written in cyrillic). Even the Finnish word for "identical" is almost identical (pun intented ): Identtinen.

[AnotherSpin]

I struggle to understand how can you believe that a digital copy is no different from an analogue original and both are identical. Would you believe that a plastic 3D copy of Michelangelo's David is identical to the marble original?

[Daverz]

I struggle to understand how can you believe that a digital copy is no different from an analogue original and both are identical. Would you believe that a plastic 3D copy of Michelangelo's David is identical to the marble original?

You're getting hung up on the word "identical".  It's clear to me what is meant: transparent enough that there are no audible differences with proper testing (outputs gain matched with a voltmeter, randomized A/B trials).  If differences remain then there is probably a problem with the ADC, the recording method (e.g. clipping), or an impedance mismatch.

[71 dB]

I struggle to understand how can you believe that a digital copy is no different from an analogue original and both are identical. Would you believe that a plastic 3D copy of Michelangelo's David is identical to the marble original?

Do you also struggle to understand how you can pay your groceries with credit card or cash? Just as paying for groceries using whatever paying method is just a sign of transfer of monetary value from you to the grocery store, digital or analog recordings are just the information/data/signal for how to create the music. The music as sound waves is created by your speakers or headphones based on the information provided by your music recording. Your speakers and headphones have no way knowing whether the information comes from a vinyl LP, a C-cassette, a mp3 file or a CD. Electric current is received from the amp and it makes the drivers to move.

As for how I can believe how the digitazed version is no different from the original analog version, I don't need to believe anything, because I know they are identical. How do I know it? Because I have education to know and understand these things. I studies electric engineering specializing to acoustics and signal processing in university. If such education doesn't give insight into these things, I don't know what does...

[Spotted Horses]

In general, I liked the sound of LPs in the 20th century much more than the digital sound of the 21st. The main problem with LPs was that they could easily be damaged mechanically. CDs and digital files eliminated the problem of mechanical damage, but in return they provided lifeless sound.

I struggle to understand how can you believe that a digital copy is no different from an analogue original and both are identical. Would you believe that a plastic 3D copy of Michelangelo's David is identical to the marble original?

That is a terrible analogy (pun not intended).

All audio starts analog. The microphone is typically a parallel plate capacitor (condenser in UK usage) where one of the plates is a flexible diaphragm. Sound waves cause the diaphragm to move, changing the capacitance and causing the voltage across the capacitor to vary. That voltage variation is proportional to the sound pressure level, and is an analog of the sound pressure.

You could send that voltage, amplified by appropriate electronic circuits, to a speaker, which causes the speaker cone to move according to the signal, reproducing the original sound. You can't get more live than that.

Now you want to record it for later playback. You sent the signal to a device that measures the voltage hundreds of thousands to times per second, with accuracy more than one in ten million, and store the numbers. Later you send the numbers to a device which performs he inverse function, producing an output voltage from the stream of numbers, updating the output more than a hundred thousand times per second with accuracy better than one in ten million. The output signal is indistinguishable from the input. Why does the output have any less life than the original input? Doesn't the life come from the sound, and the microphone signal it produces?

Now you do it a different way. You send the signal from the microphone to a cutting head which etches a groove in a rotating plate, where the wiggles of the groove correspond to the signal. They you make in impression of the plate and press it into vinyl. You play back with a stylus (a diamond chip) that rides in the groove and is connected by a reed-like cantilever to a magnet. The jiggling of the magnet produces a voltage in a pickup coil which is amplified and sent to the speaker. But on a fine scale vinyl is rough, producing background noise. The stylus has mass, it takes force to make it conform to the groove, which may cause it to bounce around in the groove, ride up the walls, and shave the sides of the groove. The cantilever is not perfectly rigid, it flexes and vibrates as it tries to transmit the motion of the stylus to the magnet. The cantilever has to hold up the tone-arm, which similarly may bounce around in response the large amplitude segment of the signal. The signal in the pickup coils is very weak and must be amplified with high gain, causing electronic noise to encroach on the signal. The output signal, viewed on an oscilloscope or analyzed by scientific equipment, is obviously different from the original, affected by background noise, distortion of various types. How does this degradation of the signal add "life?" Isn't all of the life present in the original microphone signal, which the digital recording and playback process preserves?

And what 71dB was getting it is that if you like the sound of an LP (that is, you like the obvious distortion) you can make a digital recording of the LP playback, and it will sound exactly like an LP, with the surface noise, tracking error, harmonic distortion, rendered with exquisite fidelity.

There are digital recordings that sound awful and sterile to me, but it is not because the digital process suppresses the life of the signal. It is because produces have access to equipment which allows them to record with dozens, maybe hundreds of microphones and mix the sound artificially. This fails th capture the essence of the sound the way a minimal set of microphones in a good recording venue can.

[DavidW]

I struggle to understand how can you believe that a digital copy is no different from an analogue original and both are identical. Would you believe that a plastic 3D copy of Michelangelo's David is identical to the marble original?

Both analog and digital recordings are just facsimiles of the real deal.  The original isn't the analog recording, it was the actual performance.  Which format has higher fidelity seems to be a silly debate, when so much of the sound isn't captured perfectly by the microphones, and then you play it back in a room with completely different acoustics.

Just listen to whatever format sounds good to you and enjoy what you have.

[71 dB]

Just listen to whatever format sounds good to you and enjoy what you have.

That's a good advice. 

[AnotherSpin]

That is a terrible analogy (pun not intended).

All audio starts analog. The microphone is typically a parallel plate capacitor (condenser in UK usage) where one of the plates is a flexible diaphragm. Sound waves cause the diaphragm to move, changing the capacitance and causing the voltage across the capacitor to vary. That voltage variation is proportional to the sound pressure level, and is an analog of the sound pressure.

You could send that voltage, amplified by appropriate electronic circuits, to a speaker, which causes the speaker cone to move according to the signal, reproducing the original sound. You can't get more live than that.

Now you want to record it for later playback. You sent the signal to a device that measures the voltage hundreds of thousands to times per second, with accuracy more than one in ten million, and store the numbers. Later you send the numbers to a device which performs he inverse function, producing an output voltage from the stream of numbers, updating the output more than a hundred thousand times per second with accuracy better than one in ten million. The output signal is indistinguishable from the input. Why does the output have any less life than the original input? Doesn't the life come from the sound, and the microphone signal it produces?

Now you do it a different way. You send the signal from the microphone to a cutting head which etches a groove in a rotating plate, where the wiggles of the groove correspond to the signal. They you make in impression of the plate and press it into vinyl. You play back with a stylus (a diamond chip) that rides in the groove and is connected by a reed-like cantilever to a magnet. The jiggling of the magnet produces a voltage in a pickup coil which is amplified and sent to the speaker. But on a fine scale vinyl is rough, producing background noise. The stylus has mass, it takes force to make it conform to the groove, which may cause it to bounce around in the groove, ride up the walls, and shave the sides of the groove. The cantilever is not perfectly rigid, it flexes and vibrates as it tries to transmit the motion of the stylus to the magnet. The cantilever has to hold up the tone-arm, which similarly may bounce around in response the large amplitude segment of the signal. The signal in the pickup coils is very weak and must be amplified with high gain, causing electronic noise to encroach on the signal. The output signal, viewed on an oscilloscope or analyzed by scientific equipment, is obviously different from the original, affected by background noise, distortion of various types. How does this degradation of the signal add "life?" Isn't all of the life present in the original microphone signal, which the digital recording and playback process preserves?

And what 71dB was getting it is that if you like the sound of an LP (that is, you like the obvious distortion) you can make a digital recording of the LP playback, and it will sound exactly like an LP, with the surface noise, tracking error, harmonic distortion, rendered with exquisite fidelity.

There are digital recordings that sound awful and sterile to me, but it is not because the digital process suppresses the life of the signal. It is because produces have access to equipment which allows them to record with dozens, maybe hundreds of microphones and mix the sound artificially. This fails th capture the essence of the sound the way a minimal set of microphones in a good recording venue can.

Thanks for the lecture. I didn't go into it because I'm not interested in technical details. I have my experience. Decades of listening to LPs, then decades of CDs, now files, often streamed. The latter is much more convenient than the first and second. However, when I listened to LPs, I never thought about the sound quality, I only listened to the music, and was upset if there were too many clicks and scratches. With the transition to digital, I began to miss something in the sound. Many times I have heard the argument that people like me miss distortion, and that digital copy can add the distortion and everything will be fine. This argument seems stupid to me. I think you perfectly understand why. And if not, then I can't help

[71 dB]

That is a terrible analogy (pun not intended).

It's funny nevertheless. 

All audio starts analog.

Electronic music starts as signals in electric circuits or as digitally synthesized signals in a computer/synthesizer. I make music with computer and the music has an analog form only when I play it back on speakers/headphones. Very rarely do I record acoustic sounds. That's because I can't sing or play any instrument. 

You sent the signal to a device that measures the voltage hundreds of thousands to times per second, with accuracy more than one in ten million, and store the numbers.

If we talk about PCM recording, the most common sampling rates are 44.1 kHz (music productions), 48 kHz (video productions) and 96 kHz (high-res music production). Analog to digital converter do employ extremely high sampling rates, even megahertses, but the bit depth is low, even just one bit. Typically 24 bit recording is used (PCM). Almost 17 million quantized signal levels are theoretically available, but half of those have to be allocated to negative signal values and also a lot of headroom is being used in order to avoid signal clipping. Given the headroom is 18 dB (3 bits), the effective amount of quantized voltage levels is 2^20 which is about 1 million. In practise, this is more than enough.

Later you send the numbers to a device which performs he inverse function, producing an output voltage from the stream of numbers, updating the output more than a hundred thousand times per second.

Internally digital-to-analog-converters "update" the signal level according to the sample rate which can be anything from typically 44.1 kHz (NOS) to hundreds of thousands of times per second (major oversampling). However, the output is an analog signal which is "updating" constantly as a continuous smooth band-limited signal thanks to reconstruction filter.

How does this degradation of the signal add "life?" Isn't all of the life present in the original microphone signal, which the digital recording and playback process preserves?

There are digital recordings that sound awful and sterile to me, but it is not because the digital process suppresses the life of the signal. It is because produces have access to equipment which allows them to record with dozens, maybe hundreds of microphones and mix the sound artificially. This fails th capture the essence of the sound the way a minimal set of microphones in a good recording venue can.

It may very well add "life" to the sound from subjective point of view. The microphones are typically completely still while recording whereas people listening to a concert move their heads/body. Due to how vinyl works, it introduces all kind of fluctuation to the sound which may render the listening experience more "live."

I have been writing and developing Nyquist plugins for Audacity sound editing software to make digitally synthesized sounds feel warmer and more live (but be so subtle the effect is almost subliminal). We can learn some things from analog sound, but that doesn't mean we need to get stuck with analog audio. We can take all the good digital audio has to offer and improve it (subjectively) with tricks learned from analog sound. Best of both world!