How Science Got Sound Wrong

[CN211276]

Food for thought?

https://www.fairobserver.com/more/scien ... LsMeGXKr2I

[Geoff.R.G]

Interesting and very informative. I have long been a believer in higher sample rates from a theoretical perspective but that article explains why there is more to sound than just frequency and amplitude.

The Nyquist principle has always seemed wrong, I=f/2 is questionable anyway because it assumes that frequency alone determines the fidelity of a digital sound. If we sample at 44.1KHz what happens between samples? Sure we can reproduce a 20KHz sound but what else happened in the intervening 23mS and, more to the point, is it important?

Why is listening to a concert on an FM radio more engaging than playing a CD of the same piece?

Seems that timing is the essential component and analogue recording preserves that while digital doesn't.

[Lindsayt]

There's 2 types of compression.

Compression where you use fewer bits in a digital recording to represent the signal.

Dynamic compression.

It's unclear which he's referring to in the article or whether he's referring to both.

I think that both vinyl and CD are good enough recording mediums that it's the quality of the mastering that's most important in determining which sounds better.
With 21st century commercial CD's having universally shockingly poor mastering quality from a dynamic compression point of view.

[Daniel Quinn]

I don't think we are anywhere near a full knowledge of the science of sound , because we don't understand the brain

All this arguing if this or that sounds best and we haven't got a clue how our brain works.

I used to think that we all heard the same and our 'personalities' got in the way . But 1 thing a brain injury does ,it allows you to review your recovery.

All my right side was damaged ,including my hearing . Subtle spatial clues about were things were in space were blunted on my right side. Thus although I had functional hearing, such finer functions were not there ,it is only as they have come back that I've noticed. The soundstage on my hifi as never been better .

Also ,initially my ability to hear into the passage of music and follow the musics individual parts was compromised. Had I not had a brain injury , I'd of thought my hifi was shit and been on the upgrade path .

The damage to my brain was by far the single most important factor in what I was hearing.

I recently developed the ability for my brain to move my fingers . They don't move ,but my brain thinks it's moved them . Prior to this I was unable to 'think' about moving them . This is difficult to understand unless you had a brain injury. But your brain as got to think about movement for you to do it. It becomes ingrained in you so you don't register it.

Following my brain bleed, I was unable to think about my entire right side. I was able to think about moving, but I was unable to conceive of the functional thought which moves your body .

The point being , I am able to conceive of my fingers moving and touching one another , so I am able to feel them touch one another. But they haven't .

This is why I think its all subjective.

[CN211276]

A very interesting and informative article which I have come across before. I think it hits the nail on the head about the limitations of 44kHz CD and the roll of the brain in our perception of sound and the importance of timing. The article echoes what Rob Watts wrote at around the same time, all be it with diffegrent terminology. Rob Watts referred to the brains reaction to the reproduction of transients and the effect this has on the perception of sound stage. It boils down to upsampling being required to replicate the benefits of vinyl and its emotional appeal, whilst at the same time retaining the many benefits of digital.

As the brain has such an important role to play in the perception of sound/music, it is understandable that subjectihve opinions about hifi sound vary a lot. From my individual prespective upsampling has been a game changer. It first came to my attention when Qobuz started streaming familiar classic albums at 192 kHz. The sounded so much better. As well as subjective responses the transparcy of the rest of the system is important. When I heard the Mscaler through WAF speakers it did very little for me. When I heard it at home I could not be without it. I think the regression in speaker design is putting the brakes on the take up of advances in high resolution streaming.

Having read Neil Young's autobiography last year I am well aquainted with his views on digital and in particular his loathing of compression. Many older musicians have similar views, especially concerning the recording process, preferring the more emotional sound of analogue studeo recordings.

[Theo]

I recommended this book some time ago:



It's a wonderful read about how we got to where we are today in terms of recorded sound. You'll be both amazed and horrified at decisions made along the way.

[Wonfor14]

This as been very interesting subject, I have said many time I hate Digital Lies.
My reasoning was thus
1) most sound like crap .
2) as a wee kid my Grandad taught me some electronics and said that the world and all it wonder can not be sampled with one dip of the toe.
This is basically what Digital Lies are a dip of the toe, and we have been mislead by money grabbers like Marconi, Edison, none of the money making marketing people have every had a creative thought and never will, or understand the finer detail of why a things works, and be truthful in my experience are blinkered by the signs of money.

At college (Circa 1969) we are taught many things most of which is laid down by industry who want certain pre-program but bright or quick workers. We are not taught to be creative and after getting a certificate we are allowed lose into the world with only the basic and just enough knowledge to do the job well enough to make the marketing men happy (mmmmm dollar's).

A odd few lucky one's go on the "Higher Education" and university for more pre-programing.
My point is we are taught to except what the boss men say's as gospel and never question so in the past we are told CD are better (yes as beer mats) than vinyl but this is also limited.
So get the whole foot, no body immersed go to more live concert but wear a mask and no hugging.

[Geoff.R.G]

This as been very interesting subject, I have said many time I hate Digital Lies.
My reasoning was thus
1) most sound like crap .
2) as a wee kid my Grandad taught me some electronics and said that the world and all it wonder can not be sampled with one dip of the toe.
This is basically what Digital Lies are a dip of the toe, and we have been mislead by money grabbers like Marconi, Edison, none of the money making marketing people have every had a creative thought and never will, or understand the finer detail of why a things works, and be truthful in my experience are blinkered by the signs of money.

At college (Circa 1969) we are taught many things most of which is laid down by industry who want certain pre-program but bright or quick workers. We are not taught to be creative and after getting a certificate we are allowed lose into the world with only the basic and just enough knowledge to do the job well enough to make the marketing men happy (mmmmm dollar's).

A odd few lucky one's go on the "Higher Education" and university for more pre-programing.
My point is we are taught to except what the boss men say's as gospel and never question so in the past we are told CD are better (yes as beer mats) than vinyl but this is also limited.
So get the whole foot, no body immersed go to more live concert but wear a mask and no hugging.

I rather like the metaphor of dipping a toe. Digital audio at a low sample rate is a bit like reading every odd numbered page in a book, higher sampling rates might get to reading alternate words or even letters. The brain is left to fill in the gaps with a high probability of getting it wrong.

Digital recording has its place, short talks with easy distribution being the obvious one. From an industry perspective ease of editing and convenience must be attractive but that easy editing must play havoc with the timing.

I have already had my say on dynamic compression elsewhere so I’ll say no more on that. Digital compression is another matter, throwing away data that, apparently, isn’t necessary seems wrong, a bit like abridging a book. I learned some years ago that you can’t just remove something, even if it can’t in theory be heard, without affecting the sound. An elderly friend with some hearing loss complained that I had rolled off the top end. He couldn’t hear that high but he knew when it was missing.

Clearly something is lost in the digital recording process and I don’t think we are in any position to say what it is. Combine this loss with deliberate removal of data to “save space”; either by digital compression or low sample rates, or both and it can be no surprise that the result is unsatisfying.

All analog recordings are dynamically compressed by the nature of the medium but, it seems, that isn’t as much of a problem as the missing “data” resulting from digital recording. Being involved in live sound has resulted in my rarely playing recorded music but I haven’t, until now, considered why. Despite using a digital mixer, with live sound the analog is still there which, when you think about it, must preserve the timing.

Thanks Colin for making me think.

[Wonfor14]

If the all important thing is about timing then that would align with the way RD assessed things. Not entirely on topic but it explains why some like single point source speakers, why time alignment in multiple driver speakers is important and why phase (time) shifting filters can really mess up the sonic experience for some.

[Wonfor14]

If the all important thing is about timing then that would align with the way RD assessed things. Not entirely on topic but it explains why some like single point source speakers, why time alignment in multiple driver speakers is important and why phase (time) shifting filters can really mess up the sonic experience for some.

Spot on Phase is so important, i.e. it relates to time and thus speed and frequency why do only a few people get it? And as we now understand filter not only crush/lift frequency but screws phase.
So you can have fast amp to control speakers with massive damping but a high frequency filter in the input giving -3dB roll on and about 800KHz but the cable also can not be a load to the amp especially capacitance. The same applies but less in input leads, and USB digital lies cable.
The subject has not been fully explored IMO. Let alone the effects of the environment as in RFI/EMC/temperature and air pressure, and the dreaded mother in-laws effect.