The Problem With Experts Indeed

studio


Interesting take over on RealHDAudio taking shots at a music producer.

I read and replied to his post but it’s not publishing over there, so here is:


vocalize


Timing, timbre, and room sound.
Timing, timbre, and room sound.
Timing, timbre, and room sound.

These are things that you can’t scope or measure or chart. These are the basic building blocks of music.

This is why record producers, mastering engineers, and artists with a good ear are the experts here.

They are the only ones who understand mixed music. Not test tones. Not frequencies alone and isolated. Every bit of music is a complex stew of multiple tones, some heard, some hinted, some masked, some over/under ringing.

If the people in the studio that did the session say the 16/44 version sounds the best, then it does. If they prefer the 24/88 or 24/192 versions, they are the best. Creators privilege. Only they heard it as it was being made, aka what it originally came from. (They can all be different mixes of the song too, they don’t have to tell us that.)

The rest of us just take it for granted and enjoy it. Unless you are making the mix, or making the original sound being mixed, you are a secondary expert.

Mixed music is a tremendously complex collection of tones, all affecting each other, all containing critical timing, timbre, and layers upon layers of complex sound.

That’s why it’s so powerful. The power of music is ignored in these scientific discussions. If the 16/44 version moves you emotionally, that’s good. If the 24bit version does it more so, it’s a better version. Whichever packs the most in it is the best.


Even for sparse music, acoustic music, whatever…. more data = more sound = more vibration = more enjoyment. It’s simple.

I do think there’s a limit though. I hear some advantage at 24/192 on very good rigs but it does not make 24/88 or 24/92 sound degraded.

The pointless 16/44 is the degradation that we need to remove.


 

Too many people these days try to hear with their eyes and understand with their computer screens.

 

Which is music?

This:

beatles_wave

Strawberry Fields Forever, by The Beatles

 

or the audio track in this?

 

Science On My Side

maxresdefault


Another advancement in the understanding of how we detect sound: timing is more important, and more accurately detected, than frequency. #savetheaudio

This flies in the face of hi-res skeptics who believe our horrible ears could never hear beyond that CD-quality standard from 38 years ago.


It points to my favorite quote regarding this debate:

“The whole point of science is that most of it is uncertain. That’s why science is exciting–because we don’t know. Science is all about things we don’t understand. The public, of course, imagines science is just a set of facts. But it’s not. Science is a process of exploring, which is always partial. We explore, and we find out things that we understand. We find out things we thought we understood were wrong. That’s how it makes progress.” – Freeman Dyson, 90, Mathematical Physicist

and shows me that the established science will eventually catch up with what musicians and music lovers have known inherently for a long time. Some day they will be able to accurately measure what and how we hear, and perhaps even what music does to us.


Musical-Timing-Metal

 

 

 

Hearing With Bones and Eyeballs

Wha? We hear with more than our ears?  

Our bones? Our joints? Our eyeballs? Our teeth?

But don’t they just test the ears in hearing tests?  Yep.

Don’t they just test the ears in music listening tests? Yep.


ykbbzz7j-1373648866


 

See the problem? Complete hearing is done by combining multiple inputs.

Those making measurements of hearing and sound continue to miss this basic point.

How You Hear Music

tumblr_inline_n2emxrvzxP1qh0h3e

Listening 101

1. Directionality – Where is that sound coming from? Where exactly is that hi-hat sitting in the mix? Is the band in front of me or all around me?

2. Delay/Roomspace – Am I in a large room, small room, outside? Was this music recorded in a large room or small room? Was the recording room rounded, square, long?

3. Quality/timbre of instruments – Is that an electric guitar? Do I like the tone of it? Do I like that keyboard patch? How about the singer’s voice?

4. Stereo soundstage – Do I hear 2 guitars doubling each other, or 1 guitar with a wide delay? Is the singer front and center, or is he singing 2 parts, 1 left, 1 right? How wide are the drums set in the mix?

5. Timing of musicians and recording – Do the various delays work together musically, or are they clashing and changing the feel of the song? Are the drummer and bass player locked in? Is the 2nd percussion player ahead, behind, or on the beat?

6. Quality of recording – Is this the best version of this song? Is the distortion in the track intentionally added by the artist or is it in the format?

7. Clarity and breadth of EQ – Are most of the pleasing frequencies present, and are the harsh, brittle frequencies diminished?  Do the various instruments and voices blend and work with each other as layers, or do they cover each other?

8. Noise floor – Is there a hum or buzz in this recording? Is it from a bad recording, or a loud instrument, or something wrong in my system?

9. Phase – If things were recorded in phase you don’t notice it. If things are out of phase with each other, various comb filtering and aliasing artifacts appear in your music.

10. Digital loss/compression – Has this file been reduced from the original? Did they remove things they hope I can’t hear to make the file smaller?  Are there compression artifacts in the mix.

qsm_main

There’s 10 things to listen to without even thinking about frequency range. 10 things most internet audio experts never take into account.

Healing Senses By Depriving Them

Interesting theory here that a lazy eye can be cured or corrected with several days of absolute darkness. The lazy eye is a function of the brain not being able to control and process the eye properly, not a function of a malfunctioning eyeball.

The idea is that this total darkness reboots the visual cortex and allows the subject to emerge from the darkness and view the world in a more accurate way.

This is based on the what they call plasticity of our brains when we are younger. It’s been shown that the young brain adapts and takes on new skills faster than an older brain, in particular sensory and motor skills.

 

sight-and-sound-in-the-visual-cortex-617x416

The visual cortex in the rear of the brain

 

Researchers believe prolonged darkness, while maintaining everything else in normal patterns (eating, sleeping, socializing, etc.), allows the visual cortex to reset. When first presented with light it quickly decodes it and re-learns how to see.

The theory is that those that suffer from a lazy eye will see a real improvement with this non-invasive treatment.

The doctor behind it had to do a trial run, so he used himself and a volunteer to live in a light-free apartment for a week. Since they survived the trial run with no drama a larger study is being prepared.

 

shttrstck_modified_brain

The auditory cortex is nearly wrapped around the Thalamus, which is responsible for processing all sensory input, and regulating consciousness, motor skills, and sleep.

 

It is also interesting to note that sound experiments with light depravation have shown that subjects increase their acuity to sound when in the dark – aka the blind musician concept. Researchers kept mice in the dark and did sound tests and guess what, the dark-living mice could detect more detail in sound than the lighted ones.  But in this case, once emerging from the darkness, their hearing skills went back to previous norms. They want to extend the test to see if they can permanently improve the subject’s hearing.

I generally support any therapy that is non-invasive and non-pharmaceutical, so I wish them luck in their studies.

I’d go into a dark room for a few days and have a real stay-cation especially if I believed I would emerge improved and healed. This also sounds like something someone could manage mostly by themselves with a few close companions and basic medical consultation.

The Ghost in the MP3

Excellent work by Ryan letting you hear an approximation of what they are removing from MP3 files when doing “lossy” compression.

This is what the MP3 programmers deem unimportant in your music. You can play the video with it’s own lossy audio, or go here to hear the full version of what they pull from your music to make MP3 files.

Most of what is cut out is spatial — reverbs, room sound, delays, decays, fade outs, dynamics, lots of pre-delays, layering of sounds, attacks, breathes, etc..

This is the movement and the emotional content of the song. The interacting layers is the kind of data that computer programmers (and digital internet babies) can’t quite measure, so they disregard it. That’s scientific method at work – if you can’t measure or control it, disregard it.

This is important listening and will help you to understand that hearing music is more than frequencies.

I would love to see someone do this type of experiment with a 24bit mix and a 16bit mix of the same music.

 

enhanced-buzz-11125-1347395176-28

Don’t take away my reverb and delays! The power of Bonzo is a result of decay, delay, and room sound.

The Complicated World of Hearing

What do you hear?  

Everything you can. 

How do you hear?  

Binaurally through the ears using mechanoreceptors in the form of tiny hairs called follicular receptors. Additional data is picked up by other mechanoreceptors throughout your skin including under every hair follicle and inside many of your joints. Our brains then use both heard and felt sound data to understand what we are facing.

What are properties of binaural sound?  Also known as sound localization, this details the time- and level-differences between both ears, such as

  • Spectral information
  • Timing analysis
  • Correlation analysis
  • Pattern matching

This overall intelligence of the 3-dimensional sound helps form the timbre, room ambience, and performance dynamics of the sound. This happens whether the sound is a lawnmower outside your window, an oboe solo, your cell phone buzzing, or all three at once.

This binaural data is all added to frequency when we determine a sound and its total quality.  Frequency is the note being played and the sound tone being generated on a measurable scale – basically wavelength at a particular moment in time.

The problem is that modern science only has reliable measurements for frequency. Frequency, besides being easily measured, is easily displayed on our 2D displays, and thus has been studied thoroughly for over 70 years. The field of Digital Signal Processing (DSP) is based on the study of frequencies and human’s interactions with them.

But there’s much more happening when you hear a song you love than frequency analysis and musical recognition, and modern math has yet to capture and replicate it. This leaves people that insist on having math behind everything in the natural world in a bad place, disregarding most of this sound data as being outside their realm of importance.

But the spatial and sizing data is critical, in some ways more than frequency. Examples: You always know if what you hear is real or a cover band. If it is the original artist you also know if it’s been degraded at either the source (a bad copy), or on playback (bad headphones, bad cable, bad reception, etc.).

If it’s a playback problem you start looking for the damage to correct it asap. If it’s damaged source you usually just accept it as all you have and try your best to enjoy it. This all happens without thinking and without training or education. We just understand it as how the physical world works.

When music you love comes to you, regardless of the quality, it lands. Since you can only comfortably take in 1 version at a time you eat it up.

 

There's a lot going on inside your ears.

There’s a lot going on inside your ears.

 

Complicated stuff, right? There are lots of variables. All of this independent to you and you only. I have totally different listening spots, gear, and favorite songs than you do. We all do. Emotionally we are all different, minute by minute, week after week, living within our altering moods.

I can project that this “hearing thing” is perhaps impossible to measure with all of the variables.


 

So how else to model this? The marketplace is one way, which will both benefit and hurt the argument.

The fine art market has experts, moneyed customers, and various business interests building value on pieces of art, which as you know sell  from dollars to millions of dollars per piece. These experts hunt down and remove fakes to protect the quality of the masters at any cost.

More people bought Justin Bieber posters last year than fine art. That’s the marketplace, that’s mass duplication.

We can’t let the digital generations burn and loot the art vault with resolution assumptions based on bad science. We have a century of recorded music that could be upgraded and distributed to this centuries people, and it should be done in highest quality 21st century digital can offer.

Imagine if in 1978 two paper companies invented a printing method that output 2,200 colors at 160dpi, perfectly matching their new poster stock, and then sold millions of fine art posters at that resolution. Then for the next 40 years people walked around saying “it’s the same as the original” and couldn’t even accept there was any better quality possible. This is exactly what has happened with CD audio.

 

The point of all of this: Team Ear, baby! Don’t let technology worship cloud your views of what the human body is capable of.

The human ear has an extraordinarily large sensitivity range of a trillion to one, allowing us to hear a rocket launch or the footfalls of a cat on a carpet.  According to Werner Gitt, the ear is our highest-precision sense organ, capable of responding over twelve orders of magnitude without switching (The Wonder of Man, p.21).  Some of this sensitivity is amplified by the eardrum and middle ear ossicles, but the paper reported above shows even more fine-tuning inside the cochlea.  Gitt’s book is highly recommended for generating a profound feeling of awe over the design of our senses.  Proverbs said, “The seeing eye, and the hearing ear, the Lord has made them both” (Prov. 20:12)

Read more details on one of the studies of the cochlea here.

 

 

Lecture: Quality Sound Matters

Stop listening to internet experts and listen to real experts. Here’s a panel of mastering engineers talking about quality sound and consumer trends. Interesting, informative and correct!

Bonus fun is that they invited a guy from streamer Rdio who has to defend 320k streaming in this room full of quality experts. I bet he’s the young one on the end looking exasperated 😉

The Problem With A-B’ing And Why Neil Young Is Right About Sound Quality

turntable


Great Tape Op post that’s thinking big about audio, music, and hearing.

The main crutch of the good enough team is what is called the double-blind listening test (shortened to ABX). When doing studies based on perception, it is the great measuring stick, and perhaps the only way they can start to squeeze some numbers out of human sensory perception.

It’s basic – here’s source A, here’s source B, maybe switch back and forth a couple of times, now make your decision. Which one was better? Can you hear a difference? Do you like one better than the other?

But as the article states, every ABX test is flawed because of it’s short sample time, and building out theories on these short ‘taste-test’ findings has led us to this mess of bad science and bad assumptions.

Since we live with and love music in intimate ways we cannot accurately write or describe, the author proposes that for any “double blind” tests to be valid the subjects should actually get to keep and live with their music collection for a month or two, then report their feelings towards it.

Much like how a sugary treat tastes better than anything next to it, but if you lived on sugary treats all month you would be feeling much worse than the person with the quality diet. Often the lesser files are close enough on initial inspection to fool enough people, and the ABX test stops right there. No one is doing long-term ABX tests, we all are doing taste tests, not nutrition tests.

Neil Young and the high-def audio movement is about getting the nutrition back into your music. There’s industrial white bread, and then there’s all those other breads. They both hold the sandwich together but living off the nutrition inside of it leads us to different outcomes.

 

1 Trillion Odors, or alot of Funk

12868_2007_Article_679_Fig1_HTML


Ha, Imagine That!  I’m running all over the internet fighting bad science about hearing and music, and The Journal of Science publishes a study that says scientists have really underestimated the abilities of our nose and sense of smell.

Oh those crazy scientists, always learning more about our senses. Always so amazed at what the human body and brain can do. Sometime Simpleton.

This mirrors what is happening in the audio world. I really do think we will look back at the days (decades) of claiming “humans can’t actually perceive anything beyond 16/44 digital files” as the ignorant dark ages of hearing science. Producers and musicians have been ignored and derided in the name of digital convenience for many years now.

All it takes is one scientific paper to state something about how we can sense all kinds of other tones, timbres, and frequencies throughout our bodies, and how when receiving the full spectrum of audio, human bodies react positively. Familiarity is the first stage of listening, but we must go further than that for actual enjoyment.

But that’s not science, is it? It’s just a reality that is hard to quantize.

 

 

Bad Science + Business Interests = Trouble

Computer geeks know lots of things. The sheer breadth of stuff that geeks have crammed in their head is impressive.

But their major mistake is often not acknowledging their own ignorance. Many have come up in a world so digitally driven that they forget they are analog animals.

They forget sound, light, smell, touch are all analog. These are things computers don’t do natively.

In fact it has taken 40+ years of digital advancement to even start competing with original (analog) methods of creation.


NEM U87AISETZNI

Hi there I’m analog


Most computer nerds know nothing about professional media production. They might know the basics or have clicked around a bit with an app, but they know nothing of producing high quality media for a living.

On the other hand, most producers these days have to know their computers, especially the parts critical to creating professional media. I believe some nerds don’t like the competition so they declare themselves experts on everything digital.

Experts are the people that do it for a living, not people tasked with spreading false information on the internet.

A computer programmer/nerd believes there is a digital solution to everything.

Then they build on this bad foundation the fatal flaw of believing a digital copy of something analog will somehow be superior. Many sub-measurements of that digital file might be superior to the analog, but remember to always step back and say “what is this trying to solve?”.

Music is created to get an emotional response from us and that requires as much audio data as possible.


gemini-pt-2000-ii-direct-drive-turntable


All consumer digital music, from the CD in 1978 on, has been a compromise. When you hear analog playback you are hearing a reflection of the recording, that is, an analog copy that is slightly degraded but overall intact and whole.

The original sounds hit the microphone in analog and it will hit your ears in analog.  It has not been broken up and re-assembled, and no computer decided what to keep and what to throw out.

Nature does degrade the signal to a certain extent (magnetism in a tape or physical dragging movement on vinyl), but no programmer had to determine mathematically what parts of your music to throw out.

Computer nerds trust in the computer to decide what’s important in our audio signal, more than they trust their own intuition or senses.

Computers don’t have skin, hair, ears, or emotions, so what do they know about music? Nothing. Nada.

Programmers with agendas are behind much of this nonsense, and it is all based on a total misunderstanding of how we hear, and what we actually get from music.

Familiarity is just step 1. “I can recognize that song I like!” is not the same as hearing the whole thing the way it was intended.

Check out this cool article about a guy that helped design the Pono Player.