audio file converter for music production and hi‑end audio
ISO DSF DFF WAV FLAC AIFF ALAC mp3 ...
Follow
Audiophile Inventory on Facebook Audiophile Inventory on Twitter Audiophile Inventory on Instagramm Audiophile Inventory on Pinterest Audiophile Inventory on Telegram Audiophile Inventory on YouTube

Hi-Res Audio [Explanation: Sound Quality, 7 Myths | 2019]

Share 
Audio Basis - educational articles

Read below quick guide:  what is hi-res audio explanation; High-Res vs 44.1/16; Myth debunking and other things that you want to know; Where to download free high res samples (24 bit FLAC download(s), DSD and others); Where are online streaming sites.



Watch and share: Hi-Res Audio [How it works. Sound quality. Myth debunking]
 


What is hi-res audio

High-resolution audio (hi-res audio, high-definition audio) is one of widespread audiophile and music production terms. Check free hi-res downloads (24 bit, 96 kHz, 192 kHz, DSD) >

 

Why Hi Res need?

High resolution audio is sound formats (signals) with resolutions above 16 bit / 44.1 or 48 kHz, including 1-bit DSD (Direct Stream Digital).

Hi-res audio was designed to provide higher sound quality, than standard CD-audio. Read below: is the aim achieved or not?

PCM 44.1 kHz / 16 bit vs. High resolution audio PCM and DSD

PCM 44.1 kHz /16 bit vs Hi-Res PCM and DSD

Some kinds of music with wide range of instrument loudness (like classical, jazz), demands wider dynamic range of audio system.
The range may be achieved via increasing of resolution. Read details in "Why bit depth above 16 bit is need" part below.

High resolution records was designed for music with wide loudness range

High resolution music records

There are many questions about necessity of hi resolution musical records.
One people thinks, that "excessive" resolution waste disk space only.
Other people thinks, that such records are need to produce ultrasound.
But, by author opinion, existing of high-res may be useful and have other reasons. Read "Myths" part below.

High resolution audio file formats
(sample rates)

 

High resolution audio formats (sampling rates)

 

High resolution audio file formats
(bit depths)

 

High resolution audio formats (bit depths)

Opinion, that high res give advantages always, is not correct for all cases.
Sometimes people, who play records with high sample rates, stumble with unexpected noise or other unwanted sounds.
In "Myths" part you can watch video with example.

 


Watch and share: What is DSD audio? Video explanation
 

 

Hi-res playback

Hi resolution may be played back at:

  • computers,
  • portable audio players,
  • mobile phones [smartphones],
  • tablet computers,
  • music servers,
  • SACD and BlueRay players,
  • other.

High resolution audio file player software list

 

Separate [DAC-headphone amplifier] device may be used to connect to mobile phone, portable player, computer and other.

There are high res audio players for Mac, Windows [PC], iPhone, Android.

There is Bluetooth for high resolution audio [aptX™ HD] may be used. It support 48 kHz / 24 bit LPCM audio data.

 

High-res audio converters

Audiophile and professional audio conversion software should provide wide range of sample rates and bit depth, including 32-/64-bit.

List of converter software for high resolution audio files

 

 

High resolution audio PCM

High resolution PCM is supported by lossless files: WAV, AIFF, FLAC (up to 384 kHz / 32 bit), WV (WavPack), DVD, Blue-Ray and other.

Also 32/64-bit float formats are high resolution formats. But these format are intended for music production and scientific purposes rather.

Read details below in "Bit depth" and "Sample rate" parts.

Read more about PCM formats >

 

 

DSD

DSD (Direct Stream Digital) is 1-bit audio format with very high sample rate (from 2.8 MHz) and based on sigma delta modulation.

To reduce noise level and expand dynamic range, noise shaping is used there.

Mediums as DSF/DFF/SACD ISO files, SACD optical disk.

Read:

Watch and share: What is DSD audio? Video explanation
 

 

 

DXD

DXD (Digital eXtreme Definition) is PCM format with high sample rate, that derived from DSD.
Other words, DXD contains low part of DSD spectrum with noise.
Only part of DXD spectrum contains useful audio signal.
Noise level is grow with grow of frequency.

As rule, DXD have sample rates 352 800 Hz and above and bit depth 24 bit and above.

Primarily, DXD format is intended to DSD editing purpoces.
However, there are other formats may be used too.
Read details about DSD editing >

However, not all PCM files with such sample rates are DXD. Traditional PCM have no significant noise level grow at high frequencies.

Playback devices can filter high frequency noise.

More about DXD here.

 

 

Size compression for audiophiles

 

FLAC

Lossless method to compress PCM audio up to 384 kHz / 32 bit. Read details here >

 

DST

DST (Direct Stream Transfer) is lossless compression method for DSD (1-bit audio) sound files. May be applied in *.dff files.

 

MQA

MQA is intended to PCM high res audio size compression.
As rule, it have compression ratio better, that FLAC.
But, as author know, MQA is not lossless (no losses of information after restorting audio from compressed stuff). Read details here.

Also read about MQA.

MQA official site

 

 

PCM vs DSD vs DXD sound quality

All formats give sound system designers same abilities. Final result depend on how audio system is done.

PCM have no technological noise "hump" at high frequencies. DSD and DXD have.

So there is no necessity to suppress the noise excess at PCM high frequencies.

However, we don't hear ultrasound (see details below in "Myths"). And the noise is not audible.

Nevertheless, audio system hardware and, sometimes, software have non-linear distortions.
The distortions can cause generation of audible products (noise, as rule) by inaudible noise. It's called as intermodulation distortions. See more here >

Read about DSD vs PCM >


Watch and share: DSD vs FLAC [Format Comparison]
 

 

 

Why bit depth above 16 bit is need

 

When bit depth discussed, human hearing range (about 130 dB from inaudible to pain) is taken as reference value.

16 bit have difference about 96 ... 110 dB between lowest (noise floor) and highest peak levels.
It looks like enough to cover almost full range 130 dB.

However, the range don't consider sound quality aspects.

Low hearing threshold 0 dB (quiet rustle of leaves, as example) is sound with own signal/noise ratio.

As example, we want to provide proper sound quality (like tape's 60 dB, as example) at 0 dB audible level.

So we can add about 60 dB to 130 dB (total about 180 dB) to provide allowable sound quality at lowest levels.

Why bit depth, higher than 16 bit, is need.
Look at signal/noise ratio of lowest loudness (quiet instruments)

Why bit depth, higher than 16 bit, is need

 

It is not exact numbers, because author don't have result of researches. But we can get it as hypothesis.

So bit depths, higher, than 16 bit, may be needed.

DSD solve dynamic range issue via noise shaping.

Read details about: DSD noise shaping >

More about hearing range here

 

 

Why sample rate above 44.1 kHz is need

 

Sample rate 44100 Hz:

  • cover audible band 0 ... 20 000 Hz and
  • have 2050 Hz reserve for transient band of low frequency filter (in ADC and DAC).

However, the reserve is considered abstractly. Real-life filters are analog. And have wide transient band.

To solve the wide transient band issue, digital filters and higher sample rates or oversampling are used.

But digital filters are not ideal and have:

  • ringing audio (property of digital filters);
  • limitation of minimal size of transient band (due limited computing resources of example).

To reduce necessity in "hard" efforts to proper filter transient band, higher sample rates (including DSD) may be applied.

Read details issues of sample rate and filters:

 

 

Myths

Myth #1. High resolution audio have no sense because we can't hear ultrasound

High res is not intended to recording and playback ultrasound.

ADC and DAC have analog filter at input and output respectively.

Analog filters are not steep and can't suppress aliases properly in allowable band 20 000...22 050 Hz for 44 100 Hz sample rate.

Higher sample rates allow to place aliases at higher frequencies, in better suppression range of analog filter.

The issue may be partially solved via digital filtering and decimation in ADC or oversampling and digital filtering in DAC. However, may be issue with quality of filtering due limited computing resources of hardware.

Resume: high resolution is intended to analog filters in ADC and DAC. And ultrasound playback is sequece of higher sample rates. Ultrasound playback is not aim of hi-res records.

 

Myth #2. High resolution have sense because it playback ultrasound

Practical experience, that may be checked right at home, says, that we cant hear sines with frequencies above 16 ... 22 kHz. And there no surprises for number trials.

There is study:

  1. Toshiyuki Nishiguchi, Kimio Hamasaki, Masakazu Iwaki and Akio Ando: Perceptual Discrimination between Musical Sounds with and without Very High Frequency Components

This study don't found significant difference between samples with and without very high frequency components. However, the authors of the research neither confirm nor deny general human possibility to discrimination of music with and without the components.

 

Other hand, some researches, that studied ultrasound issues:

  1. Milind N. Kunchur, 2007: Temporal resolution of hearing probed by bandwidth restriction

     
  2. Lisa Zyga, 2013: Human hearing beats the Fourier uncertainty principle

     
  3. Tsutomu Oohashi, Emi Nishina, Manabu Honda, Yoshiharu Yonekura, Yoshitaka Fuwamoto, Norie Kawai, Tadao Maekawa, Satoshi Nakamura, Hidenao Fukuyama and Hiroshi Shibasaki, 2000: Inaudible High-Frequency Sounds Affect Brain Activity: Hypersonic Effect

 

Some of the studies show, that we may have brain response to ultrasound. But, it is unknown, how ultrasound impact to sound quality.

 

Myth #3. Hi Res audio sound nicer (or closer to analog) because it cause lesser "stairs"

Digital signal don't cause "stairs" in any resolution. Because analog filter at output works as interpolator (connect momentary voltage values of digital samples via smooth curve).

Digital signal isn't stairs

Digital signal isn't stairs

For high resolutions interpolator (analog low frequency filter) is easier to implementation in better quality (see myth #1).

 

Myth #4. We can compare 24 bit / 192 kHz versus 16 bit / 44.1 kHz, DSD vs PCM, etc.

To the comparison we must have pure samples, that cleared of other factors.

As rule, such comparison considered as blind test of samples that provided by a label in different resolution. As example, one album in 24 bit/192 000 Hz and 16 bit/44 100 Hz.

However, main issue is there, that it is impossible to record "clean samples".

As example, we split sound from microphone to 2 audio interfaces, that works in different resolutions. But we can't split absolutelly identically. An identity of 2 audio interfaces is more sophisticated task, than the splitting.

When single sample (music album) is converted to test with a software, we don't know that we compare: converters or resolution.

But, even, when converted samples are played back at single DAC, it may works in different modes (electrical circuit, digital processing) for various resolutions.

Author would suggest to consider all known kinds of tests like "hi-res vs 44/16" as comparison of hardware and/or software, not as comparisons of audio resolutions.

Test standard vs high resolution audio:

  1. Joshua Reiss, 2016: A Meta-Analysis Of High Resolution Audio Perceptual Evaluation

 

Myth #5. Hi-res records always bring the best sound quality

Sound quality depend on many variables: recording, playback, mixing/post-production software and hardware tools, skills of sound engineer, recording and playback environment, quality of resolution conversion and other.

Format/resolution is one of number of these variables. No warranties, that DSD1024 is sound better than standard PCM 44/16.

 

Myth #6. Hi-res formats bring advantages only

No. High resolution formats contains ultrasound. Theoretically it is not matter for ears, because we don't hear it.

But, practically, audio hardware and, sometimes, software have non-linear distortions. It can cause audible products from inaudible high frequency components (intermodulation distortions).

At the educational video below, noise removing via ultrasound filtering is shown.

Watch and share: How to remove audible noise for high resolution audio
 

 

Read more about optimization audio >

In general case, higher sample rates and bit depth guarantee nothing. Result depend on implementation quality.

Professional audio interface at 44100 Hz can give better sound, than built-in sound card at 192000 kHz.

 

Myth #7. Resampling can't improve sound quality

If consider altering audio resolution (resampling) as itself, it lossy processing always.

After upsampling no information is added. Conversely, we have losses, as we told above.

But we can't consider audio recording without considering of playback musical system.

And the system may have different distortions for various modes (combinations of sample rates and bit depths).

So, we can optimize the recording to better playback mode: choose mode with minimal distortions.

To reduce conversion losses, audio conversion algorithm should have minimal distortions.

However, it is need to remember, that sampling rate increasing don't give advantages at its own.
Theoretically, in some cases, reducing sample rate may make sound better.
In other cases, no audible difference between resolution.
Each case should be learned individually.

Read more about sound quality improvement >

 

F.A.Q. Hi-res music

 

What is the best audio quality?

Audio quality have 2 definitions:

  1. distortion level;
  2. subjective estimation of sound "beauty".

Distortion level estimate degrading sound of original source. But not only integrated level of distortions is matter, but character of the distortions makes subjectively perceived sound quality. This phenomenon is studied by psychoacoustics.

Subjective estimation can more like more degraded sound. Because some kinds of distortions can bring "warmth", "coziness", etc. of sound.

Resume: The best sound is zero distortions. But some distortions can "enhance" sound "beauty" ("colorization").

 

What is the best audio format for sound quality? What is the highest quality digital audio format?

Lossless audio formats (WAV, FLAC, AIFF, ALAC, DSF, DFF, others) provide storing of audio stuff without losses.

So, these formats may be considered as the best by sound quality.

Read more details >

 

Is 24 bit audio better, than 16 bit?

24 bit is not better, than 16 bit at its own. Theoretically, 24 bit have higher signal/noise ratio (lesser distortions). But final result depend on implementation of software, ADC and DAC.

Read more details >

 

What is hi res music?

Hi-res music is music, produced and stored in hi-resolution formats (sample rate more 48 kHz, bit depth 24 bit and higher or DSD).

 

Is FLAC hi res audio?

FLAC can contains high-resolution audio (sample rate more 48 kHz, bit depth 24 bit and higher or DSD in DoP).

 

Is FLAC lossless?

Yes. FLAC is lossless PCM audio format, that provide binary identity of original and uncompressed information.

Read more details >

 

Is FLAC the best audio format? Does FLAC sound better?

FLAC is one of lossless formats, like WAV, AIFF, APE, DSF, SACD and other.

FLAC have sample rate and bit-depth limitations: 32 bit / 384 kHz. WAV and AIFF are more flexible.

But FLAC have wide support (including metadata) of software and hardware and lossless size compression.

Read about:

 

Is FLAC better than mp3 320 Kbps?

FLAC is lossless formats. mp3 is lossy. So, technoically, FLAC is better mp3 in any bitrate.

But the difference between FLAC and mp3 320 Kbps is very subtle, that may be inaudible.

Read more details >

 

Is FLAC better than CD?

Technically, Free Lossless Audio Codec allow to achieve better signal/noise ratio, than CD.

But actual sound quality depend on implementation of audio playback system and recordings.

 

Is FLAC better than ALAC?

For sound quality, FLAC or ALAC is not matter. But compatibility with audio player issues may be there.

 

What is the best FLAC player?

It is personal choise. But there are some general recommendations.

Read details >

 

Why are FLAC files so big?

It is technical limitation for lossless formats (original and uncompressed audio are binary identical).

 

What is DSD audio?

DSD is high resolution audio format based on 1-bit sigma delta-modulation to achieve proper signal/noise ratio.

Read more details >

 

Is DSD better than FLAC? Is DSD better than PCM?

Final result depend on implementation and played back recording.

Read more details >

 

Where can I download music legally? How can I download free music legally?

  1. Some labels (recording studios) have free samples at official sites.
  2. Sometimes coupons for free music downloads may be placed in pack of audio equipment.
  3. Audio magazins can provide samples.
  4. Services for promote musicians.
  5. Others

See list of free musical downloads >

 

Where can I download high quality music? How do I download FLAC?

See list of free and commercial audio downloads >

 

Where can I download free songs? What is the best free music download site?

See list of free music downloads >

 

 

Conclusions

 

  1. Higher sample rates solve issues of analog and digital filtering in analog-to-digital (ADC) and digital-to-analog (DAC) converters.
  2. Higher bit depths solve issues of signal/noise ratio for lowest-level musical signals.
  3. Now saving space is not so actual as before (except portable devices). But lower bit stream (bit per second) value is actual for network streaming yet. So high-resolution size compression is demanded currently.

 

Author: ,
Audiophile Inventory's developer

 

Hi-res audio

 


Read the articles

Share 

Copyright © Yuri Korzunov
["Audiophile Inventory" (since 2011), "AudiVentory", "Audiophile Inventory by AudiVentory.com".
Также "Audiophile Inventory by AudiVentory.com" (прежде "Audiophile Inventory") охраняемое коммерческое обозначение],
2010-2019. All Rights Reserved.
Information at this site is not public offer.

Site map      Terms and Conditions