What is Hi-Res audio and where to download music in this format

Glossary

Stereoland 04/08/2020

SaveSavedRemoved 4

Hi-Res Audio (High-resolution audio) is high-resolution digital audio.

Unlike HD (high-definition video), there is no single standard for high-definition audio.

In 2014, Digital Entertainment Group, Consumer Electronics Association and The Recording Academy, together with music labels, officially defined Hi-Res Audio as: “Lossless audio that is capable of reproducing the full spectrum of a recording and has been prepared and transferred from a higher quality source.” than CD."

Simply put, Hi-Res audio refers to music files that have a higher sample rate and/or bitrate than CD, which is defined as 16-bit/44.1 kHz. The sampling rate (or sampling rate) refers to the number of times the analog-to-digital conversion process takes a sample of the signal per second. The more bits, the more accurately the signal can be measured in the first case, so moving from 16 bits to 24 bits can provide a noticeable jump in quality.

Hi-Res audio files typically use a sample rate of 96 kHz or 192 kHz and 24-bit. There may also be 88.2 kHz and 176.4 kHz files.

Hi-Res has one drawback: file size. One track can be tens of megabytes in size and several albums can quickly eat up the free space on your mobile device. They can also take a long time to load over Wi-Fi and mobile Internet.

• Hi-Res audio file formats
• What is the advantage of Hi-Res audio
• What you need to play Hi-Res audio files
• What's next for high-resolution audio

What is Hi-Res audio in simple words

Unlike video and photos, no specific standard has been invented for high-resolution audio. It is generally accepted that a Hirez is an audio file with a sample rate and bit depth greater than that of a CD. The first parameter determines how many times per second the signal is sampled when converting it to digital format. And the accuracy of signal measurement directly depends on the bit depth, so a jump from 16 bits to 24 improves sound quality. Accordingly, the sampling frequency of Hirez is 96/192 kHz with a bit depth of 24 bits, but there are also indicators lower than these.

This mode includes the following file formats:

1. FLAC is a common format whose file weight is smaller than its analogues. Most devices (except Apple equipment) support this standard. A number of users consider FLAC to be the best option for downloading individual audios and albums.
2. ALAC is a format that has received certification from Apple. It has high sampling rates and serves as a FLAC replacement for opening in iTunes.
3. WAV is a format in which CDs are primarily recorded. The sound quality is good, but the file sizes are large due to unavailability of compression.
4. MQA is a format in which the main thing is rhythm. Often used by Tidal for broadcasting. The disadvantage of MQA is that it has few supported devices.
5. AIFF is an alternative format from Yabloko, which has received more complete metadata. At the same time, AIFF, like the Highrez format, is not widespread.

There are other formats in Hirez, a list of which can be found on Wikipedia. But these are the main ones.

Master Class

This may be okay for listening to Spotify playlists while commuting to work on the bus, but for true music fans this quality won't be enough. High-resolution audio will come to their aid.

To understand why it should sound better than MP3, let's compare their bitrates. The highest possible bitrate of an MP3 is 320 kbps, compared to 9216 kbps for a 24-bit/192 kHz file when broadcast and 1411 kbps for a CD recording.

Therefore, Hi-Res audio files in 24/96 or 24/192 formats should more accurately recreate the sound that musicians and engineers worked on in the studio. In some cases, you may receive these exact studio files; they are labeled "Studio Masters".

Since the file contains more information, more detail and texture can be conveyed, allowing the listener to get as close as possible to the original performance of the music - as long as their system is transparent enough.

Many new components are capable of reproducing Hi-Res audio. The choice will depend on the size of the system and the size of the budget, as well as the preferred way of listening to music.

Pros and cons of Hi-Res format

During compression, some information is lost. This means that to reduce the weight of an audio file, its signal resolution is also reduced. And this negatively affects the sound quality: in these formats the music is not fully revealed.

The advantage of Hi-Res audio files over those recorded with compression is excellent sound quality. To prove why Hi-Res sounds better than MP3, below is a comparison of their bitrates. The maximum bitrate of an MP3 file is 320 kbps, while 24-bit/129 kHz Hi-Res transmits data at 9216 kbps.

At the same time, the MP3 format is also suitable for most users (especially when listening to audio on the way to school or work). In turn, the Hi-Res format is aimed at reproducing sound the way it was recorded by musicians and sound engineers.

Despite the advantages, Hirez also has disadvantages. Firstly, not all devices have the appropriate bandwidth to play large files. In other words, you cannot listen to Hi-Res on a relatively outdated computer or smartphone.

Secondly, HRA files are larger than audio recorded in MP3. And this requires more free space in the device memory. For example, if an MP3 track weighs 5 MB, then a similar composition in Hi-Res will take up tens of megabytes in the gadget’s memory.

Well, the last drawback is that playing files in HRA heavily loads the processor, so the device may freeze.

Small loss

However, Hi-Res audio has one serious drawback - file sizes. They usually amount to tens of megabytes, and a few songs can easily take up the entire memory of your device. This also makes them difficult to transmit over Wi-Fi and mobile networks.

And that's not all: each of the Hi-Res audio file formats has certain compatibility limitations. Examples include FLAC (Free Lossless Audio Codec) and ALAC (Apple Lossless Audio Codec); both of them theoretically provide lossless transmission of musical information. In addition, there are uncompressed formats - WAV and AIFF, DSD (the format used in Super Audio CD) and the recently developed MQA (Master Quality Authenticated).

The relative merits of each format can be debated, but the first consideration will be their compatibility with audio system components and software solutions.

The main advantage of high-resolution formats over compressed ones is higher sound quality. Download sites like Amazon and iTunes and streaming services like Spotify offer relatively low bitrate compressed formats, such as Apple Music's 256 kbps AAC and Spotify's 320 kbps Ogg Vorbis.

During the compression encoding process, some information is lost; in other words, for the sake of convenience and to reduce file size, the signal resolution is reduced. This affects the sound quality: in these formats, your favorite songs will not be able to fully unfold.

What do you need to listen to Hi-Res audio?

To play Hi-Res, you need a device that supports audio of this format.

Phones

Those who like to listen to music through smartphones should pay attention to the segment of devices compatible with Hirez. We are talking about Samsung Galaxy S8, S8+ and new models. For example, the Galaxy S9 received support for Dolby Atmos and DSD. Among devices not related to this manufacturer, other Android gadgets deserve attention, for example, Sony Xperia XZ2, as well as models released in 2022 and 2022. By the way, these phones support the MQA standard.

Unfortunately, Apple devices do not support Hirez. But this can be corrected with the help of special applications. To listen, you need Lighting headphones or a converter.

Tablets

Samsung is also the leader when it comes to releasing Hirez-enabled tablets. Among the devices from the Korean manufacturer, the Tab S3 (and newer) is worth considering. Huawei also produces similar devices. We are talking about the M5 line.

DACs

You should not rely only on the DACs of your tablet or smartphone, because... their functionality is limited by column sizes and other factors. A desktop DAC or compact USB model will provide high-quality Hirez sound from gadgets that are not designed to reproduce this standard. You need to include a digital-to-analog converter in the chain between the sound source and the headset, and the sound quality of the audio file will increase. As the name suggests, the DAC takes a digital signal and outputs an analog signal.

Rating of the best DACs by sound

Players

An alternative to a smartphone is a music player with Hi-res support. The Walkman Sony Player line and Astel&Kern, Pono models are worth considering for purchase - they have more free space for storing information.

The advantage of a player over a smartphone is that the first device is aimed at playing better quality.

Computers

Windows, Mac or Linux PCs also play Hi-Res sound. But before this, the user should make sure that the software he is using supports Hi-Res.

For example, iTunes in Hirez files does not work on both iMac and MacBook, so you will need to find and download a third-party utility to listen to songs.

Systems

Hirez formats also support other types of devices, including hybrid systems (consisting of a chain: DAC-amplifier-player), as well as TV media players and AC.

Fans of vinyl records do not necessarily need to collect collections in a drawer, because... Players like the Sony PS-HX500 can copy audio from a disc to Hirez.

Why is it better to listen only to Hi-Res and how it is tortured for CD circulation

Why are highres needed? What's so special about them? Why are they if our ear hears up to 20 kHz? Answers in pictures to these and other questions will be discussed in this small investigation.

Let's not ramble on and get straight to the point. Hi-Res and 24-bit in particular are intended primarily for sound engineers. The lowest report of this value (i.e. 24 x 6 = -144 dB) lies far beyond the limits of hearing, which means it allows minimizing quantization errors when editing and resaving material. From here we move on to the next question - does the listener need high-rez?

The short answer is no need. But this is for the listener “vulgaris”, i.e. an ordinary citizen who watches TV and movies on gadgets, listens to the radio. Such a consumer may expect some bass from the subwoofer, but in principle does not think that the phonogram has shades, and the audio file container itself provides different levels of distortion of the original. But now we won’t talk about everyone, but within the framework of the Hi-Fi subculture, right?

And if first-press LPs are understandably prized in the vinyl space, then it’s obvious that the same thing should happen with digital releases. For a music collector, the authenticity of the cast is important, to be as close as possible to the source of the content. And since the vast majority of albums these days are cut digitally, the final Hi-Res mix is ​​where the audio gem collector strives to be. And it doesn’t matter that it’s easy to copy. The trouble comes when it becomes necessary to spoil the Hi-Res original for the sake of lower-level formats.

Let's say you are a publisher and there is a phonogram in pre-production with PCM stream parameters of 24 bit/96 kHz. The above parameters are set only for transfers of archival master tapes, and even then not very often. The most typical case in the music industry is generally the final 24-bit/44.1 kHz mix. But we will still consider higher matter now.

Let’s assume that you, as a publisher, want to please the Hi-Fi community and have not kept this high-res in the studio archive, but have passed it on for sale in Qobuz, HD Tracks, Bandcamp, or somewhere else. Or maybe, on the contrary, they squeezed. In any case, your task is to prepare a master for printing CDs, which everyone is used to, and to make MP3s.

Of course, no one will re-do equalization and other mastering specifically for the CD. Everything that needed to be compressed and equalized was already done in the Hi-Res master. You only need to convert it to the 16bit/44.1kHz standard. And now let's talk about the meaning of the term lossless.

In principle, historically it has been the custom that the contents of a CD in any form - AIFF/WAV or FLAC are called lossless formats. But is this true if the original soundtrack was 24 bit/96 kHz or even 24 bit/44.1 kHz? If we talk about bit precision, then it’s unlikely that a CD can be classified as a lossless original, even if it’s officially released. I propose to study what happens when Hi-Res is decimated to the format frames of the CD standard with a 40-year history.

It is in vain that some figures think that lowering the discreteness from 96 kHz to 44.1 kHz is something like throwing away the stupid emptiness of ultrasound, which no one will hear anyway. the entire signal, including in the audible spectrum, is recalculated into a new grid Moreover, from 96 kHz to 44.1 kHz is also a multiple recalculation; it would be better at 48 kHz.

Any filtering, even digital, entails disturbances in phase characteristics. And the harder the filters are cut, the more parasitic ringing there will be at the pulse fronts. If you don’t filter, then the mirror spectra of the quantization products will come to you. Now we will show you what and how.

For the experiment, I synthesized a 1 kHz square wave signal at 24 bit/96 kHz resolution. Yes, there is no square wave in music, and the recording itself has a more complex shape, but this sample will more clearly demonstrate conversion artifacts. Let's try to bring this signal to the CD standard.

We'll deal with the bits later, but now let's first reduce the resolution to 44,100 Hz. Let's throw away that stupid emptiness of ultrasound, especially since there is only one signal and at 1000 hertz, yeah, yeah.

Various audio software offers their own Sample Rate Convertor (SRC) options. In the Izotope RX interface it looks like this (see menu below). Some techies use a ready-made preset without looking. Someone more responsible (which happens much less often) will select the settings manually each time and listen to the result. Personally, this procedure always drives me crazy because I never like the result 100%.

The SRC setting is reminiscent of the digital filter parameters in DACs, only of course that the digital-to-analog conversion itself does not occur. The slope of the virtual frequency response decline, the distribution of parasitic “ringing” oscillations - before or after the pulse, etc. are set. I'll show you a few settings and the results they produce.

The filter always heals something and cripples something. A hard cutoff of the spectrum of quantization products leads to an increase in oscillations at the pulse fronts, the so-called “ringing”. Moreover, the “ringing” is located both after the impulse and before. This, as we know, is the case of a Sharp-type phase-linear filter. The slope characteristic may deviate slightly from the vertical axis, but ideally it is a “right angle” (see the picture), which is also called a Brickwall. Chord DACs love to filter this way, but not everyone is happy with it.

If you want to get rid of the pre-ringing, then you shift this energy to “later”, to the tail of the impulse, summing up the vibrations in this area. Do you recognize the minimum phase filter?

Both cases above refer to the classification of the amplitude-frequency response rolloff according to the Sharp type. If you want to reduce the “ringing”, weaken the slope of the filter, make it softer, i.e., Slow. But at the same time, the purity of the spectrum in the upper range will suffer. Place the extreme minimum point on the boundary of the range - the filter will begin to influence and partially dampen the useful signal in the operating range. Below you can see the results of the operation of the Slow filter in phase-linear and minimum-phase output. As you can see, the waveform is a little closer to the original.

The less filtering, the sooner the pulse fronts stop ringing. But don't expect any specular spectrum protection either. If you move the cutoff point forward beyond the range until the most free development of events, as in NOS DACs, more and more parasitic quantization products will penetrate into the useful signal.

Well, have you chosen your fighter? How will we filter? Or is it better not to torture the signal with processing and listen to Hi-Res in its original form?

The second stage after resampling the discreteness of 44,100 Hz is reducing the audio bit rate to 16 bits. After all, the audio stream is still 24-bit, and this is not suitable for CDs. To show the following degradation, for the second experiment I generated a 24-bit/44.1 kHz signal at minus 90 dB. This is the same 1 kHz, but already of a sinusoidal shape, in order to evaluate its gradation during conversion. Why is it so quiet in volume?

Because minus 90 dB is 6 dB from the CD dynamic range limit, that is, by the least significant bit. Don’t forget the ratio of 6 dB per 1 bit, i.e., 16 x 6 = 96 dB of theoretical Red Book resolution.

At 24-bit resolution, a sine wave of this level has a beautiful, smooth shape. You can see for yourself the points from which the sine curve is constructed.

If we try to express the same signal in 16 bits, it will look like “steps”, because one bit cannot be cut in two. You can discard the least significant bits of 24 bits or immediately try to record the sine of this volume level in a 16-bit environment. The same thing will happen if you send a 24-bit signal to a vintage 16-bit DAC. You will return to this miserable form in any case. How can we preserve (at least partially) the smoothness of 24-bit quiet transitions on lower formats?

An absurd, at first glance, method comes to the rescue. When converting from 24 to 16 bits, you should add a small (at the level of the least significant bit) portion of noise with an emphasis on the HF region - above 15 kHz, where it will not be so noticeable. What will we see after such a procedure?

Externally, the signal became “shaggy” and its amplitude increased. Pay attention to the samples, they now occupy higher levels, individual peaks reach as much as -78 dB. This result seems unsightly, but its overall silhouette already has smooth, rather than stepped, outlines. And if we profile the “hairworm” with a high-pass filter (and such a procedure will certainly be done in any DAC), then we will again see our sine, which was considered lost. Crippled and crooked, but they’re no longer “steps,” isn’t it, a miracle?

Thanks to this trick, the quiet dynamic tones of 24-bit recording are pulled into the audible realm of 16-bit resolution. In this case, the maximum signal levels remain unchanged due to the negligible amount of smoothing noise in the region of the least significant bit. This is how the noise shaping procedure works, as a special case of dithering.

This trick is not new. In the 80s, it was first used by Philips in order not to lose CD resolution on the 14-bit TDA1540 DAC. At the very end of the 80s, Technics and JVC introduced 1-bit MASH (MultistAge noise SHaping) and PEMDD (Pulse Edge Modulation Differential-linearity-error-less D/A conversion) converters using PWM modulation. Throughout the 90s, noise shaping was used to increase the dynamic range to 20 bits in HDCD and Sony's proprietary Super Bit Mapping (SBM) technology, remember CDs with such markings? And of course, noise shaping is the fundamental basis of DSD technology, which we adore.

Dithering is suitable not only for sound recording, but also for image processing. When converting a high-bit image into standard 8-bit, in order to avoid posterization (i.e., torn color gradation), smoothing noise is mixed into it. What did we get?

In this case, the amplitude of the reconstructed sine increased from -90 dB to a level of -82 dB. Please do not treat this figure as a complete formula. The amount of noise shaping/dithering is also adjusted separately in the audio editor. The larger it is, the louder (i.e., more noticeable) the low-level signals from the 24-bit original will be drawn out. But this algorithm can be used up to a certain limit, so as not to degrade the overall signal/noise of the phonogram. And you definitely shouldn’t repeat this procedure twice.

So, we have audio in a standard suitable for placement on a CD. But should we still call it lossles, i.e. lossless format of the original data table? It depends on what we consider to be the original. This can only be said in relation to more crude and destructive methods of reducing the size of an audio stream, such as MP3 and other lossy codecs.

By the way, sometimes you hear contemptuous attitudes towards MQA. Like, this is a nasty lossy codec with compression and data loss. Yes, MQA is obtained by trimming the original Hi-Res, but to a much lesser extent than CD! Even unpacked, MQA is still a step above CD as it contains audio at over 16-bit resolution, as opposed to the limits of Red Book. That is, MQA will have more reasons to be called lossless.

You can, of course, still compare all this with each other by ear, if there is nothing else to do. Hi-Res originals vs. planed 16-bit blanks for CD. Add more MQA there. In some cases, the samples will be indistinguishable from each other. In others the difference will be more obvious. But such results can be achieved even when playing at lower levels with CD and MP3 320 kbps. Then why guess every time? After all, there are a lot of influencing factors here - the repertoire, the dynamic and frequency characteristics of the phonogram and other things. And there is no single protocol for SRC algorithms for reducing the bit depth of Hi-Res. As already mentioned, this is a studio decision that is made and executed in a way that is not obvious to us.

So if we are talking about a new release (and comparing remasters with previous releases is a separate topic and we are not talking about it now), it is advisable to purchase the album in the highest possible audio bitrate, no matter how empty the spectrum hangs over it. This will help you avoid and not reflect on the topic of all these methods of degreasing the original discussed above.

Hi-Res makes it possible to have a digital first press, i.e. be at an extremely close distance to the musician. And, importantly, without violating the quarantine distance rules!

Where can I get music in Hi-Res?

Everything is clear about how to listen to music in high definition. But where to get Hi-Res files? There are many sites and services that offer the opportunity to buy and download individual songs and even albums in Hirez. They are available in Europe and the USA.

Below is a rating of such sites. Before making purchases on a specific site, it is recommended to check information about the desired songs on other platforms, because... There they may be sold cheaper or even be free.

Some of the services described below may not work within Russia, so you should download a VPN first.

Primephonic

This platform is suitable for lovers of classical compositions. The site's catalog includes thousands of works in resolutions from CD and FlAC (Lossless) to DSD. The compositions are available for download both individually and in albums.

The store developers do not give up on their narrow specialization and update the file catalog weekly. And if the user needs to find a specific song, the search icon will help do this.

7Digital

A large music platform that provides users with a selection of music in FLAC and MQA. Services for downloading are provided here. In addition, 7Digital applications are available for all operating systems. You can buy both individual songs and albums.

HDTracks

A large music store with many genres of music available in its catalog, including classical, jazz, pop, rock (for example, Anthropillo), etc. The content of the site is controlled by musician David Chesky.

Audio files on the site are available in many formats, incl. in maximum resolutions and even in CD. On the platform, the user will find any compositions, ranging from works by American and English musicians to works by Bob Marley.

However, the site also has a drawback. Users are fed up with the outdated design, which will be difficult for a beginner to understand. In addition, the design of the site is aimed at adults, so it alienates music lovers among young people.

Qobuz Sublime

A site that will delight people with a high-quality selection of tracks, as well as the ability to select music by genre, novelty, and even sampling rates. The Qobuz catalog is expanding daily with new sections. However, users note that it perfectly implements French music.

Subscription prices are comparable to those offered by competitors. But if the user signs up for a hybrid subscription (its cost is 350 euros per year), he will receive discounts on the purchase of albums in Hi-Res formats.

Tidal

Hi-Res music streaming service, launched in 2014 by the Norwegian company Aspiro. In addition to 60 million files in various formats, there are 250,000 video clips available. The advantage of subscribing to Tidal is that the platform periodically broadcasts concerts or events that are open to subscribers.

The software is available for download in the application store. For new users, there is a promotion in the form of a free 30-day subscription, which can be canceled at any time.

There are other popular music services, such as Spotify.

Where to store and how to listen to Hi-res music?

On a modern laptop or desktop computer, hard drives are used, measured in terrabytes in size. There seem to be no restrictions on volume. However, videos and music libraries on CDs, DVDs, and even Blu-rays are gradually becoming a thing of the past, and you don’t want to erase music (or a movie) once recorded, especially if it was downloaded legally, with payment of a price.

There is a solution: network data storage. These are relatively small boxes with capacious storage units inside. To access information, a regular network connection is sufficient.

Another advantage: to listen to your favorite fragment, it is not necessary to turn on a personal computer with a screen and a powerful processor. Music equipment is connected to the storage, and you can get simultaneous access to different music fragments.

Popular manufacturers of Hi-Res acoustics

In order for users to choose reliable Hi-Res speakers, below is a rating of manufacturers of wired and wireless equipment.

The principle of digitizing analog information

For the average user, the graphical representation of any analog signal looks like a sine wave.

To represent it digitally, a smooth curve should be divided into a broken line of peculiar steps. Each horizontal element determines the signal level (brightness, volume) and duration.

When digitizing, the smoothness of the sinusoid is lost. It is obvious that the more often the steps are formed, the more accurately the shape of the original signal is transmitted - the quality improves.

The downside of technology is an increase in the volume of information.

Many companies working professionally in the field of sound recording have long been looking for a compromise: how to maintain acceptable sound quality without turning musical fragments into huge repositories of digital information inaccessible for everyday use.

SONY found a middle ground, jointly with PHILIPS introducing the CD DA (Compact Disc Digital Audio) format to the world.

The duration of the digitized music was 74 minutes. Beethoven's ninth symphony was taken as a time standard: the developers set themselves the task of ensuring continuous sound of this work on one side of the disc (initially all DA CDs were single-sided, with a picture with information placed on the front surface).

The main digitization parameter is the sampling frequency: in CD format DA is 44.1 kHz: 22.05 kHz per channel in stereo. In theory, this was enough for the vast majority of listeners: a normal person without unique abilities perceives a frequency no higher than 20 kHz. Without going into too much technical detail, this sample rate requires a 16-bit recording format.

Further, the development of technology was aimed at compressing data in order to fit as much music as possible into the CD format. This made it possible to “squeeze in” a soundtrack for video fragments, and most importantly, to transmit sounds via the Internet. Of course, the inevitable decrease in sound quality caused a wave of dissatisfaction among music lovers, so new digitization formats were developed in parallel, where the size of the data did not matter.

Recommendations: 5 Best Wireless Hi-Fi Headphones

, 6 Best Hi End Headphones

6 Best Studio Headphones