lossy lossless uncompressed file formats Archives

Data compression: lossless or lossy

Free Download Mp4Gain

To be able to store large amounts of data, a huge amount of storage space is required. When the space is at some point full, a new storage space is usually made available. Mainly in the form of new hard drives, server systems or the like. But there is another way to store more data: data compression.

Lossless vs Lossy

Data compression works like a pillow you squeeze
Figuratively speaking, data compression works the same way as when you squeeze a pillow. Not all air is required between the pillowcase and the feathers. By squeezing the pillow, you compress it by removing the air (as much as you can); the pillow becomes smaller, that is, more compressed than before.

It is similar with data compression. Here also items that are not absolutely necessary, in this case data, are removed and the storage space still required by the remaining data is reduced.

Lossy vs Lossless Compression

Different lossless compression encoding methods.

When it comes to data compression, a distinction is made between lossless compression and lossy compression, so it can be further subdivided into different types of compression.
Lossless compression is generally only done with text files. If the text is saved as a character string and not as an image or the like, the storage space can be saved using the dictionary method, for example. Recurring parts of the string are simply replaced by abbreviations.

Text: Hello world. Hello user
Encoding: world X5. X5 users.

With the help of Run Length Encoding (RLE), identical text components that are placed one after the other are only saved once. This can also save storage space.

Text: Today was beautiful, beautiful, beautiful and tomorrow will be more beautiful.
Coding: Today was / 3 / beautiful, / and tomorrow will be more beautiful.

In entropy encoding, as is the case with Huffman code or arithmetic encoding, for example, text elements are sorted in binary and encoded according to their frequency, and the most frequently occurring element is given the binary number smaller.

With these numbers, the text elements are stored in a separate dictionary.

Text: IF YOU FLY BACK, FLY, FLY, FLY, FLY AFTER.
Coding: 10100 1 1 1101 1 1 1 11

Difference between lossless and lossy compression

The benefit of lossless encoding is clear: it ensures that all “compressed” data in the compressed data packet can be accessed and restored.

In addition to lossless compression, there is also lossy compression. Irrelevant information is not encoded here, but is directly removed. Therefore, we speak of a reduction in irrelevance.

Lossy compression: the MP3 format

Of course, this procedure can lead to extremely high compression rates. However, this irrelevant information is irrevocably deleted. This means that the original state cannot be restored after compression.

Lossy compressions are often used for image, audio, and video files, as they often have to handle large volumes. The MP3 format is a classic example in this context. Here, frequency patterns are removed from an audio file that are almost inaudible to the human ear. This saves storage space.

If you compress data packets, you can use your own data storage capacity better than uncompressed data packets. However, you should always consider the type of compression you are using. Because lossy compression is not always desirable.

Free Download Mp4Gain

Mp4Gain Main Window

Mp4Gain Features

Free Download Mp4Gain

Uncompressed and compressed formats

Popular audio formats

Uncompressed formats

There are audio formats that do not have compression and are the best in terms of sound quality. On the other hand, they take up much more memory space and a lower transmission speed than compressed formats. With professional software such as Pro Tools, SuperCollider or Max, you generally work with files of this type. The two main formats are:

WAV (wave). Audio format developed by Microsoft and IBM for compatible IBM PCs. In this format, musical instruments feel the same regardless of the PC on which the file is played (with the same acoustic quality of the hardware components, of course).

AIFF (Audio Interchange File Format). The format developed by Apple is also called the Apple Interchange File Format.

Audio formats

Lossy compression

It allows for more compression, but at the expense of sound quality. Lossy compression methods generally tend to discard information considered useless, keeping only the essential ones and arise from the idea that not all frequencies contained in a sound spectrum are perceived by the human ear. Then the high frequencies are cut off, which are believed to be the least distinct in our hearing. Obviously, the more frequencies you cut, the more the space occupied by the audio track decreases and with this also the quality of the result as a reconversion process. It does not allow the complete restoration of the cut-off frequencies. Let’s see what are the main audio formats of this type:

MP3 (MPEG-1/2 Audio Layer 3). Audio compression algorithm capable of drastically reducing the amount of data required to reproduce a sound, compromising obtaining an almost faithful reproduction of the original uncompressed file. The best codec for compression is Lame. maximum achievable bit rate 320 Kbps.

WMA (Windows Media Audio). Audio file standard invented by Microsoft. Audio compression format very similar to an MP3.

OGG (Vorbis). It is an open source algorithm and with the same perceived quality, it allows greater compression than the MP3 format, obtained through advanced psychoacoustic research.

AAC (advanced audio coding). It is an audio compression format officially included in MPEG-4. Provides superior audio quality to MP3 format with more compact encoding. Currently it is used mainly by Apple, which in the variant that manages copyright has a compression of 128 Kbps (the standard of the iTunes Store) and corresponds to that of an MP3 at 192 Kbps at constant bitrate. Conversion times are a bit slower than other formats.

AC3 (Dolby Digital). Dolby Digital is a multi-channel audio coding system developed by Dolby Laboratories Inc and used in film, digital TV, laser discs, DVDs, and other digital audio streaming or playback media. It works from a minimum of 96 kbps to a maximum of 640 kbps. In cinema, Dolby Digital is used with only 320 kbps of bandwidth, as it prints in the limited space available between the slide holes of the films. Generally, in DVD it is used with a bit rate of 192 kbps to encode stereo signals (2.0, 2.1) or stereo surrounds, and with a bit rate between 384 and 448 kbps for 5.1 signals.

Lossless compression

These compression methods try to reduce the space occupied by the track without going to touch the sound. The compression rate is much lower than with lossy methods, but there is no loss of quality and the sound is identical to the original when converting. Let’s see what are the main audio formats of this type:

FLAC (Free Lossless Audio Codec). The widespread open source audio codec is currently well supported by various audio software. Unlike most lossless compression algorithms (like ZIP and gzip, for example) that only achieve 10-20% compression, it achieves significant compressions, on the order of 30-50%.

APE (Monkey’s Audio). No lossy format that allows us to reduce the space our music occupies by approximately 50% (in some cases even more). Currently it is no longer developed.

ALAC (AApple Lossless Audio Codec). The format developed by Apple stores data in an MPEG-4 container with the extension .m4a. It does not have Digital Rights Management (DRM) and is now deprecated.

Files without compression, lossless and lossy formats

Files without compression, with compression but without loss and with compression and with loss … what does that mean?

There is currently a huge confusion in most people about audio formats and that is why we are going to give an explanation that is as simple and clear as possible so that most of the people who read can understand enough to be able to make decisions about it. .

We will not try to get into technicalities or knowledge What are for specialists or connoisseurs or programmers, such as those who developed the mp4gain, but rather we will talk about general concepts at the beginner level but that can allow a correct understanding of the matter.

So the first thing we must understand is that the quality of an audio file depends on the amount of details it contains and for it to contain many details, it takes up a lot of space on the hard disk.

We could say that there are three groups of audio file formats:

– audio formats that do not have compression such as WAV, AIFF, etc.

– formats that compress audio but do not have a loss of information. FLAC, MPEG-4 SSL, WAV PACK, etc.

– There are also the formats that the audio compresses but that do have a loss of information, for example MP3, Ogg, HERE, ETCÉTERA

Now let’s clarify what it means that some formats do have compression but do not have loss of information while others also have compression but do have loss of information.

The formats that are compressed but have no loss of information what they do is something very similar to zip the file. In other words, they simply compress.

The original information but they do not discard anything, therefore when unzipping the file the original file is obtained again that had all the details and therefore is of very high quality.

On the other hand there are files such as MP3 that are compressed but also to achieve greater compression what they do is that they get rid of some part of the information. For example, those frequencies that are not audible to the human ear also get rid of that information that is redundant and they also use the method called masking which part of the idea that a sound that is very loud and is very forward It will be able to mask other sounds and the human brain will not perceive those other sounds. Therefore, according to this theory, they can get rid of these sounds that have been masked and the file will sound more or less the same to the human ear.

This concept of sound masking was one of the great milestones that made MP3 become so popular because it could compress files a lot, at a time when disk space was not superfluous, and yet they sounded pretty good to that ear. era and those people.

Also the programs that made the normalization of the audio volume, achieved results that at that time sounded good, but at this time the listening ear as a has become much more demanding and sophisticated. And so we need much more modern normalizers with much more algorithms. more powerful like mp4gain.