Tag: how lossless audio works

Flac compared to Mp3

Lossy vs lossless

For an introduction to the topic, let’s start with a simple reference that connoisseurs of audio formats can safely skip … Most of us today listen to music in a car in MP3 format. It is convenient: a huge music library fits in a fist, it is not afraid of scratches, bumps, dirt, it can be easily replenished and edited on a home computer, etc. The MP3 format suits almost everyone except those who are “boosted” due to its internal nature …

And the gist is that when the music stream is compressed, those sounds and frequencies are cut off according to a special algorithm, which, according to the algorithm’s creators, ALMOST does not affect the listener’s perception of the music. This “almost” is the reason for the compromise of the MP3 format … In the music of the club genres, the difference is barely perceptible, but in the good quality rock, instrumental, serious vocals and classics reproduced through a system of decent speakers, the Many may experience the difference between the “cut” MP3 format and the “uncut” WAV source.

MP3 does not convey the depth and fullness of the sound of music, it is perceived as a “poor relative” compared to uncompressed audio formats. All these are, of course, subjective characteristics, and not everyone, due to the physiological characteristics of hearing, is able to “taste” them at all, but the truth is that MP3 and other codecs that compress sound at a loss ( WMA, AAC, and others) degrade the quality of music for the sake of ease of use.

original-3.jpg20151019-2336-1k348sb.jpg
An alternative is “lossless” formats, which are commonly referred to as lossless. Codecs like FLAC and APE also compress music, but during playback the original data is fully restored from the compressed state, unlike MP3, from which compression algorithms remove “extra notes” at their discretion and without your knowledge …

FLAC in cars
Until recently, it was possible to listen without loss in the car mainly only in multimedia centers on Android, since for Android it is natural to install various playback programs that read all formats. But the sound quality was questionable, because most of the Android-based recorder manufacturers, with dubious Chinese ancestry and a blank space in the brand’s history place … Also, the serious restrictions on the distribution of these devices are the high price and two … din format, which also does not suit all.

And now, a year and a half ago, the first FLAC models started to appear in the segment of cheap radio tape recorders from famous brands, with good quality, one-din design and the usual “radio” design. Today, this market segment has taken place: a variety of inexpensive FLAC devices from all well-known brands such as Sony, Pioneer, Kenwood, etc. already strong. Consequently, you can try to draw some conclusions.

The main one will be the answer to the question: was the introduction of FLAC support in low-cost massive devices a breakthrough in the evolution of car audio, or is lossless relevant only to serious and expensive car audio systems, and a simple user in an inexpensive car with standard acoustics does not have different advantages of FLAC will not receive?

What are we testing
Today, the cost of the most affordable car radios with FLAC support is extremely democratic, and these devices are available to everyone: in the initial segment, their prices start at 3,000 rubles. Thus, such equipment leads many budget car owners to the idea that in a budget car, with consumer standard acoustics, you can significantly improve the sound quality simply by replacing the radio recorder and using a music source from high quality: FLAC. instead of MP3.

Advantages of the FLAC audio format

Wave is uncompressed or lossless format, while MP3 is compressed or lossy.

Technically .wav is just a container format and can contain various types of compressed or uncompressed audio, but you will usually find that it contains uncompressed LPCM audio (same as on audio CDs). With .waves files you essentially end up with a raw digital representation of the audio bitstream. Analog sound produced in the real world contains essentially an infinite amount of information because it is a constantly changing wave (see below). To digitize these sounds, you must sample the signal at different intervals to get an approximate sound. For .wav, the audio signal is typically sampled at 44,100 times per second or more, and each sampled value is recorded so that the audio wave can be played:

MP3s are compressed to compress the same audio information into a smaller file size. The .wav format is great for representing an analog signal very accurately, but generally at the expense of large files, as you probably know. Compressed audio (and video similarly) is designed to reduce file size while maintaining an acceptable level of fidelity. In simple terms, compression tries to remove unnecessary data from the stream and reduce the signal to its most necessary components. In MP3, the encoding and compression algorithms use a model of how we listen to analyze the sound in the frequency domain and remove any unnecessary information. For example, due to hearing disguise, if there are two sounds at close frequencies, we will often only hear a loud sound if the difference in volume between them is significant. Therefore, for MP3s, the lower volume sound may be lost and the sound will sound essentially the same as it does to our ears. Find out more about the technical aspect of MP3 encoding here.

In practice, both .wav and MP3 have their uses. For production, .wav is the standard because it will almost always be a 100% accurate, bit-level reproduction of the source material. MP3 can be a decent alternative at high enough bit rates. Bitrate is a measure of how many bits per second the MP3 encoding will use. This means that the higher the bit rate, the closer the MP3 will be to the original uncompressed stream. Bit rate is generally measured in kilobits per second (kbps). I like high-quality sound for my digital music collection, so when I get the chance, I usually encode MP3 at a constant 256 or 320 kbps. This is the upper limit of what MP3 can do, and unfortunately most digital music isn’t encoded that high there. When the bit rate drops, it is usually heard first in the high frequencies, like the cymbals of a drum set. 160 kbps is tolerable, but anything below and you’ll really start to notice. But again, with a fairly high bitrate, the differences between MP3 and .wav subtle, especially for an inexperienced listener (most listeners).

For .wav files, we mainly look at the bit depth and the frequency or sample rate. Bit depth is the number of bits used to encode each sample value. The sample rate indicates how many times per second the audio is sampled. CD (.wav) and MP3 are encoded at a sample rate of 44100 Hz (Hertz stands for “cycles per second”). Newer computers and audio hardware / software now support higher sample rates, including 48 kHz or 96 kHz. For .wav the bit depth is usually 16 or 24 bit on newer systems. For most purposes, 16-bit and 44.1 kHz is sufficient when using .wav, but if you have this capability, it’s generally worth switching to 24-bit, 48 kHz.

Some examples of file sizes for 5 minute stereo recording:

.wav, 16-bit, 44.1 kHz: 50 MB
.wav 24 bit 48 kHz: 82 MB
.wav 24 bit 96 kHz: 164 MB
MP3, 128 kbps, 44.1 kHz: 4.5 MB
MP3, 192 kbps, 44.1 kHz: 7 MB
MP3, 320 kbps, 44.1 kHz: 11 MB
FLAC, 24-bit, 44.1 kHz: 28 MB
FLAC, 24-bit, 48 kHz: 31 MB
24-bit 96 kHz FLAC: 61 MB
There is also a variable bitrate option for MP3 encoding, which should have a slightly smaller file size for the same quality. It uses an encoding scheme that changes (alters) the bit rate for different parts of the song, depending on the complexity and how many samples are needed to accurately recreate the section.

[FLAC] There is a third category: lossless compression. FLAC is a good example of this, and has the quality and fidelity of a .wav file, but with smaller file sizes

Differences between FLAC and MP3?

Lossless vs lossy

“Here, of course, the question is not about the difference between MP3 and FLAC, it is broader: that lossy compression formats (MP3, AAC, WMA, Ogg Vorbis and others; that is, lossy) are They differ from “lossless.” ”(FLAC, ALAC, APE, WavPack and others; that is, no losses). Actually, with such wording, it becomes clear that in the first group of formats, the original data is not completely saved, and the second can be restored to the original format (for example, Wav or Aiff extracted from CD) without loss. What exactly is lost and in what proportions depends on the specific type of lossy files and their bit rate, that is, the degree of compression. But to say that all MP3s sound bad and that “flacks” are perfect is the height of arrogance and incompetence. Lossy audio formats have been developing for more than twenty years, and serious research laboratories (Fraunhofer Institute, for example, in addition to working on MP3, is also famous for the invention of the most efficient solar battery) and a group of enthusiasts. Mathematical encoding is constantly improving, and nowadays it is no longer so easy to distinguish files produced by different codecs by ear.

I would immediately make a reservation that not only the files themselves are important, but also the equipment they are to be tested on, the listening environment, and the examiner’s listening experience. In MP3 of any low bitrate, Ariel Pink will sing with the voice of Ariel Pink, of that there is no doubt. It is quite possible that a person listening to music as a melody through white headphones in a subway car will be enough for the eye, and the difference in codecs will come down to a file size comparison. A disc jockey who is embarrassed to buy or search without losses will also think that everything is in order with his MP3, as he prepares a set on the “Tractor” on the laptop’s built-in speakers. It is true that during a party in a big, loud and clear audio system of the club (sometimes they meet, believe me), it suddenly turns out that the guy who speaks immediately after for some reason, the music became big, clear and great . Lossy formats are developed for the convenience of transferring files over the Internet, for storage on portable audio players, and finally for personal playback. Okay, it’s silly to watch a gigabyte AVI movie on a big screen. Even in a home theater, this is not entirely decent. The same goes for MP3. On your iPod: listen to your health (although AAC from iTunes sounds better for sure), but if you go clubbing, don’t miss out, even if you start Skrillex. And when you’re listening to Christmas jazz with his girlfriend’s parents on their big lacquered speakers, buy FLAC or ALAC too. With MP3, you risk getting into an awkward situation. In theory, after a bit rate of 256 kbps, it will be quite difficult for your future audiophile father-in-law to know if he is lossy or not.

Usually when viewing an MP3 file, he isn’t paying attention to anything other than the bit rate. If he already considers himself a person with a taste for music and sound, he should look in the properties of the file for the data of the codec that was used during the conversion. Suppose you see “Lame 3.99” there, it means the latest MP3 codec was used and you’re in luck. But next to it is “Joint Stereo”, which is no longer great. This means that to save a couple of percent of the file size, the codec was allowed to add something to mono, although the recording is stereophonic and the sound image is slightly lost in depth and clarity. There are also fully botanical CBR or VBR, ABR and UBR, but if you’re ready to dig that seriously, do it yourself. Well, you figured out the properties of the file, everything is simple there. The difficulty is this: you hardly ever know what your 320 CBR Stereo is made of.
Scammed out of Internet radio? Made from an unremastered original Japanese CD? Recoded from 192? There is a lot of music on torrents or Soulseek, but there are few guarantees. Another complication is that lossy formats slightly increase the peak values ​​of the audio signal. The so-called overshoot: thousands of micro-overloads along the entire length of the file. Again, you won’t notice this on a train with an iPod. And the future father-in-law can hear ”. The so-called overshoot: thousands of micro-overloads along the entire length of the file. Again, you won’t notice this on a train with an iPod. And the future father-in-law can hear ”. The so-called overshoot: thousands of micro-overloads along the entire length of the file. Again, you won’t notice this on a train with an iPod. And the future father-in-law can hear. “