A layman friendly guide to Lossy and Lossless music

What is Lossless Music? (Technically)

Lossless audio is a type of audio which haven't discarded any type of frequency or audio signals and is raw or the same as its source.

Now there are two types of lossless audio too!

Compressed Lossless 
Uncompressed Lossless

That's where it gets confusing for most people, they say if it is compressed how is it lossless? as compression leads to loss of data right? You are both right and wrong. Read more to understand.

Let's first know the formats of Compressed and Uncompressed Lossless Audio.

Uncompressed Lossless Audio Formats-

WAV = Waveform Audio File Format

DSD = Direct Stream Digital

PCM = Pulse Code Modulation

AIFF = Audio Interchange File Format

Compressed Lossless Audio Formats-

FLAC = Free Lossless Audio Codec

ALAC = Apple Lossless Audio Codec

APE =  Monkey's Audio 

WMA = Windows Media Audio

Now let's understand how this lossless audio compression works? And how it doesn't affect or lose quality?

Suppose there is a file that contains alphabets like -

AAAAAAAWWWWWWWOOOOOOOLLLLLLLLLPPPPPPPPPPSSSSSSSS

That's 48 characters above, now if I want to compress it or to say make it more efficient in less space then I would write it as 

Ax7Wx7Ox7Lx9Px10Sx8

That's still 48 characters but in just 19 characters now. Not even a single character is lost in this way. 

That's how WAV and FLAC work.  

WAV saves all 48 characters without arranging them whereas FLAC or any compressed lossless audio Format arranges it to make it more space efficient. 

Now let's understand the most common things you as an audiophile should know.

First, what is lossless in actual life to us?

Anything which sounds better than the lossy versions you had already. This is a complicated thing to explain but for starters, Lossless audio ranges from 16-bit to 40 kHz and above. Anything that ranges below is not lossless. Except if the audio was recorded originally in a lower frequency, then that would be lossless. 

What is this "16 bit" or "44.1 kHz"? 

16-bit, 24-bit, or 32-bit represents the audio bit depth of audio. What does that mean? It means the audio bit depth can improve the audio's dynamic range. But now what is dynamic range?  Dynamic range means how loud or low can the audio frequency be in an audio file. 

So basically, if your audio files were recorded or processed at a higher bit depth, then that file can have sounds of instruments that range way too above in frequency reach or low too. 

44.1 kHz, 48 kHz, 96kHz or 192kHz what is this?

This is how you measure the frequency. kHz, where k represents kilo and Hz represents Hertz. What is the unit of frequency? Hertz. What is the meaning of Kilo? 1000. 

Hence when someone says this is a 96kHz audio file, check its spectrogram to know whether it has 96000Hz or not.

I see only 22 kHz in Spectrogram in a file of 44.1 kHz. Why?

Non-technical fake reason for a very layman-

Yes because the songs are generally in stereo i.e. two channels. Left and Right. Each channel carries 22kHz so, 22kHz from left and 22kHz from right = 44kHz. The Spek app only analyses the single channel's sound. For multi-channel viewing use SoX.

Actual reason - 

Nyquist theorem states that a 44.1 kHz sample rate can accurately capture frequencies up to 22.05 kHz, not that each channel carries 22 kHz. 

How to check the spectrogram? Also, what is a spectrogram?

A spectrogram is a visual representation of audio frequencies. You may have seen that in movies or shows when a person is on a ventilator and the people around him are looking at his ECG readings. That is what? Spectrogram of heart. 

How to check the spectrogram?

You can use Spek. Just open the audio file in this program. 

I have opened the spectrogram but how to read it :) ?

Here is a spectrogram of Kesariya's track from the movie Brahmastra. It was released by Sony Music in 96 kHz and 24-bit on Qobuz. Its size is above 100 MB. But this is an upscaled track. Its actual quality is 16-bit and 44.1 kHz.

How do I know that? 

See the image above and see the solid bluish frequencies ending just above the 20 kHz. After that, there are just sprinkles of that bluish-dotted frequencies. What is that? That is noise. So what they did was, they just added noise so that you people can not sue them or charge them with forgery. 

These people are very clever. How? See they have added noise just above the 22khz so that we humans won't be able to detect it. Noise is basically the hissing sound. If the hissing sound was under 20kHz we will know but after that, no frequencies or audio signals can be heard by any human ear. So they added just above that so that we won't know about the upscaled music as not everyone analyzes it. 

So, STOP ASKING OR BLINDLY DOWNLOADING THE 24 BIT 96kHz AUDIO, YOU ARE JUST WASTING YOUR STORAGE AND INTERNET DATA AS MOST OF THEM I.E. 90-95% ARE FAKE.

Why are they upscaling music?

To earn more money. They know, people just want the best 24-bit, HiRes or MQA, etc. So the moment they publish it as hires, it works as a catalyst for them. And this is not just for Indian music, international music albums are facing it too. For example, Eagles' Hotel California is also available in 24-bit 192 kHz on Qobuz but :) 

This too is upscaled.

Elvis Presley's hit song Can't Help falling in love too.

Imagine if the official providers are doing this, then the telegram channels which are providing 24-bit ALAC rips☺ RIP🙏

Are Lossy Formats any good?

MP3, AAC, OPUS, and OGG. There are the formats that are most commonly used.

AAC i.e. Advance Audio Codec is used by Apple Music, Youtube Music, and Saavn.

OPUS is used by Youtube.

OGG by Spotify.

Now the best of all these is AAC by Apple. Why only by apple? Why not YT music or Saavn? Because AAC is also of two types. AAC compression by Apple and AAC compression which YT music and Saavn use.  

AAC by Apple is just similar to any lossless format. Only 1-2% quality that also you can't hear. 

So it's better to get an authentic AppleRIP than the fake ALAC file. 

What is a fake ALAC file?

Since Apple has introduced lossless music in the Indian region, people expect ALAC rips i.e. lossless versions of albums from the labels like T-Series and Zee music as they don't release their songs on Tidal or Qobuz (as of now).  

Again now many people know that people are searching for ALAC files hence they start creating a telegram channel or forum or something like that where they upload their fake ALAC files.

These files are made up of programs like Tunefab, Sidify, or something similar like these. What these programs do is they record the playback stream of apple music. In simpler terms, when you install Tunefab on your PC and ask it to rip a song. In the background, it will play the track on apple music in 256 KBPS and record it 24-bit 1000+ KBPS. It's like storing 1L of water in a 10L container. After 1L everything will be air. 

Basically converting 256 kbps stream into FLAC. Again recording tracks like this results in losing quality. 

So that's why fake.

Is ALAC any better than FLAC?

No both are lossless compression. A track ripped from CD in ALAC and FLAC will be identical. There will be no difference between them. The only difference is that ALAC is made by Apple and FLAC was open source by Xiph org. 

Hope this post was able to clear the basics of audio. Also free to correct or give any suggestions to what I wrote. 

Thank you for reading.