Why is the sound quality of music different when you get it from different sourc

[ChunYinZi]

This question has been bothering me for a long time.

As you know, some libraries have proxies on different websites.

I did some tests. I downloaded the same piece of music on these sites.

They are all in WAV format, but I compared the spectra and they are not the same.

The best of them, in my opinion, was the CD ripped data, which had a very full spectrum and I could even hear the difference in the size of the original source (I tested the FC-E30 album from Firstcom Music).

It's my understanding that more and more video library agents are forcing the original masters to be upgraded to 24 Hz instead of the original 22 Hz. This changes their original sound.


CDs aside, the spectrum of this digital music is slightly different, so should I use the WAVs from their original websites or the WAVs from big companies like APM Music or Universal Productions?

So if there's an album I really like, I'll buy it on CD and rip it, which I understand is the best sound quality.

Do you guys have a different opinion?

[ChunYinZi]

Also, I'd like to mention something about the library music represented on this SourceAudio site

They're the last WAV choices I'd ever want to download.

For one thing, the spectrum of the WAVs downloaded from their site is very poor, barely reaching 20hz, and very little of the music reaches 22hz or more.

In addition, their website also claims to use digital watermarking technology, which is something that can't be heard by the human ear but exists, and has been tested by a team of professionals, so it can't be heard, but it does exist.

The watermark is mainly used to detect the source of the music, so that the officials can quickly recognize the music.

That's why I rarely download music from here unless I have to.

[nidostar]

Doesn't this take us back to a topic on LMT a year or so ago? I think it was @Psyclon who pointed out that although the files on these sites have a WAV extension it does not necessarily guarantee a totally lossless version of the music. A WAV file is merely a container and could consist of a lo-fi or previously compressed recording. The only way to tell is to put it through a spectrum analyser as you have done in order to establish the recording's true credentials. The expression in English comes to mind - a wolf in sheep's clothing.

[ChunYinZi]

Doesn't this take us back to a topic on LMT a year or so ago? I think it was @Psyclon who pointed out that although the files on these sites have a WAV extension it does not necessarily guarantee a totally lossless version of the music. A WAV file is merely a container and could consist of a lo-fi or previously compressed recording. The only way to tell is to put it through a spectrum analyser as you have done in order to establish the recording's true credentials. The expression in English comes to mind - a wolf in sheep's clothing.

You're right,

At present, my choice is to give priority to music that has been torn by CD if conditions permit.

If it is electronic music, I will download the same music on some different websites, and then compare the spectrum and choose the music with the best spectrum.

[Psyclon]

It is not only the spectrum that is different. Libraries treat their stock ridiculously, schockingly bad if you think their bread and butter is their, well, music library.
I am sure that any LMT user has equal or more understanding of extracting the sound information of a medium than the employees of those libraries.
We had recently two incidents that show that very clearly:

* Poorly ripped CDs with skips and outright errors that are clearly visible in the waveform such as the SONOTON albums.
* Poorly recorded LPs. Unfortunately I forgot the person who recorded the SELECTED SOUND vinyls but it sounds so fantastic, instruments super clear (such as big toms in "BIG CITY HEAT") or the outright pristine sound of "NIGHTCAT" makes me really positively shocked on how great LPs can sound. I always have prejudices against that format and there are many caveats that digital/CDs don't have, but it shows that those records are sure delivering incredibly good sound performance. If I remember the original WAV material from SONOTON which is clearly several steps below..

Maybe they even used software declicker/decrackler or EQs while recording... Uff.

Other incidents are:

* FOCUS MUSIC has chirps in their material, very subtle but you can hear the error correction trying to restore the audio. NICHION has the FOCUS music in perfect condition.
* KPM has stained, dirty inlays as their cover art that they even put up on Spotify and their distributors.

This makes me think that SONOTON and KPM have either only one copy left over and if that is damaged, it's over. Or they don't care to go into their vaults and get the masters. Either case is embarrasing.

Now about the different sound quality, please keep in mind that there are so many different interfaces and settings that can change the outcome.

CDs can be ripped with fast settings and they can bring lots of errors or chirps with them - even from a flawless disk. A few "bad sectors" can be error-corrected by stock players even in the first form of the compact disk "rulebook" (there are different "Books", e.g. "Red Book Standard" is the default CD without any perks). These can be inaudible - or not. If you rip a CD in the "fast" setting, you get small errors, it's kind of collateral damage. If I remember our old PC with its 40x speed CD drive, there were lots of bad sectors afterwards that even the burning programs warned about the questionable file integrity. You'd think the software is working, but it bites often more off than it can chew..

If you do it right, you can set up your recorder/ripper to be very careful. RETROMATIC for example does this, he has logs with every rip he does that show 100% perfect and pristine rips. I assume these libraries have their stock of 100+ disks and want them ripped. So they put a guy in an office and he picks the "fine most of the time" settings for speed reasons. There you go: A chirp here, an odd click there.

Then there is the WAV format which can be different, too. PCM audio (what you usually expect in a WAV file) can be modified a lot. Sometimes, older games had PCM music but are actually ADPCM which tries to predict the needs to save bandwidth and this can be faulty too. Sim City 3000 IIRC uses that format.

Then there is a good idea of ChunYinZi that there is tampering with frequencies and sample rates and whatnot. Not sure how much that "resampling" and "dithering" and all that have an effect, some effects may be practically reversible, some might leave inaudible but visible (in a spectrograph) differences.

I am just throwing some random ideas here that showcases all the little things where changes of the material can happen. But now imagine a combination of the above. And each modification is affecting each other until the point you can see - albeit not necessarily hear - it.

[ChunYinZi]

It is not only the spectrum that is different. Libraries treat their stock ridiculously, schockingly bad if you think their bread and butter is their, well, music library.
I am sure that any LMT user has equal or more understanding of extracting the sound information of a medium than the employees of those libraries.
We had recently two incidents that show that very clearly:

* Poorly ripped CDs with skips and outright errors that are clearly visible in the waveform such as the SONOTON albums.
* Poorly recorded LPs. Unfortunately I forgot the person who recorded the SELECTED SOUND vinyls but it sounds so fantastic, instruments super clear (such as big toms in "BIG CITY HEAT") or the outright pristine sound of "NIGHTCAT" makes me really positively shocked on how great LPs can sound. I always have prejudices against that format and there are many caveats that digital/CDs don't have, but it shows that those records are sure delivering incredibly good sound performance. If I remember the original WAV material from SONOTON which is clearly several steps below..

Maybe they even used software declicker/decrackler or EQs while recording... Uff.

Other incidents are:

* FOCUS MUSIC has chirps in their material, very subtle but you can hear the error correction trying to restore the audio. NICHION has the FOCUS music in perfect condition.
* KPM has stained, dirty inlays as their cover art that they even put up on Spotify and their distributors.

This makes me think that SONOTON and KPM have either only one copy left over and if that is damaged, it's over. Or they don't care to go into their vaults and get the masters. Either case is embarrasing.

Now about the different sound quality, please keep in mind that there are so many different interfaces and settings that can change the outcome.

CDs can be ripped with fast settings and they can bring lots of errors or chirps with them - even from a flawless disk. A few "bad sectors" can be error-corrected by stock players even in the first form of the compact disk "rulebook" (there are different "Books", e.g. "Red Book Standard" is the default CD without any perks). These can be inaudible - or not. If you rip a CD in the "fast" setting, you get small errors, it's kind of collateral damage. If I remember our old PC with its 40x speed CD drive, there were lots of bad sectors afterwards that even the burning programs warned about the questionable file integrity. You'd think the software is working, but it bites often more off than it can chew..

If you do it right, you can set up your recorder/ripper to be very careful. RETROMATIC for example does this, he has logs with every rip he does that show 100% perfect and pristine rips. I assume these libraries have their stock of 100+ disks and want them ripped. So they put a guy in an office and he picks the "fine most of the time" settings for speed reasons. There you go: A chirp here, an odd click there.

Then there is the WAV format which can be different, too. PCM audio (what you usually expect in a WAV file) can be modified a lot. Sometimes, older games had PCM music but are actually ADPCM which tries to predict the needs to save bandwidth and this can be faulty too. Sim City 3000 IIRC uses that format.

Then there is a good idea of ChunYinZi that there is tampering with frequencies and sample rates and whatnot. Not sure how much that "resampling" and "dithering" and all that have an effect, some effects may be practically reversible, some might leave inaudible but visible (in a spectrograph) differences.

I am just throwing some random ideas here that showcases all the little things where changes of the material can happen. But now imagine a combination of the above. And each modification is affecting each other until the point you can see - albeit not necessarily hear - it.

Thank you very much for your excellent answers, in fact you seem to have answered all the questions about sound quality in a very standard and excellent way.

Now, I will share with you two pieces of music

The first one is the music that you get from CD rips and the second one is a WAV of Universal Productions music downloads

I can hear the difference between the two very clearly

That's why I prioritized the CD

The music you get from ripping CDs has a sound as if you were there, as if you were playing it live.

The music from Universal's downloads, however, has a distinctly broken sound and a lot of noise.

The CDs are also better if you compare the spectrograms.

Also, the music is orchestral, so you can visualize the difference in sound quality.

hXXps://mega.nz/folder/FHUgFbIL#BZc7UI4KLXHx0QibzBSh1w

[kpmhill]

I'll just mention briefly again that WAV files have no reason to exist for sharing, and they just impede the process.

FLAC keeps all the info that exists in a WAV.

[nidostar]

I'll just mention briefly again that WAV files have no reason to exist for sharing, and they just impede the process.

FLAC keeps all the info that exists in a WAV.

I couldn't agree more. The same applies to rips taken at more than 16 bit/44.1kHz. But there are those who are tempted by the dark side and feel the bigger the file the better the sound. Wrong, of course!

[nidostar]

Thank you very much for your excellent answers, in fact you seem to have answered all the questions about sound quality in a very standard and excellent way.

Now, I will share with you two pieces of music

The first one is the music that you get from CD rips and the second one is a WAV of Universal Productions music downloads

I can hear the difference between the two very clearly

That's why I prioritized the CD

The music you get from ripping CDs has a sound as if you were there, as if you were playing it live.

The music from Universal's downloads, however, has a distinctly broken sound and a lot of noise.

The CDs are also better if you compare the spectrograms.

Also, the music is orchestral, so you can visualize the difference in sound quality.

I listened to both versions through a pair of decent Sennheiser headphones. The CD version has greater frequency response in the upper register but I found that seemed to make the whole sound appear thin. The Universal download seemed to have greater frequency response at the lower end giving a much fuller, richer sound. I couldn't detect the broken sound or noise you refer to and I reckon I have a keen musical ear. For me the Universal download is the one I'd go for.

I guess it's all down to personal taste though it also depends on what equipment one is listening to the recording. I've maintained for many years there is a point where the more expensive and exclusive audio gear is the better it is at enhancing the non-musical features of a recording thus spoiling the overall experience.

Another interesting point was that there is a glitch in the recording right at the beginning at the 2 second mark which appears on both versions.