Graphs - One is marketed by Qobuz as 24 Bit FLAC, the other is from youtube. Without looking at the file names, can you tell which is which?

[The Torrent]

 

 

I have a level 2 once you crack this one.

[.Apex.]

It says the file name in the image.. doesn't require any brain, but the FLAC is on the right because of the higher frequency range

[The Torrent]

Just now, Syn. said:

It says the file name in the image.. doesn't require any brain, but the FLAC is on the right because of the higher frequency range

Title said "Without looking at the file names"...

 

Its stil cutoff at the same frequency... the bar just extends... anything else....? Out of all the graphs?

[.Apex.]

Just now, The Torrent said:

Title said "Without looking at the file names"...

 

Its stil cutoff at the same frequency... the bar just extends... anything else....? Out of all the graphs?

I wasn't even trying to look for it, I only saw it when I glanced at the frequency which is the main differentiator between MP3 and FLAC

 

Other than that I'm not aware of any indicators that could definitely confirm that it's an MP3 or FLAC

[The Torrent]

2 minutes ago, Syn. said:

I wasn't even trying to look for it, I only saw it when I glanced at the frequency which is the main differentiator between MP3 and FLAC

 

Other than that I'm not aware of any indicators that could definitely confirm that it's an MP3 or FLAC

or even if one of them is 24 bit...

 

I'll continue waiting for others...

[geo3]

The histogram seems to indicate some clipping on the mp3.

[The Flying Sloth]

If you don't want us to look at the file names crop them out silly

[The Torrent]

2 hours ago, The Flying Sloth said:

If you don't want us to look at the file names crop them out silly

im more interested in the method than the test itself.

5 hours ago, geo3 said:

The histogram seems to indicate some clipping on the mp3.

thats what i thought but look at this:

the left is mp3 from youtube the right flac/ Why is the right clipping soooo much compared to the mp3?

[Derkoli]

Guys, please remember that a crappy master on FLAC, is 10x worse than a good master on MP3. Mastering is so much more important than file type.

[The Torrent]

14 minutes ago, Derkoli said:

Guys, please remember that a crappy master on FLAC, is 10x worse than a good master on MP3. Mastering is so much more important than file type.

Is there any way to tell if it's a good or bad master from a graph?

 

I don't even understand why they make several masters of the same song...

[Arphis]

36 minutes ago, The Torrent said:

I don't even understand why they make several masters of the same song...

Different listening environments, different sources, different audience and to re-release for extra money/legal reasons.

 

If you plan to play the song on the radio you can expect it to be running in a car or maybe at work, alongside some machines or higher ambient noise, while the radio frequency limits the bandwidth and the radio usually isnt the best quality output device. So you can degrade quality or enhance certain features to compensate for that.

 

If you you plan to release your song for streaming you look at the target audience and what they are listening with. If you see the majority will be using mobile phones with "cheap" headphones, you tune the sound to be pleasant on these devices. You may also want to compensate for the streaming services restrictions, equalizing or sound profile, etc.

 

If you go for the hi-fi audio guy to be your listener, you want to offer highest quality for perfect listening environments, like a seperate, acousticly treated room, proper sources and a good signal path with quality speakers at its end.

 

Sometimes a re-master is done to accomodate new technology, Dolby Atmos, 360° audio or just better quality devices. It didnt matter what it sounded like in the 70's because vinyl was everything, but they were a lot of improvements over time.

 

Also legal reasons can be important. Usually a song has mulitple owners over different elements, like lyrics or music and that is true in production and mastering/mixing. Sometimes tracks have to be re-recorded for artists to gain control over "their" music. They will re-master them at the same time.

 

Taylor Swift was not allowed to play her own tracks on stage, since her old label has the copyright ownership of the recordings for her first six albums. It is complex, so you may look it up yourself.

[Derkoli]

1 hour ago, The Torrent said:

Is there any way to tell if it's a good or bad master from a graph?

 

I don't even understand why they make several masters of the same song...

Really, the best way to tell a bad master is via listening.

 

There's a reason that alot of mastering engineers have incredibly accurate/high-end systems and treated rooms.

[geo3]

4 hours ago, The Torrent said:

im more interested in the method than the test itself.

thats what i thought but look at this:

the left is mp3 from youtube the right flac/ Why is the right clipping soooo much compared to the mp3?

Well these don't appear to be the same recording. Here the flac is clearly over compressed. I assumed the first pair were the same recording just different file formats.

[HM-2]

As one of the posters above says, a lossless format won't save a dog-turd mastering job. Sadly producers have taken to compressing the living shit out of everything these days.

[The Torrent]

39 minutes ago, geo3 said:

Well these don't appear to be the same recording. Here the flac is clearly over compressed. I assumed the first pair were the same recording just different file formats.

They are the same song but very possibly different masters. So you think the flac in this case is very poorly mastered?

 

[geo3]

7 minutes ago, The Torrent said:

They are the same song but very possibly different masters. So you think the flac in this case is very poorly mastered?

 

A good master should never clip. But I can't really evaluate it's quality otherwise. 

[Spuriae]

2 hours ago, The Torrent said:

They are the same song but very possibly different masters. So you think the flac in this case is very poorly mastered?

The flac is 7db louder than the mp3. The peak on the mp3 is at -5dBfs. Something is wrong with the levels in both files.

[The Torrent]

1 hour ago, Nimrodor said:

The flac is 7db louder than the mp3. The peak on the mp3 is at -5dBfs. Something is wrong with the levels in both files.

hmmm

 

Also im noticing this clipping issue on all flac songs.

 

 

e.g this above..

 

for some reason, theres always waay less clipping on the youtube compared to the FLAC versions of any songs which have loads...

[The Torrent]

3 hours ago, geo3 said:

A good master should never clip. But I can't really evaluate it's quality otherwise. 

seems that loads of flac copies of songs have clipping all over the histograms when u compare to the youtube. Any clue why this could be?

[Derkoli]

13 minutes ago, The Torrent said:

seems that loads of flac copies of songs have clipping all over the histograms when u compare to the youtube. Any clue why this could be?

It's most likely due to the levels of everything. The crest factor will have little effect on this, so ignore that. You want a crest factor of 10-20dB in most music anyway. So it's irrelevant in this case.

 

Youtube tend's to try to level everything off to one reasonable level, so user's don't get bleeding ear drums from that one a-hole youtuber who has an intro that's louder than a saturn V rocket. Try and find a video that's nice and loud to compare to.