Why do we use lossless?
Apr 9, 2011 at 9:11 PM Thread Starter Post #1 of 88

Gallictic

New Head-Fier
Joined
Apr 9, 2011
Posts
6
Likes
0
Hi guys, I'm still trying to discover the various types of sounds and how they all work together on a whole so go easy on me.
 
Throughout my quest for knowledge, I've seen thousand and thousands of posts saying the lossless (FLAC) is the best way to go and we lose sound quality with lossy music ect ect.
 
My question is how is lossless better than lossy. This is not a debate about convenience or portability. I want to know why do we need to listen to something that extends beyond the human range of hearing frequencies (20 Hz - 20 kHz).
 
To my understanding, most high end lossy algorithms cut off frequencies that cannot be heard by humans (320kbps mp3 cuts off at around 20 kHz) .
 
Is that just wrong?
 
If its right then why do we need lossless if we cannot really hear the frequencies anyway and they are just taking up space? Am I just missing something here?
 
TL;DR
 
Why lossless over lossy when looking at frequencies?

Cheers!
 
PS: If you need clarification on what im asking then I'll be glad to clarify. I wrote this in a hurry.
 
 
 
 
 
 
Apr 9, 2011 at 9:49 PM Post #2 of 88

SDDL-UP

Head-Fier
Joined
Sep 26, 2008
Posts
67
Likes
11
I'm buying a CD, and I figure why not put 100% of it into my music collection? Why would I settle for 85% of it, even if I can't tell a difference? Just getting what I paid for!

I would hate to be a person that upgrades something in their playback chain only to discover it exposes a lossy format!

There is simply NO NEED for a lossy format IMO.
 
Apr 9, 2011 at 10:24 PM Post #3 of 88

MacedonianHero

Headphone.Guru Editor
Joined
Feb 26, 2009
Posts
17,961
Likes
4,206
Location
Toronto, Canada
Lossless when re-translated back digitally it is a bit for bit identical translation to the original recording. Lossy music simply is not....it has data missing. The lower the rate 128 vs. 256 vs 320 the more data bits that are missing (and therefore worse sound quality).

Here's a good source:

http://www.highdefdigest.com/news/show/1233
 
Apr 9, 2011 at 10:31 PM Post #4 of 88

Steve Eddy

Member of the Trade: The Audio Guild
Aka: TempAccount555
Joined
Sep 28, 2003
Posts
6,609
Likes
546


Quote:
Originally Posted by Gallictic /img/forum/go_quote.gif
 
To my understanding, most high end lossy algorithms cut off frequencies that cannot be heard by humans (320kbps mp3 cuts off at around 20 kHz) .

 
Lossy compression schemes don't simply limit bandwidth (by the way, CD's cut off at around 20kHz as well). There's much more to it than that. And it's audible.
 
Lossless compression schemes allow you to retain all the information while still reducing the storage requirements, though not as much as lossy schemes.
 
se
 
 
 
 
 
Apr 9, 2011 at 10:31 PM Post #5 of 88

nick_charles

Headphoneus Supremus
Joined
Feb 26, 2008
Posts
3,180
Likes
335


 
Quote:
Hi guys, I'm still trying to discover the various types of sounds and how they all work together on a whole so go easy on me.
 
Throughout my quest for knowledge, I've seen thousand and thousands of posts saying the lossless (FLAC) is the best way to go and we lose sound quality with lossy music ect ect.
 
My question is how is lossless better than lossy. This is not a debate about convenience or portability. I want to know why do we need to listen to something that extends beyond the human range of hearing frequencies (20 Hz - 20 kHz).
 
Lossy compression is a bit more complicated than just truncating high frequencies, many codecs/bit-rates cut off long before 20K with minimal impact, but all rely heavily on masking, this on a slice by slice basis analyses the frequencies and their amplitudes and removes frequencies that are low enough in amplitude to be inaudible due to being masked by adjacent frequencies, so the dominant frequencies are passed through and others (perceptually inaudible) are hacked out - this process can be so good as to allow the codec to remove 4/5th of the data and stilll be perceptually transparent (most of the time)
 
 
 
To my understanding, most high end lossy algorithms cut off frequencies that cannot be heard by humans (320kbps mp3 cuts off at around 20 kHz) .
 
Is that just wrong?
 
If its right then why do we need lossless if we cannot really hear the frequencies anyway and they are just taking up space? Am I just missing something here?
 
TL;DR
 
Why lossless over lossy when looking at frequencies?

Cheers!
 
PS: If you need clarification on what im asking then I'll be glad to clarify. I wrote this in a hurry.
 
 
 
 
 



 
 
Apr 9, 2011 at 10:44 PM Post #6 of 88

Gallictic

New Head-Fier
Joined
Apr 9, 2011
Posts
6
Likes
0


Quote:
 
Lossy compression schemes don't simply limit bandwidth (by the way, CD's cut off at around 20kHz as well). There's much more to it than that. And it's audible.
 
Lossless compression schemes allow you to retain all the information while still reducing the storage requirements, though not as much as lossy schemes.
 
 
 
 

 
 
 
This is great info thanks!

 
 
Apr 9, 2011 at 10:50 PM Post #7 of 88

Gallictic

New Head-Fier
Joined
Apr 9, 2011
Posts
6
Likes
0
Quote:Originally Posted by Gallictic 

Hi guys, I'm still trying to discover the various types of sounds and how they all work together on a whole so go easy on me.
 
Throughout my quest for knowledge, I've seen thousand and thousands of posts saying the lossless (FLAC) is the best way to go and we lose sound quality with lossy music ect ect.
 
My question is how is lossless better than lossy. This is not a debate about convenience or portability. I want to know why do we need to listen to something that extends beyond the human range of hearing frequencies (20 Hz - 20 kHz).
 
Lossy compression is a bit more complicated than just truncating high frequencies, many codecs/bit-rates cut off long before 20K with minimal impact, but all rely heavily on masking, this on a slice by slice basis analyses the frequencies and their amplitudes and removes frequencies that are low enough in amplitude to be inaudible due to being masked by adjacent frequencies, so the dominant frequencies are passed through and others (perceptually inaudible) are hacked out - this process can be so good as to allow the codec to remove 4/5th of the data and stilll be perceptually transparent (most of the time)
 
 
 
To my understanding, most high end lossy algorithms cut off frequencies that cannot be heard by humans (320kbps mp3 cuts off at around 20 kHz) .
 
Is that just wrong?
 
If its right then why do we need lossless if we cannot really hear the frequencies anyway and they are just taking up space? Am I just missing something here?
 
TL;DR
 
Why lossless over lossy when looking at frequencies?

Cheers!
 
PS: If you need clarification on what im asking then I'll be glad to clarify. I wrote this in a hurry.
 
-
-
-
-
-
 
Im guessing that some low standard headphones will skip the missing frequencies because they are masked by the more dominant ones?
Will one notice the difference between a FLAC and 320 mp3 if they are just listening casually? or does it take a well trained ear?
 
 

 
Apr 9, 2011 at 11:56 PM Post #8 of 88

crayonhead

100+ Head-Fier
Joined
Mar 19, 2011
Posts
235
Likes
12
it really depends. I have one good pair of balanced earphones (ATH-AD700) and I use for listening on the go. If you listen to music constantly and use an amp, the difference can really be heard with the higher and lower frequencies. For example, Gorillaz... on plastic beach (flac format), I can hear minute differences from the itunes played version, and i'm also using a PA2V2 amp. But if you listen casually, not much would change in your perception. The only way to bring out the differences between lossy and lossless is to run it through an amp and then listen to it.
 
Apr 10, 2011 at 12:54 AM Post #9 of 88

Uncle Erik

Uncle Exotic
Joined
Mar 18, 2006
Posts
22,596
Likes
506
Practically speaking, when you can get 1TB of storage for around $50, there's no point in ripping lossy files. It was more important back when drive space cost more.

I've seen arguments about the audibility of various lossy schemes, but why not just put the whole disc on your computer? Storage is cheap and I don't see any drawbacks to having lossless files.
 
Apr 10, 2011 at 1:25 AM Post #11 of 88

Zabardo

New Head-Fier
Joined
Apr 7, 2011
Posts
42
Likes
13
To answer the OP I wanna say because we can but in reality we use lossless because its the closest we can get digitally to an actual CD and while lossy was great for portability and size before, right now we have the means to have everything on lossless plus, in some bass heavy genres you have a noticeable cut when playing the songs on a lossy like DnB.
 
Bottom line if you have a proper equipment you can take advantage on lossless just as much you can take on CDs.
 
Apr 11, 2011 at 7:06 PM Post #12 of 88

Gallictic

New Head-Fier
Joined
Apr 9, 2011
Posts
6
Likes
0
So how about headphones then?
 
The Shure HD 800 has a minimum frequency of 6 ​Hz and a maximum of 51 ​kHz. One cannot hear such low and high frequencies.
Is there another factor that justifies such extended frequencies again?
 
Apr 11, 2011 at 8:07 PM Post #14 of 88

tranhieu

Member of the Trade: DIY Amp Builder - AKA kytuphicanh
Caution: Incomplete Sales
Joined
Jul 15, 2008
Posts
1,150
Likes
94


Quote:
So how about headphones then?
 
The Shure HD 800 has a minimum frequency of 6 ​Hz and a maximum of 51 ​kHz. One cannot hear such low and high frequencies.
Is there another factor that justifies such extended frequencies again?



to minimize the effect of roll off at 2 ends.
that has nothing to do with frequency range of lossless file though. there are many iem out there cannot reach 20khz or even if they do, they already rolloff long before reaching 20khz. as already being said above, storing lossless files is mostly for the peace of mind.
 
Apr 11, 2011 at 8:13 PM Post #15 of 88

jman1503

New Head-Fier
Joined
Dec 2, 2010
Posts
19
Likes
0
I may be wrong, but don't we use lossless because it sounds exactly like the original? Perhaps regarding the frequencies, there are sounds that we percieve but do not hear which adds to the listening experience.  This is a high frequency eqivalent to tactile sound perhaps?
 

Users who are viewing this thread

Top