[liamstrain]

Edited the quoted post for clarity - I was referring to a modification of the metaphor for lossy formats. Moodyrn - yes, yours was simpler, but also misleading. Just trying to strike a balance.

 

[Brooko]

What amuses me with all of this debate, is the presumption that the OP (and one or two posters) can easily (think 'night and day') tell all MP3 / AAC / lossy from lossless.  In my experience - most people who claim this have never conducted a properly controlled abx test.  When confronted with true blind testing - at aac256 / 320mp3 (or V0), the ability to pick the differences 'magically' disappears.
 
What is most disappointing is that the same people will make sweeping statements without ever knowing their own limitations.  Thereby influencing a whole new unsuspecting generation of audio enthusiasts - and thus the myth is perpetuated.
 
Disclaimer - i do know of a few (and the number is very small) people who can successfully discern aac256 from lossless - but even they admit it is not 'night and day'.  The differences are very small and generally they have well trained ears (ie they know what to look for).
 
Roller's post #46 above is a very good one that explains why you should rip to a lossless codec.  For home based systems, where space is not an issue, having a lossless rip simply means that as lossy algorithms get better - you don't need to re-rip to re-encode.  And for the few who can actually discern a difference - space is cheap, why not have lossless on your main system.
 
For portable though - and this was part of the OPs question - honestly - why would you bother with lossless .... ?  Usually with portable, you are in an often noisy environment, where you are not concentrating on the music alone - and for most people, they won't be ale to tell the difference anyway.
 
Assuming the OP can hear a night and day difference between his MP3 and lossless as he states, it's probably because:
 - they were different recordings (ie one an mp3 he either downloaded or bought, and the 2nd a flac file from another source either downloaded or bought.
 - they were not level matched (a slightly louder recording will generally 'sound superior' to us - giving rise to the notion of superior dynamics / soundstage etc)  This disappears when they are the same volume.
 - the original files he had may have been really crappy 128kbps or lower with audible artifacts
However - none of these factors can lead to a claim that for him all lossless files are going to better than lossy, and that he can tell the difference.
 
Test for the OP
If the OP and anyone else who hasn't tried a proper abx - all you need is a good ripper, a good encoder, a piece of music you know really well, Foobar2000 and the comparator (abx) plugin.  Take a CD you know really well.  Rip it to lossless.  Take the lossless file, and using a good transcoder, transcode to either MP3 320 or aac256.  Open the two files in the Foobar2000 abx comparator, use the level matching tool, and perform 15-20 tests.  It's better if you turn off viewing the results real time - so you only see the results at the end of the trial.  The trial allows you to listen to A, listen to B, then listen to X.  You don't know which file is which - and you have to match X to either A or B.  Assuming that the files are from the same source CD, ripped and encoded properly so no introduced artifacts, MOST people will not be able to reliably tell the difference.  A log file is generated at the end - which you can then post if you are game.
 
I'm not trying to rain on anyone's parade here - but the notion that most people can easily discern aac256/320mp3 (or even lower a lot of the time) in a blind test is simply incorrect.  That's why aac256 and 320mp3 are good enough on portable players - and why a lot of people even use them on their main systems.
 
Ignorance is bliss - but if we are prepared to offer advice to others - we should at least test ourselves to make sure we are not perpetuating a continued myth.
 
Final note - I've tested myself many times, and I don't have 'bad' gear.  Personally aac192 is transparent enough - but I use aac256 vbr (a vbr rate that averages a bit over 200) for my iPod touch.  With 32Gb I can actually have almost all of my library with me.  Because I know I can't discern the difference personally - it's liberating.  before I tried the tests, I always wondered if I was missing something.  Now I don't.

 

[moodyrn]

Edited the quoted post for clarity - I was referring to a modification of the metaphor for lossy formats. Moodyrn - yes, yours was simpler, but also misleading. Just trying to strike a balance.

Saying my quote was misleading is entirely incorrect and uncalled for. What was misleading about it? It's one thing to put it as simpler, but to go all the way to misleading when we are basically saying the same thing is something entirely different and petty. Now this thread has been derailed for no reason at all. You took it to a whole nother level calling my quote misleading when it was anything but.

 

[Porter1]

Quote:

<snip>

 
great post.  lots of well-articulated ideas for us newcomers to consider.  i personally cannot discern between aac256 and WAV.  like you said, it's liberating to know i'm not missing out on anything.

 

[cel4145]

Maybe we should do a Head-Fi Lossy Vs. Lossy Listening Test?
 
Start a thread in the Sound Science forum with a few music files loaded with MP3, AAC, and WAV versions, but without saying which is which. Invite people to do their own listening tests and post their scores. Then after a few weeks, reveal the results. Not as perfect as ABX testing, but might be surprising to some people. And it would be fun 

 
So something like Girl Talk's "On and On" (from All Day) could be used since it's Creative Commons licensed. Just need a few other CC licensed songs. 

 

[liamstrain]

Quote:

Saying my quote was misleading is entirely incorrect and uncalled for. What was misleading about it? It's one thing to put it as simpler, but to go all the way to misleading when we are basically saying the same thing is something entirely different and petty. Now this thread has been derailed for no reason at all. You took it to a whole nother level calling my quote misleading when it was anything but.

 
To be honest here, I wasn't trying to call you out as being deliberate about being misleading. I was just trying to provide a clarification in the metaphor so as to give a more accurate picture without sacrificing its elegant simplicity. But since you're going to get touchy about it there we are. 
 
Misleading because it gives the impression that a lossy format cuts in half the audible information/that a lossless format is 50% or better sounding than a lossy one as a result. 
 
Misleading because it doesn't deal at all with what a lossy compression works with, and can lead people to erroneously conclude that they simply throw away information willy nilly.
 
Misleading precisely because it was an over simplification, which ended up giving the wrong impression of the difference between the two as more vast and damaging than it really is.  
 
Whether you have some axe to grind, or no, I feel like your metaphor was misleading to a point worth clarifying. That is all. 

 

[moodyrn]

Are you serious. It's obvious I was using an illustration in simple terms to explain the fundlemental difference between lossy and lossless compressions. I even stated as such. And also stated It wasn't in absolute terms. And also the gist of my illustration is that lossless shrinks the file without removing any data and lossy compression actually shrinks the file by removing data. Those are the two main things that separates the two. Which is a fact. Now your statement doesn't much merit. Because not all lossy compressions are created equal. Some not only removes 50% of data(which is something I "did not" say anyway). But some lossy compressions removes even more than that. For a long time people ripped files to 64kbs and I've even come across some files at 32kbs. And even 128kbs removes more than 50% of data. Why do you think to file is so small? Now most of that data can be referred to as wasted bits of information but it still gets removed nonetheless.

So that's why I used a simple illustration, because not all lossy compressions are created equal. The fact that lossless compressions removes no data and lossy compressions does is a fact that applies to all lossless and lossy compressions. Only you chose to make a big deal out of nothing while making false accusations of what I said by putting words in my mouth. Show me where I said a lossy compression actually cuts a file in half anyway. I never said that. That's you taking my statements and twisting them into meaning what you want them to mean. And show me where I said a lossless compression sounds 50% better than a lossy compressions. Now I can show you posts I made about lame v0 or 320 bitrate and lossless sounding very close and that I couldn't perceive a difference on any portable device. I can also show you where I said you need a very resolving system to hear the difference between the two. But yet and still you put words in my mouth when I was only using a piece a paper no less for an illustration. Not once did I go into the technical aspect of the differences between the two. So get your facts and straight.

 

[skamp]

And even 128kbs removes more than 50% of data.

[…]

Show me where I said a lossy compression actually cuts a file in half anyway. I never said that.

[…]

But yet and still you put words in my mouth when I was only using a piece a paper no less for an illustration.

I think liamstrain's modification of your illustration is more accurate. You completely overreacted, let it goooo!

 

[moodyrn]

That comment came "after" he accused me of saying it and was a response of his quote. Show me in the post where he accused me of saying it where I actually said it. Your quote is very misleading . And I never said anything in reference to his illustration being more or less accurate. It was in response of him saying I was being misleading when we both were saying in essence the same thing. And you tell me to let it go when you are the one keeping this going? Why don't you both try letting it go. This thread has been derailed enough. Lets see if you both can let it go.

 

[cel4145]

Quote:

It was in response of him saying I was being misleading when we both were saying in essence the same thing. 

 
Oh, come on. He tried to improve on a analogy that really doesn't work. It's a minor improvement, but the analogy is still flawed. Why act so defensive about an idea that's not that great to begin with? Your analogy is more akin to having the mp3 file compression truncate the second half of the song. 

 
So instead of all this argument over a mediocre analogy, why don't you guys try to come up with a better one? 

 

[castleofargh]

"and here we are again tonight for a new ultimate fighting championship" (voice tone not included)
 
 

 

[streetdragon]

hmm this is the analogy i can think of:
raw data: red dot,dark red dot, blue dot, light red dot, dark blue dot, red dot, dark blue dot, blue dot, light blue dot, dark blue dot, blue dot,red dot, light red dot,light blue dot.
lossless data:|light red dot @ 4,13|red dot@1,6,12|dark red dot@2|light blue dot@9,14|blue dot@3,8,11|dark blue dot@5,7,10|
lossy data: |blue dots@3,5,7,8,9,10,11,14|red dots@1,2,4,6,12,13|

well i guess the length of the text indicates how much data is used?

please don't shoot me down if this analogy is flawed, just scroll past or correct it if it is so

 

[Achmedisdead]

In relation to thread title....
Why on earth do people still use caps lock? 

 

[Happy Camper]

Gents, I guess my comment about using wav vs flac may have stirred the kettle. I am aware of flac and it's method of recording data. A friend of mine ripped all his albums to wav and looks down at flac as a lossy format. I am in disagreement with him but that's his belief and I won't change it. When I read the quoted statement, I immediately thought of my friend's opinion.

For portable players, I've settled on 192k for the majority and 320 for my reference pieces. I'll still use flac on the main rig as it's much better gear but it's still mighty hard to discern a difference. I think I may be deciding based on knowing the bitrate when I play it and trying to find that one flaw that contributes to my decision to proclaim the higher bitrate better. Hell, with my pop 60s music, 128 works just fine in the car or on a walk.

If I were to use my ipod as my top end source, I'd convert everything to alac but I don't. It's a convenience piece for lifestyle flexibility. That's what the lossy formats are for. The OP seems to wear flac as a badge of credibility by slamming MP3.

 

[chewy4]

Quote:

hmm this is the analogy i can think of:
raw data: red dot,dark red dot, blue dot, light red dot, dark blue dot, red dot, dark blue dot, blue dot, light blue dot, dark blue dot, blue dot,red dot, light red dot,light blue dot.
lossless data:|light red dot @ 4,13|red dot@1,6,12|dark red dot@2|light blue dot@9,14|blue dot@3,8,11|dark blue dot@5,7,10|
lossy data: |blue dots@3,5,7,8,9,10,11,14|red dots@1,2,4,6,12,13|

well i guess the length of the text indicates how much data is used?

please don't shoot me down if this analogy is flawed, just scroll past or correct it if it is so

That's definitely a good example for lossless compression.
 
Not quite sure how lossy compression works myself though, as in how it chooses which data to throw out. 
 
 

Gents, I guess my comment about using wav vs flac may have stirred the kettle. I am aware of flac and it's method of recording data. A friend of mine ripped all his albums to wav and looks down at flac as a lossy format. I am in disagreement with him but that's his belief and I won't change it. When I read the quoted statement, I immediately thought of my friend's opinion.
For portable players, I've settled on 192k for the majority and 320 for my reference pieces. I'll still use flac on the main rig as it's much better gear but it's still mighty hard to discern a difference. I think I may be deciding based on knowing the bitrate when I play it and trying to find that one flaw that contributes to my decision to proclaim the higher bitrate better. Hell, with my pop 60s music, 128 works just fine in the car or on a walk.
If I were to use my ipod as my top end source, I'd convert everything to alac but I don't. It's a convenience piece for lifestyle flexibility. That's what the lossy formats are for. The OP seems to wear flac as a badge of credibility by slamming MP3.

 
 
Even the differences between 128 and 320 are quite exaggerated I'd say. I can tell the difference(most the time, and only after close examination), but it's certainly not night and day.