or Connect
Head-Fi.org › Forums › Equipment Forums › Sound Science › Hi-Res 24/94 vs Flac vs CD vs Mp3 files download comparison
New Posts  All Forums:Forum Nav:

Hi-Res 24/94 vs Flac vs CD vs Mp3 files download comparison - Page 4

post #46 of 134
Quote:
Originally Posted by SilverEars View Post
 

You two bring out some good points.  Some cans actually give you no choice as they throw details in your face.  This is a flawed test, if you give people with ibuds to the most high-end setup to tell apart two formats.

 

 

that to, there are a huge number of factors to consider, gear is an imporant one, Some headphones, dacs and amps are better at pulling out mirco details than others. I prefer a very dry sounding system, I like those little itty bitty details, and I'm not a huge fan of excessive decay or "warmth" 

Quote:
Originally Posted by cjl View Post
 

See, that strikes me as an extreme exaggeration, if not an outright lie. The difference between 320kbps well encoded MP3 and wav is by no means obvious, and even on the most difficult to encode tracks, it is subtle at best. I agree that it is occasionally (with the right equipment, training in what to listen for, and certain samples) audible, but it is always a minimal and difficult to hear difference.

Now you and Silver Ears make two contrasted points, I listen with a Hifiman HE 4, Beyer Ear pads and a SUPER bright Solid State amp and dac, my set up is pretty aggressive with treble extension and decay, I have bright sounding headphones with brigh sounding amps and dacs, not to mention I know what to listen for, it's subtle but it's distinct, yea for me it's obvious since I've done... like 6 abxs now. Not to mention about every two weeks or so I find a transcode within my own collection, or I'll listen to a song that's listed as flac, playing at 600+ kbs but sounds like an mp3, throw it in the spectrum and now I'm reporting some one for uploading a transcode

 

Quote:
Originally Posted by elmoe View Post


I'm not assuming anything, its in your previous posts... Now your defensive attitude really makes me doubt you even more.
Yeah that's real cute and all but the people arguing in this thread have been at this for decade and heard systems that cost tens if not hundreds of thousands...

my point exactly, your using any and everything you can do dis validate my results 

 

 

So what we have at this very moment is a stalemate, two differing opinions and on that note I'm leaving this thread... I did my abx and as I predicted half of you accept and agree with my results, the other half of you are dead set on calling me a liar. I don't care what you think of me, since I'm the one listening to my music, I just encourage... Oh this is in the sound science forum... lol well then in that case good luck and have fun! 

post #47 of 134
Quote:
Originally Posted by cjl View Post
 

This strikes me as an extreme exaggeration, if not an outright lie. The difference between 320kbps well encoded MP3 and wav is by no means obvious, and even on the most difficult to encode tracks, it is subtle at best. I agree that it is occasionally (with the right equipment, training in what to listen for, and certain samples) audible, but it is always a minimal and difficult to hear difference.

 

I've never seen any reliable proof to refute what you just stated.  Outside of a few killer samples posted in Hydrogen Audio forums, I have seen some Foobar ABX plugin test results that seem to suggest that a listener is able to identify a difference with some regularity, but then nobody else is able to reproduce these results using the same files.  I'm left feeling that there might be something else in the audio chain, no matter how subtle, that is causing some difference to be noticed.  And when the person providing the results has a history of claiming that they can always hear a difference between such files, I become skeptical (the irony of my own personal bias) in the same way that I do when people claim to have seen ghosts.

post #48 of 134

well I don't dismiss people saying they hear it, because there is always the possibility of the source making a mess when extracting the mp3 (or the flac, or both?). then the "perhaps" very audible difference comes from the source doing a bad job instead of the file extension.

post #49 of 134
Quote:
Originally Posted by castleofargh View Post
 

well I don't dismiss people saying they hear it, because there is always the possibility of the source making a mess when extracting the mp3 (or the flac, or both?). then the "perhaps" very audible difference comes from the source doing a bad job instead of the file extension.

 

If the source did a bad job on mp3 playback, the only reason would be latency, and it would do a worse job on FLAC or whatever lossless playback you threw at it... I remain unconvinced, personally. No result whatsoever has been spoken of (did he get 7/10? 8/10? 10/10? how many sessions?) and the arguments are dodgy at best. There is such a lack of description in the abx process that I'm doubting he even tried it.

post #50 of 134

I wish I understood this stuff. Being a complete duffer at maths doesn't help much...

 

But lossy compression is based on perception. I do appreciate that lossless  compression is just maths, of the cryptographic kind. Brilliant, but bottom line is numeric and, assuming it works, the numbers in equal the numbers out. Perceptual compression requires genius of a different order: an understanding of what and how we hear, and what can be removed from it. And then comes the maths and the coding.

 

Yes, I know that every person reading this knows all that, but it seems to me that people do not really think about the psychoacoustic reality. Can't say I ever did much, and always preferring to have losslessly-compressed music didn't give me much of an incentive.

 

Something that I read recently, and have half forgotten so can't repeat properly (I don't know where: a few posts back, here, maybe :o ) is that perceptual compression must always include the perceiver and, of course, not all perceivers are the same. Never mind the golden-eared thing,  the suggestion was that someone with poor hearing, lacking certain frequencies, might be far better equipped to detect the difference between, eg, flac and 320-MP3. What they perceive as being masked will not be the same as an "average" person.

 

It seems pointless the argue the 320-MP3 versus uncompressed thing. My attitude to it is that it is an absolute genius accomplishment that music can be compressed in such a way that most people cannot detect the difference. Most is not all, even if it is a large percentage. MP3 of any bit rate, has never pretended to output the same data. We know it doesn't. Let's not argue the point with the few who claim that they can tell the difference: it should not be an issue. If we must spend our energy on fact v. fiction, we might do better to stick to sample rates, flac v wav, etc etc etc. There are plenty of choices!

 

I'll go on taking the music I want to hear in whatever format I can get it in, but I'll also go on preferring uncompressed or lossless compression when I have a choice. But I'm not claiming I can hear the difference!

post #51 of 134
Quote:
Originally Posted by ab initio View Post
 

foobar's (or any other) ABX testing plugin is the best way to determine whether or not one can actually hear a difference between any of the tracks. Simply shuffling the play list will not permit the short switching times necessary to overcome the short audio-memory that we (humans) have, nor will it prevent non-auditory cues from poisoning a fair and unbiased comparison.

 

Cheers

 

I don't exactly what you mean, but would this kind of test work for you? 

 

http://goo.gl/Dfawrb

 

Have a look at this one too (http://goo.gl/Xin6xt), thought the clips are longer and the approach is different.

______________

cdvsmp3.wordpress.com


Edited by cdvsmp3 - 6/9/14 at 1:12am
post #52 of 134
Quote:
Originally Posted by cdvsmp3 View Post
 

 

I don't exactly what you mean, but would this kind of test work for you? 

 

http://goo.gl/Dfawrb

 

Have a look at this one too (http://goo.gl/Xin6xt), thought the clips are longer and the approach is different.

foo bar has an ABX plugin, listening to files via streaming is not the right way to compare, both of those files are most likely already compressed for use online. I doubt those files are actually streaming in 16bit lossless audio, it's to large 

 

that said using an ABX plugin with FooBar would be a good start 

post #53 of 134
Quote:
foo bar has an ABX plugin, listening to files via streaming is not the right way to compare, both of those files are most likely already compressed for use online. I doubt those files are actually streaming in 16bit lossless audio, it's to large that said using an ABX plugin with FooBar would be a good start 

 

 I can guarantee the online files are in the right format, and they are not compressed, I am not that naive! The test would not make any sense otherwise! I am paying a service to save these otherwise huge files! You can download one of the samples to check this for yourself. And they are streaming .wav (in the CD vs AAC 256k) and .m4a (ALAC, i.e. Apple lossless) in the third (HD vs AAC 256k) (you just need about 200k of bandwidth to do it, which is much less than any HD viideo in YouTube.)

 

______________

cdvsmp3.wordpress.com


Edited by cdvsmp3 - 6/9/14 at 1:10am
post #54 of 134

Take a look at this experiment I did with an HD track. It certainly is revealing.

 

1. Pick an HD track that is sampled at, say, 96k/24bit.

2. Convert it to 44.1/16bit PCM (.wav)
3. Upsample the .wav file to the original 96k.
4. Invert one of the waves, say, the converted and upsampled one.
5. Mix the two tracks and listen to the result. VERY IMPORTANT! You must align the tracks PERFECTLY for the experiment to work.

I did it, and here are the results:

First, take a look at the spectrograms:

 

Original 96k/24bit track

 

Converted 44.1k/16bit track upsampled to 96k

 

Mixed track (What you see is what you cut when you downsample the track. There is certainly a huge amount of data there!)

 

Now listen to the mixed track[/URL] (that's exactly what you are losing when you convert this track from HD to CD). Surprised? If you don't believe it, try it yourself! (and remember to perfectly align the tracks when you mix them).

 

_________________

cdvsmp3.wordpress.com

post #55 of 134

That's very interesting.

 

If I were 12 I might be able to hear something in that mixed track, but even a somewhat younger man with better hearing than mine would probably also hear the silence I hear.

 

Technically, I wonder what that relatively loud (green) sound at about 22khz is?

 

Thank you for the experiment. I'm amazed that you achieved such nulling of the audible content, not because I didn't believe that would be the result, but because experiments, in practice, are not always perfect. Congatulations :)

post #56 of 134
Quote:
Originally Posted by Thad-E-Ginathom View Post
 

That's very interesting.

 

If I were 12 I might be able to hear something in that mixed track, but even a somewhat younger man with better hearing than mine would probably also hear the silence I hear.

 

Technically, I wonder what that relatively loud (green) sound at about 22khz is?

 

Thank you for the experiment. I'm amazed that you achieved such nulling of the audible content, not because I didn't believe that would be the result, but because experiments, in practice, are not always perfect. Congatulations :)

Thank you! The output was quite of a surprise for me too! How will HD audio fans manage to hide this kind of evidence?  :)

post #57 of 134

Easy. They'll repeat all the stuff about how stuff we can't actually hear might be affecting us in some way, or how they'd rather have all the music even if they can't hear it or... well, you know, the usual arguments.

 

Undoubtedly, as you say, there is a considerable amount over the 22khz line in the 96k recording. I can't help wondering what it is. I suppose one could take the difference file and play it back an octave lower, but I'm not sure that would answer the question as to whether it is just noise or some sort of useful musical content. That we can't hear anyway, so useful is the wrong word.


Edited by Thad-E-Ginathom - 6/12/14 at 9:32am
post #58 of 134

Well, I took the difference file, dropped it by 5 octaves, and it does sound like the content is at least musically related (it had some semblance of a pitch, and definitely seemed to have a beat rather than being pure noise), but there really isn't much there. I also had to really crank the volume before I heard much, even with the tone dropped 5 octaves.


Edited by cjl - 6/12/14 at 10:39am
post #59 of 134

That's interesting! I'll see if I can try something similar... just for the fun of it. We are not missing much, that's for sure.

post #60 of 134

By the way, I am engaged in personal project to see how good people are at telling CD quality and iTunes plus apart. Would you be kind enough to have a look at it? cdvsmp3.wordpress.com

New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Sound Science
Head-Fi.org › Forums › Equipment Forums › Sound Science › Hi-Res 24/94 vs Flac vs CD vs Mp3 files download comparison